**Description**
- Added the `SingleStoreDBChatMessageHistory` class that inherits
`BaseChatMessageHistory` and allows to use of a SingleStoreDB database
as a storage for chat message history.
- Added integration test to check that everything works (requires
`singlestoredb` to be installed)
- Added notebook with usage example
- Removed custom retriever for SingleStoreDB vector store (as it is
useless)
---------
Co-authored-by: Volodymyr Tkachuk <vtkachuk-ua@singlestore.com>
## Description
| Tool | Original Tool Name |
|-----------------------------|---------------------------|
| open-meteo-api | Open Meteo API |
| news-api | News API |
| tmdb-api | TMDB API |
| podcast-api | Podcast API |
| golden_query | Golden Query |
| dall-e-image-generator | Dall-E Image Generator |
| twilio | Text Message |
| searx_search_results | Searx Search Results |
| dataforseo | DataForSeo Results JSON |
When using these tools through `load_tools`, I encountered the following
validation error:
```console
openai.error.InvalidRequestError: 'TMDB API' does not match '^[a-zA-Z0-9_-]{1,64}$' - 'functions.0.name'
```
In order to avoid this error, I replaced spaces with hyphens in the tool
names:
| Tool | Corrected Tool Name |
|-----------------------------|---------------------------|
| open-meteo-api | Open-Meteo-API |
| news-api | News-API |
| tmdb-api | TMDB-API |
| podcast-api | Podcast-API |
| golden_query | Golden-Query |
| dall-e-image-generator | Dall-E-Image-Generator |
| twilio | Text-Message |
| searx_search_results | Searx-Search-Results |
| dataforseo | DataForSeo-Results-JSON |
This correction resolved the validation error.
Additionally, a unit test,
`tests/unit_tests/schema/runnable/test_runnable.py::test_stream_log_retriever`,
was failing at random. Upon further investigation, I confirmed that the
failure was not related to the above-mentioned changes. The `stream_log`
variable was generating the order of logs in two ways at random The
reason for this behavior is unclear, but in the assertion, I included
both possible orders to account for this variability.
Hello Folks,
Alibaba Cloud OpenSearch has released a new version of the vector
storage engine, which has significantly improved performance compared to
the previous version. At the same time, the sdk has also undergone
changes, requiring adjustments alibaba opensearch vector store code to
adapt.
This PR includes:
Adapt to the latest version of Alibaba Cloud OpenSearch API.
More comprehensive unit testing.
Improve documentation.
I have read your contributing guidelines. And I have passed the tests
below
- [x] make format
- [x] make lint
- [x] make coverage
- [x] make test
---------
Co-authored-by: zhaoshengbo <shengbo.zsb@alibaba-inc.com>
**Description:**
While working on the Docusaurus site loader #9138, I noticed some
outdated docs and tests for the Sitemap Loader.
**Issue:**
This is tangentially related to #6691 in reference to doc links. I plan
on digging in to a few of these issue when I find time next.
- **Description:** added examples to Vertex chat models as optional
class attributes, so that a model with examples can be used inside a
chain
- **Twitter handle:** lkuligin
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
Related to #10800
- Errors in the Docstring of GradientLLM / Gradient.ai LLM
- Renamed the `model_id` to `model` and adapting this in all tests.
Reason to so is to be in Sync with `GradientEmbeddings` and other LLM's.
- inmproving tests so they check the headers in the sent request.
- making the aiosession a private attribute in the docs, as in the
future `pip install gradientai` will be replacing aiosession.
- adding a example how to fine-tune on the Prompt Template as suggested
in #10800
- Description: Adds the ChatEverlyAI class with llama-2 7b on [EverlyAI
Hosted
Endpoints](https://everlyai.xyz/)
- It inherits from ChatOpenAI and requires openai (probably unnecessary
but it made for a quick and easy implementation)
---------
Co-authored-by: everly-studio <127131037+everly-studio@users.noreply.github.com>
- **Description:**
- If the Elasticsearch field used for Langchain > Document.page_content
is missing because the specific document is
somehow malformed fail gracefully.
