diff --git a/.github/workflows/_lint.yml b/.github/workflows/_lint.yml
index 1a01b225a0..64169ce0be 100644
--- a/.github/workflows/_lint.yml
+++ b/.github/workflows/_lint.yml
@@ -87,7 +87,7 @@ jobs:
python-version: ${{ matrix.python-version }}
poetry-version: ${{ env.POETRY_VERSION }}
working-directory: ${{ inputs.working-directory }}
- cache-key: lint
+ cache-key: lint-with-extras
- name: Check Poetry File
shell: bash
@@ -102,9 +102,17 @@ jobs:
poetry lock --check
- name: Install dependencies
+ # Also installs dev/lint/test/typing dependencies, to ensure we have
+ # type hints for as many of our libraries as possible.
+ # This helps catch errors that require dependencies to be spotted, for example:
+ # https://github.com/langchain-ai/langchain/pull/10249/files#diff-935185cd488d015f026dcd9e19616ff62863e8cde8c0bee70318d3ccbca98341
+ #
+ # If you change this configuration, make sure to change the `cache-key`
+ # in the `poetry_setup` action above to stop using the old cache.
+ # It doesn't matter how you change it, any change will cause a cache-bust.
working-directory: ${{ inputs.working-directory }}
run: |
- poetry install
+ poetry install --with dev,lint,test,typing
- name: Install langchain editable
working-directory: ${{ inputs.working-directory }}
diff --git a/.github/workflows/_pydantic_compatibility.yml b/.github/workflows/_pydantic_compatibility.yml
index 7d8fe26d92..94d362f327 100644
--- a/.github/workflows/_pydantic_compatibility.yml
+++ b/.github/workflows/_pydantic_compatibility.yml
@@ -79,3 +79,15 @@ jobs:
- name: Run pydantic compatibility tests
shell: bash
run: make test
+
+ - name: Ensure the tests did not create any additional files
+ shell: bash
+ run: |
+ set -eu
+
+ STATUS="$(git status)"
+ echo "$STATUS"
+
+ # grep will exit non-zero if the target message isn't found,
+ # and `set -e` above will cause the step to fail.
+ echo "$STATUS" | grep 'nothing to commit, working tree clean'
diff --git a/.github/workflows/_test.yml b/.github/workflows/_test.yml
index 76d86a2862..04be6a2c39 100644
--- a/.github/workflows/_test.yml
+++ b/.github/workflows/_test.yml
@@ -43,3 +43,15 @@ jobs:
- name: Run core tests
shell: bash
run: make test
+
+ - name: Ensure the tests did not create any additional files
+ shell: bash
+ run: |
+ set -eu
+
+ STATUS="$(git status)"
+ echo "$STATUS"
+
+ # grep will exit non-zero if the target message isn't found,
+ # and `set -e` above will cause the step to fail.
+ echo "$STATUS" | grep 'nothing to commit, working tree clean'
diff --git a/.github/workflows/langchain_ci.yml b/.github/workflows/langchain_ci.yml
index 125f4b3147..f184af9772 100644
--- a/.github/workflows/langchain_ci.yml
+++ b/.github/workflows/langchain_ci.yml
@@ -83,3 +83,15 @@ jobs:
- name: Run extended tests
run: make extended_tests
+
+ - name: Ensure the tests did not create any additional files
+ shell: bash
+ run: |
+ set -eu
+
+ STATUS="$(git status)"
+ echo "$STATUS"
+
+ # grep will exit non-zero if the target message isn't found,
+ # and `set -e` above will cause the step to fail.
+ echo "$STATUS" | grep 'nothing to commit, working tree clean'
diff --git a/.github/workflows/langchain_experimental_ci.yml b/.github/workflows/langchain_experimental_ci.yml
index 6e07a5ae1b..c4c4a039c1 100644
--- a/.github/workflows/langchain_experimental_ci.yml
+++ b/.github/workflows/langchain_experimental_ci.yml
@@ -115,3 +115,15 @@ jobs:
- name: Run extended tests
run: make extended_tests
+
+ - name: Ensure the tests did not create any additional files
+ shell: bash
+ run: |
+ set -eu
+
+ STATUS="$(git status)"
+ echo "$STATUS"
+
+ # grep will exit non-zero if the target message isn't found,
+ # and `set -e` above will cause the step to fail.
+ echo "$STATUS" | grep 'nothing to commit, working tree clean'
diff --git a/.github/workflows/scheduled_test.yml b/.github/workflows/scheduled_test.yml
index b71eee0592..7ce59d5b69 100644
--- a/.github/workflows/scheduled_test.yml
+++ b/.github/workflows/scheduled_test.yml
@@ -47,3 +47,15 @@ jobs:
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
run: |
make scheduled_tests
+
+ - name: Ensure the tests did not create any additional files
+ shell: bash
+ run: |
+ set -eu
+
+ STATUS="$(git status)"
+ echo "$STATUS"
+
+ # grep will exit non-zero if the target message isn't found,
+ # and `set -e` above will cause the step to fail.
+ echo "$STATUS" | grep 'nothing to commit, working tree clean'
diff --git a/docs/api_reference/guide_imports.json b/docs/api_reference/guide_imports.json
index f35c7805a0..8e4d0fed32 100644
--- a/docs/api_reference/guide_imports.json
+++ b/docs/api_reference/guide_imports.json
@@ -317,7 +317,7 @@
"Chatbots": "https://python.langchain.com/docs/use_cases/chatbots",
"Summarization": "https://python.langchain.com/docs/use_cases/summarization",
"Extraction": "https://python.langchain.com/docs/use_cases/extraction",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"Tagging": "https://python.langchain.com/docs/use_cases/tagging",
"Code Understanding": "https://python.langchain.com/docs/use_cases/code_understanding",
"AutoGPT": "https://python.langchain.com/docs/use_cases/autonomous_agents/autogpt",
@@ -400,7 +400,7 @@
"Summarization": "https://python.langchain.com/docs/use_cases/summarization",
"Extraction": "https://python.langchain.com/docs/use_cases/extraction",
"Interacting with APIs": "https://python.langchain.com/docs/use_cases/apis",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"QA over Documents": "https://python.langchain.com/docs/use_cases/question_answering/index",
"Retrieve from vector stores directly": "https://python.langchain.com/docs/use_cases/question_answering/how_to/vector_db_text_generation",
"Improve document indexing with HyDE": "https://python.langchain.com/docs/use_cases/question_answering/how_to/hyde",
@@ -641,7 +641,7 @@
"Chatbots": "https://python.langchain.com/docs/use_cases/chatbots",
"Extraction": "https://python.langchain.com/docs/use_cases/extraction",
"Interacting with APIs": "https://python.langchain.com/docs/use_cases/apis",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"HuggingGPT": "https://python.langchain.com/docs/use_cases/autonomous_agents/hugginggpt",
"Perform context-aware text splitting": "https://python.langchain.com/docs/use_cases/question_answering/how_to/document-context-aware-QA",
"Retrieve from vector stores directly": "https://python.langchain.com/docs/use_cases/question_answering/how_to/vector_db_text_generation",
@@ -1009,7 +1009,7 @@
"LangSmith Walkthrough": "https://python.langchain.com/docs/guides/langsmith/walkthrough",
"Comparing Chain Outputs": "https://python.langchain.com/docs/guides/evaluation/examples/comparisons",
"Agent Trajectory": "https://python.langchain.com/docs/guides/evaluation/trajectory/trajectory_eval",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"Multi-modal outputs: Image & Text": "https://python.langchain.com/docs/use_cases/multi_modal/image_agent",
"Agent Debates with Tools": "https://python.langchain.com/docs/use_cases/agent_simulations/two_agent_debate_tools",
"Multiple callback handlers": "https://python.langchain.com/docs/modules/callbacks/multiple_callbacks",
@@ -1268,7 +1268,7 @@
"SQL Database Agent": "https://python.langchain.com/docs/integrations/toolkits/sql_database",
"JSON Agent": "https://python.langchain.com/docs/integrations/toolkits/json",
"NIBittensorLLM": "https://python.langchain.com/docs/integrations/llms/bittensor",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"BabyAGI with Tools": "https://python.langchain.com/docs/use_cases/agents/baby_agi_with_agent",
"Conversational Retrieval Agent": "https://python.langchain.com/docs/use_cases/question_answering/how_to/conversational_retrieval_agents",
"Plug-and-Plai": "https://python.langchain.com/docs/use_cases/agents/custom_agent_with_plugin_retrieval_using_plugnplai",
@@ -1832,12 +1832,12 @@
"create_sql_agent": {
"CnosDB": "https://python.langchain.com/docs/integrations/providers/cnosdb",
"SQL Database Agent": "https://python.langchain.com/docs/integrations/toolkits/sql_database",
- "SQL": "https://python.langchain.com/docs/use_cases/sql"
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql"
},
"SQLDatabaseToolkit": {
"CnosDB": "https://python.langchain.com/docs/integrations/providers/cnosdb",
"SQL Database Agent": "https://python.langchain.com/docs/integrations/toolkits/sql_database",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"Use ToolKits with OpenAI Functions": "https://python.langchain.com/docs/modules/agents/how_to/use_toolkits_with_openai_functions"
},
"SageMakerCallbackHandler": {
@@ -1899,7 +1899,7 @@
"Rebuff": "https://python.langchain.com/docs/integrations/providers/rebuff",
"SQL Database Agent": "https://python.langchain.com/docs/integrations/toolkits/sql_database",
"Cookbook": "https://python.langchain.com/docs/guides/expression_language/cookbook",
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"Multiple Retrieval Sources": "https://python.langchain.com/docs/use_cases/question_answering/how_to/multiple_retrieval"
},
"Weaviate": {
@@ -3035,11 +3035,11 @@
"Interacting with APIs": "https://python.langchain.com/docs/use_cases/apis"
},
"create_sql_query_chain": {
- "SQL": "https://python.langchain.com/docs/use_cases/sql",
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql",
"Multiple Retrieval Sources": "https://python.langchain.com/docs/use_cases/question_answering/how_to/multiple_retrieval"
},
"ElasticsearchDatabaseChain": {
- "SQL": "https://python.langchain.com/docs/use_cases/sql"
+ "SQL": "https://python.langchain.com/docs/use_cases/qa_structured/sql"
},
"FileChatMessageHistory": {
"AutoGPT": "https://python.langchain.com/docs/use_cases/autonomous_agents/autogpt"
diff --git a/docs/docs_skeleton/docs/modules/model_io/output_parsers/index.mdx b/docs/docs_skeleton/docs/modules/model_io/output_parsers/index.mdx
index bfb4d7241a..a46031ffdd 100644
--- a/docs/docs_skeleton/docs/modules/model_io/output_parsers/index.mdx
+++ b/docs/docs_skeleton/docs/modules/model_io/output_parsers/index.mdx
@@ -12,7 +12,7 @@ Output parsers are classes that help structure language model responses. There a
And then one optional one:
-- "Parse with prompt": A method which takes in a string (assumed to be the response from a language model) and a prompt (assumed to the prompt that generated such a response) and parses it into some structure. The prompt is largely provided in the event the OutputParser wants to retry or fix the output in some way, and needs information from the prompt to do so.
+- "Parse with prompt": A method which takes in a string (assumed to be the response from a language model) and a prompt (assumed to be the prompt that generated such a response) and parses it into some structure. The prompt is largely provided in the event the OutputParser wants to retry or fix the output in some way, and needs information from the prompt to do so.
## Get started
diff --git a/docs/docs_skeleton/docs/use_cases/question_answering/_category_.yml b/docs/docs_skeleton/docs/use_cases/question_answering/_category_.yml
index 7c931cf41b..75252fdc39 100644
--- a/docs/docs_skeleton/docs/use_cases/question_answering/_category_.yml
+++ b/docs/docs_skeleton/docs/use_cases/question_answering/_category_.yml
@@ -1 +1,2 @@
position: 0
+collapsed: false
diff --git a/docs/docs_skeleton/docs/use_cases/web_scraping/index.mdx b/docs/docs_skeleton/docs/use_cases/web_scraping/index.mdx
deleted file mode 100644
index 0adf1130b7..0000000000
--- a/docs/docs_skeleton/docs/use_cases/web_scraping/index.mdx
+++ /dev/null
@@ -1,5 +0,0 @@
-# Web Scraping
-
-Web scraping has historically been a challenging endeavor due to the ever-changing nature of website structures, making it tedious for developers to maintain their scraping scripts. Traditional methods often rely on specific HTML tags and patterns which, when altered, can disrupt data extraction processes.
-
-Enter the LLM-based method for parsing HTML: By leveraging the capabilities of LLMs, and especially OpenAI Functions in LangChain's extraction chain, developers can instruct the model to extract only the desired data in a specified format. This method not only streamlines the extraction process but also significantly reduces the time spent on manual debugging and script modifications. Its adaptability means that even if websites undergo significant design changes, the extraction remains consistent and robust. This level of resilience translates to reduced maintenance efforts, cost savings, and ensures a higher quality of extracted data. Compared to its predecessors, the LLM-based approach wins out in the web scraping domain by transforming a historically cumbersome task into a more automated and efficient process.
diff --git a/docs/docs_skeleton/vercel.json b/docs/docs_skeleton/vercel.json
index dfa378952a..0c7625af99 100644
--- a/docs/docs_skeleton/vercel.json
+++ b/docs/docs_skeleton/vercel.json
@@ -1076,6 +1076,10 @@
"source": "/docs/modules/agents/tools/integrations/zapier",
"destination": "/docs/integrations/tools/zapier"
},
+ {
+ "source": "/docs/integrations/tools/sqlite",
+ "destination": "/docs/use_cases/sql/sqlite"
+ },
{
"source": "/en/latest/modules/callbacks/filecallbackhandler.html",
"destination": "/docs/modules/callbacks/how_to/filecallbackhandler"
@@ -2216,6 +2220,10 @@
"source": "/docs/modules/data_connection/text_embedding/integrations/tensorflowhub",
"destination": "/docs/integrations/text_embedding/tensorflowhub"
},
+ {
+ "source": "/docs/integrations/text_embedding/Awa",
+ "destination": "/docs/integrations/text_embedding/awadb"
+ },
{
"source": "/en/latest/modules/indexes/vectorstores/examples/analyticdb.html",
"destination": "/docs/integrations/vectorstores/analyticdb"
@@ -3178,7 +3186,11 @@
},
{
"source": "/en/latest/use_cases/tabular.html",
- "destination": "/docs/use_cases/tabular"
+ "destination": "/docs/use_cases/qa_structured"
+ },
+ {
+ "source": "/docs/use_cases/sql(/?)",
+ "destination": "/docs/use_cases/qa_structured/sql"
},
{
"source": "/en/latest/youtube.html",
@@ -3370,7 +3382,7 @@
},
{
"source": "/docs/modules/chains/popular/sqlite",
- "destination": "/docs/use_cases/tabular/sqlite"
+ "destination": "/docs/use_cases/qa_structured/sql"
},
{
"source": "/docs/modules/chains/popular/openai_functions",
@@ -3582,7 +3594,7 @@
},
{
"source": "/docs/modules/chains/additional/elasticsearch_database",
- "destination": "/docs/use_cases/tabular/elasticsearch_database"
+ "destination": "/docs/use_cases/qa_structured/integrations/elasticsearch"
},
{
"source": "/docs/modules/chains/additional/tagging",
diff --git a/docs/extras/additional_resources/youtube.mdx b/docs/extras/additional_resources/youtube.mdx
index fc266bf48b..78da30e453 100644
--- a/docs/extras/additional_resources/youtube.mdx
+++ b/docs/extras/additional_resources/youtube.mdx
@@ -1,6 +1,6 @@
# YouTube videos
-⛓ icon marks a new addition [last update 2023-06-20]
+⛓ icon marks a new addition [last update 2023-09-05]
### [Official LangChain YouTube channel](https://www.youtube.com/@LangChain)
@@ -86,20 +86,20 @@
- [`Llama Index`: Chat with Documentation using URL Loader](https://youtu.be/XJRoDEctAwA) by [Merk](https://www.youtube.com/@merksworld)
- [Using OpenAI, LangChain, and `Gradio` to Build Custom GenAI Applications](https://youtu.be/1MsmqMg3yUc) by [David Hundley](https://www.youtube.com/@dkhundley)
- [LangChain, Chroma DB, OpenAI Beginner Guide | ChatGPT with your PDF](https://youtu.be/FuqdVNB_8c0)
-- ⛓ [Build AI chatbot with custom knowledge base using OpenAI API and GPT Index](https://youtu.be/vDZAZuaXf48) by [Irina Nik](https://www.youtube.com/@irina_nik)
-- ⛓ [Build Your Own Auto-GPT Apps with LangChain (Python Tutorial)](https://youtu.be/NYSWn1ipbgg) by [Dave Ebbelaar](https://www.youtube.com/@daveebbelaar)
-- ⛓ [Chat with Multiple `PDFs` | LangChain App Tutorial in Python (Free LLMs and Embeddings)](https://youtu.be/dXxQ0LR-3Hg) by [Alejandro AO - Software & Ai](https://www.youtube.com/@alejandro_ao)
-- ⛓ [Chat with a `CSV` | `LangChain Agents` Tutorial (Beginners)](https://youtu.be/tjeti5vXWOU) by [Alejandro AO - Software & Ai](https://www.youtube.com/@alejandro_ao)
-- ⛓ [Create Your Own ChatGPT with `PDF` Data in 5 Minutes (LangChain Tutorial)](https://youtu.be/au2WVVGUvc8) by [Liam Ottley](https://www.youtube.com/@LiamOttley)
-- ⛓ [Using ChatGPT with YOUR OWN Data. This is magical. (LangChain OpenAI API)](https://youtu.be/9AXP7tCI9PI) by [TechLead](https://www.youtube.com/@TechLead)
-- ⛓ [Build a Custom Chatbot with OpenAI: `GPT-Index` & LangChain | Step-by-Step Tutorial](https://youtu.be/FIDv6nc4CgU) by [Fabrikod](https://www.youtube.com/@fabrikod)
-- ⛓ [`Flowise` is an open source no-code UI visual tool to build 🦜🔗LangChain applications](https://youtu.be/CovAPtQPU0k) by [Cobus Greyling](https://www.youtube.com/@CobusGreylingZA)
-- ⛓ [LangChain & GPT 4 For Data Analysis: The `Pandas` Dataframe Agent](https://youtu.be/rFQ5Kmkd4jc) by [Rabbitmetrics](https://www.youtube.com/@rabbitmetrics)
-- ⛓ [`GirlfriendGPT` - AI girlfriend with LangChain](https://youtu.be/LiN3D1QZGQw) by [Toolfinder AI](https://www.youtube.com/@toolfinderai)
-- ⛓ [`PrivateGPT`: Chat to your FILES OFFLINE and FREE [Installation and Tutorial]](https://youtu.be/G7iLllmx4qc) by [Prompt Engineering](https://www.youtube.com/@engineerprompt)
-- ⛓ [How to build with Langchain 10x easier | ⛓️ LangFlow & `Flowise`](https://youtu.be/Ya1oGL7ZTvU) by [AI Jason](https://www.youtube.com/@AIJasonZ)
-- ⛓ [Getting Started With LangChain In 20 Minutes- Build Celebrity Search Application](https://youtu.be/_FpT1cwcSLg) by [Krish Naik](https://www.youtube.com/@krishnaik06)
-
+- [Build AI chatbot with custom knowledge base using OpenAI API and GPT Index](https://youtu.be/vDZAZuaXf48) by [Irina Nik](https://www.youtube.com/@irina_nik)
+- [Build Your Own Auto-GPT Apps with LangChain (Python Tutorial)](https://youtu.be/NYSWn1ipbgg) by [Dave Ebbelaar](https://www.youtube.com/@daveebbelaar)
+- [Chat with Multiple `PDFs` | LangChain App Tutorial in Python (Free LLMs and Embeddings)](https://youtu.be/dXxQ0LR-3Hg) by [Alejandro AO - Software & Ai](https://www.youtube.com/@alejandro_ao)
+- [Chat with a `CSV` | `LangChain Agents` Tutorial (Beginners)](https://youtu.be/tjeti5vXWOU) by [Alejandro AO - Software & Ai](https://www.youtube.com/@alejandro_ao)
+- [Create Your Own ChatGPT with `PDF` Data in 5 Minutes (LangChain Tutorial)](https://youtu.be/au2WVVGUvc8) by [Liam Ottley](https://www.youtube.com/@LiamOttley)
+- [Using ChatGPT with YOUR OWN Data. This is magical. (LangChain OpenAI API)](https://youtu.be/9AXP7tCI9PI) by [TechLead](https://www.youtube.com/@TechLead)
+- [Build a Custom Chatbot with OpenAI: `GPT-Index` & LangChain | Step-by-Step Tutorial](https://youtu.be/FIDv6nc4CgU) by [Fabrikod](https://www.youtube.com/@fabrikod)
+- [`Flowise` is an open source no-code UI visual tool to build 🦜🔗LangChain applications](https://youtu.be/CovAPtQPU0k) by [Cobus Greyling](https://www.youtube.com/@CobusGreylingZA)
+- [LangChain & GPT 4 For Data Analysis: The `Pandas` Dataframe Agent](https://youtu.be/rFQ5Kmkd4jc) by [Rabbitmetrics](https://www.youtube.com/@rabbitmetrics)
+- [`GirlfriendGPT` - AI girlfriend with LangChain](https://youtu.be/LiN3D1QZGQw) by [Toolfinder AI](https://www.youtube.com/@toolfinderai)
+- [`PrivateGPT`: Chat to your FILES OFFLINE and FREE [Installation and Tutorial]](https://youtu.be/G7iLllmx4qc) by [Prompt Engineering](https://www.youtube.com/@engineerprompt)
+- [How to build with Langchain 10x easier | ⛓️ LangFlow & `Flowise`](https://youtu.be/Ya1oGL7ZTvU) by [AI Jason](https://www.youtube.com/@AIJasonZ)
+- [Getting Started With LangChain In 20 Minutes- Build Celebrity Search Application](https://youtu.be/_FpT1cwcSLg) by [Krish Naik](https://www.youtube.com/@krishnaik06)
+- ⛓ [LangChain HowTo and Guides YouTube playlist](https://www.youtube.com/playlist?list=PL8motc6AQftk1Bs42EW45kwYbyJ4jOdiZ) by [Sam Witteveen](https://www.youtube.com/@samwitteveenai/)
### [Prompt Engineering and LangChain](https://www.youtube.com/watch?v=muXbPpG_ys4&list=PLEJK-H61Xlwzm5FYLDdKt_6yibO33zoMW) by [Venelin Valkov](https://www.youtube.com/@venelin_valkov)
diff --git a/docs/extras/expression_language/cookbook.ipynb b/docs/extras/expression_language/cookbook.ipynb
deleted file mode 100644
index 04b74164dd..0000000000
--- a/docs/extras/expression_language/cookbook.ipynb
+++ /dev/null
@@ -1,1664 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "id": "9a9acd2e",
- "metadata": {},
- "source": [
- "# Cookbook\n",
- "\n",
- "In this notebook we'll take a look at a few common types of sequences to create."
- ]
- },
- {
- "cell_type": "markdown",
- "id": "93aa2c87",
- "metadata": {},
- "source": [
- "## PromptTemplate + LLM\n",
- "\n",
- "A PromptTemplate -> LLM is a core chain that is used in most other larger chains/systems."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 1,
- "id": "466b65b3",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.prompts import ChatPromptTemplate\n",
- "from langchain.chat_models import ChatOpenAI"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 11,
- "id": "3c634ef0",
- "metadata": {},
- "outputs": [],
- "source": [
- "model = ChatOpenAI()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "id": "d1850a1f",
- "metadata": {},
- "outputs": [],
- "source": [
- "prompt = ChatPromptTemplate.from_template(\"tell me a joke about {foo}\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "id": "56d0669f",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = prompt | model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "id": "e3d0a6cd",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='Why don\\'t bears use cell phones? \\n\\nBecause they always get terrible \"grizzly\" reception!', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 5,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bears\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "7eb9ef50",
- "metadata": {},
- "source": [
- "Often times we want to attach kwargs to the model that's passed in. Here's a few examples of that:"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "0b1d8f88",
- "metadata": {},
- "source": [
- "### Attaching Stop Sequences"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "id": "562a06bf",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = prompt | model.bind(stop=[\"\\n\"])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "id": "43f5d04c",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content=\"Why don't bears use cell phones?\", additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 7,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bears\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "f3eaf88a",
- "metadata": {},
- "source": [
- "### Attaching Function Call information"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "id": "f94b71b2",
- "metadata": {},
- "outputs": [],
- "source": [
- "functions = [\n",
- " {\n",
- " \"name\": \"joke\",\n",
- " \"description\": \"A joke\",\n",
- " \"parameters\": {\n",
- " \"type\": \"object\",\n",
- " \"properties\": {\n",
- " \"setup\": {\n",
- " \"type\": \"string\",\n",
- " \"description\": \"The setup for the joke\"\n",
- " },\n",
- " \"punchline\": {\n",
- " \"type\": \"string\",\n",
- " \"description\": \"The punchline for the joke\"\n",
- " }\n",
- " },\n",
- " \"required\": [\"setup\", \"punchline\"]\n",
- " }\n",
- " }\n",
- " ]\n",
- "chain = prompt | model.bind(function_call= {\"name\": \"joke\"}, functions= functions)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 9,
- "id": "decf7710",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='', additional_kwargs={'function_call': {'name': 'joke', 'arguments': '{\\n \"setup\": \"Why don\\'t bears wear shoes?\",\\n \"punchline\": \"Because they have bear feet!\"\\n}'}}, example=False)"
- ]
- },
- "execution_count": 9,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bears\"}, config={})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "9098c5ed",
- "metadata": {},
- "source": [
- "## PromptTemplate + LLM + OutputParser\n",
- "\n",
- "We can also add in an output parser to easily trasform the raw LLM/ChatModel output into a more workable format"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 10,
- "id": "f799664d",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.schema.output_parser import StrOutputParser"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 12,
- "id": "cc194c78",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = prompt | model | StrOutputParser()"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "77acf448",
- "metadata": {},
- "source": [
- "Notice that this now returns a string - a much more workable format for downstream tasks"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 13,
- "id": "e3d69a18",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "\"Why don't bears wear shoes?\\n\\nBecause they have bear feet!\""
- ]
- },
- "execution_count": 13,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bears\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "c01864e5",
- "metadata": {},
- "source": [
- "### Functions Output Parser\n",
- "\n",
- "When you specify the function to return, you may just want to parse that directly"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 14,
- "id": "ad0dd88e",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.output_parsers.openai_functions import JsonOutputFunctionsParser\n",
- "chain = (\n",
- " prompt \n",
- " | model.bind(function_call= {\"name\": \"joke\"}, functions= functions) \n",
- " | JsonOutputFunctionsParser()\n",
- ")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 15,
- "id": "1e7aa8eb",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "{'setup': \"Why don't bears wear shoes?\",\n",
- " 'punchline': 'Because they have bear feet!'}"
- ]
- },
- "execution_count": 15,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bears\"})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
- "id": "d4aa1a01",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.output_parsers.openai_functions import JsonKeyOutputFunctionsParser\n",
- "chain = (\n",
- " prompt \n",
- " | model.bind(function_call= {\"name\": \"joke\"}, functions= functions) \n",
- " | JsonKeyOutputFunctionsParser(key_name=\"setup\")\n",
- ")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 18,
- "id": "8b6df9ba",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "\"Why don't bears like fast food?\""
- ]
- },
- "execution_count": 18,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bears\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "2ed58136",
- "metadata": {},
- "source": [
- "## Passthroughs and itemgetter\n",
- "\n",
- "Often times when constructing a chain you may want to pass along original input variables to future steps in the chain. How exactly you do this depends on what exactly the input is:\n",
- "\n",
- "- If the original input was a string, then you likely just want to pass along the string. This can be done with `RunnablePassthrough`. For an example of this, see `LLMChain + Retriever`\n",
- "- If the original input was a dictionary, then you likely want to pass along specific keys. This can be done with `itemgetter`. For an example of this see `Multiple LLM Chains`"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 12,
- "id": "5d3d8ffe",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.schema.runnable import RunnablePassthrough\n",
- "from operator import itemgetter"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "91c5ef3d",
- "metadata": {},
- "source": [
- "## LLMChain + Retriever\n",
- "\n",
- "Let's now look at adding in a retrieval step, which adds up to a \"retrieval-augmented generation\" chain"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "id": "33be32af",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.vectorstores import Chroma\n",
- "from langchain.embeddings import OpenAIEmbeddings\n",
- "from langchain.schema.runnable import RunnablePassthrough"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "id": "df3f3fa2",
- "metadata": {},
- "outputs": [],
- "source": [
- "# Create the retriever\n",
- "vectorstore = Chroma.from_texts([\"harrison worked at kensho\"], embedding=OpenAIEmbeddings())\n",
- "retriever = vectorstore.as_retriever()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 16,
- "id": "bfc47ec1",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"Answer the question based only on the following context:\n",
- "{context}\n",
- "\n",
- "Question: {question}\n",
- "\"\"\"\n",
- "prompt = ChatPromptTemplate.from_template(template)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
- "id": "eae31755",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = (\n",
- " {\"context\": retriever, \"question\": RunnablePassthrough()} \n",
- " | prompt \n",
- " | model \n",
- " | StrOutputParser()\n",
- ")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 18,
- "id": "f3040b0c",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Number of requested results 4 is greater than number of elements in index 1, updating n_results = 1\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "'Harrison worked at Kensho.'"
- ]
- },
- "execution_count": 18,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke(\"where did harrison work?\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 19,
- "id": "e1d20c7c",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"Answer the question based only on the following context:\n",
- "{context}\n",
- "\n",
- "Question: {question}\n",
- "\n",
- "Answer in the following language: {language}\n",
- "\"\"\"\n",
- "prompt = ChatPromptTemplate.from_template(template)\n",
- "\n",
- "chain = {\n",
- " \"context\": itemgetter(\"question\") | retriever, \n",
- " \"question\": itemgetter(\"question\"), \n",
- " \"language\": itemgetter(\"language\")\n",
- "} | prompt | model | StrOutputParser()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 20,
- "id": "7ee8b2d4",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Number of requested results 4 is greater than number of elements in index 1, updating n_results = 1\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "'Harrison ha lavorato a Kensho.'"
- ]
- },
- "execution_count": 20,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"question\": \"where did harrison work\", \"language\": \"italian\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "f007669c",
- "metadata": {},
- "source": [
- "## Conversational Retrieval Chain\n",
- "\n",
- "We can easily add in conversation history. This primarily means adding in chat_message_history"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "id": "3f30c348",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.schema.runnable import RunnableMap\n",
- "from langchain.schema import format_document"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "id": "64ab1dbf",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.prompts.prompt import PromptTemplate\n",
- "\n",
- "_template = \"\"\"Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question, in its original language.\n",
- "\n",
- "Chat History:\n",
- "{chat_history}\n",
- "Follow Up Input: {question}\n",
- "Standalone question:\"\"\"\n",
- "CONDENSE_QUESTION_PROMPT = PromptTemplate.from_template(_template)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "id": "7d628c97",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"Answer the question based only on the following context:\n",
- "{context}\n",
- "\n",
- "Question: {question}\n",
- "\"\"\"\n",
- "ANSWER_PROMPT = ChatPromptTemplate.from_template(template)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "id": "f60a5d0f",
- "metadata": {},
- "outputs": [],
- "source": [
- "DEFAULT_DOCUMENT_PROMPT = PromptTemplate.from_template(template=\"{page_content}\")\n",
- "def _combine_documents(docs, document_prompt = DEFAULT_DOCUMENT_PROMPT, document_separator=\"\\n\\n\"):\n",
- " doc_strings = [format_document(doc, document_prompt) for doc in docs]\n",
- " return document_separator.join(doc_strings)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "id": "7d007db6",
- "metadata": {},
- "outputs": [],
- "source": [
- "from typing import Tuple, List\n",
- "def _format_chat_history(chat_history: List[Tuple]) -> str:\n",
- " buffer = \"\"\n",
- " for dialogue_turn in chat_history:\n",
- " human = \"Human: \" + dialogue_turn[0]\n",
- " ai = \"Assistant: \" + dialogue_turn[1]\n",
- " buffer += \"\\n\" + \"\\n\".join([human, ai])\n",
- " return buffer"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 16,
- "id": "5c32cc89",
- "metadata": {},
- "outputs": [],
- "source": [
- "_inputs = RunnableMap(\n",
- " {\n",
- " \"standalone_question\": {\n",
- " \"question\": lambda x: x[\"question\"],\n",
- " \"chat_history\": lambda x: _format_chat_history(x['chat_history'])\n",
- " } | CONDENSE_QUESTION_PROMPT | ChatOpenAI(temperature=0) | StrOutputParser(),\n",
- " }\n",
- ")\n",
- "_context = {\n",
- " \"context\": itemgetter(\"standalone_question\") | retriever | _combine_documents,\n",
- " \"question\": lambda x: x[\"standalone_question\"]\n",
- "}\n",
- "conversational_qa_chain = _inputs | _context | ANSWER_PROMPT | ChatOpenAI()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
- "id": "135c8205",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Number of requested results 4 is greater than number of elements in index 1, updating n_results = 1\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "AIMessage(content='Harrison was employed at Kensho.', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 17,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "conversational_qa_chain.invoke({\n",
- " \"question\": \"where did harrison work?\",\n",
- " \"chat_history\": [],\n",
- "})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 15,
- "id": "424e7e7a",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Number of requested results 4 is greater than number of elements in index 1, updating n_results = 1\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "AIMessage(content='Harrison worked at Kensho.', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 15,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "conversational_qa_chain.invoke({\n",
- " \"question\": \"where did he work?\",\n",
- " \"chat_history\": [(\"Who wrote this notebook?\", \"Harrison\")],\n",
- "})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "c5543183",
- "metadata": {},
- "source": [
- "### With Memory and returning source documents\n",
- "\n",
- "This shows how to use memory with the above. For memory, we need to manage that outside at the memory. For returning the retrieved documents, we just need to pass them through all the way."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 18,
- "id": "e31dd17c",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.memory import ConversationBufferMemory"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 44,
- "id": "d4bffe94",
- "metadata": {},
- "outputs": [],
- "source": [
- "memory = ConversationBufferMemory(return_messages=True, output_key=\"answer\", input_key=\"question\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 45,
- "id": "733be985",
- "metadata": {},
- "outputs": [],
- "source": [
- "# First we add a step to load memory\n",
- "# This needs to be a RunnableMap because its the first input\n",
- "loaded_memory = RunnableMap(\n",
- " {\n",
- " \"question\": itemgetter(\"question\"),\n",
- " \"memory\": memory.load_memory_variables,\n",
- " }\n",
- ")\n",
- "# Next we add a step to expand memory into the variables\n",
- "expanded_memory = {\n",
- " \"question\": itemgetter(\"question\"),\n",
- " \"chat_history\": lambda x: x[\"memory\"][\"history\"]\n",
- "}\n",
- "\n",
- "# Now we calculate the standalone question\n",
- "standalone_question = {\n",
- " \"standalone_question\": {\n",
- " \"question\": lambda x: x[\"question\"],\n",
- " \"chat_history\": lambda x: _format_chat_history(x['chat_history'])\n",
- " } | CONDENSE_QUESTION_PROMPT | ChatOpenAI(temperature=0) | StrOutputParser(),\n",
- "}\n",
- "# Now we retrieve the documents\n",
- "retrieved_documents = {\n",
- " \"docs\": itemgetter(\"standalone_question\") | retriever,\n",
- " \"question\": lambda x: x[\"standalone_question\"]\n",
- "}\n",
- "# Now we construct the inputs for the final prompt\n",
- "final_inputs = {\n",
- " \"context\": lambda x: _combine_documents(x[\"docs\"]),\n",
- " \"question\": itemgetter(\"question\")\n",
- "}\n",
- "# And finally, we do the part that returns the answers\n",
- "answer = {\n",
- " \"answer\": final_inputs | ANSWER_PROMPT | ChatOpenAI(),\n",
- " \"docs\": itemgetter(\"docs\"),\n",
- "}\n",
- "# And now we put it all together!\n",
- "final_chain = loaded_memory | expanded_memory | standalone_question | retrieved_documents | answer"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 46,
- "id": "806e390c",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Number of requested results 4 is greater than number of elements in index 1, updating n_results = 1\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "{'answer': AIMessage(content='Harrison was employed at Kensho.', additional_kwargs={}, example=False),\n",
- " 'docs': [Document(page_content='harrison worked at kensho', metadata={})]}"
- ]
- },
- "execution_count": 46,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "inputs = {\"question\": \"where did harrison work?\"}\n",
- "result = final_chain.invoke(inputs)\n",
- "result"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 47,
- "id": "977399fd",
- "metadata": {},
- "outputs": [],
- "source": [
- "# Note that the memory does not save automatically\n",
- "# This will be improved in the future\n",
- "# For now you need to save it yourself\n",
- "memory.save_context(inputs, {\"answer\": result[\"answer\"].content})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 48,
- "id": "f94f7de4",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "{'history': [HumanMessage(content='where did harrison work?', additional_kwargs={}, example=False),\n",
- " AIMessage(content='Harrison was employed at Kensho.', additional_kwargs={}, example=False)]}"
- ]
- },
- "execution_count": 48,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "memory.load_memory_variables({})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "0f2bf8d3",
- "metadata": {},
- "source": [
- "## Multiple LLM Chains\n",
- "\n",
- "This can also be used to string together multiple LLMChains"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 31,
- "id": "d65d4e9e",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "'El país en el que nació la ciudad de Honolulu, Hawái, donde nació Barack Obama, el 44º presidente de los Estados Unidos, es Estados Unidos.'"
- ]
- },
- "execution_count": 31,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "from operator import itemgetter\n",
- "\n",
- "prompt1 = ChatPromptTemplate.from_template(\"what is the city {person} is from?\")\n",
- "prompt2 = ChatPromptTemplate.from_template(\"what country is the city {city} in? respond in {language}\")\n",
- "\n",
- "chain1 = prompt1 | model | StrOutputParser()\n",
- "\n",
- "chain2 = {\"city\": chain1, \"language\": itemgetter(\"language\")} | prompt2 | model | StrOutputParser()\n",
- "\n",
- "chain2.invoke({\"person\": \"obama\", \"language\": \"spanish\"})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 32,
- "id": "878f8176",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.schema.runnable import RunnableMap\n",
- "prompt1 = ChatPromptTemplate.from_template(\"generate a random color\")\n",
- "prompt2 = ChatPromptTemplate.from_template(\"what is a fruit of color: {color}\")\n",
- "prompt3 = ChatPromptTemplate.from_template(\"what is countries flag that has the color: {color}\")\n",
- "prompt4 = ChatPromptTemplate.from_template(\"What is the color of {fruit} and {country}\")\n",
- "chain1 = prompt1 | model | StrOutputParser()\n",
- "chain2 = RunnableMap(steps={\"color\": chain1}) | {\n",
- " \"fruit\": prompt2 | model | StrOutputParser(),\n",
- " \"country\": prompt3 | model | StrOutputParser(),\n",
- "} | prompt4"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 33,
- "id": "d621a870",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "ChatPromptValue(messages=[HumanMessage(content=\"What is the color of A fruit that has a color similar to #7E7DE6 is the Peruvian Apple Cactus (Cereus repandus). It is a tropical fruit with a vibrant purple or violet exterior. and The country's flag that has the color #7E7DE6 is North Macedonia.\", additional_kwargs={}, example=False)])"
- ]
- },
- "execution_count": 33,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain2.invoke({})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "d094d637",
- "metadata": {},
- "source": [
- "## Router\n",
- "\n",
- "You can also use the router runnable to conditionally route inputs to different runnables."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "id": "252625fd",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.chains import create_tagging_chain_pydantic\n",
- "from pydantic import BaseModel, Field\n",
- "\n",
- "class PromptToUse(BaseModel):\n",
- " \"\"\"Used to determine which prompt to use to answer the user's input.\"\"\"\n",
- " \n",
- " name: str = Field(description=\"Should be one of `math` or `english`\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "id": "57886e84",
- "metadata": {},
- "outputs": [],
- "source": [
- "tagger = create_tagging_chain_pydantic(PromptToUse, ChatOpenAI(temperature=0))"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "id": "a303b089",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain1 = ChatPromptTemplate.from_template(\"You are a math genius. Answer the question: {question}\") | ChatOpenAI()\n",
- "chain2 = ChatPromptTemplate.from_template(\"You are an english major. Answer the question: {question}\") | ChatOpenAI()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "id": "7aa9ea06",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.schema.runnable import RouterRunnable\n",
- "router = RouterRunnable({\"math\": chain1, \"english\": chain2})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "id": "6a3d3f5d",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = {\n",
- " \"key\": {\"input\": lambda x: x[\"question\"]} | tagger | (lambda x: x['text'].name),\n",
- " \"input\": {\"question\": lambda x: x[\"question\"]}\n",
- "} | router"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 9,
- "id": "8aeda930",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='Thank you for the compliment! The sum of 2 + 2 is equal to 4.', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 9,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"question\": \"whats 2 + 2\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "29781123",
- "metadata": {},
- "source": [
- "## Tools\n",
- "\n",
- "You can use any LangChain tool easily"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 1,
- "id": "9232d2a9",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "/Users/harrisonchase/.pyenv/versions/3.9.1/envs/langchain/lib/python3.9/site-packages/deeplake/util/check_latest_version.py:32: UserWarning: A newer version of deeplake (3.6.14) is available. It's recommended that you update to the latest version using `pip install -U deeplake`.\n",
- " warnings.warn(\n"
- ]
- }
- ],
- "source": [
- "from langchain.tools import DuckDuckGoSearchRun"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "id": "a0c64d2c",
- "metadata": {},
- "outputs": [],
- "source": [
- "search = DuckDuckGoSearchRun()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "id": "391969b6",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"turn the following user input into a search query for a search engine:\n",
- "\n",
- "{input}\"\"\"\n",
- "prompt = ChatPromptTemplate.from_template(template)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 9,
- "id": "e3d9d20d",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = prompt | model | StrOutputParser() | search"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 10,
- "id": "55f2967d",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "\"What sports games are on TV today & tonight? Watch and stream live sports on TV today, tonight, tomorrow. Today's 2023 sports TV schedule includes football, basketball, baseball, hockey, motorsports, soccer and more. Watch on TV or stream online on ESPN, FOX, FS1, CBS, NBC, ABC, Peacock, Paramount+, fuboTV, local channels and many other networks. Weather Alerts Alerts Bar. Not all offers available in all states, please visit BetMGM for the latest promotions for your area. Must be 21+ to gamble, please wager responsibly. If you or someone ... Speak of the Devils. Good Morning Arizona. Happy Hour Spots. Jaime's Local Love. Surprise Squad. Silver Apple. Field Trip Friday. Seen on TV. Arizona Highways TV. MLB Games Tonight: How to Watch on TV, Streaming & Odds - Friday, July 28. San Diego Padres' Juan Soto plays during the first baseball game in a doubleheader, Saturday, July 15, 2023, in Philadelphia. (AP Photo/Matt Slocum) (APMedia) Today's MLB schedule features top teams in action. Among those games is the Texas Rangers playing the San Diego ... TV. Cleveland at Chi. White Sox. 1:10pm. Bally Sports. NBCS-CHI. Cleveland Guardians (50-51) are second place in AL Central and Chicago White Sox (41-61) are fourth place in AL Central. The Guardians are 23-27 on the road this season and White Sox are 21-26 at home. Chi. Cubs at St. Louis.\""
- ]
- },
- "execution_count": 10,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"input\": \"I'd like to figure out what games are tonight\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "fbc4bf6e",
- "metadata": {},
- "source": [
- "## Arbitrary Functions\n",
- "\n",
- "You can use arbitrary functions in the pipeline\n",
- "\n",
- "Note that all inputs to these functions need to be a SINGLE argument. If you have a function that accepts multiple arguments, you should write a wrapper that accepts a single input and unpacks it into multiple argument."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 35,
- "id": "6bb221b3",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.schema.runnable import RunnableLambda\n",
- "\n",
- "def length_function(text):\n",
- " return len(text)\n",
- "\n",
- "def _multiple_length_function(text1, text2):\n",
- " return len(text1) * len(text2)\n",
- "\n",
- "def multiple_length_function(_dict):\n",
- " return _multiple_length_function(_dict[\"text1\"], _dict[\"text2\"])\n",
- "\n",
- "prompt = ChatPromptTemplate.from_template(\"what is {a} + {b}\")\n",
- "\n",
- "chain1 = prompt | model\n",
- "\n",
- "chain = {\n",
- " \"a\": itemgetter(\"foo\") | RunnableLambda(length_function),\n",
- " \"b\": {\"text1\": itemgetter(\"foo\"), \"text2\": itemgetter(\"bar\")} | RunnableLambda(multiple_length_function)\n",
- "} | prompt | model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 36,
- "id": "5488ec85",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='3 + 9 is equal to 12.', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 36,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"foo\": \"bar\", \"bar\": \"gah\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "506e9636",
- "metadata": {},
- "source": [
- "## SQL Database\n",
- "\n",
- "We can also try to replicate our SQLDatabaseChain using this style."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 37,
- "id": "7a927516",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"Based on the table schema below, write a SQL query that would answer the user's question:\n",
- "{schema}\n",
- "\n",
- "Question: {question}\"\"\"\n",
- "prompt = ChatPromptTemplate.from_template(template)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 38,
- "id": "3f51f386",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.utilities import SQLDatabase"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 41,
- "id": "2ccca6fc",
- "metadata": {},
- "outputs": [],
- "source": [
- "db = SQLDatabase.from_uri(\"sqlite:///../../../../notebooks/Chinook.db\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 42,
- "id": "05ba88ee",
- "metadata": {},
- "outputs": [],
- "source": [
- "def get_schema(_):\n",
- " return db.get_table_info()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 43,
- "id": "a4eda902",
- "metadata": {},
- "outputs": [],
- "source": [
- "def run_query(query):\n",
- " return db.run(query)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 47,
- "id": "5046cb17",
- "metadata": {},
- "outputs": [],
- "source": [
- "inputs = {\n",
- " \"schema\": RunnableLambda(get_schema),\n",
- " \"question\": itemgetter(\"question\")\n",
- "}\n",
- "sql_response = (\n",
- " RunnableMap(inputs)\n",
- " | prompt\n",
- " | model.bind(stop=[\"\\nSQLResult:\"])\n",
- " | StrOutputParser()\n",
- " )"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 48,
- "id": "a5552039",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "'SELECT COUNT(*) \\nFROM Employee;'"
- ]
- },
- "execution_count": 48,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "sql_response.invoke({\"question\": \"How many employees are there?\"})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 49,
- "id": "d6fee130",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"Based on the table schema below, question, sql query, and sql response, write a natural language response:\n",
- "{schema}\n",
- "\n",
- "Question: {question}\n",
- "SQL Query: {query}\n",
- "SQL Response: {response}\"\"\"\n",
- "prompt_response = ChatPromptTemplate.from_template(template)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 52,
- "id": "923aa634",
- "metadata": {},
- "outputs": [],
- "source": [
- "full_chain = (\n",
- " RunnableMap({\n",
- " \"question\": itemgetter(\"question\"),\n",
- " \"query\": sql_response,\n",
- " }) \n",
- " | {\n",
- " \"schema\": RunnableLambda(get_schema),\n",
- " \"question\": itemgetter(\"question\"),\n",
- " \"query\": itemgetter(\"query\"),\n",
- " \"response\": lambda x: db.run(x[\"query\"]) \n",
- " } \n",
- " | prompt_response \n",
- " | model\n",
- ")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 53,
- "id": "e94963d8",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='There are 8 employees.', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 53,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "full_chain.invoke({\"question\": \"How many employees are there?\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "f09fd305",
- "metadata": {},
- "source": [
- "## Code Writing"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 57,
- "id": "bd7c259a",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.utilities import PythonREPL\n",
- "from langchain.prompts import SystemMessagePromptTemplate, HumanMessagePromptTemplate"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 58,
- "id": "73795d2d",
- "metadata": {},
- "outputs": [],
- "source": [
- "template = \"\"\"Write some python code to solve the user's problem. \n",
- "\n",
- "Return only python code in Markdown format, e.g.:\n",
- "\n",
- "```python\n",
- "....\n",
- "```\"\"\"\n",
- "prompt = ChatPromptTemplate(messages=[\n",
- " SystemMessagePromptTemplate.from_template(template),\n",
- " HumanMessagePromptTemplate.from_template(\"{input}\")\n",
- "])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 64,
- "id": "42859e8a",
- "metadata": {},
- "outputs": [],
- "source": [
- "def _sanitize_output(text: str):\n",
- " _, after = text.split(\"```python\")\n",
- " return after.split(\"```\")[0]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 67,
- "id": "5ded1a86",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = prompt | model | StrOutputParser() | _sanitize_output | PythonREPL().run"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 68,
- "id": "208c2b75",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Python REPL can execute arbitrary code. Use with caution.\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "'4\\n'"
- ]
- },
- "execution_count": 68,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"input\": \"whats 2 plus 2\"})"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "5062941a",
- "metadata": {},
- "source": [
- "## Memory\n",
- "\n",
- "This shows how to add memory to an arbitrary chain. Right now, you can use the memory classes but need to hook it up manually"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 99,
- "id": "7998efd8",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.memory import ConversationBufferMemory\n",
- "from langchain.schema.runnable import RunnableMap\n",
- "from langchain.prompts import MessagesPlaceholder\n",
- "model = ChatOpenAI()\n",
- "prompt = ChatPromptTemplate.from_messages([\n",
- " (\"system\", \"You are a helpful chatbot\"),\n",
- " MessagesPlaceholder(variable_name=\"history\"),\n",
- " (\"human\", \"{input}\")\n",
- "])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 100,
- "id": "fa0087f3",
- "metadata": {},
- "outputs": [],
- "source": [
- "memory = ConversationBufferMemory(return_messages=True)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 101,
- "id": "06b531ae",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "{'history': []}"
- ]
- },
- "execution_count": 101,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "memory.load_memory_variables({})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 102,
- "id": "d9437af6",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = RunnableMap({\n",
- " \"input\": lambda x: x[\"input\"],\n",
- " \"memory\": memory.load_memory_variables\n",
- "}) | {\n",
- " \"input\": lambda x: x[\"input\"],\n",
- " \"history\": lambda x: x[\"memory\"][\"history\"]\n",
- "} | prompt | model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 103,
- "id": "bed1e260",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='Hello Bob! How can I assist you today?', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 103,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "inputs = {\"input\": \"hi im bob\"}\n",
- "response = chain.invoke(inputs)\n",
- "response"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 104,
- "id": "890475b4",
- "metadata": {},
- "outputs": [],
- "source": [
- "memory.save_context(inputs, {\"output\": response.content})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 105,
- "id": "e8fcb77f",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "{'history': [HumanMessage(content='hi im bob', additional_kwargs={}, example=False),\n",
- " AIMessage(content='Hello Bob! How can I assist you today?', additional_kwargs={}, example=False)]}"
- ]
- },
- "execution_count": 105,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "memory.load_memory_variables({})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 106,
- "id": "d837d5c3",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "AIMessage(content='Your name is Bob. You mentioned it in your previous message. Is there anything else I can help you with, Bob?', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 106,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "inputs = {\"input\": \"whats my name\"}\n",
- "response = chain.invoke(inputs)\n",
- "response"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "4927a727-b4c8-453c-8c83-bd87b4fcac14",
- "metadata": {},
- "source": [
- "## Moderation\n",
- "\n",
- "This shows how to add in moderation (or other safeguards) around your LLM application."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 26,
- "id": "4f5f6449-940a-4f5c-97c0-39b71c3e2a68",
- "metadata": {},
- "outputs": [],
- "source": [
- "from langchain.chains import OpenAIModerationChain\n",
- "from langchain.llms import OpenAI"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 35,
- "id": "fcb8312b-7e7a-424f-a3ec-76738c9a9d21",
- "metadata": {},
- "outputs": [],
- "source": [
- "moderate = OpenAIModerationChain()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 32,
- "id": "b24b9148-f6b0-4091-8ea8-d3fb281bd950",
- "metadata": {},
- "outputs": [],
- "source": [
- "model = OpenAI()\n",
- "prompt = ChatPromptTemplate.from_messages([\n",
- " (\"system\", \"repeat after me: {input}\")\n",
- "])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 33,
- "id": "1c8ed87c-9ca6-4559-bf60-d40e94a0af08",
- "metadata": {},
- "outputs": [],
- "source": [
- "chain = prompt | model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 34,
- "id": "5256b9bd-381a-42b0-bfa8-7e6d18f853cb",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "'\\n\\nYou are stupid.'"
- ]
- },
- "execution_count": 34,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chain.invoke({\"input\": \"you are stupid\"})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 36,
- "id": "fe6e3b33-dc9a-49d5-b194-ba750c58a628",
- "metadata": {},
- "outputs": [],
- "source": [
- "moderated_chain = chain | moderate"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 37,
- "id": "d8ba0cbd-c739-4d23-be9f-6ae092bd5ffb",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "{'input': '\\n\\nYou are stupid.',\n",
- " 'output': \"Text was found that violates OpenAI's content policy.\"}"
- ]
- },
- "execution_count": 37,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "moderated_chain.invoke({\"input\": \"you are stupid\"})"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Python 3 (ipykernel)",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.10.1"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/docs/extras/expression_language/cookbook/code_writing.ipynb b/docs/extras/expression_language/cookbook/code_writing.ipynb
new file mode 100644
index 0000000000..25b039ce44
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/code_writing.ipynb
@@ -0,0 +1,119 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "f09fd305",
+ "metadata": {},
+ "source": [
+ "# Code writing\n",
+ "\n",
+ "Example of how to use LCEL to write Python code."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "bd7c259a",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.prompts import ChatPromptTemplate, SystemMessagePromptTemplate, HumanMessagePromptTemplate\n",
+ "from langchain.schema.output_parser import StrOutputParser\n",
+ "from langchain.utilities import PythonREPL"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "id": "73795d2d",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "template = \"\"\"Write some python code to solve the user's problem. \n",
+ "\n",
+ "Return only python code in Markdown format, e.g.:\n",
+ "\n",
+ "```python\n",
+ "....\n",
+ "```\"\"\"\n",
+ "prompt = ChatPromptTemplate.from_messages(\n",
+ " [(\"system\", template), (\"human\", \"{input}\")]\n",
+ ")\n",
+ "\n",
+ "model = ChatOpenAI()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "id": "42859e8a",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def _sanitize_output(text: str):\n",
+ " _, after = text.split(\"```python\")\n",
+ " return after.split(\"```\")[0]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "id": "5ded1a86",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = prompt | model | StrOutputParser() | _sanitize_output | PythonREPL().run"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "id": "208c2b75",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Python REPL can execute arbitrary code. Use with caution.\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'4\\n'"
+ ]
+ },
+ "execution_count": 15,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"input\": \"whats 2 plus 2\"})"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/index.mdx b/docs/extras/expression_language/cookbook/index.mdx
new file mode 100644
index 0000000000..6310fd50b9
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/index.mdx
@@ -0,0 +1,11 @@
+---
+sidebar_position: 2
+---
+
+# Cookbook
+
+import DocCardList from "@theme/DocCardList";
+
+Example code for accomplishing common tasks with the LangChain Expression Language (LCEL). These examples show how to compose different Runnable (the core LCEL interface) components to achieve various tasks. If you're just getting acquainted with LCEL, the [Prompt + LLM](/docs/expression_language/cookbook/prompt_llm_parser) page is a good place to start.
+
+
\ No newline at end of file
diff --git a/docs/extras/expression_language/cookbook/memory.ipynb b/docs/extras/expression_language/cookbook/memory.ipynb
new file mode 100644
index 0000000000..bef7e5ed01
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/memory.ipynb
@@ -0,0 +1,180 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "5062941a",
+ "metadata": {},
+ "source": [
+ "# Adding memory\n",
+ "\n",
+ "This shows how to add memory to an arbitrary chain. Right now, you can use the memory classes but need to hook it up manually"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "id": "7998efd8",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.memory import ConversationBufferMemory\n",
+ "from langchain.schema.runnable import RunnableMap\n",
+ "from langchain.prompts import ChatPromptTemplate, MessagesPlaceholder\n",
+ "\n",
+ "model = ChatOpenAI()\n",
+ "prompt = ChatPromptTemplate.from_messages([\n",
+ " (\"system\", \"You are a helpful chatbot\"),\n",
+ " MessagesPlaceholder(variable_name=\"history\"),\n",
+ " (\"human\", \"{input}\")\n",
+ "])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "fa0087f3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "memory = ConversationBufferMemory(return_messages=True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "06b531ae",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'history': []}"
+ ]
+ },
+ "execution_count": 3,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "memory.load_memory_variables({})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "d9437af6",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = RunnableMap({\n",
+ " \"input\": lambda x: x[\"input\"],\n",
+ " \"memory\": memory.load_memory_variables\n",
+ "}) | {\n",
+ " \"input\": lambda x: x[\"input\"],\n",
+ " \"history\": lambda x: x[\"memory\"][\"history\"]\n",
+ "} | prompt | model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "bed1e260",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='Hello Bob! How can I assist you today?', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "inputs = {\"input\": \"hi im bob\"}\n",
+ "response = chain.invoke(inputs)\n",
+ "response"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "890475b4",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "memory.save_context(inputs, {\"output\": response.content})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "e8fcb77f",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'history': [HumanMessage(content='hi im bob', additional_kwargs={}, example=False),\n",
+ " AIMessage(content='Hello Bob! How can I assist you today?', additional_kwargs={}, example=False)]}"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "memory.load_memory_variables({})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "d837d5c3",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='Your name is Bob.', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "inputs = {\"input\": \"whats my name\"}\n",
+ "response = chain.invoke(inputs)\n",
+ "response"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/moderation.ipynb b/docs/extras/expression_language/cookbook/moderation.ipynb
new file mode 100644
index 0000000000..cb4114d8e9
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/moderation.ipynb
@@ -0,0 +1,133 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "4927a727-b4c8-453c-8c83-bd87b4fcac14",
+ "metadata": {},
+ "source": [
+ "# Adding moderation\n",
+ "\n",
+ "This shows how to add in moderation (or other safeguards) around your LLM application."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "id": "4f5f6449-940a-4f5c-97c0-39b71c3e2a68",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chains import OpenAIModerationChain\n",
+ "from langchain.llms import OpenAI\n",
+ "from langchain.prompts import ChatPromptTemplate"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "fcb8312b-7e7a-424f-a3ec-76738c9a9d21",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "moderate = OpenAIModerationChain()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "b24b9148-f6b0-4091-8ea8-d3fb281bd950",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "model = OpenAI()\n",
+ "prompt = ChatPromptTemplate.from_messages([\n",
+ " (\"system\", \"repeat after me: {input}\")\n",
+ "])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "id": "1c8ed87c-9ca6-4559-bf60-d40e94a0af08",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = prompt | model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "id": "5256b9bd-381a-42b0-bfa8-7e6d18f853cb",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'\\n\\nYou are stupid.'"
+ ]
+ },
+ "execution_count": 23,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"input\": \"you are stupid\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "id": "fe6e3b33-dc9a-49d5-b194-ba750c58a628",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "moderated_chain = chain | moderate"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "id": "d8ba0cbd-c739-4d23-be9f-6ae092bd5ffb",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'input': '\\n\\nYou are stupid',\n",
+ " 'output': \"Text was found that violates OpenAI's content policy.\"}"
+ ]
+ },
+ "execution_count": 25,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "moderated_chain.invoke({\"input\": \"you are stupid\"})"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/multiple_chains.ipynb b/docs/extras/expression_language/cookbook/multiple_chains.ipynb
new file mode 100644
index 0000000000..7db06a85f5
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/multiple_chains.ipynb
@@ -0,0 +1,240 @@
+{
+ "cells": [
+ {
+ "cell_type": "raw",
+ "id": "877102d1-02ea-4fa3-8ec7-a08e242b95b3",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 2\n",
+ "title: Multiple chains\n",
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0f2bf8d3",
+ "metadata": {},
+ "source": [
+ "Runnables can easily be used to string together multiple Chains"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "d65d4e9e",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'El país donde se encuentra la ciudad de Honolulu, donde nació Barack Obama, el 44º Presidente de los Estados Unidos, es Estados Unidos. Honolulu se encuentra en la isla de Oahu, en el estado de Hawái.'"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from operator import itemgetter\n",
+ "\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.prompts import ChatPromptTemplate\n",
+ "from langchain.schema import StrOutputParser\n",
+ "\n",
+ "prompt1 = ChatPromptTemplate.from_template(\"what is the city {person} is from?\")\n",
+ "prompt2 = ChatPromptTemplate.from_template(\"what country is the city {city} in? respond in {language}\")\n",
+ "\n",
+ "model = ChatOpenAI()\n",
+ "\n",
+ "chain1 = prompt1 | model | StrOutputParser()\n",
+ "\n",
+ "chain2 = {\"city\": chain1, \"language\": itemgetter(\"language\")} | prompt2 | model | StrOutputParser()\n",
+ "\n",
+ "chain2.invoke({\"person\": \"obama\", \"language\": \"spanish\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "878f8176",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.schema.runnable import RunnableMap, RunnablePassthrough\n",
+ "\n",
+ "prompt1 = ChatPromptTemplate.from_template(\"generate a {attribute} color. Return the name of the color and nothing else:\")\n",
+ "prompt2 = ChatPromptTemplate.from_template(\"what is a fruit of color: {color}. Return the name of the fruit and nothing else:\")\n",
+ "prompt3 = ChatPromptTemplate.from_template(\"what is a country with a flag that has the color: {color}. Return the name of the country and nothing else:\")\n",
+ "prompt4 = ChatPromptTemplate.from_template(\"What is the color of {fruit} and the flag of {country}?\")\n",
+ "\n",
+ "model_parser = model | StrOutputParser()\n",
+ "\n",
+ "color_generator = {\"attribute\": RunnablePassthrough()} | prompt1 | {\"color\": model_parser}\n",
+ "color_to_fruit = prompt2 | model_parser\n",
+ "color_to_country = prompt3 | model_parser\n",
+ "question_generator = color_generator | {\"fruit\": color_to_fruit, \"country\": color_to_country} | prompt4"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "d621a870",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "ChatPromptValue(messages=[HumanMessage(content='What is the color of strawberry and the flag of China?', additional_kwargs={}, example=False)])"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "question_generator.invoke({\"warm\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "b4a9812b-bead-4fd9-ae27-0b8be57e5dc1",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='The color of an apple is typically red or green. The flag of China is predominantly red with a large yellow star in the upper left corner and four smaller yellow stars surrounding it.', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 10,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "prompt = question_generator.invoke({\"warm\"})\n",
+ "model.invoke(prompt)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "6d75a313-f1c8-4e94-9a17-24e0bf4a2bdc",
+ "metadata": {},
+ "source": [
+ "### Branching and Merging\n",
+ "\n",
+ "You may want the output of one component to be processed by 2 or more other components. [RunnableMaps](https://api.python.langchain.com/en/latest/schema/langchain.schema.runnable.base.RunnableMap.html) let you split or fork the chain so multiple components can process the input in parallel. Later, other components can join or merge the results to synthesize a final response. This type of chain creates a computation graph that looks like the following:\n",
+ "\n",
+ "```text\n",
+ " Input\n",
+ " / \\\n",
+ " / \\\n",
+ " Branch1 Branch2\n",
+ " \\ /\n",
+ " \\ /\n",
+ " Combine\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "247fa0bd-4596-4063-8cb3-1d7fc119d982",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "planner = (\n",
+ " ChatPromptTemplate.from_template(\n",
+ " \"Generate an argument about: {input}\"\n",
+ " )\n",
+ " | ChatOpenAI()\n",
+ " | StrOutputParser()\n",
+ " | {\"base_response\": RunnablePassthrough()}\n",
+ ")\n",
+ "\n",
+ "arguments_for = (\n",
+ " ChatPromptTemplate.from_template(\n",
+ " \"List the pros or positive aspects of {base_response}\"\n",
+ " )\n",
+ " | ChatOpenAI()\n",
+ " | StrOutputParser()\n",
+ ")\n",
+ "arguments_against = (\n",
+ " ChatPromptTemplate.from_template(\n",
+ " \"List the cons or negative aspects of {base_response}\"\n",
+ " )\n",
+ " | ChatOpenAI()\n",
+ " | StrOutputParser()\n",
+ ")\n",
+ "\n",
+ "final_responder = (\n",
+ " ChatPromptTemplate.from_messages(\n",
+ " [\n",
+ " (\"ai\", \"{original_response}\"),\n",
+ " (\"human\", \"Pros:\\n{results_1}\\n\\nCons:\\n{results_2}\"),\n",
+ " (\"system\", \"Generate a final response given the critique\"),\n",
+ " ]\n",
+ " )\n",
+ " | ChatOpenAI()\n",
+ " | StrOutputParser()\n",
+ ")\n",
+ "\n",
+ "chain = (\n",
+ " planner \n",
+ " | {\n",
+ " \"results_1\": arguments_for,\n",
+ " \"results_2\": arguments_against,\n",
+ " \"original_response\": itemgetter(\"base_response\"),\n",
+ " }\n",
+ " | final_responder\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "id": "2564f310-0674-4bb1-9c4e-d7848ca73511",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'While Scrum has its potential cons and challenges, many organizations have successfully embraced and implemented this project management framework to great effect. The cons mentioned above can be mitigated or overcome with proper training, support, and a commitment to continuous improvement. It is also important to note that not all cons may be applicable to every organization or project.\\n\\nFor example, while Scrum may be complex initially, with proper training and guidance, teams can quickly grasp the concepts and practices. The lack of predictability can be mitigated by implementing techniques such as velocity tracking and release planning. The limited documentation can be addressed by maintaining a balance between lightweight documentation and clear communication among team members. The dependency on team collaboration can be improved through effective communication channels and regular team-building activities.\\n\\nScrum can be scaled and adapted to larger projects by using frameworks like Scrum of Scrums or LeSS (Large Scale Scrum). Concerns about speed versus quality can be addressed by incorporating quality assurance practices, such as continuous integration and automated testing, into the Scrum process. Scope creep can be managed by having a well-defined and prioritized product backlog, and a strong product owner can be developed through training and mentorship.\\n\\nResistance to change can be overcome by providing proper education and communication to stakeholders and involving them in the decision-making process. Ultimately, the cons of Scrum can be seen as opportunities for growth and improvement, and with the right mindset and support, they can be effectively managed.\\n\\nIn conclusion, while Scrum may have its challenges and potential cons, the benefits and advantages it offers in terms of collaboration, flexibility, adaptability, transparency, and customer satisfaction make it a widely adopted and successful project management framework. With proper implementation and continuous improvement, organizations can leverage Scrum to drive innovation, efficiency, and project success.'"
