Community: Fuse HuggingFace Endpoint-related classes into one #17254
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
dosubot
bot
added
size:L
aymeric-roucher
force-pushed
the
master
branch
from
2dcee49
to
5b02bc6
Compare
|
@andrewrreed @Jofthomas if you want to take a look at the PR! |
aymeric-roucher
force-pushed
the
master
branch
from
edf5b20
to
a55b398
Compare
aymeric-roucher
changed the title
|
@baskaryan @efriis what do you think: should I remove the huggingface_hub.py and |
no we should leave them for backwards compatibility but mark them as deprecated. here's an example of how to do that for a class |
aymeric-roucher
force-pushed
the
master
branch
from
747272b
to
36f5074
Compare
aymeric-roucher
force-pushed
the
master
branch
from
36f5074
to
e53c11f
Compare
|
Thank you @baskaryan ! ➡️ For the release number, I just went to the root repo and did |
baskaryan
reviewed
Feb 13, 2024
libs/community/langchain_community/llms/huggingface_text_gen_inference.py
Show resolved
Hide resolved
| logger = logging.getLogger(__name__) | ||
|
|
||
|
|
||
| @deprecated("0.1.7", alternative="HuggingFaceEndppoint") |
baskaryan
approved these changes
Feb 14, 2024
| @@ -41,7 +38,7 @@ class ChatHuggingFace(BaseChatModel): | |||
| Adapted from: https://python.langchain.com/docs/integrations/chat/llama2_chat | |||
| """ | |||
|
|
|||
| llm: Union[HuggingFaceTextGenInference, HuggingFaceEndpoint, HuggingFaceHub] | |||
There was a problem hiding this comment.
this is also backwards incompatible, can we revert?
There was a problem hiding this comment.
Only problem is when I add this line back, it seems to confuse the deprecation manager:
Now running this script
from langchain_community.llms import HuggingFaceEndpoint
from langchain.chat_models import ChatHuggingFace
llm = HuggingFaceEndpoint(repo_id="mistralai/Mixtral-8x7B-Instruct-v0.1")
agent = ChatHuggingFace(llm=llm)
Triggers this error:
---------------------------------------------------------------------------
NotImplementedError Traceback (most recent call last)
[/home/ubuntu/benchmark_agents/test_frameworks.ipynb](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/test_frameworks.ipynb) Cell 1 line 5
[2](vscode-notebook-cell://ssh-remote%2Bec2/home/ubuntu/benchmark_agents/test_frameworks.ipynb#X56sdnNjb2RlLXJlbW90ZQ%3D%3D?line=1) from langchain.chat_models import ChatHuggingFace
[4](vscode-notebook-cell://ssh-remote%2Bec2/home/ubuntu/benchmark_agents/test_frameworks.ipynb#X56sdnNjb2RlLXJlbW90ZQ%3D%3D?line=3) llm = HuggingFaceEndpoint(repo_id="mistralai/Mixtral-8x7B-Instruct-v0.1")
----> [5](vscode-notebook-cell://ssh-remote%2Bec2/home/ubuntu/benchmark_agents/test_frameworks.ipynb#X56sdnNjb2RlLXJlbW90ZQ%3D%3D?line=4) agent = ChatHuggingFace(llm=llm)
File [~/langchain_fuse_hf_endpoints/libs/community/langchain_community/chat_models/huggingface.py:50](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/langchain_fuse_hf_endpoints/libs/community/langchain_community/chat_models/huggingface.py:50), in ChatHuggingFace.__init__(self, **kwargs)
[49](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/langchain_fuse_hf_endpoints/libs/community/langchain_community/chat_models/huggingface.py:49) def __init__(self, **kwargs: Any):
---> [50](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/langchain_fuse_hf_endpoints/libs/community/langchain_community/chat_models/huggingface.py:50) super().__init__(**kwargs)
[52](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/langchain_fuse_hf_endpoints/libs/community/langchain_community/chat_models/huggingface.py:52) from transformers import AutoTokenizer
[54](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/langchain_fuse_hf_endpoints/libs/community/langchain_community/chat_models/huggingface.py:54) self._resolve_model_id()
File [~/venv/agents/lib/python3.10/site-packages/langchain_core/load/serializable.py:107](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/load/serializable.py:107), in Serializable.__init__(self, **kwargs)
[106](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/load/serializable.py:106) def __init__(self, **kwargs: Any) -> None:
--> [107](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/load/serializable.py:107) super().__init__(**kwargs)
[108](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/load/serializable.py:108) self._lc_kwargs = kwargs
File [~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:339](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:339), in BaseModel.__init__(__pydantic_self__, **data)
[333](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:333) """
[334](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:334) Create a new model by parsing and validating input data from keyword arguments.