- **Tag maintainer:**
- @joemcelroy
Reverts langchain-ai/langchain#11714
This has linting and formatting issues, plus it's added to chat models
folder but doesn't subclass Chat Model base class
Motivation and Context
At present, the Baichuan Large Language Model is relatively popular and
efficient in performance. Due to widespread market recognition, this
model has been added to enhance the scalability of Langchain's ability
to access the big language model, so as to facilitate application access
and usage for interested users.
System Info
langchain： 0.0.295
python：3.8.3
IDE：vs code
Description
Add the following files:
1. Add baichuan_baichuaninc_endpoint.py in the
libs/langchain/langchain/chat_models
2. Modify the __init__.py file,which is located in the
libs/langchain/langchain/chat_models/__init__.py：
a. Add "from langchain.chat_models.baichuan_baichuaninc_endpoint import
BaichuanChatEndpoint"
b. Add "BaichuanChatEndpoint" In the file's __ All__ method
Your contribution
I am willing to help implement this feature and submit a PR, but I would
appreciate guidance from the maintainers or community to ensure the
changes are made correctly and in line with the project's standards and
practices.
- **Description:** Add `TrainableLLM` for those LLM support fine-tuning
- **Tag maintainer:** @hwchase17
This PR add training methods to `GradientLLM`
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
Hi there
This PR is aim to implement chat model for Alibaba Tongyi LLM model. It
contains work below:
1.Implement ChatTongyi chat model in langchain.chat_models.tongyi. Note
this is different with tongyi llm model to another PR
https://github.com/langchain-ai/langchain/pull/10878.
For detail it implements _generate() and _stream() function in
ChatTongyi.
2. Add some examples in chat/tongyi.ipynb.
3. Add integration test in chat_models/test_tongyi.py
Note async completion for the Text API is not yet supported.
Dependencies: dashscope. It will be installed manually cause it is not
need by everyone.
**Description**
This PR adds the `ElasticsearchChatMessageHistory` implementation that
stores chat message history in the configured
[Elasticsearch](https://www.elastic.co/elasticsearch/) deployment.
```python
from langchain.memory.chat_message_histories import ElasticsearchChatMessageHistory
history = ElasticsearchChatMessageHistory(
es_url="https://my-elasticsearch-deployment-url:9200", index="chat-history-index", session_id="123"
)
history.add_ai_message("This is me, the AI")
history.add_user_message("This is me, the human")
```
**Dependencies**
- [elasticsearch client](https://elasticsearch-py.readthedocs.io/)
required
Co-authored-by: Bagatur <baskaryan@gmail.com>
Instead of accessing `langchain.debug`, `langchain.verbose`, or
`langchain.llm_cache`, please use the new getter/setter functions in
`langchain.globals`:
- `langchain.globals.set_debug()` and `langchain.globals.get_debug()`
- `langchain.globals.set_verbose()` and
`langchain.globals.get_verbose()`
- `langchain.globals.set_llm_cache()` and
`langchain.globals.get_llm_cache()`
Using the old globals directly will now raise a warning.
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
**Description:**
Add a document loader for the RSpace Electronic Lab Notebook
(www.researchspace.com), so that scientific documents and research notes
can be easily pulled into Langchain pipelines.
**Issue**
This is an new contribution, rather than an issue fix.
**Dependencies:**
There are no new required dependencies.
In order to use the loader, clients will need to install rspace_client
SDK using `pip install rspace_client`
---------
Co-authored-by: richarda23 <richard.c.adams@infinityworks.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
**Description:** Update Indexing API docs to specify vectorstores that
are compatible with the Indexing API. I add a unit test to remind
developers to update the documentation whenever they add or change a
vectorstore in a way that affects compatibility. For the unit test I
repurposed existing code from
[here](https://github.com/langchain-ai/langchain/blob/v0.0.311/libs/langchain/langchain/indexes/_api.py#L245-L257).
This is my first PR to an open source project. This is a trivially
simple PR whose main purpose is to make me more comfortable submitting
Langchain PRs. If this PR goes through I plan to submit PRs with more
substantive changes in the near future.
**Issue:** Resolves
[10482](https://github.com/langchain-ai/langchain/discussions/10482).