+ ]
+ },
+ "execution_count": 12,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"input\": \"scrum\"})"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "poetry-venv",
+ "language": "python",
+ "name": "poetry-venv"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/prompt_llm_parser.ipynb b/docs/extras/expression_language/cookbook/prompt_llm_parser.ipynb
new file mode 100644
index 0000000000..1b670904d5
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/prompt_llm_parser.ipynb
@@ -0,0 +1,431 @@
+{
+ "cells": [
+ {
+ "cell_type": "raw",
+ "id": "abf7263d-3a62-4016-b5d5-b157f92f2070",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 0\n",
+ "title: Prompt + LLM\n",
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "9a434f2b-9405-468c-9dfd-254d456b57a6",
+ "metadata": {},
+ "source": [
+ "The most common and valuable composition is taking:\n",
+ "\n",
+ "``PromptTemplate`` / ``ChatPromptTemplate`` -> ``LLM`` / ``ChatModel`` -> ``OutputParser``\n",
+ "\n",
+ "Almost any other chains you build will use this building block."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "93aa2c87",
+ "metadata": {},
+ "source": [
+ "## PromptTemplate + LLM\n",
+ "\n",
+ "The simplest composition is just combing a prompt and model to create a chain that takes user input, adds it to a prompt, passes it to a model, and returns the raw model input.\n",
+ "\n",
+ "Note, you can mix and match PromptTemplate/ChatPromptTemplates and LLMs/ChatModels as you like here."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "id": "466b65b3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.prompts import ChatPromptTemplate\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "\n",
+ "prompt = ChatPromptTemplate.from_template(\"tell me a joke about {foo}\")\n",
+ "model = ChatOpenAI()\n",
+ "chain = prompt | model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "e3d0a6cd",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content=\"Why don't bears wear shoes?\\n\\nBecause they have bear feet!\", additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 2,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bears\"})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "7eb9ef50",
+ "metadata": {},
+ "source": [
+ "Often times we want to attach kwargs that'll be passed to each model call. Here's a few examples of that:"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0b1d8f88",
+ "metadata": {},
+ "source": [
+ "### Attaching Stop Sequences"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "562a06bf",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = prompt | model.bind(stop=[\"\\n\"])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "43f5d04c",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='Why did the bear never wear shoes?', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bears\"})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f3eaf88a",
+ "metadata": {},
+ "source": [
+ "### Attaching Function Call information"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "f94b71b2",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "functions = [\n",
+ " {\n",
+ " \"name\": \"joke\",\n",
+ " \"description\": \"A joke\",\n",
+ " \"parameters\": {\n",
+ " \"type\": \"object\",\n",
+ " \"properties\": {\n",
+ " \"setup\": {\n",
+ " \"type\": \"string\",\n",
+ " \"description\": \"The setup for the joke\"\n",
+ " },\n",
+ " \"punchline\": {\n",
+ " \"type\": \"string\",\n",
+ " \"description\": \"The punchline for the joke\"\n",
+ " }\n",
+ " },\n",
+ " \"required\": [\"setup\", \"punchline\"]\n",
+ " }\n",
+ " }\n",
+ " ]\n",
+ "chain = prompt | model.bind(function_call= {\"name\": \"joke\"}, functions= functions)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "decf7710",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='', additional_kwargs={'function_call': {'name': 'joke', 'arguments': '{\\n \"setup\": \"Why don\\'t bears wear shoes?\",\\n \"punchline\": \"Because they have bear feet!\"\\n}'}}, example=False)"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bears\"}, config={})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "9098c5ed",
+ "metadata": {},
+ "source": [
+ "## PromptTemplate + LLM + OutputParser\n",
+ "\n",
+ "We can also add in an output parser to easily trasform the raw LLM/ChatModel output into a more workable format"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "cc194c78",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.schema.output_parser import StrOutputParser\n",
+ "\n",
+ "chain = prompt | model | StrOutputParser()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "77acf448",
+ "metadata": {},
+ "source": [
+ "Notice that this now returns a string - a much more workable format for downstream tasks"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "e3d69a18",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "\"Why don't bears wear shoes?\\n\\nBecause they have bear feet!\""
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bears\"})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "c01864e5",
+ "metadata": {},
+ "source": [
+ "### Functions Output Parser\n",
+ "\n",
+ "When you specify the function to return, you may just want to parse that directly"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "ad0dd88e",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.output_parsers.openai_functions import JsonOutputFunctionsParser\n",
+ "\n",
+ "chain = (\n",
+ " prompt \n",
+ " | model.bind(function_call= {\"name\": \"joke\"}, functions= functions) \n",
+ " | JsonOutputFunctionsParser()\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "1e7aa8eb",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'setup': \"Why don't bears like fast food?\",\n",
+ " 'punchline': \"Because they can't catch it!\"}"
+ ]
+ },
+ "execution_count": 10,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bears\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "d4aa1a01",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.output_parsers.openai_functions import JsonKeyOutputFunctionsParser\n",
+ "\n",
+ "chain = (\n",
+ " prompt \n",
+ " | model.bind(function_call= {\"name\": \"joke\"}, functions= functions) \n",
+ " | JsonKeyOutputFunctionsParser(key_name=\"setup\")\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "id": "8b6df9ba",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "\"Why don't bears wear shoes?\""
+ ]
+ },
+ "execution_count": 12,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bears\"})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "023fbccb-ef7d-489e-a9ba-f98e17283d51",
+ "metadata": {},
+ "source": [
+ "## Simplifying input\n",
+ "\n",
+ "To make invocation even simpler, we can add a `RunnableMap` to take care of creating the prompt input dict for us:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "id": "9601c0f0-71f9-4bd4-a672-7bd04084b018",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.schema.runnable import RunnableMap, RunnablePassthrough\n",
+ "\n",
+ "map_ = RunnableMap({\"foo\": RunnablePassthrough()})\n",
+ "chain = (\n",
+ " map_ \n",
+ " | prompt\n",
+ " | model.bind(function_call= {\"name\": \"joke\"}, functions= functions) \n",
+ " | JsonKeyOutputFunctionsParser(key_name=\"setup\")\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "id": "7ec4f154-fda5-4847-9220-41aa902fdc33",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "\"Why don't bears wear shoes?\""
+ ]
+ },
+ "execution_count": 14,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke(\"bears\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "def00bfe-0f83-4805-8c8f-8a53f99fa8ea",
+ "metadata": {},
+ "source": [
+ "Since we're composing our map with another Runnable, we can even use some syntactic sugar and just use a dict:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "7bf3846a-02ee-41a3-ba1b-a708827d4f3a",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = (\n",
+ " {\"foo\": RunnablePassthrough()} \n",
+ " | prompt\n",
+ " | model.bind(function_call= {\"name\": \"joke\"}, functions= functions) \n",
+ " | JsonKeyOutputFunctionsParser(key_name=\"setup\")\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "id": "e566d6a1-538d-4cb5-a210-a63e082e4c74",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "\"Why don't bears like fast food?\""
+ ]
+ },
+ "execution_count": 22,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke(\"bears\")"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/retrieval.ipynb b/docs/extras/expression_language/cookbook/retrieval.ipynb
new file mode 100644
index 0000000000..6579b1c7f3
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/retrieval.ipynb
@@ -0,0 +1,461 @@
+{
+ "cells": [
+ {
+ "cell_type": "raw",
+ "id": "abe47592-909c-4844-bf44-9e55c2fb4bfa",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: RAG\n",
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "91c5ef3d",
+ "metadata": {},
+ "source": [
+ "Let's look at adding in a retrieval step to a prompt and LLM, which adds up to a \"retrieval-augmented generation\" chain"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "7f25d9e9-d192-42e9-af50-5660a4bfb0d9",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip install langchain openai faiss-cpu"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "33be32af",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from operator import itemgetter\n",
+ "\n",
+ "from langchain.prompts import ChatPromptTemplate\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.embeddings import OpenAIEmbeddings\n",
+ "from langchain.schema.output_parser import StrOutputParser\n",
+ "from langchain.schema.runnable import RunnablePassthrough\n",
+ "from langchain.vectorstores import FAISS"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "bfc47ec1",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "vectorstore = FAISS.from_texts([\"harrison worked at kensho\"], embedding=OpenAIEmbeddings())\n",
+ "retriever = vectorstore.as_retriever()\n",
+ "\n",
+ "template = \"\"\"Answer the question based only on the following context:\n",
+ "{context}\n",
+ "\n",
+ "Question: {question}\n",
+ "\"\"\"\n",
+ "prompt = ChatPromptTemplate.from_template(template)\n",
+ "\n",
+ "model = ChatOpenAI()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "eae31755",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = (\n",
+ " {\"context\": retriever, \"question\": RunnablePassthrough()} \n",
+ " | prompt \n",
+ " | model \n",
+ " | StrOutputParser()\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "f3040b0c",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'Harrison worked at Kensho.'"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke(\"where did harrison work?\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "e1d20c7c",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "template = \"\"\"Answer the question based only on the following context:\n",
+ "{context}\n",
+ "\n",
+ "Question: {question}\n",
+ "\n",
+ "Answer in the following language: {language}\n",
+ "\"\"\"\n",
+ "prompt = ChatPromptTemplate.from_template(template)\n",
+ "\n",
+ "chain = {\n",
+ " \"context\": itemgetter(\"question\") | retriever, \n",
+ " \"question\": itemgetter(\"question\"), \n",
+ " \"language\": itemgetter(\"language\")\n",
+ "} | prompt | model | StrOutputParser()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "7ee8b2d4",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'Harrison ha lavorato a Kensho.'"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"question\": \"where did harrison work\", \"language\": \"italian\"})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f007669c",
+ "metadata": {},
+ "source": [
+ "## Conversational Retrieval Chain\n",
+ "\n",
+ "We can easily add in conversation history. This primarily means adding in chat_message_history"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "3f30c348",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.schema.runnable import RunnableMap\n",
+ "from langchain.schema import format_document"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "64ab1dbf",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.prompts.prompt import PromptTemplate\n",
+ "\n",
+ "_template = \"\"\"Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question, in its original language.\n",
+ "\n",
+ "Chat History:\n",
+ "{chat_history}\n",
+ "Follow Up Input: {question}\n",
+ "Standalone question:\"\"\"\n",
+ "CONDENSE_QUESTION_PROMPT = PromptTemplate.from_template(_template)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "7d628c97",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "template = \"\"\"Answer the question based only on the following context:\n",
+ "{context}\n",
+ "\n",
+ "Question: {question}\n",
+ "\"\"\"\n",
+ "ANSWER_PROMPT = ChatPromptTemplate.from_template(template)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "f60a5d0f",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "DEFAULT_DOCUMENT_PROMPT = PromptTemplate.from_template(template=\"{page_content}\")\n",
+ "def _combine_documents(docs, document_prompt = DEFAULT_DOCUMENT_PROMPT, document_separator=\"\\n\\n\"):\n",
+ " doc_strings = [format_document(doc, document_prompt) for doc in docs]\n",
+ " return document_separator.join(doc_strings)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "id": "7d007db6",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from typing import Tuple, List\n",
+ "def _format_chat_history(chat_history: List[Tuple]) -> str:\n",
+ " buffer = \"\"\n",
+ " for dialogue_turn in chat_history:\n",
+ " human = \"Human: \" + dialogue_turn[0]\n",
+ " ai = \"Assistant: \" + dialogue_turn[1]\n",
+ " buffer += \"\\n\" + \"\\n\".join([human, ai])\n",
+ " return buffer"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "id": "5c32cc89",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "_inputs = RunnableMap(\n",
+ " {\n",
+ " \"standalone_question\": {\n",
+ " \"question\": lambda x: x[\"question\"],\n",
+ " \"chat_history\": lambda x: _format_chat_history(x['chat_history'])\n",
+ " } | CONDENSE_QUESTION_PROMPT | ChatOpenAI(temperature=0) | StrOutputParser(),\n",
+ " }\n",
+ ")\n",
+ "_context = {\n",
+ " \"context\": itemgetter(\"standalone_question\") | retriever | _combine_documents,\n",
+ " \"question\": lambda x: x[\"standalone_question\"]\n",
+ "}\n",
+ "conversational_qa_chain = _inputs | _context | ANSWER_PROMPT | ChatOpenAI()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "id": "135c8205",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='Harrison was employed at Kensho.', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 14,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "conversational_qa_chain.invoke({\n",
+ " \"question\": \"where did harrison work?\",\n",
+ " \"chat_history\": [],\n",
+ "})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "id": "424e7e7a",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='Harrison worked at Kensho.', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 15,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "conversational_qa_chain.invoke({\n",
+ " \"question\": \"where did he work?\",\n",
+ " \"chat_history\": [(\"Who wrote this notebook?\", \"Harrison\")],\n",
+ "})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "c5543183",
+ "metadata": {},
+ "source": [
+ "### With Memory and returning source documents\n",
+ "\n",
+ "This shows how to use memory with the above. For memory, we need to manage that outside at the memory. For returning the retrieved documents, we just need to pass them through all the way."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "id": "e31dd17c",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.memory import ConversationBufferMemory"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "id": "d4bffe94",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "memory = ConversationBufferMemory(return_messages=True, output_key=\"answer\", input_key=\"question\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "id": "733be985",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# First we add a step to load memory\n",
+ "# This needs to be a RunnableMap because its the first input\n",
+ "loaded_memory = RunnableMap(\n",
+ " {\n",
+ " \"question\": itemgetter(\"question\"),\n",
+ " \"memory\": memory.load_memory_variables,\n",
+ " }\n",
+ ")\n",
+ "# Next we add a step to expand memory into the variables\n",
+ "expanded_memory = {\n",
+ " \"question\": itemgetter(\"question\"),\n",
+ " \"chat_history\": lambda x: x[\"memory\"][\"history\"]\n",
+ "}\n",
+ "\n",
+ "# Now we calculate the standalone question\n",
+ "standalone_question = {\n",
+ " \"standalone_question\": {\n",
+ " \"question\": lambda x: x[\"question\"],\n",
+ " \"chat_history\": lambda x: _format_chat_history(x['chat_history'])\n",
+ " } | CONDENSE_QUESTION_PROMPT | ChatOpenAI(temperature=0) | StrOutputParser(),\n",
+ "}\n",
+ "# Now we retrieve the documents\n",
+ "retrieved_documents = {\n",
+ " \"docs\": itemgetter(\"standalone_question\") | retriever,\n",
+ " \"question\": lambda x: x[\"standalone_question\"]\n",
+ "}\n",
+ "# Now we construct the inputs for the final prompt\n",
+ "final_inputs = {\n",
+ " \"context\": lambda x: _combine_documents(x[\"docs\"]),\n",
+ " \"question\": itemgetter(\"question\")\n",
+ "}\n",
+ "# And finally, we do the part that returns the answers\n",
+ "answer = {\n",
+ " \"answer\": final_inputs | ANSWER_PROMPT | ChatOpenAI(),\n",
+ " \"docs\": itemgetter(\"docs\"),\n",
+ "}\n",
+ "# And now we put it all together!\n",
+ "final_chain = loaded_memory | expanded_memory | standalone_question | retrieved_documents | answer"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "id": "806e390c",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'answer': AIMessage(content='Harrison was employed at Kensho.', additional_kwargs={}, example=False),\n",
+ " 'docs': [Document(page_content='harrison worked at kensho', metadata={})]}"
+ ]
+ },
+ "execution_count": 19,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "inputs = {\"question\": \"where did harrison work?\"}\n",
+ "result = final_chain.invoke(inputs)\n",
+ "result"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "id": "977399fd",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Note that the memory does not save automatically\n",
+ "# This will be improved in the future\n",
+ "# For now you need to save it yourself\n",
+ "memory.save_context(inputs, {\"answer\": result[\"answer\"].content})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "f94f7de4",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'history': [HumanMessage(content='where did harrison work?', additional_kwargs={}, example=False),\n",
+ " AIMessage(content='Harrison was employed at Kensho.', additional_kwargs={}, example=False)]}"
+ ]
+ },
+ "execution_count": 21,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "memory.load_memory_variables({})"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "poetry-venv",
+ "language": "python",
+ "name": "poetry-venv"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/sql_db.ipynb b/docs/extras/expression_language/cookbook/sql_db.ipynb
new file mode 100644
index 0000000000..0cf0748009
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/sql_db.ipynb
@@ -0,0 +1,227 @@
+{
+ "cells": [
+ {
+ "cell_type": "raw",
+ "id": "c14da114-1a4a-487d-9cff-e0e8c30ba366",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 3\n",
+ "title: Querying a SQL DB\n",
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "506e9636",
+ "metadata": {},
+ "source": [
+ "We can replicate our SQLDatabaseChain with Runnables."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "id": "7a927516",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.prompts import ChatPromptTemplate\n",
+ "\n",
+ "template = \"\"\"Based on the table schema below, write a SQL query that would answer the user's question:\n",
+ "{schema}\n",
+ "\n",
+ "Question: {question}\n",
+ "SQL Query:\"\"\"\n",
+ "prompt = ChatPromptTemplate.from_template(template)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "3f51f386",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.utilities import SQLDatabase"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "7c3449d6-684b-416e-ba16-90a035835a88",
+ "metadata": {},
+ "source": [
+ "We'll need the Chinook sample DB for this example. There's many places to download it from, e.g. https://database.guide/2-sample-databases-sqlite/"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "id": "2ccca6fc",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "db = SQLDatabase.from_uri(\"sqlite:///./Chinook.db\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "05ba88ee",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def get_schema(_):\n",
+ " return db.get_table_info()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "id": "a4eda902",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def run_query(query):\n",
+ " return db.run(query)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "id": "5046cb17",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from operator import itemgetter\n",
+ "\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.schema.output_parser import StrOutputParser\n",
+ "from langchain.schema.runnable import RunnableLambda, RunnableMap\n",
+ "\n",
+ "model = ChatOpenAI()\n",
+ "\n",
+ "inputs = {\n",
+ " \"schema\": RunnableLambda(get_schema),\n",
+ " \"question\": itemgetter(\"question\")\n",
+ "}\n",
+ "sql_response = (\n",
+ " RunnableMap(inputs)\n",
+ " | prompt\n",
+ " | model.bind(stop=[\"\\nSQLResult:\"])\n",
+ " | StrOutputParser()\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "id": "a5552039",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'SELECT COUNT(*) FROM Employee'"
+ ]
+ },
+ "execution_count": 24,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "sql_response.invoke({\"question\": \"How many employees are there?\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "id": "d6fee130",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "template = \"\"\"Based on the table schema below, question, sql query, and sql response, write a natural language response:\n",
+ "{schema}\n",
+ "\n",
+ "Question: {question}\n",
+ "SQL Query: {query}\n",
+ "SQL Response: {response}\"\"\"\n",
+ "prompt_response = ChatPromptTemplate.from_template(template)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "id": "923aa634",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "full_chain = (\n",
+ " RunnableMap({\n",
+ " \"question\": itemgetter(\"question\"),\n",
+ " \"query\": sql_response,\n",
+ " }) \n",
+ " | {\n",
+ " \"schema\": RunnableLambda(get_schema),\n",
+ " \"question\": itemgetter(\"question\"),\n",
+ " \"query\": itemgetter(\"query\"),\n",
+ " \"response\": lambda x: db.run(x[\"query\"]) \n",
+ " } \n",
+ " | prompt_response \n",
+ " | model\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 27,
+ "id": "e94963d8",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='There are 8 employees.', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 27,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "full_chain.invoke({\"question\": \"How many employees are there?\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "4f358d7b-a721-4db3-9f92-f06913428afc",
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/cookbook/tools.ipynb b/docs/extras/expression_language/cookbook/tools.ipynb
new file mode 100644
index 0000000000..d13dece3c9
--- /dev/null
+++ b/docs/extras/expression_language/cookbook/tools.ipynb
@@ -0,0 +1,122 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "29781123",
+ "metadata": {},
+ "source": [
+ "# Using tools\n",
+ "\n",
+ "You can use any Tools with Runnables easily."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "a5c579dd-2e22-41b0-a789-346dfdecb5a2",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip install duckduckgo-search"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "9232d2a9",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.prompts import ChatPromptTemplate\n",
+ "from langchain.schema.output_parser import StrOutputParser\n",
+ "from langchain.tools import DuckDuckGoSearchRun"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "a0c64d2c",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "search = DuckDuckGoSearchRun()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "391969b6",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "template = \"\"\"turn the following user input into a search query for a search engine:\n",
+ "\n",
+ "{input}\"\"\"\n",
+ "prompt = ChatPromptTemplate.from_template(template)\n",
+ "\n",
+ "model = ChatOpenAI()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "e3d9d20d",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "chain = prompt | model | StrOutputParser() | search"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "55f2967d",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'What sports games are on TV today & tonight? Watch and stream live sports on TV today, tonight, tomorrow. Today\\'s 2023 sports TV schedule includes football, basketball, baseball, hockey, motorsports, soccer and more. Watch on TV or stream online on ESPN, FOX, FS1, CBS, NBC, ABC, Peacock, Paramount+, fuboTV, local channels and many other networks. MLB Games Tonight: How to Watch on TV, Streaming & Odds - Thursday, September 7. Seattle Mariners\\' Julio Rodriguez greets teammates in the dugout after scoring against the Oakland Athletics in a ... Circle - Country Music and Lifestyle. Live coverage of all the MLB action today is available to you, with the information provided below. The Brewers will look to pick up a road win at PNC Park against the Pirates on Wednesday at 12:35 PM ET. Check out the latest odds and with BetMGM Sportsbook. Use bonus code \"GNPLAY\" for special offers! MLB Games Tonight: How to Watch on TV, Streaming & Odds - Tuesday, September 5. Houston Astros\\' Kyle Tucker runs after hitting a double during the fourth inning of a baseball game against the Los Angeles Angels, Sunday, Aug. 13, 2023, in Houston. (AP Photo/Eric Christian Smith) (APMedia) The Houston Astros versus the Texas Rangers is one of ... The second half of tonight\\'s college football schedule still has some good games remaining to watch on your television.. We\\'ve already seen an exciting one when Colorado upset TCU. And we saw some ...'"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"input\": \"I'd like to figure out what games are tonight\"})"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "a16949cf-00ea-43c6-a6aa-797ad4f6918d",
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "poetry-venv",
+ "language": "python",
+ "name": "poetry-venv"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/how_to/_category_.yml b/docs/extras/expression_language/how_to/_category_.yml
new file mode 100644
index 0000000000..39fa22bfbf
--- /dev/null
+++ b/docs/extras/expression_language/how_to/_category_.yml
@@ -0,0 +1,2 @@
+label: 'How to'
+position: 1
\ No newline at end of file
diff --git a/docs/extras/expression_language/how_to/functions.ipynb b/docs/extras/expression_language/how_to/functions.ipynb
new file mode 100644
index 0000000000..fc2f0a2962
--- /dev/null
+++ b/docs/extras/expression_language/how_to/functions.ipynb
@@ -0,0 +1,158 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "fbc4bf6e",
+ "metadata": {},
+ "source": [
+ "# Run arbitrary functions\n",
+ "\n",
+ "You can use arbitrary functions in the pipeline\n",
+ "\n",
+ "Note that all inputs to these functions need to be a SINGLE argument. If you have a function that accepts multiple arguments, you should write a wrapper that accepts a single input and unpacks it into multiple argument."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 77,
+ "id": "6bb221b3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.schema.runnable import RunnableLambda\n",
+ "\n",
+ "def length_function(text):\n",
+ " return len(text)\n",
+ "\n",
+ "def _multiple_length_function(text1, text2):\n",
+ " return len(text1) * len(text2)\n",
+ "\n",
+ "def multiple_length_function(_dict):\n",
+ " return _multiple_length_function(_dict[\"text1\"], _dict[\"text2\"])\n",
+ "\n",
+ "prompt = ChatPromptTemplate.from_template(\"what is {a} + {b}\")\n",
+ "\n",
+ "chain1 = prompt | model\n",
+ "\n",
+ "chain = {\n",
+ " \"a\": itemgetter(\"foo\") | RunnableLambda(length_function),\n",
+ " \"b\": {\"text1\": itemgetter(\"foo\"), \"text2\": itemgetter(\"bar\")} | RunnableLambda(multiple_length_function)\n",
+ "} | prompt | model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 78,
+ "id": "5488ec85",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "AIMessage(content='3 + 9 equals 12.', additional_kwargs={}, example=False)"
+ ]
+ },
+ "execution_count": 78,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.invoke({\"foo\": \"bar\", \"bar\": \"gah\"})"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "4728ddd9-914d-42ce-ae9b-72c9ce8ec940",
+ "metadata": {},
+ "source": [
+ "## Accepting a Runnable Config\n",
+ "\n",
+ "Runnable lambdas can optionally accept a [RunnableConfig](https://api.python.langchain.com/en/latest/schema/langchain.schema.runnable.config.RunnableConfig.html?highlight=runnableconfig#langchain.schema.runnable.config.RunnableConfig), which they can use to pass callbacks, tags, and other configuration information to nested runs."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 139,
+ "id": "80b3b5f6-5d58-44b9-807e-cce9a46bf49f",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.schema.runnable import RunnableConfig"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 149,
+ "id": "ff0daf0c-49dd-4d21-9772-e5fa133c5f36",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import json\n",
+ "\n",
+ "def parse_or_fix(text: str, config: RunnableConfig):\n",
+ " fixing_chain = (\n",
+ " ChatPromptTemplate.from_template(\n",
+ " \"Fix the following text:\\n\\n```text\\n{input}\\n```\\nError: {error}\"\n",
+ " \" Don't narrate, just respond with the fixed data.\"\n",
+ " )\n",
+ " | ChatOpenAI()\n",
+ " | StrOutputParser()\n",
+ " )\n",
+ " for _ in range(3):\n",
+ " try:\n",
+ " return json.loads(text)\n",
+ " except Exception as e:\n",
+ " text = fixing_chain.invoke({\"input\": text, \"error\": e}, config)\n",
+ " return \"Failed to parse\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 152,
+ "id": "1a5e709e-9d75-48c7-bb9c-503251990505",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Tokens Used: 65\n",
+ "\tPrompt Tokens: 56\n",
+ "\tCompletion Tokens: 9\n",
+ "Successful Requests: 1\n",
+ "Total Cost (USD): $0.00010200000000000001\n"
+ ]
+ }
+ ],
+ "source": [
+ "from langchain.callbacks import get_openai_callback\n",
+ "\n",
+ "with get_openai_callback() as cb:\n",
+ " RunnableLambda(parse_or_fix).invoke(\"{foo: bar}\", {\"tags\": [\"my-tag\"], \"callbacks\": [cb]})\n",
+ " print(cb)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/expression_language/interface.ipynb b/docs/extras/expression_language/interface.ipynb
index cf19bfe4db..c47800ecad 100644
--- a/docs/extras/expression_language/interface.ipynb
+++ b/docs/extras/expression_language/interface.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "366a0e68-fd67-4fe5-a292-5c33733339ea",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 0\n",
+ "title: Interface\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "9a9acd2e",
"metadata": {},
"source": [
- "# Interface\n",
- "\n",
"In an effort to make it as easy as possible to create custom chains, we've implemented a [\"Runnable\"](https://api.python.langchain.com/en/latest/schema/langchain.schema.runnable.Runnable.html#langchain.schema.runnable.Runnable) protocol that most components implement. This is a standard interface with a few different methods, which makes it easy to define custom chains as well as making it possible to invoke them in a standard way. The standard interface exposed includes:\n",
"\n",
"- `stream`: stream back chunks of the response\n",
@@ -429,7 +438,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.10.1"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/extras/guides/privacy/presidio_data_anonymization.ipynb b/docs/extras/guides/privacy/presidio_data_anonymization/index.ipynb
similarity index 85%
rename from docs/extras/guides/privacy/presidio_data_anonymization.ipynb
rename to docs/extras/guides/privacy/presidio_data_anonymization/index.ipynb
index faa9929259..2502a45092 100644
--- a/docs/extras/guides/privacy/presidio_data_anonymization.ipynb
+++ b/docs/extras/guides/privacy/presidio_data_anonymization/index.ipynb
@@ -6,7 +6,7 @@
"source": [
"# Data anonymization with Microsoft Presidio\n",
"\n",
- "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/guides/privacy/presidio_data_anonymization.ipynb)\n",
+ "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/guides/privacy/presidio_data_anonymization/index.ipynb)\n",
"\n",
"## Use case\n",
"\n",
@@ -28,7 +28,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
@@ -47,16 +47,16 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 2,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'My name is Mrs. Rachel Chen DDS, call me at 849-829-7628x073 or email me at christopherfrey@example.org'"
+ "'My name is Laura Ruiz, call me at +1-412-982-8374x13414 or email me at javierwatkins@example.net'"
]
},
- "execution_count": 14,
+ "execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
@@ -82,7 +82,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
@@ -94,35 +94,53 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "text = f\"\"\"Slim Shady recently lost his wallet. \n",
+ "Inside is some cash and his credit card with the number 4916 0387 9536 0861. \n",
+ "If you would find it, please call at 313-666-7440 or write an email here: real.slim.shady@gmail.com.\"\"\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
"metadata": {},
"outputs": [
{
- "data": {
- "text/plain": [
- "AIMessage(content='You can find our super secret data at https://www.ross.com/', additional_kwargs={}, example=False)"
- ]
- },
- "execution_count": 16,
- "metadata": {},
- "output_type": "execute_result"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Dear Sir/Madam,\n",
+ "\n",
+ "We regret to inform you that Richard Fields has recently misplaced his wallet, which contains a sum of cash and his credit card bearing the number 30479847307774. \n",
+ "\n",
+ "Should you happen to come across it, we kindly request that you contact us immediately at 6439182672 or via email at frank45@example.com.\n",
+ "\n",
+ "Thank you for your attention to this matter.\n",
+ "\n",
+ "Yours faithfully,\n",
+ "\n",
+ "[Your Name]\n"
+ ]
}
],
"source": [
"from langchain.prompts.prompt import PromptTemplate\n",
"from langchain.chat_models import ChatOpenAI\n",
- "from langchain.schema.runnable import RunnablePassthrough\n",
"\n",
- "template = \"\"\"According to this text, where can you find our super secret data?\n",
+ "anonymizer = PresidioAnonymizer()\n",
"\n",
- "{anonymized_text}\n",
+ "template = \"\"\"Rewrite this text into an official, short email:\n",
"\n",
- "Answer:\"\"\"\n",
+ "{anonymized_text}\"\"\"\n",
"prompt = PromptTemplate.from_template(template)\n",
- "llm = ChatOpenAI()\n",
+ "llm = ChatOpenAI(temperature=0)\n",
"\n",
"chain = {\"anonymized_text\": anonymizer.anonymize} | prompt | llm\n",
- "chain.invoke(\"You can find our super secret data at https://supersecretdata.com\")"
+ "response = chain.invoke(text)\n",
+ "print(response.content)"
]
},
{
@@ -135,16 +153,16 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'My name is Gabrielle Edwards, call me at 313-666-7440 or email me at real.slim.shady@gmail.com'"
+ "'My name is Adrian Fleming, call me at 313-666-7440 or email me at real.slim.shady@gmail.com'"
]
},
- "execution_count": 18,
+ "execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
@@ -166,16 +184,16 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'My name is Victoria Mckinney, call me at 713-549-8623 or email me at real.slim.shady@gmail.com'"
+ "'My name is Justin Miller, call me at 761-824-1889 or email me at real.slim.shady@gmail.com'"
]
},
- "execution_count": 3,
+ "execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
@@ -201,16 +219,16 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'My name is Billy Russo, call me at 970-996-9453x038 or email me at jamie80@example.org'"
+ "'My name is Dr. Jennifer Baker, call me at (508)839-9329x232 or email me at ehamilton@example.com'"
]
},
- "execution_count": 4,
+ "execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
@@ -232,16 +250,16 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'My polish phone number is EVIA70648911396944'"
+ "'My polish phone number is NRGN41434238921378'"
]
},
- "execution_count": 5,
+ "execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
@@ -261,7 +279,7 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
@@ -291,7 +309,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
@@ -308,7 +326,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 12,
"metadata": {},
"outputs": [
{
@@ -337,16 +355,16 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'+48 533 220 543'"
+ "'511 622 683'"
]
},
- "execution_count": 9,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
@@ -374,7 +392,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
@@ -389,7 +407,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
@@ -398,16 +416,16 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "'My polish phone number is +48 692 715 636'"
+ "'My polish phone number is +48 734 630 977'"
]
},
- "execution_count": 12,
+ "execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
@@ -421,8 +439,6 @@
"metadata": {},
"source": [
"## Future works\n",
- "\n",
- "- **deanonymization** - add the ability to reverse anonymization. For example, the workflow could look like this: `anonymize -> LLMChain -> deanonymize`. By doing this, we will retain anonymity in requests to, for example, OpenAI, and then be able restore the original data.\n",
"- **instance anonymization** - at this point, each occurrence of PII is treated as a separate entity and separately anonymized. Therefore, two occurrences of the name John Doe in the text will be changed to two different names. It is therefore worth introducing support for full instance detection, so that repeated occurrences are treated as a single object."
]
}
diff --git a/docs/extras/guides/privacy/presidio_data_anonymization/multi_language.ipynb b/docs/extras/guides/privacy/presidio_data_anonymization/multi_language.ipynb
new file mode 100644
index 0000000000..63ba8931a6
--- /dev/null
+++ b/docs/extras/guides/privacy/presidio_data_anonymization/multi_language.ipynb
@@ -0,0 +1,520 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Mutli-language data anonymization with Microsoft Presidio\n",
+ "\n",
+ "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/guides/privacy/presidio_data_anonymization/multi_language.ipynb)\n",
+ "\n",
+ "\n",
+ "## Use case\n",
+ "\n",
+ "Multi-language support in data pseudonymization is essential due to differences in language structures and cultural contexts. Different languages may have varying formats for personal identifiers. For example, the structure of names, locations and dates can differ greatly between languages and regions. Furthermore, non-alphanumeric characters, accents, and the direction of writing can impact pseudonymization processes. Without multi-language support, data could remain identifiable or be misinterpreted, compromising data privacy and accuracy. Hence, it enables effective and precise pseudonymization suited for global operations.\n",
+ "\n",
+ "## Overview\n",
+ "\n",
+ "PII detection in Microsoft Presidio relies on several components - in addition to the usual pattern matching (e.g. using regex), the analyser uses a model for Named Entity Recognition (NER) to extract entities such as:\n",
+ "- `PERSON`\n",
+ "- `LOCATION`\n",
+ "- `DATE_TIME`\n",
+ "- `NRP`\n",
+ "- `ORGANIZATION`\n",
+ "\n",
+ "[[Source]](https://github.com/microsoft/presidio/blob/main/presidio-analyzer/presidio_analyzer/predefined_recognizers/spacy_recognizer.py)\n",
+ "\n",
+ "To handle NER in specific languages, we utilize unique models from the `spaCy` library, recognized for its extensive selection covering multiple languages and sizes. However, it's not restrictive, allowing for integration of alternative frameworks such as [Stanza](https://microsoft.github.io/presidio/analyzer/nlp_engines/spacy_stanza/) or [transformers](https://microsoft.github.io/presidio/analyzer/nlp_engines/transformers/) when necessary.\n",
+ "\n",
+ "\n",
+ "## Quickstart\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Install necessary packages\n",
+ "# ! pip install langchain langchain-experimental openai presidio-analyzer presidio-anonymizer spacy Faker\n",
+ "# ! python -m spacy download en_core_web_lg"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer\n",
+ "\n",
+ "anonymizer = PresidioReversibleAnonymizer(\n",
+ " analyzed_fields=[\"PERSON\"],\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "By default, `PresidioAnonymizer` and `PresidioReversibleAnonymizer` use a model trained on English texts, so they handle other languages moderately well. \n",
+ "\n",
+ "For example, here the model did not detect the person:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'Me llamo Sofía'"
+ ]
+ },
+ "execution_count": 3,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "anonymizer.anonymize(\"Me llamo Sofía\") # \"My name is Sofía\" in Spanish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "They may also take words from another language as actual entities. Here, both the word *'Yo'* (*'I'* in Spanish) and *Sofía* have been classified as `PERSON`:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'Bridget Kirk soy Sally Knight'"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "anonymizer.anonymize(\"Yo soy Sofía\") # \"I am Sofía\" in Spanish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "If you want to anonymise texts from other languages, you need to download other models and add them to the anonymiser configuration:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Download the models for the languages you want to use\n",
+ "# ! python -m spacy download en_core_web_md\n",
+ "# ! python -m spacy download es_core_news_md"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "nlp_config = {\n",
+ " \"nlp_engine_name\": \"spacy\",\n",
+ " \"models\": [\n",
+ " {\"lang_code\": \"en\", \"model_name\": \"en_core_web_md\"},\n",
+ " {\"lang_code\": \"es\", \"model_name\": \"es_core_news_md\"},\n",
+ " ],\n",
+ "}"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We have therefore added a Spanish language model. Note also that we have downloaded an alternative model for English as well - in this case we have replaced the large model `en_core_web_lg` (560MB) with its smaller version `en_core_web_md` (40MB) - the size is therefore reduced by 14 times! If you care about the speed of anonymisation, it is worth considering it.\n",
+ "\n",
+ "All models for the different languages can be found in the [spaCy documentation](https://spacy.io/usage/models).\n",
+ "\n",
+ "Now pass the configuration as the `languages_config` parameter to Anonymiser. As you can see, both previous examples work flawlessly:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Me llamo Michelle Smith\n",
+ "Yo soy Rachel Wright\n"
+ ]
+ }
+ ],
+ "source": [
+ "anonymizer = PresidioReversibleAnonymizer(\n",
+ " analyzed_fields=[\"PERSON\"],\n",
+ " languages_config=nlp_config,\n",
+ ")\n",
+ "\n",
+ "print(\n",
+ " anonymizer.anonymize(\"Me llamo Sofía\", language=\"es\")\n",
+ ") # \"My name is Sofía\" in Spanish\n",
+ "print(anonymizer.anonymize(\"Yo soy Sofía\", language=\"es\")) # \"I am Sofía\" in Spanish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "By default, the language indicated first in the configuration will be used when anonymising text (in this case English):"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "My name is Ronnie Ayala\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(anonymizer.anonymize(\"My name is John\"))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Advanced usage\n",
+ "\n",
+ "### Custom labels in NER model"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "It may be that the spaCy model has different class names than those supported by the Microsoft Presidio by default. Take Polish, for example:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Text: Wiktoria, Start: 12, End: 20, Label: persName\n"
+ ]
+ }
+ ],
+ "source": [
+ "# ! python -m spacy download pl_core_news_md\n",
+ "\n",
+ "import spacy\n",
+ "\n",
+ "nlp = spacy.load(\"pl_core_news_md\")\n",
+ "doc = nlp(\"Nazywam się Wiktoria\") # \"My name is Wiktoria\" in Polish\n",
+ "\n",
+ "for ent in doc.ents:\n",
+ " print(\n",
+ " f\"Text: {ent.text}, Start: {ent.start_char}, End: {ent.end_char}, Label: {ent.label_}\"\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The name *Victoria* was classified as `persName`, which does not correspond to the default class names `PERSON`/`PER` implemented in Microsoft Presidio (look for `CHECK_LABEL_GROUPS` in [SpacyRecognizer implementation](https://github.com/microsoft/presidio/blob/main/presidio-analyzer/presidio_analyzer/predefined_recognizers/spacy_recognizer.py)). \n",
+ "\n",
+ "You can find out more about custom labels in spaCy models (including your own, trained ones) in [this thread](https://github.com/microsoft/presidio/issues/851).\n",
+ "\n",
+ "That's why our sentence will not be anonymized:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Nazywam się Wiktoria\n"
+ ]
+ }
+ ],
+ "source": [
+ "nlp_config = {\n",
+ " \"nlp_engine_name\": \"spacy\",\n",
+ " \"models\": [\n",
+ " {\"lang_code\": \"en\", \"model_name\": \"en_core_web_md\"},\n",
+ " {\"lang_code\": \"es\", \"model_name\": \"es_core_news_md\"},\n",
+ " {\"lang_code\": \"pl\", \"model_name\": \"pl_core_news_md\"},\n",
+ " ],\n",
+ "}\n",
+ "\n",
+ "anonymizer = PresidioReversibleAnonymizer(\n",
+ " analyzed_fields=[\"PERSON\", \"LOCATION\", \"DATE_TIME\"],\n",
+ " languages_config=nlp_config,\n",
+ ")\n",
+ "\n",
+ "print(\n",
+ " anonymizer.anonymize(\"Nazywam się Wiktoria\", language=\"pl\")\n",
+ ") # \"My name is Wiktoria\" in Polish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "To address this, create your own `SpacyRecognizer` with your own class mapping and add it to the anonymizer:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from presidio_analyzer.predefined_recognizers import SpacyRecognizer\n",
+ "\n",
+ "polish_check_label_groups = [\n",
+ " ({\"LOCATION\"}, {\"placeName\", \"geogName\"}),\n",
+ " ({\"PERSON\"}, {\"persName\"}),\n",
+ " ({\"DATE_TIME\"}, {\"date\", \"time\"}),\n",
+ "]\n",
+ "\n",
+ "spacy_recognizer = SpacyRecognizer(\n",
+ " supported_language=\"pl\",\n",
+ " check_label_groups=polish_check_label_groups,\n",
+ ")\n",
+ "\n",
+ "anonymizer.add_recognizer(spacy_recognizer)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now everything works smoothly:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Nazywam się Morgan Walters\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(\n",
+ " anonymizer.anonymize(\"Nazywam się Wiktoria\", language=\"pl\")\n",
+ ") # \"My name is Wiktoria\" in Polish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Let's try on more complex example:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Nazywam się Ernest Liu. New Taylorburgh to moje miasto rodzinne. Urodziłam się 1987-01-19\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(\n",
+ " anonymizer.anonymize(\n",
+ " \"Nazywam się Wiktoria. Płock to moje miasto rodzinne. Urodziłam się dnia 6 kwietnia 2001 roku\",\n",
+ " language=\"pl\",\n",
+ " )\n",
+ ") # \"My name is Wiktoria. Płock is my home town. I was born on 6 April 2001\" in Polish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "As you can see, thanks to class mapping, the anonymiser can cope with different types of entities. "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Custom language-specific operators\n",
+ "\n",
+ "In the example above, the sentence has been anonymised correctly, but the fake data does not fit the Polish language at all. Custom operators can therefore be added, which will resolve the issue:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from faker import Faker\n",
+ "from presidio_anonymizer.entities import OperatorConfig\n",
+ "\n",
+ "fake = Faker(locale=\"pl_PL\") # Setting faker to provide Polish data\n",
+ "\n",
+ "new_operators = {\n",
+ " \"PERSON\": OperatorConfig(\"custom\", {\"lambda\": lambda _: fake.first_name_female()}),\n",
+ " \"LOCATION\": OperatorConfig(\"custom\", {\"lambda\": lambda _: fake.city()}),\n",
+ "}\n",
+ "\n",
+ "anonymizer.add_operators(new_operators)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Nazywam się Marianna. Szczecin to moje miasto rodzinne. Urodziłam się 1976-11-16\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(\n",
+ " anonymizer.anonymize(\n",
+ " \"Nazywam się Wiktoria. Płock to moje miasto rodzinne. Urodziłam się dnia 6 kwietnia 2001 roku\",\n",
+ " language=\"pl\",\n",
+ " )\n",
+ ") # \"My name is Wiktoria. Płock is my home town. I was born on 6 April 2001\" in Polish"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Limitations\n",
+ "\n",
+ "Remember - results are as good as your recognizers and as your NER models!\n",
+ "\n",
+ "Look at the example below - we downloaded the small model for Spanish (12MB) and it no longer performs as well as the medium version (40MB):"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Model: es_core_news_sm. Result: Me llamo Sofía\n",
+ "Model: es_core_news_md. Result: Me llamo Lawrence Davis\n"
+ ]
+ }
+ ],
+ "source": [
+ "# ! python -m spacy download es_core_news_sm\n",
+ "\n",
+ "for model in [\"es_core_news_sm\", \"es_core_news_md\"]:\n",
+ " nlp_config = {\n",
+ " \"nlp_engine_name\": \"spacy\",\n",
+ " \"models\": [\n",
+ " {\"lang_code\": \"es\", \"model_name\": model},\n",
+ " ],\n",
+ " }\n",
+ "\n",
+ " anonymizer = PresidioReversibleAnonymizer(\n",
+ " analyzed_fields=[\"PERSON\"],\n",
+ " languages_config=nlp_config,\n",
+ " )\n",
+ "\n",
+ " print(\n",
+ " f\"Model: {model}. Result: {anonymizer.anonymize('Me llamo Sofía', language='es')}\"\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "In many cases, even the larger models from spaCy will not be sufficient - there are already other, more complex and better methods of detecting named entities, based on transformers. You can read more about this [here](https://microsoft.github.io/presidio/analyzer/nlp_engines/transformers/)."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Future works\n",
+ "\n",
+ "- **automatic language detection** - instead of passing the language as a parameter in `anonymizer.anonymize`, we could detect the language/s beforehand and then use the corresponding NER model."
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/docs/extras/guides/privacy/presidio_data_anonymization/reversible.ipynb b/docs/extras/guides/privacy/presidio_data_anonymization/reversible.ipynb
new file mode 100644
index 0000000000..de5655ba1e
--- /dev/null
+++ b/docs/extras/guides/privacy/presidio_data_anonymization/reversible.ipynb
@@ -0,0 +1,461 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Reversible data anonymization with Microsoft Presidio\n",
+ "\n",
+ "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/guides/privacy/presidio_data_anonymization/reversible.ipynb)\n",
+ "\n",
+ "\n",
+ "## Use case\n",
+ "\n",
+ "We have already written about the importance of anonymizing sensitive data in the previous section. **Reversible Anonymization** is an equally essential technology while sharing information with language models, as it balances data protection with data usability. This technique involves masking sensitive personally identifiable information (PII), yet it can be reversed and original data can be restored when authorized users need it. Its main advantage lies in the fact that while it conceals individual identities to prevent misuse, it also allows the concealed data to be accurately unmasked should it be necessary for legal or compliance purposes. \n",
+ "\n",
+ "## Overview\n",
+ "\n",
+ "We implemented the `PresidioReversibleAnonymizer`, which consists of two parts:\n",
+ "\n",
+ "1. anonymization - it works the same way as `PresidioAnonymizer`, plus the object itself stores a mapping of made-up values to original ones, for example:\n",
+ "```\n",
+ " {\n",
+ " \"PERSON\": {\n",
+ " \"\": \"\",\n",
+ " \"John Doe\": \"Slim Shady\"\n",
+ " },\n",
+ " \"PHONE_NUMBER\": {\n",
+ " \"111-111-1111\": \"555-555-5555\"\n",
+ " }\n",
+ " ...\n",
+ " }\n",
+ "```\n",
+ "\n",
+ "2. deanonymization - using the mapping described above, it matches fake data with original data and then substitutes it.\n",
+ "\n",
+ "Between anonymization and deanonymization user can perform different operations, for example, passing the output to LLM.\n",
+ "\n",
+ "## Quickstart\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Install necessary packages\n",
+ "# ! pip install langchain langchain-experimental openai presidio-analyzer presidio-anonymizer spacy Faker\n",
+ "# ! python -m spacy download en_core_web_lg"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "`PresidioReversibleAnonymizer` is not significantly different from its predecessor (`PresidioAnonymizer`) in terms of anonymization:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'My name is Maria Lynch, call me at 7344131647 or email me at jamesmichael@example.com. By the way, my card number is: 4838637940262'"
+ ]
+ },
+ "execution_count": 2,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer\n",
+ "\n",
+ "anonymizer = PresidioReversibleAnonymizer(\n",
+ " analyzed_fields=[\"PERSON\", \"PHONE_NUMBER\", \"EMAIL_ADDRESS\", \"CREDIT_CARD\"],\n",
+ " # Faker seed is used here to make sure the same fake data is generated for the test purposes\n",
+ " # In production, it is recommended to remove the faker_seed parameter (it will default to None)\n",
+ " faker_seed=42,\n",
+ ")\n",
+ "\n",
+ "anonymizer.anonymize(\n",
+ " \"My name is Slim Shady, call me at 313-666-7440 or email me at real.slim.shady@gmail.com. \"\n",
+ " \"By the way, my card number is: 4916 0387 9536 0861\"\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "This is what the full string we want to deanonymize looks like:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Maria Lynch recently lost his wallet. \n",
+ "Inside is some cash and his credit card with the number 4838637940262. \n",
+ "If you would find it, please call at 7344131647 or write an email here: jamesmichael@example.com.\n",
+ "Maria Lynch would be very grateful!\n"
+ ]
+ }
+ ],
+ "source": [
+ "# We know this data, as we set the faker_seed parameter\n",
+ "fake_name = \"Maria Lynch\"\n",
+ "fake_phone = \"7344131647\"\n",
+ "fake_email = \"jamesmichael@example.com\"\n",
+ "fake_credit_card = \"4838637940262\"\n",
+ "\n",
+ "anonymized_text = f\"\"\"{fake_name} recently lost his wallet. \n",
+ "Inside is some cash and his credit card with the number {fake_credit_card}. \n",
+ "If you would find it, please call at {fake_phone} or write an email here: {fake_email}.\n",
+ "{fake_name} would be very grateful!\"\"\"\n",
+ "\n",
+ "print(anonymized_text)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "And now, using the `deanonymize` method, we can reverse the process:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Slim Shady recently lost his wallet. \n",
+ "Inside is some cash and his credit card with the number 4916 0387 9536 0861. \n",
+ "If you would find it, please call at 313-666-7440 or write an email here: real.slim.shady@gmail.com.\n",
+ "Slim Shady would be very grateful!\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(anonymizer.deanonymize(anonymized_text))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Using with LangChain Expression Language\n",
+ "\n",
+ "With LCEL we can easily chain together anonymization and deanonymization with the rest of our application. This is an example of using the anonymization mechanism with a query to LLM (without deanonymization for now):"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "text = f\"\"\"Slim Shady recently lost his wallet. \n",
+ "Inside is some cash and his credit card with the number 4916 0387 9536 0861. \n",
+ "If you would find it, please call at 313-666-7440 or write an email here: real.slim.shady@gmail.com.\"\"\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Dear Sir/Madam,\n",
+ "\n",
+ "We regret to inform you that Mr. Dana Rhodes has reported the loss of his wallet. The wallet contains a sum of cash and his credit card, bearing the number 4397528473885757. \n",
+ "\n",
+ "If you happen to come across the aforementioned wallet, we kindly request that you contact us immediately at 258-481-7074x714 or via email at laurengoodman@example.com.\n",
+ "\n",
+ "Your prompt assistance in this matter would be greatly appreciated.\n",
+ "\n",
+ "Yours faithfully,\n",
+ "\n",
+ "[Your Name]\n"
+ ]
+ }
+ ],
+ "source": [
+ "from langchain.prompts.prompt import PromptTemplate\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "\n",
+ "anonymizer = PresidioReversibleAnonymizer()\n",
+ "\n",
+ "template = \"\"\"Rewrite this text into an official, short email:\n",
+ "\n",
+ "{anonymized_text}\"\"\"\n",
+ "prompt = PromptTemplate.from_template(template)\n",
+ "llm = ChatOpenAI(temperature=0)\n",
+ "\n",
+ "chain = {\"anonymized_text\": anonymizer.anonymize} | prompt | llm\n",
+ "response = chain.invoke(text)\n",
+ "print(response.content)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now, let's add **deanonymization step** to our sequence:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Dear Sir/Madam,\n",
+ "\n",
+ "We regret to inform you that Mr. Slim Shady has recently misplaced his wallet. The wallet contains a sum of cash and his credit card, bearing the number 4916 0387 9536 0861. \n",
+ "\n",
+ "If by any chance you come across the lost wallet, kindly contact us immediately at 313-666-7440 or send an email to real.slim.shady@gmail.com.\n",
+ "\n",
+ "Your prompt assistance in this matter would be greatly appreciated.\n",
+ "\n",
+ "Yours faithfully,\n",
+ "\n",
+ "[Your Name]\n"
+ ]
+ }
+ ],
+ "source": [
+ "chain = chain | (lambda ai_message: anonymizer.deanonymize(ai_message.content))\n",
+ "response = chain.invoke(text)\n",
+ "print(response)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Anonymized data was given to the model itself, and therefore it was protected from being leaked to the outside world. Then, the model's response was processed, and the factual value was replaced with the real one."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Extra knowledge"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "`PresidioReversibleAnonymizer` stores the mapping of the fake values to the original values in the `deanonymizer_mapping` parameter, where key is fake PII and value is the original one: "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'PERSON': {'Maria Lynch': 'Slim Shady'},\n",
+ " 'PHONE_NUMBER': {'7344131647': '313-666-7440'},\n",
+ " 'EMAIL_ADDRESS': {'jamesmichael@example.com': 'real.slim.shady@gmail.com'},\n",
+ " 'CREDIT_CARD': {'4838637940262': '4916 0387 9536 0861'}}"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer\n",
+ "\n",
+ "anonymizer = PresidioReversibleAnonymizer(\n",
+ " analyzed_fields=[\"PERSON\", \"PHONE_NUMBER\", \"EMAIL_ADDRESS\", \"CREDIT_CARD\"],\n",
+ " # Faker seed is used here to make sure the same fake data is generated for the test purposes\n",
+ " # In production, it is recommended to remove the faker_seed parameter (it will default to None)\n",
+ " faker_seed=42,\n",
+ ")\n",
+ "\n",
+ "anonymizer.anonymize(\n",
+ " \"My name is Slim Shady, call me at 313-666-7440 or email me at real.slim.shady@gmail.com. \"\n",
+ " \"By the way, my card number is: 4916 0387 9536 0861\"\n",
+ ")\n",
+ "\n",
+ "anonymizer.deanonymizer_mapping"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Anonymizing more texts will result in new mapping entries:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Do you have his VISA card number? Yep, it's 3537672423884966. I'm William Bowman by the way.\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "{'PERSON': {'Maria Lynch': 'Slim Shady', 'William Bowman': 'John Doe'},\n",
+ " 'PHONE_NUMBER': {'7344131647': '313-666-7440'},\n",
+ " 'EMAIL_ADDRESS': {'jamesmichael@example.com': 'real.slim.shady@gmail.com'},\n",
+ " 'CREDIT_CARD': {'4838637940262': '4916 0387 9536 0861',\n",
+ " '3537672423884966': '4001 9192 5753 7193'}}"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "print(\n",
+ " anonymizer.anonymize(\n",
+ " \"Do you have his VISA card number? Yep, it's 4001 9192 5753 7193. I'm John Doe by the way.\"\n",
+ " )\n",
+ ")\n",
+ "\n",
+ "anonymizer.deanonymizer_mapping"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can save the mapping itself to a file for future use: "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# We can save the deanonymizer mapping as a JSON or YAML file\n",
+ "\n",
+ "anonymizer.save_deanonymizer_mapping(\"deanonymizer_mapping.json\")\n",
+ "# anonymizer.save_deanonymizer_mapping(\"deanonymizer_mapping.yaml\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "And then, load it in another `PresidioReversibleAnonymizer` instance:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{}"
+ ]
+ },
+ "execution_count": 11,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "anonymizer = PresidioReversibleAnonymizer()\n",
+ "\n",
+ "anonymizer.deanonymizer_mapping"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "{'PERSON': {'Maria Lynch': 'Slim Shady', 'William Bowman': 'John Doe'},\n",
+ " 'PHONE_NUMBER': {'7344131647': '313-666-7440'},\n",
+ " 'EMAIL_ADDRESS': {'jamesmichael@example.com': 'real.slim.shady@gmail.com'},\n",
+ " 'CREDIT_CARD': {'4838637940262': '4916 0387 9536 0861',\n",
+ " '3537672423884966': '4001 9192 5753 7193'}}"
+ ]
+ },
+ "execution_count": 12,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "anonymizer.load_deanonymizer_mapping(\"deanonymizer_mapping.json\")\n",
+ "\n",
+ "anonymizer.deanonymizer_mapping"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Future works\n",
+ "\n",
+ "- **instance anonymization** - at this point, each occurrence of PII is treated as a separate entity and separately anonymized. Therefore, two occurrences of the name John Doe in the text will be changed to two different names. It is therefore worth introducing support for full instance detection, so that repeated occurrences are treated as a single object.\n",
+ "- **better matching and substitution of fake values for real ones** - currently the strategy is based on matching full strings and then substituting them. Due to the indeterminism of language models, it may happen that the value in the answer is slightly changed (e.g. *John Doe* -> *John* or *Main St, New York* -> *New York*) and such a substitution is then no longer possible. Therefore, it is worth adjusting the matching for your needs."
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/docs/extras/guides/safety/amazon_comprehend_chain.ipynb b/docs/extras/guides/safety/amazon_comprehend_chain.ipynb
index e7e1961d42..69117b8257 100644
--- a/docs/extras/guides/safety/amazon_comprehend_chain.ipynb
+++ b/docs/extras/guides/safety/amazon_comprehend_chain.ipynb
@@ -512,9 +512,9 @@
"# Examples\n",
"---\n",
"\n",
- "## With HuggingFace Hub Models\n",
+ "## With Hugging Face Hub Models\n",
"\n",
- "Get your API Key from Huggingface hub - https://huggingface.co/docs/api-inference/quicktour#get-your-api-token"
+ "Get your API Key from Hugging Face hub - https://huggingface.co/docs/api-inference/quicktour#get-your-api-token"
]
},
{
diff --git a/docs/extras/integrations/document_transformers/nuclia_transformer.ipynb b/docs/extras/integrations/document_transformers/nuclia_transformer.ipynb
index d4317c9bba..468be00d94 100644
--- a/docs/extras/integrations/document_transformers/nuclia_transformer.ipynb
+++ b/docs/extras/integrations/document_transformers/nuclia_transformer.ipynb
@@ -18,7 +18,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
@@ -93,8 +93,22 @@
}
],
"metadata": {
+ "kernelspec": {
+ "display_name": "langchain",
+ "language": "python",
+ "name": "python3"
+ },
"language_info": {
- "name": "python"
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.5"
},
"orig_nbformat": 4
},
diff --git a/docs/extras/integrations/llms/banana.ipynb b/docs/extras/integrations/llms/banana.ipynb
index 44e51faafa..b92db8daba 100644
--- a/docs/extras/integrations/llms/banana.ipynb
+++ b/docs/extras/integrations/llms/banana.ipynb
@@ -31,11 +31,16 @@
"outputs": [],
"source": [
"# get new tokens: https://app.banana.dev/\n",
- "# We need two tokens, not just an `api_key`: `BANANA_API_KEY` and `YOUR_MODEL_KEY`\n",
+ "# We need three parameters to make a Banana.dev API call:\n",
+ "# * a team api key\n",
+ "# * the model's unique key\n",
+ "# * the model's url slug\n",
"\n",
"import os\n",
"from getpass import getpass\n",
"\n",
+ "# You can get this from the main dashboard\n",
+ "# at https://app.banana.dev\n",
"os.environ[\"BANANA_API_KEY\"] = \"YOUR_API_KEY\"\n",
"# OR\n",
"# BANANA_API_KEY = getpass()"
@@ -70,7 +75,9 @@
"metadata": {},
"outputs": [],
"source": [
- "llm = Banana(model_key=\"YOUR_MODEL_KEY\")"
+ "# Both of these are found in your model's \n",
+ "# detail page in https://app.banana.dev\n",
+ "llm = Banana(model_key=\"YOUR_MODEL_KEY\", model_url_slug=\"YOUR_MODEL_URL_SLUG\")"
]
},
{
diff --git a/docs/extras/integrations/providers/awadb.md b/docs/extras/integrations/providers/awadb.md
index 7c2e9943f5..be6d4d66fe 100644
--- a/docs/extras/integrations/providers/awadb.md
+++ b/docs/extras/integrations/providers/awadb.md
@@ -9,13 +9,20 @@ pip install awadb
```
-## VectorStore
+## Vector Store
-There exists a wrapper around AwaDB vector databases, allowing you to use it as a vectorstore,
-whether for semantic search or example selection.
```python
from langchain.vectorstores import AwaDB
```
-For a more detailed walkthrough of the AwaDB wrapper, see [here](/docs/integrations/vectorstores/awadb.html).
+See a [usage example](/docs/integrations/vectorstores/awadb).
+
+
+## Text Embedding Model
+
+```python
+from langchain.embeddings import AwaEmbeddings
+```
+
+See a [usage example](/docs/integrations/text_embedding/awadb).
diff --git a/docs/extras/integrations/providers/bananadev.mdx b/docs/extras/integrations/providers/bananadev.mdx
index 4961e5f88b..ee7992be74 100644
--- a/docs/extras/integrations/providers/bananadev.mdx
+++ b/docs/extras/integrations/providers/bananadev.mdx
@@ -1,79 +1,72 @@
# Banana
-This page covers how to use the Banana ecosystem within LangChain.
-It is broken into two parts: installation and setup, and then references to specific Banana wrappers.
+Banana provided serverless GPU inference for AI models, including a CI/CD build pipeline and a simple Python framework (Potassium) to server your models.
+
+This page covers how to use the [Banana](https://www.banana.dev) ecosystem within LangChain.
+
+It is broken into two parts:
+* installation and setup,
+* and then references to specific Banana wrappers.
## Installation and Setup
- Install with `pip install banana-dev`
-- Get an Banana api key and set it as an environment variable (`BANANA_API_KEY`)
+- Get an Banana api key from the [Banana.dev dashboard](https://app.banana.dev) and set it as an environment variable (`BANANA_API_KEY`)
+- Get your model's key and url slug from the model's details page
## Define your Banana Template
-If you want to use an available language model template you can find one [here](https://app.banana.dev/templates/conceptofmind/serverless-template-palmyra-base).
-This template uses the Palmyra-Base model by [Writer](https://writer.com/product/api/).
-You can check out an example Banana repository [here](https://github.com/conceptofmind/serverless-template-palmyra-base).
+You'll need to set up a Github repo for your Banana app. You can get started in 5 minutes using [this guide](https://docs.banana.dev/banana-docs/).
+
+Alternatively, for a ready-to-go LLM example, you can check out Banana's [CodeLlama-7B-Instruct-GPTQ](https://github.com/bananaml/demo-codellama-7b-instruct-gptq) GitHub repository. Just fork it and deploy it within Banana.
+
+Other starter repos are available [here](https://github.com/orgs/bananaml/repositories?q=demo-&type=all&language=&sort=).
## Build the Banana app
-Banana Apps must include the "output" key in the return json.
-There is a rigid response structure.
+To use Banana apps within Langchain, they must include the `outputs` key
+in the returned json, and the value must be a string.
```python
# Return the results as a dictionary
-result = {'output': result}
+result = {'outputs': result}
```
An example inference function would be:
```python
-def inference(model_inputs:dict) -> dict:
- global model
- global tokenizer
-
- # Parse out your arguments
- prompt = model_inputs.get('prompt', None)
- if prompt == None:
- return {'message': "No prompt provided"}
-
- # Run the model
- input_ids = tokenizer.encode(prompt, return_tensors='pt').cuda()
- output = model.generate(
- input_ids,
- max_length=100,
- do_sample=True,
- top_k=50,
- top_p=0.95,
- num_return_sequences=1,
- temperature=0.9,
- early_stopping=True,
- no_repeat_ngram_size=3,
- num_beams=5,
- length_penalty=1.5,
- repetition_penalty=1.5,
- bad_words_ids=[[tokenizer.encode(' ', add_prefix_space=True)[0]]]
- )
-
- result = tokenizer.decode(output[0], skip_special_tokens=True)
- # Return the results as a dictionary
- result = {'output': result}
- return result
+@app.handler("/")
+def handler(context: dict, request: Request) -> Response:
+ """Handle a request to generate code from a prompt."""
+ model = context.get("model")
+ tokenizer = context.get("tokenizer")
+ max_new_tokens = request.json.get("max_new_tokens", 512)
+ temperature = request.json.get("temperature", 0.7)
+ prompt = request.json.get("prompt")
+ prompt_template=f'''[INST] Write code to solve the following coding problem that obeys the constraints and passes the example test cases. Please wrap your code answer using ```:
+ {prompt}
+ [/INST]
+ '''
+ input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
+ output = model.generate(inputs=input_ids, temperature=temperature, max_new_tokens=max_new_tokens)
+ result = tokenizer.decode(output[0])
+ return Response(json={"outputs": result}, status=200)
```
-You can find a full example of a Banana app [here](https://github.com/conceptofmind/serverless-template-palmyra-base/blob/main/app.py).