[335](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:335)
[336](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:336) Raises ValidationError if the input data cannot be parsed to form a valid model.
[337](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:337) """
[338](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/pydantic/v1/main.py:338) # Uses something other than `self` the first arg to allow "self" as a settable attribute
...
[339](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/_api/deprecation.py:339) )
[340](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/_api/deprecation.py:340) else:
[341](https://vscode-remote+ssh-002dremote-002bec2.vscode-resource.vscode-cdn.net/home/ubuntu/benchmark_agents/~/venv/agents/lib/python3.10/site-packages/langchain_core/_api/deprecation.py:341) removal = f"in {removal}"
NotImplementedError: Need to determine which default deprecation schedule to use. within ?? minor releases
There was a problem hiding this comment.
The error seems to be triggered here and seems due to the fact that I have not setup neither pending nor removal argument.
@baskaryan : do you know the proper configuration I should use in my @deprecated decorator?
There was a problem hiding this comment.
ah yea let's add a removal argument, can set removal="0.2.0"
dosubot
bot
added
the
lgtm
aymeric-roucher
force-pushed
the
master
branch
from
2eaead5
to
0c6411e
Compare
This was referenced Feb 15, 2024
|
@baskaryan seems like we're good to merge? |
|
@baskaryan thanks a lot for this review! I added follow-up on a very strange issue here: the order of the type hints in |
haydeniw
pushed a commit
to haydeniw/langchain
that referenced
this pull request
Feb 27, 2024
…ain-ai#17254) ## Description Fuse HuggingFace Endpoint-related classes into one: - [HuggingFaceHub](https://github.com/langchain-ai/langchain/blob/5ceaf784f324064b868a3cfed3fab7554173e7b3/libs/community/langchain_community/llms/huggingface_hub.py) - [HuggingFaceTextGenInference](https://github.com/langchain-ai/langchain/blob/5ceaf784f324064b868a3cfed3fab7554173e7b3/libs/community/langchain_community/llms/huggingface_text_gen_inference.py) - and [HuggingFaceEndpoint](https://github.com/langchain-ai/langchain/blob/5ceaf784f324064b868a3cfed3fab7554173e7b3/libs/community/langchain_community/llms/huggingface_endpoint.py) Are fused into - HuggingFaceEndpoint ## Issue The deduplication of classes was creating a lack of clarity, and additional effort to develop classes leads to issues like [this hack](https://github.com/langchain-ai/langchain/blob/5ceaf784f324064b868a3cfed3fab7554173e7b3/libs/community/langchain_community/llms/huggingface_endpoint.py#L159). ## Dependancies None, this removes dependancies. ## Twitter handle If you want to post about this: @AymericRoucher --------- Co-authored-by: Bagatur <baskaryan@gmail.com>
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Fuse HuggingFace Endpoint-related classes into one:
Are fused into
Issue
The deduplication of classes was creating a lack of clarity, and additional effort to develop classes leads to issues like this hack.
Dependancies
None, this removes dependancies.
Twitter handle
If you want to post about this: @AymericRoucher