**Dependencies:** No new dependencies.
**Twitter handle:** None.
Allows MMR functionality only for the case where we have access to the
embedding function. Also allows for users to request for fields from
elasticsearch store. These are added to the document metadata.
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
Description: Introducing an ability to load a transcription document of
audio file using [Yandex
SpeechKit](https://cloud.yandex.com/en-ru/services/speechkit)
Issue: None
Dependencies: yandex-speechkit
Tag maintainer: @rlancemartin, @eyurtsev
**Description**
This PR implements the usage of the correct tokenizer in Bedrock LLMs,
if using anthropic models.
**Issue:** #11560
**Dependencies:** optional dependency on `anthropic` python library.
**Twitter handle:** jtolgyesi
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
**Description:** Modify Anyscale integration to work with [Anyscale
Endpoint](https://docs.endpoints.anyscale.com/)
and it supports invoke, async invoke, stream and async invoke features
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
Should delegate to parse_result, not to aparse, as parse_result is a
method that some output parsers override
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
**Description:** Avoid huggingfacepipeline to truncate the response if
user setup return_full_text as False within huggingface pipeline.
**Dependencies:** : None
**Tag maintainer:** Maybe @sam-h-bean ?
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
- **Description:** implements a retriever on top of DocAI Warehouse (to
interact with existing enterprise documents)
https://cloud.google.com/document-ai-warehouse?hl=en
- **Issue:** new functionality
@baskaryan
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
No relevant documents may be found for a given question. In some use
cases, we could directly respond with a fixed message instead of doing
an LLM call with an empty context. This PR exposes this as an option:
response_if_no_docs_found.
---------
Co-authored-by: Sudharsan Rangarajan <sudranga@nile-global.com>
Replace this entire comment with:
- **Description:** In this modified version of the function, if the
metadatas parameter is not None, the function includes the corresponding
metadata in the JSON object for each text. This allows the metadata to
be stored alongside the text's embedding in the vector store.
-
- **Issue:** #10924
- **Dependencies:** None
- **Tag maintainer:** @hwchase17
@agola11
- **Twitter handle:** @MelliJoaco
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
- **Description:** fixed a bug in pal-chain when it reports Python
code validation errors. When node.func does not have any ids, the
original code tried to print node.func.id in raising ValueError.
- **Issue:** n/a,
- **Dependencies:** no dependencies,
- **Tag maintainer:** @hazzel-cn, @eyurtsev
- **Twitter handle:** @lazyswamp
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
I am merely making some minor adjustments to the function documentation.
I hope to provide a small assistance to LangChain.
- **Description:** Change the docs of JSONAgentOutputParser. It will be
`JSON` better,
- **Issue:** no,
- **Dependencies:** no,
- **Tag maintainer:** @hwchase17,
- **Twitter handle:** Not worth mentioning.
**Description:** This PR adds support for ChatOpenAI models in the
Infino callback handler. In particular, this PR implements
`on_chat_model_start` callback, so that ChatOpenAI models are supported.
With this change, Infino callback handler can be used to track latency,
errors, and prompt tokens for ChatOpenAI models too (in addition to the
support for OpenAI and other non-chat models it has today). The existing
example notebook is updated to show how to use this integration as well.
cc/ @naman-modi @savannahar68
**Issue:** https://github.com/langchain-ai/langchain/issues/11607
**Dependencies:** None
**Tag maintainer:** @hwchase17
**Twitter handle:** [@vkakade](https://twitter.com/vkakade)
This PR adds support for the Azure Cosmos DB MongoDB vCore Vector Store
https://learn.microsoft.com/en-us/azure/cosmos-db/mongodb/vcore/https://learn.microsoft.com/en-us/azure/cosmos-db/mongodb/vcore/vector-search
Summary:
- **Description:** added vector store integration for Azure Cosmos DB
MongoDB vCore Vector Store,
- **Issue:** the issue # it fixes#11627,
- **Dependencies:** pymongo dependency,
- **Tag maintainer:** @hwchase17,
- **Twitter handle:** @izzyacademy
---------
Co-authored-by: Israel Ekpo <israel.ekpo@gmail.com>
Co-authored-by: Israel Ekpo <44282278+izzyacademy@users.noreply.github.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
* Should use non chunked messages for Invoke/Batch
* After this PR, stream output type is not represented, do we want to
use the union?