+This example is from the `app.py` file in [CodeLlama-7B-Instruct-GPTQ](https://github.com/bananaml/demo-codellama-7b-instruct-gptq).
## Wrappers
### LLM
-There exists an Banana LLM wrapper, which you can access with
+Within Langchain, there exists a Banana LLM wrapper, which you can access with
```python
from langchain.llms import Banana
```
-You need to provide a model key located in the dashboard:
+You need to provide a model key and model url slug, which you can get from the model's details page in the [Banana.dev dashboard](https://app.banana.dev).
```python
-llm = Banana(model_key="YOUR_MODEL_KEY")
+llm = Banana(model_key="YOUR_MODEL_KEY", model_url_slug="YOUR_MODEL_URL_SLUG")
```
diff --git a/docs/extras/integrations/providers/modelscope.mdx b/docs/extras/integrations/providers/modelscope.mdx
index c37c5f60c4..df6add2bb1 100644
--- a/docs/extras/integrations/providers/modelscope.mdx
+++ b/docs/extras/integrations/providers/modelscope.mdx
@@ -1,20 +1,24 @@
# ModelScope
+>[ModelScope](https://www.modelscope.cn/home) is a big repository of the models and datasets.
+
This page covers how to use the modelscope ecosystem within LangChain.
It is broken into two parts: installation and setup, and then references to specific modelscope wrappers.
## Installation and Setup
-* Install the Python SDK with `pip install modelscope`
+Install the `modelscope` package.
+
+```bash
+pip install modelscope
+```
-## Wrappers
-### Embeddings
+## Text Embedding Models
-There exists a modelscope Embeddings wrapper, which you can access with
```python
from langchain.embeddings import ModelScopeEmbeddings
```
-For a more detailed walkthrough of this, see [this notebook](/docs/integrations/text_embedding/modelscope_hub.html)
+For a more detailed walkthrough of this, see [this notebook](/docs/integrations/text_embedding/modelscope_hub)
diff --git a/docs/extras/integrations/providers/nlpcloud.mdx b/docs/extras/integrations/providers/nlpcloud.mdx
index 050da5af04..e401faeb5a 100644
--- a/docs/extras/integrations/providers/nlpcloud.mdx
+++ b/docs/extras/integrations/providers/nlpcloud.mdx
@@ -1,17 +1,31 @@
# NLPCloud
-This page covers how to use the NLPCloud ecosystem within LangChain.
-It is broken into two parts: installation and setup, and then references to specific NLPCloud wrappers.
+>[NLP Cloud](https://docs.nlpcloud.com/#introduction) is an artificial intelligence platform that allows you to use the most advanced AI engines, and even train your own engines with your own data.
+
## Installation and Setup
-- Install the Python SDK with `pip install nlpcloud`
+
+- Install the `nlpcloud` package.
+
+```bash
+pip install nlpcloud
+```
+
- Get an NLPCloud api key and set it as an environment variable (`NLPCLOUD_API_KEY`)
-## Wrappers
-### LLM
+## LLM
+
+See a [usage example](/docs/integrations/llms/nlpcloud).
-There exists an NLPCloud LLM wrapper, which you can access with
```python
from langchain.llms import NLPCloud
```
+
+## Text Embedding Models
+
+See a [usage example](/docs/integrations/text_embedding/nlp_cloud)
+
+```python
+from langchain.embeddings import NLPCloudEmbeddings
+```
diff --git a/docs/extras/integrations/providers/portkey/index.md b/docs/extras/integrations/providers/portkey/index.md
index 51a9962386..daefe35a22 100644
--- a/docs/extras/integrations/providers/portkey/index.md
+++ b/docs/extras/integrations/providers/portkey/index.md
@@ -1,4 +1,10 @@
# Portkey
+
+>[Portkey](https://docs.portkey.ai/overview/introduction) is a platform designed to streamline the deployment
+> and management of Generative AI applications.
+> It provides comprehensive features for monitoring, managing models,
+> and improving the performance of your AI applications.
+
## LLMOps for Langchain
Portkey brings production readiness to Langchain. With Portkey, you can
diff --git a/docs/extras/integrations/providers/portkey/logging_tracing_portkey.ipynb b/docs/extras/integrations/providers/portkey/logging_tracing_portkey.ipynb
index e26fabd659..7fd2cd4161 100644
--- a/docs/extras/integrations/providers/portkey/logging_tracing_portkey.ipynb
+++ b/docs/extras/integrations/providers/portkey/logging_tracing_portkey.ipynb
@@ -1,19 +1,14 @@
{
"cells": [
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": []
- },
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Log, Trace, and Monitor Langchain LLM Calls\n",
+ "# Log, Trace, and Monitor\n",
"\n",
"When building apps or agents using Langchain, you end up making multiple API calls to fulfill a single user request. However, these requests are not chained when you want to analyse them. With [**Portkey**](/docs/ecosystem/integrations/portkey), all the embeddings, completion, and other requests from a single user request will get logged and traced to a common ID, enabling you to gain full visibility of user interactions.\n",
"\n",
- "This notebook serves as a step-by-step guide on how to integrate and use Portkey in your Langchain app."
+ "This notebook serves as a step-by-step guide on how to log, trace, and monitor Langchain LLM calls using `Portkey` in your Langchain app."
]
},
{
@@ -234,9 +229,9 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.11.3"
+ "version": "3.10.12"
}
},
"nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
}
diff --git a/docs/extras/integrations/providers/spacy.mdx b/docs/extras/integrations/providers/spacy.mdx
index f4d49497dd..ab9b685898 100644
--- a/docs/extras/integrations/providers/spacy.mdx
+++ b/docs/extras/integrations/providers/spacy.mdx
@@ -18,3 +18,11 @@ See a [usage example](/docs/modules/data_connection/document_transformers/text_s
```python
from langchain.text_splitter import SpacyTextSplitter
```
+
+## Text Embedding Models
+
+See a [usage example](/docs/integrations/text_embedding/spacy_embedding)
+
+```python
+from langchain.embeddings.spacy_embeddings import SpacyEmbeddings
+```
diff --git a/docs/extras/integrations/providers/vectara/index.mdx b/docs/extras/integrations/providers/vectara/index.mdx
index 0e30735f0b..abd8283735 100644
--- a/docs/extras/integrations/providers/vectara/index.mdx
+++ b/docs/extras/integrations/providers/vectara/index.mdx
@@ -11,9 +11,10 @@ What is Vectara?
- You can use Vectara's integration with LangChain as a Vector store or using the Retriever abstraction.
## Installation and Setup
-To use Vectara with LangChain no special installation steps are required. You just have to provide your customer_id, corpus ID, and an API key created within the Vectara console to enable indexing and searching.
+To use Vectara with LangChain no special installation steps are required.
+To get started, follow our [quickstart](https://docs.vectara.com/docs/quickstart) guide to create an account, a corpus and an API key.
+Once you have these, you can provide them as arguments to the Vectara vectorstore, or you can set them as environment variables.
-Alternatively these can be provided as environment variables
- export `VECTARA_CUSTOMER_ID`="your_customer_id"
- export `VECTARA_CORPUS_ID`="your_corpus_id"
- export `VECTARA_API_KEY`="your-vectara-api-key"
diff --git a/docs/extras/integrations/text_embedding/Awa.ipynb b/docs/extras/integrations/text_embedding/awadb.ipynb
similarity index 89%
rename from docs/extras/integrations/text_embedding/Awa.ipynb
rename to docs/extras/integrations/text_embedding/awadb.ipynb
index 1fb7ddca6f..f2c1e73392 100644
--- a/docs/extras/integrations/text_embedding/Awa.ipynb
+++ b/docs/extras/integrations/text_embedding/awadb.ipynb
@@ -5,9 +5,11 @@
"id": "b14a24db",
"metadata": {},
"source": [
- "# AwaEmbedding\n",
+ "# AwaDB\n",
"\n",
- "This notebook explains how to use AwaEmbedding, which is included in [awadb](https://github.com/awa-ai/awadb), to embedding texts in langchain."
+ ">[AwaDB](https://github.com/awa-ai/awadb) is an AI Native database for the search and storage of embedding vectors used by LLM Applications.\n",
+ "\n",
+ "This notebook explains how to use `AwaEmbeddings` in LangChain."
]
},
{
@@ -101,7 +103,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.11.4"
+ "version": "3.10.12"
}
},
"nbformat": 4,
diff --git a/docs/extras/integrations/text_embedding/bedrock.ipynb b/docs/extras/integrations/text_embedding/bedrock.ipynb
index 7c16cb8ead..0dbbcd080f 100644
--- a/docs/extras/integrations/text_embedding/bedrock.ipynb
+++ b/docs/extras/integrations/text_embedding/bedrock.ipynb
@@ -5,7 +5,9 @@
"id": "75e378f5-55d7-44b6-8e2e-6d7b8b171ec4",
"metadata": {},
"source": [
- "# Bedrock Embeddings"
+ "# Bedrock\n",
+ "\n",
+ ">[Amazon Bedrock](https://aws.amazon.com/bedrock/) is a fully managed service that makes FMs from leading AI startups and Amazon available via an API, so you can choose from a wide range of FMs to find the model that is best suited for your use case.\n"
]
},
{
@@ -91,7 +93,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.13"
+ "version": "3.10.12"
}
},
"nbformat": 4,
diff --git a/docs/extras/integrations/text_embedding/bge_huggingface.ipynb b/docs/extras/integrations/text_embedding/bge_huggingface.ipynb
index bcf196fc20..923ba92874 100644
--- a/docs/extras/integrations/text_embedding/bge_huggingface.ipynb
+++ b/docs/extras/integrations/text_embedding/bge_huggingface.ipynb
@@ -5,26 +5,29 @@
"id": "719619d3",
"metadata": {},
"source": [
- "# BGE Hugging Face Embeddings\n",
+ "# BGE on Hugging Face\n",
"\n",
- "This notebook shows how to use BGE Embeddings through Hugging Face"
+ ">[BGE models on the HuggingFace](https://huggingface.co/BAAI/bge-large-en) are [the best open-source embedding models](https://huggingface.co/spaces/mteb/leaderboard).\n",
+ ">BGE model is created by the [Beijing Academy of Artificial Intelligence (BAAI)](https://www.baai.ac.cn/english.html). `BAAI` is a private non-profit organization engaged in AI research and development.\n",
+ "\n",
+ "This notebook shows how to use `BGE Embeddings` through `Hugging Face`"
]
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": null,
"id": "f7a54279",
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
- "# !pip install sentence_transformers"
+ "#!pip install sentence_transformers"
]
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": null,
"id": "9e1d5b6b",
"metadata": {},
"outputs": [],
@@ -43,12 +46,24 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 5,
"id": "e59d1a89",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "384"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
- "embedding = hf.embed_query(\"hi this is harrison\")"
+ "embedding = hf.embed_query(\"hi this is harrison\")\n",
+ "len(embedding)"
]
},
{
@@ -76,7 +91,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.10.1"
+ "version": "3.10.12"
}
},
"nbformat": 4,
diff --git a/docs/extras/integrations/text_embedding/google_vertex_ai_palm.ipynb b/docs/extras/integrations/text_embedding/google_vertex_ai_palm.ipynb
index ea607467fb..4c0c515e80 100644
--- a/docs/extras/integrations/text_embedding/google_vertex_ai_palm.ipynb
+++ b/docs/extras/integrations/text_embedding/google_vertex_ai_palm.ipynb
@@ -1,13 +1,14 @@
{
"cells": [
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Google Cloud Platform Vertex AI PaLM \n",
+ "# Google Vertex AI PaLM \n",
"\n",
- "Note: This is seperate from the Google PaLM integration, it exposes [Vertex AI PaLM API](https://cloud.google.com/vertex-ai/docs/generative-ai/learn/overview) on Google Cloud. \n",
+ ">[Vertex AI PaLM API](https://cloud.google.com/vertex-ai/docs/generative-ai/learn/overview) is a service on Google Cloud exposing the embedding models. \n",
+ "\n",
+ "Note: This integration is seperate from the Google PaLM integration.\n",
"\n",
"By default, Google Cloud [does not use](https://cloud.google.com/vertex-ai/docs/generative-ai/data-governance#foundation_model_development) Customer Data to train its foundation models as part of Google Cloud`s AI/ML Privacy Commitment. More details about how Google processes data can also be found in [Google's Customer Data Processing Addendum (CDPA)](https://cloud.google.com/terms/data-processing-addendum).\n",
"\n",
@@ -96,7 +97,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.1"
+ "version": "3.10.12"
},
"vscode": {
"interpreter": {
diff --git a/docs/extras/integrations/text_embedding/huggingfacehub.ipynb b/docs/extras/integrations/text_embedding/huggingfacehub.ipynb
index a86df86d74..cb897f8693 100644
--- a/docs/extras/integrations/text_embedding/huggingfacehub.ipynb
+++ b/docs/extras/integrations/text_embedding/huggingfacehub.ipynb
@@ -5,13 +5,23 @@
"id": "ed47bb62",
"metadata": {},
"source": [
- "# Hugging Face Hub\n",
+ "# Hugging Face\n",
"Let's load the Hugging Face Embedding class."
]
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": null,
+ "id": "16b20335-da1d-46ba-aa23-fbf3e2c6fe60",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip install langchain sentence_transformers"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
"id": "861521a9",
"metadata": {},
"outputs": [],
@@ -21,7 +31,7 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 3,
"id": "ff9be586",
"metadata": {},
"outputs": [],
@@ -31,7 +41,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 3,
"id": "d0a98ae9",
"metadata": {},
"outputs": [],
@@ -41,7 +51,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 5,
"id": "5d6c682b",
"metadata": {},
"outputs": [],
@@ -51,7 +61,28 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 6,
+ "id": "b57b8ce9-ef7d-4e63-979e-aa8763d1f9a8",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[-0.04895168915390968, -0.03986193612217903, -0.021562768146395683]"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "query_result[:3]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
"id": "bb5e74c0",
"metadata": {},
"outputs": [],
@@ -59,20 +90,72 @@
"doc_result = embeddings.embed_documents([text])"
]
},
+ {
+ "cell_type": "markdown",
+ "id": "92019ef1-5d30-4985-b4e6-c0d98bdfe265",
+ "metadata": {},
+ "source": [
+ "## Hugging Face Inference API\n",
+ "We can also access embedding models via the Hugging Face Inference API, which does not require us to install ``sentence_transformers`` and download models locally."
+ ]
+ },
{
"cell_type": "code",
- "execution_count": null,
- "id": "aaad49f8",
+ "execution_count": 1,
+ "id": "66f5c6ba-1446-43e1-b012-800d17cef300",
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter your HF Inference API Key:\n",
+ "\n",
+ " ········\n"
+ ]
+ }
+ ],
+ "source": [
+ "import getpass\n",
+ "\n",
+ "inference_api_key = getpass.getpass(\"Enter your HF Inference API Key:\\n\\n\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "d0623c1f-cd82-4862-9bce-3655cb9b66ac",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[-0.038338541984558105, 0.1234646737575531, -0.028642963618040085]"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from langchain.embeddings import HuggingFaceInferenceAPIEmbeddings\n",
+ "\n",
+ "embeddings = HuggingFaceInferenceAPIEmbeddings(\n",
+ " api_key=inference_api_key,\n",
+ " model_name=\"sentence-transformers/all-MiniLM-l6-v2\"\n",
+ ")\n",
+ "\n",
+ "query_result = embeddings.embed_query(text)\n",
+ "query_result[:3]"
+ ]
}
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3 (ipykernel)",
+ "display_name": "poetry-venv",
"language": "python",
- "name": "python3"
+ "name": "poetry-venv"
},
"language_info": {
"codemirror_mode": {
diff --git a/docs/extras/integrations/text_embedding/modelscope_hub.ipynb b/docs/extras/integrations/text_embedding/modelscope_hub.ipynb
index 765d46769c..e2f47c4f3a 100644
--- a/docs/extras/integrations/text_embedding/modelscope_hub.ipynb
+++ b/docs/extras/integrations/text_embedding/modelscope_hub.ipynb
@@ -1,12 +1,13 @@
{
"cells": [
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"# ModelScope\n",
"\n",
+ ">[ModelScope](https://www.modelscope.cn/home) is big repository of the models and datasets.\n",
+ "\n",
"Let's load the ModelScope Embedding class."
]
},
@@ -67,16 +68,23 @@
],
"metadata": {
"kernelspec": {
- "display_name": "chatgpt",
+ "display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
"name": "python",
- "version": "3.9.15"
- },
- "orig_nbformat": 4
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.12"
+ }
},
"nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
}
diff --git a/docs/extras/integrations/text_embedding/mosaicml.ipynb b/docs/extras/integrations/text_embedding/mosaicml.ipynb
index 2d91c8d9c5..24d7aecb72 100644
--- a/docs/extras/integrations/text_embedding/mosaicml.ipynb
+++ b/docs/extras/integrations/text_embedding/mosaicml.ipynb
@@ -1,15 +1,14 @@
{
"cells": [
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "# MosaicML embeddings\n",
+ "# MosaicML\n",
"\n",
- "[MosaicML](https://docs.mosaicml.com/en/latest/inference.html) offers a managed inference service. You can either use a variety of open source models, or deploy your own.\n",
+ ">[MosaicML](https://docs.mosaicml.com/en/latest/inference.html) offers a managed inference service. You can either use a variety of open source models, or deploy your own.\n",
"\n",
- "This example goes over how to use LangChain to interact with MosaicML Inference for text embedding."
+ "This example goes over how to use LangChain to interact with `MosaicML` Inference for text embedding."
]
},
{
@@ -94,6 +93,11 @@
}
],
"metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
"language_info": {
"codemirror_mode": {
"name": "ipython",
@@ -103,9 +107,10 @@
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
- "pygments_lexer": "ipython3"
+ "pygments_lexer": "ipython3",
+ "version": "3.10.12"
}
},
"nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
}
diff --git a/docs/extras/integrations/text_embedding/nlp_cloud.ipynb b/docs/extras/integrations/text_embedding/nlp_cloud.ipynb
index 73ae71fe0f..9567d59c4b 100644
--- a/docs/extras/integrations/text_embedding/nlp_cloud.ipynb
+++ b/docs/extras/integrations/text_embedding/nlp_cloud.ipynb
@@ -7,7 +7,7 @@
"source": [
"# NLP Cloud\n",
"\n",
- "NLP Cloud is an artificial intelligence platform that allows you to use the most advanced AI engines, and even train your own engines with your own data. \n",
+ ">[NLP Cloud](https://docs.nlpcloud.com/#introduction) is an artificial intelligence platform that allows you to use the most advanced AI engines, and even train your own engines with your own data. \n",
"\n",
"The [embeddings](https://docs.nlpcloud.com/#embeddings) endpoint offers the following model:\n",
"\n",
@@ -80,7 +80,7 @@
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3.11.2 64-bit",
+ "display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
@@ -94,7 +94,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.11.2"
+ "version": "3.10.12"
},
"vscode": {
"interpreter": {
diff --git a/docs/extras/integrations/text_embedding/sagemaker-endpoint.ipynb b/docs/extras/integrations/text_embedding/sagemaker-endpoint.ipynb
index fe5299ae6f..ec80112e10 100644
--- a/docs/extras/integrations/text_embedding/sagemaker-endpoint.ipynb
+++ b/docs/extras/integrations/text_embedding/sagemaker-endpoint.ipynb
@@ -5,11 +5,13 @@
"id": "1f83f273",
"metadata": {},
"source": [
- "# SageMaker Endpoint Embeddings\n",
+ "# SageMaker\n",
"\n",
- "Let's load the SageMaker Endpoints Embeddings class. The class can be used if you host, e.g. your own Hugging Face model on SageMaker.\n",
+ "Let's load the `SageMaker Endpoints Embeddings` class. The class can be used if you host, e.g. your own Hugging Face model on SageMaker.\n",
"\n",
- "For instructions on how to do this, please see [here](https://www.philschmid.de/custom-inference-huggingface-sagemaker). **Note**: In order to handle batched requests, you will need to adjust the return line in the `predict_fn()` function within the custom `inference.py` script:\n",
+ "For instructions on how to do this, please see [here](https://www.philschmid.de/custom-inference-huggingface-sagemaker). \n",
+ "\n",
+ "**Note**: In order to handle batched requests, you will need to adjust the return line in the `predict_fn()` function within the custom `inference.py` script:\n",
"\n",
"Change from\n",
"\n",
@@ -143,7 +145,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.1"
+ "version": "3.10.12"
},
"vscode": {
"interpreter": {
diff --git a/docs/extras/integrations/text_embedding/self-hosted.ipynb b/docs/extras/integrations/text_embedding/self-hosted.ipynb
index 00c497220e..47faa6bf2d 100644
--- a/docs/extras/integrations/text_embedding/self-hosted.ipynb
+++ b/docs/extras/integrations/text_embedding/self-hosted.ipynb
@@ -5,8 +5,8 @@
"id": "eec4efda",
"metadata": {},
"source": [
- "# Self Hosted Embeddings\n",
- "Let's load the SelfHostedEmbeddings, SelfHostedHuggingFaceEmbeddings, and SelfHostedHuggingFaceInstructEmbeddings classes."
+ "# Self Hosted\n",
+ "Let's load the `SelfHostedEmbeddings`, `SelfHostedHuggingFaceEmbeddings`, and `SelfHostedHuggingFaceInstructEmbeddings` classes."
]
},
{
@@ -149,9 +149,7 @@
"cell_type": "code",
"execution_count": null,
"id": "fc1bfd0f",
- "metadata": {
- "scrolled": false
- },
+ "metadata": {},
"outputs": [],
"source": [
"query_result = embeddings.embed_query(text)"
@@ -182,7 +180,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.1"
+ "version": "3.10.12"
},
"vscode": {
"interpreter": {
diff --git a/docs/extras/integrations/text_embedding/sentence_transformers.ipynb b/docs/extras/integrations/text_embedding/sentence_transformers.ipynb
index 67eb83ab7c..e4649e6b71 100644
--- a/docs/extras/integrations/text_embedding/sentence_transformers.ipynb
+++ b/docs/extras/integrations/text_embedding/sentence_transformers.ipynb
@@ -1,16 +1,15 @@
{
"cells": [
{
- "attachments": {},
"cell_type": "markdown",
"id": "ed47bb62",
"metadata": {},
"source": [
- "# Sentence Transformers Embeddings\n",
+ "# Sentence Transformers\n",
"\n",
- "[SentenceTransformers](https://www.sbert.net/) embeddings are called using the `HuggingFaceEmbeddings` integration. We have also added an alias for `SentenceTransformerEmbeddings` for users who are more familiar with directly using that package.\n",
+ ">[SentenceTransformers](https://www.sbert.net/) embeddings are called using the `HuggingFaceEmbeddings` integration. We have also added an alias for `SentenceTransformerEmbeddings` for users who are more familiar with directly using that package.\n",
"\n",
- "SentenceTransformers is a python package that can generate text and image embeddings, originating from [Sentence-BERT](https://arxiv.org/abs/1908.10084)"
+ "`SentenceTransformers` is a python package that can generate text and image embeddings, originating from [Sentence-BERT](https://arxiv.org/abs/1908.10084)"
]
},
{
@@ -109,7 +108,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.8.16"
+ "version": "3.10.12"
},
"vscode": {
"interpreter": {
diff --git a/docs/extras/integrations/text_embedding/spacy_embedding.ipynb b/docs/extras/integrations/text_embedding/spacy_embedding.ipynb
index bfea82d5d4..edda4828b4 100644
--- a/docs/extras/integrations/text_embedding/spacy_embedding.ipynb
+++ b/docs/extras/integrations/text_embedding/spacy_embedding.ipynb
@@ -1,21 +1,31 @@
{
"cells": [
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Spacy Embedding\n",
+ "# SpaCy\n",
"\n",
- "### Loading the Spacy embedding class to generate and query embeddings"
+ ">[spaCy](https://spacy.io/) is an open-source software library for advanced natural language processing, written in the programming languages Python and Cython.\n",
+ " \n",
+ "\n",
+ "## Installation and Setup"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#!pip install spacy"
]
},
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "#### Import the necessary classes"
+ "Import the necessary classes"
]
},
{
@@ -28,11 +38,12 @@
]
},
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "#### Initialize SpacyEmbeddings.This will load the Spacy model into memory."
+ "## Example\n",
+ "\n",
+ "Initialize SpacyEmbeddings.This will load the Spacy model into memory."
]
},
{
@@ -45,11 +56,10 @@
]
},
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "#### Define some example texts . These could be any documents that you want to analyze - for example, news articles, social media posts, or product reviews."
+ "Define some example texts . These could be any documents that you want to analyze - for example, news articles, social media posts, or product reviews."
]
},
{
@@ -67,11 +77,10 @@
]
},
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "#### Generate and print embeddings for the texts . The SpacyEmbeddings class generates an embedding for each document, which is a numerical representation of the document's content. These embeddings can be used for various natural language processing tasks, such as document similarity comparison or text classification."
+ "Generate and print embeddings for the texts . The SpacyEmbeddings class generates an embedding for each document, which is a numerical representation of the document's content. These embeddings can be used for various natural language processing tasks, such as document similarity comparison or text classification."
]
},
{
@@ -86,11 +95,10 @@
]
},
{
- "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "#### Generate and print an embedding for a single piece of text. You can also generate an embedding for a single piece of text, such as a search query. This can be useful for tasks like information retrieval, where you want to find documents that are similar to a given query."
+ "Generate and print an embedding for a single piece of text. You can also generate an embedding for a single piece of text, such as a search query. This can be useful for tasks like information retrieval, where you want to find documents that are similar to a given query."
]
},
{
@@ -106,11 +114,24 @@
}
],
"metadata": {
- "language_info": {
- "name": "python"
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
},
- "orig_nbformat": 4
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.12"
+ }
},
"nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
}
diff --git a/docs/extras/integrations/vectorstores/nucliadb.ipynb b/docs/extras/integrations/vectorstores/nucliadb.ipynb
new file mode 100644
index 0000000000..f5fe2299c8
--- /dev/null
+++ b/docs/extras/integrations/vectorstores/nucliadb.ipynb
@@ -0,0 +1,126 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# NucliaDB\n",
+ "\n",
+ "You can use a local NucliaDB instance or use [Nuclia Cloud](https://nuclia.cloud).\n",
+ "\n",
+ "When using a local instance, you need a Nuclia Understanding API key, so your texts are properly vectorized and indexed. You can get a key by creating a free account at [https://nuclia.cloud](https://nuclia.cloud), and then [create a NUA key](https://docs.nuclia.dev/docs/docs/using/understanding/intro)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#!pip install langchain nuclia"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Usage with nuclia.cloud"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.vectorstores.nucliadb import NucliaDB\n",
+ "API_KEY = \"YOUR_API_KEY\"\n",
+ "\n",
+ "ndb = NucliaDB(knowledge_box=\"YOUR_KB_ID\", local=False, api_key=API_KEY)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Usage with a local instance\n",
+ "\n",
+ "Note: By default `backend` is set to `http://localhost:8080`."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.vectorstores.nucliadb import NucliaDB\n",
+ "\n",
+ "ndb = NucliaDB(knowledge_box=\"YOUR_KB_ID\", local=True, backend=\"http://my-local-server\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Add and delete texts to your Knowledge Box"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "ids = ndb.add_texts([\"This is a new test\", \"This is a second test\"])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "ndb.delete(ids=ids)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Search in your Knowledge Box"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "results = ndb.similarity_search(\"Who was inspired by Ada Lovelace?\")\n",
+ "print(res.page_content)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/docs/extras/integrations/vectorstores/sqlitevss.ipynb b/docs/extras/integrations/vectorstores/sqlitevss.ipynb
new file mode 100644
index 0000000000..e670d5683f
--- /dev/null
+++ b/docs/extras/integrations/vectorstores/sqlitevss.ipynb
@@ -0,0 +1,207 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# sqlite-vss\n",
+ "\n",
+ ">[sqlite-vss](https://alexgarcia.xyz/sqlite-vss/) is an SQLite extension designed for vector search, emphasizing local-first operations and easy integration into applications without external servers. Leveraging the Faiss library, it offers efficient similarity search and clustering capabilities.\n",
+ "\n",
+ "This notebook shows how to use the `SQLiteVSS` vector database."
+ ],
+ "metadata": {
+ "collapsed": false
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "outputs": [],
+ "source": [
+ "# You need to install sqlite-vss as a dependency.\n",
+ "%pip install sqlite-vss"
+ ],
+ "metadata": {
+ "collapsed": false
+ }
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "### Quickstart"
+ ],
+ "metadata": {
+ "collapsed": false
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "outputs": [
+ {
+ "data": {
+ "text/plain": "'Tonight. I call on the Senate to: Pass the Freedom to Vote Act. Pass the John Lewis Voting Rights Act. And while you’re at it, pass the Disclose Act so Americans can know who is funding our elections. \\n\\nTonight, I’d like to honor someone who has dedicated his life to serve this country: Justice Stephen Breyer—an Army veteran, Constitutional scholar, and retiring Justice of the United States Supreme Court. Justice Breyer, thank you for your service. \\n\\nOne of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. \\n\\nAnd I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence.'"
+ },
+ "execution_count": 2,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from langchain.embeddings.sentence_transformer import SentenceTransformerEmbeddings\n",
+ "from langchain.text_splitter import CharacterTextSplitter\n",
+ "from langchain.vectorstores import SQLiteVSS\n",
+ "from langchain.document_loaders import TextLoader\n",
+ "\n",
+ "# load the document and split it into chunks\n",
+ "loader = TextLoader(\"../../../state_of_the_union.txt\")\n",
+ "documents = loader.load()\n",
+ "\n",
+ "# split it into chunks\n",
+ "text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)\n",
+ "docs = text_splitter.split_documents(documents)\n",
+ "texts = [doc.page_content for doc in docs]\n",
+ "\n",
+ "\n",
+ "# create the open-source embedding function\n",
+ "embedding_function = SentenceTransformerEmbeddings(model_name=\"all-MiniLM-L6-v2\")\n",
+ "\n",
+ "\n",
+ "# load it in sqlite-vss in a table named state_union.\n",
+ "# the db_file parameter is the name of the file you want\n",
+ "# as your sqlite database.\n",
+ "db = SQLiteVSS.from_texts(\n",
+ " texts=texts,\n",
+ " embedding=embedding_function,\n",
+ " table=\"state_union\",\n",
+ " db_file=\"/tmp/vss.db\"\n",
+ ")\n",
+ "\n",
+ "# query it\n",
+ "query = \"What did the president say about Ketanji Brown Jackson\"\n",
+ "data = db.similarity_search(query)\n",
+ "\n",
+ "# print results\n",
+ "data[0].page_content"
+ ],
+ "metadata": {
+ "collapsed": false,
+ "ExecuteTime": {
+ "end_time": "2023-09-06T14:55:55.370351Z",
+ "start_time": "2023-09-06T14:55:53.547755Z"
+ }
+ }
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "### Using existing sqlite connection"
+ ],
+ "metadata": {
+ "collapsed": false
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "outputs": [
+ {
+ "data": {
+ "text/plain": "'Ketanji Brown Jackson is awesome'"
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from langchain.embeddings.sentence_transformer import SentenceTransformerEmbeddings\n",
+ "from langchain.text_splitter import CharacterTextSplitter\n",
+ "from langchain.vectorstores import SQLiteVSS\n",
+ "from langchain.document_loaders import TextLoader\n",
+ "\n",
+ "# load the document and split it into chunks\n",
+ "loader = TextLoader(\"../../../state_of_the_union.txt\")\n",
+ "documents = loader.load()\n",
+ "\n",
+ "# split it into chunks\n",
+ "text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)\n",
+ "docs = text_splitter.split_documents(documents)\n",
+ "texts = [doc.page_content for doc in docs]\n",
+ "\n",
+ "\n",
+ "# create the open-source embedding function\n",
+ "embedding_function = SentenceTransformerEmbeddings(model_name=\"all-MiniLM-L6-v2\")\n",
+ "connection = SQLiteVSS.create_connection(db_file=\"/tmp/vss.db\")\n",
+ "\n",
+ "db1 = SQLiteVSS(\n",
+ " table=\"state_union\",\n",
+ " embedding=embedding_function,\n",
+ " connection=connection\n",
+ ")\n",
+ "\n",
+ "db1.add_texts([\"Ketanji Brown Jackson is awesome\"])\n",
+ "# query it again\n",
+ "query = \"What did the president say about Ketanji Brown Jackson\"\n",
+ "data = db1.similarity_search(query)\n",
+ "\n",
+ "# print results\n",
+ "data[0].page_content"
+ ],
+ "metadata": {
+ "collapsed": false,
+ "ExecuteTime": {
+ "end_time": "2023-09-06T14:59:22.086252Z",
+ "start_time": "2023-09-06T14:59:21.693237Z"
+ }
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "outputs": [],
+ "source": [
+ "# Cleaning up\n",
+ "import os\n",
+ "os.remove(\"/tmp/vss.db\")"
+ ],
+ "metadata": {
+ "collapsed": false,
+ "ExecuteTime": {
+ "end_time": "2023-09-06T15:01:15.550318Z",
+ "start_time": "2023-09-06T15:01:15.546428Z"
+ }
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "outputs": [],
+ "source": [],
+ "metadata": {
+ "collapsed": false
+ }
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 2
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython2",
+ "version": "2.7.6"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}
diff --git a/docs/extras/integrations/vectorstores/supabase.ipynb b/docs/extras/integrations/vectorstores/supabase.ipynb
index 9ba5dacd08..9a5f583adb 100644
--- a/docs/extras/integrations/vectorstores/supabase.ipynb
+++ b/docs/extras/integrations/vectorstores/supabase.ipynb
@@ -28,43 +28,41 @@
"The following function determines cosine similarity, but you can adjust to your needs.\n",
"\n",
"```sql\n",
- " -- Enable the pgvector extension to work with embedding vectors\n",
- " create extension vector;\n",
+ "-- Enable the pgvector extension to work with embedding vectors\n",
+ "create extension if not exists vector;\n",
"\n",
- " -- Create a table to store your documents\n",
- " create table documents (\n",
- " id uuid primary key,\n",
- " content text, -- corresponds to Document.pageContent\n",
- " metadata jsonb, -- corresponds to Document.metadata\n",
- " embedding vector(1536) -- 1536 works for OpenAI embeddings, change if needed\n",
- " );\n",
+ "-- Create a table to store your documents\n",
+ "create table\n",
+ " documents (\n",
+ " id uuid primary key,\n",
+ " content text, -- corresponds to Document.pageContent\n",
+ " metadata jsonb, -- corresponds to Document.metadata\n",
+ " embedding vector (1536) -- 1536 works for OpenAI embeddings, change if needed\n",
+ " );\n",
"\n",
- " CREATE FUNCTION match_documents(query_embedding vector(1536), match_count int)\n",
- " RETURNS TABLE(\n",
- " id uuid,\n",
- " content text,\n",
- " metadata jsonb,\n",
- " -- we return matched vectors to enable maximal marginal relevance searches\n",
- " embedding vector(1536),\n",
- " similarity float)\n",
- " LANGUAGE plpgsql\n",
- " AS $$\n",
- " # variable_conflict use_column\n",
- " BEGIN\n",
- " RETURN query\n",
- " SELECT\n",
- " id,\n",
- " content,\n",
- " metadata,\n",
- " embedding,\n",
- " 1 -(documents.embedding <=> query_embedding) AS similarity\n",
- " FROM\n",
- " documents\n",
- " ORDER BY\n",
- " documents.embedding <=> query_embedding\n",
- " LIMIT match_count;\n",
- " END;\n",
- " $$;\n",
+ "-- Create a function to search for documents\n",
+ "create function match_documents (\n",
+ " query_embedding vector (1536),\n",
+ " filter jsonb default '{}'\n",
+ ") returns table (\n",
+ " id uuid,\n",
+ " content text,\n",
+ " metadata jsonb,\n",
+ " similarity float\n",
+ ") language plpgsql as $$\n",
+ "#variable_conflict use_column\n",
+ "begin\n",
+ " return query\n",
+ " select\n",
+ " id,\n",
+ " content,\n",
+ " metadata,\n",
+ " 1 - (documents.embedding <=> query_embedding) as similarity\n",
+ " from documents\n",
+ " where metadata @> filter\n",
+ " order by documents.embedding <=> query_embedding;\n",
+ "end;\n",
+ "$$;\n",
"```"
]
},
diff --git a/docs/extras/integrations/vectorstores/vectara.ipynb b/docs/extras/integrations/vectorstores/vectara.ipynb
index 0741c1b199..e95504860b 100644
--- a/docs/extras/integrations/vectorstores/vectara.ipynb
+++ b/docs/extras/integrations/vectorstores/vectara.ipynb
@@ -26,7 +26,7 @@
"source": [
"# Setup\n",
"\n",
- "You will need a Vectara account to use Vectara with LangChain. To get started, use the following steps:\n",
+ "You will need a Vectara account to use Vectara with LangChain. To get started, use the following steps (see our [quickstart](https://docs.vectara.com/docs/quickstart) guide):\n",
"1. [Sign up](https://console.vectara.com/signup) for a Vectara account if you don't already have one. Once you have completed your sign up you will have a Vectara customer ID. You can find your customer ID by clicking on your name, on the top-right of the Vectara console window.\n",
"2. Within your account you can create one or more corpora. Each corpus represents an area that stores text data upon ingest from input documents. To create a corpus, use the **\"Create Corpus\"** button. You then provide a name to your corpus as well as a description. Optionally you can define filtering attributes and apply some advanced options. If you click on your created corpus, you can see its name and corpus ID right on the top.\n",
"3. Next you'll need to create API keys to access the corpus. Click on the **\"Authorization\"** tab in the corpus view and then the **\"Create API Key\"** button. Give your key a name, and choose whether you want query only or query+index for your key. Click \"Create\" and you now have an active API key. Keep this key confidential. \n",
@@ -47,7 +47,7 @@
"os.environ[\"VECTARA_API_KEY\"] = getpass.getpass(\"Vectara API Key:\")\n",
"```\n",
"\n",
- "2. Add them to the Vectara vectorstore constructor:\n",
+ "1. Provide them as arguments when creating the Vectara vectorstore object:\n",
"\n",
"```python\n",
"vectorstore = Vectara(\n",
@@ -65,13 +65,22 @@
"source": [
"## Connecting to Vectara from LangChain\n",
"\n",
- "To get started, let's ingest the documents using the from_documents() method.\n",
- "We assume here that you've added your VECTARA_CUSTOMER_ID, VECTARA_CORPUS_ID and query+indexing VECTARA_API_KEY as environment variables."
+ "In this example, we assume that you've created an account and a corpus, and added your VECTARA_CUSTOMER_ID, VECTARA_CORPUS_ID and VECTARA_API_KEY (created with permissions for both indexing and query) as environment variables.\n",
+ "\n",
+ "The corpus has 3 fields defined as metadata for filtering:\n",
+ "* url: a string field containing the source URL of the document (where relevant)\n",
+ "* speech: a string field containing the name of the speech\n",
+ "* author: the name of the author\n",
+ "\n",
+ "Let's start by ingesting 3 documents into the corpus:\n",
+ "1. The State of the Union speech from 2022, available in the LangChain repository as a text file\n",
+ "2. The \"I have a dream\" speech by Dr. Kind\n",
+ "3. The \"We shall Fight on the Beaches\" speech by Winston Churchil"
]
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": 2,
"id": "04a1f1a0",
"metadata": {},
"outputs": [],
@@ -79,12 +88,17 @@
"from langchain.embeddings import FakeEmbeddings\n",
"from langchain.text_splitter import CharacterTextSplitter\n",
"from langchain.vectorstores import Vectara\n",
- "from langchain.document_loaders import TextLoader"
+ "from langchain.document_loaders import TextLoader\n",
+ "\n",
+ "from langchain.llms import OpenAI\n",
+ "from langchain.chains import ConversationalRetrievalChain\n",
+ "from langchain.retrievers.self_query.base import SelfQueryRetriever\n",
+ "from langchain.chains.query_constructor.base import AttributeInfo"
]
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 3,
"id": "be0a4973",
"metadata": {},
"outputs": [],
@@ -97,7 +111,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 4,
"id": "8429667e",
"metadata": {
"ExecuteTime": {
@@ -111,7 +125,7 @@
"vectara = Vectara.from_documents(\n",
" docs,\n",
" embedding=FakeEmbeddings(size=768),\n",
- " doc_metadata={\"speech\": \"state-of-the-union\"},\n",
+ " doc_metadata={\"speech\": \"state-of-the-union\", \"author\": \"Biden\"},\n",
")"
]
},
@@ -130,7 +144,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 5,
"id": "85ef3468",
"metadata": {},
"outputs": [],
@@ -142,14 +156,16 @@
" [\n",
" \"https://www.gilderlehrman.org/sites/default/files/inline-pdfs/king.dreamspeech.excerpts.pdf\",\n",
" \"I-have-a-dream\",\n",
+ " \"Dr. King\"\n",
" ],\n",
" [\n",
" \"https://www.parkwayschools.net/cms/lib/MO01931486/Centricity/Domain/1578/Churchill_Beaches_Speech.pdf\",\n",
" \"we shall fight on the beaches\",\n",
+ " \"Churchil\"\n",
" ],\n",
"]\n",
"files_list = []\n",
- "for url, _ in urls:\n",
+ "for url, _, _ in urls:\n",
" name = tempfile.NamedTemporaryFile().name\n",
" urllib.request.urlretrieve(url, name)\n",
" files_list.append(name)\n",
@@ -157,7 +173,7 @@
"docsearch: Vectara = Vectara.from_files(\n",
" files=files_list,\n",
" embedding=FakeEmbeddings(size=768),\n",
- " metadatas=[{\"url\": url, \"speech\": title} for url, title in urls],\n",
+ " metadatas=[{\"url\": url, \"speech\": title, \"author\": author} for url, title, author in urls],\n",
")"
]
},
@@ -178,7 +194,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 6,
"id": "a8c513ab",
"metadata": {
"ExecuteTime": {
@@ -197,7 +213,7 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 7,
"id": "fc516993",
"metadata": {
"ExecuteTime": {
@@ -231,7 +247,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 8,
"id": "8804a21d",
"metadata": {
"ExecuteTime": {
@@ -249,7 +265,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 9,
"id": "756a6887",
"metadata": {
"ExecuteTime": {
@@ -264,7 +280,7 @@
"text": [
"Justice Breyer, thank you for your service. One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence. A former top litigator in private practice.\n",
"\n",
- "Score: 0.786569\n"
+ "Score: 0.8299499\n"
]
}
],
@@ -284,7 +300,7 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 10,
"id": "47784de5",
"metadata": {},
"outputs": [
@@ -307,7 +323,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 11,
"id": "3e22949f",
"metadata": {},
"outputs": [
@@ -315,7 +331,7 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "With this threshold of 0.2 we have 3 documents\n"
+ "With this threshold of 0.2 we have 5 documents\n"
]
}
],
@@ -340,7 +356,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 12,
"id": "9427195f",
"metadata": {
"ExecuteTime": {
@@ -352,10 +368,10 @@
{
"data": {
"text/plain": [
- "VectaraRetriever(tags=['Vectara'], metadata=None, vectorstore=, search_type='similarity', search_kwargs={'lambda_val': 0.025, 'k': 5, 'filter': '', 'n_sentence_context': '2'})"
+ "VectaraRetriever(tags=['Vectara'], metadata=None, vectorstore=, search_type='similarity', search_kwargs={'lambda_val': 0.025, 'k': 5, 'filter': '', 'n_sentence_context': '2'})"
]
},
- "execution_count": 11,
+ "execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
@@ -367,7 +383,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 13,
"id": "f3c70c31",
"metadata": {
"ExecuteTime": {
@@ -379,10 +395,10 @@
{
"data": {
"text/plain": [
- "Document(page_content='Justice Breyer, thank you for your service. One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence. A former top litigator in private practice.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '596', 'len': '97', 'speech': 'state-of-the-union'})"
+ "Document(page_content='Justice Breyer, thank you for your service. One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence. A former top litigator in private practice.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '596', 'len': '97', 'speech': 'state-of-the-union', 'author': 'Biden'})"
]
},
- "execution_count": 12,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
@@ -392,10 +408,118 @@
"retriever.get_relevant_documents(query)[0]"
]
},
+ {
+ "cell_type": "markdown",
+ "id": "e944c26a",
+ "metadata": {},
+ "source": [
+ "## Using Vectara as a SelfQuery Retriever"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "id": "8be674de",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "metadata_field_info = [\n",
+ " AttributeInfo(\n",
+ " name=\"speech\",\n",
+ " description=\"what name of the speech\",\n",
+ " type=\"string or list[string]\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"author\",\n",
+ " description=\"author of the speech\",\n",
+ " type=\"string or list[string]\",\n",
+ " ),\n",
+ "]\n",
+ "document_content_description = \"the text of the speech\"\n",
+ "\n",
+ "vectordb = Vectara()\n",
+ "llm = OpenAI(temperature=0)\n",
+ "retriever = SelfQueryRetriever.from_llm(llm, vectara, \n",
+ " document_content_description, metadata_field_info, \n",
+ " verbose=True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "id": "f8938999",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/ofer/dev/langchain/libs/langchain/langchain/chains/llm.py:278: UserWarning: The predict_and_parse method is deprecated, instead pass an output parser directly to LLMChain.\n",
+ " warnings.warn(\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='freedom' filter=Comparison(comparator=, attribute='author', value='Biden') limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Well I know this nation. We will meet the test. To protect freedom and liberty, to expand fairness and opportunity. We will save democracy. As hard as these times have been, I am more optimistic about America today than I have been my whole life.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '346', 'len': '67', 'speech': 'state-of-the-union', 'author': 'Biden'}),\n",
+ " Document(page_content='To our fellow Ukrainian Americans who forge a deep bond that connects our two nations we stand with you. Putin may circle Kyiv with tanks, but he will never gain the hearts and souls of the Ukrainian people. He will never extinguish their love of freedom. He will never weaken the resolve of the free world. We meet tonight in an America that has lived through two of the hardest years this nation has ever faced.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '740', 'len': '47', 'speech': 'state-of-the-union', 'author': 'Biden'}),\n",
+ " Document(page_content='But most importantly as Americans. With a duty to one another to the American people to the Constitution. And with an unwavering resolve that freedom will always triumph over tyranny. Six days ago, Russia’s Vladimir Putin sought to shake the foundations of the free world thinking he could make it bend to his menacing ways. But he badly miscalculated.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '413', 'len': '77', 'speech': 'state-of-the-union', 'author': 'Biden'}),\n",
+ " Document(page_content='We can do this. \\n\\nMy fellow Americans—tonight , we have gathered in a sacred space—the citadel of our democracy. In this Capitol, generation after generation, Americans have debated great questions amid great strife, and have done great things. We have fought for freedom, expanded liberty, defeated totalitarianism and terror. And built the strongest, freest, and most prosperous nation the world has ever known. Now is the hour. \\n\\nOur moment of responsibility.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '906', 'len': '82', 'speech': 'state-of-the-union', 'author': 'Biden'}),\n",
+ " Document(page_content='In state after state, new laws have been passed, not only to suppress the vote, but to subvert entire elections. We cannot let this happen. Tonight. I call on the Senate to: Pass the Freedom to Vote Act. Pass the John Lewis Voting Rights Act. And while you’re at it, pass the Disclose Act so Americans can know who is funding our elections.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '0', 'len': '63', 'speech': 'state-of-the-union', 'author': 'Biden'})]"
+ ]
+ },
+ "execution_count": 16,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "retriever.get_relevant_documents(\"what did Biden say about the freedom?\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "id": "a97037fb",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='freedom' filter=Comparison(comparator=, attribute='author', value='Dr. King') limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='And if America is to be a great nation, this must become true. So\\nlet freedom ring from the prodigious hilltops of New Hampshire. Let freedom ring from the mighty\\nmountains of New York. Let freedom ring from the heightening Alleghenies of Pennsylvania. Let\\nfreedom ring from the snowcapped Rockies of Colorado.', metadata={'lang': 'eng', 'section': '3', 'offset': '1534', 'len': '55', 'CreationDate': '1424880481', 'Producer': 'Adobe PDF Library 10.0', 'Author': 'Sasha Rolon-Pereira', 'Title': 'Martin Luther King Jr.pdf', 'Creator': 'Acrobat PDFMaker 10.1 for Word', 'ModDate': '1424880524', 'url': 'https://www.gilderlehrman.org/sites/default/files/inline-pdfs/king.dreamspeech.excerpts.pdf', 'speech': 'I-have-a-dream', 'author': 'Dr. King', 'title': 'Martin Luther King Jr.pdf'}),\n",
+ " Document(page_content='And if America is to be a great nation, this must become true. So\\nlet freedom ring from the prodigious hilltops of New Hampshire. Let freedom ring from the mighty\\nmountains of New York. Let freedom ring from the heightening Alleghenies of Pennsylvania. Let\\nfreedom ring from the snowcapped Rockies of Colorado.', metadata={'lang': 'eng', 'section': '3', 'offset': '1534', 'len': '55', 'CreationDate': '1424880481', 'Producer': 'Adobe PDF Library 10.0', 'Author': 'Sasha Rolon-Pereira', 'Title': 'Martin Luther King Jr.pdf', 'Creator': 'Acrobat PDFMaker 10.1 for Word', 'ModDate': '1424880524', 'url': 'https://www.gilderlehrman.org/sites/default/files/inline-pdfs/king.dreamspeech.excerpts.pdf', 'speech': 'I-have-a-dream', 'author': 'Dr. King', 'title': 'Martin Luther King Jr.pdf'}),\n",
+ " Document(page_content='Let freedom ring from the curvaceous slopes of\\nCalifornia. But not only that. Let freedom ring from Stone Mountain of Georgia. Let freedom ring from Lookout\\nMountain of Tennessee. Let freedom ring from every hill and molehill of Mississippi, from every\\nmountain side. Let freedom ring . . .\\nWhen we allow freedom to ring—when we let it ring from every city and every hamlet, from every state\\nand every city, we will be able to speed up that day when all of God’s children, black men and white\\nmen, Jews and Gentiles, Protestants and Catholics, will be able to join hands and sing in the words of the\\nold Negro spiritual, “Free at last, Free at last, Great God a-mighty, We are free at last.”', metadata={'lang': 'eng', 'section': '3', 'offset': '1842', 'len': '52', 'CreationDate': '1424880481', 'Producer': 'Adobe PDF Library 10.0', 'Author': 'Sasha Rolon-Pereira', 'Title': 'Martin Luther King Jr.pdf', 'Creator': 'Acrobat PDFMaker 10.1 for Word', 'ModDate': '1424880524', 'url': 'https://www.gilderlehrman.org/sites/default/files/inline-pdfs/king.dreamspeech.excerpts.pdf', 'speech': 'I-have-a-dream', 'author': 'Dr. King', 'title': 'Martin Luther King Jr.pdf'}),\n",
+ " Document(page_content='Let freedom ring from the curvaceous slopes of\\nCalifornia. But not only that. Let freedom ring from Stone Mountain of Georgia. Let freedom ring from Lookout\\nMountain of Tennessee. Let freedom ring from every hill and molehill of Mississippi, from every\\nmountain side. Let freedom ring . . .\\nWhen we allow freedom to ring—when we let it ring from every city and every hamlet, from every state\\nand every city, we will be able to speed up that day when all of God’s children, black men and white\\nmen, Jews and Gentiles, Protestants and Catholics, will be able to join hands and sing in the words of the\\nold Negro spiritual, “Free at last, Free at last, Great God a-mighty, We are free at last.”', metadata={'lang': 'eng', 'section': '3', 'offset': '1842', 'len': '52', 'CreationDate': '1424880481', 'Producer': 'Adobe PDF Library 10.0', 'Author': 'Sasha Rolon-Pereira', 'Title': 'Martin Luther King Jr.pdf', 'Creator': 'Acrobat PDFMaker 10.1 for Word', 'ModDate': '1424880524', 'url': 'https://www.gilderlehrman.org/sites/default/files/inline-pdfs/king.dreamspeech.excerpts.pdf', 'speech': 'I-have-a-dream', 'author': 'Dr. King', 'title': 'Martin Luther King Jr.pdf'}),\n",
+ " Document(page_content='Let freedom ring from the mighty\\nmountains of New York. Let freedom ring from the heightening Alleghenies of Pennsylvania. Let\\nfreedom ring from the snowcapped Rockies of Colorado. Let freedom ring from the curvaceous slopes of\\nCalifornia. But not only that. Let freedom ring from Stone Mountain of Georgia.', metadata={'lang': 'eng', 'section': '3', 'offset': '1657', 'len': '57', 'CreationDate': '1424880481', 'Producer': 'Adobe PDF Library 10.0', 'Author': 'Sasha Rolon-Pereira', 'Title': 'Martin Luther King Jr.pdf', 'Creator': 'Acrobat PDFMaker 10.1 for Word', 'ModDate': '1424880524', 'url': 'https://www.gilderlehrman.org/sites/default/files/inline-pdfs/king.dreamspeech.excerpts.pdf', 'speech': 'I-have-a-dream', 'author': 'Dr. King', 'title': 'Martin Luther King Jr.pdf'})]"
+ ]
+ },
+ "execution_count": 17,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "retriever.get_relevant_documents(\"what did Dr. King say about the freedom?\")"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
- "id": "2300e785",
+ "id": "f6d17e90",
"metadata": {},
"outputs": [],
"source": []
diff --git a/docs/extras/modules/data_connection/retrievers/self_query/supabase_self_query.ipynb b/docs/extras/modules/data_connection/retrievers/self_query/supabase_self_query.ipynb
new file mode 100644
index 0000000000..1414f70d38
--- /dev/null
+++ b/docs/extras/modules/data_connection/retrievers/self_query/supabase_self_query.ipynb
@@ -0,0 +1,587 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "13afcae7",
+ "metadata": {},
+ "source": [
+ "# Supabase Vector self-querying \n",
+ "\n",
+ ">[Supabase](https://supabase.com/docs) is an open source `Firebase` alternative. \n",
+ "> `Supabase` is built on top of `PostgreSQL`, which offers strong `SQL` \n",
+ "> querying capabilities and enables a simple interface with already-existing tools and frameworks.\n",
+ "\n",
+ ">[PostgreSQL](https://en.wikipedia.org/wiki/PostgreSQL) also known as `Postgres`,\n",
+ "> is a free and open-source relational database management system (RDBMS) \n",
+ "> emphasizing extensibility and `SQL` compliance.\n",
+ "\n",
+ "In the notebook we'll demo the `SelfQueryRetriever` wrapped around a Supabase vector store.\n",
+ "\n",
+ "Specifically we will:\n",
+ "1. Create a Supabase database\n",
+ "2. Enable the `pgvector` extension\n",
+ "3. Create a `documents` table and `match_documents` function that will be used by `SupabaseVectorStore`\n",
+ "4. Load sample documents into the vector store (database table)\n",
+ "5. Build and test a self-querying retriever"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "347935ad",
+ "metadata": {},
+ "source": [
+ "## Setup Supabase Database\n",
+ "\n",
+ "1. Head over to https://database.new to provision your Supabase database.\n",
+ "2. In the studio, jump to the [SQL editor](https://supabase.com/dashboard/project/_/sql/new) and run the following script to enable `pgvector` and setup your database as a vector store:\n",
+ " ```sql\n",
+ " -- Enable the pgvector extension to work with embedding vectors\n",
+ " create extension if not exists vector;\n",
+ "\n",
+ " -- Create a table to store your documents\n",
+ " create table\n",
+ " documents (\n",
+ " id uuid primary key,\n",
+ " content text, -- corresponds to Document.pageContent\n",
+ " metadata jsonb, -- corresponds to Document.metadata\n",
+ " embedding vector (1536) -- 1536 works for OpenAI embeddings, change if needed\n",
+ " );\n",
+ "\n",
+ " -- Create a function to search for documents\n",
+ " create function match_documents (\n",
+ " query_embedding vector (1536),\n",
+ " filter jsonb default '{}'\n",
+ " ) returns table (\n",
+ " id uuid,\n",
+ " content text,\n",
+ " metadata jsonb,\n",
+ " similarity float\n",
+ " ) language plpgsql as $$\n",
+ " #variable_conflict use_column\n",
+ " begin\n",
+ " return query\n",
+ " select\n",
+ " id,\n",
+ " content,\n",
+ " metadata,\n",
+ " 1 - (documents.embedding <=> query_embedding) as similarity\n",
+ " from documents\n",
+ " where metadata @> filter\n",
+ " order by documents.embedding <=> query_embedding;\n",
+ " end;\n",
+ " $$;\n",
+ " ```"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "68e75fb9",
+ "metadata": {},
+ "source": [
+ "## Creating a Supabase vector store\n",
+ "Next we'll want to create a Supabase vector store and seed it with some data. We've created a small demo set of documents that contain summaries of movies.\n",
+ "\n",
+ "Be sure to install the latest version of `langchain`:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "78546fd7",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "%pip install langchain"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e06df198",
+ "metadata": {},
+ "source": [
+ "The self-query retriever requires you to have `lark` installed:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "63a8af5b",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "%pip install lark"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "114f768f",
+ "metadata": {},
+ "source": [
+ "We also need the `openai` and `supabase` packages:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "434ae558",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "%pip install openai"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "22431060-52c4-48a7-a97b-9f542b8b0928",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "%pip install supabase==1.0.0"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "83811610-7df3-4ede-b268-68a6a83ba9e2",
+ "metadata": {},
+ "source": [
+ "Since we are using `SupabaseVectorStore` and `OpenAIEmbeddings`, we have to load their API keys.\n",
+ "\n",
+ "- To find your `SUPABASE_URL` and `SUPABASE_SERVICE_KEY`, head to your Supabase project's [API settings](https://supabase.com/dashboard/project/_/settings/api).\n",
+ " - `SUPABASE_URL` corresponds to the Project URL\n",
+ " - `SUPABASE_SERVICE_KEY` corresponds to the `service_role` API key\n",
+ "\n",
+ "- To get your `OPENAI_API_KEY`, navigate to [API keys](https://platform.openai.com/account/api-keys) on your OpenAI account and create a new secret key."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "id": "dd01b61b-7d32-4a55-85d6-b2d2d4f18840",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "import os\n",
+ "import getpass\n",
+ "\n",
+ "os.environ[\"SUPABASE_URL\"] = getpass.getpass(\"Supabase URL:\")\n",
+ "os.environ[\"SUPABASE_SERVICE_KEY\"] = getpass.getpass(\"Supabase Service Key:\")\n",
+ "os.environ[\"OPENAI_API_KEY\"] = getpass.getpass(\"OpenAI API Key:\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3aaf5075",
+ "metadata": {},
+ "source": [
+ "_Optional:_ If you're storing your Supabase and OpenAI API keys in a `.env` file, you can load them with [`dotenv`](https://github.com/theskumar/python-dotenv)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "e0089221",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "%pip install python-dotenv"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "3d56c5ef",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from dotenv import load_dotenv\n",
+ "\n",
+ "load_dotenv()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f6dd9aef",
+ "metadata": {},
+ "source": [
+ "First we'll create a Supabase client and instantiate a OpenAI embeddings class."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "cb4a5787",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "import os\n",
+ "from supabase.client import Client, create_client\n",
+ "from langchain.schema import Document\n",
+ "from langchain.embeddings.openai import OpenAIEmbeddings\n",
+ "from langchain.vectorstores import SupabaseVectorStore\n",
+ "\n",
+ "supabase_url = os.environ.get(\"SUPABASE_URL\")\n",
+ "supabase_key = os.environ.get(\"SUPABASE_SERVICE_KEY\")\n",
+ "supabase: Client = create_client(supabase_url, supabase_key)\n",
+ "\n",
+ "embeddings = OpenAIEmbeddings()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0fca9b0b",
+ "metadata": {},
+ "source": [
+ "Next let's create our documents."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "bcbe04d9",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "docs = [\n",
+ " Document(\n",
+ " page_content=\"A bunch of scientists bring back dinosaurs and mayhem breaks loose\",\n",
+ " metadata={\"year\": 1993, \"rating\": 7.7, \"genre\": \"science fiction\"},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"Leo DiCaprio gets lost in a dream within a dream within a dream within a ...\",\n",
+ " metadata={\"year\": 2010, \"director\": \"Christopher Nolan\", \"rating\": 8.2},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea\",\n",
+ " metadata={\"year\": 2006, \"director\": \"Satoshi Kon\", \"rating\": 8.6},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"A bunch of normal-sized women are supremely wholesome and some men pine after them\",\n",
+ " metadata={\"year\": 2019, \"director\": \"Greta Gerwig\", \"rating\": 8.3},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"Toys come alive and have a blast doing so\",\n",
+ " metadata={\"year\": 1995, \"genre\": \"animated\"},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"Three men walk into the Zone, three men walk out of the Zone\",\n",
+ " metadata={\n",
+ " \"year\": 1979,\n",
+ " \"rating\": 9.9,\n",
+ " \"director\": \"Andrei Tarkovsky\",\n",
+ " \"genre\": \"science fiction\",\n",
+ " \"rating\": 9.9,\n",
+ " },\n",
+ " ),\n",
+ "]\n",
+ "\n",
+ "vectorstore = SupabaseVectorStore.from_documents(docs, embeddings, client=supabase, table_name=\"documents\", query_name=\"match_documents\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "5ecaab6d",
+ "metadata": {},
+ "source": [
+ "## Creating our self-querying retriever\n",
+ "Now we can instantiate our retriever. To do this we'll need to provide some information upfront about the metadata fields that our documents support and a short description of the document contents."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "86e34dbf",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "from langchain.llms import OpenAI\n",
+ "from langchain.retrievers.self_query.base import SelfQueryRetriever\n",
+ "from langchain.chains.query_constructor.base import AttributeInfo\n",
+ "\n",
+ "metadata_field_info = [\n",
+ " AttributeInfo(\n",
+ " name=\"genre\",\n",
+ " description=\"The genre of the movie\",\n",
+ " type=\"string or list[string]\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"year\",\n",
+ " description=\"The year the movie was released\",\n",
+ " type=\"integer\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"director\",\n",
+ " description=\"The name of the movie director\",\n",
+ " type=\"string\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"rating\", description=\"A 1-10 rating for the movie\", type=\"float\"\n",
+ " ),\n",
+ "]\n",
+ "document_content_description = \"Brief summary of a movie\"\n",
+ "llm = OpenAI(temperature=0)\n",
+ "retriever = SelfQueryRetriever.from_llm(\n",
+ " llm, vectorstore, document_content_description, metadata_field_info, verbose=True\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ea9df8d4",
+ "metadata": {},
+ "source": [
+ "## Testing it out\n",
+ "And now we can try actually using our retriever!"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "38a126e9",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='dinosaur' filter=None limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='A bunch of scientists bring back dinosaurs and mayhem breaks loose', metadata={'year': 1993, 'genre': 'science fiction', 'rating': 7.7}),\n",
+ " Document(page_content='Toys come alive and have a blast doing so', metadata={'year': 1995, 'genre': 'animated'}),\n",
+ " Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'year': 1979, 'genre': 'science fiction', 'rating': 9.9, 'director': 'Andrei Tarkovsky'}),\n",
+ " Document(page_content='A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea', metadata={'year': 2006, 'rating': 8.6, 'director': 'Satoshi Kon'})]"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example only specifies a relevant query\n",
+ "retriever.get_relevant_documents(\"What are some movies about dinosaurs\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "fc3f1e6e",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query=' ' filter=Comparison(comparator=, attribute='rating', value=8.5) limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'year': 1979, 'genre': 'science fiction', 'rating': 9.9, 'director': 'Andrei Tarkovsky'}),\n",
+ " Document(page_content='A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea', metadata={'year': 2006, 'rating': 8.6, 'director': 'Satoshi Kon'})]"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example only specifies a filter\n",
+ "retriever.get_relevant_documents(\"I want to watch a movie rated higher than 8.5\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "b19d4da0",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='women' filter=Comparison(comparator=, attribute='director', value='Greta Gerwig') limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='A bunch of normal-sized women are supremely wholesome and some men pine after them', metadata={'year': 2019, 'rating': 8.3, 'director': 'Greta Gerwig'})]"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example specifies a query and a filter\n",
+ "retriever.get_relevant_documents(\"Has Greta Gerwig directed any movies about women?\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "f900e40e",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query=' ' filter=Operation(operator=, arguments=[Comparison(comparator=, attribute='rating', value=8.5), Comparison(comparator=, attribute='genre', value='science fiction')]) limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'year': 1979, 'genre': 'science fiction', 'rating': 9.9, 'director': 'Andrei Tarkovsky'})]"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example specifies a composite filter\n",
+ "retriever.get_relevant_documents(\n",
+ " \"What's a highly rated (above 8.5) science fiction film?\"\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "12a51522",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='toys' filter=Operation(operator=, arguments=[Comparison(comparator=, attribute='year', value=1990), Comparison(comparator=, attribute='year', value=2005), Comparison(comparator=, attribute='genre', value='animated')]) limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Toys come alive and have a blast doing so', metadata={'year': 1995, 'genre': 'animated'})]"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example specifies a query and composite filter\n",
+ "retriever.get_relevant_documents(\n",
+ " \"What's a movie after 1990 but before (or on) 2005 that's all about toys, and preferably is animated\"\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "39bd1de1-b9fe-4a98-89da-58d8a7a6ae51",
+ "metadata": {},
+ "source": [
+ "## Filter k\n",
+ "\n",
+ "We can also use the self query retriever to specify `k`: the number of documents to fetch.\n",
+ "\n",
+ "We can do this by passing `enable_limit=True` to the constructor."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "bff36b88-b506-4877-9c63-e5a1a8d78e64",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "retriever = SelfQueryRetriever.from_llm(\n",
+ " llm,\n",
+ " vectorstore,\n",
+ " document_content_description,\n",
+ " metadata_field_info,\n",
+ " enable_limit=True,\n",
+ " verbose=True,\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "2758d229-4f97-499c-819f-888acaf8ee10",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='dinosaur' filter=None limit=2\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='A bunch of scientists bring back dinosaurs and mayhem breaks loose', metadata={'year': 1993, 'genre': 'science fiction', 'rating': 7.7}),\n",
+ " Document(page_content='Toys come alive and have a blast doing so', metadata={'year': 1995, 'genre': 'animated'})]"
+ ]
+ },
+ "execution_count": 11,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example only specifies a relevant query\n",
+ "retriever.get_relevant_documents(\"what are two movies about dinosaurs\")"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.12"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/modules/data_connection/retrievers/self_query/vectara_self_query.ipynb b/docs/extras/modules/data_connection/retrievers/self_query/vectara_self_query.ipynb
new file mode 100644
index 0000000000..1e9128dc6f
--- /dev/null
+++ b/docs/extras/modules/data_connection/retrievers/self_query/vectara_self_query.ipynb
@@ -0,0 +1,440 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "13afcae7",
+ "metadata": {},
+ "source": [
+ "# Vectara self-querying \n",
+ "\n",
+ ">[Vectara](https://docs.vectara.com/docs/) is a GenAI platform for developers. It provides a simple API to build Grounded Generation (aka Retrieval-augmented-generation) applications.\n",
+ "\n",
+ "In the notebook we'll demo the `SelfQueryRetriever` wrapped around a Vectara vector store. "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "68e75fb9",
+ "metadata": {},
+ "source": [
+ "# Setup\n",
+ "\n",
+ "You will need a Vectara account to use Vectara with LangChain. To get started, use the following steps (see our [quickstart](https://docs.vectara.com/docs/quickstart) guide):\n",
+ "1. [Sign up](https://console.vectara.com/signup) for a Vectara account if you don't already have one. Once you have completed your sign up you will have a Vectara customer ID. You can find your customer ID by clicking on your name, on the top-right of the Vectara console window.\n",
+ "2. Within your account you can create one or more corpora. Each corpus represents an area that stores text data upon ingest from input documents. To create a corpus, use the **\"Create Corpus\"** button. You then provide a name to your corpus as well as a description. Optionally you can define filtering attributes and apply some advanced options. If you click on your created corpus, you can see its name and corpus ID right on the top.\n",
+ "3. Next you'll need to create API keys to access the corpus. Click on the **\"Authorization\"** tab in the corpus view and then the **\"Create API Key\"** button. Give your key a name, and choose whether you want query only or query+index for your key. Click \"Create\" and you now have an active API key. Keep this key confidential. \n",
+ "\n",
+ "To use LangChain with Vectara, you'll need to have these three values: customer ID, corpus ID and api_key.\n",
+ "You can provide those to LangChain in two ways:\n",
+ "\n",
+ "1. Include in your environment these three variables: `VECTARA_CUSTOMER_ID`, `VECTARA_CORPUS_ID` and `VECTARA_API_KEY`.\n",
+ "\n",
+ "> For example, you can set these variables using os.environ and getpass as follows:\n",
+ "\n",
+ "```python\n",
+ "import os\n",
+ "import getpass\n",
+ "\n",
+ "os.environ[\"VECTARA_CUSTOMER_ID\"] = getpass.getpass(\"Vectara Customer ID:\")\n",
+ "os.environ[\"VECTARA_CORPUS_ID\"] = getpass.getpass(\"Vectara Corpus ID:\")\n",
+ "os.environ[\"VECTARA_API_KEY\"] = getpass.getpass(\"Vectara API Key:\")\n",
+ "```\n",
+ "\n",
+ "1. Provide them as arguments when creating the Vectara vectorstore object:\n",
+ "\n",
+ "```python\n",
+ "vectorstore = Vectara(\n",
+ " vectara_customer_id=vectara_customer_id,\n",
+ " vectara_corpus_id=vectara_corpus_id,\n",
+ " vectara_api_key=vectara_api_key\n",
+ " )\n",
+ "```\n",
+ "\n",
+ "**Note:** The self-query retriever requires you to have `lark` installed (`pip install lark`). "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "742ac16d",
+ "metadata": {},
+ "source": [
+ "## Connecting to Vectara from LangChain\n",
+ "\n",
+ "In this example, we assume that you've created an account and a corpus, and added your VECTARA_CUSTOMER_ID, VECTARA_CORPUS_ID and VECTARA_API_KEY (created with permissions for both indexing and query) as environment variables.\n",
+ "\n",
+ "The corpus has 4 fields defined as metadata for filtering: year, director, rating, and genre\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "cb4a5787",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "from langchain.embeddings import FakeEmbeddings\n",
+ "from langchain.schema import Document\n",
+ "from langchain.text_splitter import CharacterTextSplitter\n",
+ "from langchain.vectorstores import Vectara\n",
+ "from langchain.document_loaders import TextLoader\n",
+ "\n",
+ "from langchain.llms import OpenAI\n",
+ "from langchain.chains import ConversationalRetrievalChain\n",
+ "from langchain.retrievers.self_query.base import SelfQueryRetriever\n",
+ "from langchain.chains.query_constructor.base import AttributeInfo\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "bcbe04d9",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "docs = [\n",
+ " Document(\n",
+ " page_content=\"A bunch of scientists bring back dinosaurs and mayhem breaks loose\",\n",
+ " metadata={\"year\": 1993, \"rating\": 7.7, \"genre\": \"science fiction\"},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"Leo DiCaprio gets lost in a dream within a dream within a dream within a ...\",\n",
+ " metadata={\"year\": 2010, \"director\": \"Christopher Nolan\", \"rating\": 8.2},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea\",\n",
+ " metadata={\"year\": 2006, \"director\": \"Satoshi Kon\", \"rating\": 8.6},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"A bunch of normal-sized women are supremely wholesome and some men pine after them\",\n",
+ " metadata={\"year\": 2019, \"director\": \"Greta Gerwig\", \"rating\": 8.3},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"Toys come alive and have a blast doing so\",\n",
+ " metadata={\"year\": 1995, \"genre\": \"animated\"},\n",
+ " ),\n",
+ " Document(\n",
+ " page_content=\"Three men walk into the Zone, three men walk out of the Zone\",\n",
+ " metadata={\n",
+ " \"year\": 1979,\n",
+ " \"rating\": 9.9,\n",
+ " \"director\": \"Andrei Tarkovsky\",\n",
+ " \"genre\": \"science fiction\",\n",
+ " },\n",
+ " ),\n",
+ "]\n",
+ "\n",
+ "vectara = Vectara()\n",
+ "for doc in docs:\n",
+ " vectara.add_texts([doc.page_content], embedding=FakeEmbeddings(size=768), doc_metadata=doc.metadata)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "5ecaab6d",
+ "metadata": {},
+ "source": [
+ "## Creating our self-querying retriever\n",
+ "Now we can instantiate our retriever. To do this we'll need to provide some information upfront about the metadata fields that our documents support and a short description of the document contents."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "86e34dbf",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "from langchain.llms import OpenAI\n",
+ "from langchain.retrievers.self_query.base import SelfQueryRetriever\n",
+ "from langchain.chains.query_constructor.base import AttributeInfo\n",
+ "\n",
+ "metadata_field_info = [\n",
+ " AttributeInfo(\n",
+ " name=\"genre\",\n",
+ " description=\"The genre of the movie\",\n",
+ " type=\"string or list[string]\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"year\",\n",
+ " description=\"The year the movie was released\",\n",
+ " type=\"integer\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"director\",\n",
+ " description=\"The name of the movie director\",\n",
+ " type=\"string\",\n",
+ " ),\n",
+ " AttributeInfo(\n",
+ " name=\"rating\", description=\"A 1-10 rating for the movie\", type=\"float\"\n",
+ " ),\n",
+ "]\n",
+ "document_content_description = \"Brief summary of a movie\"\n",
+ "llm = OpenAI(temperature=0)\n",
+ "retriever = SelfQueryRetriever.from_llm(\n",
+ " llm, vectara, document_content_description, metadata_field_info, verbose=True\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ea9df8d4",
+ "metadata": {},
+ "source": [
+ "## Testing it out\n",
+ "And now we can try actually using our retriever!"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "38a126e9",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/ofer/dev/langchain/libs/langchain/langchain/chains/llm.py:278: UserWarning: The predict_and_parse method is deprecated, instead pass an output parser directly to LLMChain.\n",
+ " warnings.warn(\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='dinosaur' filter=None limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='A bunch of scientists bring back dinosaurs and mayhem breaks loose', metadata={'lang': 'eng', 'offset': '0', 'len': '66', 'year': '1993', 'rating': '7.7', 'genre': 'science fiction', 'source': 'langchain'}),\n",
+ " Document(page_content='Toys come alive and have a blast doing so', metadata={'lang': 'eng', 'offset': '0', 'len': '41', 'year': '1995', 'genre': 'animated', 'source': 'langchain'}),\n",
+ " Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'lang': 'eng', 'offset': '0', 'len': '60', 'year': '1979', 'rating': '9.9', 'director': 'Andrei Tarkovsky', 'genre': 'science fiction', 'source': 'langchain'}),\n",
+ " Document(page_content='Leo DiCaprio gets lost in a dream within a dream within a dream within a ...', metadata={'lang': 'eng', 'offset': '0', 'len': '76', 'year': '2010', 'director': 'Christopher Nolan', 'rating': '8.2', 'source': 'langchain'}),\n",
+ " Document(page_content='A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea', metadata={'lang': 'eng', 'offset': '0', 'len': '116', 'year': '2006', 'director': 'Satoshi Kon', 'rating': '8.6', 'source': 'langchain'})]"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example only specifies a relevant query\n",
+ "retriever.get_relevant_documents(\"What are some movies about dinosaurs\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "fc3f1e6e",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query=' ' filter=Comparison(comparator=, attribute='rating', value=8.5) limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'lang': 'eng', 'offset': '0', 'len': '60', 'year': '1979', 'rating': '9.9', 'director': 'Andrei Tarkovsky', 'genre': 'science fiction', 'source': 'langchain'}),\n",
+ " Document(page_content='A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea', metadata={'lang': 'eng', 'offset': '0', 'len': '116', 'year': '2006', 'director': 'Satoshi Kon', 'rating': '8.6', 'source': 'langchain'})]"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example only specifies a filter\n",
+ "retriever.get_relevant_documents(\"I want to watch a movie rated higher than 8.5\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "b19d4da0",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='women' filter=Comparison(comparator=, attribute='director', value='Greta Gerwig') limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='A bunch of normal-sized women are supremely wholesome and some men pine after them', metadata={'lang': 'eng', 'offset': '0', 'len': '82', 'year': '2019', 'director': 'Greta Gerwig', 'rating': '8.3', 'source': 'langchain'})]"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example specifies a query and a filter\n",
+ "retriever.get_relevant_documents(\"Has Greta Gerwig directed any movies about women\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "f900e40e",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query=' ' filter=Operation(operator=, arguments=[Comparison(comparator=, attribute='rating', value=8.5), Comparison(comparator=, attribute='genre', value='science fiction')]) limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'lang': 'eng', 'offset': '0', 'len': '60', 'year': '1979', 'rating': '9.9', 'director': 'Andrei Tarkovsky', 'genre': 'science fiction', 'source': 'langchain'})]"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example specifies a composite filter\n",
+ "retriever.get_relevant_documents(\n",
+ " \"What's a highly rated (above 8.5) science fiction film?\"\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "12a51522",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='toys' filter=Operation(operator=, arguments=[Comparison(comparator=, attribute='year', value=1990), Comparison(comparator=, attribute='year', value=2005), Comparison(comparator=, attribute='genre', value='animated')]) limit=None\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='Toys come alive and have a blast doing so', metadata={'lang': 'eng', 'offset': '0', 'len': '41', 'year': '1995', 'genre': 'animated', 'source': 'langchain'})]"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example specifies a query and composite filter\n",
+ "retriever.get_relevant_documents(\n",
+ " \"What's a movie after 1990 but before 2005 that's all about toys, and preferably is animated\"\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "39bd1de1-b9fe-4a98-89da-58d8a7a6ae51",
+ "metadata": {},
+ "source": [
+ "## Filter k\n",
+ "\n",
+ "We can also use the self query retriever to specify `k`: the number of documents to fetch.\n",
+ "\n",
+ "We can do this by passing `enable_limit=True` to the constructor."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "bff36b88-b506-4877-9c63-e5a1a8d78e64",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "retriever = SelfQueryRetriever.from_llm(\n",
+ " llm,\n",
+ " vectara,\n",
+ " document_content_description,\n",
+ " metadata_field_info,\n",
+ " enable_limit=True,\n",
+ " verbose=True,\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "2758d229-4f97-499c-819f-888acaf8ee10",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "query='dinosaur' filter=None limit=2\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "[Document(page_content='A bunch of scientists bring back dinosaurs and mayhem breaks loose', metadata={'lang': 'eng', 'offset': '0', 'len': '66', 'year': '1993', 'rating': '7.7', 'genre': 'science fiction', 'source': 'langchain'}),\n",
+ " Document(page_content='Toys come alive and have a blast doing so', metadata={'lang': 'eng', 'offset': '0', 'len': '41', 'year': '1995', 'genre': 'animated', 'source': 'langchain'})]"
+ ]
+ },
+ "execution_count": 11,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# This example only specifies a relevant query\n",
+ "retriever.get_relevant_documents(\"what are two movies about dinosaurs\")"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.9"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/use_cases/apis.ipynb b/docs/extras/use_cases/apis.ipynb
index 1af0a7f3ce..8d9259c3ca 100644
--- a/docs/extras/use_cases/apis.ipynb
+++ b/docs/extras/use_cases/apis.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "ea5c61b2-8b52-4270-bdb0-c4df88608f15",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Interacting with APIs\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "a15e6a18",
"metadata": {},
"source": [
- "# Interacting with APIs\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/apis.ipynb)\n",
"\n",
"## Use case \n",
@@ -69,9 +78,7 @@
"cell_type": "code",
"execution_count": 2,
"id": "30b780e3",
- "metadata": {
- "scrolled": false
- },
+ "metadata": {},
"outputs": [
{
"name": "stderr",
@@ -415,7 +422,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.16"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/extras/use_cases/chatbots.ipynb b/docs/extras/use_cases/chatbots.ipynb
index 58e3ce5317..c67d595c9f 100644
--- a/docs/extras/use_cases/chatbots.ipynb
+++ b/docs/extras/use_cases/chatbots.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "22fd28c9-9b48-476c-bca8-20efef7fdb14",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Chatbots\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "ee7f95e4",
"metadata": {},
"source": [
- "# Chatbots\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/chatbots.ipynb)\n",
"\n",
"## Use case\n",
diff --git a/docs/extras/use_cases/code_understanding.ipynb b/docs/extras/use_cases/code_understanding.ipynb
index 60a02b9bb3..df0cfbf9d1 100644
--- a/docs/extras/use_cases/code_understanding.ipynb
+++ b/docs/extras/use_cases/code_understanding.ipynb
@@ -1,11 +1,19 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Code understanding\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Code Understanding\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/code_understanding.ipynb)\n",
"\n",
"## Use case\n",
@@ -1047,7 +1055,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.16"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/extras/use_cases/extraction.ipynb b/docs/extras/use_cases/extraction.ipynb
index 7aaa37f046..628026127a 100644
--- a/docs/extras/use_cases/extraction.ipynb
+++ b/docs/extras/use_cases/extraction.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "df29b30a-fd27-4e08-8269-870df5631f9e",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Extraction\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "b84edb4e",
"metadata": {},
"source": [
- "# Extraction\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/extraction.ipynb)\n",
"\n",
"## Use case\n",
@@ -589,7 +598,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.16"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/extras/use_cases/more/_category_.yml b/docs/extras/use_cases/more/_category_.yml
index ee76dee18f..53055fb940 100644
--- a/docs/extras/use_cases/more/_category_.yml
+++ b/docs/extras/use_cases/more/_category_.yml
@@ -1 +1,2 @@
label: 'More'
+position: 2
\ No newline at end of file
diff --git a/docs/extras/use_cases/more/agents/agents.ipynb b/docs/extras/use_cases/more/agents/agents.ipynb
index 98b65d1bbe..54ba5c29db 100644
--- a/docs/extras/use_cases/more/agents/agents.ipynb
+++ b/docs/extras/use_cases/more/agents/agents.ipynb
@@ -584,7 +584,7 @@
"\n",
"Collectivly, this tells us: carefully inspect Agent traces and tool outputs. \n",
"\n",
- "As we saw with the [SQL use case](/docs/use_cases/sql), `ReAct agents` can be work very well for specific problems. \n",
+ "As we saw with the [SQL use case](/docs/use_cases/qa_structured/sql), `ReAct agents` can be work very well for specific problems. \n",
"\n",
"But, as shown here, the result is degraded relative to what we see with the OpenAI agent."
