- **Description:** Some text-generation models on huggingface repeat the
prompt in their generated response, but not all do! The tests use "gpt2"
which DOES repeat the prompt and as such, the HuggingFaceHub class is
hardcoded to remove the first few characters of the response (to match
the len(prompt)). However, if you are using a model (such as the very
popular "meta-llama/Llama-2-7b-chat-hf") that DOES NOT repeat the prompt
in it's generated text, then the beginning of the generated text will be
cut off. This code change fixes that bug by first checking whether the
prompt is repeated in the generated response and removing it
conditionally.
- **Issue:** #16232
- **Dependencies:** N/A
- **Twitter handle:** N/A
This PR adds `astream_events` method to Runnables to make it easier to
stream data from arbitrary chains.
* Streaming only works properly in async right now
* One should use `astream()` with if mixing in imperative code as might
be done with tool implementations
* Astream_log has been modified with minimal additive changes, so no
breaking changes are expected
* Underlying callback code / tracing code should be refactored at some
point to handle things more consistently (OK for now)
- ~~[ ] verify event for on_retry~~ does not work until we implement
streaming for retry
- ~~[ ] Any rrenaming? Should we rename "event" to "hook"?~~
- [ ] Any other feedback from community?
- [x] throw NotImplementedError for `RunnableEach` for now
## Example
See this [Example
Notebook](dbbc7fa0d6/docs/docs/modules/agents/how_to/streaming_events.ipynb)
for an example with streaming in the context of an Agent
## Event Hooks Reference
Here is a reference table that shows some events that might be emitted
by the various Runnable objects.
Definitions for some of the Runnable are included after the table.
| event | name | chunk | input | output |
|----------------------|------------------|---------------------------------|-----------------------------------------------|-------------------------------------------------|
| on_chat_model_start | [model name] | | {"messages": [[SystemMessage,
HumanMessage]]} | |
| on_chat_model_stream | [model name] | AIMessageChunk(content="hello")
| | |
| on_chat_model_end | [model name] | | {"messages": [[SystemMessage,
HumanMessage]]} | {"generations": [...], "llm_output": None, ...} |
| on_llm_start | [model name] | | {'input': 'hello'} | |
| on_llm_stream | [model name] | 'Hello' | | |
| on_llm_end | [model name] | | 'Hello human!' |
| on_chain_start | format_docs | | | |
| on_chain_stream | format_docs | "hello world!, goodbye world!" | | |
| on_chain_end | format_docs | | [Document(...)] | "hello world!,
goodbye world!" |
| on_tool_start | some_tool | | {"x": 1, "y": "2"} | |
| on_tool_stream | some_tool | {"x": 1, "y": "2"} | | |
| on_tool_end | some_tool | | | {"x": 1, "y": "2"} |
| on_retriever_start | [retriever name] | | {"query": "hello"} | |
| on_retriever_chunk | [retriever name] | {documents: [...]} | | |
| on_retriever_end | [retriever name] | | {"query": "hello"} |
{documents: [...]} |
| on_prompt_start | [template_name] | | {"question": "hello"} | |
| on_prompt_end | [template_name] | | {"question": "hello"} |
ChatPromptValue(messages: [SystemMessage, ...]) |
Here are declarations associated with the events shown above:
`format_docs`:
```python
def format_docs(docs: List[Document]) -> str:
'''Format the docs.'''
return ", ".join([doc.page_content for doc in docs])
format_docs = RunnableLambda(format_docs)
```
`some_tool`:
```python
@tool
def some_tool(x: int, y: str) -> dict:
'''Some_tool.'''
return {"x": x, "y": y}
```
`prompt`:
```python
template = ChatPromptTemplate.from_messages(
[("system", "You are Cat Agent 007"), ("human", "{question}")]
).with_config({"run_name": "my_template", "tags": ["my_template"]})
```
<!-- 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:** In Google Vertex AI, Gemini Chat models currently
doesn't have a support for SystemMessage. This PR adds support for it
only if a user provides additional convert_system_message_to_human flag
during model initialization (in this case, SystemMessage would be
prepended to the first HumanMessage). **NOTE:** The implementation is
similar to #14824
- **Twitter handle:** rajesh_thallam
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
- **Description**: Updated doc for llm/google_vertex_ai_palm with new
functions: `invoke`, `stream`... Changed structure of the document to
match the required one.
- **Issue**: #15664
- **Dependencies**: None
- **Twitter handle**: None
---------
Co-authored-by: Jorge Zaldívar <jzaldivar@google.com>
**Description:** Gemini model has quite annoying default safety_settings
settings. In addition, current VertexAI class doesn't provide a property
to override such settings.
So, this PR aims to
- add safety_settings property to VertexAI
- fix issue with incorrect LLM output parsing when LLM responds with
appropriate 'blocked' response
- fix issue with incorrect parsing LLM output when Gemini API blocks
prompt itself as inappropriate
- add safety_settings related tests
I'm not enough familiar with langchain code base and guidelines. So, any
comments and/or suggestions are very welcome.
**Issue:** it will likely fix#14841
---------
Co-authored-by: Erick Friis <erick@langchain.dev>
* Removed some env vars not used in langchain package IT
* Added Astra DB env vars in langchain package, used for cache tests
* Added conftest.py to load env vars in langchain_community IT
* Added .env.example in langchain_community IT
<!-- 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.
-->
The timeout function comes in handy when you want to kill longrunning
queries.
The value sanitization removes all lists that are larger than 128
elements. The idea here is to remove embedding properties from results.
- **Description:** As Shell tool is very versatile, while integrating it
into applications as openai functions, developers have no clue about
what command is being executed using the ShellTool. All one can see is:
Summarising my feature request:
1. There's no visibility about what command was executed.
2. There's no mechanism to prevent a command to be executed using
ShellTool, like a y/n human input which can be accepted from user to
proceed with executing the command.,
- **Issue:** the issue #15931 it fixes if applicable,
- **Dependencies:** There isn't any dependancy,
- **Twitter handle:** @krishnashed
- **Description:** Made a small fix for the `SQLDatabase` highlighted in
an issue. The issue pertains to switching schema for different SQL
engines.
- **Issue:** #16023
@baskaryan
- **Description:** Support IN and LIKE comparators in Milvus
self-querying retriever, based on [Boolean Expression
Rules](https://milvus.io/docs/boolean.md)
- **Issue:** No
- **Dependencies:** No
- **Twitter handle:** No
Signed-off-by: ChengZi <chen.zhang@zilliz.com>
