- New pin-to-side (button). This functionality allows you to search the
docs while asking the AI for questions
- Fixed the search bar in Firefox that won't detect a mouse click
- Fixes and improvements overall in the model's performance
Description: Added debugging output in DirectoryLoader to identify the
file being processed.
Issue: [Need a trace or debug feature in Lanchain DirectoryLoader
#7725](https://github.com/hwchase17/langchain/issues/7725)
Dependencies: No additional dependencies are required.
Tag maintainer: @rlancemartin, @eyurtsev
This PR enhances the DirectoryLoader with debugging output to help
diagnose issues when loading documents. This new feature does not add
any dependencies and has been tested on a local machine.
Inspired by #5550, I implemented full async API support in Qdrant. The
docs were extended to mention the existence of asynchronous operations
in Langchain. I also used that chance to restructure the tests of Qdrant
and provided a suite of tests for the async version. Async API requires
the GRPC protocol to be enabled. Thus, it doesn't work on local mode
yet, but we're considering including the support to be consistent.
<!-- Thank you for contributing to LangChain!
Replace this 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!
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.
Maintainer responsibilities:
- General / Misc / if you don't know who to tag: @baskaryan
- DataLoaders / VectorStores / Retrievers: @rlancemartin, @eyurtsev
- Models / Prompts: @hwchase17, @baskaryan
- Memory: @hwchase17
- Agents / Tools / Toolkits: @hinthornw
- Tracing / Callbacks: @agola11
- Async: @agola11
If no one reviews your PR within a few days, feel free to @-mention the
same people again.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/hwchase17/langchain/blob/master/.github/CONTRIBUTING.md
-->
Integrate [Rockset](https://rockset.com/docs/) as a document loader.
Issue: None
Dependencies: Nothing new (rockset's dependency was already added
[here](https://github.com/hwchase17/langchain/pull/6216))
Tag maintainer: @rlancemartin
I have added a test for the integration and an example notebook showing
its use. I ran `make lint` and everything looks good.
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
This pull request adds a ElasticsearchDatabaseChain chain for
interacting with analytics database, in the manner of the
SQLDatabaseChain.
Maintainer: @samber
Twitter handler: samuelberthe
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
- Description: This allows passing auth objects in request wrappers.
Currently, we can handle auth by editing headers in the
RequestsWrappers, but more complex auth methods, such as Kerberos, could
be handled better by using existing functionality within the requests
library. There are many authentication options supported both natively
and by extensions, such as requests-kerberos or requests-ntlm.
- Issue: Fixes#7542
- Dependencies: none
Co-authored-by: eric.speidel@de.bosch.com <eric.speidel@de.bosch.com>
<!-- Thank you for contributing to LangChain!
Replace this 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!
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.
Maintainer responsibilities:
- General / Misc / if you don't know who to tag: @baskaryan
- DataLoaders / VectorStores / Retrievers: @rlancemartin, @eyurtsev
- Models / Prompts: @hwchase17, @baskaryan
- Memory: @hwchase17
- Agents / Tools / Toolkits: @hinthornw
- Tracing / Callbacks: @agola11
- Async: @agola11
If no one reviews your PR within a few days, feel free to @-mention the
same people again.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/hwchase17/langchain/blob/master/.github/CONTRIBUTING.md
-->
- Add langchain.llms.Tonyi for text completion, in examples into the
Tonyi Text API,
- Add system tests.
Note async completion for the Text API is not yet supported and will be
included in a future PR.
Dependencies: dashscope. It will be installed manually cause it is not
need by everyone.
Happy for feedback on any aspect of this PR @hwchase17 @baskaryan.
Multiple people have asked in #5081 for a way to limit the documents
returned from an AzureCognitiveSearchRetriever. This PR adds the `top_n`
parameter to allow that.
Twitter handle:
[@UmerHAdil](twitter.com/umerHAdil)
Fix for Serializable class to include name, used in FileCallbackHandler
as same issue #7524
Description: Fixes the Serializable class to include 'name' attribute
(class_name) in the dict created,
This is used in Callbacks, specifically the StdOutCallbackHandler,
FileCallbackHandler.
Issue: As described in issue #7524
Dependencies: None
Tag maintainer: SInce this is related to the callback module, tagging
@agola11 @idoru
Comments:
Glad to see issue #7524 fixed in pull #6124, but you forget to change
the same place in FileCallbackHandler
When a custom Embeddings object is set, embed all given texts in a batch
instead of passing them through individually. Any code calling add_texts
can then appropriately size the chunks of texts that are passed through
to take full advantage of the hardware it's running on.
Fixes#6198
ElasticKnnSearch.from_texts is actually ElasticVectorSearch.from_texts
and throws because it calls ElasticKnnSearch constructor with the wrong
arguments.
Now ElasticKnnSearch has its own from_texts, which constructs a proper
ElasticKnnSearch.