]
diff --git a/docs/extras/use_cases/more/graph/diffbot_graphtransformer.ipynb b/docs/extras/use_cases/more/graph/diffbot_graphtransformer.ipynb
new file mode 100644
index 0000000000..f8961174ce
--- /dev/null
+++ b/docs/extras/use_cases/more/graph/diffbot_graphtransformer.ipynb
@@ -0,0 +1,307 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "7f0b0c06-ee70-468c-8bf5-b023f9e5e0a2",
+ "metadata": {},
+ "source": [
+ "# Diffbot Graph Transformer\n",
+ "\n",
+ "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/more/graph/diffbot_graphtransformer.ipynb)\n",
+ "\n",
+ "## Use case\n",
+ "\n",
+ "Text data often contain rich relationships and insights that can be useful for various analytics, recommendation engines, or knowledge management applications.\n",
+ "\n",
+ "Diffbot's NLP API allows for the extraction of entities, relationships, and semantic meaning from unstructured text data.\n",
+ "\n",
+ "By coupling Diffbot's NLP API with Neo4j, a graph database, you can create powerful, dynamic graph structures based on the information extracted from text. These graph structures are fully queryable and can be integrated into various applications.\n",
+ "\n",
+ "This combination allows for use cases such as:\n",
+ "\n",
+ "* Building knowledge graphs from textual documents, websites, or social media feeds.\n",
+ "* Generating recommendations based on semantic relationships in the data.\n",
+ "* Creating advanced search features that understand the relationships between entities.\n",
+ "* Building analytics dashboards that allow users to explore the hidden relationships in data.\n",
+ "\n",
+ "## Overview\n",
+ "\n",
+ "LangChain provides tools to interact with Graph Databases:\n",
+ "\n",
+ "1. `Construct knowledge graphs from text` using graph transformer and store integrations \n",
+ "2. `Query a graph database` using chains for query creation and execution\n",
+ "3. `Interact with a graph database` using agents for robust and flexible querying \n",
+ "\n",
+ "## Quickstart\n",
+ "\n",
+ "First, get required packages and set environment variables:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "975648da-b24f-4164-a671-6772179e12df",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip install langchain langchain-experimental openai neo4j wikipedia"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "77718977-629e-46c2-b091-f9191b9ec569",
+ "metadata": {},
+ "source": [
+ "## Diffbot NLP Service\n",
+ "\n",
+ "Diffbot's NLP service is a tool for extracting entities, relationships, and semantic context from unstructured text data.\n",
+ "This extracted information can be used to construct a knowledge graph.\n",
+ "To use their service, you'll need to obtain an API key from [Diffbot](https://www.diffbot.com/products/natural-language/)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "2cbf97d0-3682-439b-8750-b695ff726789",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain_experimental.graph_transformers.diffbot import DiffbotGraphTransformer\n",
+ "\n",
+ "diffbot_api_key = \"DIFFBOT_API_KEY\"\n",
+ "diffbot_nlp = DiffbotGraphTransformer(diffbot_api_key=diffbot_api_key)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "5e3b894a-e3ee-46c7-8116-f8377f8f0159",
+ "metadata": {},
+ "source": [
+ "This code fetches Wikipedia articles about \"Warren Buffett\" and then uses `DiffbotGraphTransformer` to extract entities and relationships.\n",
+ "The `DiffbotGraphTransformer` outputs a structured data `GraphDocument`, which can be used to populate a graph database.\n",
+ "Note that text chunking is avoided due to Diffbot's [character limit per API request](https://docs.diffbot.com/reference/introduction-to-natural-language-api)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "53f8df86-47a1-44a1-9a0f-6725b90703bc",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.document_loaders import WikipediaLoader\n",
+ "\n",
+ "query = \"Warren Buffett\"\n",
+ "raw_documents = WikipediaLoader(query=query).load()\n",
+ "graph_documents = diffbot_nlp.convert_to_graph_documents(raw_documents)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "31bb851a-aab4-4b97-a6b7-fce397d32b47",
+ "metadata": {},
+ "source": [
+ "## Loading the data into a knowledge graph\n",
+ "\n",
+ "You will need to have a running Neo4j instance. One option is to create a [free Neo4j database instance in their Aura cloud service](https://neo4j.com/cloud/platform/aura-graph-database/). You can also run the database locally using the [Neo4j Desktop application](https://neo4j.com/download/), or running a docker container. You can run a local docker container by running the executing the following script:\n",
+ "```\n",
+ "docker run \\\n",
+ " --name neo4j \\\n",
+ " -p 7474:7474 -p 7687:7687 \\\n",
+ " -d \\\n",
+ " -e NEO4J_AUTH=neo4j/pleaseletmein \\\n",
+ " -e NEO4J_PLUGINS=\\[\\\"apoc\\\"\\] \\\n",
+ " neo4j:latest\n",
+ "``` \n",
+ "If you are using the docker container, you need to wait a couple of second for the database to start."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "0b2b6641-5a5d-467c-b148-e6aad5e4baa7",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.graphs import Neo4jGraph\n",
+ "\n",
+ "url=\"bolt://localhost:7687\"\n",
+ "username=\"neo4j\"\n",
+ "password=\"pleaseletmein\"\n",
+ "\n",
+ "graph = Neo4jGraph(\n",
+ " url=url,\n",
+ " username=username, \n",
+ " password=password\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0b15e840-fe6f-45db-9193-1b4e2df5c12c",
+ "metadata": {},
+ "source": [
+ "The `GraphDocuments` can be loaded into a knowledge graph using the `add_graph_documents` method."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "1a67c4a8-955c-42a2-9c5d-de3ac0e640ec",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "graph.add_graph_documents(graph_documents)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ed411e05-2b03-460d-997e-938482774f40",
+ "metadata": {},
+ "source": [
+ "## Refresh graph schema information\n",
+ "If the schema of database changes, you can refresh the schema information needed to generate Cypher statements"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "904c9ee3-787c-403f-857d-459ce5ad5a1b",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "graph.refresh_schema()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f19d1387-5899-4258-8c94-8ef5fa7db464",
+ "metadata": {},
+ "source": [
+ "## Querying the graph\n",
+ "We can now use the graph cypher QA chain to ask question of the graph. It is advisable to use **gpt-4** to construct Cypher queries to get the best experience."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "9393b732-67c8-45c1-9ec2-089f49c62448",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chains import GraphCypherQAChain\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "\n",
+ "chain = GraphCypherQAChain.from_llm(\n",
+ " cypher_llm=ChatOpenAI(temperature=0, model_name=\"gpt-4\"),\n",
+ " qa_llm=ChatOpenAI(temperature=0, model_name=\"gpt-3.5-turbo\"),\n",
+ " graph=graph, verbose=True,\n",
+ " \n",
+ ")\n",
+ " "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "id": "1a9b3652-b436-404d-aa25-5fb576f23dc0",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new GraphCypherQAChain chain...\u001b[0m\n",
+ "Generated Cypher:\n",
+ "\u001b[32;1m\u001b[1;3mMATCH (p:Person {name: \"Warren Buffett\"})-[:EDUCATED_AT]->(o:Organization)\n",
+ "RETURN o.name\u001b[0m\n",
+ "Full Context:\n",
+ "\u001b[32;1m\u001b[1;3m[{'o.name': 'New York Institute of Finance'}, {'o.name': 'Alice Deal Junior High School'}, {'o.name': 'Woodrow Wilson High School'}, {'o.name': 'University of Nebraska'}]\u001b[0m\n",
+ "\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'Warren Buffett attended the University of Nebraska.'"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.run(\"Which university did Warren Buffett attend?\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "id": "adc0ba0f-a62c-4875-89ce-da717f3ab148",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new GraphCypherQAChain chain...\u001b[0m\n",
+ "Generated Cypher:\n",
+ "\u001b[32;1m\u001b[1;3mMATCH (p:Person)-[r:EMPLOYEE_OR_MEMBER_OF]->(o:Organization) WHERE o.name = 'Berkshire Hathaway' RETURN p.name\u001b[0m\n",
+ "Full Context:\n",
+ "\u001b[32;1m\u001b[1;3m[{'p.name': 'Charlie Munger'}, {'p.name': 'Oliver Chace'}, {'p.name': 'Howard Buffett'}, {'p.name': 'Howard'}, {'p.name': 'Susan Buffett'}, {'p.name': 'Warren Buffett'}]\u001b[0m\n",
+ "\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'Charlie Munger, Oliver Chace, Howard Buffett, Susan Buffett, and Warren Buffett are or were working at Berkshire Hathaway.'"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chain.run(\"Who is or was working at Berkshire Hathaway?\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "d636954b-d967-4e96-9489-92e11c74af35",
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.11.4"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/use_cases/qa_structured/_category_.yml b/docs/extras/use_cases/qa_structured/_category_.yml
new file mode 100644
index 0000000000..209e3895ff
--- /dev/null
+++ b/docs/extras/use_cases/qa_structured/_category_.yml
@@ -0,0 +1,3 @@
+label: 'QA over structured data'
+collapsed: false
+position: 0.5
diff --git a/docs/extras/use_cases/qa_structured/integrations/_category_.yml b/docs/extras/use_cases/qa_structured/integrations/_category_.yml
new file mode 100644
index 0000000000..4a4b0b2f28
--- /dev/null
+++ b/docs/extras/use_cases/qa_structured/integrations/_category_.yml
@@ -0,0 +1 @@
+label: 'Integration-specific'
diff --git a/docs/extras/use_cases/qa_structured/integrations/elasticsearch.ipynb b/docs/extras/use_cases/qa_structured/integrations/elasticsearch.ipynb
new file mode 100644
index 0000000000..e28bc6bf61
--- /dev/null
+++ b/docs/extras/use_cases/qa_structured/integrations/elasticsearch.ipynb
@@ -0,0 +1,158 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Elasticsearch\n",
+ "\n",
+ "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/qa_structured/integrations/elasticsearch.ipynb)\n",
+ "\n",
+ "We can use LLMs to interact with Elasticsearch analytics databases in natural language.\n",
+ "\n",
+ "This chain builds search queries via the Elasticsearch DSL API (filters and aggregations).\n",
+ "\n",
+ "The Elasticsearch client must have permissions for index listing, mapping description and search queries.\n",
+ "\n",
+ "See [here](https://www.elastic.co/guide/en/elasticsearch/reference/current/docker.html) for instructions on how to run Elasticsearch locally."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "! pip install langchain langchain-experimental openai elasticsearch\n",
+ "\n",
+ "# Set env var OPENAI_API_KEY or load from a .env file\n",
+ "# import dotenv\n",
+ "\n",
+ "# dotenv.load_dotenv()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from elasticsearch import Elasticsearch\n",
+ "\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.chains.elasticsearch_database import ElasticsearchDatabaseChain"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Initialize Elasticsearch python client.\n",
+ "# See https://elasticsearch-py.readthedocs.io/en/v8.8.2/api.html#elasticsearch.Elasticsearch\n",
+ "ELASTIC_SEARCH_SERVER = \"https://elastic:pass@localhost:9200\"\n",
+ "db = Elasticsearch(ELASTIC_SEARCH_SERVER)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Uncomment the next cell to initially populate your db."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# customers = [\n",
+ "# {\"firstname\": \"Jennifer\", \"lastname\": \"Walters\"},\n",
+ "# {\"firstname\": \"Monica\",\"lastname\":\"Rambeau\"},\n",
+ "# {\"firstname\": \"Carol\",\"lastname\":\"Danvers\"},\n",
+ "# {\"firstname\": \"Wanda\",\"lastname\":\"Maximoff\"},\n",
+ "# {\"firstname\": \"Jennifer\",\"lastname\":\"Takeda\"},\n",
+ "# ]\n",
+ "# for i, customer in enumerate(customers):\n",
+ "# db.create(index=\"customers\", document=customer, id=i)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "llm = ChatOpenAI(model_name=\"gpt-4\", temperature=0)\n",
+ "chain = ElasticsearchDatabaseChain.from_llm(llm=llm, database=db, verbose=True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "question = \"What are the first names of all the customers?\"\n",
+ "chain.run(question)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can customize the prompt."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chains.elasticsearch_database.prompts import DEFAULT_DSL_TEMPLATE\n",
+ "from langchain.prompts.prompt import PromptTemplate\n",
+ "\n",
+ "PROMPT_TEMPLATE = \"\"\"Given an input question, create a syntactically correct Elasticsearch query to run. Unless the user specifies in their question a specific number of examples they wish to obtain, always limit your query to at most {top_k} results. You can order the results by a relevant column to return the most interesting examples in the database.\n",
+ "\n",
+ "Unless told to do not query for all the columns from a specific index, only ask for a the few relevant columns given the question.\n",
+ "\n",
+ "Pay attention to use only the column names that you can see in the mapping description. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which index. Return the query as valid json.\n",
+ "\n",
+ "Use the following format:\n",
+ "\n",
+ "Question: Question here\n",
+ "ESQuery: Elasticsearch Query formatted as json\n",
+ "\"\"\"\n",
+ "\n",
+ "PROMPT = PromptTemplate.from_template(\n",
+ " PROMPT_TEMPLATE,\n",
+ ")\n",
+ "chain = ElasticsearchDatabaseChain.from_llm(llm=llm, database=db, query_prompt=PROMPT)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/docs/extras/use_cases/qa_structured/integrations/myscale_vector_sql.ipynb b/docs/extras/use_cases/qa_structured/integrations/myscale_vector_sql.ipynb
new file mode 100644
index 0000000000..65bd8323ed
--- /dev/null
+++ b/docs/extras/use_cases/qa_structured/integrations/myscale_vector_sql.ipynb
@@ -0,0 +1,200 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "245065c6",
+ "metadata": {},
+ "source": [
+ "# Vector SQL Retriever with MyScale\n",
+ "\n",
+ ">[MyScale](https://docs.myscale.com/en/) is an integrated vector database. You can access your database in SQL and also from here, LangChain. MyScale can make a use of [various data types and functions for filters](https://blog.myscale.com/2023/06/06/why-integrated-database-solution-can-boost-your-llm-apps/#filter-on-anything-without-constraints). It will boost up your LLM app no matter if you are scaling up your data or expand your system to broader application."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "0246c5bf",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip3 install clickhouse-sqlalchemy InstructorEmbedding sentence_transformers openai langchain-experimental"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "7585d2c3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "\n",
+ "from os import environ\n",
+ "import getpass\n",
+ "from typing import Dict, Any\n",
+ "from langchain import OpenAI, SQLDatabase, LLMChain\n",
+ "from langchain_experimental.sql.vector_sql import VectorSQLDatabaseChain\n",
+ "from sqlalchemy import create_engine, Column, MetaData\n",
+ "from langchain import PromptTemplate\n",
+ "\n",
+ "\n",
+ "from sqlalchemy import create_engine\n",
+ "\n",
+ "MYSCALE_HOST = \"msc-1decbcc9.us-east-1.aws.staging.myscale.cloud\"\n",
+ "MYSCALE_PORT = 443\n",
+ "MYSCALE_USER = \"chatdata\"\n",
+ "MYSCALE_PASSWORD = \"myscale_rocks\"\n",
+ "OPENAI_API_KEY = getpass.getpass(\"OpenAI API Key:\")\n",
+ "\n",
+ "engine = create_engine(\n",
+ " f\"clickhouse://{MYSCALE_USER}:{MYSCALE_PASSWORD}@{MYSCALE_HOST}:{MYSCALE_PORT}/default?protocol=https\"\n",
+ ")\n",
+ "metadata = MetaData(bind=engine)\n",
+ "environ[\"OPENAI_API_KEY\"] = OPENAI_API_KEY"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "e08d9ddc",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.embeddings import HuggingFaceInstructEmbeddings\n",
+ "from langchain_experimental.sql.vector_sql import VectorSQLOutputParser\n",
+ "\n",
+ "output_parser = VectorSQLOutputParser.from_embeddings(\n",
+ " model=HuggingFaceInstructEmbeddings(\n",
+ " model_name=\"hkunlp/instructor-xl\", model_kwargs={\"device\": \"cpu\"}\n",
+ " )\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "84b705b2",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "\n",
+ "from langchain.llms import OpenAI\n",
+ "from langchain.callbacks import StdOutCallbackHandler\n",
+ "\n",
+ "from langchain.utilities.sql_database import SQLDatabase\n",
+ "from langchain_experimental.sql.prompt import MYSCALE_PROMPT\n",
+ "from langchain_experimental.sql.vector_sql import VectorSQLDatabaseChain\n",
+ "\n",
+ "chain = VectorSQLDatabaseChain(\n",
+ " llm_chain=LLMChain(\n",
+ " llm=OpenAI(openai_api_key=OPENAI_API_KEY, temperature=0),\n",
+ " prompt=MYSCALE_PROMPT,\n",
+ " ),\n",
+ " top_k=10,\n",
+ " return_direct=True,\n",
+ " sql_cmd_parser=output_parser,\n",
+ " database=SQLDatabase(engine, None, metadata),\n",
+ ")\n",
+ "\n",
+ "import pandas as pd\n",
+ "\n",
+ "pd.DataFrame(\n",
+ " chain.run(\n",
+ " \"Please give me 10 papers to ask what is PageRank?\",\n",
+ " callbacks=[StdOutCallbackHandler()],\n",
+ " )\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "6c09cda0",
+ "metadata": {},
+ "source": [
+ "## SQL Database as Retriever"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "734d7ff5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.chains.qa_with_sources.retrieval import RetrievalQAWithSourcesChain\n",
+ "\n",
+ "from langchain_experimental.sql.vector_sql import VectorSQLDatabaseChain\n",
+ "from langchain_experimental.retrievers.vector_sql_database \\\n",
+ " import VectorSQLDatabaseChainRetriever\n",
+ "from langchain_experimental.sql.prompt import MYSCALE_PROMPT\n",
+ "from langchain_experimental.sql.vector_sql import VectorSQLRetrieveAllOutputParser\n",
+ "\n",
+ "output_parser_retrieve_all = VectorSQLRetrieveAllOutputParser.from_embeddings(\n",
+ " output_parser.model\n",
+ ")\n",
+ "\n",
+ "chain = VectorSQLDatabaseChain.from_llm(\n",
+ " llm=OpenAI(openai_api_key=OPENAI_API_KEY, temperature=0),\n",
+ " prompt=MYSCALE_PROMPT,\n",
+ " top_k=10,\n",
+ " return_direct=True,\n",
+ " db=SQLDatabase(engine, None, metadata),\n",
+ " sql_cmd_parser=output_parser_retrieve_all,\n",
+ " native_format=True,\n",
+ ")\n",
+ "\n",
+ "# You need all those keys to get docs\n",
+ "retriever = VectorSQLDatabaseChainRetriever(sql_db_chain=chain, page_content_key=\"abstract\")\n",
+ "\n",
+ "document_with_metadata_prompt = PromptTemplate(\n",
+ " input_variables=[\"page_content\", \"id\", \"title\", \"authors\", \"pubdate\", \"categories\"],\n",
+ " template=\"Content:\\n\\tTitle: {title}\\n\\tAbstract: {page_content}\\n\\tAuthors: {authors}\\n\\tDate of Publication: {pubdate}\\n\\tCategories: {categories}\\nSOURCE: {id}\",\n",
+ ")\n",
+ "\n",
+ "chain = RetrievalQAWithSourcesChain.from_chain_type(\n",
+ " ChatOpenAI(\n",
+ " model_name=\"gpt-3.5-turbo-16k\", openai_api_key=OPENAI_API_KEY, temperature=0.6\n",
+ " ),\n",
+ " retriever=retriever,\n",
+ " chain_type=\"stuff\",\n",
+ " chain_type_kwargs={\n",
+ " \"document_prompt\": document_with_metadata_prompt,\n",
+ " },\n",
+ " return_source_documents=True,\n",
+ ")\n",
+ "ans = chain(\"Please give me 10 papers to ask what is PageRank?\",\n",
+ " callbacks=[StdOutCallbackHandler()])\n",
+ "print(ans[\"answer\"])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "4948ff25",
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.11.3"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/extras/use_cases/qa_structured/sql.ipynb b/docs/extras/use_cases/qa_structured/sql.ipynb
new file mode 100644
index 0000000000..23bde6a2a5
--- /dev/null
+++ b/docs/extras/use_cases/qa_structured/sql.ipynb
@@ -0,0 +1,1268 @@
+{
+ "cells": [
+ {
+ "cell_type": "raw",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "title: SQL\n",
+ "sidebar_position: 2\n",
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/qa_structured/sql.ipynb)\n",
+ "\n",
+ "## Use case\n",
+ "\n",
+ "Enterprise data is often stored in SQL databases.\n",
+ "\n",
+ "LLMs make it possible to interact with SQL databases using natural langugae.\n",
+ "\n",
+ "LangChain offers SQL Chains and Agents to build and run SQL queries based on natural language prompts. \n",
+ "\n",
+ "These are compatible with any SQL dialect supported by SQLAlchemy (e.g., MySQL, PostgreSQL, Oracle SQL, Databricks, SQLite).\n",
+ "\n",
+ "They enable use cases such as:\n",
+ "\n",
+ "- Generating queries that will be run based on natural language questions\n",
+ "- Creating chatbots that can answer questions based on database data\n",
+ "- Building custom dashboards based on insights a user wants to analyze\n",
+ "\n",
+ "## Overview\n",
+ "\n",
+ "LangChain provides tools to interact with SQL Databases:\n",
+ "\n",
+ "1. `Build SQL queries` based on natural language user questions\n",
+ "2. `Query a SQL database` using chains for query creation and execution\n",
+ "3. `Interact with a SQL database` using agents for robust and flexible querying \n",
+ "\n",
+ "\n",
+ "\n",
+ "## Quickstart\n",
+ "\n",
+ "First, get required packages and set environment variables:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "! pip install langchain langchain-experimental openai\n",
+ "\n",
+ "# Set env var OPENAI_API_KEY or load from a .env file\n",
+ "# import dotenv\n",
+ "\n",
+ "# dotenv.load_dotenv()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The below example will use a SQLite connection with Chinook database. \n",
+ " \n",
+ "Follow [installation steps](https://database.guide/2-sample-databases-sqlite/) to create `Chinook.db` in the same directory as this notebook:\n",
+ "\n",
+ "* Save [this file](https://raw.githubusercontent.com/lerocha/chinook-database/master/ChinookDatabase/DataSources/Chinook_Sqlite.sql) to the directory as `Chinook_Sqlite.sql`\n",
+ "* Run `sqlite3 Chinook.db`\n",
+ "* Run `.read Chinook_Sqlite.sql`\n",
+ "* Test `SELECT * FROM Artist LIMIT 10;`\n",
+ "\n",
+ "Now, `Chinhook.db` is in our directory.\n",
+ "\n",
+ "Let's create a `SQLDatabaseChain` to create and execute SQL queries."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.utilities import SQLDatabase\n",
+ "from langchain.llms import OpenAI\n",
+ "from langchain_experimental.sql import SQLDatabaseChain\n",
+ "\n",
+ "db = SQLDatabase.from_uri(\"sqlite:///Chinook.db\")\n",
+ "llm = OpenAI(temperature=0, verbose=True)\n",
+ "db_chain = SQLDatabaseChain.from_llm(llm, db, verbose=True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new SQLDatabaseChain chain...\u001b[0m\n",
+ "How many employees are there?\n",
+ "SQLQuery:\u001b[32;1m\u001b[1;3mSELECT COUNT(*) FROM \"Employee\";\u001b[0m\n",
+ "SQLResult: \u001b[33;1m\u001b[1;3m[(8,)]\u001b[0m\n",
+ "Answer:\u001b[32;1m\u001b[1;3mThere are 8 employees.\u001b[0m\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'There are 8 employees.'"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "db_chain.run(\"How many employees are there?\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Note that this both creates and executes the query. \n",
+ "\n",
+ "In the following sections, we will cover the 3 different use cases mentioned in the overview.\n",
+ "\n",
+ "### Go deeper\n",
+ "\n",
+ "You can load tabular data from other sources other than SQL Databases.\n",
+ "For example:\n",
+ "- [Loading a CSV file](/docs/integrations/document_loaders/csv)\n",
+ "- [Loading a Pandas DataFrame](/docs/integrations/document_loaders/pandas_dataframe)\n",
+ "Here you can [check full list of Document Loaders](/docs/integrations/document_loaders/)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Case 1: Text-to-SQL query\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.chains import create_sql_query_chain"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Let's create the chain that will build the SQL Query:\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "SELECT COUNT(*) FROM Employee\n"
+ ]
+ }
+ ],
+ "source": [
+ "chain = create_sql_query_chain(ChatOpenAI(temperature=0), db)\n",
+ "response = chain.invoke({\"question\":\"How many employees are there\"})\n",
+ "print(response)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "After building the SQL query based on a user question, we can execute the query:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'[(8,)]'"
+ ]
+ },
+ "execution_count": 10,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "db.run(response)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "As we can see, the SQL Query Builder chain **only created** the query, and we handled the **query execution separately**."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Go deeper\n",
+ "\n",
+ "**Looking under the hood**\n",
+ "\n",
+ "We can look at the [LangSmith trace](https://smith.langchain.com/public/c8fa52ea-be46-4829-bde2-52894970b830/r) to unpack this:\n",
+ "\n",
+ "[Some papers](https://arxiv.org/pdf/2204.00498.pdf) have reported good performance when prompting with:\n",
+ " \n",
+ "* A `CREATE TABLE` description for each table, which include column names, their types, etc\n",
+ "* Followed by three example rows in a `SELECT` statement\n",
+ "\n",
+ "`create_sql_query_chain` adopts this the best practice (see more in this [blog](https://blog.langchain.dev/llms-and-sql/)). \n",
+ "\n",
+ "\n",
+ "**Improvements**\n",
+ "\n",
+ "The query builder can be improved in several ways, such as (but not limited to):\n",
+ "\n",
+ "- Customizing database description to your specific use case\n",
+ "- Hardcoding a few examples of questions and their corresponding SQL query in the prompt\n",
+ "- Using a vector database to include dynamic examples that are relevant to the specific user question\n",
+ "\n",
+ "All these examples involve customizing the chain's prompt. \n",
+ "\n",
+ "For example, we can include a few examples in our prompt like so:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.prompts import PromptTemplate\n",
+ "\n",
+ "TEMPLATE = \"\"\"Given an input question, first create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer.\n",
+ "Use the following format:\n",
+ "\n",
+ "Question: \"Question here\"\n",
+ "SQLQuery: \"SQL Query to run\"\n",
+ "SQLResult: \"Result of the SQLQuery\"\n",
+ "Answer: \"Final answer here\"\n",
+ "\n",
+ "Only use the following tables:\n",
+ "\n",
+ "{table_info}.\n",
+ "\n",
+ "Some examples of SQL queries that corrsespond to questions are:\n",
+ "\n",
+ "{few_shot_examples}\n",
+ "\n",
+ "Question: {input}\"\"\"\n",
+ "\n",
+ "CUSTOM_PROMPT = PromptTemplate(\n",
+ " input_variables=[\"input\", \"few_shot_examples\", \"table_info\", \"dialect\"], template=TEMPLATE\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can also access this [prompt](https://smith.langchain.com/hub/rlm/text-to-sql) in the LangChain prompt hub.\n",
+ "\n",
+ "This will work with your [LangSmith API key](https://docs.smith.langchain.com/)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain import hub\n",
+ "CUSTOM_PROMPT = hub.pull(\"rlm/text-to-sql\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Case 2: Text-to-SQL query and execution\n",
+ "\n",
+ "We can use `SQLDatabaseChain` from `langchain_experimental` to create and run SQL queries."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.llms import OpenAI\n",
+ "from langchain_experimental.sql import SQLDatabaseChain\n",
+ "\n",
+ "llm = OpenAI(temperature=0, verbose=True)\n",
+ "db_chain = SQLDatabaseChain.from_llm(llm, db, verbose=True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new SQLDatabaseChain chain...\u001b[0m\n",
+ "How many employees are there?\n",
+ "SQLQuery:\u001b[32;1m\u001b[1;3mSELECT COUNT(*) FROM \"Employee\";\u001b[0m\n",
+ "SQLResult: \u001b[33;1m\u001b[1;3m[(8,)]\u001b[0m\n",
+ "Answer:\u001b[32;1m\u001b[1;3mThere are 8 employees.\u001b[0m\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'There are 8 employees.'"
+ ]
+ },
+ "execution_count": 13,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "db_chain.run(\"How many employees are there?\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "As we can see, we get the same result as the previous case.\n",
+ "\n",
+ "Here, the chain **also handles the query execution** and provides a final answer based on the user question and the query result.\n",
+ "\n",
+ "**Be careful** while using this approach as it is susceptible to `SQL Injection`:\n",
+ "\n",
+ "* The chain is executing queries that are created by an LLM, and weren't validated\n",
+ "* e.g. records may be created, modified or deleted unintentionally_\n",
+ "\n",
+ "This is why we see the `SQLDatabaseChain` is inside `langchain_experimental`."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Go deeper\n",
+ "\n",
+ "**Looking under the hood**\n",
+ "\n",
+ "We can use the [LangSmith trace](https://smith.langchain.com/public/7f202a0c-1e35-42d6-a84a-6c2a58f697ef/r) to see what is happening under the hood:\n",
+ "\n",
+ "* As discussed above, first we create the query:\n",
+ "\n",
+ "```\n",
+ "text: ' SELECT COUNT(*) FROM \"Employee\";'\n",
+ "```\n",
+ "\n",
+ "* Then, it executes the query and passes the results to an LLM for synthesis.\n",
+ "\n",
+ "\n",
+ "\n",
+ "**Improvements**\n",
+ "\n",
+ "The performance of the `SQLDatabaseChain` can be enhanced in several ways:\n",
+ "\n",
+ "- [Adding sample rows](#adding-sample-rows)\n",
+ "- [Specifying custom table information](/docs/integrations/tools/sqlite#custom-table-info)\n",
+ "- [Using Query Checker](/docs/integrations/tools/sqlite#use-query-checker) self-correct invalid SQL using parameter `use_query_checker=True`\n",
+ "- [Customizing the LLM Prompt](/docs/integrations/tools/sqlite#customize-prompt) include specific instructions or relevant information, using parameter `prompt=CUSTOM_PROMPT`\n",
+ "- [Get intermediate steps](/docs/integrations/tools/sqlite#return-intermediate-steps) access the SQL statement as well as the final result using parameter `return_intermediate_steps=True`\n",
+ "- [Limit the number of rows](/docs/integrations/tools/sqlite#choosing-how-to-limit-the-number-of-rows-returned) a query will return using parameter `top_k=5`\n",
+ "\n",
+ "You might find [SQLDatabaseSequentialChain](/docs/integrations/tools/sqlite#sqldatabasesequentialchain)\n",
+ "useful for cases in which the number of tables in the database is large.\n",
+ "\n",
+ "This `Sequential Chain` handles the process of:\n",
+ "\n",
+ "1. Determining which tables to use based on the user question\n",
+ "2. Calling the normal SQL database chain using only relevant tables\n",
+ "\n",
+ "**Adding Sample Rows**\n",
+ "\n",
+ "Providing sample data can help the LLM construct correct queries when the data format is not obvious. \n",
+ "\n",
+ "For example, we can tell LLM that artists are saved with their full names by providing two rows from the Track table.\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "db = SQLDatabase.from_uri(\n",
+ " \"sqlite:///Chinook.db\",\n",
+ " include_tables=['Track'], # we include only one table to save tokens in the prompt :)\n",
+ " sample_rows_in_table_info=2)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The sample rows are added to the prompt after each corresponding table's column information.\n",
+ "\n",
+ "We can use `db.table_info` and check which sample rows are included:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "CREATE TABLE \"Track\" (\n",
+ "\t\"TrackId\" INTEGER NOT NULL, \n",
+ "\t\"Name\" NVARCHAR(200) NOT NULL, \n",
+ "\t\"AlbumId\" INTEGER, \n",
+ "\t\"MediaTypeId\" INTEGER NOT NULL, \n",
+ "\t\"GenreId\" INTEGER, \n",
+ "\t\"Composer\" NVARCHAR(220), \n",
+ "\t\"Milliseconds\" INTEGER NOT NULL, \n",
+ "\t\"Bytes\" INTEGER, \n",
+ "\t\"UnitPrice\" NUMERIC(10, 2) NOT NULL, \n",
+ "\tPRIMARY KEY (\"TrackId\"), \n",
+ "\tFOREIGN KEY(\"MediaTypeId\") REFERENCES \"MediaType\" (\"MediaTypeId\"), \n",
+ "\tFOREIGN KEY(\"GenreId\") REFERENCES \"Genre\" (\"GenreId\"), \n",
+ "\tFOREIGN KEY(\"AlbumId\") REFERENCES \"Album\" (\"AlbumId\")\n",
+ ")\n",
+ "\n",
+ "/*\n",
+ "2 rows from Track table:\n",
+ "TrackId\tName\tAlbumId\tMediaTypeId\tGenreId\tComposer\tMilliseconds\tBytes\tUnitPrice\n",
+ "1\tFor Those About To Rock (We Salute You)\t1\t1\t1\tAngus Young, Malcolm Young, Brian Johnson\t343719\t11170334\t0.99\n",
+ "2\tBalls to the Wall\t2\t2\t1\tNone\t342562\t5510424\t0.99\n",
+ "*/\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(db.table_info)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Case 3: SQL agents\n",
+ "\n",
+ "LangChain has an SQL Agent which provides a more flexible way of interacting with SQL Databases than the `SQLDatabaseChain`.\n",
+ "\n",
+ "The main advantages of using the SQL Agent are:\n",
+ "\n",
+ "- It can answer questions based on the databases' schema as well as on the databases' content (like describing a specific table)\n",
+ "- It can recover from errors by running a generated query, catching the traceback and regenerating it correctly\n",
+ "\n",
+ "To initialize the agent, we use `create_sql_agent` function. \n",
+ "\n",
+ "This agent contains the `SQLDatabaseToolkit` which contains tools to: \n",
+ "\n",
+ "* Create and execute queries\n",
+ "* Check query syntax\n",
+ "* Retrieve table descriptions\n",
+ "* ... and more"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.agents import create_sql_agent\n",
+ "from langchain.agents.agent_toolkits import SQLDatabaseToolkit\n",
+ "# from langchain.agents import AgentExecutor\n",
+ "from langchain.agents.agent_types import AgentType\n",
+ "\n",
+ "db = SQLDatabase.from_uri(\"sqlite:///Chinook.db\")\n",
+ "llm = OpenAI(temperature=0, verbose=True)\n",
+ "\n",
+ "agent_executor = create_sql_agent(\n",
+ " llm=OpenAI(temperature=0),\n",
+ " toolkit=SQLDatabaseToolkit(db=db, llm=OpenAI(temperature=0)),\n",
+ " verbose=True,\n",
+ " agent_type=AgentType.ZERO_SHOT_REACT_DESCRIPTION,\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Agent task example #1 - Running queries\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
+ "\u001b[32;1m\u001b[1;3mAction: sql_db_list_tables\n",
+ "Action Input: \u001b[0m\n",
+ "Observation: \u001b[38;5;200m\u001b[1;3mAlbum, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track\u001b[0m\n",
+ "Thought:\u001b[32;1m\u001b[1;3m I should query the schema of the Invoice and Customer tables.\n",
+ "Action: sql_db_schema\n",
+ "Action Input: Invoice, Customer\u001b[0m\n",
+ "Observation: \u001b[33;1m\u001b[1;3m\n",
+ "CREATE TABLE \"Customer\" (\n",
+ "\t\"CustomerId\" INTEGER NOT NULL, \n",
+ "\t\"FirstName\" NVARCHAR(40) NOT NULL, \n",
+ "\t\"LastName\" NVARCHAR(20) NOT NULL, \n",
+ "\t\"Company\" NVARCHAR(80), \n",
+ "\t\"Address\" NVARCHAR(70), \n",
+ "\t\"City\" NVARCHAR(40), \n",
+ "\t\"State\" NVARCHAR(40), \n",
+ "\t\"Country\" NVARCHAR(40), \n",
+ "\t\"PostalCode\" NVARCHAR(10), \n",
+ "\t\"Phone\" NVARCHAR(24), \n",
+ "\t\"Fax\" NVARCHAR(24), \n",
+ "\t\"Email\" NVARCHAR(60) NOT NULL, \n",
+ "\t\"SupportRepId\" INTEGER, \n",
+ "\tPRIMARY KEY (\"CustomerId\"), \n",
+ "\tFOREIGN KEY(\"SupportRepId\") REFERENCES \"Employee\" (\"EmployeeId\")\n",
+ ")\n",
+ "\n",
+ "/*\n",
+ "3 rows from Customer table:\n",
+ "CustomerId\tFirstName\tLastName\tCompany\tAddress\tCity\tState\tCountry\tPostalCode\tPhone\tFax\tEmail\tSupportRepId\n",
+ "1\tLuís\tGonçalves\tEmbraer - Empresa Brasileira de Aeronáutica S.A.\tAv. Brigadeiro Faria Lima, 2170\tSão José dos Campos\tSP\tBrazil\t12227-000\t+55 (12) 3923-5555\t+55 (12) 3923-5566\tluisg@embraer.com.br\t3\n",
+ "2\tLeonie\tKöhler\tNone\tTheodor-Heuss-Straße 34\tStuttgart\tNone\tGermany\t70174\t+49 0711 2842222\tNone\tleonekohler@surfeu.de\t5\n",
+ "3\tFrançois\tTremblay\tNone\t1498 rue Bélanger\tMontréal\tQC\tCanada\tH2G 1A7\t+1 (514) 721-4711\tNone\tftremblay@gmail.com\t3\n",
+ "*/\n",
+ "\n",
+ "\n",
+ "CREATE TABLE \"Invoice\" (\n",
+ "\t\"InvoiceId\" INTEGER NOT NULL, \n",
+ "\t\"CustomerId\" INTEGER NOT NULL, \n",
+ "\t\"InvoiceDate\" DATETIME NOT NULL, \n",
+ "\t\"BillingAddress\" NVARCHAR(70), \n",
+ "\t\"BillingCity\" NVARCHAR(40), \n",
+ "\t\"BillingState\" NVARCHAR(40), \n",
+ "\t\"BillingCountry\" NVARCHAR(40), \n",
+ "\t\"BillingPostalCode\" NVARCHAR(10), \n",
+ "\t\"Total\" NUMERIC(10, 2) NOT NULL, \n",
+ "\tPRIMARY KEY (\"InvoiceId\"), \n",
+ "\tFOREIGN KEY(\"CustomerId\") REFERENCES \"Customer\" (\"CustomerId\")\n",
+ ")\n",
+ "\n",
+ "/*\n",
+ "3 rows from Invoice table:\n",
+ "InvoiceId\tCustomerId\tInvoiceDate\tBillingAddress\tBillingCity\tBillingState\tBillingCountry\tBillingPostalCode\tTotal\n",
+ "1\t2\t2009-01-01 00:00:00\tTheodor-Heuss-Straße 34\tStuttgart\tNone\tGermany\t70174\t1.98\n",
+ "2\t4\t2009-01-02 00:00:00\tUllevålsveien 14\tOslo\tNone\tNorway\t0171\t3.96\n",
+ "3\t8\t2009-01-03 00:00:00\tGrétrystraat 63\tBrussels\tNone\tBelgium\t1000\t5.94\n",
+ "*/\u001b[0m\n",
+ "Thought:\u001b[32;1m\u001b[1;3m I should query the total sales per country.\n",
+ "Action: sql_db_query\n",
+ "Action Input: SELECT Country, SUM(Total) AS TotalSales FROM Invoice INNER JOIN Customer ON Invoice.CustomerId = Customer.CustomerId GROUP BY Country ORDER BY TotalSales DESC LIMIT 10\u001b[0m\n",
+ "Observation: \u001b[36;1m\u001b[1;3m[('USA', 523.0600000000003), ('Canada', 303.9599999999999), ('France', 195.09999999999994), ('Brazil', 190.09999999999997), ('Germany', 156.48), ('United Kingdom', 112.85999999999999), ('Czech Republic', 90.24000000000001), ('Portugal', 77.23999999999998), ('India', 75.25999999999999), ('Chile', 46.62)]\u001b[0m\n",
+ "Thought:\u001b[32;1m\u001b[1;3m I now know the final answer\n",
+ "Final Answer: The country with the highest total sales is the USA, with a total of $523.06.\u001b[0m\n",
+ "\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'The country with the highest total sales is the USA, with a total of $523.06.'"
+ ]
+ },
+ "execution_count": 15,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "agent_executor.run(\n",
+ " \"List the total sales per country. Which country's customers spent the most?\"\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Looking at the [LangSmith trace](https://smith.langchain.com/public/a86dbe17-5782-4020-bce6-2de85343168a/r), we can see:\n",
+ "\n",
+ "* The agent is using a ReAct style prompt\n",
+ "* First, it will look at the tables: `Action: sql_db_list_tables` using tool `sql_db_list_tables`\n",
+ "* Given the tables as an observation, it `thinks` and then determinates the next `action`:\n",
+ "\n",
+ "```\n",
+ "Observation: Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track\n",
+ "Thought: I should query the schema of the Invoice and Customer tables.\n",
+ "Action: sql_db_schema\n",
+ "Action Input: Invoice, Customer\n",
+ "```\n",
+ "\n",
+ "* It then formulates the query using the schema from tool `sql_db_schema`\n",
+ "\n",
+ "```\n",
+ "Thought: I should query the total sales per country.\n",
+ "Action: sql_db_query\n",
+ "Action Input: SELECT Country, SUM(Total) AS TotalSales FROM Invoice INNER JOIN Customer ON Invoice.CustomerId = Customer.CustomerId GROUP BY Country ORDER BY TotalSales DESC LIMIT 10\n",
+ "```\n",
+ "\n",
+ "* It finally executes the generated query using tool `sql_db_query`\n",
+ "\n",
+ ""
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Agent task example #2 - Describing a Table"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
+ "\u001b[32;1m\u001b[1;3mAction: sql_db_list_tables\n",
+ "Action Input: \u001b[0m\n",
+ "Observation: \u001b[38;5;200m\u001b[1;3mAlbum, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track\u001b[0m\n",
+ "Thought:\u001b[32;1m\u001b[1;3m The PlaylistTrack table is the most relevant to the question.\n",
+ "Action: sql_db_schema\n",
+ "Action Input: PlaylistTrack\u001b[0m\n",
+ "Observation: \u001b[33;1m\u001b[1;3m\n",
+ "CREATE TABLE \"PlaylistTrack\" (\n",
+ "\t\"PlaylistId\" INTEGER NOT NULL, \n",
+ "\t\"TrackId\" INTEGER NOT NULL, \n",
+ "\tPRIMARY KEY (\"PlaylistId\", \"TrackId\"), \n",
+ "\tFOREIGN KEY(\"TrackId\") REFERENCES \"Track\" (\"TrackId\"), \n",
+ "\tFOREIGN KEY(\"PlaylistId\") REFERENCES \"Playlist\" (\"PlaylistId\")\n",
+ ")\n",
+ "\n",
+ "/*\n",
+ "3 rows from PlaylistTrack table:\n",
+ "PlaylistId\tTrackId\n",
+ "1\t3402\n",
+ "1\t3389\n",
+ "1\t3390\n",
+ "*/\u001b[0m\n",
+ "Thought:\u001b[32;1m\u001b[1;3m I now know the final answer\n",
+ "Final Answer: The PlaylistTrack table contains two columns, PlaylistId and TrackId, which are both integers and form a primary key. It also has two foreign keys, one to the Track table and one to the Playlist table.\u001b[0m\n",
+ "\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'The PlaylistTrack table contains two columns, PlaylistId and TrackId, which are both integers and form a primary key. It also has two foreign keys, one to the Track table and one to the Playlist table.'"
+ ]
+ },
+ "execution_count": 14,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "agent_executor.run(\"Describe the playlisttrack table\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Extending the SQL Toolkit\n",
+ "\n",
+ "Although the out-of-the-box SQL Toolkit contains the necessary tools to start working on a database, it is often the case that some extra tools may be useful for extending the agent's capabilities. This is particularly useful when trying to use **domain specific knowledge** in the solution, in order to improve its overall performance.\n",
+ "\n",
+ "Some examples include:\n",
+ "\n",
+ "- Including dynamic few shot examples\n",
+ "- Finding misspellings in proper nouns to use as column filters\n",
+ "\n",
+ "We can create separate tools which tackle these specific use cases and include them as a complement to the standard SQL Toolkit. Let's see how to include these two custom tools.\n",
+ "\n",
+ "#### Including dynamic few-shot examples\n",
+ "\n",
+ "In order to include dynamic few-shot examples, we need a custom **Retriever Tool** that handles the vector database in order to retrieve the examples that are semantically similar to the user’s question.\n",
+ "\n",
+ "Let's start by creating a dictionary with some examples: "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# few_shots = {'List all artists.': 'SELECT * FROM artists;',\n",
+ "# \"Find all albums for the artist 'AC/DC'.\": \"SELECT * FROM albums WHERE ArtistId = (SELECT ArtistId FROM artists WHERE Name = 'AC/DC');\",\n",
+ "# \"List all tracks in the 'Rock' genre.\": \"SELECT * FROM tracks WHERE GenreId = (SELECT GenreId FROM genres WHERE Name = 'Rock');\",\n",
+ "# 'Find the total duration of all tracks.': 'SELECT SUM(Milliseconds) FROM tracks;',\n",
+ "# 'List all customers from Canada.': \"SELECT * FROM customers WHERE Country = 'Canada';\",\n",
+ "# 'How many tracks are there in the album with ID 5?': 'SELECT COUNT(*) FROM tracks WHERE AlbumId = 5;',\n",
+ "# 'Find the total number of invoices.': 'SELECT COUNT(*) FROM invoices;',\n",
+ "# 'List all tracks that are longer than 5 minutes.': 'SELECT * FROM tracks WHERE Milliseconds > 300000;',\n",
+ "# 'Who are the top 5 customers by total purchase?': 'SELECT CustomerId, SUM(Total) AS TotalPurchase FROM invoices GROUP BY CustomerId ORDER BY TotalPurchase DESC LIMIT 5;',\n",
+ "# 'Which albums are from the year 2000?': \"SELECT * FROM albums WHERE strftime('%Y', ReleaseDate) = '2000';\",\n",
+ "# 'How many employees are there': 'SELECT COUNT(*) FROM \"employee\"'\n",
+ "# }"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can then create a retriever using the list of questions, assigning the target SQL query as metadata:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.embeddings.openai import OpenAIEmbeddings\n",
+ "from langchain.vectorstores import FAISS\n",
+ "from langchain.schema import Document\n",
+ "\n",
+ "embeddings = OpenAIEmbeddings()\n",
+ "\n",
+ "few_shot_docs = [Document(page_content=question, metadata={'sql_query': few_shots[question]}) for question in few_shots.keys()]\n",
+ "vector_db = FAISS.from_documents(few_shot_docs, embeddings)\n",
+ "retriever = vector_db.as_retriever()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now we can create our own custom tool and append it as a new tool in the `create_sql_agent` function:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.agents.agent_toolkits import create_retriever_tool\n",
+ "\n",
+ "tool_description = \"\"\"\n",
+ "This tool will help you understand similar examples to adapt them to the user question.\n",
+ "Input to this tool should be the user question.\n",
+ "\"\"\"\n",
+ "\n",
+ "retriever_tool = create_retriever_tool(\n",
+ " retriever,\n",
+ " name='sql_get_similar_examples',\n",
+ " description=tool_description\n",
+ " )\n",
+ "custom_tool_list = [retriever_tool]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now we can create the agent, adjusting the standard SQL Agent suffix to consider our use case. Although the most straightforward way to handle this would be to include it just in the tool description, this is often not enough and we need to specify it in the agent prompt using the `suffix` argument in the constructor."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.agents import create_sql_agent, AgentType\n",
+ "from langchain.agents.agent_toolkits import SQLDatabaseToolkit\n",
+ "from langchain.utilities import SQLDatabase\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "\n",
+ "db = SQLDatabase.from_uri(\"sqlite:///Chinook.db\")\n",
+ "llm = ChatOpenAI(model_name='gpt-4',temperature=0)\n",
+ "\n",
+ "toolkit = SQLDatabaseToolkit(db=db, llm=llm)\n",
+ "\n",
+ "custom_suffix = \"\"\"\n",
+ "I should first get the similar examples I know.\n",
+ "If the examples are enough to construct the query, I can build it.\n",
+ "Otherwise, I can then look at the tables in the database to see what I can query.\n",
+ "Then I should query the schema of the most relevant tables\n",
+ "\"\"\"\n",
+ "\n",
+ "agent = create_sql_agent(llm=llm,\n",
+ " toolkit=toolkit,\n",
+ " verbose=True,\n",
+ " agent_type=AgentType.OPENAI_FUNCTIONS,\n",
+ " extra_tools=custom_tool_list,\n",
+ " suffix=custom_suffix\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Let's try it out:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
+ "\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_get_similar_examples` with `How many employees do we have?`\n",
+ "\n",
+ "\n",
+ "\u001b[0m\u001b[33;1m\u001b[1;3m[Document(page_content='How many employees are there', metadata={'sql_query': 'SELECT COUNT(*) FROM \"employee\"'}), Document(page_content='Find the total number of invoices.', metadata={'sql_query': 'SELECT COUNT(*) FROM invoices;'})]\u001b[0m\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_db_query_checker` with `SELECT COUNT(*) FROM employee`\n",
+ "responded: {content}\n",
+ "\n",
+ "\u001b[0m\u001b[36;1m\u001b[1;3mSELECT COUNT(*) FROM employee\u001b[0m\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_db_query` with `SELECT COUNT(*) FROM employee`\n",
+ "\n",
+ "\n",
+ "\u001b[0m\u001b[36;1m\u001b[1;3m[(8,)]\u001b[0m\u001b[32;1m\u001b[1;3mWe have 8 employees.\u001b[0m\n",
+ "\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'We have 8 employees.'"
