Add unit tests that show differences between sync / async versions when
streaming.
The inner on_chain_chunk event is missing if mixing sync and async
functionality. Likely due to missing tap_output_iter implementation on
the sync variant of `_transform_stream_with_config`
0.2 is not a breaking release for core (but it is for langchain and
community)
To keep the core+langchain+community packages in sync at 0.2, we will
relax deps throughout the ecosystem to tolerate `langchain-core` 0.2
## Description
This PR introduces the new `langchain-qdrant` partner package, intending
to deprecate the community package.
## Changes
- Moved the Qdrant vector store implementation `/libs/partners/qdrant`
with integration tests.
- The conditional imports of the client library are now regular with
minor implementation improvements.
- Added a deprecation warning to
`langchain_community.vectorstores.qdrant.Qdrant`.
- Replaced references/imports from `langchain_community` with either
`langchain_core` or by moving the definitions to the `langchain_qdrant`
package itself.
- Updated the Qdrant vector store documentation to reflect the changes.
## Testing
- `QDRANT_URL` and
[`QDRANT_API_KEY`](583e36bf6b)
env values need to be set to [run integration
tests](d608c93d1f)
in the [cloud](https://cloud.qdrant.tech).
- If a Qdrant instance is running at `http://localhost:6333`, the
integration tests will use it too.
- By default, tests use an
[`in-memory`](https://github.com/qdrant/qdrant-client?tab=readme-ov-file#local-mode)
instance(Not comprehensive).
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
Co-authored-by: Erick Friis <erickfriis@gmail.com>
This PR makes some small updates for `KuzuQAChain` for graph QA.
- Updated Cypher generation prompt (we now support `WHERE EXISTS`) and
generalize it more
- Support different LLMs for Cypher generation and QA
- Update docs and examples
First Pr for the langchain_huggingface partner Package
- Moved some of the hugging face related class from `community` to the
new `partner package`
Still needed :
- Documentation
- Tests
- Support for the new apply_chat_template in `ChatHuggingFace`
- Confirm choice of class to support for embeddings witht he
sentence-transformer team.
cc : @efriis
---------
Co-authored-by: Cyril Kondratenko <kkn1993@gmail.com>
Co-authored-by: Erick Friis <erick@langchain.dev>
- Introduce the `merge_and_split` function in the
`UpstageLayoutAnalysisLoader`.
- The `merge_and_split` function takes a list of documents and a
splitter as inputs.
- This function merges all documents and then divides them using the
`split_documents` method, which is a proprietary function of the
splitter.
- If the provided splitter is `None` (which is the default setting), the
function will simply merge the documents without splitting them.
Adds a Python REPL that executes code in a code interpreter session
using Azure Container Apps dynamic sessions.
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
Thank you for contributing to LangChain!
- [X] **PR title**: "community: Add source metadata to bedrock retriever
response"
- [X] **PR message**:
- **Description:** Bedrock retrieve API returns extra metadata in the
response which is currently not returned in the retriever response
- **Issue:** The change adds the metadata from bedrock retrieve API
response to the bedrock retriever in a backward compatible way. Renamed
metadata to sourceMetadata as metadata term is being used in the
Document already. This is in sync with what we are doing in llama-index
as well.
- **Dependencies:** No
- [X] **Add tests and docs**:
1. Added unit tests
2. Notebook already exists and does not need any change
3. Response from end to end testing, just to ensure backward
compatibility: `[Document(page_content='Exoplanets.',
metadata={'location': {'s3Location': {'uri':
's3://bucket/file_name.txt'}, 'type': 'S3'}, 'score': 0.46886647,
'source_metadata': {'x-amz-bedrock-kb-source-uri':
's3://bucket/file_name.txt', 'tag': 'space', 'team': 'Nasa', 'year':
1946.0}})]`
- [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.
---------
Co-authored-by: Piyush Jain <piyushjain@duck.com>
**Description:** Added a few additional arguments to the whisper parser,
which can be consumed by the underlying API.
The prompt is especially important to fine-tune transcriptions.