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
Adds standard `type` field for all messages that will be
serialized/validated by pydantic.
* The presence of `type` makes it easier for developers consuming
schemas to write client code to serialize/deserialize.
* In LangServe `type` will be used for both validation and will appear
in the generated openapi specs
Preventing error caused by attempting to move the model that was already
loaded on the GPU using the Accelerate module to the same or another
device. It is not possible to load model with Accelerate/PEFT to CPU for
now
Addresses:
[#10985](https://github.com/langchain-ai/langchain/issues/10985)
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
- **Description:** This is an update to OctoAI LLM provider that adds
support for llama2 endpoints hosted on OctoAI and updates MPT-7b url
with the current one.
@baskaryan
Thanks!
---------
Co-authored-by: ML Wiz <bassemgeorgi@gmail.com>
**Description:** I noticed the metadata returned by the url_selenium
loader was missing several values included by the web_base loader. (The
former returned `{source: ...}`, the latter returned `{source: ...,
title: ..., description: ..., language: ...}`.) This change fixes it so
both loaders return all 4 key value pairs.
Files have been properly formatted and all tests are passing. Note,
however, that I am not much of a python expert, so that whole "Adding
the imports inside the code so that tests pass" thing seems weird to me.
Please LMK if I did anything wrong.
- **Description:** Assigning the custom_llm_provider to the default
params function so that it will be passed to the litellm
- **Issue:** Even though the custom_llm_provider argument is being
defined it's not being assigned anywhere in the code and hence its not
being passed to litellm, therefore any litellm call which uses the
custom_llm_provider as required parameter is being failed. This
parameter is mainly used by litellm when we are doing inference via
Custom API server.
https://docs.litellm.ai/docs/providers/custom_openai_proxy
- **Dependencies:** No dependencies are required
@krrishdholakia , @baskaryan
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
- **Description:** This PR introduces a new LLM and Retriever API to
https://arcee.ai for the python client
- **Issue:** implements the integrations as requested in #11578 ,
- **Dependencies:** no dependencies are required,
- **Tag maintainer:** @hwchase17
- **Twitter handle:** shwooobham
**✅ `make format`, `make lint` and `make test` runs locally.**
```shell
=========== 1245 passed, 277 skipped, 20 warnings in 16.26s ===========
./scripts/check_pydantic.sh .
./scripts/check_imports.sh
poetry run ruff .
[ "." = "" ] || poetry run black . --check
All done! ✨🍰✨
1818 files would be left unchanged.
[ "." = "" ] || poetry run mypy .
Success: no issues found in 1815 source files
[ "." = "" ] || poetry run black .
All done! ✨🍰✨
1818 files left unchanged.
[ "." = "" ] || poetry run ruff --select I --fix .
poetry run codespell --toml pyproject.toml
poetry run codespell --toml pyproject.toml -w
```
**Contributions**
1. Arcee (langchain/llms), ArceeRetriever (langchain/retrievers),
ArceeWrapper (langchain/utilities)
2. docs for Arcee (llms/arcee.py) and
ArceeRetriever(retrievers/arcee.py)
3.
cc: @jacobsolawetz @ben-epstein
---------
Co-authored-by: Shubham <shubham@sORo.local>
jinja2 templates are not sandboxed and are at risk for arbitrary code
execution. To mitigate this risk:
- We no longer support loading jinja2-formatted prompt template files.
- `PromptTemplate` with jinja2 may still be constructed manually, but
the class carries a security warning reminding the user to not pass
untrusted input into it.
Resolves#4394.
<!-- Thank you for contributing to LangChain!
Replace this entire comment with:
- **Description:** a description of the change,
- **Issue:** the issue # it fixes (if applicable),
- **Dependencies:** any dependencies required for this change,
- **Tag maintainer:** for a quicker response, tag the relevant
maintainer (see below),
- **Twitter handle:** we announce bigger features on Twitter. If your PR
gets announced, and you'd like a mention, we'll gladly shout you out!