+ ]
+ },
+ "execution_count": 23,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "agent.run(\"How many employees do we have?\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "As we can see, the agent first used the `sql_get_similar_examples` tool in order to retrieve similar examples. As the question was very similar to other few shot examples, the agent **didn't need to use any other tool** from the standard Toolkit, thus **saving time and tokens**."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Finding and correcting misspellings for proper nouns\n",
+ "\n",
+ "In order to filter columns that contain proper nouns such as addresses, song names or artists, we first need to double-check the spelling in order to filter the data correctly. \n",
+ "\n",
+ "We can achieve this by creating a vector store using all the distinct proper nouns that exist in the database. We can then have the agent query that vector store each time the user includes a proper noun in their question, to find the correct spelling for that word. In this way, the agent can make sure it understands which entity the user is referring to before building the target query.\n",
+ "\n",
+ "Let's follow a similar approach to the few shots, but without metadata: just embedding the proper nouns and then querying to get the most similar one to the misspelled user question.\n",
+ "\n",
+ "First we need the unique values for each entity we want, for which we define a function that parses the result into a list of elements:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 37,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import ast\n",
+ "import re\n",
+ "\n",
+ "def run_query_save_results(db, query):\n",
+ " res = db.run(query)\n",
+ " res = [el for sub in ast.literal_eval(res) for el in sub if el]\n",
+ " res = [re.sub(r'\\b\\d+\\b', '', string).strip() for string in res]\n",
+ " return res\n",
+ "\n",
+ "artists = run_query_save_results(db, \"SELECT Name FROM Artist\")\n",
+ "albums = run_query_save_results(db, \"SELECT Title FROM Album\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Now we can proceed with creating the custom **retreiver tool** and the final agent:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 51,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.agents.agent_toolkits import create_retriever_tool\n",
+ "from langchain.embeddings.openai import OpenAIEmbeddings\n",
+ "from langchain.vectorstores import FAISS\n",
+ "\n",
+ "\n",
+ "texts = (artists + albums)\n",
+ "\n",
+ "embeddings = OpenAIEmbeddings()\n",
+ "vector_db = FAISS.from_texts(texts, embeddings)\n",
+ "retriever = vector_db.as_retriever()\n",
+ "\n",
+ "retriever_tool = create_retriever_tool(\n",
+ " retriever,\n",
+ " name='name_search',\n",
+ " description='use to learn how a piece of data is actually written, can be from names, surnames addresses etc'\n",
+ " )\n",
+ "\n",
+ "custom_tool_list = [retriever_tool]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 54,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.agents import create_sql_agent, AgentType\n",
+ "from langchain.agents.agent_toolkits import SQLDatabaseToolkit\n",
+ "from langchain.utilities import SQLDatabase\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "\n",
+ "# db = SQLDatabase.from_uri(\"sqlite:///Chinook.db\")\n",
+ "llm = ChatOpenAI(model_name='gpt-4', temperature=0)\n",
+ "\n",
+ "toolkit = SQLDatabaseToolkit(db=db, llm=llm)\n",
+ "\n",
+ "custom_suffix = \"\"\"\n",
+ "If a user asks for me to filter based on proper nouns, I should first check the spelling using the name_search tool.\n",
+ "Otherwise, I can then look at the tables in the database to see what I can query.\n",
+ "Then I should query the schema of the most relevant tables\n",
+ "\"\"\"\n",
+ "\n",
+ "agent = create_sql_agent(llm=llm,\n",
+ " toolkit=toolkit,\n",
+ " verbose=True,\n",
+ " agent_type=AgentType.OPENAI_FUNCTIONS,\n",
+ " extra_tools=custom_tool_list,\n",
+ " suffix=custom_suffix\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Let's try it out:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 55,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
+ "\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `name_search` with `alis in pains`\n",
+ "\n",
+ "\n",
+ "\u001b[0m\u001b[33;1m\u001b[1;3m[Document(page_content='House of Pain', metadata={}), Document(page_content='Alice In Chains', metadata={}), Document(page_content='Aisha Duo', metadata={}), Document(page_content='House Of Pain', metadata={})]\u001b[0m\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_db_list_tables` with ``\n",
+ "responded: {content}\n",
+ "\n",
+ "\u001b[0m\u001b[38;5;200m\u001b[1;3mAlbum, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track\u001b[0m\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_db_schema` with `Album, Artist`\n",
+ "responded: {content}\n",
+ "\n",
+ "\u001b[0m\u001b[33;1m\u001b[1;3m\n",
+ "CREATE TABLE \"Album\" (\n",
+ "\t\"AlbumId\" INTEGER NOT NULL, \n",
+ "\t\"Title\" NVARCHAR(160) NOT NULL, \n",
+ "\t\"ArtistId\" INTEGER NOT NULL, \n",
+ "\tPRIMARY KEY (\"AlbumId\"), \n",
+ "\tFOREIGN KEY(\"ArtistId\") REFERENCES \"Artist\" (\"ArtistId\")\n",
+ ")\n",
+ "\n",
+ "/*\n",
+ "3 rows from Album table:\n",
+ "AlbumId\tTitle\tArtistId\n",
+ "1\tFor Those About To Rock We Salute You\t1\n",
+ "2\tBalls to the Wall\t2\n",
+ "3\tRestless and Wild\t2\n",
+ "*/\n",
+ "\n",
+ "\n",
+ "CREATE TABLE \"Artist\" (\n",
+ "\t\"ArtistId\" INTEGER NOT NULL, \n",
+ "\t\"Name\" NVARCHAR(120), \n",
+ "\tPRIMARY KEY (\"ArtistId\")\n",
+ ")\n",
+ "\n",
+ "/*\n",
+ "3 rows from Artist table:\n",
+ "ArtistId\tName\n",
+ "1\tAC/DC\n",
+ "2\tAccept\n",
+ "3\tAerosmith\n",
+ "*/\u001b[0m\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_db_query_checker` with `SELECT COUNT(*) FROM Album JOIN Artist ON Album.ArtistId = Artist.ArtistId WHERE Artist.Name = 'Alice In Chains'`\n",
+ "responded: {content}\n",
+ "\n",
+ "\u001b[0m\u001b[36;1m\u001b[1;3mSELECT COUNT(*) FROM Album JOIN Artist ON Album.ArtistId = Artist.ArtistId WHERE Artist.Name = 'Alice In Chains'\u001b[0m\u001b[32;1m\u001b[1;3m\n",
+ "Invoking: `sql_db_query` with `SELECT COUNT(*) FROM Album JOIN Artist ON Album.ArtistId = Artist.ArtistId WHERE Artist.Name = 'Alice In Chains'`\n",
+ "\n",
+ "\n",
+ "\u001b[0m\u001b[36;1m\u001b[1;3m[(1,)]\u001b[0m\u001b[32;1m\u001b[1;3mAlice In Chains has 1 album in the database.\u001b[0m\n",
+ "\n",
+ "\u001b[1m> Finished chain.\u001b[0m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'Alice In Chains has 1 album in the database.'"
+ ]
+ },
+ "execution_count": 55,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "agent.run(\"How many albums does alis in pains have?\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "As we can see, the agent used the `name_search` tool in order to check how to correctly query the database for this specific artist."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Go deeper\n",
+ "\n",
+ "To learn more about the SQL Agent and how it works we refer to the [SQL Agent Toolkit](/docs/integrations/toolkits/sql_database) documentation.\n",
+ "\n",
+ "You can also check Agents for other document types:\n",
+ "- [Pandas Agent](/docs/integrations/toolkits/pandas.html)\n",
+ "- [CSV Agent](/docs/integrations/toolkits/csv.html)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Elastic Search\n",
+ "\n",
+ "Going beyond the above use-case, there are integrations with other databases.\n",
+ "\n",
+ "For example, we can interact with Elasticsearch analytics database. \n",
+ "\n",
+ "This chain builds search queries via the Elasticsearch DSL API (filters and aggregations).\n",
+ "\n",
+ "The Elasticsearch client must have permissions for index listing, mapping description and search queries.\n",
+ "\n",
+ "See [here](https://www.elastic.co/guide/en/elasticsearch/reference/current/docker.html) for instructions on how to run Elasticsearch locally.\n",
+ "\n",
+ "Make sure to install the Elasticsearch Python client before:\n",
+ "\n",
+ "```sh\n",
+ "pip install elasticsearch\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from elasticsearch import Elasticsearch\n",
+ "from langchain.chat_models import ChatOpenAI\n",
+ "from langchain.chains.elasticsearch_database import ElasticsearchDatabaseChain"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Initialize Elasticsearch python client.\n",
+ "# See https://elasticsearch-py.readthedocs.io/en/v8.8.2/api.html#elasticsearch.Elasticsearch\n",
+ "ELASTIC_SEARCH_SERVER = \"https://elastic:pass@localhost:9200\"\n",
+ "db = Elasticsearch(ELASTIC_SEARCH_SERVER)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Uncomment the next cell to initially populate your db."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# customers = [\n",
+ "# {\"firstname\": \"Jennifer\", \"lastname\": \"Walters\"},\n",
+ "# {\"firstname\": \"Monica\",\"lastname\":\"Rambeau\"},\n",
+ "# {\"firstname\": \"Carol\",\"lastname\":\"Danvers\"},\n",
+ "# {\"firstname\": \"Wanda\",\"lastname\":\"Maximoff\"},\n",
+ "# {\"firstname\": \"Jennifer\",\"lastname\":\"Takeda\"},\n",
+ "# ]\n",
+ "# for i, customer in enumerate(customers):\n",
+ "# db.create(index=\"customers\", document=customer, id=i)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "llm = ChatOpenAI(model_name=\"gpt-4\", temperature=0)\n",
+ "chain = ElasticsearchDatabaseChain.from_llm(llm=llm, database=db, verbose=True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "question = \"What are the first names of all the customers?\"\n",
+ "chain.run(question)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can customize the prompt."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from langchain.chains.elasticsearch_database.prompts import DEFAULT_DSL_TEMPLATE\n",
+ "from langchain.prompts.prompt import PromptTemplate\n",
+ "\n",
+ "PROMPT_TEMPLATE = \"\"\"Given an input question, create a syntactically correct Elasticsearch query to run. Unless the user specifies in their question a specific number of examples they wish to obtain, always limit your query to at most {top_k} results. You can order the results by a relevant column to return the most interesting examples in the database.\n",
+ "\n",
+ "Unless told to do not query for all the columns from a specific index, only ask for a the few relevant columns given the question.\n",
+ "\n",
+ "Pay attention to use only the column names that you can see in the mapping description. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which index. Return the query as valid json.\n",
+ "\n",
+ "Use the following format:\n",
+ "\n",
+ "Question: Question here\n",
+ "ESQuery: Elasticsearch Query formatted as json\n",
+ "\"\"\"\n",
+ "\n",
+ "PROMPT = PromptTemplate.from_template(\n",
+ " PROMPT_TEMPLATE,\n",
+ ")\n",
+ "chain = ElasticsearchDatabaseChain.from_llm(llm=llm, database=db, query_prompt=PROMPT)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.9.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/docs/extras/use_cases/sql.ipynb b/docs/extras/use_cases/sql/sql.ipynb
similarity index 100%
rename from docs/extras/use_cases/sql.ipynb
rename to docs/extras/use_cases/sql/sql.ipynb
diff --git a/docs/extras/integrations/tools/sqlite.mdx b/docs/extras/use_cases/sql/sqlite.mdx
similarity index 100%
rename from docs/extras/integrations/tools/sqlite.mdx
rename to docs/extras/use_cases/sql/sqlite.mdx
diff --git a/docs/extras/use_cases/summarization.ipynb b/docs/extras/use_cases/summarization.ipynb
index 000ba48124..6d7e118ab7 100644
--- a/docs/extras/use_cases/summarization.ipynb
+++ b/docs/extras/use_cases/summarization.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "2aca8168-62ec-4bba-93f0-73da08cd1920",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Summarization\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "cf13f702",
"metadata": {},
"source": [
- "# Summarization\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/summarization.ipynb)\n",
"\n",
"## Use case\n",
@@ -548,7 +557,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.16"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/extras/use_cases/tagging.ipynb b/docs/extras/use_cases/tagging.ipynb
index 235f9d06cb..37242a84f5 100644
--- a/docs/extras/use_cases/tagging.ipynb
+++ b/docs/extras/use_cases/tagging.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "cb6f552e-775f-4d84-bc7c-dca94c06a33c",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Tagging\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "a0507a4b",
"metadata": {},
"source": [
- "# Tagging\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/tagging.ipynb)\n",
"\n",
"## Use case\n",
@@ -408,7 +417,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.16"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/extras/use_cases/web_scraping.ipynb b/docs/extras/use_cases/web_scraping.ipynb
index 57c9e8387a..41bb28703e 100644
--- a/docs/extras/use_cases/web_scraping.ipynb
+++ b/docs/extras/use_cases/web_scraping.ipynb
@@ -1,12 +1,21 @@
{
"cells": [
+ {
+ "cell_type": "raw",
+ "id": "e254cf03-49fc-4051-a4df-3a8e4e7d2688",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "sidebar_position: 1\n",
+ "title: Web scraping\n",
+ "---"
+ ]
+ },
{
"cell_type": "markdown",
"id": "6605e7f7",
"metadata": {},
"source": [
- "# Web Scraping\n",
- "\n",
"[](https://colab.research.google.com/github/langchain-ai/langchain/blob/master/docs/extras/use_cases/web_scraping.ipynb)\n",
"\n",
"## Use case\n",
@@ -306,9 +315,7 @@
"cell_type": "code",
"execution_count": 7,
"id": "977560ba",
- "metadata": {
- "scrolled": false
- },
+ "metadata": {},
"outputs": [
{
"name": "stdout",
@@ -591,7 +598,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.16"
+ "version": "3.9.1"
}
},
"nbformat": 4,
diff --git a/docs/snippets/modules/agents/agent_types/openai_functions_agent.mdx b/docs/snippets/modules/agents/agent_types/openai_functions_agent.mdx
index a04e2ac8c7..aaf208dc05 100644
--- a/docs/snippets/modules/agents/agent_types/openai_functions_agent.mdx
+++ b/docs/snippets/modules/agents/agent_types/openai_functions_agent.mdx
@@ -5,10 +5,12 @@ pip install openai google-search-results
```
```python
-from langchain import LLMMathChain, OpenAI, SerpAPIWrapper, SQLDatabase, SQLDatabaseChain
-from langchain.agents import initialize_agent, Tool
-from langchain.agents import AgentType
+from langchain.agents import initialize_agent, AgentType, Tool
+from langchain.chains import LLMMathChain
from langchain.chat_models import ChatOpenAI
+from langchain.llms import OpenAI
+from langchain.utilities import SerpAPIWrapper, SQLDatabase
+from langchain_experimental.sql import SQLDatabaseChain
```
diff --git a/libs/experimental/langchain_experimental/data_anonymizer/__init__.py b/libs/experimental/langchain_experimental/data_anonymizer/__init__.py
index 69babad859..f43d6d98df 100644
--- a/libs/experimental/langchain_experimental/data_anonymizer/__init__.py
+++ b/libs/experimental/langchain_experimental/data_anonymizer/__init__.py
@@ -1,4 +1,7 @@
"""Data anonymizer package"""
-from langchain_experimental.data_anonymizer.presidio import PresidioAnonymizer
+from langchain_experimental.data_anonymizer.presidio import (
+ PresidioAnonymizer,
+ PresidioReversibleAnonymizer,
+)
-__all__ = ["PresidioAnonymizer"]
+__all__ = ["PresidioAnonymizer", "PresidioReversibleAnonymizer"]
diff --git a/libs/experimental/langchain_experimental/data_anonymizer/base.py b/libs/experimental/langchain_experimental/data_anonymizer/base.py
index 3f9905375e..292d2a2a0f 100644
--- a/libs/experimental/langchain_experimental/data_anonymizer/base.py
+++ b/libs/experimental/langchain_experimental/data_anonymizer/base.py
@@ -1,4 +1,5 @@
from abc import ABC, abstractmethod
+from typing import Optional
class AnonymizerBase(ABC):
@@ -8,10 +9,24 @@ class AnonymizerBase(ABC):
wrapping the behavior for all methods in a base class.
"""
- def anonymize(self, text: str) -> str:
+ def anonymize(self, text: str, language: Optional[str] = None) -> str:
"""Anonymize text"""
- return self._anonymize(text)
+ return self._anonymize(text, language)
@abstractmethod
- def _anonymize(self, text: str) -> str:
+ def _anonymize(self, text: str, language: Optional[str]) -> str:
"""Abstract method to anonymize text"""
+
+
+class ReversibleAnonymizerBase(AnonymizerBase):
+ """
+ Base abstract class for reversible anonymizers.
+ """
+
+ def deanonymize(self, text: str) -> str:
+ """Deanonymize text"""
+ return self._deanonymize(text)
+
+ @abstractmethod
+ def _deanonymize(self, text: str) -> str:
+ """Abstract method to deanonymize text"""
diff --git a/libs/experimental/langchain_experimental/data_anonymizer/deanonymizer_mapping.py b/libs/experimental/langchain_experimental/data_anonymizer/deanonymizer_mapping.py
new file mode 100644
index 0000000000..2ee03eb208
--- /dev/null
+++ b/libs/experimental/langchain_experimental/data_anonymizer/deanonymizer_mapping.py
@@ -0,0 +1,21 @@
+from collections import defaultdict
+from dataclasses import dataclass, field
+from typing import Dict
+
+MappingDataType = Dict[str, Dict[str, str]]
+
+
+@dataclass
+class DeanonymizerMapping:
+ mapping: MappingDataType = field(
+ default_factory=lambda: defaultdict(lambda: defaultdict(str))
+ )
+
+ @property
+ def data(self) -> MappingDataType:
+ """Return the deanonymizer mapping"""
+ return {k: dict(v) for k, v in self.mapping.items()}
+
+ def update(self, new_mapping: MappingDataType) -> None:
+ for entity_type, values in new_mapping.items():
+ self.mapping[entity_type].update(values)
diff --git a/libs/experimental/langchain_experimental/data_anonymizer/deanonymizer_matching_strategies.py b/libs/experimental/langchain_experimental/data_anonymizer/deanonymizer_matching_strategies.py
new file mode 100644
index 0000000000..e5d9e8581b
--- /dev/null
+++ b/libs/experimental/langchain_experimental/data_anonymizer/deanonymizer_matching_strategies.py
@@ -0,0 +1,17 @@
+from langchain_experimental.data_anonymizer.presidio import MappingDataType
+
+
+def default_matching_strategy(text: str, deanonymizer_mapping: MappingDataType) -> str:
+ """
+ Default matching strategy for deanonymization.
+ It replaces all the anonymized entities with the original ones.
+
+ Args:
+ text: text to deanonymize
+ deanonymizer_mapping: mapping between anonymized entities and original ones"""
+
+ # Iterate over all the entities (PERSON, EMAIL_ADDRESS, etc.)
+ for entity_type in deanonymizer_mapping:
+ for anonymized, original in deanonymizer_mapping[entity_type].items():
+ text = text.replace(anonymized, original)
+ return text
diff --git a/libs/experimental/langchain_experimental/data_anonymizer/faker_presidio_mapping.py b/libs/experimental/langchain_experimental/data_anonymizer/faker_presidio_mapping.py
index 8db4f94c2f..9015679f20 100644
--- a/libs/experimental/langchain_experimental/data_anonymizer/faker_presidio_mapping.py
+++ b/libs/experimental/langchain_experimental/data_anonymizer/faker_presidio_mapping.py
@@ -1,8 +1,8 @@
import string
-from typing import Callable, Dict
+from typing import Callable, Dict, Optional
-def get_pseudoanonymizer_mapping() -> Dict[str, Callable]:
+def get_pseudoanonymizer_mapping(seed: Optional[int] = None) -> Dict[str, Callable]:
try:
from faker import Faker
except ImportError as e:
@@ -11,6 +11,7 @@ def get_pseudoanonymizer_mapping() -> Dict[str, Callable]:
) from e
fake = Faker()
+ fake.seed_instance(seed)
# Listed entities supported by Microsoft Presidio (for now, global and US only)
# Source: https://microsoft.github.io/presidio/supported_entities/
@@ -26,8 +27,8 @@ def get_pseudoanonymizer_mapping() -> Dict[str, Callable]:
fake.random_choices(string.ascii_lowercase + string.digits, length=26)
),
"IP_ADDRESS": lambda _: fake.ipv4_public(),
- "LOCATION": lambda _: fake.address(),
- "DATE_TIME": lambda _: fake.iso8601(),
+ "LOCATION": lambda _: fake.city(),
+ "DATE_TIME": lambda _: fake.date(),
"NRP": lambda _: str(fake.random_number(digits=8, fix_len=True)),
"MEDICAL_LICENSE": lambda _: fake.bothify(text="??######").upper(),
"URL": lambda _: fake.url(),
diff --git a/libs/experimental/langchain_experimental/data_anonymizer/presidio.py b/libs/experimental/langchain_experimental/data_anonymizer/presidio.py
index 298e3de1d5..b2be1dc5a1 100644
--- a/libs/experimental/langchain_experimental/data_anonymizer/presidio.py
+++ b/libs/experimental/langchain_experimental/data_anonymizer/presidio.py
@@ -1,24 +1,75 @@
from __future__ import annotations
-from typing import TYPE_CHECKING, Dict, List, Optional
+import json
+from collections import defaultdict
+from pathlib import Path
+from typing import TYPE_CHECKING, Callable, Dict, List, Optional, Union
-from langchain_experimental.data_anonymizer.base import AnonymizerBase
+import yaml
+
+from langchain_experimental.data_anonymizer.base import (
+ AnonymizerBase,
+ ReversibleAnonymizerBase,
+)
+from langchain_experimental.data_anonymizer.deanonymizer_mapping import (
+ DeanonymizerMapping,
+ MappingDataType,
+)
+from langchain_experimental.data_anonymizer.deanonymizer_matching_strategies import (
+ default_matching_strategy,
+)
from langchain_experimental.data_anonymizer.faker_presidio_mapping import (
get_pseudoanonymizer_mapping,
)
-if TYPE_CHECKING:
- from presidio_analyzer import EntityRecognizer
+try:
+ from presidio_analyzer import AnalyzerEngine
+ from presidio_analyzer.nlp_engine import NlpEngineProvider
+
+except ImportError as e:
+ raise ImportError(
+ "Could not import presidio_analyzer, please install with "
+ "`pip install presidio-analyzer`. You will also need to download a "
+ "spaCy model to use the analyzer, e.g. "
+ "`python -m spacy download en_core_web_lg`."
+ ) from e
+try:
+ from presidio_anonymizer import AnonymizerEngine
from presidio_anonymizer.entities import OperatorConfig
+except ImportError as e:
+ raise ImportError(
+ "Could not import presidio_anonymizer, please install with "
+ "`pip install presidio-anonymizer`."
+ ) from e
+if TYPE_CHECKING:
+ from presidio_analyzer import EntityRecognizer, RecognizerResult
+ from presidio_anonymizer.entities import EngineResult
-class PresidioAnonymizer(AnonymizerBase):
- """Anonymizer using Microsoft Presidio."""
+# Configuring Anonymizer for multiple languages
+# Detailed description and examples can be found here:
+# langchain/docs/extras/guides/privacy/multi_language_anonymization.ipynb
+DEFAULT_LANGUAGES_CONFIG = {
+ # You can also use Stanza or transformers library.
+ # See https://microsoft.github.io/presidio/analyzer/customizing_nlp_models/
+ "nlp_engine_name": "spacy",
+ "models": [
+ {"lang_code": "en", "model_name": "en_core_web_lg"},
+ # {"lang_code": "de", "model_name": "de_core_news_md"},
+ # {"lang_code": "es", "model_name": "es_core_news_md"},
+ # ...
+ # List of available models: https://spacy.io/usage/models
+ ],
+}
+
+class PresidioAnonymizerBase(AnonymizerBase):
def __init__(
self,
analyzed_fields: Optional[List[str]] = None,
operators: Optional[Dict[str, OperatorConfig]] = None,
+ languages_config: Dict = DEFAULT_LANGUAGES_CONFIG,
+ faker_seed: Optional[int] = None,
):
"""
Args:
@@ -28,25 +79,15 @@ class PresidioAnonymizer(AnonymizerBase):
Operators allow for custom anonymization of detected PII.
Learn more:
https://microsoft.github.io/presidio/tutorial/10_simple_anonymization/
+ languages_config: Configuration for the NLP engine.
+ First language in the list will be used as the main language
+ in self.anonymize(...) when no language is specified.
+ Learn more:
+ https://microsoft.github.io/presidio/analyzer/customizing_nlp_models/
+ faker_seed: Seed used to initialize faker.
+ Defaults to None, in which case faker will be seeded randomly
+ and provide random values.
"""
- try:
- from presidio_analyzer import AnalyzerEngine
- except ImportError as e:
- raise ImportError(
- "Could not import presidio_analyzer, please install with "
- "`pip install presidio-analyzer`. You will also need to download a "
- "spaCy model to use the analyzer, e.g. "
- "`python -m spacy download en_core_web_lg`."
- ) from e
- try:
- from presidio_anonymizer import AnonymizerEngine
- from presidio_anonymizer.entities import OperatorConfig
- except ImportError as e:
- raise ImportError(
- "Could not import presidio_anonymizer, please install with "
- "`pip install presidio-anonymizer`."
- ) from e
-
self.analyzed_fields = (
analyzed_fields
if analyzed_fields is not None
@@ -59,17 +100,66 @@ class PresidioAnonymizer(AnonymizerBase):
field: OperatorConfig(
operator_name="custom", params={"lambda": faker_function}
)
- for field, faker_function in get_pseudoanonymizer_mapping().items()
+ for field, faker_function in get_pseudoanonymizer_mapping(
+ faker_seed
+ ).items()
}
)
- self._analyzer = AnalyzerEngine()
+
+ provider = NlpEngineProvider(nlp_configuration=languages_config)
+ nlp_engine = provider.create_engine()
+
+ self.supported_languages = list(nlp_engine.nlp.keys())
+
+ self._analyzer = AnalyzerEngine(
+ supported_languages=self.supported_languages, nlp_engine=nlp_engine
+ )
self._anonymizer = AnonymizerEngine()
- def _anonymize(self, text: str) -> str:
+ def add_recognizer(self, recognizer: EntityRecognizer) -> None:
+ """Add a recognizer to the analyzer
+
+ Args:
+ recognizer: Recognizer to add to the analyzer.
+ """
+ self._analyzer.registry.add_recognizer(recognizer)
+ self.analyzed_fields.extend(recognizer.supported_entities)
+
+ def add_operators(self, operators: Dict[str, OperatorConfig]) -> None:
+ """Add operators to the anonymizer
+
+ Args:
+ operators: Operators to add to the anonymizer.
+ """
+ self.operators.update(operators)
+
+
+class PresidioAnonymizer(PresidioAnonymizerBase):
+ def _anonymize(self, text: str, language: Optional[str] = None) -> str:
+ """Anonymize text.
+ Each PII entity is replaced with a fake value.
+ Each time fake values will be different, as they are generated randomly.
+
+ Args:
+ text: text to anonymize
+ language: language to use for analysis of PII
+ If None, the first (main) language in the list
+ of languages specified in the configuration will be used.
+ """
+ if language is None:
+ language = self.supported_languages[0]
+
+ if language not in self.supported_languages:
+ raise ValueError(
+ f"Language '{language}' is not supported. "
+ f"Supported languages are: {self.supported_languages}. "
+ "Change your language configuration file to add more languages."
+ )
+
results = self._analyzer.analyze(
text,
entities=self.analyzed_fields,
- language="en",
+ language=language,
)
return self._anonymizer.anonymize(
@@ -78,11 +168,199 @@ class PresidioAnonymizer(AnonymizerBase):
operators=self.operators,
).text
- def add_recognizer(self, recognizer: EntityRecognizer) -> None:
- """Add a recognizer to the analyzer"""
- self._analyzer.registry.add_recognizer(recognizer)
- self.analyzed_fields.extend(recognizer.supported_entities)
- def add_operators(self, operators: Dict[str, OperatorConfig]) -> None:
- """Add operators to the anonymizer"""
- self.operators.update(operators)
+class PresidioReversibleAnonymizer(PresidioAnonymizerBase, ReversibleAnonymizerBase):
+ def __init__(
+ self,
+ analyzed_fields: Optional[List[str]] = None,
+ operators: Optional[Dict[str, OperatorConfig]] = None,
+ languages_config: Dict = DEFAULT_LANGUAGES_CONFIG,
+ faker_seed: Optional[int] = None,
+ ):
+ super().__init__(analyzed_fields, operators, languages_config, faker_seed)
+ self._deanonymizer_mapping = DeanonymizerMapping()
+
+ @property
+ def deanonymizer_mapping(self) -> MappingDataType:
+ """Return the deanonymizer mapping"""
+ return self._deanonymizer_mapping.data
+
+ def _update_deanonymizer_mapping(
+ self,
+ original_text: str,
+ analyzer_results: List[RecognizerResult],
+ anonymizer_results: EngineResult,
+ ) -> None:
+ """Creates or updates the mapping used to de-anonymize text.
+
+ This method exploits the results returned by the
+ analysis and anonymization processes.
+
+ It constructs a mapping from each anonymized entity
+ back to its original text value.
+
+ Mapping will be stored as "deanonymizer_mapping" property.
+
+ Example of "deanonymizer_mapping":
+ {
+ "PERSON": {
+ "": "",
+ "John Doe": "Slim Shady"
+ },
+ "PHONE_NUMBER": {
+ "111-111-1111": "555-555-5555"
+ }
+ ...
+ }
+ """
+
+ # We are able to zip and loop through both lists because we expect
+ # them to return corresponding entities for each identified piece
+ # of analyzable data from our input.
+
+ # We sort them by their 'start' attribute because it allows us to
+ # match corresponding entities by their position in the input text.
+ analyzer_results = sorted(analyzer_results, key=lambda d: d.start)
+ anonymizer_results.items = sorted(
+ anonymizer_results.items, key=lambda d: d.start
+ )
+
+ new_deanonymizer_mapping: MappingDataType = defaultdict(dict)
+
+ for analyzed_entity, anonymized_entity in zip(
+ analyzer_results, anonymizer_results.items
+ ):
+ original_value = original_text[analyzed_entity.start : analyzed_entity.end]
+ new_deanonymizer_mapping[anonymized_entity.entity_type][
+ anonymized_entity.text
+ ] = original_value
+
+ self._deanonymizer_mapping.update(new_deanonymizer_mapping)
+
+ def _anonymize(self, text: str, language: Optional[str] = None) -> str:
+ """Anonymize text.
+ Each PII entity is replaced with a fake value.
+ Each time fake values will be different, as they are generated randomly.
+ At the same time, we will create a mapping from each anonymized entity
+ back to its original text value.
+
+ Args:
+ text: text to anonymize
+ language: language to use for analysis of PII
+ If None, the first (main) language in the list
+ of languages specified in the configuration will be used.
+ """
+ if language is None:
+ language = self.supported_languages[0]
+
+ if language not in self.supported_languages:
+ raise ValueError(
+ f"Language '{language}' is not supported. "
+ f"Supported languages are: {self.supported_languages}. "
+ "Change your language configuration file to add more languages."
+ )
+
+ analyzer_results = self._analyzer.analyze(
+ text,
+ entities=self.analyzed_fields,
+ language=language,
+ )
+
+ filtered_analyzer_results = (
+ self._anonymizer._remove_conflicts_and_get_text_manipulation_data(
+ analyzer_results
+ )
+ )
+
+ anonymizer_results = self._anonymizer.anonymize(
+ text,
+ analyzer_results=analyzer_results,
+ operators=self.operators,
+ )
+
+ self._update_deanonymizer_mapping(
+ text, filtered_analyzer_results, anonymizer_results
+ )
+
+ return anonymizer_results.text
+
+ def _deanonymize(
+ self,
+ text_to_deanonymize: str,
+ deanonymizer_matching_strategy: Callable[
+ [str, MappingDataType], str
+ ] = default_matching_strategy,
+ ) -> str:
+ """Deanonymize text.
+ Each anonymized entity is replaced with its original value.
+ This method exploits the mapping created during the anonymization process.
+
+ Args:
+ text_to_deanonymize: text to deanonymize
+ deanonymizer_matching_strategy: function to use to match
+ anonymized entities with their original values and replace them.
+ """
+ if not self._deanonymizer_mapping:
+ raise ValueError(
+ "Deanonymizer mapping is empty.",
+ "Please call anonymize() and anonymize some text first.",
+ )
+
+ text_to_deanonymize = deanonymizer_matching_strategy(
+ text_to_deanonymize, self.deanonymizer_mapping
+ )
+
+ return text_to_deanonymize
+
+ def save_deanonymizer_mapping(self, file_path: Union[Path, str]) -> None:
+ """Save the deanonymizer mapping to a JSON or YAML file.
+
+ Args:
+ file_path: Path to file to save the mapping to.
+
+ Example:
+ .. code-block:: python
+
+ anonymizer.save_deanonymizer_mapping(file_path="path/mapping.json")
+ """
+
+ save_path = Path(file_path)
+
+ if save_path.suffix not in [".json", ".yaml"]:
+ raise ValueError(f"{save_path} must have an extension of .json or .yaml")
+
+ # Make sure parent directories exist
+ save_path.parent.mkdir(parents=True, exist_ok=True)
+
+ if save_path.suffix == ".json":
+ with open(save_path, "w") as f:
+ json.dump(self.deanonymizer_mapping, f, indent=2)
+ elif save_path.suffix == ".yaml":
+ with open(save_path, "w") as f:
+ yaml.dump(self.deanonymizer_mapping, f, default_flow_style=False)
+
+ def load_deanonymizer_mapping(self, file_path: Union[Path, str]) -> None:
+ """Load the deanonymizer mapping from a JSON or YAML file.
+
+ Args:
+ file_path: Path to file to load the mapping from.
+
+ Example:
+ .. code-block:: python
+
+ anonymizer.load_deanonymizer_mapping(file_path="path/mapping.json")
+ """
+
+ load_path = Path(file_path)
+
+ if load_path.suffix not in [".json", ".yaml"]:
+ raise ValueError(f"{load_path} must have an extension of .json or .yaml")
+
+ if load_path.suffix == ".json":
+ with open(load_path, "r") as f:
+ loaded_mapping = json.load(f)
+ elif load_path.suffix == ".yaml":
+ with open(load_path, "r") as f:
+ loaded_mapping = yaml.load(f, Loader=yaml.FullLoader)
+
+ self._deanonymizer_mapping.update(loaded_mapping)
diff --git a/libs/experimental/langchain_experimental/graph_transformers/__init__.py b/libs/experimental/langchain_experimental/graph_transformers/__init__.py
new file mode 100644
index 0000000000..3f6c8a665e
--- /dev/null
+++ b/libs/experimental/langchain_experimental/graph_transformers/__init__.py
@@ -0,0 +1,5 @@
+from langchain_experimental.graph_transformers.diffbot import DiffbotGraphTransformer
+
+__all__ = [
+ "DiffbotGraphTransformer",
+]
diff --git a/libs/experimental/langchain_experimental/graph_transformers/diffbot.py b/libs/experimental/langchain_experimental/graph_transformers/diffbot.py
new file mode 100644
index 0000000000..000c70de4b
--- /dev/null
+++ b/libs/experimental/langchain_experimental/graph_transformers/diffbot.py
@@ -0,0 +1,316 @@
+from typing import Any, Dict, List, Optional, Sequence, Tuple, Union
+
+import requests
+from langchain.graphs.graph_document import GraphDocument, Node, Relationship
+from langchain.schema import Document
+from langchain.utils import get_from_env
+
+
+def format_property_key(s: str) -> str:
+ words = s.split()
+ if not words:
+ return s
+ first_word = words[0].lower()
+ capitalized_words = [word.capitalize() for word in words[1:]]
+ return "".join([first_word] + capitalized_words)
+
+
+class NodesList:
+ """
+ Manages a list of nodes with associated properties.
+
+ Attributes:
+ nodes (Dict[Tuple, Any]): Stores nodes as keys and their properties as values.
+ Each key is a tuple where the first element is the
+ node ID and the second is the node type.
+ """
+
+ def __init__(self) -> None:
+ self.nodes: Dict[Tuple[Union[str, int], str], Any] = dict()
+
+ def add_node_property(
+ self, node: Tuple[Union[str, int], str], properties: Dict[str, Any]
+ ) -> None:
+ """
+ Adds or updates node properties.
+
+ If the node does not exist in the list, it's added along with its properties.
+ If the node already exists, its properties are updated with the new values.
+
+ Args:
+ node (Tuple): A tuple containing the node ID and node type.
+ properties (Dict): A dictionary of properties to add or update for the node.
+ """
+ if node not in self.nodes:
+ self.nodes[node] = properties
+ else:
+ self.nodes[node].update(properties)
+
+ def return_node_list(self) -> List[Node]:
+ """
+ Returns the nodes as a list of Node objects.
+
+ Each Node object will have its ID, type, and properties populated.
+
+ Returns:
+ List[Node]: A list of Node objects.
+ """
+ nodes = [
+ Node(id=key[0], type=key[1], properties=self.nodes[key])
+ for key in self.nodes
+ ]
+ return nodes
+
+
+# Properties that should be treated as node properties instead of relationships
+FACT_TO_PROPERTY_TYPE = [
+ "Date",
+ "Number",
+ "Job title",
+ "Cause of death",
+ "Organization type",
+ "Academic title",
+]
+
+
+schema_mapping = [
+ ("HEADQUARTERS", "ORGANIZATION_LOCATIONS"),
+ ("RESIDENCE", "PERSON_LOCATION"),
+ ("ALL_PERSON_LOCATIONS", "PERSON_LOCATION"),
+ ("CHILD", "HAS_CHILD"),
+ ("PARENT", "HAS_PARENT"),
+ ("CUSTOMERS", "HAS_CUSTOMER"),
+ ("SKILLED_AT", "INTERESTED_IN"),
+]
+
+
+class SimplifiedSchema:
+ """
+ Provides functionality for working with a simplified schema mapping.
+
+ Attributes:
+ schema (Dict): A dictionary containing the mapping to simplified schema types.
+ """
+
+ def __init__(self) -> None:
+ """Initializes the schema dictionary based on the predefined list."""
+ self.schema = dict()
+ for row in schema_mapping:
+ self.schema[row[0]] = row[1]
+
+ def get_type(self, type: str) -> str:
+ """
+ Retrieves the simplified schema type for a given original type.
+
+ Args:
+ type (str): The original schema type to find the simplified type for.
+
+ Returns:
+ str: The simplified schema type if it exists;
+ otherwise, returns the original type.
+ """
+ try:
+ return self.schema[type]
+ except KeyError:
+ return type
+
+
+class DiffbotGraphTransformer:
+ """Transforms documents into graph documents using Diffbot's NLP API.
+
+ A graph document transformation system takes a sequence of Documents and returns a
+ sequence of Graph Documents.
+
+ Example:
+ .. code-block:: python
+
+ class DiffbotGraphTransformer(BaseGraphDocumentTransformer):
+
+ def transform_documents(
+ self, documents: Sequence[Document], **kwargs: Any
+ ) -> Sequence[GraphDocument]:
+ results = []
+
+ for document in documents:
+ raw_results = self.nlp_request(document.page_content)
+ graph_document = self.process_response(raw_results, document)
+ results.append(graph_document)
+ return results
+
+ async def atransform_documents(
+ self, documents: Sequence[Document], **kwargs: Any
+ ) -> Sequence[Document]:
+ raise NotImplementedError
+ """
+
+ def __init__(
+ self,
+ diffbot_api_key: Optional[str] = None,
+ fact_confidence_threshold: float = 0.7,
+ include_qualifiers: bool = True,
+ include_evidence: bool = True,
+ simplified_schema: bool = True,
+ ) -> None:
+ """
+ Initialize the graph transformer with various options.
+
+ Args:
+ diffbot_api_key (str):
+ The API key for Diffbot's NLP services.
+
+ fact_confidence_threshold (float):
+ Minimum confidence level for facts to be included.
+ include_qualifiers (bool):
+ Whether to include qualifiers in the relationships.
+ include_evidence (bool):
+ Whether to include evidence for the relationships.
+ simplified_schema (bool):
+ Whether to use a simplified schema for relationships.
+ """
+ self.diffbot_api_key = diffbot_api_key or get_from_env(
+ "diffbot_api_key", "DIFFBOT_API_KEY"
+ )
+ self.fact_threshold_confidence = fact_confidence_threshold
+ self.include_qualifiers = include_qualifiers
+ self.include_evidence = include_evidence
+ self.simplified_schema = None
+ if simplified_schema:
+ self.simplified_schema = SimplifiedSchema()
+
+ def nlp_request(self, text: str) -> Dict[str, Any]:
+ """
+ Make an API request to the Diffbot NLP endpoint.
+
+ Args:
+ text (str): The text to be processed.
+
+ Returns:
+ Dict[str, Any]: The JSON response from the API.
+ """
+
+ # Relationship extraction only works for English
+ payload = {
+ "content": text,
+ "lang": "en",
+ }
+
+ FIELDS = "facts"
+ HOST = "nl.diffbot.com"
+ url = (
+ f"https://{HOST}/v1/?fields={FIELDS}&"
+ f"token={self.diffbot_api_key}&language=en"
+ )
+ result = requests.post(url, data=payload)
+ return result.json()
+
+ def process_response(
+ self, payload: Dict[str, Any], document: Document
+ ) -> GraphDocument:
+ """
+ Transform the Diffbot NLP response into a GraphDocument.
+
+ Args:
+ payload (Dict[str, Any]): The JSON response from Diffbot's NLP API.
+ document (Document): The original document.
+
+ Returns:
+ GraphDocument: The transformed document as a graph.
+ """
+
+ # Return empty result if there are no facts
+ if "facts" not in payload or not payload["facts"]:
+ return GraphDocument(nodes=[], relationships=[], source=document)
+
+ # Nodes are a custom class because we need to deduplicate
+ nodes_list = NodesList()
+ # Relationships are a list because we don't deduplicate nor anything else
+ relationships = list()
+ for record in payload["facts"]:
+ # Skip if the fact is below the threshold confidence
+ if record["confidence"] < self.fact_threshold_confidence:
+ continue
+
+ # TODO: It should probably be treated as a node property
+ if not record["value"]["allTypes"]:
+ continue
+
+ # Define source node
+ source_id = (
+ record["entity"]["allUris"][0]
+ if record["entity"]["allUris"]
+ else record["entity"]["name"]
+ )
+ source_label = record["entity"]["allTypes"][0]["name"].capitalize()
+ source_name = record["entity"]["name"]
+ source_node = Node(id=source_id, type=source_label)
+ nodes_list.add_node_property(
+ (source_id, source_label), {"name": source_name}
+ )
+
+ # Define target node
+ target_id = (
+ record["value"]["allUris"][0]
+ if record["value"]["allUris"]
+ else record["value"]["name"]
+ )
+ target_label = record["value"]["allTypes"][0]["name"].capitalize()
+ target_name = record["value"]["name"]
+ # Some facts are better suited as node properties
+ if target_label in FACT_TO_PROPERTY_TYPE:
+ nodes_list.add_node_property(
+ (source_id, source_label),
+ {format_property_key(record["property"]["name"]): target_name},
+ )
+ else: # Define relationship
+ # Define target node object
+ target_node = Node(id=target_id, type=target_label)
+ nodes_list.add_node_property(
+ (target_id, target_label), {"name": target_name}
+ )
+ # Define relationship type
+ rel_type = record["property"]["name"].replace(" ", "_").upper()
+ if self.simplified_schema:
+ rel_type = self.simplified_schema.get_type(rel_type)
+
+ # Relationship qualifiers/properties
+ rel_properties = dict()
+ relationship_evidence = [el["passage"] for el in record["evidence"]][0]
+ if self.include_evidence:
+ rel_properties.update({"evidence": relationship_evidence})
+ if self.include_qualifiers and record.get("qualifiers"):
+ for property in record["qualifiers"]:
+ prop_key = format_property_key(property["property"]["name"])
+ rel_properties[prop_key] = property["value"]["name"]
+
+ relationship = Relationship(
+ source=source_node,
+ target=target_node,
+ type=rel_type,
+ properties=rel_properties,
+ )
+ relationships.append(relationship)
+
+ return GraphDocument(
+ nodes=nodes_list.return_node_list(),
+ relationships=relationships,
+ source=document,
+ )
+
+ def convert_to_graph_documents(
+ self, documents: Sequence[Document]
+ ) -> List[GraphDocument]:
+ """Convert a sequence of documents into graph documents.
+
+ Args:
+ documents (Sequence[Document]): The original documents.
+ **kwargs: Additional keyword arguments.
+
+ Returns:
+ Sequence[GraphDocument]: The transformed documents as graphs.
+ """
+ results = []
+ for document in documents:
+ raw_results = self.nlp_request(document.page_content)
+ graph_document = self.process_response(raw_results, document)
+ results.append(graph_document)
+ return results
diff --git a/libs/experimental/langchain_experimental/retrievers/vector_sql_database.py b/libs/experimental/langchain_experimental/retrievers/vector_sql_database.py
new file mode 100644
index 0000000000..1ec088dbc5
--- /dev/null
+++ b/libs/experimental/langchain_experimental/retrievers/vector_sql_database.py
@@ -0,0 +1,38 @@
+"""Vector SQL Database Chain Retriever"""
+from typing import Any, Dict, List
+
+from langchain.callbacks.manager import (
+ AsyncCallbackManagerForRetrieverRun,
+ CallbackManagerForRetrieverRun,
+)
+from langchain.schema import BaseRetriever, Document
+
+from langchain_experimental.sql.vector_sql import VectorSQLDatabaseChain
+
+
+class VectorSQLDatabaseChainRetriever(BaseRetriever):
+ """Retriever that uses SQLDatabase as Retriever"""
+
+ sql_db_chain: VectorSQLDatabaseChain
+ """SQL Database Chain"""
+ page_content_key: str = "content"
+ """column name for page content of documents"""
+
+ def _get_relevant_documents(
+ self,
+ query: str,
+ *,
+ run_manager: CallbackManagerForRetrieverRun,
+ **kwargs: Any,
+ ) -> List[Document]:
+ ret: List[Dict[str, Any]] = self.sql_db_chain(
+ query, callbacks=run_manager.get_child(), **kwargs
+ )["result"]
+ return [
+ Document(page_content=r[self.page_content_key], metadata=r) for r in ret
+ ]
+
+ async def _aget_relevant_documents(
+ self, query: str, *, run_manager: AsyncCallbackManagerForRetrieverRun
+ ) -> List[Document]:
+ raise NotImplementedError
diff --git a/libs/experimental/langchain_experimental/sql/prompt.py b/libs/experimental/langchain_experimental/sql/prompt.py
new file mode 100644
index 0000000000..5f4c9b8a4f
--- /dev/null
+++ b/libs/experimental/langchain_experimental/sql/prompt.py
@@ -0,0 +1,85 @@
+# flake8: noqa
+from langchain.prompts.prompt import PromptTemplate
+
+
+PROMPT_SUFFIX = """Only use the following tables:
+{table_info}
+
+Question: {input}"""
+
+_VECTOR_SQL_DEFAULT_TEMPLATE = """You are a {dialect} expert. Given an input question, first create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer to the input question.
+{dialect} queries has a vector distance function called `DISTANCE(column, array)` to compute relevance to the user's question and sort the feature array column by the relevance.
+When the query is asking for {top_k} closest row, you have to use this distance function to calculate distance to entity's array on vector column and order by the distance to retrieve relevant rows.
+
+*NOTICE*: `DISTANCE(column, array)` only accept an array column as its first argument and a `NeuralArray(entity)` as its second argument. You also need a user defined function called `NeuralArray(entity)` to retrieve the entity's array.
+
+Unless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per {dialect}. You should only order according to the distance function.
+Never query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (") to denote them as delimited identifiers.
+Pay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.
+Pay attention to use today() function to get the current date, if the question involves "today". `ORDER BY` clause should always be after `WHERE` clause. DO NOT add semicolon to the end of SQL. Pay attention to the comment in table schema.
+
+Use the following format:
+
+Question: "Question here"
+SQLQuery: "SQL Query to run"
+SQLResult: "Result of the SQLQuery"
+Answer: "Final answer here"
+"""
+
+VECTOR_SQL_PROMPT = PromptTemplate(
+ input_variables=["input", "table_info", "dialect", "top_k"],
+ template=_VECTOR_SQL_DEFAULT_TEMPLATE + PROMPT_SUFFIX,
+)
+
+
+_myscale_prompt = """You are a MyScale expert. Given an input question, first create a syntactically correct MyScale query to run, then look at the results of the query and return the answer to the input question.
+MyScale queries has a vector distance function called `DISTANCE(column, array)` to compute relevance to the user's question and sort the feature array column by the relevance.
+When the query is asking for {top_k} closest row, you have to use this distance function to calculate distance to entity's array on vector column and order by the distance to retrieve relevant rows.
+
+*NOTICE*: `DISTANCE(column, array)` only accept an array column as its first argument and a `NeuralArray(entity)` as its second argument. You also need a user defined function called `NeuralArray(entity)` to retrieve the entity's array.
+
+Unless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per MyScale. You should only order according to the distance function.
+Never query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (") to denote them as delimited identifiers.
+Pay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.
+Pay attention to use today() function to get the current date, if the question involves "today". `ORDER BY` clause should always be after `WHERE` clause. DO NOT add semicolon to the end of SQL. Pay attention to the comment in table schema.
+
+Use the following format:
+
+======== table info ========
+
+
+Question: "Question here"
+SQLQuery: "SQL Query to run"
+
+
+Here are some examples:
+
+======== table info ========
+CREATE TABLE "ChatPaper" (
+ abstract String,
+ id String,
+ vector Array(Float32),
+) ENGINE = ReplicatedReplacingMergeTree()
+ ORDER BY id
+ PRIMARY KEY id
+
+Question: What is Feartue Pyramid Network?
+SQLQuery: SELECT ChatPaper.title, ChatPaper.id, ChatPaper.authors FROM ChatPaper ORDER BY DISTANCE(vector, NeuralArray(PaperRank contribution)) LIMIT {top_k}
+
+
+Let's begin:
+======== table info ========
+{table_info}
+
+Question: {input}
+SQLQuery: """
+
+MYSCALE_PROMPT = PromptTemplate(
+ input_variables=["input", "table_info", "top_k"],
+ template=_myscale_prompt + PROMPT_SUFFIX,
+)
+
+
+VECTOR_SQL_PROMPTS = {
+ "myscale": MYSCALE_PROMPT,
+}
diff --git a/libs/experimental/langchain_experimental/sql/vector_sql.py b/libs/experimental/langchain_experimental/sql/vector_sql.py
new file mode 100644
index 0000000000..98f3c2dee0
--- /dev/null
+++ b/libs/experimental/langchain_experimental/sql/vector_sql.py
@@ -0,0 +1,237 @@
+"""Vector SQL Database Chain Retriever"""
+from __future__ import annotations
+
+from typing import Any, Dict, List, Optional, Union
+
+from langchain.callbacks.manager import CallbackManagerForChainRun
+from langchain.chains.llm import LLMChain
+from langchain.chains.sql_database.prompt import PROMPT, SQL_PROMPTS
+from langchain.embeddings.base import Embeddings
+from langchain.prompts.prompt import PromptTemplate
+from langchain.schema import BaseOutputParser, BasePromptTemplate
+from langchain.schema.language_model import BaseLanguageModel
+from langchain.tools.sql_database.prompt import QUERY_CHECKER
+from langchain.utilities.sql_database import SQLDatabase
+
+from langchain_experimental.sql.base import INTERMEDIATE_STEPS_KEY, SQLDatabaseChain
+
+
+class VectorSQLOutputParser(BaseOutputParser[str]):
+ """Output Parser for Vector SQL
+ 1. finds for `NeuralArray()` and replace it with the embedding
+ 2. finds for `DISTANCE()` and replace it with the distance name in backend SQL
+ """
+
+ model: Embeddings
+ """Embedding model to extract embedding for entity"""
+ distance_func_name: str = "distance"
+ """Distance name for Vector SQL"""
+
+ class Config:
+ arbitrary_types_allowed = 1
+
+ @property
+ def _type(self) -> str:
+ return "vector_sql_parser"
+
+ @classmethod
+ def from_embeddings(
+ cls, model: Embeddings, distance_func_name: str = "distance", **kwargs: Any
+ ) -> BaseOutputParser:
+ return cls(model=model, distance_func_name=distance_func_name, **kwargs)
+
+ def parse(self, text: str) -> str:
+ text = text.strip()
+ start = text.find("NeuralArray(")
+ _sql_str_compl = text
+ if start > 0:
+ _matched = text[text.find("NeuralArray(") + len("NeuralArray(") :]
+ end = _matched.find(")") + start + len("NeuralArray(") + 1
+ entity = _matched[: _matched.find(")")]
+ vecs = self.model.embed_query(entity)
+ vecs_str = "[" + ",".join(map(str, vecs)) + "]"
+ _sql_str_compl = text.replace("DISTANCE", self.distance_func_name).replace(
+ text[start:end], vecs_str
+ )
+ if _sql_str_compl[-1] == ";":
+ _sql_str_compl = _sql_str_compl[:-1]
+ return _sql_str_compl
+
+
+class VectorSQLRetrieveAllOutputParser(VectorSQLOutputParser):
+ """Based on VectorSQLOutputParser
+ It also modify the SQL to get all columns
+ """
+
+ @property
+ def _type(self) -> str:
+ return "vector_sql_retrieve_all_parser"
+
+ def parse(self, text: str) -> str:
+ text = text.strip()
+ start = text.upper().find("SELECT")
+ if start >= 0:
+ end = text.upper().find("FROM")
+ text = text.replace(text[start + len("SELECT") + 1 : end - 1], "*")
+ return super().parse(text)
+
+
+def _try_eval(x: Any) -> Any:
+ try:
+ return eval(x)
+ except Exception:
+ return x
+
+
+def get_result_from_sqldb(
+ db: SQLDatabase, cmd: str
+) -> Union[str, List[Dict[str, Any]], Dict[str, Any]]:
+ result = db._execute(cmd, fetch="all") # type: ignore
+ if isinstance(result, list):
+ return [{k: _try_eval(v) for k, v in dict(d._asdict()).items()} for d in result]
+ else:
+ return {
+ k: _try_eval(v) for k, v in dict(result._asdict()).items() # type: ignore
+ }
+
+
+class VectorSQLDatabaseChain(SQLDatabaseChain):
+ """Chain for interacting with Vector SQL Database.
+
+ Example:
+ .. code-block:: python
+
+ from langchain_experimental.sql import SQLDatabaseChain
+ from langchain import OpenAI, SQLDatabase, OpenAIEmbeddings
+ db = SQLDatabase(...)
+ db_chain = VectorSQLDatabaseChain.from_llm(OpenAI(), db, OpenAIEmbeddings())
+
+ *Security note*: Make sure that the database connection uses credentials
+ that are narrowly-scoped to only include the permissions this chain needs.
+ Failure to do so may result in data corruption or loss, since this chain may
+ attempt commands like `DROP TABLE` or `INSERT` if appropriately prompted.
+ The best way to guard against such negative outcomes is to (as appropriate)
+ limit the permissions granted to the credentials used with this chain.
+ This issue shows an example negative outcome if these steps are not taken:
+ https://github.com/langchain-ai/langchain/issues/5923
+ """
+
+ sql_cmd_parser: VectorSQLOutputParser
+ """Parser for Vector SQL"""
+ native_format: bool = False
+ """If return_direct, controls whether to return in python native format"""
+
+ def _call(
+ self,
+ inputs: Dict[str, Any],
+ run_manager: Optional[CallbackManagerForChainRun] = None,
+ ) -> Dict[str, Any]:
+ _run_manager = run_manager or CallbackManagerForChainRun.get_noop_manager()
+ input_text = f"{inputs[self.input_key]}\nSQLQuery:"
+ _run_manager.on_text(input_text, verbose=self.verbose)
+ # If not present, then defaults to None which is all tables.
+ table_names_to_use = inputs.get("table_names_to_use")
+ table_info = self.database.get_table_info(table_names=table_names_to_use)
+ llm_inputs = {
+ "input": input_text,
+ "top_k": str(self.top_k),
+ "dialect": self.database.dialect,
+ "table_info": table_info,
+ "stop": ["\nSQLResult:"],
+ }
+ intermediate_steps: List = []
+ try:
+ intermediate_steps.append(llm_inputs) # input: sql generation
+ llm_out = self.llm_chain.predict(
+ callbacks=_run_manager.get_child(),
+ **llm_inputs,
+ )
+ sql_cmd = self.sql_cmd_parser.parse(llm_out)
+ if self.return_sql:
+ return {self.output_key: sql_cmd}
+ if not self.use_query_checker:
+ _run_manager.on_text(llm_out, color="green", verbose=self.verbose)
+ intermediate_steps.append(
+ llm_out
+ ) # output: sql generation (no checker)
+ intermediate_steps.append({"sql_cmd": llm_out}) # input: sql exec
+ result = get_result_from_sqldb(self.database, sql_cmd)
+ intermediate_steps.append(str(result)) # output: sql exec
+ else:
+ query_checker_prompt = self.query_checker_prompt or PromptTemplate(
+ template=QUERY_CHECKER, input_variables=["query", "dialect"]
+ )
+ query_checker_chain = LLMChain(
+ llm=self.llm_chain.llm,
+ prompt=query_checker_prompt,
+ output_parser=self.llm_chain.output_parser,
+ )
+ query_checker_inputs = {
+ "query": llm_out,
+ "dialect": self.database.dialect,
+ }
+ checked_llm_out = query_checker_chain.predict(
+ callbacks=_run_manager.get_child(), **query_checker_inputs
+ )
+ checked_sql_command = self.sql_cmd_parser.parse(checked_llm_out)
+ intermediate_steps.append(
+ checked_llm_out
+ ) # output: sql generation (checker)
+ _run_manager.on_text(
+ checked_llm_out, color="green", verbose=self.verbose
+ )
+ intermediate_steps.append(
+ {"sql_cmd": checked_llm_out}
+ ) # input: sql exec
+ result = get_result_from_sqldb(self.database, checked_sql_command)
+ intermediate_steps.append(str(result)) # output: sql exec
+ llm_out = checked_llm_out
+ sql_cmd = checked_sql_command
+
+ _run_manager.on_text("\nSQLResult: ", verbose=self.verbose)
+ _run_manager.on_text(str(result), color="yellow", verbose=self.verbose)
+ # If return direct, we just set the final result equal to
+ # the result of the sql query result, otherwise try to get a human readable
+ # final answer
+ if self.return_direct:
+ final_result = result
+ else:
+ _run_manager.on_text("\nAnswer:", verbose=self.verbose)
+ input_text += f"{llm_out}\nSQLResult: {result}\nAnswer:"
+ llm_inputs["input"] = input_text
+ intermediate_steps.append(llm_inputs) # input: final answer
+ final_result = self.llm_chain.predict(
+ callbacks=_run_manager.get_child(),
+ **llm_inputs,
+ ).strip()
+ intermediate_steps.append(final_result) # output: final answer
+ _run_manager.on_text(final_result, color="green", verbose=self.verbose)
+ chain_result: Dict[str, Any] = {self.output_key: final_result}
+ if self.return_intermediate_steps:
+ chain_result[INTERMEDIATE_STEPS_KEY] = intermediate_steps
+ return chain_result
+ except Exception as exc:
+ # Append intermediate steps to exception, to aid in logging and later
+ # improvement of few shot prompt seeds
+ exc.intermediate_steps = intermediate_steps # type: ignore
+ raise exc
+
+ @property
+ def _chain_type(self) -> str:
+ return "vector_sql_database_chain"
+
+ @classmethod
+ def from_llm(
+ cls,
+ llm: BaseLanguageModel,
+ db: SQLDatabase,
+ prompt: Optional[BasePromptTemplate] = None,
+ sql_cmd_parser: Optional[VectorSQLOutputParser] = None,
+ **kwargs: Any,
+ ) -> VectorSQLDatabaseChain:
+ assert sql_cmd_parser, "`sql_cmd_parser` must be set in VectorSQLDatabaseChain."
+ prompt = prompt or SQL_PROMPTS.get(db.dialect, PROMPT)
+ llm_chain = LLMChain(llm=llm, prompt=prompt)
+ return cls(
+ llm_chain=llm_chain, database=db, sql_cmd_parser=sql_cmd_parser, **kwargs
+ )
diff --git a/libs/experimental/poetry.lock b/libs/experimental/poetry.lock
index b0d5b9139a..9e8cf9f1af 100644
--- a/libs/experimental/poetry.lock
+++ b/libs/experimental/poetry.lock
@@ -1245,6 +1245,7 @@ optional = false
python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, !=3.5.*, !=3.6.*"
files = [
{file = "jsonpointer-2.4-py2.py3-none-any.whl", hash = "sha256:15d51bba20eea3165644553647711d150376234112651b4f1811022aecad7d7a"},
+ {file = "jsonpointer-2.4.tar.gz", hash = "sha256:585cee82b70211fa9e6043b7bb89db6e1aa49524340dde8ad6b63206ea689d88"},
]
[[package]]
@@ -3752,6 +3753,31 @@ files = [
{file = "types_PyYAML-6.0.12.11-py3-none-any.whl", hash = "sha256:a461508f3096d1d5810ec5ab95d7eeecb651f3a15b71959999988942063bf01d"},
]
+[[package]]
+name = "types-requests"
+version = "2.31.0.2"
+description = "Typing stubs for requests"
+optional = false
+python-versions = "*"
+files = [
+ {file = "types-requests-2.31.0.2.tar.gz", hash = "sha256:6aa3f7faf0ea52d728bb18c0a0d1522d9bfd8c72d26ff6f61bfc3d06a411cf40"},
+ {file = "types_requests-2.31.0.2-py3-none-any.whl", hash = "sha256:56d181c85b5925cbc59f4489a57e72a8b2166f18273fd8ba7b6fe0c0b986f12a"},
+]
+
+[package.dependencies]
+types-urllib3 = "*"
+
+[[package]]
+name = "types-urllib3"
+version = "1.26.25.14"
+description = "Typing stubs for urllib3"
+optional = false
+python-versions = "*"
+files = [
+ {file = "types-urllib3-1.26.25.14.tar.gz", hash = "sha256:229b7f577c951b8c1b92c1bc2b2fdb0b49847bd2af6d1cc2a2e3dd340f3bda8f"},
+ {file = "types_urllib3-1.26.25.14-py3-none-any.whl", hash = "sha256:9683bbb7fb72e32bfe9d2be6e04875fbe1b3eeec3cbb4ea231435aa7fd6b4f0e"},
+]
+
[[package]]
name = "typing-extensions"
version = "4.7.1"
@@ -3995,4 +4021,4 @@ extended-testing = ["faker", "presidio-analyzer", "presidio-anonymizer"]
[metadata]
lock-version = "2.0"
python-versions = ">=3.8.1,<4.0"
-content-hash = "66ac482bd05eb74414210ac28fc1e8dae1a9928a4a1314e1326fada3551aa8ad"
+content-hash = "443e88f690572715cf58671e4480a006574c7141a1258dff0a0818b954184901"
diff --git a/libs/experimental/pyproject.toml b/libs/experimental/pyproject.toml
index 8e876c392c..272e5b6ad2 100644
--- a/libs/experimental/pyproject.toml
+++ b/libs/experimental/pyproject.toml
@@ -1,6 +1,6 @@
[tool.poetry]
name = "langchain-experimental"
-version = "0.0.14"
+version = "0.0.16"
description = "Building applications with LLMs through composability"
authors = []
license = "MIT"
@@ -23,6 +23,7 @@ black = "^23.1.0"
[tool.poetry.group.typing.dependencies]
mypy = "^0.991"
types-pyyaml = "^6.0.12.2"
+types-requests = "^2.28.11.5"
[tool.poetry.group.dev.dependencies]
jupyter = "^1.0.0"
diff --git a/libs/experimental/tests/unit_tests/test_reversible_data_anonymizer.py b/libs/experimental/tests/unit_tests/test_reversible_data_anonymizer.py
new file mode 100644
index 0000000000..9484a0e9dc
--- /dev/null
+++ b/libs/experimental/tests/unit_tests/test_reversible_data_anonymizer.py
@@ -0,0 +1,154 @@
+import os
+from typing import Iterator, List
+
+import pytest
+
+
+@pytest.fixture(scope="module", autouse=True)
+def check_spacy_model() -> Iterator[None]:
+ import spacy
+
+ if not spacy.util.is_package("en_core_web_lg"):
+ pytest.skip(reason="Spacy model 'en_core_web_lg' not installed")
+ yield
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+@pytest.mark.parametrize(
+ "analyzed_fields,should_contain",
+ [(["PERSON"], False), (["PHONE_NUMBER"], True), (None, False)],
+)
+def test_anonymize(analyzed_fields: List[str], should_contain: bool) -> None:
+ """Test anonymizing a name in a simple sentence"""
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ text = "Hello, my name is John Doe."
