This PR introduces adjustments to ensure compatibility with the recently
released preview version of [TiDB Serverless Vector
Search](https://tidb.cloud/ai), aiming to prevent user confusion.
- TiDB Vector now supports vector indexing with cosine and l2 distance
strategies, although inner_product remains unsupported.
- Changing the distance strategy is currently not supported, so the test
cased should be adjusted.
Issue: the `service` optional parameter was mentioned but not used.
Fix: added this parameter.
---------
Co-authored-by: Chester Curme <chester.curme@gmail.com>
## Description
There is a bug in the concatenation of embeddings obtained from MLflow
that does not conform to the type hint requested by the function.
``` python
def _query(self, texts: List[str]) -> List[List[float]]:
```
It is logical to expect a **List[List[float]]** for a **List[str]**.
However, the append method encapsulates the response in a global List.
To avoid this, the extend method should be used, which will add the
embeddings of all strings at the same list level.
## Testing
I have tried using OpenAI-ADA to obtain the embeddings, and the result
of executing this snippet is as follows:
``` python
embeds = await MlflowAIGatewayEmbeddings().aembed_documents(texts=["hi", "how are you?"])
print(embeds)
```
``` python
[[[-0.03512698, -0.020624293, -0.015343423, ...], [-0.021260535, -0.011461929, -0.00033121882, ...]]]
```
When in reality, the expected result should be:
``` python
[[-0.03512698, -0.020624293, -0.015343423, ...], [-0.021260535, -0.011461929, -0.00033121882, ...]]
```
The above result complies with the expected type hint:
**List[List[float]]** . As I mentioned, we can achieve that by using the
extend method instead of the append method.
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
Co-authored-by: ccurme <chester.curme@gmail.com>
Thank you for contributing to LangChain!
- [x] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core,
experimental, etc. is being modified. Use "docs: ..." for purely docs
changes, "templates: ..." for template changes, "infra: ..." for CI
changes.
- Example: "community: add foobar LLM"
Description: Simply pass kwargs to allow arguments like "where" to be
propagated
Issue: Previously, db.delete(where={}) wouldn't work for chroma
vectorstores
Dependencies: N/A
Twitter handle: N/A
- [x] **Add tests and docs**: 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.
- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/
Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.
If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, hwchase17.
Description: Send both the query and query_embedding to the Databricks
index for hybrid search.
Issue: When using hybrid search with non-Databricks managed embedding we
currently don't pass both the embedding and query_text to the index.
Hybrid search requires both of these. This change fixes this issue for
both `similarity_search` and `similarity_search_by_vector`.
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
# Issue
As of late July, Perplexity [no longer supports Llama 3
models](https://docs.perplexity.ai/changelog/introducing-new-and-improved-sonar-models).
# Description
This PR updates the default model and doc examples to reflect their
latest supported model. (Mostly updating the same places changed by
#23723.)
# Twitter handle
`@acompa_` on behalf of the team at Not Diamond. Check us out
[here](https://notdiamond.ai).
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Hello.
First of all, thank you for maintaining such a great project.
## Description
In https://github.com/langchain-ai/langchain/pull/25123, support for
structured_output is added. However, `"additionalProperties": false`
needs to be included at all levels when a nested object is generated.
error from current code:
https://gist.github.com/fufufukakaka/e9b475300e6934853d119428e390f204
```
BadRequestError: Error code: 400 - {'error': {'message': "Invalid schema for response_format 'JokeWithEvaluation': In context=('properties', 'self_evaluation'), 'additionalProperties' is required to be supplied and to be false", 'type': 'invalid_request_error', 'param': 'response_format', 'code': None}}
```
Reference: [Introducing Structured Outputs in the
API](https://openai.com/index/introducing-structured-outputs-in-the-api/)
```json
{
"model": "gpt-4o-2024-08-06",
"messages": [
{
"role": "system",
"content": "You are a helpful math tutor."
},
{
"role": "user",
"content": "solve 8x + 31 = 2"
}
],
"response_format": {
"type": "json_schema",
"json_schema": {
"name": "math_response",
"strict": true,
"schema": {
"type": "object",
"properties": {
"steps": {
"type": "array",
"items": {
"type": "object",
"properties": {
"explanation": {
"type": "string"
},
"output": {
"type": "string"
}
},
"required": ["explanation", "output"],
"additionalProperties": false
}
},
"final_answer": {
"type": "string"
}
},
"required": ["steps", "final_answer"],
"additionalProperties": false
}
}
}
}
```
In the current code, `"additionalProperties": false` is only added at
the last level.
This PR introduces the `_add_additional_properties_key` function, which
recursively adds `"additionalProperties": false` to the entire JSON
schema for the request.
Twitter handle: `@fukkaa1225`
Thank you!
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
Previously the code was able to only handle a single level of nesting
for subgraphs in mermaid. This change adds support for arbitrary nesting
of subgraphs.
This PR adds tiny improvements to the `GithubFileLoader` document loader
and its code sample, addressing the following issues:
1. Currently, the `file_extension` argument of `GithubFileLoader` does
not change its behavior at all.
1. The `GithubFileLoader` sample code in
`docs/docs/integrations/document_loaders/github.ipynb` does not work as
it stands.
The respective solutions I propose are the following:
1. Remove `file_extension` argument from `GithubFileLoader`.
1. Specify the branch as `master` (not the default `main`) and rename
`documents` as `document`.
---------
Co-authored-by: Isaac Francisco <78627776+isahers1@users.noreply.github.com>
When I used the Neo4JGraph enhanced_schema=True option, I ran into an
error because a prop min_size of None was compared numerically with an
int.
The fix I applied is similar to the pattern of skipping embeddings
elsewhere in the file.