Please make sure your PR is passing linting and testing before
submitting. Run `make format`, `make lint` and `make test` to check this
locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md
If you're adding a new integration, please include:
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in `docs/extras`
directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
**Description:** CohereRerank is missing `cohere_api_key` as a field and
since extras are forbidden, it is not possible to pass-in the key. The
only way is to use an env variable named `COHERE_API_KEY`.
For example, if trying to create a compressor like this:
```python
cohere_api_key = "......Cohere api key......"
compressor = CohereRerank(cohere_api_key=cohere_api_key)
```
you will get the following error:
```
File "/langchain/.venv/lib/python3.10/site-packages/pydantic/v1/main.py", line 341, in __init__
raise validation_error
pydantic.v1.error_wrappers.ValidationError: 1 validation error for CohereRerank
cohere_api_key
extra fields not permitted (type=value_error.extra)
```
- **Description:** Fixes minor typo for the
query_sql_database_tool_description in the db toolkit
- **Issue:** N/A
- **Dependencies:** N/A
- **Tag maintainer:** @nfcampos
- **Twitter handle:** N/A
LangChain relies on NumPy to compute cosine distances, which becomes a
bottleneck with the growing dimensionality and number of embeddings. To
avoid this bottleneck, in our libraries at
[Unum](https://github.com/unum-cloud), we have created a specialized
package - [SimSIMD](https://github.com/ashvardanian/simsimd), that knows
how to use newer hardware capabilities. Compared to SciPy and NumPy, it
reaches 3x-200x performance for various data types. Since publication,
several LangChain users have asked me if I can integrate it into
LangChain to accelerate their workflows, so here I am 🤗
## Benchmarking
To conduct benchmarks locally, run this in your Jupyter:
```py
import numpy as np
import scipy as sp
import simsimd as simd
import timeit as tt
def cosine_similarity_np(X: np.ndarray, Y: np.ndarray) -> np.ndarray:
X_norm = np.linalg.norm(X, axis=1)
Y_norm = np.linalg.norm(Y, axis=1)
with np.errstate(divide="ignore", invalid="ignore"):
similarity = np.dot(X, Y.T) / np.outer(X_norm, Y_norm)
similarity[np.isnan(similarity) | np.isinf(similarity)] = 0.0
return similarity
def cosine_similarity_sp(X: np.ndarray, Y: np.ndarray) -> np.ndarray:
return 1 - sp.spatial.distance.cdist(X, Y, metric='cosine')
def cosine_similarity_simd(X: np.ndarray, Y: np.ndarray) -> np.ndarray:
return 1 - simd.cdist(X, Y, metric='cosine')
X = np.random.randn(1, 1536).astype(np.float32)
Y = np.random.randn(1, 1536).astype(np.float32)
repeat = 1000
print("NumPy: {:,.0f} ops/s, SciPy: {:,.0f} ops/s, SimSIMD: {:,.0f} ops/s".format(
repeat / tt.timeit(lambda: cosine_similarity_np(X, Y), number=repeat),
repeat / tt.timeit(lambda: cosine_similarity_sp(X, Y), number=repeat),
repeat / tt.timeit(lambda: cosine_similarity_simd(X, Y), number=repeat),
))
```
## Results
I ran this on an M2 Pro Macbook for various data types and different
number of rows in `X` and reformatted the results as a table for
readability:
| Data Type | NumPy | SciPy | SimSIMD |
| :--- | ---: | ---: | ---: |
| `f32, 1` | 59,114 ops/s | 80,330 ops/s | 475,351 ops/s |
| `f16, 1` | 32,880 ops/s | 82,420 ops/s | 650,177 ops/s |
| `i8, 1` | 47,916 ops/s | 115,084 ops/s | 866,958 ops/s |
| `f32, 10` | 40,135 ops/s | 24,305 ops/s | 185,373 ops/s |
| `f16, 10` | 7,041 ops/s | 17,596 ops/s | 192,058 ops/s |
| `f16, 10` | 21,989 ops/s | 25,064 ops/s | 619,131 ops/s |
| `f32, 100` | 3,536 ops/s | 3,094 ops/s | 24,206 ops/s |
| `f16, 100` | 900 ops/s | 2,014 ops/s | 23,364 ops/s |
| `i8, 100` | 5,510 ops/s | 3,214 ops/s | 143,922 ops/s |
It's important to note that SimSIMD will underperform if both matrices
are huge.