+ anonymizer = PresidioReversibleAnonymizer(analyzed_fields=analyzed_fields)
+ anonymized_text = anonymizer.anonymize(text)
+ assert ("John Doe" in anonymized_text) == should_contain
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+def test_anonymize_multiple() -> None:
+ """Test anonymizing multiple items in a sentence"""
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ text = "John Smith's phone number is 313-666-7440 and email is johnsmith@gmail.com"
+ anonymizer = PresidioReversibleAnonymizer()
+ anonymized_text = anonymizer.anonymize(text)
+ for phrase in ["John Smith", "313-666-7440", "johnsmith@gmail.com"]:
+ assert phrase not in anonymized_text
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+def test_anonymize_with_custom_operator() -> None:
+ """Test anonymize a name with a custom operator"""
+ from presidio_anonymizer.entities import OperatorConfig
+
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ custom_operator = {"PERSON": OperatorConfig("replace", {"new_value": ""})}
+ anonymizer = PresidioReversibleAnonymizer(operators=custom_operator)
+
+ text = "Jane Doe was here."
+
+ anonymized_text = anonymizer.anonymize(text)
+ assert anonymized_text == " was here."
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+def test_add_recognizer_operator() -> None:
+ """
+ Test add recognizer and anonymize a new type of entity and with a custom operator
+ """
+ from presidio_analyzer import PatternRecognizer
+ from presidio_anonymizer.entities import OperatorConfig
+
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ anonymizer = PresidioReversibleAnonymizer(analyzed_fields=[])
+ titles_list = ["Sir", "Madam", "Professor"]
+ custom_recognizer = PatternRecognizer(
+ supported_entity="TITLE", deny_list=titles_list
+ )
+ anonymizer.add_recognizer(custom_recognizer)
+
+ # anonymizing with custom recognizer
+ text = "Madam Jane Doe was here."
+ anonymized_text = anonymizer.anonymize(text)
+ assert anonymized_text == " Jane Doe was here."
+
+ # anonymizing with custom recognizer and operator
+ custom_operator = {"TITLE": OperatorConfig("replace", {"new_value": "Dear"})}
+ anonymizer.add_operators(custom_operator)
+ anonymized_text = anonymizer.anonymize(text)
+ assert anonymized_text == "Dear Jane Doe was here."
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+def test_deanonymizer_mapping() -> None:
+ """Test if deanonymizer mapping is correctly populated"""
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ anonymizer = PresidioReversibleAnonymizer(
+ analyzed_fields=["PERSON", "PHONE_NUMBER", "EMAIL_ADDRESS", "CREDIT_CARD"]
+ )
+
+ anonymizer.anonymize("Hello, my name is John Doe and my number is 444 555 6666.")
+
+ # ["PERSON", "PHONE_NUMBER"]
+ assert len(anonymizer.deanonymizer_mapping.keys()) == 2
+ assert "John Doe" in anonymizer.deanonymizer_mapping.get("PERSON", {}).values()
+ assert (
+ "444 555 6666"
+ in anonymizer.deanonymizer_mapping.get("PHONE_NUMBER", {}).values()
+ )
+
+ text_to_anonymize = (
+ "And my name is Jane Doe, my email is jane@gmail.com and "
+ "my credit card is 4929 5319 6292 5362."
+ )
+ anonymizer.anonymize(text_to_anonymize)
+
+ # ["PERSON", "PHONE_NUMBER", "EMAIL_ADDRESS", "CREDIT_CARD"]
+ assert len(anonymizer.deanonymizer_mapping.keys()) == 4
+ assert "Jane Doe" in anonymizer.deanonymizer_mapping.get("PERSON", {}).values()
+ assert (
+ "jane@gmail.com"
+ in anonymizer.deanonymizer_mapping.get("EMAIL_ADDRESS", {}).values()
+ )
+ assert (
+ "4929 5319 6292 5362"
+ in anonymizer.deanonymizer_mapping.get("CREDIT_CARD", {}).values()
+ )
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+def test_deanonymize() -> None:
+ """Test deanonymizing a name in a simple sentence"""
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ text = "Hello, my name is John Doe."
+ anonymizer = PresidioReversibleAnonymizer(analyzed_fields=["PERSON"])
+ anonymized_text = anonymizer.anonymize(text)
+ deanonymized_text = anonymizer.deanonymize(anonymized_text)
+ assert deanonymized_text == text
+
+
+@pytest.mark.requires("presidio_analyzer", "presidio_anonymizer", "faker")
+def test_save_load_deanonymizer_mapping() -> None:
+ from langchain_experimental.data_anonymizer import PresidioReversibleAnonymizer
+
+ anonymizer = PresidioReversibleAnonymizer(analyzed_fields=["PERSON"])
+ anonymizer.anonymize("Hello, my name is John Doe.")
+ try:
+ anonymizer.save_deanonymizer_mapping("test_file.json")
+ assert os.path.isfile("test_file.json")
+
+ anonymizer = PresidioReversibleAnonymizer()
+ anonymizer.load_deanonymizer_mapping("test_file.json")
+
+ assert "John Doe" in anonymizer.deanonymizer_mapping.get("PERSON", {}).values()
+
+ finally:
+ os.remove("test_file.json")
diff --git a/libs/langchain/langchain/callbacks/tracers/evaluation.py b/libs/langchain/langchain/callbacks/tracers/evaluation.py
index 5b178c84e6..1cf205e3d2 100644
--- a/libs/langchain/langchain/callbacks/tracers/evaluation.py
+++ b/libs/langchain/langchain/callbacks/tracers/evaluation.py
@@ -2,29 +2,20 @@
from __future__ import annotations
import logging
-from concurrent.futures import Future, ThreadPoolExecutor, wait
+from concurrent.futures import Future, ThreadPoolExecutor
from typing import Any, Dict, List, Optional, Sequence, Set, Union
from uuid import UUID
import langsmith
from langsmith import schemas as langsmith_schemas
-from langchain.callbacks.manager import tracing_v2_enabled
+from langchain.callbacks import manager
+from langchain.callbacks.tracers import langchain as langchain_tracer
from langchain.callbacks.tracers.base import BaseTracer
-from langchain.callbacks.tracers.langchain import _get_client
from langchain.callbacks.tracers.schemas import Run
logger = logging.getLogger(__name__)
-_TRACERS: List[EvaluatorCallbackHandler] = []
-
-
-def wait_for_all_evaluators() -> None:
- """Wait for all tracers to finish."""
- global _TRACERS
- for tracer in _TRACERS:
- tracer.wait_for_futures()
-
class EvaluatorCallbackHandler(BaseTracer):
"""A tracer that runs a run evaluator whenever a run is persisted.
@@ -79,17 +70,13 @@ class EvaluatorCallbackHandler(BaseTracer):
self.example_id = (
UUID(example_id) if isinstance(example_id, str) else example_id
)
- self.client = client or _get_client()
+ self.client = client or langchain_tracer.get_client()
self.evaluators = evaluators
- self.executor = ThreadPoolExecutor(
- max_workers=max(max_workers or len(evaluators), 1)
- )
+ self.max_workers = max_workers or len(evaluators)
self.futures: Set[Future] = set()
self.skip_unfinished = skip_unfinished
self.project_name = project_name
self.logged_feedback: Dict[str, List[langsmith_schemas.Feedback]] = {}
- global _TRACERS
- _TRACERS.append(self)
def _evaluate_in_project(self, run: Run, evaluator: langsmith.RunEvaluator) -> None:
"""Evaluate the run in the project.
@@ -105,7 +92,7 @@ class EvaluatorCallbackHandler(BaseTracer):
try:
if self.project_name is None:
feedback = self.client.evaluate_run(run, evaluator)
- with tracing_v2_enabled(
+ with manager.tracing_v2_enabled(
project_name=self.project_name, tags=["eval"], client=self.client
):
feedback = self.client.evaluate_run(run, evaluator)
@@ -133,14 +120,15 @@ class EvaluatorCallbackHandler(BaseTracer):
return
run_ = run.copy()
run_.reference_example_id = self.example_id
- for evaluator in self.evaluators:
- self.futures.add(
- self.executor.submit(self._evaluate_in_project, run_, evaluator)
- )
-
- def wait_for_futures(self) -> None:
- """Wait for all futures to complete."""
- futures = list(self.futures)
- wait(futures)
- for future in futures:
- self.futures.remove(future)
+ if self.max_workers > 0:
+ with ThreadPoolExecutor(max_workers=self.max_workers) as executor:
+ list(
+ executor.map(
+ self._evaluate_in_project,
+ [run_ for _ in range(len(self.evaluators))],
+ self.evaluators,
+ )
+ )
+ else:
+ for evaluator in self.evaluators:
+ self._evaluate_in_project(run_, evaluator)
diff --git a/libs/langchain/langchain/callbacks/tracers/langchain.py b/libs/langchain/langchain/callbacks/tracers/langchain.py
index 0f57697721..0e6393c78b 100644
--- a/libs/langchain/langchain/callbacks/tracers/langchain.py
+++ b/libs/langchain/langchain/callbacks/tracers/langchain.py
@@ -42,7 +42,7 @@ def wait_for_all_tracers() -> None:
tracer.wait_for_futures()
-def _get_client() -> Client:
+def get_client() -> Client:
"""Get the client."""
global _CLIENT
if _CLIENT is None:
@@ -83,7 +83,7 @@ class LangChainTracer(BaseTracer):
_EXECUTORS.append(self.executor)
else:
self.executor = None
- self.client = client or _get_client()
+ self.client = client or get_client()
self._futures: Set[Future] = set()
self.tags = tags or []
global _TRACERS
diff --git a/libs/langchain/langchain/chat_loaders/__init__.py b/libs/langchain/langchain/chat_loaders/__init__.py
index 594d87344d..7547ddcecc 100644
--- a/libs/langchain/langchain/chat_loaders/__init__.py
+++ b/libs/langchain/langchain/chat_loaders/__init__.py
@@ -1,6 +1,19 @@
-"""Load chat messages from common communications platforms for finetuning.
+"""**Chat Loaders** load chat messages from common communications platforms.
-This module provides functions to load chat messages from various
+Load chat messages from various
communications platforms such as Facebook Messenger, Telegram, and
-WhatsApp. The loaded chat messages can be used for finetuning models.
-"""
+WhatsApp. The loaded chat messages can be used for fine-tuning models.
+
+**Class hierarchy:**
+
+.. code-block::
+
+ BaseChatLoader --> <name>ChatLoader # Examples: WhatsAppChatLoader, IMessageChatLoader
+
+**Main helpers:**
+
+.. code-block::
+
+ ChatSession
+
+""" # noqa: E501
diff --git a/libs/langchain/langchain/chat_loaders/base.py b/libs/langchain/langchain/chat_loaders/base.py
index 418ba15d2f..6e1f37ca9a 100644
--- a/libs/langchain/langchain/chat_loaders/base.py
+++ b/libs/langchain/langchain/chat_loaders/base.py
@@ -1,10 +1,3 @@
-"""Base definitions for chat loaders.
-
-A chat loader is a class that loads chat messages from an external
-source such as a file or a database. The chat messages can then be
-used for finetuning.
-"""
-
from abc import ABC, abstractmethod
from typing import Iterator, List, Sequence, TypedDict
@@ -12,7 +5,7 @@ from langchain.schema.messages import BaseMessage
class ChatSession(TypedDict):
- """A chat session represents a single
+ """Chat Session represents a single
conversation, channel, or other group of messages."""
messages: Sequence[BaseMessage]
diff --git a/libs/langchain/langchain/chat_loaders/facebook_messenger.py b/libs/langchain/langchain/chat_loaders/facebook_messenger.py
index 5864c32740..bfdc0155c7 100644
--- a/libs/langchain/langchain/chat_loaders/facebook_messenger.py
+++ b/libs/langchain/langchain/chat_loaders/facebook_messenger.py
@@ -10,7 +10,7 @@ logger = logging.getLogger(__file__)
class SingleFileFacebookMessengerChatLoader(BaseChatLoader):
- """A chat loader for loading Facebook Messenger chat data from a single file.
+ """Load `Facebook Messenger` chat data from a single file.
Args:
path (Union[Path, str]): The path to the chat file.
@@ -45,7 +45,7 @@ class SingleFileFacebookMessengerChatLoader(BaseChatLoader):
class FolderFacebookMessengerChatLoader(BaseChatLoader):
- """A chat loader for loading Facebook Messenger chat data from a folder.
+ """Load `Facebook Messenger` chat data from a folder.
Args:
path (Union[str, Path]): The path to the directory
diff --git a/libs/langchain/langchain/chat_loaders/gmail.py b/libs/langchain/langchain/chat_loaders/gmail.py
index 4e88accdee..94a3c5617e 100644
--- a/libs/langchain/langchain/chat_loaders/gmail.py
+++ b/libs/langchain/langchain/chat_loaders/gmail.py
@@ -62,7 +62,7 @@ def _get_message_data(service: Any, message: Any) -> ChatSession:
class GMailLoader(BaseChatLoader):
- """This loader goes over how to load data from GMail.
+ """Load data from `GMail`.
There are many ways you could want to load data from GMail.
This loader is currently fairly opinionated in how to do so.
diff --git a/libs/langchain/langchain/chat_loaders/imessage.py b/libs/langchain/langchain/chat_loaders/imessage.py
index ff9a06142c..d6c02f1e53 100644
--- a/libs/langchain/langchain/chat_loaders/imessage.py
+++ b/libs/langchain/langchain/chat_loaders/imessage.py
@@ -1,14 +1,3 @@
-"""IMessage Chat Loader.
-
-This class is used to load chat sessions from the iMessage chat.db SQLite file.
-It only works on macOS when you have iMessage enabled and have the chat.db file.
-
-The chat.db file is likely located at ~/Library/Messages/chat.db. However, your
-terminal may not have permission to access this file. To resolve this, you can
-copy the file to a different location, change the permissions of the file, or
-grant full disk access for your terminal emulator in System Settings > Security
-and Privacy > Full Disk Access.
-"""
from __future__ import annotations
from pathlib import Path
@@ -22,6 +11,17 @@ if TYPE_CHECKING:
class IMessageChatLoader(chat_loaders.BaseChatLoader):
+ """Load chat sessions from the `iMessage` chat.db SQLite file.
+
+ It only works on macOS when you have iMessage enabled and have the chat.db file.
+
+ The chat.db file is likely located at ~/Library/Messages/chat.db. However, your
+ terminal may not have permission to access this file. To resolve this, you can
+ copy the file to a different location, change the permissions of the file, or
+ grant full disk access for your terminal emulator in System Settings > Security
+ and Privacy > Full Disk Access.
+ """
+
def __init__(self, path: Optional[Union[str, Path]] = None):
"""
Initialize the IMessageChatLoader.
diff --git a/libs/langchain/langchain/chat_loaders/slack.py b/libs/langchain/langchain/chat_loaders/slack.py
index 261289bb43..0bbd503979 100644
--- a/libs/langchain/langchain/chat_loaders/slack.py
+++ b/libs/langchain/langchain/chat_loaders/slack.py
@@ -12,6 +12,8 @@ logger = logging.getLogger(__name__)
class SlackChatLoader(chat_loaders.BaseChatLoader):
+ """Load `Slack` conversations from a dump zip file."""
+
def __init__(
self,
path: Union[str, Path],
diff --git a/libs/langchain/langchain/chat_loaders/telegram.py b/libs/langchain/langchain/chat_loaders/telegram.py
index 786dad7278..5f0bbfa324 100644
--- a/libs/langchain/langchain/chat_loaders/telegram.py
+++ b/libs/langchain/langchain/chat_loaders/telegram.py
@@ -1,6 +1,7 @@
import json
import logging
import os
+import tempfile
import zipfile
from pathlib import Path
from typing import Iterator, List, Union
@@ -12,8 +13,7 @@ logger = logging.getLogger(__name__)
class TelegramChatLoader(chat_loaders.BaseChatLoader):
- """A loading utility for converting telegram conversations
- to LangChain chat messages.
+ """Load `telegram` conversations to LangChain chat messages.
To export, use the Telegram Desktop app from
https://desktop.telegram.org/, select a conversation, click the three dots
@@ -136,7 +136,8 @@ class TelegramChatLoader(chat_loaders.BaseChatLoader):
with zipfile.ZipFile(path) as zip_file:
for file in zip_file.namelist():
if file.endswith((".html", ".json")):
- yield zip_file.extract(file)
+ with tempfile.TemporaryDirectory() as temp_dir:
+ yield zip_file.extract(file, path=temp_dir)
def lazy_load(self) -> Iterator[chat_loaders.ChatSession]:
"""Lazy load the messages from the chat file and yield them
diff --git a/libs/langchain/langchain/chat_loaders/whatsapp.py b/libs/langchain/langchain/chat_loaders/whatsapp.py
index c911e262c6..e2518ab44d 100644
--- a/libs/langchain/langchain/chat_loaders/whatsapp.py
+++ b/libs/langchain/langchain/chat_loaders/whatsapp.py
@@ -12,6 +12,8 @@ logger = logging.getLogger(__name__)
class WhatsAppChatLoader(chat_loaders.BaseChatLoader):
+ """Load `WhatsApp` conversations from a dump zip file or directory."""
+
def __init__(self, path: str):
"""Initialize the WhatsAppChatLoader.
diff --git a/libs/langchain/langchain/document_loaders/url_playwright.py b/libs/langchain/langchain/document_loaders/url_playwright.py
index 16f5b00fd3..7aa60cf188 100644
--- a/libs/langchain/langchain/document_loaders/url_playwright.py
+++ b/libs/langchain/langchain/document_loaders/url_playwright.py
@@ -8,7 +8,9 @@ from langchain.docstore.document import Document
from langchain.document_loaders.base import BaseLoader
if TYPE_CHECKING:
- from playwright.async_api import AsyncBrowser, AsyncPage, AsyncResponse
+ from playwright.async_api import Browser as AsyncBrowser
+ from playwright.async_api import Page as AsyncPage
+ from playwright.async_api import Response as AsyncResponse
from playwright.sync_api import Browser, Page, Response
@@ -155,6 +157,9 @@ class PlaywrightURLLoader(BaseLoader):
try:
page = browser.new_page()
response = page.goto(url)
+ if response is None:
+ raise ValueError(f"page.goto() returned None for url {url}")
+
text = self.evaluator.evaluate(page, browser, response)
metadata = {"source": url}
docs.append(Document(page_content=text, metadata=metadata))
@@ -185,6 +190,9 @@ class PlaywrightURLLoader(BaseLoader):
try:
page = await browser.new_page()
response = await page.goto(url)
+ if response is None:
+ raise ValueError(f"page.goto() returned None for url {url}")
+
text = await self.evaluator.evaluate_async(page, browser, response)
metadata = {"source": url}
docs.append(Document(page_content=text, metadata=metadata))
diff --git a/libs/langchain/langchain/embeddings/__init__.py b/libs/langchain/langchain/embeddings/__init__.py
index 87cb5e90d5..e8aa683a9a 100644
--- a/libs/langchain/langchain/embeddings/__init__.py
+++ b/libs/langchain/langchain/embeddings/__init__.py
@@ -35,6 +35,7 @@ from langchain.embeddings.gpt4all import GPT4AllEmbeddings
from langchain.embeddings.huggingface import (
HuggingFaceBgeEmbeddings,
HuggingFaceEmbeddings,
+ HuggingFaceInferenceAPIEmbeddings,
HuggingFaceInstructEmbeddings,
)
from langchain.embeddings.huggingface_hub import HuggingFaceHubEmbeddings
@@ -69,6 +70,7 @@ __all__ = [
"CohereEmbeddings",
"ElasticsearchEmbeddings",
"HuggingFaceEmbeddings",
+ "HuggingFaceInferenceAPIEmbeddings",
"JinaEmbeddings",
"LlamaCppEmbeddings",
"HuggingFaceHubEmbeddings",
diff --git a/libs/langchain/langchain/embeddings/huggingface.py b/libs/langchain/langchain/embeddings/huggingface.py
index 52afabd79b..a91d643793 100644
--- a/libs/langchain/langchain/embeddings/huggingface.py
+++ b/libs/langchain/langchain/embeddings/huggingface.py
@@ -1,5 +1,7 @@
from typing import Any, Dict, List, Optional
+import requests
+
from langchain.embeddings.base import Embeddings
from langchain.pydantic_v1 import BaseModel, Extra, Field
@@ -58,7 +60,7 @@ class HuggingFaceEmbeddings(BaseModel, Embeddings):
except ImportError as exc:
raise ImportError(
"Could not import sentence_transformers python package. "
- "Please install it with `pip install sentence_transformers`."
+ "Please install it with `pip install sentence-transformers`."
) from exc
self.client = sentence_transformers.SentenceTransformer(
@@ -266,3 +268,71 @@ class HuggingFaceBgeEmbeddings(BaseModel, Embeddings):
self.query_instruction + text, **self.encode_kwargs
)
return embedding.tolist()
+
+
+class HuggingFaceInferenceAPIEmbeddings(BaseModel, Embeddings):
+ """Embed texts using the HuggingFace API.
+
+ Requires a HuggingFace Inference API key and a model name.
+ """
+
+ api_key: str
+ """Your API key for the HuggingFace Inference API."""
+ model_name: str = "sentence-transformers/all-MiniLM-L6-v2"
+ """The name of the model to use for text embeddings."""
+
+ @property
+ def _api_url(self) -> str:
+ return (
+ "https://api-inference.huggingface.co"
+ "/pipeline"
+ "/feature-extraction"
+ f"/{self.model_name}"
+ )
+
+ @property
+ def _headers(self) -> dict:
+ return {"Authorization": f"Bearer {self.api_key}"}
+
+ def embed_documents(self, texts: List[str]) -> List[List[float]]:
+ """Get the embeddings for a list of texts.
+
+ Args:
+ texts (Documents): A list of texts to get embeddings for.
+
+ Returns:
+ Embedded texts as List[List[float]], where each inner List[float]
+ corresponds to a single input text.
+
+ Example:
+ .. code-block:: python
+
+ from langchain.embeddings import HuggingFaceInferenceAPIEmbeddings
+
+ hf_embeddings = HuggingFaceInferenceAPIEmbeddings(
+ api_key="your_api_key",
+ model_name="sentence-transformers/all-MiniLM-l6-v2"
+ )
+ texts = ["Hello, world!", "How are you?"]
+ hf_embeddings.embed_documents(texts)
+ """
+ response = requests.post(
+ self._api_url,
+ headers=self._headers,
+ json={
+ "inputs": texts,
+ "options": {"wait_for_model": True, "use_cache": True},
+ },
+ )
+ return response.json()
+
+ def embed_query(self, text: str) -> List[float]:
+ """Compute query embeddings using a HuggingFace transformer model.
+
+ Args:
+ text: The text to embed.
+
+ Returns:
+ Embeddings for the text.
+ """
+ return self.embed_documents([text])[0]
diff --git a/libs/langchain/langchain/graphs/graph_document.py b/libs/langchain/langchain/graphs/graph_document.py
new file mode 100644
index 0000000000..9f72a3ad8e
--- /dev/null
+++ b/libs/langchain/langchain/graphs/graph_document.py
@@ -0,0 +1,51 @@
+from __future__ import annotations
+
+from typing import List, Union
+
+from langchain.load.serializable import Serializable
+from langchain.pydantic_v1 import Field
+from langchain.schema import Document
+
+
+class Node(Serializable):
+ """Represents a node in a graph with associated properties.
+
+ Attributes:
+ id (Union[str, int]): A unique identifier for the node.
+ type (str): The type or label of the node, default is "Node".
+ properties (dict): Additional properties and metadata associated with the node.
+ """
+
+ id: Union[str, int]
+ type: str = "Node"
+ properties: dict = Field(default_factory=dict)
+
+
+class Relationship(Serializable):
+ """Represents a directed relationship between two nodes in a graph.
+
+ Attributes:
+ source (Node): The source node of the relationship.
+ target (Node): The target node of the relationship.
+ type (str): The type of the relationship.
+ properties (dict): Additional properties associated with the relationship.
+ """
+
+ source: Node
+ target: Node
+ type: str
+ properties: dict = Field(default_factory=dict)
+
+
+class GraphDocument(Serializable):
+ """Represents a graph document consisting of nodes and relationships.
+
+ Attributes:
+ nodes (List[Node]): A list of nodes in the graph.
+ relationships (List[Relationship]): A list of relationships in the graph.
+ source (Document): The document from which the graph information is derived.
+ """
+
+ nodes: List[Node]
+ relationships: List[Relationship]
+ source: Document
diff --git a/libs/langchain/langchain/graphs/neo4j_graph.py b/libs/langchain/langchain/graphs/neo4j_graph.py
index 02572b2d1a..256df9d26b 100644
--- a/libs/langchain/langchain/graphs/neo4j_graph.py
+++ b/libs/langchain/langchain/graphs/neo4j_graph.py
@@ -1,5 +1,7 @@
from typing import Any, Dict, List
+from langchain.graphs.graph_document import GraphDocument
+
node_properties_query = """
CALL apoc.meta.data()
YIELD label, other, elementType, type, property
@@ -99,3 +101,56 @@ class Neo4jGraph:
The relationships are the following:
{[el['output'] for el in relationships]}
"""
+
+ def add_graph_documents(
+ self, graph_documents: List[GraphDocument], include_source: bool = False
+ ) -> None:
+ """
+ Take GraphDocument as input as uses it to construct a graph.
+ """
+ for document in graph_documents:
+ include_docs_query = (
+ "CREATE (d:Document) "
+ "SET d.text = $document.page_content "
+ "SET d += $document.metadata "
+ "WITH d "
+ )
+ # Import nodes
+ self.query(
+ (
+ f"{include_docs_query if include_source else ''}"
+ "UNWIND $data AS row "
+ "CALL apoc.merge.node([row.type], {id: row.id}, "
+ "row.properties, {}) YIELD node "
+ f"{'MERGE (d)-[:MENTIONS]->(node) ' if include_source else ''}"
+ "RETURN distinct 'done' AS result"
+ ),
+ {
+ "data": [el.__dict__ for el in document.nodes],
+ "document": document.source.__dict__,
+ },
+ )
+ # Import relationships
+ self.query(
+ "UNWIND $data AS row "
+ "CALL apoc.merge.node([row.source_label], {id: row.source},"
+ "{}, {}) YIELD node as source "
+ "CALL apoc.merge.node([row.target_label], {id: row.target},"
+ "{}, {}) YIELD node as target "
+ "CALL apoc.merge.relationship(source, row.type, "
+ "{}, row.properties, target) YIELD rel "
+ "RETURN distinct 'done'",
+ {
+ "data": [
+ {
+ "source": el.source.id,
+ "source_label": el.source.type,
+ "target": el.target.id,
+ "target_label": el.target.type,
+ "type": el.type.replace(" ", "_").upper(),
+ "properties": el.properties,
+ }
+ for el in document.relationships
+ ]
+ },
+ )
diff --git a/libs/langchain/langchain/llms/bananadev.py b/libs/langchain/langchain/llms/bananadev.py
index f0659118d6..3a984a3cb2 100644
--- a/libs/langchain/langchain/llms/bananadev.py
+++ b/libs/langchain/langchain/llms/bananadev.py
@@ -15,6 +15,7 @@ class Banana(LLM):
To use, you should have the ``banana-dev`` python package installed,
and the environment variable ``BANANA_API_KEY`` set with your API key.
+ This is the team API key available in the Banana dashboard.
Any parameters that are valid to be passed to the call can be passed
in, even if not explicitly saved on this class.
@@ -23,10 +24,13 @@ class Banana(LLM):
.. code-block:: python
from langchain.llms import Banana
- banana = Banana(model_key="")
+ banana = Banana(model_key="", model_url_slug="")
"""
model_key: str = ""
+ """model key to use"""
+
+ model_url_slug: str = ""
"""model endpoint to use"""
model_kwargs: Dict[str, Any] = Field(default_factory=dict)
@@ -72,6 +76,7 @@ class Banana(LLM):
"""Get the identifying parameters."""
return {
**{"model_key": self.model_key},
+ **{"model_url_slug": self.model_url_slug},
**{"model_kwargs": self.model_kwargs},
}
@@ -89,7 +94,7 @@ class Banana(LLM):
) -> str:
"""Call to Banana endpoint."""
try:
- import banana_dev as banana
+ from banana_dev import Client
except ImportError:
raise ImportError(
"Could not import banana-dev python package. "
@@ -99,19 +104,25 @@ class Banana(LLM):
params = {**params, **kwargs}
api_key = self.banana_api_key
model_key = self.model_key
+ model_url_slug = self.model_url_slug
model_inputs = {
# a json specific to your model.
"prompt": prompt,
**params,
}
- response = banana.run(api_key, model_key, model_inputs)
+ model = Client(
+ # Found in main dashboard
+ api_key=api_key,
+ # Both found in model details page
+ model_key=model_key,
+ url=f"https://{model_url_slug}.run.banana.dev",
+ )
+ response, meta = model.call("/", model_inputs)
try:
- text = response["modelOutputs"][0]["output"]
+ text = response["outputs"]
except (KeyError, TypeError):
- returned = response["modelOutputs"][0]
raise ValueError(
- "Response should be of schema: {'output': 'text'}."
- f"\nResponse was: {returned}"
+ "Response should be of schema: {'outputs': 'text'}."
"\nTo fix this:"
"\n- fork the source repo of the Banana model"
"\n- modify app.py to return the above schema"
diff --git a/libs/langchain/langchain/llms/rwkv.py b/libs/langchain/langchain/llms/rwkv.py
index bb54c9d5de..8072b2b91b 100644
--- a/libs/langchain/langchain/llms/rwkv.py
+++ b/libs/langchain/langchain/llms/rwkv.py
@@ -121,7 +121,7 @@ class RWKV(LLM, BaseModel):
values["pipeline"] = PIPELINE(values["client"], values["tokens_path"])
except ImportError:
- raise ValueError(
+ raise ImportError(
"Could not import rwkv python package. "
"Please install it with `pip install rwkv`."
)
diff --git a/libs/langchain/langchain/memory/chat_message_histories/dynamodb.py b/libs/langchain/langchain/memory/chat_message_histories/dynamodb.py
index 704efa9ea7..06d7897dbd 100644
--- a/libs/langchain/langchain/memory/chat_message_histories/dynamodb.py
+++ b/libs/langchain/langchain/memory/chat_message_histories/dynamodb.py
@@ -1,5 +1,7 @@
+from __future__ import annotations
+
import logging
-from typing import Dict, List, Optional
+from typing import TYPE_CHECKING, Dict, List, Optional
from langchain.schema import (
BaseChatMessageHistory,
@@ -11,6 +13,9 @@ from langchain.schema.messages import (
messages_to_dict,
)
+if TYPE_CHECKING:
+ from boto3.session import Session
+
logger = logging.getLogger(__name__)
@@ -42,13 +47,21 @@ class DynamoDBChatMessageHistory(BaseChatMessageHistory):
endpoint_url: Optional[str] = None,
primary_key_name: str = "SessionId",
key: Optional[Dict[str, str]] = None,
+ boto3_session: Optional[Session] = None,
):
- import boto3
-
- if endpoint_url:
- client = boto3.resource("dynamodb", endpoint_url=endpoint_url)
+ if boto3_session:
+ client = boto3_session.resource("dynamodb")
else:
- client = boto3.resource("dynamodb")
+ try:
+ import boto3
+ except ImportError as e:
+ raise ImportError(
+ "Unable to import boto3, please install with `pip install boto3`."
+ ) from e
+ if endpoint_url:
+ client = boto3.resource("dynamodb", endpoint_url=endpoint_url)
+ else:
+ client = boto3.resource("dynamodb")
self.table = client.Table(table_name)
self.session_id = session_id
self.key: Dict = key or {primary_key_name: session_id}
@@ -56,7 +69,12 @@ class DynamoDBChatMessageHistory(BaseChatMessageHistory):
@property
def messages(self) -> List[BaseMessage]: # type: ignore
"""Retrieve the messages from DynamoDB"""
- from botocore.exceptions import ClientError
+ try:
+ from botocore.exceptions import ClientError
+ except ImportError as e:
+ raise ImportError(
+ "Unable to import botocore, please install with `pip install botocore`."
+ ) from e
response = None
try:
@@ -77,7 +95,12 @@ class DynamoDBChatMessageHistory(BaseChatMessageHistory):
def add_message(self, message: BaseMessage) -> None:
"""Append the message to the record in DynamoDB"""
- from botocore.exceptions import ClientError
+ try:
+ from botocore.exceptions import ClientError
+ except ImportError as e:
+ raise ImportError(
+ "Unable to import botocore, please install with `pip install botocore`."
+ ) from e
messages = messages_to_dict(self.messages)
_message = _message_to_dict(message)
@@ -90,7 +113,12 @@ class DynamoDBChatMessageHistory(BaseChatMessageHistory):
def clear(self) -> None:
"""Clear session memory from DynamoDB"""
- from botocore.exceptions import ClientError
+ try:
+ from botocore.exceptions import ClientError
+ except ImportError as e:
+ raise ImportError(
+ "Unable to import botocore, please install with `pip install botocore`."
+ ) from e
try:
self.table.delete_item(self.key)
diff --git a/libs/langchain/langchain/retrievers/self_query/base.py b/libs/langchain/langchain/retrievers/self_query/base.py
index 0251bff52c..62bce34847 100644
--- a/libs/langchain/langchain/retrievers/self_query/base.py
+++ b/libs/langchain/langchain/retrievers/self_query/base.py
@@ -16,6 +16,8 @@ from langchain.retrievers.self_query.milvus import MilvusTranslator
from langchain.retrievers.self_query.myscale import MyScaleTranslator
from langchain.retrievers.self_query.pinecone import PineconeTranslator
from langchain.retrievers.self_query.qdrant import QdrantTranslator
+from langchain.retrievers.self_query.supabase import SupabaseVectorTranslator
+from langchain.retrievers.self_query.vectara import VectaraTranslator
from langchain.retrievers.self_query.weaviate import WeaviateTranslator
from langchain.schema import BaseRetriever, Document
from langchain.schema.language_model import BaseLanguageModel
@@ -28,6 +30,8 @@ from langchain.vectorstores import (
MyScale,
Pinecone,
Qdrant,
+ SupabaseVectorStore,
+ Vectara,
VectorStore,
Weaviate,
)
@@ -41,11 +45,13 @@ def _get_builtin_translator(vectorstore: VectorStore) -> Visitor:
Chroma: ChromaTranslator,
DashVector: DashvectorTranslator,
Weaviate: WeaviateTranslator,
+ Vectara: VectaraTranslator,
Qdrant: QdrantTranslator,
MyScale: MyScaleTranslator,
DeepLake: DeepLakeTranslator,
ElasticsearchStore: ElasticsearchTranslator,
Milvus: MilvusTranslator,
+ SupabaseVectorStore: SupabaseVectorTranslator,
}
if vectorstore_cls not in BUILTIN_TRANSLATORS:
raise ValueError(
diff --git a/libs/langchain/langchain/retrievers/self_query/supabase.py b/libs/langchain/langchain/retrievers/self_query/supabase.py
new file mode 100644
index 0000000000..267e228fcd
--- /dev/null
+++ b/libs/langchain/langchain/retrievers/self_query/supabase.py
@@ -0,0 +1,97 @@
+from typing import Any, Dict, Tuple
+
+from langchain.chains.query_constructor.ir import (
+ Comparator,
+ Comparison,
+ Operation,
+ Operator,
+ StructuredQuery,
+ Visitor,
+)
+
+
+class SupabaseVectorTranslator(Visitor):
+ """Translate Langchain filters to Supabase PostgREST filters."""
+
+ allowed_operators = [Operator.AND, Operator.OR]
+ """Subset of allowed logical operators."""
+
+ allowed_comparators = [
+ Comparator.EQ,
+ Comparator.NE,
+ Comparator.GT,
+ Comparator.GTE,
+ Comparator.LT,
+ Comparator.LTE,
+ Comparator.LIKE,
+ ]
+ """Subset of allowed logical comparators."""
+
+ metadata_column = "metadata"
+
+ def _map_comparator(self, comparator: Comparator) -> str:
+ """
+ Maps Langchain comparator to PostgREST comparator:
+
+ https://postgrest.org/en/stable/references/api/tables_views.html#operators
+ """
+ postgrest_comparator = {
+ Comparator.EQ: "eq",
+ Comparator.NE: "neq",
+ Comparator.GT: "gt",
+ Comparator.GTE: "gte",
+ Comparator.LT: "lt",
+ Comparator.LTE: "lte",
+ Comparator.LIKE: "like",
+ }.get(comparator)
+
+ if postgrest_comparator is None:
+ raise Exception(
+ f"Comparator '{comparator}' is not currently "
+ "supported in Supabase Vector"
+ )
+
+ return postgrest_comparator
+
+ def _get_json_operator(self, value: Any) -> str:
+ if isinstance(value, str):
+ return "->>"
+ else:
+ return "->"
+
+ def visit_operation(self, operation: Operation) -> str:
+ args = [arg.accept(self) for arg in operation.arguments]
+ return f"{operation.operator.value}({','.join(args)})"
+
+ def visit_comparison(self, comparison: Comparison) -> str:
+ if isinstance(comparison.value, list):
+ return self.visit_operation(
+ Operation(
+ operator=Operator.AND,
+ arguments=(
+ Comparison(
+ comparator=comparison.comparator,
+ attribute=comparison.attribute,
+ value=value,
+ )
+ for value in comparison.value
+ ),
+ )
+ )
+
+ return ".".join(
+ [
+ f"{self.metadata_column}{self._get_json_operator(comparison.value)}{comparison.attribute}",
+ f"{self._map_comparator(comparison.comparator)}",
+ f"{comparison.value}",
+ ]
+ )
+
+ def visit_structured_query(
+ self, structured_query: StructuredQuery
+ ) -> Tuple[str, Dict[str, str]]:
+ if structured_query.filter is None:
+ kwargs = {}
+ else:
+ kwargs = {"postgrest_filter": structured_query.filter.accept(self)}
+ return structured_query.query, kwargs
diff --git a/libs/langchain/langchain/retrievers/self_query/vectara.py b/libs/langchain/langchain/retrievers/self_query/vectara.py
new file mode 100644
index 0000000000..73dc46ff59
--- /dev/null
+++ b/libs/langchain/langchain/retrievers/self_query/vectara.py
@@ -0,0 +1,69 @@
+from typing import Tuple, Union
+
+from langchain.chains.query_constructor.ir import (
+ Comparator,
+ Comparison,
+ Operation,
+ Operator,
+ StructuredQuery,
+ Visitor,
+)
+
+
+def process_value(value: Union[int, float, str]) -> str:
+ if isinstance(value, str):
+ return f"'{value}'"
+ else:
+ return str(value)
+
+
+class VectaraTranslator(Visitor):
+ """Translate `Vectara` internal query language elements to valid filters."""
+
+ allowed_operators = [Operator.AND, Operator.OR]
+ """Subset of allowed logical operators."""
+ allowed_comparators = [
+ Comparator.EQ,
+ Comparator.NE,
+ Comparator.GT,
+ Comparator.GTE,
+ Comparator.LT,
+ Comparator.LTE,
+ ]
+ """Subset of allowed logical comparators."""
+
+ def _format_func(self, func: Union[Operator, Comparator]) -> str:
+ map_dict = {
+ Operator.AND: " and ",
+ Operator.OR: " or ",
+ Comparator.EQ: "=",
+ Comparator.NE: "!=",
+ Comparator.GT: ">",
+ Comparator.GTE: ">=",
+ Comparator.LT: "<",
+ Comparator.LTE: "<=",
+ }
+ self._validate_func(func)
+ return map_dict[func]
+
+ def visit_operation(self, operation: Operation) -> str:
+ args = [arg.accept(self) for arg in operation.arguments]
+ operator = self._format_func(operation.operator)
+ return "( " + operator.join(args) + " )"
+
+ def visit_comparison(self, comparison: Comparison) -> str:
+ comparator = self._format_func(comparison.comparator)
+ processed_value = process_value(comparison.value)
+ attribute = comparison.attribute
+ return (
+ "( " + "doc." + attribute + " " + comparator + " " + processed_value + " )"
+ )
+
+ def visit_structured_query(
+ self, structured_query: StructuredQuery
+ ) -> Tuple[str, dict]:
+ if structured_query.filter is None:
+ kwargs = {}
+ else:
+ kwargs = {"filter": structured_query.filter.accept(self)}
+ return structured_query.query, kwargs
diff --git a/libs/langchain/langchain/schema/runnable/base.py b/libs/langchain/langchain/schema/runnable/base.py
index 9b23e50438..51ccc58f9f 100644
--- a/libs/langchain/langchain/schema/runnable/base.py
+++ b/libs/langchain/langchain/schema/runnable/base.py
@@ -39,6 +39,8 @@ from langchain.load.serializable import Serializable
from langchain.pydantic_v1 import Field
from langchain.schema.runnable.config import (
RunnableConfig,
+ acall_func_with_variable_args,
+ call_func_with_variable_args,
ensure_config,
get_async_callback_manager_for_config,
get_callback_manager_for_config,
@@ -47,16 +49,15 @@ from langchain.schema.runnable.config import (
patch_config,
)
from langchain.schema.runnable.utils import (
+ Input,
+ Output,
+ accepts_config,
accepts_run_manager,
- accepts_run_manager_and_config,
gather_with_concurrency,
)
from langchain.utils.aiter import atee, py_anext
from langchain.utils.iter import safetee
-Input = TypeVar("Input")
-# Output type should implement __concat__, as eg str, list, dict do
-Output = TypeVar("Output")
Other = TypeVar("Other")
@@ -253,7 +254,7 @@ class Runnable(Generic[Input, Output], ABC):
def with_retry(
self,
*,
- retry_if_exception_type: Tuple[Type[BaseException]] = (Exception,),
+ retry_if_exception_type: Tuple[Type[BaseException], ...] = (Exception,),
wait_exponential_jitter: bool = True,
stop_after_attempt: int = 3,
) -> Runnable[Input, Output]:
@@ -279,7 +280,7 @@ class Runnable(Generic[Input, Output], ABC):
self,
fallbacks: Sequence[Runnable[Input, Output]],
*,
- exceptions_to_handle: Tuple[Type[BaseException]] = (Exception,),
+ exceptions_to_handle: Tuple[Type[BaseException], ...] = (Exception,),
) -> RunnableWithFallbacks[Input, Output]:
return RunnableWithFallbacks(
runnable=self,
@@ -311,16 +312,7 @@ class Runnable(Generic[Input, Output], ABC):
name=config.get("run_name"),
)
try:
- if accepts_run_manager_and_config(func):
- output = func(
- input,
- run_manager=run_manager,
- config=config,
- ) # type: ignore[call-arg]
- elif accepts_run_manager(func):
- output = func(input, run_manager=run_manager) # type: ignore[call-arg]
- else:
- output = func(input) # type: ignore[call-arg]
+ output = call_func_with_variable_args(func, input, run_manager, config)
except Exception as e:
run_manager.on_chain_error(e)
raise
@@ -353,19 +345,9 @@ class Runnable(Generic[Input, Output], ABC):
name=config.get("run_name"),
)
try:
- if accepts_run_manager_and_config(func):
- output = await func(
- input,
- run_manager=run_manager,
- config=config,
- ) # type: ignore[call-arg]
- elif accepts_run_manager(func):
- output = await func(
- input,
- run_manager=run_manager,
- ) # type: ignore[call-arg]
- else:
- output = await func(input) # type: ignore[call-arg]
+ output = await acall_func_with_variable_args(
+ func, input, run_manager, config
+ )
except Exception as e:
await run_manager.on_chain_error(e)
raise
@@ -408,16 +390,15 @@ class Runnable(Generic[Input, Output], ABC):
)
]
try:
- if accepts_run_manager_and_config(func):
- output = func(
- input,
- run_manager=run_managers,
- config=configs,
- ) # type: ignore[call-arg]
- elif accepts_run_manager(func):
- output = func(input, run_manager=run_managers) # type: ignore[call-arg]
- else:
- output = func(input) # type: ignore[call-arg]
+ kwargs: Dict[str, Any] = {}
+ if accepts_config(func):
+ kwargs["config"] = [
+ patch_config(c, callbacks=rm.get_child())
+ for c, rm in zip(configs, run_managers)
+ ]
+ if accepts_run_manager(func):
+ kwargs["run_manager"] = run_managers
+ output = func(input, **kwargs) # type: ignore[call-arg]
except Exception as e:
for run_manager in run_managers:
run_manager.on_chain_error(e)
@@ -479,16 +460,15 @@ class Runnable(Generic[Input, Output], ABC):
)
)
try:
- if accepts_run_manager_and_config(func):
- output = await func(
- input,
- run_manager=run_managers,
- config=configs,
- ) # type: ignore[call-arg]
- elif accepts_run_manager(func):
- output = await func(input, run_manager=run_managers) # type: ignore
- else:
- output = await func(input) # type: ignore[call-arg]
+ kwargs: Dict[str, Any] = {}
+ if accepts_config(func):
+ kwargs["config"] = [
+ patch_config(c, callbacks=rm.get_child())
+ for c, rm in zip(configs, run_managers)
+ ]
+ if accepts_run_manager(func):
+ kwargs["run_manager"] = run_managers
+ output = await func(input, **kwargs) # type: ignore[call-arg]
except Exception as e:
await asyncio.gather(
*(run_manager.on_chain_error(e) for run_manager in run_managers)
@@ -550,19 +530,16 @@ class Runnable(Generic[Input, Output], ABC):
name=config.get("run_name"),
)
try:
- if accepts_run_manager_and_config(transformer):
- iterator = transformer(
- input_for_transform,
- run_manager=run_manager,
- config=config,
- ) # type: ignore[call-arg]
- elif accepts_run_manager(transformer):
- iterator = transformer(
- input_for_transform,
- run_manager=run_manager,
- ) # type: ignore[call-arg]
- else:
- iterator = transformer(input_for_transform) # type: ignore[call-arg]
+ kwargs: Dict[str, Any] = {}
+ if accepts_config(transformer):
+ kwargs["config"] = patch_config(
+ config, callbacks=run_manager.get_child()
+ )
+ if accepts_run_manager(transformer):
+ kwargs["run_manager"] = run_manager
+ iterator = transformer(
+ input_for_transform, **kwargs
+ ) # type: ignore[call-arg]
for chunk in iterator:
yield chunk
if final_output_supported:
@@ -631,21 +608,16 @@ class Runnable(Generic[Input, Output], ABC):
name=config.get("run_name"),
)
try:
- # mypy can't quite work out thew type guard here, but this is safe,
- # check implementations of the accepts_* functions
- if accepts_run_manager_and_config(transformer):
- iterator = transformer(
- input_for_transform,
- run_manager=run_manager,
- config=config,
- ) # type: ignore[call-arg]
- elif accepts_run_manager(transformer):
- iterator = transformer(
- input_for_transform,
- run_manager=run_manager,
- ) # type: ignore[call-arg]
- else:
- iterator = transformer(input_for_transform) # type: ignore[call-arg]
+ kwargs: Dict[str, Any] = {}
+ if accepts_config(transformer):
+ kwargs["config"] = patch_config(
+ config, callbacks=run_manager.get_child()
+ )
+ if accepts_run_manager(transformer):
+ kwargs["run_manager"] = run_manager
+ iterator = transformer(
+ input_for_transform, **kwargs
+ ) # type: ignore[call-arg]
async for chunk in iterator:
yield chunk
if final_output_supported:
@@ -681,7 +653,7 @@ class RunnableWithFallbacks(Serializable, Runnable[Input, Output]):
runnable: Runnable[Input, Output]
fallbacks: Sequence[Runnable[Input, Output]]
- exceptions_to_handle: Tuple[Type[BaseException]] = (Exception,)
+ exceptions_to_handle: Tuple[Type[BaseException], ...] = (Exception,)
class Config:
arbitrary_types_allowed = True
@@ -1756,7 +1728,7 @@ class RunnableLambda(Runnable[Input, Output]):
run_manager: CallbackManagerForChainRun,
config: RunnableConfig,
) -> Output:
- output = self.func(input)
+ output = call_func_with_variable_args(self.func, input, run_manager, config)
# If the output is a runnable, invoke it
if isinstance(output, Runnable):
recursion_limit = config["recursion_limit"]
@@ -1780,7 +1752,9 @@ class RunnableLambda(Runnable[Input, Output]):
run_manager: AsyncCallbackManagerForChainRun,
config: RunnableConfig,
) -> Output:
- output = await self.afunc(input)
+ output = await acall_func_with_variable_args(
+ self.afunc, input, run_manager, config
+ )
# If the output is a runnable, invoke it
if isinstance(output, Runnable):
recursion_limit = config["recursion_limit"]
@@ -1798,6 +1772,21 @@ class RunnableLambda(Runnable[Input, Output]):
)
return output
+ def _config(
+ self, config: Optional[RunnableConfig], callable: Callable[..., Any]
+ ) -> RunnableConfig:
+ config = config or {}
+
+ if config.get("run_name") is None:
+ try:
+ run_name = callable.__name__
+ except AttributeError:
+ run_name = None
+ if run_name is not None:
+ return patch_config(config, run_name=run_name)
+
+ return config
+
def invoke(
self,
input: Input,
@@ -1805,7 +1794,11 @@ class RunnableLambda(Runnable[Input, Output]):
**kwargs: Optional[Any],
) -> Output:
if hasattr(self, "func"):
- return self._call_with_config(self._invoke, input, config)
+ return self._call_with_config(
+ self._invoke,
+ input,
+ self._config(config, self.func),
+ )
else:
raise TypeError(
"Cannot invoke a coroutine function synchronously."
@@ -1819,7 +1812,11 @@ class RunnableLambda(Runnable[Input, Output]):
**kwargs: Optional[Any],
) -> Output:
if hasattr(self, "afunc"):
- return await self._acall_with_config(self._ainvoke, input, config)
+ return await self._acall_with_config(
+ self._ainvoke,
+ input,
+ self._config(config, self.afunc),
+ )
else:
# Delegating to super implementation of ainvoke.
# Uses asyncio executor to run the sync version (invoke)
diff --git a/libs/langchain/langchain/schema/runnable/config.py b/libs/langchain/langchain/schema/runnable/config.py
index 3f87f04403..987a2c7d2f 100644
--- a/libs/langchain/langchain/schema/runnable/config.py
+++ b/libs/langchain/langchain/schema/runnable/config.py
@@ -3,13 +3,35 @@ from __future__ import annotations
from concurrent.futures import Executor, ThreadPoolExecutor
from contextlib import contextmanager
from copy import deepcopy
-from typing import TYPE_CHECKING, Any, Dict, Generator, List, Optional, Union
+from typing import (
+ TYPE_CHECKING,
+ Any,
+ Awaitable,
+ Callable,
+ Dict,
+ Generator,
+ List,
+ Optional,
+ Union,
+)
from typing_extensions import TypedDict
+from langchain.schema.runnable.utils import (
+ Input,
+ Output,
+ accepts_config,
+ accepts_run_manager,
+)
+
if TYPE_CHECKING:
from langchain.callbacks.base import BaseCallbackManager, Callbacks
- from langchain.callbacks.manager import AsyncCallbackManager, CallbackManager
+ from langchain.callbacks.manager import (
+ AsyncCallbackManager,
+ AsyncCallbackManagerForChainRun,
+ CallbackManager,
+ CallbackManagerForChainRun,
+ )
class RunnableConfig(TypedDict, total=False):
@@ -117,6 +139,47 @@ def patch_config(
return config
+def call_func_with_variable_args(
+ func: Union[
+ Callable[[Input], Output],
+ Callable[[Input, CallbackManagerForChainRun], Output],
+ Callable[[Input, CallbackManagerForChainRun, RunnableConfig], Output],
+ ],
+ input: Input,
+ run_manager: CallbackManagerForChainRun,
+ config: RunnableConfig,
+) -> Output:
+ """Call function that may optionally accept a run_manager and/or config."""
+ kwargs: Dict[str, Any] = {}
+ if accepts_config(func):
+ kwargs["config"] = patch_config(config, callbacks=run_manager.get_child())
+ if accepts_run_manager(func):
+ kwargs["run_manager"] = run_manager
+ return func(input, **kwargs) # type: ignore[call-arg]
+
+
+async def acall_func_with_variable_args(
+ func: Union[
+ Callable[[Input], Awaitable[Output]],
+ Callable[[Input, AsyncCallbackManagerForChainRun], Awaitable[Output]],
+ Callable[
+ [Input, AsyncCallbackManagerForChainRun, RunnableConfig],
+ Awaitable[Output],
+ ],
+ ],
+ input: Input,
+ run_manager: AsyncCallbackManagerForChainRun,
+ config: RunnableConfig,
+) -> Output:
+ """Call function that may optionally accept a run_manager and/or config."""
+ kwargs: Dict[str, Any] = {}
+ if accepts_config(func):
+ kwargs["config"] = patch_config(config, callbacks=run_manager.get_child())
+ if accepts_run_manager(func):
+ kwargs["run_manager"] = run_manager
+ return await func(input, **kwargs) # type: ignore[call-arg]
+
+
def get_callback_manager_for_config(config: RunnableConfig) -> CallbackManager:
from langchain.callbacks.manager import CallbackManager
diff --git a/libs/langchain/langchain/schema/runnable/retry.py b/libs/langchain/langchain/schema/runnable/retry.py
index 37de03f600..b41f74583b 100644
--- a/libs/langchain/langchain/schema/runnable/retry.py
+++ b/libs/langchain/langchain/schema/runnable/retry.py
@@ -24,7 +24,7 @@ U = TypeVar("U")
class RunnableRetry(RunnableBinding[Input, Output]):
"""Retry a Runnable if it fails."""
- retry_exception_types: Tuple[Type[BaseException]] = (Exception,)
+ retry_exception_types: Tuple[Type[BaseException], ...] = (Exception,)
wait_exponential_jitter: bool = True
diff --git a/libs/langchain/langchain/schema/runnable/utils.py b/libs/langchain/langchain/schema/runnable/utils.py
index 2afa3705c4..43d9b325fd 100644
--- a/libs/langchain/langchain/schema/runnable/utils.py
+++ b/libs/langchain/langchain/schema/runnable/utils.py
@@ -2,7 +2,11 @@ from __future__ import annotations
import asyncio
from inspect import signature
-from typing import Any, Callable, Coroutine, Union
+from typing import Any, Callable, Coroutine, TypeVar, Union
+
+Input = TypeVar("Input")
+# Output type should implement __concat__, as eg str, list, dict do
+Output = TypeVar("Output")
async def gated_coro(semaphore: asyncio.Semaphore, coro: Coroutine) -> Any:
@@ -26,8 +30,8 @@ def accepts_run_manager(callable: Callable[..., Any]) -> bool:
return False
-def accepts_run_manager_and_config(callable: Callable[..., Any]) -> bool:
- return (
- accepts_run_manager(callable)
- and signature(callable).parameters.get("config") is not None
- )
+def accepts_config(callable: Callable[..., Any]) -> bool:
+ try:
+ return signature(callable).parameters.get("config") is not None
+ except ValueError:
+ return False
diff --git a/libs/langchain/langchain/smith/evaluation/name_generation.py b/libs/langchain/langchain/smith/evaluation/name_generation.py
new file mode 100644
index 0000000000..33ac3a6753
--- /dev/null
+++ b/libs/langchain/langchain/smith/evaluation/name_generation.py
@@ -0,0 +1,729 @@
+import random
+
+adjectives = [
+ "abandoned",
+ "aching",
+ "advanced",
+ "ample",
+ "artistic",
+ "back",
+ "best",
+ "bold",
+ "brief",
+ "clear",
+ "cold",
+ "complicated",
+ "cooked",
+ "crazy",
+ "crushing",
+ "damp",
+ "dear",
+ "definite",
+ "dependable",
+ "diligent",
+ "drab",
+ "earnest",
+ "elderly",
+ "enchanted",
+ "essential",
+ "excellent",
+ "extraneous",
+ "fixed",
+ "flowery",
+ "formal",
+ "fresh",
+ "frosty",
+ "giving",
+ "glossy",
+ "healthy",
+ "helpful",
+ "impressionable",
+ "kind",
+ "large",
+ "left",
+ "long",
+ "loyal",
+ "mealy",
+ "memorable",
+ "monthly",
+ "new",
+ "notable",
+ "only",
+ "ordinary",
+ "passionate",
+ "perfect",
+ "pertinent",
+ "proper",
+ "puzzled",
+ "reflecting",
+ "respectful",
+ "roasted",
+ "scholarly",
+ "shiny",
+ "slight",
+ "sparkling",
+ "spotless",
+ "stupendous",
+ "sunny",
+ "tart",
+ "terrific",
+ "timely",
+ "unique",
+ "upbeat",
+ "vacant",
+ "virtual",
+ "warm",
+ "weary",
+ "whispered",
+ "worthwhile",
+ "yellow",
+]
+
+nouns = [
+ "account",
+ "acknowledgment",
+ "address",
+ "advertising",
+ "airplane",
+ "animal",
+ "appointment",
+ "arrival",
+ "artist",
+ "attachment",
+ "attitude",
+ "availability",
+ "backpack",
+ "bag",
+ "balance",
+ "bass",
+ "bean",
+ "beauty",
+ "bibliography",
+ "bill",
+ "bite",
+ "blossom",
+ "boat",
+ "book",
+ "box",
+ "boy",
+ "bread",
+ "bridge",
+ "broccoli",
+ "building",
+ "butter",
+ "button",
+ "cabbage",
+ "cake",
+ "camera",
+ "camp",
+ "candle",
+ "candy",
+ "canvas",
+ "car",
+ "card",
+ "carrot",
+ "cart",
+ "case",
+ "cat",
+ "chain",
+ "chair",
+ "chalk",
+ "chance",
+ "change",
+ "channel",
+ "character",
+ "charge",
+ "charm",
+ "chart",
+ "check",
+ "cheek",
+ "cheese",
+ "chef",
+ "cherry",
+ "chicken",
+ "child",
+ "church",
+ "circle",
+ "class",
+ "clay",
+ "click",
+ "clock",
+ "cloth",
+ "cloud",
+ "clove",
+ "club",
+ "coach",
+ "coal",
+ "coast",
+ "coat",
+ "cod",
+ "coffee",
+ "collar",
+ "color",
+ "comb",
+ "comfort",
+ "comic",
+ "committee",
+ "community",
+ "company",
+ "comparison",
+ "competition",
+ "condition",
+ "connection",
+ "control",
+ "cook",
+ "copper",
+ "copy",
+ "corn",
+ "cough",
+ "country",
+ "cover",
+ "crate",
+ "crayon",
+ "cream",
+ "creator",
+ "crew",
+ "crown",
+ "current",
+ "curtain",
+ "curve",
+ "cushion",
+ "dad",
+ "daughter",
+ "day",
+ "death",
+ "debt",
+ "decision",
+ "deer",
+ "degree",
+ "design",
+ "desire",
+ "desk",
+ "detail",
+ "development",
+ "digestion",
+ "dime",
+ "dinner",
+ "direction",
+ "dirt",
+ "discovery",
+ "discussion",
+ "disease",
+ "disgust",
+ "distance",
+ "distribution",
+ "division",
+ "doctor",
+ "dog",
+ "door",
+ "drain",
+ "drawer",
+ "dress",
+ "drink",
+ "driving",
+ "dust",
+ "ear",
+ "earth",
+ "edge",
+ "education",
+ "effect",
+ "egg",
+ "end",
+ "energy",
+ "engine",
+ "error",
+ "event",
+ "example",
+ "exchange",
+ "existence",
+ "expansion",
+ "experience",
+ "expert",
+ "eye",
+ "face",
+ "fact",
+ "fall",
+ "family",
+ "farm",
+ "father",
+ "fear",
+ "feeling",
+ "field",
+ "finger",
+ "fire",
+ "fish",
+ "flag",
+ "flight",
+ "floor",
+ "flower",
+ "fold",
+ "food",
+ "football",
+ "force",
+ "form",
+ "frame",
+ "friend",
+ "frog",
+ "fruit",
+ "fuel",
+ "furniture",
+ "game",
+ "garden",
+ "gate",
+ "girl",
+ "glass",
+ "glove",
+ "goat",
+ "gold",
+ "government",
+ "grade",
+ "grain",
+ "grass",
+ "green",
+ "grip",
+ "group",
+ "growth",
+ "guide",
+ "guitar",
+ "hair",
+ "hall",
+ "hand",
+ "harbor",
+ "harmony",
+ "hat",
+ "head",
+ "health",
+ "heart",
+ "heat",
+ "hill",
+ "history",
+ "hobbies",
+ "hole",
+ "hope",
+ "horn",
+ "horse",
+ "hospital",
+ "hour",
+ "house",
+ "humor",
+ "idea",
+ "impulse",
+ "income",
+ "increase",
+ "industry",
+ "ink",
+ "insect",
+ "instrument",
+ "insurance",
+ "interest",
+ "invention",
+ "iron",
+ "island",
+ "jelly",
+ "jet",
+ "jewel",
+ "join",
+ "judge",
+ "juice",
+ "jump",
+ "kettle",
+ "key",
+ "kick",
+ "kiss",
+ "kitten",
+ "knee",
+ "knife",
+ "knowledge",
+ "land",
+ "language",
+ "laugh",
+ "law",
+ "lead",
+ "learning",
+ "leather",
+ "leg",
+ "lettuce",
+ "level",
+ "library",
+ "lift",
+ "light",
+ "limit",
+ "line",
+ "linen",
+ "lip",
+ "liquid",
+ "list",
+ "look",
+ "loss",
+ "love",
+ "lunch",
+ "machine",
+ "man",
+ "manager",
+ "map",
+ "marble",
+ "mark",
+ "market",
+ "mass",
+ "match",
+ "meal",
+ "measure",
+ "meat",
+ "meeting",
+ "memory",
+ "metal",
+ "middle",
+ "milk",
+ "mind",
+ "mine",
+ "minute",
+ "mist",
+ "mitten",
+ "mom",
+ "money",
+ "monkey",
+ "month",
+ "moon",
+ "morning",
+ "mother",
+ "motion",
+ "mountain",
+ "mouth",
+ "muscle",
+ "music",
+ "nail",
+ "name",
+ "nation",
+ "neck",
+ "need",
+ "news",
+ "night",
+ "noise",
+ "note",
+ "number",
+ "nut",
+ "observation",
+ "offer",
+ "oil",
+ "operation",
+ "opinion",
+ "orange",
+ "order",
+ "organization",
+ "ornament",
+ "oven",
+ "page",
+ "pail",
+ "pain",
+ "paint",
+ "pan",
+ "pancake",
+ "paper",
+ "parcel",
+ "parent",
+ "part",
+ "passenger",
+ "paste",
+ "payment",
+ "peace",
+ "pear",
+ "pen",
+ "pencil",
+ "person",
+ "pest",
+ "pet",
+ "picture",
+ "pie",
+ "pin",
+ "pipe",
+ "pizza",
+ "place",
+ "plane",
+ "plant",
+ "plastic",
+ "plate",
+ "play",
+ "pleasure",
+ "plot",
+ "plough",
+ "pocket",
+ "point",
+ "poison",
+ "police",
+ "pollution",
+ "popcorn",
+ "porter",
+ "position",
+ "pot",
+ "potato",
+ "powder",
+ "power",
+ "price",
+ "print",
+ "process",
+ "produce",
+ "product",
+ "profit",
+ "property",
+ "prose",
+ "protest",
+ "pull",
+ "pump",
+ "punishment",
+ "purpose",
+ "push",
+ "quarter",
+ "question",
+ "quiet",
+ "quill",
+ "quilt",
+ "quince",
+ "rabbit",
+ "rail",
+ "rain",
+ "range",
+ "rat",
+ "rate",
+ "ray",
+ "reaction",
+ "reading",
+ "reason",
+ "record",
+ "regret",
+ "relation",
+ "religion",
+ "representative",
+ "request",
+ "respect",
+ "rest",
+ "reward",
+ "rhythm",
+ "rice",
+ "river",
+ "road",
+ "roll",
+ "room",
+ "root",
+ "rose",
+ "route",
+ "rub",
+ "rule",
+ "run",
+ "sack",
+ "sail",
+ "salt",
+ "sand",
+ "scale",
+ "scarecrow",
+ "scarf",
+ "scene",
+ "scent",
+ "school",
+ "science",
+ "scissors",
+ "screw",
+ "sea",
+ "seat",
+ "secretary",
+ "seed",
+ "selection",
+ "self",
+ "sense",
+ "servant",
+ "shade",
+ "shake",
+ "shame",
+ "shape",
+ "sheep",
+ "sheet",
+ "shelf",
+ "ship",
+ "shirt",
+ "shock",
+ "shoe",
+ "shop",
+ "show",
+ "side",
+ "sign",
+ "silk",
+ "sink",
+ "sister",
+ "size",
+ "sky",
+ "slave",
+ "sleep",
+ "smash",
+ "smell",
+ "smile",
+ "smoke",
+ "snail",
+ "snake",
+ "sneeze",
+ "snow",
+ "soap",
+ "society",
+ "sock",
+ "soda",
+ "sofa",
+ "son",
+ "song",
+ "sort",
+ "sound",
+ "soup",
+ "space",
+ "spark",
+ "speed",
+ "sponge",
+ "spoon",
+ "spray",
+ "spring",
+ "spy",
+ "square",
+ "stamp",
+ "star",
+ "start",
+ "statement",
+ "station",
+ "steam",
+ "steel",
+ "stem",
+ "step",
+ "stew",
+ "stick",
+ "stitch",
+ "stocking",
+ "stomach",
+ "stone",
+ "stop",
+ "store",
+ "story",
+ "stove",
+ "stranger",
+ "straw",
+ "stream",
+ "street",
+ "stretch",
+ "string",
+ "structure",
+ "substance",
+ "sugar",
+ "suggestion",
+ "suit",
+ "summer",
+ "sun",
+ "support",
+ "surprise",
+ "sweater",
+ "swim",
+ "system",
+ "table",
+ "tail",
+ "talk",
+ "tank",
+ "taste",
+ "tax",
+ "tea",
+ "teaching",
+ "team",
+ "tendency",
+ "test",
+ "texture",
+ "theory",
+ "thing",
+ "thought",
+ "thread",
+ "throat",
+ "thumb",
+ "thunder",
+ "ticket",
+ "time",
+ "tin",
+ "title",
+ "toad",
+ "toe",
+ "tooth",
+ "toothpaste",
+ "touch",
+ "town",
+ "toy",
+ "trade",
+ "train",
+ "transport",
+ "tray",
+ "treatment",
+ "tree",
+ "trick",
+ "trip",
+ "trouble",
+ "trousers",
+ "truck",
+ "tub",
+ "turkey",
+ "turn",
+ "twist",
+ "umbrella",
+ "uncle",
+ "underwear",
+ "unit",
+ "use",
+ "vacation",
+ "value",
+ "van",
+ "vase",
+ "vegetable",
+ "veil",
+ "vein",
+ "verse",
+ "vessel",
+ "view",
+ "visitor",
+ "voice",
+ "volcano",
+ "walk",
+ "wall",
+ "war",
+ "wash",
+ "waste",
+ "watch",
+ "water",
+ "wave",
+ "wax",
+ "way",
+ "wealth",
+ "weather",
+ "week",
+ "weight",
+ "wheel",
+ "whip",
+ "whistle",
+ "window",
+ "wine",
+ "wing",
+ "winter",
+ "wire",
+ "wish",
+ "woman",
+ "wood",
+ "wool",
+ "word",
+ "work",
+ "worm",
+ "wound",
+ "wrist",
+ "writer",
+ "yard",
+ "yoke",
+ "zebra",
+ "zinc",
+ "zipper",
+ "zone",
+]
+
+
+def random_name(prefix: str = "test") -> str:
+ """Generate a random name."""
