This PR replaces the memory stream implementation used by the
LogStreamCallbackHandler.
This implementation resolves an issue in which streamed logs and
streamed events originating from sync code would arrive only after the
entire sync code would finish execution (rather than arriving in real
time as they're generated).
One example is if trying to stream tokens from an llm within a tool. If
the tool was an async tool, but the llm was invoked via stream (sync
variant) rather than astream (async variant), then the tokens would fail
to stream in real time and would all arrived bunched up after the tool
invocation completed.
**Description:** Invoke callback prior to yielding token in stream
method for watsonx.
**Issue:** [Callback for on_llm_new_token should be invoked before the
token is yielded by the model
#16913](https://github.com/langchain-ai/langchain/issues/16913)
Co-authored-by: Robby <h0rv@users.noreply.github.com>
**Description:** changed filtering so that failed filter doesn't add
document to results. Currently filtering is entirely broken and all
documents are returned whether or not they pass the filter.
fixes issue introduced in
https://github.com/langchain-ai/langchain/pull/16190
- **Description:** Adds the document loader for [AWS
Athena](https://aws.amazon.com/athena/), a serverless and interactive
analytics service.
- **Dependencies:** Added boto3 as a dependency
- **Description:** This PR adds support for `search_types="mmr"` and
`search_type="similarity_score_threshold"` to retrievers using
`DatabricksVectorSearch`,
- **Issue:**
- **Dependencies:**
- **Twitter handle:**
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
Ref: https://openai.com/pricing
<!-- Thank you for contributing to LangChain!
Please title your PR "<package>: <description>", where <package> is
whichever of langchain, community, core, experimental, etc. is being
modified.
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,
- **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` from the root
of the package you've modified to check this locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc: https://python.langchain.com/docs/contributing/
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/docs/integrations` directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
Unlike vector results, the LLM has to completely trust the context of a
graph database result, even if it doesn't provide whole context. We
tried with instructions, but it seems that adding a single example is
the way to go to solve this issue.
### This pull request makes the following changes:
* Fixed issue #16913
Fixed the google gen ai chat_models.py code to make sure that the
callback is called before the token is yielded
<!-- Thank you for contributing to LangChain!
Please title your PR "<package>: <description>", where <package> is
whichever of langchain, community, core, experimental, etc. is being
modified.
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,
- **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` from the root
of the package you've modified to check this locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc: https://python.langchain.com/docs/contributing/
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/docs/integrations` directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
<!-- Thank you for contributing to LangChain!
Please title your PR "<package>: <description>", where <package> is
whichever of langchain, community, core, experimental, etc. is being
modified.
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,
- **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` from the root
of the package you've modified to check this locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc: https://python.langchain.com/docs/contributing/
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/docs/integrations` directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
Pydantic's `dict()` function raises an error here if you pass in a
generator. We have a more robust serialization function in lagnsmith
that we will use instead.
**Description**
Make some functions work with Milvus:
1. get_ids: Get primary keys by field in the metadata
2. delete: Delete one or more entities by ids
3. upsert: Update/Insert one or more entities
**Issue**
None
**Dependencies**
None
**Tag maintainer:**
@hwchase17
**Twitter handle:**
None
---------
Co-authored-by: HoaNQ9 <hoanq.1811@gmail.com>
Co-authored-by: Erick Friis <erick@langchain.dev>
<!-- 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:**
1. Modify LLMs/Anyscale to work with OAI v1
2. Get rid of openai_ prefixed variables in Chat_model/ChatAnyscale
3. Modify `anyscale_api_base` to `anyscale_base_url` to follow OAI name
convention (reverted)
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
## Summary
This PR upgrades LangChain's Ruff configuration in preparation for
Ruff's v0.2.0 release. (The changes are compatible with Ruff v0.1.5,
which LangChain uses today.) Specifically, we're now warning when
linter-only options are specified under `[tool.ruff]` instead of
`[tool.ruff.lint]`.
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
Co-authored-by: Bagatur <baskaryan@gmail.com>
- **Issue:** Issue with model argument support (been there for a while
actually):
- Non-specially-handled arguments like temperature don't work when
passed through constructor.
- Such arguments DO work quite well with `bind`, but also do not abide
by field requirements.
- Since initial push, server-side error messages have gotten better and
v0.0.2 raises better exceptions. So maybe it's better to let server-side
handle such issues?
- **Description:**
- Removed ChatNVIDIA's argument fields in favor of
`model_kwargs`/`model_kws` arguments which aggregates constructor kwargs
(from constructor pathway) and merges them with call kwargs (bind
pathway).
- Shuffled a few functions from `_NVIDIAClient` to `ChatNVIDIA` to
streamline construction for future integrations.