---------
Co-authored-by: Roi Perlman <roi@fivesigmalabs.com>
Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
**Description:**
This PR introduces chunking logic to the `DeepInfraEmbeddings` class to
handle large batch sizes without exceeding maximum batch size of the
backend. This enhancement ensures that embedding generation processes
large batches by breaking them down into smaller, manageable chunks,
each conforming to the maximum batch size limit.
**Issue:**
Fixes#21189
**Dependencies:**
No new dependencies introduced.
- Added new document_transformer: MarkdonifyTransformer, that uses
`markdonify` package with customizable options to convert HTML to
Markdown. It's similar to Html2TextTransformer, but has more flexible
options and also I've noticed that sometimes MarkdownifyTransformer
performs better than html2text one, so that's why I use markdownify on
my project.
- Added docs and tests
- Usage:
```python
from langchain_community.document_transformers import MarkdownifyTransformer
markdownify = MarkdownifyTransformer()
docs_transform = markdownify.transform_documents(docs)
```
- Example of better performance on simple task, that I've noticed:
```
<html>
<head><title>Reports on product movement</title></head>
<body>
<p data-block-key="2wst7">The reports on product movement will be useful for forming supplier orders and controlling outcomes.</p>
</body>
```
**Html2TextTransformer**:
```python
[Document(page_content='The reports on product movement will be useful for forming supplier orders and\ncontrolling outcomes.\n\n')]
# Here we can see 'and\ncontrolling', which has extra '\n' in it
```
**MarkdownifyTranformer**:
```python
[Document(page_content='Reports on product movement\n\nThe reports on product movement will be useful for forming supplier orders and controlling outcomes.')]
```
---------
Co-authored-by: Sokolov Fedor <f.sokolov@sokolov-macbook.bbrouter>
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Co-authored-by: Sokolov Fedor <f.sokolov@sokolov-macbook.local>
Co-authored-by: Sokolov Fedor <f.sokolov@192.168.1.6>
### GPT4AllEmbeddings parameters
---
**Description:**
As of right now the **Embed4All** class inside _GPT4AllEmbeddings_ is
instantiated as it's default which leaves no room to customize the
chosen model and it's behavior. Thus:
- GPT4AllEmbeddings can now be instantiated with custom parameters like
a different model that shall be used.
---------
Co-authored-by: AlexJauchWalser <alexander.jauch-walser@knime.com>
The `_amake_session()` method does not allow modifying the
`self.session_factory` with
anything other than `async_sessionmaker`. This prohibits advanced uses
of `index()`.
In a RAG architecture, it is necessary to import document chunks.
To keep track of the links between chunks and documents, we can use the
`index()` API.
This API proposes to use an SQL-type record manager.
In a classic use case, using `SQLRecordManager` and a vector database,
it is impossible
to guarantee the consistency of the import. Indeed, if a crash occurs
during the import
(problem with the network, ...)
there is an inconsistency between the SQL database and the vector
database.
With the
[PR](https://github.com/langchain-ai/langchain-postgres/pull/32) we are
proposing for `langchain-postgres`,
it is now possible to guarantee the consistency of the import of chunks
into
a vector database. It's possible only if the outer session is built
with the connection.
```python
def main():
db_url = "postgresql+psycopg://postgres:password_postgres@localhost:5432/"
engine = create_engine(db_url, echo=True)
embeddings = FakeEmbeddings()
pgvector:VectorStore = PGVector(
embeddings=embeddings,
connection=engine,
)
record_manager = SQLRecordManager(
namespace="namespace",
engine=engine,
)
record_manager.create_schema()
with engine.connect() as connection:
session_maker = scoped_session(sessionmaker(bind=connection))
# NOTE: Update session_factories
record_manager.session_factory = session_maker
pgvector.session_maker = session_maker
with connection.begin():
loader = CSVLoader(
"data/faq/faq.csv",
source_column="source",
autodetect_encoding=True,
)
result = index(
source_id_key="source",
docs_source=loader.load()[:1],
cleanup="incremental",
vector_store=pgvector,
record_manager=record_manager,
)
print(result)
```
The same thing is possible asynchronously, but a bug in
`sql_record_manager.py`
in `_amake_session()` must first be fixed.