+ adjective = random.choice(adjectives)
+ noun = random.choice(nouns)
+ number = random.randint(1, 100)
+
+ return f"{prefix}-{adjective}-{noun}-{number}"
diff --git a/libs/langchain/langchain/smith/evaluation/progress.py b/libs/langchain/langchain/smith/evaluation/progress.py
new file mode 100644
index 0000000000..1ea51eee42
--- /dev/null
+++ b/libs/langchain/langchain/smith/evaluation/progress.py
@@ -0,0 +1,82 @@
+"""A simple progress bar for the console."""
+import threading
+from typing import Any, Dict, Optional, Sequence
+from uuid import UUID
+
+from langchain.callbacks import base as base_callbacks
+from langchain.schema.document import Document
+from langchain.schema.output import LLMResult
+
+
+class ProgressBarCallback(base_callbacks.BaseCallbackHandler):
+ """A simple progress bar for the console."""
+
+ def __init__(self, total: int, ncols: int = 50, **kwargs: Any):
+ """Initialize the progress bar.
+
+ Args:
+ total: int, the total number of items to be processed.
+ ncols: int, the character width of the progress bar.
+ """
+ self.total = total
+ self.ncols = ncols
+ self.counter = 0
+ self.lock = threading.Lock()
+ self._print_bar()
+
+ def increment(self) -> None:
+ """Increment the counter and update the progress bar."""
+ with self.lock:
+ self.counter += 1
+ self._print_bar()
+
+ def _print_bar(self) -> None:
+ """Print the progress bar to the console."""
+ progress = self.counter / self.total
+ arrow = "-" * int(round(progress * self.ncols) - 1) + ">"
+ spaces = " " * (self.ncols - len(arrow))
+ print(f"\r[{arrow + spaces}] {self.counter}/{self.total}", end="")
+
+ def on_chain_end(
+ self,
+ outputs: Dict[str, Any],
+ *,
+ run_id: UUID,
+ parent_run_id: Optional[UUID] = None,
+ **kwargs: Any,
+ ) -> Any:
+ if parent_run_id is None:
+ self.increment()
+
+ def on_retriever_end(
+ self,
+ documents: Sequence[Document],
+ *,
+ run_id: UUID,
+ parent_run_id: Optional[UUID] = None,
+ **kwargs: Any,
+ ) -> Any:
+ if parent_run_id is None:
+ self.increment()
+
+ def on_llm_end(
+ self,
+ response: LLMResult,
+ *,
+ run_id: UUID,
+ parent_run_id: Optional[UUID] = None,
+ **kwargs: Any,
+ ) -> Any:
+ if parent_run_id is None:
+ self.increment()
+
+ def on_tool_end(
+ self,
+ output: str,
+ *,
+ run_id: UUID,
+ parent_run_id: Optional[UUID] = None,
+ **kwargs: Any,
+ ) -> Any:
+ if parent_run_id is None:
+ self.increment()
diff --git a/libs/langchain/langchain/smith/evaluation/runner_utils.py b/libs/langchain/langchain/smith/evaluation/runner_utils.py
index 1f432053e2..e6dfe827f6 100644
--- a/libs/langchain/langchain/smith/evaluation/runner_utils.py
+++ b/libs/langchain/langchain/smith/evaluation/runner_utils.py
@@ -2,21 +2,16 @@
from __future__ import annotations
-import asyncio
import functools
import inspect
-import itertools
import logging
-import uuid
import warnings
from enum import Enum
from typing import (
TYPE_CHECKING,
Any,
Callable,
- Coroutine,
Dict,
- Iterator,
List,
Optional,
Sequence,
@@ -24,16 +19,13 @@ from typing import (
Union,
cast,
)
-from urllib.parse import urlparse, urlunparse
from langsmith import Client, RunEvaluator
from langsmith.schemas import Dataset, DataType, Example
-from langchain.callbacks.base import BaseCallbackHandler
from langchain.callbacks.manager import Callbacks
-from langchain.callbacks.tracers.base import BaseTracer
from langchain.callbacks.tracers.evaluation import EvaluatorCallbackHandler
-from langchain.callbacks.tracers.langchain import LangChainTracer
+from langchain.callbacks.tracers.langchain import LangChainTracer, wait_for_all_tracers
from langchain.chains.base import Chain
from langchain.evaluation.loading import load_evaluator
from langchain.evaluation.schema import EvaluatorType, StringEvaluator
@@ -41,8 +33,11 @@ from langchain.schema import ChatResult, LLMResult
from langchain.schema.language_model import BaseLanguageModel
from langchain.schema.messages import BaseMessage, messages_from_dict
from langchain.schema.runnable import Runnable, RunnableConfig, RunnableLambda
-from langchain.smith.evaluation.config import EvalConfig, RunEvalConfig
-from langchain.smith.evaluation.string_run_evaluator import StringRunEvaluatorChain
+from langchain.schema.runnable import config as runnable_config
+from langchain.schema.runnable import utils as runnable_utils
+from langchain.smith import evaluation as smith_eval
+from langchain.smith.evaluation import config as smith_eval_config
+from langchain.smith.evaluation import name_generation, progress
if TYPE_CHECKING:
import pandas as pd
@@ -69,6 +64,26 @@ class InputFormatError(Exception):
class TestResult(dict):
"""A dictionary of the results of a single test run."""
+ def get_aggregate_feedback(
+ self, quantiles: Optional[Sequence[float]] = None
+ ) -> pd.DataFrame:
+ """Return quantiles for the feedback scores.
+
+ This method calculates and prints the quantiles for the feedback scores
+ across all feedback keys.
+
+ Returns:
+ A DataFrame containing the quantiles for each feedback key.
+ """
+ df = self.to_dataframe()
+ feedback_cols = [
+ col for col in df.columns if col not in ["input", "output", "reference"]
+ ]
+ _quantiles = df[feedback_cols].quantile(
+ quantiles or [0.25, 0.5, 0.75], numeric_only=True
+ )
+ return _quantiles.transpose()
+
def to_dataframe(self) -> pd.DataFrame:
"""Convert the results to a dataframe."""
try:
@@ -83,27 +98,19 @@ class TestResult(dict):
records = []
for example_id, result in self["results"].items():
feedback = result["feedback"]
- records.append(
- {**{f.key: f.score for f in feedback}, "output": result["output"]}
- )
+ r = {
+ **{f.key: f.score for f in feedback},
+ "input": result["input"],
+ "output": result["output"],
+ }
+ if "reference" in result:
+ r["reference"] = result["reference"]
+ records.append(r)
indices.append(example_id)
return pd.DataFrame(records, index=indices)
-def _get_eval_project_url(api_url: str, project_id: str) -> str:
- """Get the project url from the api url."""
- parsed = urlparse(api_url)
- hostname = parsed.hostname or ""
- if "api." in hostname:
- hostname = hostname.replace("api.", "", 1)
- if "localhost" in hostname:
- # Remove the port
- hostname = "localhost"
- url = urlunparse(parsed._replace(netloc=hostname))
- return f"{url}/projects/p/{project_id}?eval=true"
-
-
def _wrap_in_chain_factory(
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
dataset_name: str = "<my_dataset>",
@@ -172,15 +179,6 @@ def _wrap_in_chain_factory(
return llm_or_chain_factory
-def _first_example(examples: Iterator[Example]) -> Tuple[Example, Iterator[Example]]:
- """Get the first example while chaining it back and preserving the iterator."""
- try:
- example: Example = next(examples)
- except StopIteration:
- raise ValueError("No examples provided.")
- return example, itertools.chain([example], examples)
-
-
def _get_prompt(inputs: Dict[str, Any]) -> str:
"""Get prompt from inputs.
@@ -277,31 +275,7 @@ def _get_messages(inputs: Dict[str, Any]) -> List[BaseMessage]:
)
-def _get_project_name(
- project_name: Optional[str],
- llm_or_chain_factory: MCF,
-) -> str:
- """
- Get the project name.
-
- Args:
- project_name: The project name if manually specified.
- llm_or_chain_factory: The Chain or language model constructor.
-
- Returns:
- The project name.
- """
- if project_name is not None:
- return project_name
- if isinstance(llm_or_chain_factory, BaseLanguageModel):
- model_name = llm_or_chain_factory.__class__.__name__
- else:
- model_name = llm_or_chain_factory().__class__.__name__
- hex = uuid.uuid4().hex
- return f"{hex}-{model_name}"
-
-
-## Shared Validation Utilities
+## Shared data validation utilities
def _validate_example_inputs_for_language_model(
first_example: Example,
input_mapper: Optional[Callable[[Dict], Any]],
@@ -373,22 +347,20 @@ def _validate_example_inputs_for_chain(
def _validate_example_inputs(
- examples: Iterator[Example],
+ example: Example,
llm_or_chain_factory: MCF,
input_mapper: Optional[Callable[[Dict], Any]],
-) -> Iterator[Example]:
+) -> None:
"""Validate that the example inputs are valid for the model."""
- first_example, examples = _first_example(examples)
if isinstance(llm_or_chain_factory, BaseLanguageModel):
- _validate_example_inputs_for_language_model(first_example, input_mapper)
+ _validate_example_inputs_for_language_model(example, input_mapper)
else:
chain = llm_or_chain_factory()
if isinstance(chain, Chain):
# Otherwise it's a runnable
- _validate_example_inputs_for_chain(first_example, chain, input_mapper)
+ _validate_example_inputs_for_chain(example, chain, input_mapper)
elif isinstance(chain, Runnable):
logger.debug(f"Skipping input validation for {chain}")
- return examples
## Shared Evaluator Setup Utilities
@@ -396,13 +368,12 @@ def _validate_example_inputs(
def _setup_evaluation(
llm_or_chain_factory: MCF,
- examples: Iterator[Example],
- evaluation: Optional[RunEvalConfig],
+ examples: List[Example],
+ evaluation: Optional[smith_eval.RunEvalConfig],
data_type: DataType,
-) -> Tuple[Optional[List[RunEvaluator]], Iterator[Example]]:
+) -> Optional[List[RunEvaluator]]:
"""Configure the evaluators to run on the results of the chain."""
if evaluation:
- first_example, examples = _first_example(examples)
if isinstance(llm_or_chain_factory, BaseLanguageModel):
run_inputs, run_outputs = None, None
run_type = "llm"
@@ -422,18 +393,18 @@ def _setup_evaluation(
evaluation,
run_type,
data_type,
- list(first_example.outputs) if first_example.outputs else None,
+ list(examples[0].outputs) if examples[0].outputs else None,
run_inputs,
run_outputs,
)
else:
# TODO: Create a default helpfulness evaluator
run_evaluators = None
- return run_evaluators, examples
+ return run_evaluators
def _determine_input_key(
- config: RunEvalConfig,
+ config: smith_eval.RunEvalConfig,
run_inputs: Optional[List[str]],
) -> Optional[str]:
input_key = None
@@ -452,7 +423,7 @@ def _determine_input_key(
def _determine_prediction_key(
- config: RunEvalConfig,
+ config: smith_eval.RunEvalConfig,
run_outputs: Optional[List[str]],
) -> Optional[str]:
prediction_key = None
@@ -473,7 +444,7 @@ def _determine_prediction_key(
def _determine_reference_key(
- config: RunEvalConfig,
+ config: smith_eval.RunEvalConfig,
example_outputs: Optional[List[str]],
) -> Optional[str]:
if config.reference_key:
@@ -491,7 +462,7 @@ def _determine_reference_key(
def _construct_run_evaluator(
- eval_config: Union[EvaluatorType, str, EvalConfig],
+ eval_config: Union[EvaluatorType, str, smith_eval_config.EvalConfig],
eval_llm: Optional[BaseLanguageModel],
run_type: str,
data_type: DataType,
@@ -513,11 +484,11 @@ def _construct_run_evaluator(
if isinstance(evaluator_, StringEvaluator):
if evaluator_.requires_reference and reference_key is None:
raise ValueError(
- f"Must specify reference_key in RunEvalConfig to use"
+ f"Must specify reference_key in smith_eval.RunEvalConfig to use"
f" evaluator of type {eval_type_tag} with"
f" dataset with multiple output keys: {example_outputs}."
)
- run_evaluator = StringRunEvaluatorChain.from_run_and_data_type(
+ run_evaluator = smith_eval.StringRunEvaluatorChain.from_run_and_data_type(
evaluator_,
run_type,
data_type,
@@ -534,7 +505,7 @@ def _construct_run_evaluator(
def _get_keys(
- config: RunEvalConfig,
+ config: smith_eval.RunEvalConfig,
run_inputs: Optional[List[str]],
run_outputs: Optional[List[str]],
example_outputs: Optional[List[str]],
@@ -546,7 +517,7 @@ def _get_keys(
def _load_run_evaluators(
- config: RunEvalConfig,
+ config: smith_eval.RunEvalConfig,
run_type: str,
data_type: DataType,
example_outputs: Optional[List[str]],
@@ -593,7 +564,7 @@ def _load_run_evaluators(
run_evaluators.append(custom_evaluator)
elif isinstance(custom_evaluator, StringEvaluator):
run_evaluators.append(
- StringRunEvaluatorChain.from_run_and_data_type(
+ smith_eval.StringRunEvaluatorChain.from_run_and_data_type(
custom_evaluator,
run_type,
data_type,
@@ -694,10 +665,9 @@ async def _arun_chain(
async def _arun_llm_or_chain(
example: Example,
- llm_or_chain_factory: MCF,
+ config: RunnableConfig,
*,
- tags: Optional[List[str]] = None,
- callbacks: Optional[List[BaseCallbackHandler]] = None,
+ llm_or_chain_factory: MCF,
input_mapper: Optional[Callable[[Dict], Any]] = None,
) -> Union[dict, str, LLMResult, ChatResult]:
"""Asynchronously run the Chain or language model.
@@ -712,15 +682,6 @@ async def _arun_llm_or_chain(
Returns:
A list of outputs.
"""
- if callbacks:
- previous_example_ids = [
- getattr(tracer, "example_id", None) for tracer in callbacks
- ]
- for tracer in callbacks:
- if hasattr(tracer, "example_id"):
- tracer.example_id = example.id
- else:
- previous_example_ids = None
chain_or_llm = (
"LLM" if isinstance(llm_or_chain_factory, BaseLanguageModel) else "Chain"
)
@@ -730,8 +691,8 @@ async def _arun_llm_or_chain(
output: Any = await _arun_llm(
llm_or_chain_factory,
example.inputs,
- tags=tags,
- callbacks=callbacks,
+ tags=config["tags"],
+ callbacks=config["callbacks"],
input_mapper=input_mapper,
)
else:
@@ -739,198 +700,19 @@ async def _arun_llm_or_chain(
output = await _arun_chain(
chain,
example.inputs,
- tags=tags,
- callbacks=callbacks,
+ tags=config["tags"],
+ callbacks=config["callbacks"],
input_mapper=input_mapper,
)
result = output
except Exception as e:
- logger.warning(f"{chain_or_llm} failed for example {example.id}. Error: {e}")
- result = {"Error": str(e)}
- if callbacks and previous_example_ids:
- for example_id, tracer in zip(previous_example_ids, callbacks):
- if hasattr(tracer, "example_id"):
- tracer.example_id = example_id
- return result
-
-
-async def _gather_with_concurrency(
- n: int,
- initializer: Callable[[], Coroutine[Any, Any, Any]],
- *async_funcs: Callable[
- [Sequence[BaseCallbackHandler], Dict], Coroutine[Any, Any, Any]
- ],
-) -> List[Any]:
- """Run coroutines with a concurrency limit.
-
- Args:
- n: The maximum number of concurrent tasks.
- initializer: A coroutine that initializes shared resources for the tasks.
- async_funcs: The async_funcs to be run concurrently.
-
- Returns:
- A list of results from the coroutines.
- """
- semaphore = asyncio.Semaphore(n)
- job_state = {"num_processed": 0}
-
- callback_queue: asyncio.Queue[Sequence[BaseCallbackHandler]] = asyncio.Queue()
- for _ in range(n):
- callback_queue.put_nowait(await initializer())
-
- async def run_coroutine_with_semaphore(
- async_func: Callable[
- [Sequence[BaseCallbackHandler], Dict], Coroutine[Any, Any, Any]
- ]
- ) -> Any:
- async with semaphore:
- callbacks = await callback_queue.get()
- try:
- result = await async_func(callbacks, job_state)
- finally:
- callback_queue.put_nowait(callbacks)
- return result
-
- results = await asyncio.gather(
- *(run_coroutine_with_semaphore(function) for function in async_funcs)
- )
- while callback_queue:
- try:
- callbacks = callback_queue.get_nowait()
- except asyncio.QueueEmpty:
- break
- for callback in callbacks:
- if isinstance(callback, (LangChainTracer, EvaluatorCallbackHandler)):
- callback.wait_for_futures()
- return results
-
-
-async def _callbacks_initializer(
- project_name: Optional[str],
- client: Client,
- run_evaluators: Sequence[RunEvaluator],
- evaluation_handler_collector: List[EvaluatorCallbackHandler],
-) -> List[BaseTracer]:
- """
- Initialize a tracer to share across tasks.
-
- Args:
- project_name: The project name for the tracer.
- client: The client to use for the tracer.
- run_evaluators: The evaluators to run.
- evaluation_handler_collector: A list to collect the evaluators.
- Used to wait for the evaluators to finish.
-
- Returns:
- The callbacks for this thread.
- """
- callbacks: List[BaseTracer] = []
- if project_name:
- callbacks.append(
- LangChainTracer(
- project_name=project_name, client=client, use_threading=False
- )
- )
- if run_evaluators:
- callback = EvaluatorCallbackHandler(
- client=client,
- evaluators=run_evaluators,
- # We already have concurrency, don't want to overload the machine
- max_workers=1,
- )
- callbacks.append(callback)
- evaluation_handler_collector.append(callback)
- return callbacks
-
-
-async def _arun_on_examples(
- client: Client,
- examples: Iterator[Example],
- llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
- *,
- evaluation: Optional[RunEvalConfig] = None,
- concurrency_level: int = 5,
- project_name: Optional[str] = None,
- verbose: bool = False,
- tags: Optional[List[str]] = None,
- input_mapper: Optional[Callable[[Dict], Any]] = None,
- data_type: DataType = DataType.kv,
-) -> Dict[str, Any]:
- """
- Asynchronously run the chain on examples and store traces
- to the specified project name.
-
- Args:
- client: LangSmith client to use to log feedback and runs.
- examples: Examples to run the model or chain over.
- llm_or_chain_factory: Language model or Chain constructor to run
- over the dataset. The Chain constructor is used to permit
- independent calls on each example without carrying over state.
- evaluation: Optional evaluation configuration to use when evaluating
- concurrency_level: The number of async tasks to run concurrently.
- project_name: Project name to use when tracing runs.
- Defaults to {dataset_name}-{chain class name}-{datetime}.
- verbose: Whether to print progress.
- tags: Tags to add to each run in the project.
- input_mapper: function to map to the inputs dictionary from an Example
- to the format expected by the model to be evaluated. This is useful if
- your model needs to deserialize more complex schema or if your dataset
- has inputs with keys that differ from what is expected by your chain
- or agent.
- data_type: The dataset's data type. This is used to determine determine
- how to deserialize the reference data and model compatibility.
- Returns:
- A dictionary mapping example ids to the model outputs.
- """
- wrapped_model = _wrap_in_chain_factory(llm_or_chain_factory)
- project_name = _get_project_name(project_name, wrapped_model)
- run_evaluators, examples = _setup_evaluation(
- wrapped_model, examples, evaluation, data_type
- )
- examples = _validate_example_inputs(examples, wrapped_model, input_mapper)
- results: Dict[str, dict] = {}
-
- async def process_example(
- example: Example, callbacks: List[BaseCallbackHandler], job_state: dict
- ) -> None:
- """Process a single example."""
- result = await _arun_llm_or_chain(
- example,
- wrapped_model,
- tags=tags,
- callbacks=callbacks,
- input_mapper=input_mapper,
+ logger.warning(
+ f"{chain_or_llm} failed for example {example.id} "
+ f"with inputs {example.inputs}"
+ f"\n{repr(e)}"
)
- results[str(example.id)] = {"output": result}
- job_state["num_processed"] += 1
- if verbose:
- print(
- f"Processed examples: {job_state['num_processed']}",
- end="\r",
- flush=True,
- )
-
- evaluation_handlers: List[EvaluatorCallbackHandler] = []
- await _gather_with_concurrency(
- concurrency_level,
- functools.partial(
- _callbacks_initializer,
- project_name=project_name,
- client=client,
- evaluation_handler_collector=evaluation_handlers,
- run_evaluators=run_evaluators or [],
- ),
- *(functools.partial(process_example, e) for e in examples),
- )
- all_feedback = {}
- for handler in evaluation_handlers:
- handler.wait_for_futures()
- all_feedback.update(handler.logged_feedback)
- # join the results and feedback on the example id
- for example_id, output_dict in results.items():
- feedback = all_feedback.get(example_id, [])
- output_dict["feedback"] = feedback
- return results
+ result = {"Error": repr(e)}
+ return result
## Sync Utilities
@@ -1011,10 +793,9 @@ def _run_chain(
def _run_llm_or_chain(
example: Example,
- llm_or_chain_factory: MCF,
+ config: RunnableConfig,
*,
- tags: Optional[List[str]] = None,
- callbacks: Optional[List[BaseCallbackHandler]] = None,
+ llm_or_chain_factory: MCF,
input_mapper: Optional[Callable[[Dict], Any]] = None,
) -> Union[dict, str, LLMResult, ChatResult]:
"""
@@ -1030,15 +811,6 @@ def _run_llm_or_chain(
Union[List[dict], List[str], List[LLMResult], List[ChatResult]]:
The outputs of the model or chain.
"""
- if callbacks:
- previous_example_ids = [
- getattr(tracer, "example_id", None) for tracer in callbacks
- ]
- for tracer in callbacks:
- if hasattr(tracer, "example_id"):
- tracer.example_id = example.id
- else:
- previous_example_ids = None
chain_or_llm = (
"LLM" if isinstance(llm_or_chain_factory, BaseLanguageModel) else "Chain"
)
@@ -1048,8 +820,8 @@ def _run_llm_or_chain(
output: Any = _run_llm(
llm_or_chain_factory,
example.inputs,
- callbacks,
- tags=tags,
+ config["callbacks"],
+ tags=config["tags"],
input_mapper=input_mapper,
)
else:
@@ -1057,98 +829,22 @@ def _run_llm_or_chain(
output = _run_chain(
chain,
example.inputs,
- callbacks,
- tags=tags,
+ config["callbacks"],
+ tags=config["tags"],
input_mapper=input_mapper,
)
result = output
except Exception as e:
+ error_type = type(e).__name__
logger.warning(
- f"{chain_or_llm} failed for example {example.id} with inputs:"
- f" {example.inputs}.\nError: {e}",
+ f"{chain_or_llm} failed for example {example.id} "
+ f"with inputs {example.inputs}"
+ f"\nError Type: {error_type}, Message: {e}"
)
- result = {"Error": str(e)}
- if callbacks and previous_example_ids:
- for example_id, tracer in zip(previous_example_ids, callbacks):
- if hasattr(tracer, "example_id"):
- tracer.example_id = example_id
+ result = {"Error": repr(e)}
return result
-def _run_on_examples(
- client: Client,
- examples: Iterator[Example],
- llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
- *,
- evaluation: Optional[RunEvalConfig] = None,
- project_name: Optional[str] = None,
- verbose: bool = False,
- tags: Optional[List[str]] = None,
- input_mapper: Optional[Callable[[Dict], Any]] = None,
- data_type: DataType = DataType.kv,
-) -> Dict[str, Any]:
- """
- Run the Chain or language model on examples and store
- traces to the specified project name.
-
- Args:
- client: LangSmith client to use to log feedback and runs.
- examples: Examples to run the model or chain over.
- llm_or_chain_factory: Language model or Chain constructor to run
- over the dataset. The Chain constructor is used to permit
- independent calls on each example without carrying over state.
- evaluation: Optional evaluation configuration to use when evaluating
- project_name: Name of the project to store the traces in.
- Defaults to {dataset_name}-{chain class name}-{datetime}.
- verbose: Whether to print progress.
- tags: Tags to add to each run in the project.
- input_mapper: A function to map to the inputs dictionary from an Example
- to the format expected by the model to be evaluated. This is useful if
- your model needs to deserialize more complex schema or if your dataset
- has inputs with keys that differ from what is expected by your chain
- or agent.
- data_type: The dataset's data type. This is used to determine determine
- how to deserialize the reference data and model compatibility.
-
- Returns:
- A dictionary mapping example ids to the model outputs.
- """
- results: Dict[str, dict] = {}
- wrapped_model = _wrap_in_chain_factory(llm_or_chain_factory)
- project_name = _get_project_name(project_name, wrapped_model)
- tracer = LangChainTracer(
- project_name=project_name, client=client, use_threading=False
- )
- run_evaluators, examples = _setup_evaluation(
- wrapped_model, examples, evaluation, data_type
- )
- examples = _validate_example_inputs(examples, wrapped_model, input_mapper)
- evaluation_handler = EvaluatorCallbackHandler(
- evaluators=run_evaluators or [],
- client=client,
- )
- callbacks: List[BaseCallbackHandler] = [tracer, evaluation_handler]
- for i, example in enumerate(examples):
- result = _run_llm_or_chain(
- example,
- wrapped_model,
- tags=tags,
- callbacks=callbacks,
- input_mapper=input_mapper,
- )
- if verbose:
- print(f"{i+1} processed", flush=True, end="\r")
- results[str(example.id)] = {"output": result}
- tracer.wait_for_futures()
- evaluation_handler.wait_for_futures()
- all_feedback = evaluation_handler.logged_feedback
- # join the results and feedback on the example id
- for example_id, output_dict in results.items():
- feedback = all_feedback.get(example_id, [])
- output_dict["feedback"] = feedback
- return results
-
-
## Public API
@@ -1156,10 +852,9 @@ def _prepare_eval_run(
client: Client,
dataset_name: str,
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
- project_name: Optional[str],
-) -> Tuple[MCF, str, Dataset, Iterator[Example]]:
+ project_name: str,
+) -> Tuple[MCF, str, Dataset, List[Example]]:
wrapped_model = _wrap_in_chain_factory(llm_or_chain_factory, dataset_name)
- project_name = _get_project_name(project_name, wrapped_model)
try:
project = client.create_project(project_name)
except ValueError as e:
@@ -1168,21 +863,95 @@ def _prepare_eval_run(
raise ValueError(
f"Project {project_name} already exists. Please use a different name."
)
- project_url = _get_eval_project_url(client.api_url, project.id)
print(
- f"View the evaluation results for project '{project_name}' at:\n{project_url}"
+ f"View the evaluation results for project '{project_name}' at:\n{project.url}"
)
dataset = client.read_dataset(dataset_name=dataset_name)
- examples = client.list_examples(dataset_id=str(dataset.id))
+ examples = list(client.list_examples(dataset_id=dataset.id))
+ if not examples:
+ raise ValueError(f"Dataset {dataset_name} has no example rows.")
return wrapped_model, project_name, dataset, examples
+def _prepare_run_on_dataset(
+ client: Client,
+ dataset_name: str,
+ llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
+ project_name: Optional[str],
+ evaluation: Optional[smith_eval.RunEvalConfig] = None,
+ tags: Optional[List[str]] = None,
+ input_mapper: Optional[Callable[[Dict], Any]] = None,
+ concurrency_level: int = 5,
+) -> Tuple[MCF, str, List[Example], List[RunnableConfig]]:
+ project_name = project_name or name_generation.random_name()
+ wrapped_model, project_name, dataset, examples = _prepare_eval_run(
+ client, dataset_name, llm_or_chain_factory, project_name
+ )
+ wrapped_model = _wrap_in_chain_factory(llm_or_chain_factory)
+ run_evaluators = _setup_evaluation(
+ wrapped_model, examples, evaluation, dataset.data_type
+ )
+ _validate_example_inputs(examples[0], wrapped_model, input_mapper)
+ progress_bar = progress.ProgressBarCallback(len(examples))
+ configs = [
+ RunnableConfig(
+ callbacks=[
+ LangChainTracer(
+ project_name=project_name,
+ client=client,
+ use_threading=False,
+ example_id=example.id,
+ ),
+ EvaluatorCallbackHandler(
+ evaluators=run_evaluators or [],
+ client=client,
+ max_workers=0,
+ example_id=example.id,
+ ),
+ progress_bar,
+ ],
+ tags=tags or [],
+ max_concurrency=concurrency_level,
+ )
+ for example in examples
+ ]
+ return wrapped_model, project_name, examples, configs
+
+
+def _collect_test_results(
+ examples: List[Example],
+ batch_results: List[Union[dict, str, LLMResult, ChatResult]],
+ configs: List[RunnableConfig],
+ project_name: str,
+) -> TestResult:
+ wait_for_all_tracers()
+ all_feedback = {}
+ for c in configs:
+ for callback in cast(list, c["callbacks"]):
+ if isinstance(callback, EvaluatorCallbackHandler):
+ all_feedback.update(callback.logged_feedback)
+ results = {}
+ for example, output in zip(examples, batch_results):
+ feedback = all_feedback.get(str(example.id), [])
+ results[str(example.id)] = {
+ "output": output,
+ "input": example.inputs,
+ "feedback": feedback,
+ }
+ if example.outputs:
+ results[str(example.id)]["reference"] = example.outputs
+ return TestResult(
+ project_name=project_name,
+ results=results,
+ )
+
+
async def arun_on_dataset(
client: Client,
dataset_name: str,
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
*,
- evaluation: Optional[RunEvalConfig] = None,
+ evaluation: Optional[smith_eval.RunEvalConfig] = None,
concurrency_level: int = 5,
project_name: Optional[str] = None,
verbose: bool = False,
@@ -1227,7 +996,7 @@ async def arun_on_dataset(
from langsmith import Client
from langchain.chat_models import ChatOpenAI
from langchain.chains import LLMChain
- from langchain.smith import RunEvalConfig, arun_on_dataset
+ from langchain.smith import smith_eval.RunEvalConfig, arun_on_dataset
# Chains may have memory. Passing in a constructor function lets the
# evaluation framework avoid cross-contamination between runs.
@@ -1240,12 +1009,12 @@ async def arun_on_dataset(
return chain
# Load off-the-shelf evaluators via config or the EvaluatorType (string or enum)
- evaluation_config = RunEvalConfig(
+ evaluation_config = smith_eval.RunEvalConfig(
evaluators=[
"qa", # "Correctness" against a reference answer
"embedding_distance",
- RunEvalConfig.Criteria("helpfulness"),
- RunEvalConfig.Criteria({
+ smith_eval.RunEvalConfig.Criteria("helpfulness"),
+ smith_eval.RunEvalConfig.Criteria({
"fifth-grader-score": "Do you have to be smarter than a fifth grader to answer this question?"
}),
]
@@ -1286,7 +1055,7 @@ async def arun_on_dataset(
return {"score": prediction == reference}
- evaluation_config = RunEvalConfig(
+ evaluation_config = smith_eval.RunEvalConfig(
custom_evaluators = [MyStringEvaluator()],
)
@@ -1299,51 +1068,43 @@ async def arun_on_dataset(
""" # noqa: E501
if kwargs:
warnings.warn(
- "The following arguments are deprecated and will "
- "be removed in a future release: "
+ "The following arguments are deprecated and "
+ "will be removed in a future release: "
f"{kwargs.keys()}.",
DeprecationWarning,
)
- wrapped_model, project_name, dataset, examples = _prepare_eval_run(
- client, dataset_name, llm_or_chain_factory, project_name
- )
- results = await _arun_on_examples(
+ wrapped_model, project_name, examples, configs = _prepare_run_on_dataset(
client,
- examples,
- wrapped_model,
- concurrency_level=concurrency_level,
- project_name=project_name,
- verbose=verbose,
- tags=tags,
- evaluation=evaluation,
- input_mapper=input_mapper,
- data_type=dataset.data_type,
- )
- return TestResult(
- project_name=project_name,
- results=results,
+ dataset_name,
+ llm_or_chain_factory,
+ project_name,
+ evaluation,
+ tags,
+ input_mapper,
+ concurrency_level,
)
-
-def _handle_coroutine(coro: Coroutine) -> Any:
- """
- Handles a coroutine from a sync context.
-
- Args:
- coro (asyncio.coroutine): The coroutine to be handled.
-
- Returns:
- any: The result of the executed coroutine.
- """
- # Check if there's a running event loop
- try:
- loop = asyncio.get_event_loop()
- except RuntimeError: # No event loop
- return asyncio.run(coro)
- if loop.is_running():
- return loop.run_until_complete(coro)
- else:
- return asyncio.run(coro)
+ batch_results = await runnable_utils.gather_with_concurrency(
+ configs[0].get("max_concurrency"),
+ *map(
+ functools.partial(
+ _arun_llm_or_chain,
+ llm_or_chain_factory=wrapped_model,
+ input_mapper=input_mapper,
+ ),
+ examples,
+ configs,
+ ),
+ )
+ results = _collect_test_results(examples, batch_results, configs, project_name)
+ if verbose:
+ try:
+ agg_feedback = results.get_aggregate_feedback()
+ print("\n Eval quantiles:")
+ print(agg_feedback)
+ except Exception as e:
+ logger.debug(f"Failed to print aggregate feedback: {repr(e)}")
+ return results
def run_on_dataset(
@@ -1351,7 +1112,7 @@ def run_on_dataset(
dataset_name: str,
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
*,
- evaluation: Optional[RunEvalConfig] = None,
+ evaluation: Optional[smith_eval.RunEvalConfig] = None,
concurrency_level: int = 5,
project_name: Optional[str] = None,
verbose: bool = False,
@@ -1397,7 +1158,7 @@ def run_on_dataset(
from langsmith import Client
from langchain.chat_models import ChatOpenAI
from langchain.chains import LLMChain
- from langchain.smith import RunEvalConfig, run_on_dataset
+ from langchain.smith import smith_eval.RunEvalConfig, run_on_dataset
# Chains may have memory. Passing in a constructor function lets the
# evaluation framework avoid cross-contamination between runs.
@@ -1410,12 +1171,12 @@ def run_on_dataset(
return chain
# Load off-the-shelf evaluators via config or the EvaluatorType (string or enum)
- evaluation_config = RunEvalConfig(
+ evaluation_config = smith_eval.RunEvalConfig(
evaluators=[
"qa", # "Correctness" against a reference answer
"embedding_distance",
- RunEvalConfig.Criteria("helpfulness"),
- RunEvalConfig.Criteria({
+ smith_eval.RunEvalConfig.Criteria("helpfulness"),
+ smith_eval.RunEvalConfig.Criteria({
"fifth-grader-score": "Do you have to be smarter than a fifth grader to answer this question?"
}),
]
@@ -1456,7 +1217,7 @@ def run_on_dataset(
return {"score": prediction == reference}
- evaluation_config = RunEvalConfig(
+ evaluation_config = smith_eval.RunEvalConfig(
custom_evaluators = [MyStringEvaluator()],
)
@@ -1474,37 +1235,35 @@ def run_on_dataset(
f"{kwargs.keys()}.",
DeprecationWarning,
)
- wrapped_model, project_name, dataset, examples = _prepare_eval_run(
- client, dataset_name, llm_or_chain_factory, project_name
+ wrapped_model, project_name, examples, configs = _prepare_run_on_dataset(
+ client,
+ dataset_name,
+ llm_or_chain_factory,
+ project_name,
+ evaluation,
+ tags,
+ input_mapper,
+ concurrency_level,
)
- if concurrency_level in (0, 1):
- results = _run_on_examples(
- client,
- examples,
- wrapped_model,
- project_name=project_name,
- verbose=verbose,
- tags=tags,
- evaluation=evaluation,
- input_mapper=input_mapper,
- data_type=dataset.data_type,
- )
- else:
- # TODO: Use runnables and the batch method
- coro = _arun_on_examples(
- client,
- examples,
- wrapped_model,
- concurrency_level=concurrency_level,
- project_name=project_name,
- verbose=verbose,
- tags=tags,
- evaluation=evaluation,
- input_mapper=input_mapper,
- data_type=dataset.data_type,
+ with runnable_config.get_executor_for_config(configs[0]) as executor:
+ batch_results = list(
+ executor.map(
+ functools.partial(
+ _run_llm_or_chain,
+ llm_or_chain_factory=wrapped_model,
+ input_mapper=input_mapper,
+ ),
+ examples,
+ configs,
+ )
)
- results = _handle_coroutine(coro)
- return TestResult(
- project_name=project_name,
- results=results,
- )
+
+ results = _collect_test_results(examples, batch_results, configs, project_name)
+ if verbose:
+ try:
+ agg_feedback = results.get_aggregate_feedback()
+ print("\n Eval quantiles:")
+ print(agg_feedback)
+ except Exception as e:
+ logger.debug(f"Failed to print aggregate feedback: {repr(e)}")
+ return results
diff --git a/libs/langchain/langchain/smith/evaluation/string_run_evaluator.py b/libs/langchain/langchain/smith/evaluation/string_run_evaluator.py
index d8133b19d8..1c14017673 100644
--- a/libs/langchain/langchain/smith/evaluation/string_run_evaluator.py
+++ b/libs/langchain/langchain/smith/evaluation/string_run_evaluator.py
@@ -148,13 +148,27 @@ class ChainStringRunMapper(StringRunMapper):
def map(self, run: Run) -> Dict[str, str]:
"""Maps the Run to a dictionary."""
if not run.outputs:
- raise ValueError(f"Run {run.id} has no outputs to evaluate.")
+ raise ValueError(
+ f"Run with ID {run.id} lacks outputs required for evaluation."
+ " Ensure the Run has valid outputs."
+ )
if self.input_key is not None and self.input_key not in run.inputs:
- raise ValueError(f"Run {run.id} does not have input key {self.input_key}.")
+ raise ValueError(
+ f"Run with ID {run.id} is missing the expected input key"
+ f" '{self.input_key}'.\nAvailable input keys in this Run"
+ f" are: {run.inputs.keys()}.\nAdjust the evaluator's"
+ f" input_key or ensure your input data includes key"
+ f" '{self.input_key}'."
+ )
elif self.prediction_key is not None and self.prediction_key not in run.outputs:
+ available_keys = ", ".join(run.outputs.keys())
raise ValueError(
- f"Run {run.id} does not have prediction key {self.prediction_key}."
+ f"Run with ID {run.id} doesn't have the expected prediction key"
+ f" '{self.prediction_key}'. Available prediction keys in this Run are:"
+ f" {available_keys}. Adjust the evaluator's prediction_key or"
+ " ensure the Run object's outputs the expected key."
)
+
else:
input_ = self._get_key(run.inputs, self.input_key, "input")
prediction = self._get_key(run.outputs, self.prediction_key, "prediction")
diff --git a/libs/langchain/langchain/text_splitter.py b/libs/langchain/langchain/text_splitter.py
index e804b93be9..d0bf6fca1b 100644
--- a/libs/langchain/langchain/text_splitter.py
+++ b/libs/langchain/langchain/text_splitter.py
@@ -100,6 +100,7 @@ class TextSplitter(BaseDocumentTransformer, ABC):
length_function: Callable[[str], int] = len,
keep_separator: bool = False,
add_start_index: bool = False,
+ strip_whitespace: bool = True,
) -> None:
"""Create a new TextSplitter.
@@ -109,6 +110,8 @@ class TextSplitter(BaseDocumentTransformer, ABC):
length_function: Function that measures the length of given chunks
keep_separator: Whether to keep the separator in the chunks
add_start_index: If `True`, includes chunk's start index in metadata
+ strip_whitespace: If `True`, strips whitespace from the start and end of
+ every document
"""
if chunk_overlap > chunk_size:
raise ValueError(
@@ -120,6 +123,7 @@ class TextSplitter(BaseDocumentTransformer, ABC):
self._length_function = length_function
self._keep_separator = keep_separator
self._add_start_index = add_start_index
+ self._strip_whitespace = strip_whitespace
@abstractmethod
def split_text(self, text: str) -> List[str]:
@@ -152,7 +156,8 @@ class TextSplitter(BaseDocumentTransformer, ABC):
def _join_docs(self, docs: List[str], separator: str) -> Optional[str]:
text = separator.join(docs)
- text = text.strip()
+ if self._strip_whitespace:
+ text = text.strip()
if text == "":
return None
else:
diff --git a/libs/langchain/langchain/tools/base.py b/libs/langchain/langchain/tools/base.py
index 9ad81033d5..69597cd903 100644
--- a/libs/langchain/langchain/tools/base.py
+++ b/libs/langchain/langchain/tools/base.py
@@ -592,7 +592,7 @@ class StructuredTool(BaseTool):
None, partial(self.invoke, input, config, **kwargs)
)
- return super().ainvoke(input, config, **kwargs)
+ return await super().ainvoke(input, config, **kwargs)
# --- Tool ---
diff --git a/libs/langchain/langchain/tools/sql_database/tool.py b/libs/langchain/langchain/tools/sql_database/tool.py
index f60275bcaf..75f45c7b9e 100644
--- a/libs/langchain/langchain/tools/sql_database/tool.py
+++ b/libs/langchain/langchain/tools/sql_database/tool.py
@@ -93,7 +93,7 @@ class QuerySQLCheckerTool(BaseSQLDatabaseTool, BaseTool):
name: str = "sql_db_query_checker"
description: str = """
Use this tool to double check if your query is correct before executing it.
- Always use this tool before executing a query with query_sql_db!
+ Always use this tool before executing a query with sql_db_query!
"""
@root_validator(pre=True)
diff --git a/libs/langchain/langchain/utilities/sql_database.py b/libs/langchain/langchain/utilities/sql_database.py
index 110f081d3c..13718c8c0c 100644
--- a/libs/langchain/langchain/utilities/sql_database.py
+++ b/libs/langchain/langchain/utilities/sql_database.py
@@ -9,6 +9,7 @@ from sqlalchemy import MetaData, Table, create_engine, inspect, select, text
from sqlalchemy.engine import Engine
from sqlalchemy.exc import ProgrammingError, SQLAlchemyError
from sqlalchemy.schema import CreateTable
+from sqlalchemy.types import NullType
from langchain.utils import get_from_env
@@ -314,6 +315,11 @@ class SQLDatabase:
tables.append(self._custom_table_info[table.name])
continue
+ # Ignore JSON datatyped columns
+ for k, v in table.columns.items():
+ if type(v.type) is NullType:
+ table._columns.remove(v)
+
# add create table command
create_table = str(CreateTable(table).compile(self._engine))
table_info = f"{create_table.rstrip()}"
@@ -384,6 +390,8 @@ class SQLDatabase:
connection.exec_driver_sql(f"SET @@dataset_id='{self._schema}'")
elif self.dialect == "mssql":
pass
+ elif self.dialect == "trino":
+ connection.exec_driver_sql(f"USE {self._schema}")
else: # postgresql and compatible dialects
connection.exec_driver_sql(f"SET search_path TO {self._schema}")
cursor = connection.execute(text(command))
diff --git a/libs/langchain/langchain/vectorstores/nucliadb.py b/libs/langchain/langchain/vectorstores/nucliadb.py
new file mode 100644
index 0000000000..8ba9d4454c
--- /dev/null
+++ b/libs/langchain/langchain/vectorstores/nucliadb.py
@@ -0,0 +1,159 @@
+import os
+from typing import Any, Dict, Iterable, List, Optional, Type
+
+from langchain.embeddings.base import Embeddings
+from langchain.schema.document import Document
+from langchain.vectorstores.base import VST, VectorStore
+
+FIELD_TYPES = {
+ "f": "files",
+ "t": "texts",
+ "l": "links",
+}
+
+
+class NucliaDB(VectorStore):
+ """NucliaDB vector store."""
+
+ _config: Dict[str, Any] = {}
+
+ def __init__(
+ self,
+ knowledge_box: str,
+ local: bool,
+ api_key: Optional[str] = None,
+ backend: Optional[str] = None,
+ ) -> None:
+ """Initialize the NucliaDB client.
+
+ Args:
+ knowledge_box: the Knowledge Box id.
+ local: Whether to use a local NucliaDB instance or Nuclia Cloud
+ api_key: A contributor API key for the kb (needed when local is False)
+ backend: The backend url to use when local is True, defaults to
+ http://localhost:8080
+ """
+ try:
+ from nuclia.sdk import NucliaAuth
+ except ImportError:
+ raise ValueError(
+ "nuclia python package not found. "
+ "Please install it with `pip install nuclia`."
+ )
+ self._config["LOCAL"] = local
+ zone = os.environ.get("NUCLIA_ZONE", "europe-1")
+ self._kb = knowledge_box
+ if local:
+ if not backend:
+ backend = "http://localhost:8080"
+ self._config["BACKEND"] = f"{backend}/api/v1"
+ self._config["TOKEN"] = None
+ NucliaAuth().nucliadb(url=backend)
+ NucliaAuth().kb(url=self.kb_url, interactive=False)
+ else:
+ self._config["BACKEND"] = f"https://{zone}.nuclia.cloud/api/v1"
+ self._config["TOKEN"] = api_key
+ NucliaAuth().kb(
+ url=self.kb_url, token=self._config["TOKEN"], interactive=False
+ )
+
+ @property
+ def is_local(self) -> str:
+ return self._config["LOCAL"]
+
+ @property
+ def kb_url(self) -> str:
+ return f"{self._config['BACKEND']}/kb/{self._kb}"
+
+ def add_texts(
+ self,
+ texts: Iterable[str],
+ metadatas: Optional[List[dict]] = None,
+ **kwargs: Any,
+ ) -> List[str]:
+ """Upload texts to NucliaDB"""
+ ids = []
+ from nuclia.sdk import NucliaResource
+
+ factory = NucliaResource()
+ for i, text in enumerate(texts):
+ extra: Dict[str, Any] = {"metadata": ""}
+ if metadatas:
+ extra = {"metadata": metadatas[i]}
+ id = factory.create(
+ texts={"text": {"body": text}},
+ extra=extra,
+ url=self.kb_url,
+ api_key=self._config["TOKEN"],
+ )
+ ids.append(id)
+ return ids
+
+ def delete(self, ids: Optional[List[str]] = None, **kwargs: Any) -> Optional[bool]:
+ if not ids:
+ return None
+ from nuclia.sdk import NucliaResource
+
+ factory = NucliaResource()
+ results: List[bool] = []
+ for id in ids:
+ try:
+ factory.delete(rid=id, url=self.kb_url, api_key=self._config["TOKEN"])
+ results.append(True)
+ except ValueError:
+ results.append(False)
+ return all(results)
+
+ def similarity_search(
+ self, query: str, k: int = 4, **kwargs: Any
+ ) -> List[Document]:
+ from nuclia.sdk import NucliaSearch
+ from nucliadb_models.search import FindRequest, ResourceProperties
+
+ request = FindRequest(
+ query=query,
+ page_size=k,
+ show=[ResourceProperties.VALUES, ResourceProperties.EXTRA],
+ )
+ search = NucliaSearch()
+ results = search.find(
+ query=request, url=self.kb_url, api_key=self._config["TOKEN"]
+ )
+ paragraphs = []
+ for resource in results.resources.values():
+ for field in resource.fields.values():
+ for paragraph_id, paragraph in field.paragraphs.items():
+ info = paragraph_id.split("/")
+ field_type = FIELD_TYPES.get(info[1], None)
+ field_id = info[2]
+ if not field_type:
+ continue
+ value = getattr(resource.data, field_type, {}).get(field_id, None)
+ paragraphs.append(
+ {
+ "text": paragraph.text,
+ "metadata": {
+ "extra": getattr(
+ getattr(resource, "extra", {}), "metadata", None
+ ),
+ "value": value,
+ },
+ "order": paragraph.order,
+ }
+ )
+ sorted_paragraphs = sorted(paragraphs, key=lambda x: x["order"])
+ return [
+ Document(page_content=paragraph["text"], metadata=paragraph["metadata"])
+ for paragraph in sorted_paragraphs
+ ]
+
+ @classmethod
+ def from_texts(
+ cls: Type[VST],
+ texts: List[str],
+ embedding: Embeddings,
+ metadatas: Optional[List[dict]] = None,
+ **kwargs: Any,
+ ) -> VST:
+ """Return VectorStore initialized from texts and embeddings."""
+ raise NotImplementedError
diff --git a/libs/langchain/langchain/vectorstores/pgvector.py b/libs/langchain/langchain/vectorstores/pgvector.py
index 6b02fc19c0..2fc66c0a0e 100644
--- a/libs/langchain/langchain/vectorstores/pgvector.py
+++ b/libs/langchain/langchain/vectorstores/pgvector.py
@@ -349,16 +349,16 @@ class PGVector(VectorStore):
@property
def distance_strategy(self) -> Any:
- if self._distance_strategy == "l2":
+ if self._distance_strategy == DistanceStrategy.EUCLIDEAN:
return self.EmbeddingStore.embedding.l2_distance
- elif self._distance_strategy == "cosine":
+ elif self._distance_strategy == DistanceStrategy.COSINE:
return self.EmbeddingStore.embedding.cosine_distance
- elif self._distance_strategy == "inner":
+ elif self._distance_strategy == DistanceStrategy.MAX_INNER_PRODUCT:
return self.EmbeddingStore.embedding.max_inner_product
else:
raise ValueError(
f"Got unexpected value for distance: {self._distance_strategy}. "
- f"Should be one of `l2`, `cosine`, `inner`."
+ f"Should be one of {', '.join([ds.value for ds in DistanceStrategy])}."