- Minor/Optional: Old services didn't have stop support, so client-side
stopping was implemented. Now do both.
- **Any Breaking Changes:** Minor breaking changes if you strongly rely
on chat_model.temperature, etc. This is captured by
chat_model.model_kwargs.
PR passes tests and example notebooks and example testing. Still gonna
chat with some people, so leaving as draft for now.
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
Description: Missing _identifying_params create issues when dealing with
callbacks to get current run model parameters.
All other model partners implementation provide this property and also
provide _default_params. I'm not sure about the default values to
include or if we can re-use the same as for _VertexAICommon(), this
change allows you to access the model parameters correctly.
Issue: Not exactly this issue but could be related
https://github.com/langchain-ai/langchain/issues/14711
Twitter handle:@musicaoriginal2
The streaming API doesn't separate safety_settings from the
generation_config payload. As the result the following error is observed
when using `stream` API. The functionality is correct with `invoke` API.
The fix separates the `safety_settings` from params and sets it as
argument to the `send_message` method.
```
ERROR: Unknown field for GenerationConfig: safety_settings
Traceback (most recent call last):
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/langchain_core/language_models/chat_models.py", line 250, in stream
raise e
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/langchain_core/language_models/chat_models.py", line 234, in stream
for chunk in self._stream(
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/langchain_google_vertexai/chat_models.py", line 501, in _stream
for response in responses:
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/vertexai/generative_models/_generative_models.py", line 921, in _send_message_streaming
for chunk in stream:
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/vertexai/generative_models/_generative_models.py", line 514, in _generate_content_streaming
request = self._prepare_request(
^^^^^^^^^^^^^^^^^^^^^^
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/vertexai/generative_models/_generative_models.py", line 256, in _prepare_request
gapic_generation_config = gapic_content_types.GenerationConfig(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/user/Library/Caches/pypoetry/virtualenvs/chatbot-worker-main-Ju-qIM-X-py3.12/lib/python3.12/site-packages/proto/message.py", line 576, in __init__
raise ValueError(
ValueError: Unknown field for GenerationConfig: safety_settings
```
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
I noticed that RunnableConfigurableAlternatives which is an important
composition in LCEL has no Docstring. Therefore I added the detailed
Docstring for it.
@baskaryan, @eyurtsev, @hwchase17 please have a look and let me if the
docstring is looking good.
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
This PR enables changing the behaviour of huggingface pipeline between
different calls. For example, before this PR there's no way of changing
maximum generation length between different invocations of the chain.
This is desirable in cases, such as when we want to scale the maximum
output size depending on a dynamic prompt size.
Usage example:
```python
from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
model_id = "gpt2"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
hf = HuggingFacePipeline(pipeline=pipe)
hf("Say foo:", pipeline_kwargs={"max_new_tokens": 42})
```
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
<!-- Thank you for contributing to LangChain!
Please title your PR "<package>: <description>", where <package> is
whichever of langchain, community, core, experimental, etc. is being
modified.
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,
- **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` from the root
of the package you've modified to check this locally.
See contribution guidelines for more information on how to write/run
tests, lint, etc: https://python.langchain.com/docs/contributing/
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/docs/integrations` directory.
If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.
-->
- **Description: changes to you.com files**
- general cleanup
- adds community/utilities/you.py, moving bulk of code from retriever ->
utility
- removes `snippet` as endpoint
- adds `news` as endpoint
- adds more tests
<s>**Description: update community MAKE file**
- adds `integration_tests`
- adds `coverage`</s>
- **Issue:** the issue # it fixes if applicable,
- [For New Contributors: Update Integration
Documentation](https://github.com/langchain-ai/langchain/issues/15664#issuecomment-1920099868)
- **Dependencies:** n/a
- **Twitter handle:** @scottnath
- **Mastodon handle:** scottnath@mastodon.social
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
- **Description:** This adds a recursive json splitter class to the
existing text_splitters as well as unit tests
- **Issue:** splitting text from structured data can cause issues if you
have a large nested json object and you split it as regular text you may
end up losing the structure of the json. To mitigate against this you
can split the nested json into large chunks and overlap them, but this
causes unnecessary text processing and there will still be times where
the nested json is so big that the chunks get separated from the parent
keys.