```python
async def _amake_session(self) -> AsyncGenerator[AsyncSession, None]:
"""Create a session and close it after use."""
# FIXME: REMOVE if not isinstance(self.session_factory, async_sessionmaker):~~
if not isinstance(self.engine, AsyncEngine):
raise AssertionError("This method is not supported for sync engines.")
async with self.session_factory() as session:
yield session
```
Then, it is possible to do the same thing asynchronously:
```python
async def main():
db_url = "postgresql+psycopg://postgres:password_postgres@localhost:5432/"
engine = create_async_engine(db_url, echo=True)
embeddings = FakeEmbeddings()
pgvector:VectorStore = PGVector(
embeddings=embeddings,
connection=engine,
)
record_manager = SQLRecordManager(
namespace="namespace",
engine=engine,
async_mode=True,
)
await record_manager.acreate_schema()
async with engine.connect() as connection:
session_maker = async_scoped_session(
async_sessionmaker(bind=connection),
scopefunc=current_task)
record_manager.session_factory = session_maker
pgvector.session_maker = session_maker
async with connection.begin():
loader = CSVLoader(
"data/faq/faq.csv",
source_column="source",
autodetect_encoding=True,
)
result = await aindex(
source_id_key="source",
docs_source=loader.load()[:1],
cleanup="incremental",
vector_store=pgvector,
record_manager=record_manager,
)
print(result)
asyncio.run(main())
```
---------
Signed-off-by: Rahul Tripathi <rauhl.psit.ec@gmail.com>
Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
Co-authored-by: Sean <sean@upstage.ai>
Co-authored-by: JuHyung-Son <sonju0427@gmail.com>
Co-authored-by: Erick Friis <erick@langchain.dev>
Co-authored-by: YISH <mokeyish@hotmail.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
Co-authored-by: Jason_Chen <820542443@qq.com>
Co-authored-by: Joan Fontanals <joan.fontanals.martinez@jina.ai>
Co-authored-by: Pavlo Paliychuk <pavlo.paliychuk.ca@gmail.com>
Co-authored-by: fzowl <160063452+fzowl@users.noreply.github.com>
Co-authored-by: samanhappy <samanhappy@gmail.com>
Co-authored-by: Lei Zhang <zhanglei@apache.org>
Co-authored-by: Tomaz Bratanic <bratanic.tomaz@gmail.com>
Co-authored-by: merdan <48309329+merdan-9@users.noreply.github.com>
Co-authored-by: ccurme <chester.curme@gmail.com>
Co-authored-by: Andres Algaba <andresalgaba@gmail.com>
Co-authored-by: davidefantiniIntel <115252273+davidefantiniIntel@users.noreply.github.com>
Co-authored-by: Jingpan Xiong <71321890+klaus-xiong@users.noreply.github.com>
Co-authored-by: kaka <kaka@zbyte-inc.cloud>
Co-authored-by: jingsi <jingsi@leadincloud.com>
Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
Co-authored-by: Rahul Triptahi <rahul.psit.ec@gmail.com>
Co-authored-by: Rahul Tripathi <rauhl.psit.ec@gmail.com>
Co-authored-by: Shengsheng Huang <shannie.huang@gmail.com>
Co-authored-by: Michael Schock <mjschock@users.noreply.github.com>
Co-authored-by: Anish Chakraborty <anish749@users.noreply.github.com>
Co-authored-by: am-kinetica <85610855+am-kinetica@users.noreply.github.com>
Co-authored-by: Dristy Srivastava <58721149+dristysrivastava@users.noreply.github.com>
Co-authored-by: Matt <matthew.gotteiner@microsoft.com>
Co-authored-by: William FH <13333726+hinthornw@users.noreply.github.com>
- **Description:** Fix import class name exporeted from
'playwright.async_api' and 'playwright.sync_api' to match the correct
name in playwright tool. Change import from inline guard_import to
helper function that calls guard_import to make code more readable in
gmail tool. Upgrade playwright version to 1.43.0
- **Issue:** #21354
- **Dependencies:** upgrade playwright version(this is not required for
the bugfix itself, just trying to keep dependencies fresh. I can remove
the playwright version upgrade if you want.)