)
def similarity_search_with_score_by_vector(
diff --git a/libs/langchain/langchain/vectorstores/supabase.py b/libs/langchain/langchain/vectorstores/supabase.py
index d911965346..71b7c2cc8b 100644
--- a/libs/langchain/langchain/vectorstores/supabase.py
+++ b/libs/langchain/langchain/vectorstores/supabase.py
@@ -168,10 +168,8 @@ class SupabaseVectorStore(VectorStore):
filter: Optional[Dict[str, Any]] = None,
**kwargs: Any,
) -> List[Document]:
- vectors = self._embedding.embed_documents([query])
- return self.similarity_search_by_vector(
- vectors[0], k=k, filter=filter, **kwargs
- )
+ vector = self._embedding.embed_query(query)
+ return self.similarity_search_by_vector(vector, k=k, filter=filter, **kwargs)
def similarity_search_by_vector(
self,
@@ -195,24 +193,37 @@ class SupabaseVectorStore(VectorStore):
filter: Optional[Dict[str, Any]] = None,
**kwargs: Any,
) -> List[Tuple[Document, float]]:
- vectors = self._embedding.embed_documents([query])
+ vector = self._embedding.embed_query(query)
return self.similarity_search_by_vector_with_relevance_scores(
- vectors[0], k=k, filter=filter
+ vector, k=k, filter=filter
)
def match_args(
- self, query: List[float], k: int, filter: Optional[Dict[str, Any]]
+ self, query: List[float], filter: Optional[Dict[str, Any]]
) -> Dict[str, Any]:
- ret = dict(query_embedding=query, match_count=k)
+ ret: Dict[str, Any] = dict(query_embedding=query)
if filter:
ret["filter"] = filter
return ret
def similarity_search_by_vector_with_relevance_scores(
- self, query: List[float], k: int, filter: Optional[Dict[str, Any]] = None
+ self,
+ query: List[float],
+ k: int,
+ filter: Optional[Dict[str, Any]] = None,
+ postgrest_filter: Optional[str] = None,
) -> List[Tuple[Document, float]]:
- match_documents_params = self.match_args(query, k, filter)
- res = self._client.rpc(self.query_name, match_documents_params).execute()
+ match_documents_params = self.match_args(query, filter)
+ query_builder = self._client.rpc(self.query_name, match_documents_params)
+
+ if postgrest_filter:
+ query_builder.params = query_builder.params.set(
+ "and", f"({postgrest_filter})"
+ )
+
+ query_builder.params = query_builder.params.set("limit", k)
+
+ res = query_builder.execute()
match_result = [
(
@@ -229,10 +240,23 @@ class SupabaseVectorStore(VectorStore):
return match_result
def similarity_search_by_vector_returning_embeddings(
- self, query: List[float], k: int, filter: Optional[Dict[str, Any]] = None
+ self,
+ query: List[float],
+ k: int,
+ filter: Optional[Dict[str, Any]] = None,
+ postgrest_filter: Optional[str] = None,
) -> List[Tuple[Document, float, np.ndarray[np.float32, Any]]]:
- match_documents_params = self.match_args(query, k, filter)
- res = self._client.rpc(self.query_name, match_documents_params).execute()
+ match_documents_params = self.match_args(query, filter)
+ query_builder = self._client.rpc(self.query_name, match_documents_params)
+
+ if postgrest_filter:
+ query_builder.params = query_builder.params.set(
+ "and", f"({postgrest_filter})"
+ )
+
+ query_builder.params = query_builder.params.set("limit", k)
+
+ res = query_builder.execute()
match_result = [
(
@@ -407,9 +431,9 @@ class SupabaseVectorStore(VectorStore):
$$;
```
"""
- embedding = self._embedding.embed_documents([query])
+ embedding = self._embedding.embed_query(query)
docs = self.max_marginal_relevance_search_by_vector(
- embedding[0], k, fetch_k, lambda_mult=lambda_mult
+ embedding, k, fetch_k, lambda_mult=lambda_mult
)
return docs
diff --git a/libs/langchain/langchain/vectorstores/vectara.py b/libs/langchain/langchain/vectorstores/vectara.py
index 457511b104..3e8a2549e2 100644
--- a/libs/langchain/langchain/vectorstores/vectara.py
+++ b/libs/langchain/langchain/vectorstores/vectara.py
@@ -396,8 +396,12 @@ class Vectara(VectorStore):
vectara_api_key=api_key,
)
"""
- # Note: Vectara generates its own embeddings, so we ignore the provided
- # embeddings (required by interface)
+ # Notes:
+ # * Vectara generates its own embeddings, so we ignore the provided
+ # embeddings (required by interface)
+ # * when metadatas[] are provided they are associated with each "part"
+ # in Vectara. doc_metadata can be used to provide additional metadata
+ # for the document itself (applies to all "texts" in this call)
doc_metadata = kwargs.pop("doc_metadata", {})
vectara = cls(**kwargs)
vectara.add_texts(texts, metadatas, doc_metadata=doc_metadata, **kwargs)
diff --git a/libs/langchain/poetry.lock b/libs/langchain/poetry.lock
index 8f31c88d0e..399dcc2cab 100644
--- a/libs/langchain/poetry.lock
+++ b/libs/langchain/poetry.lock
@@ -2780,7 +2780,6 @@ files = [
{file = "greenlet-2.0.2-cp27-cp27m-win32.whl", hash = "sha256:6c3acb79b0bfd4fe733dff8bc62695283b57949ebcca05ae5c129eb606ff2d74"},
{file = "greenlet-2.0.2-cp27-cp27m-win_amd64.whl", hash = "sha256:283737e0da3f08bd637b5ad058507e578dd462db259f7f6e4c5c365ba4ee9343"},
{file = "greenlet-2.0.2-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:d27ec7509b9c18b6d73f2f5ede2622441de812e7b1a80bbd446cb0633bd3d5ae"},
- {file = "greenlet-2.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d967650d3f56af314b72df7089d96cda1083a7fc2da05b375d2bc48c82ab3f3c"},
{file = "greenlet-2.0.2-cp310-cp310-macosx_11_0_x86_64.whl", hash = "sha256:30bcf80dda7f15ac77ba5af2b961bdd9dbc77fd4ac6105cee85b0d0a5fcf74df"},
{file = "greenlet-2.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:26fbfce90728d82bc9e6c38ea4d038cba20b7faf8a0ca53a9c07b67318d46088"},
{file = "greenlet-2.0.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9190f09060ea4debddd24665d6804b995a9c122ef5917ab26e1566dcc712ceeb"},
@@ -2789,7 +2788,6 @@ files = [
{file = "greenlet-2.0.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:76ae285c8104046b3a7f06b42f29c7b73f77683df18c49ab5af7983994c2dd91"},
{file = "greenlet-2.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:2d4686f195e32d36b4d7cf2d166857dbd0ee9f3d20ae349b6bf8afc8485b3645"},
{file = "greenlet-2.0.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:c4302695ad8027363e96311df24ee28978162cdcdd2006476c43970b384a244c"},
- {file = "greenlet-2.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:d4606a527e30548153be1a9f155f4e283d109ffba663a15856089fb55f933e47"},
{file = "greenlet-2.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c48f54ef8e05f04d6eff74b8233f6063cb1ed960243eacc474ee73a2ea8573ca"},
{file = "greenlet-2.0.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a1846f1b999e78e13837c93c778dcfc3365902cfb8d1bdb7dd73ead37059f0d0"},
{file = "greenlet-2.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3a06ad5312349fec0ab944664b01d26f8d1f05009566339ac6f63f56589bc1a2"},
@@ -2819,7 +2817,6 @@ files = [
{file = "greenlet-2.0.2-cp37-cp37m-win32.whl", hash = "sha256:3f6ea9bd35eb450837a3d80e77b517ea5bc56b4647f5502cd28de13675ee12f7"},
{file = "greenlet-2.0.2-cp37-cp37m-win_amd64.whl", hash = "sha256:7492e2b7bd7c9b9916388d9df23fa49d9b88ac0640db0a5b4ecc2b653bf451e3"},
{file = "greenlet-2.0.2-cp38-cp38-macosx_10_15_x86_64.whl", hash = "sha256:b864ba53912b6c3ab6bcb2beb19f19edd01a6bfcbdfe1f37ddd1778abfe75a30"},
- {file = "greenlet-2.0.2-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:1087300cf9700bbf455b1b97e24db18f2f77b55302a68272c56209d5587c12d1"},
{file = "greenlet-2.0.2-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:ba2956617f1c42598a308a84c6cf021a90ff3862eddafd20c3333d50f0edb45b"},
{file = "greenlet-2.0.2-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fc3a569657468b6f3fb60587e48356fe512c1754ca05a564f11366ac9e306526"},
{file = "greenlet-2.0.2-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8eab883b3b2a38cc1e050819ef06a7e6344d4a990d24d45bc6f2cf959045a45b"},
@@ -2828,7 +2825,6 @@ files = [
{file = "greenlet-2.0.2-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:b0ef99cdbe2b682b9ccbb964743a6aca37905fda5e0452e5ee239b1654d37f2a"},
{file = "greenlet-2.0.2-cp38-cp38-win32.whl", hash = "sha256:b80f600eddddce72320dbbc8e3784d16bd3fb7b517e82476d8da921f27d4b249"},
{file = "greenlet-2.0.2-cp38-cp38-win_amd64.whl", hash = "sha256:4d2e11331fc0c02b6e84b0d28ece3a36e0548ee1a1ce9ddde03752d9b79bba40"},
- {file = "greenlet-2.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:8512a0c38cfd4e66a858ddd1b17705587900dd760c6003998e9472b77b56d417"},
{file = "greenlet-2.0.2-cp39-cp39-macosx_11_0_x86_64.whl", hash = "sha256:88d9ab96491d38a5ab7c56dd7a3cc37d83336ecc564e4e8816dbed12e5aaefc8"},
{file = "greenlet-2.0.2-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:561091a7be172ab497a3527602d467e2b3fbe75f9e783d8b8ce403fa414f71a6"},
{file = "greenlet-2.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:971ce5e14dc5e73715755d0ca2975ac88cfdaefcaab078a284fea6cfabf866df"},
@@ -3565,7 +3561,6 @@ optional = false
python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, !=3.5.*, !=3.6.*"
files = [
{file = "jsonpointer-2.4-py2.py3-none-any.whl", hash = "sha256:15d51bba20eea3165644553647711d150376234112651b4f1811022aecad7d7a"},
- {file = "jsonpointer-2.4.tar.gz", hash = "sha256:585cee82b70211fa9e6043b7bb89db6e1aa49524340dde8ad6b63206ea689d88"},
]
[[package]]
@@ -3888,20 +3883,6 @@ files = [
{file = "keras-2.13.1.tar.gz", hash = "sha256:5df12cc241a015a11b65ddb452c0eeb2744fce21d9b54ba48db87492568ccc68"},
]
-[[package]]
-name = "konko"
-version = "0.4.0"
-description = "\"Konko AI Python SDK: A client library to interact with Konko AI's API. With this SDK, developers can integrate, evaluate, fine-tune, and deploy Large Language Models (LLMs) using Konko AI's fully-managed, SOC 2 compliant platform. Ideal for enterprise use-cases, it allows you to focus on building great products without worrying about infrastructure.\""
-optional = true
-python-versions = ">=3.8"
-files = [
- {file = "konko-0.4.0-py3-none-any.whl", hash = "sha256:b51564f3b51687ba59d136673e8ea40040df02bcc4b9978c2266656f79f6ff51"},
- {file = "konko-0.4.0.tar.gz", hash = "sha256:787717e89c0fb3e74164a600a611c80e7fea2500040504c944b10f7668a15472"},
-]
-
-[package.dependencies]
-openai = ">=0.27.9,<0.28.0"
-
[[package]]
name = "lancedb"
version = "0.1.16"
@@ -5537,13 +5518,13 @@ signedtoken = ["cryptography (>=3.0.0)", "pyjwt (>=2.0.0,<3)"]
[[package]]
name = "openai"
-version = "0.27.10"
+version = "0.27.8"
description = "Python client library for the OpenAI API"
optional = false
python-versions = ">=3.7.1"
files = [
- {file = "openai-0.27.10-py3-none-any.whl", hash = "sha256:beabd1757e3286fa166dde3b70ebb5ad8081af046876b47c14c41e203ed22a14"},
- {file = "openai-0.27.10.tar.gz", hash = "sha256:60e09edf7100080283688748c6803b7b3b52d5a55d21890f3815292a0552d83b"},
+ {file = "openai-0.27.8-py3-none-any.whl", hash = "sha256:e0a7c2f7da26bdbe5354b03c6d4b82a2f34bd4458c7a17ae1a7092c3e397e03c"},
+ {file = "openai-0.27.8.tar.gz", hash = "sha256:2483095c7db1eee274cebac79e315a986c4e55207bb4fa7b82d185b3a2ed9536"},
]
[package.dependencies]
@@ -5910,67 +5891,77 @@ files = [
[[package]]
name = "pillow"
-version = "10.0.0"
+version = "9.5.0"
description = "Python Imaging Library (Fork)"
optional = true
-python-versions = ">=3.8"
+python-versions = ">=3.7"
files = [
- {file = "Pillow-10.0.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:1f62406a884ae75fb2f818694469519fb685cc7eaff05d3451a9ebe55c646891"},
- {file = "Pillow-10.0.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d5db32e2a6ccbb3d34d87c87b432959e0db29755727afb37290e10f6e8e62614"},
- {file = "Pillow-10.0.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:edf4392b77bdc81f36e92d3a07a5cd072f90253197f4a52a55a8cec48a12483b"},
- {file = "Pillow-10.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:520f2a520dc040512699f20fa1c363eed506e94248d71f85412b625026f6142c"},
- {file = "Pillow-10.0.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:8c11160913e3dd06c8ffdb5f233a4f254cb449f4dfc0f8f4549eda9e542c93d1"},
- {file = "Pillow-10.0.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:a74ba0c356aaa3bb8e3eb79606a87669e7ec6444be352870623025d75a14a2bf"},
- {file = "Pillow-10.0.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:d5d0dae4cfd56969d23d94dc8e89fb6a217be461c69090768227beb8ed28c0a3"},
- {file = "Pillow-10.0.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:22c10cc517668d44b211717fd9775799ccec4124b9a7f7b3635fc5386e584992"},
- {file = "Pillow-10.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:dffe31a7f47b603318c609f378ebcd57f1554a3a6a8effbc59c3c69f804296de"},
- {file = "Pillow-10.0.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:9fb218c8a12e51d7ead2a7c9e101a04982237d4855716af2e9499306728fb485"},
- {file = "Pillow-10.0.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:d35e3c8d9b1268cbf5d3670285feb3528f6680420eafe35cccc686b73c1e330f"},
- {file = "Pillow-10.0.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3ed64f9ca2f0a95411e88a4efbd7a29e5ce2cea36072c53dd9d26d9c76f753b3"},
- {file = "Pillow-10.0.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0b6eb5502f45a60a3f411c63187db83a3d3107887ad0d036c13ce836f8a36f1d"},
- {file = "Pillow-10.0.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:c1fbe7621c167ecaa38ad29643d77a9ce7311583761abf7836e1510c580bf3dd"},
- {file = "Pillow-10.0.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:cd25d2a9d2b36fcb318882481367956d2cf91329f6892fe5d385c346c0649629"},
- {file = "Pillow-10.0.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:3b08d4cc24f471b2c8ca24ec060abf4bebc6b144cb89cba638c720546b1cf538"},
- {file = "Pillow-10.0.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:d737a602fbd82afd892ca746392401b634e278cb65d55c4b7a8f48e9ef8d008d"},
- {file = "Pillow-10.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:3a82c40d706d9aa9734289740ce26460a11aeec2d9c79b7af87bb35f0073c12f"},
- {file = "Pillow-10.0.0-cp311-cp311-win_arm64.whl", hash = "sha256:bc2ec7c7b5d66b8ec9ce9f720dbb5fa4bace0f545acd34870eff4a369b44bf37"},
- {file = "Pillow-10.0.0-cp312-cp312-macosx_10_10_x86_64.whl", hash = "sha256:d80cf684b541685fccdd84c485b31ce73fc5c9b5d7523bf1394ce134a60c6883"},
- {file = "Pillow-10.0.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:76de421f9c326da8f43d690110f0e79fe3ad1e54be811545d7d91898b4c8493e"},
- {file = "Pillow-10.0.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81ff539a12457809666fef6624684c008e00ff6bf455b4b89fd00a140eecd640"},
- {file = "Pillow-10.0.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ce543ed15570eedbb85df19b0a1a7314a9c8141a36ce089c0a894adbfccb4568"},
- {file = "Pillow-10.0.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:685ac03cc4ed5ebc15ad5c23bc555d68a87777586d970c2c3e216619a5476223"},
- {file = "Pillow-10.0.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:d72e2ecc68a942e8cf9739619b7f408cc7b272b279b56b2c83c6123fcfa5cdff"},
- {file = "Pillow-10.0.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:d50b6aec14bc737742ca96e85d6d0a5f9bfbded018264b3b70ff9d8c33485551"},
- {file = "Pillow-10.0.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:00e65f5e822decd501e374b0650146063fbb30a7264b4d2744bdd7b913e0cab5"},
- {file = "Pillow-10.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:f31f9fdbfecb042d046f9d91270a0ba28368a723302786c0009ee9b9f1f60199"},
- {file = "Pillow-10.0.0-cp312-cp312-win_arm64.whl", hash = "sha256:1ce91b6ec08d866b14413d3f0bbdea7e24dfdc8e59f562bb77bc3fe60b6144ca"},
- {file = "Pillow-10.0.0-cp38-cp38-macosx_10_10_x86_64.whl", hash = "sha256:349930d6e9c685c089284b013478d6f76e3a534e36ddfa912cde493f235372f3"},
- {file = "Pillow-10.0.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:3a684105f7c32488f7153905a4e3015a3b6c7182e106fe3c37fbb5ef3e6994c3"},
- {file = "Pillow-10.0.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b4f69b3700201b80bb82c3a97d5e9254084f6dd5fb5b16fc1a7b974260f89f43"},
- {file = "Pillow-10.0.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3f07ea8d2f827d7d2a49ecf1639ec02d75ffd1b88dcc5b3a61bbb37a8759ad8d"},
- {file = "Pillow-10.0.0-cp38-cp38-manylinux_2_28_aarch64.whl", hash = "sha256:040586f7d37b34547153fa383f7f9aed68b738992380ac911447bb78f2abe530"},
- {file = "Pillow-10.0.0-cp38-cp38-manylinux_2_28_x86_64.whl", hash = "sha256:f88a0b92277de8e3ca715a0d79d68dc82807457dae3ab8699c758f07c20b3c51"},
- {file = "Pillow-10.0.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:c7cf14a27b0d6adfaebb3ae4153f1e516df54e47e42dcc073d7b3d76111a8d86"},
- {file = "Pillow-10.0.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:3400aae60685b06bb96f99a21e1ada7bc7a413d5f49bce739828ecd9391bb8f7"},
- {file = "Pillow-10.0.0-cp38-cp38-win_amd64.whl", hash = "sha256:dbc02381779d412145331789b40cc7b11fdf449e5d94f6bc0b080db0a56ea3f0"},
- {file = "Pillow-10.0.0-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:9211e7ad69d7c9401cfc0e23d49b69ca65ddd898976d660a2fa5904e3d7a9baa"},
- {file = "Pillow-10.0.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:faaf07ea35355b01a35cb442dd950d8f1bb5b040a7787791a535de13db15ed90"},
- {file = "Pillow-10.0.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c9f72a021fbb792ce98306ffb0c348b3c9cb967dce0f12a49aa4c3d3fdefa967"},
- {file = "Pillow-10.0.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9f7c16705f44e0504a3a2a14197c1f0b32a95731d251777dcb060aa83022cb2d"},
- {file = "Pillow-10.0.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:76edb0a1fa2b4745fb0c99fb9fb98f8b180a1bbceb8be49b087e0b21867e77d3"},
- {file = "Pillow-10.0.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:368ab3dfb5f49e312231b6f27b8820c823652b7cd29cfbd34090565a015e99ba"},
- {file = "Pillow-10.0.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:608bfdee0d57cf297d32bcbb3c728dc1da0907519d1784962c5f0c68bb93e5a3"},
- {file = "Pillow-10.0.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:5c6e3df6bdd396749bafd45314871b3d0af81ff935b2d188385e970052091017"},
- {file = "Pillow-10.0.0-cp39-cp39-win_amd64.whl", hash = "sha256:7be600823e4c8631b74e4a0d38384c73f680e6105a7d3c6824fcf226c178c7e6"},
- {file = "Pillow-10.0.0-pp310-pypy310_pp73-macosx_10_10_x86_64.whl", hash = "sha256:92be919bbc9f7d09f7ae343c38f5bb21c973d2576c1d45600fce4b74bafa7ac0"},
- {file = "Pillow-10.0.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8f8182b523b2289f7c415f589118228d30ac8c355baa2f3194ced084dac2dbba"},
- {file = "Pillow-10.0.0-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:38250a349b6b390ee6047a62c086d3817ac69022c127f8a5dc058c31ccef17f3"},
- {file = "Pillow-10.0.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:88af2003543cc40c80f6fca01411892ec52b11021b3dc22ec3bc9d5afd1c5334"},
- {file = "Pillow-10.0.0-pp39-pypy39_pp73-macosx_10_10_x86_64.whl", hash = "sha256:c189af0545965fa8d3b9613cfdb0cd37f9d71349e0f7750e1fd704648d475ed2"},
- {file = "Pillow-10.0.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ce7b031a6fc11365970e6a5686d7ba8c63e4c1cf1ea143811acbb524295eabed"},
- {file = "Pillow-10.0.0-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:db24668940f82321e746773a4bc617bfac06ec831e5c88b643f91f122a785684"},
- {file = "Pillow-10.0.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:efe8c0681042536e0d06c11f48cebe759707c9e9abf880ee213541c5b46c5bf3"},
- {file = "Pillow-10.0.0.tar.gz", hash = "sha256:9c82b5b3e043c7af0d95792d0d20ccf68f61a1fec6b3530e718b688422727396"},
+ {file = "Pillow-9.5.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:ace6ca218308447b9077c14ea4ef381ba0b67ee78d64046b3f19cf4e1139ad16"},
+ {file = "Pillow-9.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d3d403753c9d5adc04d4694d35cf0391f0f3d57c8e0030aac09d7678fa8030aa"},
+ {file = "Pillow-9.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5ba1b81ee69573fe7124881762bb4cd2e4b6ed9dd28c9c60a632902fe8db8b38"},
+ {file = "Pillow-9.5.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fe7e1c262d3392afcf5071df9afa574544f28eac825284596ac6db56e6d11062"},
+ {file = "Pillow-9.5.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8f36397bf3f7d7c6a3abdea815ecf6fd14e7fcd4418ab24bae01008d8d8ca15e"},
+ {file = "Pillow-9.5.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:252a03f1bdddce077eff2354c3861bf437c892fb1832f75ce813ee94347aa9b5"},
+ {file = "Pillow-9.5.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:85ec677246533e27770b0de5cf0f9d6e4ec0c212a1f89dfc941b64b21226009d"},
+ {file = "Pillow-9.5.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:b416f03d37d27290cb93597335a2f85ed446731200705b22bb927405320de903"},
+ {file = "Pillow-9.5.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:1781a624c229cb35a2ac31cc4a77e28cafc8900733a864870c49bfeedacd106a"},
+ {file = "Pillow-9.5.0-cp310-cp310-win32.whl", hash = "sha256:8507eda3cd0608a1f94f58c64817e83ec12fa93a9436938b191b80d9e4c0fc44"},
+ {file = "Pillow-9.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:d3c6b54e304c60c4181da1c9dadf83e4a54fd266a99c70ba646a9baa626819eb"},
+ {file = "Pillow-9.5.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:7ec6f6ce99dab90b52da21cf0dc519e21095e332ff3b399a357c187b1a5eee32"},
+ {file = "Pillow-9.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:560737e70cb9c6255d6dcba3de6578a9e2ec4b573659943a5e7e4af13f298f5c"},
+ {file = "Pillow-9.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:96e88745a55b88a7c64fa49bceff363a1a27d9a64e04019c2281049444a571e3"},
+ {file = "Pillow-9.5.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d9c206c29b46cfd343ea7cdfe1232443072bbb270d6a46f59c259460db76779a"},
+ {file = "Pillow-9.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cfcc2c53c06f2ccb8976fb5c71d448bdd0a07d26d8e07e321c103416444c7ad1"},
+ {file = "Pillow-9.5.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:a0f9bb6c80e6efcde93ffc51256d5cfb2155ff8f78292f074f60f9e70b942d99"},
+ {file = "Pillow-9.5.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:8d935f924bbab8f0a9a28404422da8af4904e36d5c33fc6f677e4c4485515625"},
+ {file = "Pillow-9.5.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:fed1e1cf6a42577953abbe8e6cf2fe2f566daebde7c34724ec8803c4c0cda579"},
+ {file = "Pillow-9.5.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:c1170d6b195555644f0616fd6ed929dfcf6333b8675fcca044ae5ab110ded296"},
+ {file = "Pillow-9.5.0-cp311-cp311-win32.whl", hash = "sha256:54f7102ad31a3de5666827526e248c3530b3a33539dbda27c6843d19d72644ec"},
+ {file = "Pillow-9.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:cfa4561277f677ecf651e2b22dc43e8f5368b74a25a8f7d1d4a3a243e573f2d4"},
+ {file = "Pillow-9.5.0-cp311-cp311-win_arm64.whl", hash = "sha256:965e4a05ef364e7b973dd17fc765f42233415974d773e82144c9bbaaaea5d089"},
+ {file = "Pillow-9.5.0-cp312-cp312-win32.whl", hash = "sha256:22baf0c3cf0c7f26e82d6e1adf118027afb325e703922c8dfc1d5d0156bb2eeb"},
+ {file = "Pillow-9.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:432b975c009cf649420615388561c0ce7cc31ce9b2e374db659ee4f7d57a1f8b"},
+ {file = "Pillow-9.5.0-cp37-cp37m-macosx_10_10_x86_64.whl", hash = "sha256:5d4ebf8e1db4441a55c509c4baa7a0587a0210f7cd25fcfe74dbbce7a4bd1906"},
+ {file = "Pillow-9.5.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:375f6e5ee9620a271acb6820b3d1e94ffa8e741c0601db4c0c4d3cb0a9c224bf"},
+ {file = "Pillow-9.5.0-cp37-cp37m-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:99eb6cafb6ba90e436684e08dad8be1637efb71c4f2180ee6b8f940739406e78"},
+ {file = "Pillow-9.5.0-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2dfaaf10b6172697b9bceb9a3bd7b951819d1ca339a5ef294d1f1ac6d7f63270"},
+ {file = "Pillow-9.5.0-cp37-cp37m-manylinux_2_28_aarch64.whl", hash = "sha256:763782b2e03e45e2c77d7779875f4432e25121ef002a41829d8868700d119392"},
+ {file = "Pillow-9.5.0-cp37-cp37m-manylinux_2_28_x86_64.whl", hash = "sha256:35f6e77122a0c0762268216315bf239cf52b88865bba522999dc38f1c52b9b47"},
+ {file = "Pillow-9.5.0-cp37-cp37m-win32.whl", hash = "sha256:aca1c196f407ec7cf04dcbb15d19a43c507a81f7ffc45b690899d6a76ac9fda7"},
+ {file = "Pillow-9.5.0-cp37-cp37m-win_amd64.whl", hash = "sha256:322724c0032af6692456cd6ed554bb85f8149214d97398bb80613b04e33769f6"},
+ {file = "Pillow-9.5.0-cp38-cp38-macosx_10_10_x86_64.whl", hash = "sha256:a0aa9417994d91301056f3d0038af1199eb7adc86e646a36b9e050b06f526597"},
+ {file = "Pillow-9.5.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:f8286396b351785801a976b1e85ea88e937712ee2c3ac653710a4a57a8da5d9c"},
+ {file = "Pillow-9.5.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c830a02caeb789633863b466b9de10c015bded434deb3ec87c768e53752ad22a"},
+ {file = "Pillow-9.5.0-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fbd359831c1657d69bb81f0db962905ee05e5e9451913b18b831febfe0519082"},
+ {file = "Pillow-9.5.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f8fc330c3370a81bbf3f88557097d1ea26cd8b019d6433aa59f71195f5ddebbf"},
+ {file = "Pillow-9.5.0-cp38-cp38-manylinux_2_28_aarch64.whl", hash = "sha256:7002d0797a3e4193c7cdee3198d7c14f92c0836d6b4a3f3046a64bd1ce8df2bf"},
+ {file = "Pillow-9.5.0-cp38-cp38-manylinux_2_28_x86_64.whl", hash = "sha256:229e2c79c00e85989a34b5981a2b67aa079fd08c903f0aaead522a1d68d79e51"},
+ {file = "Pillow-9.5.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:9adf58f5d64e474bed00d69bcd86ec4bcaa4123bfa70a65ce72e424bfb88ed96"},
+ {file = "Pillow-9.5.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:662da1f3f89a302cc22faa9f14a262c2e3951f9dbc9617609a47521c69dd9f8f"},
+ {file = "Pillow-9.5.0-cp38-cp38-win32.whl", hash = "sha256:6608ff3bf781eee0cd14d0901a2b9cc3d3834516532e3bd673a0a204dc8615fc"},
+ {file = "Pillow-9.5.0-cp38-cp38-win_amd64.whl", hash = "sha256:e49eb4e95ff6fd7c0c402508894b1ef0e01b99a44320ba7d8ecbabefddcc5569"},
+ {file = "Pillow-9.5.0-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:482877592e927fd263028c105b36272398e3e1be3269efda09f6ba21fd83ec66"},
+ {file = "Pillow-9.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3ded42b9ad70e5f1754fb7c2e2d6465a9c842e41d178f262e08b8c85ed8a1d8e"},
+ {file = "Pillow-9.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c446d2245ba29820d405315083d55299a796695d747efceb5717a8b450324115"},
+ {file = "Pillow-9.5.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8aca1152d93dcc27dc55395604dcfc55bed5f25ef4c98716a928bacba90d33a3"},
+ {file = "Pillow-9.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:608488bdcbdb4ba7837461442b90ea6f3079397ddc968c31265c1e056964f1ef"},
+ {file = "Pillow-9.5.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:60037a8db8750e474af7ffc9faa9b5859e6c6d0a50e55c45576bf28be7419705"},
+ {file = "Pillow-9.5.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:07999f5834bdc404c442146942a2ecadd1cb6292f5229f4ed3b31e0a108746b1"},
+ {file = "Pillow-9.5.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:a127ae76092974abfbfa38ca2d12cbeddcdeac0fb71f9627cc1135bedaf9d51a"},
+ {file = "Pillow-9.5.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:489f8389261e5ed43ac8ff7b453162af39c3e8abd730af8363587ba64bb2e865"},
+ {file = "Pillow-9.5.0-cp39-cp39-win32.whl", hash = "sha256:9b1af95c3a967bf1da94f253e56b6286b50af23392a886720f563c547e48e964"},
+ {file = "Pillow-9.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:77165c4a5e7d5a284f10a6efaa39a0ae8ba839da344f20b111d62cc932fa4e5d"},
+ {file = "Pillow-9.5.0-pp38-pypy38_pp73-macosx_10_10_x86_64.whl", hash = "sha256:833b86a98e0ede388fa29363159c9b1a294b0905b5128baf01db683672f230f5"},
+ {file = "Pillow-9.5.0-pp38-pypy38_pp73-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:aaf305d6d40bd9632198c766fb64f0c1a83ca5b667f16c1e79e1661ab5060140"},
+ {file = "Pillow-9.5.0-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0852ddb76d85f127c135b6dd1f0bb88dbb9ee990d2cd9aa9e28526c93e794fba"},
+ {file = "Pillow-9.5.0-pp38-pypy38_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:91ec6fe47b5eb5a9968c79ad9ed78c342b1f97a091677ba0e012701add857829"},
+ {file = "Pillow-9.5.0-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:cb841572862f629b99725ebaec3287fc6d275be9b14443ea746c1dd325053cbd"},
+ {file = "Pillow-9.5.0-pp39-pypy39_pp73-macosx_10_10_x86_64.whl", hash = "sha256:c380b27d041209b849ed246b111b7c166ba36d7933ec6e41175fd15ab9eb1572"},
+ {file = "Pillow-9.5.0-pp39-pypy39_pp73-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:7c9af5a3b406a50e313467e3565fc99929717f780164fe6fbb7704edba0cebbe"},
+ {file = "Pillow-9.5.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5671583eab84af046a397d6d0ba25343c00cd50bce03787948e0fff01d4fd9b1"},
+ {file = "Pillow-9.5.0-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:84a6f19ce086c1bf894644b43cd129702f781ba5751ca8572f08aa40ef0ab7b7"},
+ {file = "Pillow-9.5.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:1e7723bd90ef94eda669a3c2c19d549874dd5badaeefabefd26053304abe5799"},
+ {file = "Pillow-9.5.0.tar.gz", hash = "sha256:bf548479d336726d7a0eceb6e767e179fbde37833ae42794602631a070d630f1"},
]
[package.extras]
@@ -7377,7 +7368,6 @@ files = [
{file = "PyYAML-6.0.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:69b023b2b4daa7548bcfbd4aa3da05b3a74b772db9e23b982788168117739938"},
{file = "PyYAML-6.0.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:81e0b275a9ecc9c0c0c07b4b90ba548307583c125f54d5b6946cfee6360c733d"},
{file = "PyYAML-6.0.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ba336e390cd8e4d1739f42dfe9bb83a3cc2e80f567d8805e11b46f4a943f5515"},
- {file = "PyYAML-6.0.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:326c013efe8048858a6d312ddd31d56e468118ad4cdeda36c719bf5bb6192290"},
{file = "PyYAML-6.0.1-cp310-cp310-win32.whl", hash = "sha256:bd4af7373a854424dabd882decdc5579653d7868b8fb26dc7d0e99f823aa5924"},
{file = "PyYAML-6.0.1-cp310-cp310-win_amd64.whl", hash = "sha256:fd1592b3fdf65fff2ad0004b5e363300ef59ced41c2e6b3a99d4089fa8c5435d"},
{file = "PyYAML-6.0.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:6965a7bc3cf88e5a1c3bd2e0b5c22f8d677dc88a455344035f03399034eb3007"},
@@ -7385,15 +7375,8 @@ files = [
{file = "PyYAML-6.0.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:42f8152b8dbc4fe7d96729ec2b99c7097d656dc1213a3229ca5383f973a5ed6d"},
{file = "PyYAML-6.0.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:062582fca9fabdd2c8b54a3ef1c978d786e0f6b3a1510e0ac93ef59e0ddae2bc"},
{file = "PyYAML-6.0.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d2b04aac4d386b172d5b9692e2d2da8de7bfb6c387fa4f801fbf6fb2e6ba4673"},
- {file = "PyYAML-6.0.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:e7d73685e87afe9f3b36c799222440d6cf362062f78be1013661b00c5c6f678b"},
{file = "PyYAML-6.0.1-cp311-cp311-win32.whl", hash = "sha256:1635fd110e8d85d55237ab316b5b011de701ea0f29d07611174a1b42f1444741"},
{file = "PyYAML-6.0.1-cp311-cp311-win_amd64.whl", hash = "sha256:bf07ee2fef7014951eeb99f56f39c9bb4af143d8aa3c21b1677805985307da34"},
- {file = "PyYAML-6.0.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:855fb52b0dc35af121542a76b9a84f8d1cd886ea97c84703eaa6d88e37a2ad28"},
- {file = "PyYAML-6.0.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:40df9b996c2b73138957fe23a16a4f0ba614f4c0efce1e9406a184b6d07fa3a9"},
- {file = "PyYAML-6.0.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6c22bec3fbe2524cde73d7ada88f6566758a8f7227bfbf93a408a9d86bcc12a0"},
- {file = "PyYAML-6.0.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:8d4e9c88387b0f5c7d5f281e55304de64cf7f9c0021a3525bd3b1c542da3b0e4"},
- {file = "PyYAML-6.0.1-cp312-cp312-win32.whl", hash = "sha256:d483d2cdf104e7c9fa60c544d92981f12ad66a457afae824d146093b8c294c54"},
- {file = "PyYAML-6.0.1-cp312-cp312-win_amd64.whl", hash = "sha256:0d3304d8c0adc42be59c5f8a4d9e3d7379e6955ad754aa9d6ab7a398b59dd1df"},
{file = "PyYAML-6.0.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:50550eb667afee136e9a77d6dc71ae76a44df8b3e51e41b77f6de2932bfe0f47"},
{file = "PyYAML-6.0.1-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1fe35611261b29bd1de0070f0b2f47cb6ff71fa6595c077e42bd0c419fa27b98"},
{file = "PyYAML-6.0.1-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:704219a11b772aea0d8ecd7058d0082713c3562b4e271b849ad7dc4a5c90c13c"},
@@ -7410,7 +7393,6 @@ files = [
{file = "PyYAML-6.0.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a0cd17c15d3bb3fa06978b4e8958dcdc6e0174ccea823003a106c7d4d7899ac5"},
{file = "PyYAML-6.0.1-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:28c119d996beec18c05208a8bd78cbe4007878c6dd15091efb73a30e90539696"},
{file = "PyYAML-6.0.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7e07cbde391ba96ab58e532ff4803f79c4129397514e1413a7dc761ccd755735"},
- {file = "PyYAML-6.0.1-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:49a183be227561de579b4a36efbb21b3eab9651dd81b1858589f796549873dd6"},
{file = "PyYAML-6.0.1-cp38-cp38-win32.whl", hash = "sha256:184c5108a2aca3c5b3d3bf9395d50893a7ab82a38004c8f61c258d4428e80206"},
{file = "PyYAML-6.0.1-cp38-cp38-win_amd64.whl", hash = "sha256:1e2722cc9fbb45d9b87631ac70924c11d3a401b2d7f410cc0e3bbf249f2dca62"},
{file = "PyYAML-6.0.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:9eb6caa9a297fc2c2fb8862bc5370d0303ddba53ba97e71f08023b6cd73d16a8"},
@@ -7418,7 +7400,6 @@ files = [
{file = "PyYAML-6.0.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5773183b6446b2c99bb77e77595dd486303b4faab2b086e7b17bc6bef28865f6"},
{file = "PyYAML-6.0.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b786eecbdf8499b9ca1d697215862083bd6d2a99965554781d0d8d1ad31e13a0"},
{file = "PyYAML-6.0.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc1bf2925a1ecd43da378f4db9e4f799775d6367bdb94671027b73b393a7c42c"},
- {file = "PyYAML-6.0.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:04ac92ad1925b2cff1db0cfebffb6ffc43457495c9b3c39d3fcae417d7125dc5"},
{file = "PyYAML-6.0.1-cp39-cp39-win32.whl", hash = "sha256:faca3bdcf85b2fc05d06ff3fbc1f83e1391b3e724afa3feba7d13eeab355484c"},
{file = "PyYAML-6.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:510c9deebc5c0225e8c96813043e62b680ba2f9c50a08d3724c7f28a747d1486"},
{file = "PyYAML-6.0.1.tar.gz", hash = "sha256:bfdf460b1736c775f2ba9f6a92bca30bc2095067b8a9d77876d1fad6cc3b4a43"},
@@ -8811,40 +8792,40 @@ tests = ["cython", "littleutils", "pygments", "pytest", "typeguard"]
[[package]]
name = "streamlit"
-version = "1.22.0"
+version = "1.26.0"
description = "A faster way to build and share data apps"
optional = true
-python-versions = ">=3.7, !=3.9.7"
-files = [
- {file = "streamlit-1.22.0-py2.py3-none-any.whl", hash = "sha256:520dd9b9e6efb559b5a9a22feadb48b1e6f0340ec83da3514810059fdecd4167"},
- {file = "streamlit-1.22.0.tar.gz", hash = "sha256:5bef9bf8deef32814d9565c9df48331e6357eb0b90dabc3ec4f53c44fb34fc73"},
-]
-
-[package.dependencies]
-altair = ">=3.2.0,<5"
-blinker = ">=1.0.0"
-cachetools = ">=4.0"
-click = ">=7.0"
-gitpython = "!=3.1.19"
-importlib-metadata = ">=1.4"
-numpy = "*"
-packaging = ">=14.1"
-pandas = ">=0.25,<3"
-pillow = ">=6.2.0"
-protobuf = ">=3.12,<4"
-pyarrow = ">=4.0"
-pydeck = ">=0.1.dev5"
-pympler = ">=0.9"
-python-dateutil = "*"
-requests = ">=2.4"
-rich = ">=10.11.0"
-tenacity = ">=8.0.0,<9"
-toml = "*"
-tornado = ">=6.0.3"
-typing-extensions = ">=3.10.0.0"
-tzlocal = ">=1.1"
-validators = ">=0.2"
-watchdog = {version = "*", markers = "platform_system != \"Darwin\""}
+python-versions = ">=3.8, !=3.9.7"
+files = [
+ {file = "streamlit-1.26.0-py2.py3-none-any.whl", hash = "sha256:2bfdac041816e2e1ba27f061d40112afe61e0d4e72d25f354b38ba81107b4cb3"},
+ {file = "streamlit-1.26.0.tar.gz", hash = "sha256:25475fb15a3cc9fb184945f3fc936f011998bd8386e0c892febe14c9625bf47a"},
+]
+
+[package.dependencies]
+altair = ">=4.0,<6"
+blinker = ">=1.0.0,<2"
+cachetools = ">=4.0,<6"
+click = ">=7.0,<9"
+gitpython = ">=3.0.7,<3.1.19 || >3.1.19,<4"
+importlib-metadata = ">=1.4,<7"
+numpy = ">=1.19.3,<2"
+packaging = ">=16.8,<24"
+pandas = ">=1.3.0,<3"
+pillow = ">=7.1.0,<10"
+protobuf = ">=3.20,<5"
+pyarrow = ">=6.0"
+pydeck = ">=0.8,<1"
+pympler = ">=0.9,<2"
+python-dateutil = ">=2.7.3,<3"
+requests = ">=2.18,<3"
+rich = ">=10.14.0,<14"
+tenacity = ">=8.1.0,<9"
+toml = ">=0.10.1,<2"
+tornado = ">=6.0.3,<7"
+typing-extensions = ">=4.1.0,<5"
+tzlocal = ">=1.1,<5"
+validators = ">=0.2,<1"
+watchdog = {version = ">=2.1.5", markers = "platform_system != \"Darwin\""}
[package.extras]
snowflake = ["snowflake-snowpark-python"]
@@ -10494,15 +10475,15 @@ cffi = {version = ">=1.11", markers = "platform_python_implementation == \"PyPy\
cffi = ["cffi (>=1.11)"]
[extras]
-all = ["O365", "aleph-alpha-client", "amadeus", "arxiv", "atlassian-python-api", "awadb", "azure-ai-formrecognizer", "azure-ai-vision", "azure-cognitiveservices-speech", "azure-cosmos", "azure-identity", "beautifulsoup4", "clarifai", "clickhouse-connect", "cohere", "deeplake", "docarray", "duckduckgo-search", "elasticsearch", "esprima", "faiss-cpu", "google-api-python-client", "google-auth", "google-search-results", "gptcache", "html2text", "huggingface_hub", "jinja2", "jq", "konko", "lancedb", "langkit", "lark", "libdeeplake", "librosa", "lxml", "manifest-ml", "marqo", "momento", "nebula3-python", "neo4j", "networkx", "nlpcloud", "nltk", "nomic", "openai", "openlm", "opensearch-py", "pdfminer-six", "pexpect", "pgvector", "pinecone-client", "pinecone-text", "psycopg2-binary", "pymongo", "pyowm", "pypdf", "pytesseract", "python-arango", "pyvespa", "qdrant-client", "rdflib", "redis", "requests-toolbelt", "sentence-transformers", "singlestoredb", "tensorflow-text", "tigrisdb", "tiktoken", "torch", "transformers", "weaviate-client", "wikipedia", "wolframalpha"]
+all = ["O365", "aleph-alpha-client", "amadeus", "arxiv", "atlassian-python-api", "awadb", "azure-ai-formrecognizer", "azure-ai-vision", "azure-cognitiveservices-speech", "azure-cosmos", "azure-identity", "beautifulsoup4", "clarifai", "clickhouse-connect", "cohere", "deeplake", "docarray", "duckduckgo-search", "elasticsearch", "esprima", "faiss-cpu", "google-api-python-client", "google-auth", "google-search-results", "gptcache", "html2text", "huggingface_hub", "jinja2", "jq", "lancedb", "langkit", "lark", "libdeeplake", "librosa", "lxml", "manifest-ml", "marqo", "momento", "nebula3-python", "neo4j", "networkx", "nlpcloud", "nltk", "nomic", "openai", "openlm", "opensearch-py", "pdfminer-six", "pexpect", "pgvector", "pinecone-client", "pinecone-text", "psycopg2-binary", "pymongo", "pyowm", "pypdf", "pytesseract", "python-arango", "pyvespa", "qdrant-client", "rdflib", "redis", "requests-toolbelt", "sentence-transformers", "singlestoredb", "tensorflow-text", "tigrisdb", "tiktoken", "torch", "transformers", "weaviate-client", "wikipedia", "wolframalpha"]
azure = ["azure-ai-formrecognizer", "azure-ai-vision", "azure-cognitiveservices-speech", "azure-core", "azure-cosmos", "azure-identity", "azure-search-documents", "openai"]
clarifai = ["clarifai"]
cohere = ["cohere"]
docarray = ["docarray"]
embeddings = ["sentence-transformers"]
-extended-testing = ["amazon-textract-caller", "assemblyai", "atlassian-python-api", "beautifulsoup4", "bibtexparser", "cassio", "chardet", "dashvector", "esprima", "faiss-cpu", "feedparser", "geopandas", "gitpython", "gql", "html2text", "jinja2", "jq", "konko", "lxml", "markdownify", "mwparserfromhell", "mwxml", "newspaper3k", "openai", "openai", "openapi-schema-pydantic", "pandas", "pdfminer-six", "pgvector", "psychicapi", "py-trello", "pymupdf", "pypdf", "pypdfium2", "pyspark", "rank-bm25", "rapidfuzz", "requests-toolbelt", "scikit-learn", "sqlite-vss", "streamlit", "sympy", "telethon", "tqdm", "xata", "xmltodict"]
+extended-testing = ["amazon-textract-caller", "assemblyai", "atlassian-python-api", "beautifulsoup4", "bibtexparser", "cassio", "chardet", "dashvector", "esprima", "faiss-cpu", "feedparser", "geopandas", "gitpython", "gql", "html2text", "jinja2", "jq", "lxml", "markdownify", "mwparserfromhell", "mwxml", "newspaper3k", "openai", "openai", "openapi-schema-pydantic", "pandas", "pdfminer-six", "pgvector", "psychicapi", "py-trello", "pymupdf", "pypdf", "pypdfium2", "pyspark", "rank-bm25", "rapidfuzz", "requests-toolbelt", "scikit-learn", "sqlite-vss", "streamlit", "sympy", "telethon", "tqdm", "xata", "xmltodict"]
javascript = ["esprima"]
-llms = ["clarifai", "cohere", "huggingface_hub", "konko", "manifest-ml", "nlpcloud", "openai", "openlm", "torch", "transformers"]
+llms = ["clarifai", "cohere", "huggingface_hub", "manifest-ml", "nlpcloud", "openai", "openlm", "torch", "transformers"]
openai = ["openai", "tiktoken"]
qdrant = ["qdrant-client"]
text-helpers = ["chardet"]
@@ -10510,4 +10491,4 @@ text-helpers = ["chardet"]
[metadata]
lock-version = "2.0"
python-versions = ">=3.8.1,<4.0"
-content-hash = "47412cb06a11e383815fbefec7855a841701ed3e385da048d3b72351bd195cdf"
+content-hash = "b63078268a80c07577b432114302f4f86d47be25b83a245affb0dbc999fb2c1f"
diff --git a/libs/langchain/pyproject.toml b/libs/langchain/pyproject.toml
index cd296c39d2..1851b018df 100644
--- a/libs/langchain/pyproject.toml
+++ b/libs/langchain/pyproject.toml
@@ -1,6 +1,6 @@
[tool.poetry]
name = "langchain"
-version = "0.0.282"
+version = "0.0.285"
description = "Building applications with LLMs through composability"
authors = []
license = "MIT"
@@ -129,7 +129,6 @@ markdownify = {version = "^0.11.6", optional = true}
assemblyai = {version = "^0.17.0", optional = true}
dashvector = {version = "^1.0.1", optional = true}
sqlite-vss = {version = "^0.1.2", optional = true}
-konko = {version = "^0.4.0", optional = true}
[tool.poetry.group.test.dependencies]
@@ -202,7 +201,7 @@ playwright = "^1.28.0"
setuptools = "^67.6.1"
[tool.poetry.extras]
-llms = ["clarifai", "cohere", "openai", "openlm", "nlpcloud", "huggingface_hub", "manifest-ml", "torch", "transformers", "konko"]
+llms = ["clarifai", "cohere", "openai", "openlm", "nlpcloud", "huggingface_hub", "manifest-ml", "torch", "transformers"]
qdrant = ["qdrant-client"]
openai = ["openai", "tiktoken"]
text_helpers = ["chardet"]
@@ -295,7 +294,6 @@ all = [
"amadeus",
"librosa",
"python-arango",
- "konko",
]
# An extra used to be able to add extended testing.
@@ -347,7 +345,6 @@ extended_testing = [
"markdownify",
"dashvector",
"sqlite-vss",
- "konko",
]
[tool.ruff]
diff --git a/libs/langchain/tests/integration_tests/chat_models/test_konko.py b/libs/langchain/tests/integration_tests/chat_models/test_konko.py
index 576f163498..c47bbbb3f0 100644
--- a/libs/langchain/tests/integration_tests/chat_models/test_konko.py
+++ b/libs/langchain/tests/integration_tests/chat_models/test_konko.py
@@ -14,7 +14,6 @@ from langchain.schema.messages import BaseMessage, HumanMessage, SystemMessage
from tests.unit_tests.callbacks.fake_callback_handler import FakeCallbackHandler
-@pytest.mark.scheduled
def test_konko_chat_test() -> None:
"""Evaluate basic ChatKonko functionality."""
chat_instance = ChatKonko(max_tokens=10)
@@ -58,7 +57,6 @@ def test_konko_system_msg_test() -> None:
assert isinstance(chat_response.content, str)
-@pytest.mark.scheduled
def test_konko_generation_test() -> None:
"""Check ChatKonko's generation ability."""
chat_instance = ChatKonko(max_tokens=10, n=2)
@@ -74,7 +72,6 @@ def test_konko_generation_test() -> None:
assert gen.text == gen.message.content
-@pytest.mark.scheduled
def test_konko_multiple_outputs_test() -> None:
"""Test multiple completions with ChatKonko."""
chat_instance = ChatKonko(max_tokens=10, n=5)
@@ -87,7 +84,6 @@ def test_konko_multiple_outputs_test() -> None:
assert isinstance(gen.message.content, str)
-@pytest.mark.scheduled
def test_konko_streaming_callback_test() -> None:
"""Evaluate streaming's token callback functionality."""
callback_instance = FakeCallbackHandler()
@@ -105,7 +101,6 @@ def test_konko_streaming_callback_test() -> None:
assert isinstance(chat_response, BaseMessage)
-@pytest.mark.scheduled
def test_konko_streaming_info_test() -> None:
"""Ensure generation details are retained during streaming."""
@@ -175,7 +170,6 @@ def test_konko_additional_args_test() -> None:
ChatKonko(model_kwargs={"model": "text-davinci-003"})
-@pytest.mark.scheduled
def test_konko_token_streaming_test() -> None:
"""Check token streaming for ChatKonko."""
chat_instance = ChatKonko(max_tokens=10)
diff --git a/libs/langchain/tests/integration_tests/document_loaders/test_polars_dataframe.py b/libs/langchain/tests/integration_tests/document_loaders/test_polars_dataframe.py
index 03f5070120..2858b41e8e 100644
--- a/libs/langchain/tests/integration_tests/document_loaders/test_polars_dataframe.py
+++ b/libs/langchain/tests/integration_tests/document_loaders/test_polars_dataframe.py
@@ -34,8 +34,6 @@ def test_load_returns_list_of_documents(sample_data_frame: pl.DataFrame) -> None
def test_load_converts_dataframe_columns_to_document_metadata(
sample_data_frame: pl.DataFrame,
) -> None:
- import polars as pl
-
loader = PolarsDataFrameLoader(sample_data_frame)
docs = loader.load()
diff --git a/libs/langchain/tests/integration_tests/document_loaders/test_url_playwright.py b/libs/langchain/tests/integration_tests/document_loaders/test_url_playwright.py
index 7bea1c6dee..eb53682d75 100644
--- a/libs/langchain/tests/integration_tests/document_loaders/test_url_playwright.py
+++ b/libs/langchain/tests/integration_tests/document_loaders/test_url_playwright.py
@@ -7,7 +7,9 @@ from langchain.document_loaders import PlaywrightURLLoader
from langchain.document_loaders.url_playwright import PlaywrightEvaluator
if TYPE_CHECKING:
- from playwright.async_api import AsyncBrowser, AsyncPage, AsyncResponse
+ from playwright.async_api import Browser as AsyncBrowser
+ from playwright.async_api import Page as AsyncPage
+ from playwright.async_api import Response as AsyncResponse
from playwright.sync_api import Browser, Page, Response
diff --git a/libs/langchain/tests/integration_tests/smith/evaluation/test_runner_utils.py b/libs/langchain/tests/integration_tests/smith/evaluation/test_runner_utils.py
index 9696515a9c..4db2e88c97 100644
--- a/libs/langchain/tests/integration_tests/smith/evaluation/test_runner_utils.py
+++ b/libs/langchain/tests/integration_tests/smith/evaluation/test_runner_utils.py
@@ -5,7 +5,6 @@ import pytest
from langsmith import Client as Client
from langsmith.schemas import DataType
-from langchain.callbacks.tracers.evaluation import wait_for_all_evaluators
from langchain.chains.llm import LLMChain
from langchain.chat_models import ChatOpenAI
from langchain.evaluation import EvaluatorType
@@ -22,7 +21,6 @@ def _check_all_feedback_passed(_project_name: str, client: Client) -> None:
# chain or llm passes for the feedback provided.
runs = list(client.list_runs(project_name=_project_name, execution_order=1))
assert len(runs) == 4
- wait_for_all_evaluators()
feedback = list(client.list_feedback(run_ids=[run.id for run in runs]))
assert len(feedback) == 8
assert all([f.score == 1 for f in feedback])
diff --git a/libs/langchain/tests/integration_tests/vectorstores/test_nucliadb.py b/libs/langchain/tests/integration_tests/vectorstores/test_nucliadb.py
new file mode 100644
index 0000000000..1cfeea0da4
--- /dev/null
+++ b/libs/langchain/tests/integration_tests/vectorstores/test_nucliadb.py
@@ -0,0 +1,98 @@
+from typing import Any
+from unittest import mock
+
+from langchain.vectorstores.nucliadb import NucliaDB
+
+
+class attrdict(dict):
+ def __getitem__(self, key: str) -> Any:
+ value = dict.__getitem__(self, key)
+ return attrdict(value) if isinstance(value, dict) else value
+
+ __getattr__ = __getitem__
+
+
+def FakeCreate(**args: Any) -> Any:
+ def fn(self: Any, **kwargs: Any) -> str:
+ return "fake_uuid"
+
+ return fn
+
+
+def FakeDelete(**args: Any) -> Any:
+ def fn(self: Any, **kwargs: Any) -> None:
+ return None
+
+ return fn
+
+
+def FakeFind(**args: Any) -> Any:
+ def fn(self: Any, **kwargs: Any) -> Any:
+ return attrdict(
+ {
+ "resources": {
+ "123": attrdict(
+ {
+ "fields": {
+ "456": attrdict(
+ {
+ "paragraphs": {
+ "123/t/text/0-14": attrdict(
+ {
+ "text": "This is a test",
+ "order": 0,
+ }
+ ),
+ }
+ }
+ )
+ },
+ "data": {
+ "texts": {
+ "text": {
+ "body": "This is a test",
+ }
+ }
+ },
+ "extra": attrdict({"metadata": {"some": "metadata"}}),
+ }
+ )
+ }
+ }
+ )
+
+ return fn
+
+
+def test_add_texts() -> None:
+ with mock.patch(
+ "nuclia.sdk.resource.NucliaResource.create",
+ new_callable=FakeCreate,
+ ):
+ ndb = NucliaDB(knowledge_box="YOUR_KB_ID", local=False, api_key="YOUR_API_KEY")
+ assert ndb.is_local is False
+ ids = ndb.add_texts(["This is a new test", "This is a second test"])
+ assert len(ids) == 2
+
+
+def test_delete() -> None:
+ with mock.patch(
+ "nuclia.sdk.resource.NucliaResource.delete",
+ new_callable=FakeDelete,
+ ):
+ ndb = NucliaDB(knowledge_box="YOUR_KB_ID", local=False, api_key="YOUR_API_KEY")
+ success = ndb.delete(["123", "456"])
+ assert success
+
+
+def test_search() -> None:
+ with mock.patch(
+ "nuclia.sdk.search.NucliaSearch.find",
+ new_callable=FakeFind,
+ ):
+ ndb = NucliaDB(knowledge_box="YOUR_KB_ID", local=False, api_key="YOUR_API_KEY")
+ results = ndb.similarity_search("Who was inspired by Ada Lovelace?")
+ assert len(results) == 1
+ assert results[0].page_content == "This is a test"
+ assert results[0].metadata["extra"]["some"] == "metadata"
+ assert results[0].metadata["value"]["body"] == "This is a test"
diff --git a/libs/langchain/tests/unit_tests/retrievers/self_query/test_supabase.py b/libs/langchain/tests/unit_tests/retrievers/self_query/test_supabase.py
new file mode 100644
index 0000000000..de9b04fabf
--- /dev/null
+++ b/libs/langchain/tests/unit_tests/retrievers/self_query/test_supabase.py
@@ -0,0 +1,85 @@
+from typing import Dict, Tuple
+
+from langchain.chains.query_constructor.ir import (
+ Comparator,
+ Comparison,
+ Operation,
+ Operator,
+ StructuredQuery,
+)
+from langchain.retrievers.self_query.supabase import SupabaseVectorTranslator
+
+DEFAULT_TRANSLATOR = SupabaseVectorTranslator()
+
+
+def test_visit_comparison() -> None:
+ comp = Comparison(comparator=Comparator.LT, attribute="foo", value=["1", "2"])
+ expected = "and(metadata->>foo.lt.1,metadata->>foo.lt.2)"
+ actual = DEFAULT_TRANSLATOR.visit_comparison(comp)
+ assert expected == actual
+
+
+def test_visit_operation() -> None:
+ op = Operation(
+ operator=Operator.AND,
+ arguments=[
+ Comparison(comparator=Comparator.LT, attribute="foo", value=2),
+ Comparison(comparator=Comparator.EQ, attribute="bar", value="baz"),
+ Comparison(comparator=Comparator.LT, attribute="abc", value=["1", "2"]),
+ ],
+ )
+ expected = (
+ "and("
+ "metadata->foo.lt.2,"
+ "metadata->>bar.eq.baz,"
+ "and(metadata->>abc.lt.1,metadata->>abc.lt.2)"
+ ")"
+ )
+ actual = DEFAULT_TRANSLATOR.visit_operation(op)
+ assert expected == actual
+
+
+def test_visit_structured_query() -> None:
+ query = "What is the capital of France?"
+ structured_query = StructuredQuery(
+ query=query,
+ filter=None,
+ )
+ expected: Tuple[str, Dict] = (query, {})
+ actual = DEFAULT_TRANSLATOR.visit_structured_query(structured_query)
+ assert expected == actual
+
+ comp = Comparison(comparator=Comparator.LT, attribute="foo", value=["1", "2"])
+ expected = (
+ query,
+ {"postgrest_filter": "and(metadata->>foo.lt.1,metadata->>foo.lt.2)"},
+ )
+ structured_query = StructuredQuery(
+ query=query,
+ filter=comp,
+ )
+ actual = DEFAULT_TRANSLATOR.visit_structured_query(structured_query)
+ assert expected == actual
+
+ op = Operation(
+ operator=Operator.AND,
+ arguments=[
+ Comparison(comparator=Comparator.LT, attribute="foo", value=2),
+ Comparison(comparator=Comparator.EQ, attribute="bar", value="baz"),
+ Comparison(comparator=Comparator.LT, attribute="abc", value=["1", "2"]),
+ ],
+ )
+ structured_query = StructuredQuery(
+ query=query,
+ filter=op,
+ )
+ expected = (
+ query,
+ {
+ "postgrest_filter": (
+ "and(metadata->foo.lt.2,metadata->>bar.eq.baz,and(metadata->>abc.lt.1,metadata->>abc.lt.2))"
+ )
+ },
+ )
+ actual = DEFAULT_TRANSLATOR.visit_structured_query(structured_query)
+ assert expected == actual
diff --git a/libs/langchain/tests/unit_tests/retrievers/self_query/test_vectara.py b/libs/langchain/tests/unit_tests/retrievers/self_query/test_vectara.py
new file mode 100644
index 0000000000..05c15f26ac
--- /dev/null
+++ b/libs/langchain/tests/unit_tests/retrievers/self_query/test_vectara.py
@@ -0,0 +1,71 @@
+from typing import Dict, Tuple
+
+from langchain.chains.query_constructor.ir import (
+ Comparator,
+ Comparison,
+ Operation,
+ Operator,
+ StructuredQuery,
+)
+from langchain.retrievers.self_query.vectara import VectaraTranslator
+
+DEFAULT_TRANSLATOR = VectaraTranslator()
+
+
+def test_visit_comparison() -> None:
+ comp = Comparison(comparator=Comparator.LT, attribute="foo", value="1")
+ expected = "( doc.foo < '1' )"
+ actual = DEFAULT_TRANSLATOR.visit_comparison(comp)
+ assert expected == actual
+
+
+def test_visit_operation() -> None:
+ op = Operation(
+ operator=Operator.AND,
+ arguments=[
+ Comparison(comparator=Comparator.LT, attribute="foo", value=2),
+ Comparison(comparator=Comparator.EQ, attribute="bar", value="baz"),
+ Comparison(comparator=Comparator.LT, attribute="abc", value=1),
+ ],
+ )
+ expected = "( ( doc.foo < 2 ) and ( doc.bar = 'baz' ) and ( doc.abc < 1 ) )"
+ actual = DEFAULT_TRANSLATOR.visit_operation(op)
+ assert expected == actual
+
+
+def test_visit_structured_query() -> None:
+ query = "What is the capital of France?"
+ structured_query = StructuredQuery(
+ query=query,
+ filter=None,
+ limit=None,
+ )
+ expected: Tuple[str, Dict] = (query, {})
+ actual = DEFAULT_TRANSLATOR.visit_structured_query(structured_query)
+ assert expected == actual
+
+ comp = Comparison(comparator=Comparator.LT, attribute="foo", value=1)
+ expected = (query, {"filter": "( doc.foo < 1 )"})
+ structured_query = StructuredQuery(
+ query=query,
+ filter=comp,
+ limit=None,
+ )
+ actual = DEFAULT_TRANSLATOR.visit_structured_query(structured_query)
+ assert expected == actual
+
+ op = Operation(
+ operator=Operator.AND,
+ arguments=[
+ Comparison(comparator=Comparator.LT, attribute="foo", value=2),
+ Comparison(comparator=Comparator.EQ, attribute="bar", value="baz"),
+ Comparison(comparator=Comparator.LT, attribute="abc", value=1),
+ ],
+ )
+ structured_query = StructuredQuery(query=query, filter=op, limit=None)
+ expected = (
+ query,
+ {"filter": "( ( doc.foo < 2 ) and ( doc.bar = 'baz' ) and ( doc.abc < 1 ) )"},
+ )
+ actual = DEFAULT_TRANSLATOR.visit_structured_query(structured_query)
+ assert expected == actual
diff --git a/libs/langchain/tests/unit_tests/schema/runnable/__snapshots__/test_runnable.ambr b/libs/langchain/tests/unit_tests/schema/runnable/__snapshots__/test_runnable.ambr
index 5ea21b13d6..63c0acc38d 100644
--- a/libs/langchain/tests/unit_tests/schema/runnable/__snapshots__/test_runnable.ambr
+++ b/libs/langchain/tests/unit_tests/schema/runnable/__snapshots__/test_runnable.ambr
@@ -467,7 +467,7 @@
# ---
# name: test_combining_sequences.3
list([
- Run(id=UUID('00000000-0000-4000-8000-000000000000'), name='RunnableSequence', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'runnable', 'RunnableSequence'], 'kwargs': {'first': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, 'middle': [{'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo, bar'], sleep=None, i=0)"}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}, {'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nicer assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, {'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['baz, qux'], sleep=None, i=0)"}], 'last': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'output': ['baz', 'qux']}, reference_example_id=None, parent_run_id=None, tags=[], execution_order=None, child_execution_order=None, child_runs=[Run(id=UUID('00000000-0000-4000-8000-000000000001'), name='ChatPromptTemplate', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='prompt', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptValue'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'SystemMessage'], 'kwargs': {'content': 'You are a nice assistant.', 'additional_kwargs': {}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'HumanMessage'], 'kwargs': {'content': 'What is your name?', 'additional_kwargs': {}}}]}}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:1'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000002'), name='FakeListChatModel', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='llm', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={'invocation_params': {'responses': ['foo, bar'], '_type': 'fake-list-chat-model', 'stop': None}, 'options': {'stop': None}}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo, bar'], sleep=None, i=0)"}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'prompts': ['System: You are a nice assistant.\nHuman: What is your name?']}, outputs={'generations': [[{'text': 'foo, bar', 'generation_info': None, 'message': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'AIMessage'], 'kwargs': {'content': 'foo, bar'}}}]], 'llm_output': None, 'run': None}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:2'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000003'), name='CommaSeparatedListOutputParser', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='parser', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': AIMessage(content='foo, bar', additional_kwargs={}, example=False)}, outputs={'output': ['foo', 'bar']}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:3'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000004'), name='RunnableLambda', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': ['foo', 'bar']}, outputs={'question': 'foobar'}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:4'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000005'), name='ChatPromptTemplate', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='prompt', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nicer assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'foobar'}, outputs={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptValue'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'SystemMessage'], 'kwargs': {'content': 'You are a nicer assistant.', 'additional_kwargs': {}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'HumanMessage'], 'kwargs': {'content': 'foobar', 'additional_kwargs': {}}}]}}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:5'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000006'), name='FakeListChatModel', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='llm', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={'invocation_params': {'responses': ['baz, qux'], '_type': 'fake-list-chat-model', 'stop': None}, 'options': {'stop': None}}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['baz, qux'], sleep=None, i=0)"}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'prompts': ['System: You are a nicer assistant.\nHuman: foobar']}, outputs={'generations': [[{'text': 'baz, qux', 'generation_info': None, 'message': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'AIMessage'], 'kwargs': {'content': 'baz, qux'}}}]], 'llm_output': None, 'run': None}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:6'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000007'), name='CommaSeparatedListOutputParser', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='parser', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': AIMessage(content='baz, qux', additional_kwargs={}, example=False)}, outputs={'output': ['baz', 'qux']}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:7'], execution_order=None, child_execution_order=None, child_runs=[])]),
+ Run(id=UUID('00000000-0000-4000-8000-000000000000'), name='RunnableSequence', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'runnable', 'RunnableSequence'], 'kwargs': {'first': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, 'middle': [{'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo, bar'], sleep=None, i=0)"}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}, {'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nicer assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, {'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['baz, qux'], sleep=None, i=0)"}], 'last': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'output': ['baz', 'qux']}, reference_example_id=None, parent_run_id=None, tags=[], execution_order=None, child_execution_order=None, child_runs=[Run(id=UUID('00000000-0000-4000-8000-000000000001'), name='ChatPromptTemplate', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='prompt', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptValue'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'SystemMessage'], 'kwargs': {'content': 'You are a nice assistant.', 'additional_kwargs': {}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'HumanMessage'], 'kwargs': {'content': 'What is your name?', 'additional_kwargs': {}}}]}}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:1'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000002'), name='FakeListChatModel', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='llm', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={'invocation_params': {'responses': ['foo, bar'], '_type': 'fake-list-chat-model', 'stop': None}, 'options': {'stop': None}}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo, bar'], sleep=None, i=0)"}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'prompts': ['System: You are a nice assistant.\nHuman: What is your name?']}, outputs={'generations': [[{'text': 'foo, bar', 'generation_info': None, 'message': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'AIMessage'], 'kwargs': {'content': 'foo, bar'}}}]], 'llm_output': None, 'run': None}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:2'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000003'), name='CommaSeparatedListOutputParser', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='parser', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': AIMessage(content='foo, bar', additional_kwargs={}, example=False)}, outputs={'output': ['foo', 'bar']}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:3'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000004'), name='<lambda>', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': ['foo', 'bar']}, outputs={'question': 'foobar'}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:4'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000005'), name='ChatPromptTemplate', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='prompt', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nicer assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'foobar'}, outputs={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptValue'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'SystemMessage'], 'kwargs': {'content': 'You are a nicer assistant.', 'additional_kwargs': {}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'HumanMessage'], 'kwargs': {'content': 'foobar', 'additional_kwargs': {}}}]}}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:5'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000006'), name='FakeListChatModel', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='llm', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={'invocation_params': {'responses': ['baz, qux'], '_type': 'fake-list-chat-model', 'stop': None}, 'options': {'stop': None}}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'chat_models', 'fake', 'FakeListChatModel'], 'repr': "FakeListChatModel(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['baz, qux'], sleep=None, i=0)"}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'prompts': ['System: You are a nicer assistant.\nHuman: foobar']}, outputs={'generations': [[{'text': 'baz, qux', 'generation_info': None, 'message': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'AIMessage'], 'kwargs': {'content': 'baz, qux'}}}]], 'llm_output': None, 'run': None}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:6'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000007'), name='CommaSeparatedListOutputParser', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='parser', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'output_parsers', 'list', 'CommaSeparatedListOutputParser'], 'kwargs': {}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': AIMessage(content='baz, qux', additional_kwargs={}, example=False)}, outputs={'output': ['baz', 'qux']}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:7'], execution_order=None, child_execution_order=None, child_runs=[])]),
])
# ---
# name: test_each
@@ -1407,7 +1407,7 @@
# ---
# name: test_prompt_with_llm_and_async_lambda.1
list([
- Run(id=UUID('00000000-0000-4000-8000-000000000000'), name='RunnableSequence', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'runnable', 'RunnableSequence'], 'kwargs': {'first': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, 'middle': [{'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'llms', 'fake', 'FakeListLLM'], 'repr': "FakeListLLM(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo', 'bar'], sleep=None, i=0)"}], 'last': {'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'output': 'foo'}, reference_example_id=None, parent_run_id=None, tags=[], execution_order=None, child_execution_order=None, child_runs=[Run(id=UUID('00000000-0000-4000-8000-000000000001'), name='ChatPromptTemplate', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='prompt', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptValue'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'SystemMessage'], 'kwargs': {'content': 'You are a nice assistant.', 'additional_kwargs': {}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'HumanMessage'], 'kwargs': {'content': 'What is your name?', 'additional_kwargs': {}}}]}}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:1'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000002'), name='FakeListLLM', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='llm', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={'invocation_params': {'responses': ['foo', 'bar'], '_type': 'fake-list', 'stop': None}, 'options': {'stop': None}}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'llms', 'fake', 'FakeListLLM'], 'repr': "FakeListLLM(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo', 'bar'], sleep=None, i=0)"}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'prompts': ['System: You are a nice assistant.\nHuman: What is your name?']}, outputs={'generations': [[{'text': 'foo', 'generation_info': None}]], 'llm_output': None, 'run': None}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:2'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000003'), name='RunnableLambda', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': 'foo'}, outputs={'output': 'foo'}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:3'], execution_order=None, child_execution_order=None, child_runs=[])]),
+ Run(id=UUID('00000000-0000-4000-8000-000000000000'), name='RunnableSequence', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'runnable', 'RunnableSequence'], 'kwargs': {'first': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, 'middle': [{'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'llms', 'fake', 'FakeListLLM'], 'repr': "FakeListLLM(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo', 'bar'], sleep=None, i=0)"}], 'last': {'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'output': 'foo'}, reference_example_id=None, parent_run_id=None, tags=[], execution_order=None, child_execution_order=None, child_runs=[Run(id=UUID('00000000-0000-4000-8000-000000000001'), name='ChatPromptTemplate', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='prompt', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptTemplate'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'SystemMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': [], 'template': 'You are a nice assistant.', 'template_format': 'f-string', 'partial_variables': {}}}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'HumanMessagePromptTemplate'], 'kwargs': {'prompt': {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'prompt', 'PromptTemplate'], 'kwargs': {'input_variables': ['question'], 'template': '{question}', 'template_format': 'f-string', 'partial_variables': {}}}}}], 'input_variables': ['question']}}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'question': 'What is your name?'}, outputs={'lc': 1, 'type': 'constructor', 'id': ['langchain', 'prompts', 'chat', 'ChatPromptValue'], 'kwargs': {'messages': [{'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'SystemMessage'], 'kwargs': {'content': 'You are a nice assistant.', 'additional_kwargs': {}}}, {'lc': 1, 'type': 'constructor', 'id': ['langchain', 'schema', 'messages', 'HumanMessage'], 'kwargs': {'content': 'What is your name?', 'additional_kwargs': {}}}]}}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:1'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000002'), name='FakeListLLM', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='llm', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={'invocation_params': {'responses': ['foo', 'bar'], '_type': 'fake-list', 'stop': None}, 'options': {'stop': None}}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'llms', 'fake', 'FakeListLLM'], 'repr': "FakeListLLM(cache=None, verbose=False, callbacks=None, callback_manager=None, tags=None, metadata=None, responses=['foo', 'bar'], sleep=None, i=0)"}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'prompts': ['System: You are a nice assistant.\nHuman: What is your name?']}, outputs={'generations': [[{'text': 'foo', 'generation_info': None}]], 'llm_output': None, 'run': None}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:2'], execution_order=None, child_execution_order=None, child_runs=[]), Run(id=UUID('00000000-0000-4000-8000-000000000003'), name='passthrough', start_time=FakeDatetime(2023, 1, 1, 0, 0), run_type='chain', end_time=FakeDatetime(2023, 1, 1, 0, 0), extra={}, error=None, serialized={'lc': 1, 'type': 'not_implemented', 'id': ['langchain', 'schema', 'runnable', 'base', 'RunnableLambda'], 'repr': 'RunnableLambda(...)'}, events=[{'name': 'start', 'time': FakeDatetime(2023, 1, 1, 0, 0)}, {'name': 'end', 'time': FakeDatetime(2023, 1, 1, 0, 0)}], inputs={'input': 'foo'}, outputs={'output': 'foo'}, reference_example_id=None, parent_run_id=UUID('00000000-0000-4000-8000-000000000000'), tags=['seq:step:3'], execution_order=None, child_execution_order=None, child_runs=[])]),
])
# ---
# name: test_router_runnable
diff --git a/libs/langchain/tests/unit_tests/schema/runnable/test_runnable.py b/libs/langchain/tests/unit_tests/schema/runnable/test_runnable.py
index 2e0be35ddc..98bf284fd0 100644
--- a/libs/langchain/tests/unit_tests/schema/runnable/test_runnable.py
+++ b/libs/langchain/tests/unit_tests/schema/runnable/test_runnable.py
@@ -948,7 +948,7 @@ async def test_higher_order_lambda_runnable(
parent_run = next(r for r in tracer.runs if r.parent_run_id is None)
assert len(parent_run.child_runs) == 2
router_run = parent_run.child_runs[1]
- assert router_run.name == "RunnableLambda"
+ assert router_run.name == "router"
assert len(router_run.child_runs) == 1
math_run = router_run.child_runs[0]
assert math_run.name == "RunnableSequence"
@@ -980,7 +980,7 @@ async def test_higher_order_lambda_runnable(
parent_run = next(r for r in tracer.runs if r.parent_run_id is None)
assert len(parent_run.child_runs) == 2
router_run = parent_run.child_runs[1]
- assert router_run.name == "RunnableLambda"
+ assert router_run.name == "arouter"
assert len(router_run.child_runs) == 1
math_run = router_run.child_runs[0]
assert math_run.name == "RunnableSequence"
@@ -1507,7 +1507,7 @@ async def test_async_retrying(mocker: MockerFixture) -> None:
with pytest.raises(ValueError):
await runnable.with_retry(
stop_after_attempt=2,
- retry_if_exception_type=(ValueError,),
+ retry_if_exception_type=(ValueError, KeyError),
).ainvoke(1)
assert _lambda_mock.call_count == 2 # retried
diff --git a/libs/langchain/tests/unit_tests/smith/evaluation/test_runner_utils.py b/libs/langchain/tests/unit_tests/smith/evaluation/test_runner_utils.py
index 914958031d..825c61e74e 100644
--- a/libs/langchain/tests/unit_tests/smith/evaluation/test_runner_utils.py
+++ b/libs/langchain/tests/unit_tests/smith/evaluation/test_runner_utils.py
@@ -181,11 +181,15 @@ def test_run_llm_or_chain_with_input_mapper() -> None:
assert "the wrong input" in inputs
return {"the right input": inputs["the wrong input"]}
- result = _run_llm_or_chain(example, lambda: mock_chain, input_mapper=input_mapper)
+ result = _run_llm_or_chain(
+ example,
+ {"callbacks": [], "tags": []},
+ llm_or_chain_factory=lambda: mock_chain,
+ input_mapper=input_mapper,
+ )
assert result == {"output": "2", "the right input": "1"}
bad_result = _run_llm_or_chain(
- example,
- lambda: mock_chain,
+ example, {"callbacks": [], "tags": []}, llm_or_chain_factory=lambda: mock_chain
)
assert "Error" in bad_result
@@ -195,7 +199,12 @@ def test_run_llm_or_chain_with_input_mapper() -> None:
return "the right input"
mock_llm = FakeLLM(queries={"the right input": "somenumber"})
- llm_result = _run_llm_or_chain(example, mock_llm, input_mapper=llm_input_mapper)
+ llm_result = _run_llm_or_chain(
+ example,
+ {"callbacks": [], "tags": []},
+ llm_or_chain_factory=mock_llm,
+ input_mapper=llm_input_mapper,
+ )
assert isinstance(llm_result, str)
assert llm_result == "somenumber"
@@ -324,10 +333,14 @@ async def test_arun_on_dataset(monkeypatch: pytest.MonkeyPatch) -> None:
)
expected = {
- uuid_: {
- "output": {"result": f"Result for example {uuid.UUID(uuid_)}"},
+ str(example.id): {
+ "output": {
+ "result": f"Result for example {uuid.UUID(str(example.id))}"
+ },
+ "input": {"input": example.inputs["input"]},
+ "reference": {"output": example.outputs["output"]},
"feedback": [],
}
- for uuid_ in uuids
+ for example in examples
}
assert results["results"] == expected