---------
Co-authored-by: Charles Parker <charlesparker@FiltaMacbook.local>
Co-authored-by: Bagatur <baskaryan@gmail.com>
<!-- Thank you for contributing to LangChain!
Replace this comment with:
- Description:
- 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!
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.
Maintainer responsibilities:
- General / Misc / if you don't know who to tag: @baskaryan
- DataLoaders / VectorStores / Retrievers: @rlancemartin, @eyurtsev
- Models / Prompts: @hwchase17, @baskaryan
- Memory: @hwchase17
- Agents / Tools / Toolkits: @hinthornw
- Tracing / Callbacks: @agola11
- Async: @agola11
If no one reviews your PR within a few days, feel free to @-mention the
same people again.
See contribution guidelines for more information on how to write/run
tests, lint, etc:
https://github.com/hwchase17/langchain/blob/master/.github/CONTRIBUTING.md
-->
## Description
This PR addresses a bug in the RecursiveUrlLoader class where absolute
URLs were being treated as relative URLs, causing malformed URLs to be
produced. The fix involves using the urljoin function from the
urllib.parse module to correctly handle both absolute and relative URLs.
@rlancemartin @eyurtsev
---------
Co-authored-by: Lance Martin <lance@langchain.dev>
Fixes # (issue)
The existing PlaywrightURLLoader load() function uses a synchronous
browser which is not compatible with jupyter.
This PR adds a sister function aload() which can be run insisde a
notebook.
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Mainline the tracer to avoid calling feedback before run is posted.
Chose a bool over `max_workers` arg for configuring since we don't want
to support > 1 for now anyway. At some point may want to manage the pool
ourselves (ordering only really matters within a run and with parent
runs)
# Browserless
Added support for Browserless' `/content` endpoint as a document loader.
### About Browserless
Browserless is a cloud service that provides access to headless Chrome
browsers via a REST API. It allows developers to automate Chromium in a
serverless fashion without having to configure and maintain their own
Chrome infrastructure.
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Co-authored-by: Lance Martin <lance@langchain.dev>
With AzureOpenAI openai_api_type defaulted to "azure" the logic in
utils' get_from_dict_or_env() function triggered by the root validator
never looks to environment for the user's runtime openai_api_type
values. This inhibits folks using token-based auth, or really any auth
model other than "azure."
By removing the "default" value, this allows environment variables to be
pulled at runtime for the openai_api_type and thus enables the other
api_types which are expected to work.
---------
Co-authored-by: Ebo <mebstyne@microsoft.com>
Co-authored-by: Dev 2049 <dev.dev2049@gmail.com>
This PR is aimed at enhancing the clarity of the documentation in the
langchain project.
**Description**:
In the graphql.ipynb file, I have removed the unnecessary 'llm' argument
from the initialization process of the GraphQL tool (of type
_EXTRA_OPTIONAL_TOOLS). The 'llm' argument is not required for this
process. Its presence could potentially confuse users. This modification
simplifies the understanding of tool initialization and minimizes
potential confusion.
**Issue**: Not applicable, as this is a documentation improvement.
**Dependencies**: None.
**I kindly request a review from the following maintainer**: @hinthornw,
who is responsible for Agents / Tools / Toolkits.
No new integration is being added in this PR, hence no need for a test
or an example notebook.
Please see the changes for more detail and let me know if any further
modification is necessary.
- Migrate from deprecated langchainplus_sdk to `langsmith` package
- Update the `run_on_dataset()` API to use an eval config
- Update a number of evaluators, as well as the loading logic
- Update docstrings / reference docs
- Update tracer to share single HTTP session
Retrieves the name of the class from new location as of commit
18af149e91
Co-authored-by: Zander Chase <130414180+vowelparrot@users.noreply.github.com>
Sometimes the score responded by chatgpt would be like 'Respone
example\nScore: 90 (fully answers the question, but could provide more
detail on the specific error message)'
For the score contains not only numbers, it raise a ValueError like
Update the RegexParser from `.*` to `\d*` would help us to ignore the
text after number.
Co-authored-by: Bagatur <baskaryan@gmail.com>
Fixed#6768.
This is a workaround only. I think a better longer-term solution is for
chains to declare how many input variables they *actually* need (as
opposed to ones that are in the prompt, where some may be satisfied by
the memory). Then, a wrapping chain can check the input match against
the actual input variables.
@hwchase17
Added fix to avoid irrelevant attributes being returned plus an example
of extracting unrelated entities and an exampe of using an 'extra_info'
attribute to extract unstructured data for an entity.
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
Added an option to trim intermediate steps to last N steps. This is
especially useful for long-running agents. Users can explicitly specify
N or provide a function that does custom trimming/manipulation on
intermediate steps. I've mimicked the API of the `handle_parsing_errors`
parameter.