Co-authored-by: ccurme <chester.curme@gmail.com>
**Description:**
LLM will stop generating text even in the middle of a sentence if
`finish_reason` is `length` (for OpenAI) or `stop_reason` is
`max_tokens` (for Anthropic).
To obtain longer outputs from LLM, we should call the message generation
API multiple times and merge the results into the text to circumvent the
API's output token limit.
The extra line breaks forced by the `merge_message_runs` function when
seamlessly merging messages can be annoying, so I added the option to
specify the chunk separator.
**Issue:**
No corresponding issues.
**Dependencies:**
No dependencies required.
**Twitter handle:**
@hanama_chem
https://x.com/hanama_chem
---------
Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
parsed_json is expected to be a list of dictionaries, but it seems to…
be a single dictionary instead.
This is at
libs/experimental/langchain_experimental/graph_transformers/llm.py
process process_response
Thank you for contributing to LangChain!
- [ ] **Bugfix**: "experimental: bugfix"
---------
Co-authored-by: based <basir.sedighi@nris.no>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
- Cause chunks are joined by space, so they can't be found in text, and
the final `start_index` is very possibility to be -1.
- The simplest way is to use the natural index of the chunk as
`start_index`.
- **Description:** This change adds the ID field that's required in
Pinecone to the result documents of the similarity search method.
- **Issue:** Lack of document metadata namely the ID field
- [x] **Add tests and docs**: 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.
- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/
Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.
If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.
---------
Co-authored-by: Chester Curme <chester.curme@gmail.com>
[langchain_core] Fix UnionType type var replacement
- Added types.UnionType to typing.Union mapping
Type replacement cause `TypeError: 'type' object is not subscriptable`
if any of union type comes as function `_py_38_safe_origin` return
`types.UnionType` instead of `typing.Union`
```python
>>> from types import UnionType
>>> from typing import Union, get_origin
>>> type_ = get_origin(str | None)
>>> type_
<class 'types.UnionType'>
>>> UnionType[(str, None)]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: 'type' object is not subscriptable
>>> Union[(str, None)]
typing.Optional[str]
```
---------
Co-authored-by: Chester Curme <chester.curme@gmail.com>
Description: DeepInfra 500 errors have useful information in the text
field that isn't being exposed to the user. I updated the error message
to fix this.
As an example, this code
```
from langchain_community.chat_models import ChatDeepInfra
from langchain_core.messages import HumanMessage
model = "meta-llama/Meta-Llama-3-70B-Instruct"
deepinfra_api_token = "..."
model = ChatDeepInfra(model=model, deepinfra_api_token=deepinfra_api_token)
messages = [HumanMessage("All work and no play makes Jack a dull boy\n" * 9000)]
response = model.invoke(messages)
```
Currently gives this error:
```
langchain_community.chat_models.deepinfra.ChatDeepInfraException: DeepInfra Server: Error 500
```
This change would give the following error:
```
langchain_community.chat_models.deepinfra.ChatDeepInfraException: DeepInfra Server error status 500: {"error":{"message":"Requested input length 99009 exceeds maximum input length 8192"}}
```
**Refactor PebbloRetrievalQA**
- Created `APIWrapper` and moved API logic into it.
- Created smaller functions/methods for better readability.
- Properly read environment variables.
- Removed unused code.
- Updated models
**Issue:** NA
**Dependencies:** NA
**tests**: NA
**Refactor PebbloSafeLoader**
- Created `APIWrapper` and moved API logic into it.
- Moved helper functions to the utility file.
- Created smaller functions and methods for better readability.
- Properly read environment variables.
- Removed unused code.
**Issue:** NA
**Dependencies:** NA
**tests**: Updated
limit the most recent documents to fetch from MongoDB database.
Thank you for contributing to LangChain!
- [ ] **limit the most recent documents to fetch from MongoDB
database.**: "langchain_mongodb: limit the most recent documents to
fetch from MongoDB database."
- [ ] **PR message**: ***Delete this entire checklist*** and replace
with
- **Description:** Added a doc_limit parameter which enables the limit
for the documents to fetch from MongoDB database
- **Issue:**
- **Dependencies:** None
---------
Co-authored-by: Chester Curme <chester.curme@gmail.com>
Description: The neo4j driver can raise a SessionExpired error, which is
considered a retriable error. If a query fails with a SessionExpired
error, this change retries every query once. This change will make the
neo4j integration less flaky.
Twitter handle: noahmay_
### Summary
Create `langchain-databricks` as a new partner packages. This PR does
not migrate all existing Databricks integration, but the package will
eventually contain:
* `ChatDatabricks` (implemented in this PR)
* `DatabricksVectorSearch`
* `DatabricksEmbeddings`
* ~`UCFunctionToolkit`~ (will be done after UC SDK work which
drastically simplify implementation)
Also, this PR does not add integration tests yet. This will be added
once the Databricks test workspace is ready.
Tagging @efriis as POC
### Tracker
[✍️] Create a package and imgrate ChatDatabricks
[ ] Migrate DatabricksVectorSearch, DatabricksEmbeddings, and their docs
~[ ] Migrate UCFunctionToolkit and its doc~
[ ] Add provider document and update README.md
[ ] Add integration tests and set up secrets (after moved to an external
package)
[ ] Add deprecation note to the community implementations.
---------
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
Co-authored-by: Erick Friis <erick@langchain.dev>
Thank you for contributing to LangChain!
**Description:** Adding `BoxRetriever` for langchain_box. This retriever
handles two use cases:
* Retrieve all documents that match a full-text search
* Retrieve the answer to a Box AI prompt as a Document
**Twitter handle:** @BoxPlatform
- [x] **Add tests and docs**: 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.
- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/
Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.
If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