That, however, seems to be an uncommon usage pattern for LangChain
users.
You can find a much more detailed performance report for different
hardware models here:
- [Apple M2
Pro](https://ashvardanian.com/posts/simsimd-faster-scipy/#appendix-1-performance-on-apple-m2-pro).
- [4th Gen Intel Xeon
Platinum](https://ashvardanian.com/posts/simsimd-faster-scipy/#appendix-2-performance-on-4th-gen-intel-xeon-platinum-8480).
- [AWS Graviton
3](https://ashvardanian.com/posts/simsimd-faster-scipy/#appendix-3-performance-on-aws-graviton-3).
## Additional Notes
1. Previous version used `X = np.array(X)`, to repackage lists of lists.
It's an anti-pattern, as it will use double-precision floating-point
numbers, which are slow on both CPUs and GPUs. I have replaced it with
`X = np.array(X, dtype=np.float32)`, but a more selective approach
should be discussed.
2. In numerical computations, it's recommended to explicitly define
tolerance levels, which were previously avoided in
`np.allclose(expected, actual)` calls. For now, I've set absolute
tolerance to distance computation errors as 0.01: `np.allclose(expected,
actual, atol=1e-2)`.
---
- **Dependencies:** adds `simsimd` dependency
- **Tag maintainer:** @hwchase17
- **Twitter handle:** @ashvardanian
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
#### Description
This PR adds the option to specify additional metadata columns in the
CSVLoader beyond just `Source`.
The current CSV loader includes all columns in `page_content` and if we
want to have columns specified for `page_content` and `metadata` we have
to do something like the below.:
```
csv = pd.read_csv(
"path_to_csv"
).to_dict("records")
documents = [
Document(
page_content=doc["content"],
metadata={
"last_modified_by": doc["last_modified_by"],
"point_of_contact": doc["point_of_contact"],
}
) for doc in csv
]
```
#### Usage
Example Usage:
```
csv_test = CSVLoader(
file_path="path_to_csv",
metadata_columns=["last_modified_by", "point_of_contact"]
)
```
Example CSV:
```
content, last_modified_by, point_of_contact
"hello world", "Person A", "Person B"
```
Example Result:
```
Document {
page_content: "hello world"
metadata: {
row: '0',
source: 'path_to_csv',
last_modified_by: 'Person A',
point_of_contact: 'Person B',
}
```
---------
Co-authored-by: Ben Chello <bchello@dropbox.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
- **Description:** Fixes the comments in the ConvoOutputParser. Because
the \\\\ is escaping a single \\, they render something like:
`"action_input": string \ The input to the action` in the prompt.
Changing this to \\\\\\\\ lets it escape two slashes so that it renders
a proper comment: `"action_input": string \\ The input to the action`
- **Issue:** N/A
- **Dependencies:**
- **Tag maintainer:** @hwchase17
- **Twitter handle:**
**Description**:
- Added Momento Vector Index (MVI) as a vector store provider. This
includes an implementation with docstrings, integration tests, a
notebook, and documentation on the docs pages.
- Updated the Momento dependency in pyproject.toml and the lock file to
enable access to MVI.
- Refactored the Momento cache and chat history session store to prefer
using "MOMENTO_API_KEY" over "MOMENTO_AUTH_TOKEN" for consistency with
MVI. This change is backwards compatible with the previous "auth_token"
variable usage. Updated the code and tests accordingly.
**Dependencies**:
- Updated Momento dependency in pyproject.toml.
**Testing**:
- Run the integration tests with a Momento API key. Get one at the
[Momento Console](https://console.gomomento.com) for free. MVI is
available in AWS us-west-2 with a superuser key.
- `MOMENTO_API_KEY=<your key> poetry run pytest
tests/integration_tests/vectorstores/test_momento_vector_index.py`
**Tag maintainer:**
@eyurtsev
**Twitter handle**:
Please mention @momentohq for this addition to langchain. With the
integration of Momento Vector Index, Momento caching, and session store,
Momento provides serverless support for the core langchain data needs.
Also mention @mlonml for the integration.
