phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

fix hf conversational task bug

Browse source
This commit is contained in:
Krrish Dholakia 2023-09-18 11:46:14 -07:00
parent 154aa83b5b
commit 5b294c704e
4 changed files with 11 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

BIN
litellm/__pycache__/main.cpython-311.pyc
View file

Binary file not shown.

BIN
litellm/__pycache__/utils.cpython-311.pyc
View file

Binary file not shown.

1
litellm/llms/huggingface_restapi.py
View file

@ -56,6 +56,7 @@ def completion(
if task == "conversational": if task == "conversational":
inference_params = copy.deepcopy(optional_params) inference_params = copy.deepcopy(optional_params)
inference_params.pop("details") inference_params.pop("details")
inference_params.pop("return_full_text")
past_user_inputs = [] past_user_inputs = []
generated_responses = [] generated_responses = []
text = "" text = ""

10
litellm/tests/test_completion.py
View file

@ -420,6 +420,16 @@ def test_completion_azure_deployment_id():
pytest.fail(f"Error occurred: {e}") pytest.fail(f"Error occurred: {e}")
# test_completion_azure_deployment_id() # test_completion_azure_deployment_id()
# def test_hf_conversational_task():
# try:
# messages = [{ "content": "There's a llama in my garden 😱 What should I do?","role": "user"}]
# # e.g. Call 'facebook/blenderbot-400M-distill' hosted on HF Inference endpoints
# response = completion(model="huggingface/facebook/blenderbot-400M-distill", messages=messages, task="conversational")
# print(f"response: {response}")
# except Exception as e:
# pytest.fail(f"Error occurred: {e}")
# test_hf_conversational_task()
# Replicate API endpoints are unstable -> throw random CUDA errors -> this means our tests can fail even if our tests weren't incorrect. # Replicate API endpoints are unstable -> throw random CUDA errors -> this means our tests can fail even if our tests weren't incorrect.
# def test_completion_replicate_llama_2(): # def test_completion_replicate_llama_2():