As an example you wouldn't want the following to be split in half:
```shell
{'val0': 'DFWeNdWhapbR',
'val1': {'val10': 'QdJo',
'val11': 'FWSDVFHClW',
'val12': 'bkVnXMMlTiQh',
'val13': 'tdDMKRrOY',
'val14': 'zybPALvL',
'val15': 'JMzGMNH',
'val16': {'val160': 'qLuLKusFw',
'val161': 'DGuotLh',
'val162': 'KztlcSBropT',
-----------------------------------------------------------------------split-----
'val163': 'YlHHDrN',
'val164': 'CtzsxlGBZKf',
'val165': 'bXzhcrWLmBFp',
'val166': 'zZAqC',
'val167': 'ZtyWno',
'val168': 'nQQZRsLnaBhb',
'val169': 'gSpMbJwA'},
'val17': 'JhgiyF',
'val18': 'aJaqjUSFFrI',
'val19': 'glqNSvoyxdg'}}
```
Any llm processing the second chunk of text may not have the context of
val1, and val16 reducing accuracy. Embeddings will also lack this
context and this makes retrieval less accurate.
Instead you want it to be split into chunks that retain the json
structure.
```shell
{'val0': 'DFWeNdWhapbR',
'val1': {'val10': 'QdJo',
'val11': 'FWSDVFHClW',
'val12': 'bkVnXMMlTiQh',
'val13': 'tdDMKRrOY',
'val14': 'zybPALvL',
'val15': 'JMzGMNH',
'val16': {'val160': 'qLuLKusFw',
'val161': 'DGuotLh',
'val162': 'KztlcSBropT',
'val163': 'YlHHDrN',
'val164': 'CtzsxlGBZKf'}}}
```
and
```shell
{'val1':{'val16':{
'val165': 'bXzhcrWLmBFp',
'val166': 'zZAqC',
'val167': 'ZtyWno',
'val168': 'nQQZRsLnaBhb',
'val169': 'gSpMbJwA'},
'val17': 'JhgiyF',
'val18': 'aJaqjUSFFrI',
'val19': 'glqNSvoyxdg'}}
```
This recursive json text splitter does this. Values that contain a list
can be converted to dict first by using split(... convert_lists=True)
otherwise long lists will not be split and you may end up with chunks
larger than the max chunk.
In my testing large json objects could be split into small chunks with
✅ Increased question answering accuracy
✅ The ability to split into smaller chunks meant retrieval queries can
use fewer tokens
- **Dependencies:** json import added to text_splitter.py, and random
added to the unit test
- **Twitter handle:** @joelsprunger
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
**Description:** Databricks LLM does not support SerDe the
transform_input_fn and transform_output_fn. After saving and loading,
the LLM will be broken. This PR serialize these functions into a hex
string using pickle, and saving the hex string in the yaml file. Using
pickle to serialize a function can be flaky, but this is a simple
workaround that unblocks many use cases. If more sophisticated SerDe is
needed, we can improve it later.
Test:
Added a simple unit test.
I did manual test on Databricks and it works well.
The saved yaml looks like:
```
llm:
_type: databricks
cluster_driver_port: null
cluster_id: null
databricks_uri: databricks
endpoint_name: databricks-mixtral-8x7b-instruct
extra_params: {}
host: e2-dogfood.staging.cloud.databricks.com
max_tokens: null
model_kwargs: null
n: 1
stop: null
task: null
temperature: 0.0
transform_input_fn: 80049520000000000000008c085f5f6d61696e5f5f948c0f7472616e73666f726d5f696e7075749493942e
transform_output_fn: null
```
@baskaryan
```python
from langchain_community.embeddings import DatabricksEmbeddings
from langchain_community.llms import Databricks
from langchain.chains import RetrievalQA
from langchain.document_loaders import TextLoader
from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores import FAISS
import mlflow
embeddings = DatabricksEmbeddings(endpoint="databricks-bge-large-en")
def transform_input(**request):
request["messages"] = [
{
"role": "user",
"content": request["prompt"]
}
]
del request["prompt"]
return request
llm = Databricks(endpoint_name="databricks-mixtral-8x7b-instruct", transform_input_fn=transform_input)
persist_dir = "faiss_databricks_embedding"
# Create the vector db, persist the db to a local fs folder
loader = TextLoader("state_of_the_union.txt")
documents = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
docs = text_splitter.split_documents(documents)
db = FAISS.from_documents(docs, embeddings)
db.save_local(persist_dir)
def load_retriever(persist_directory):
embeddings = DatabricksEmbeddings(endpoint="databricks-bge-large-en")
vectorstore = FAISS.load_local(persist_directory, embeddings)
return vectorstore.as_retriever()
retriever = load_retriever(persist_dir)
retrievalQA = RetrievalQA.from_llm(llm=llm, retriever=retriever)
with mlflow.start_run() as run:
logged_model = mlflow.langchain.log_model(
retrievalQA,
artifact_path="retrieval_qa",
loader_fn=load_retriever,
persist_dir=persist_dir,
)
# Load the retrievalQA chain
loaded_model = mlflow.pyfunc.load_model(logged_model.model_uri)
print(loaded_model.predict([{"query": "What did the president say about Ketanji Brown Jackson"}]))
```
