rlancemartin / auto-evaluator Goto Github PK

Use langchain-serve to expose auto-evaluator as an API to enable integration with external services.

Currently SVM retriever has a create_index def that will embed a list of chunks:

def create_index(contexts: List[str], embeddings: Embeddings) -> np.ndarray:
    return np.array([embeddings.embed_query(split) for split in contexts])

See code:
https://github.com/hwchase17/langchain/blob/f1d15b4a75238eb56e386608b7f631a853a803e0/langchain/retrievers/svm.py#L17

This is quite slow (> 10 min on ~2k splits).

Tried multiprocessing, but appears to hang w/:

Process SpawnPoolWorker-74:
Traceback (most recent call last):
  File "/Users/31treehaus/opt/anaconda3/envs/ml/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
    self.run()
  File "/Users/31treehaus/opt/anaconda3/envs/ml/lib/python3.9/multiprocessing/process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "/Users/31treehaus/opt/anaconda3/envs/ml/lib/python3.9/multiprocessing/pool.py", line 114, in worker
    task = get()
  File "/Users/31treehaus/opt/anaconda3/envs/ml/lib/python3.9/multiprocessing/queues.py", line 367, in get
    return _ForkingPickler.loads(res)
AttributeError: Can't get attribute 'embed_query' on <module '__main__' (built-in)>

Tried a dataframe apply to parallelize, but no latency improvement.

Need to carve out more time to look into this.

https://huggingface.co/spaces

This way its easy to use for anyone. Happy to help if you are interested and have any questions!

Add llama-index evaluation:
https://gpt-index.readthedocs.io/en/latest/how_to/evaluation/evaluation.html

Traceback (most recent call last):
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 565, in _run_script
exec(code, module.dict)
File "/Users/helen/code/xiaomi/auto-evaluator/auto-evaluator.py", line 430, in
graded_answers, graded_retrieval, latency, predictions = run_evaluation(qa_chain, retriever, eval_set, grade_prompt,
File "/Users/helen/code/xiaomi/auto-evaluator/auto-evaluator.py", line 342, in run_evaluation
retrieval_grade = grade_model_retrieval(gt_dataset, retrieved_docs, grade_prompt)
File "/Users/helen/code/xiaomi/auto-evaluator/auto-evaluator.py", line 277, in grade_model_retrieval
graded_outputs = eval_chain.evaluate(
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/evaluation/qa/eval_chain.py", line 60, in evaluate
return self.apply(inputs)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chains/llm.py", line 118, in apply
response = self.generate(input_list)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chains/llm.py", line 62, in generate
return self.llm.generate_prompt(prompts, stop)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/base.py", line 82, in generate_prompt
raise e
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/base.py", line 79, in generate_prompt
output = self.generate(prompt_messages, stop=stop)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/base.py", line 54, in generate
results = [self._generate(m, stop=stop) for m in messages]
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/base.py", line 54, in
results = [self._generate(m, stop=stop) for m in messages]
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/openai.py", line 266, in _generate
response = self.completion_with_retry(messages=message_dicts, **params)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/openai.py", line 228, in completion_with_retry
return _completion_with_retry(**kwargs)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/tenacity/init.py", line 289, in wrapped_f
return self(f, *args, **kw)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/tenacity/init.py", line 379, in call
do = self.iter(retry_state=retry_state)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/tenacity/init.py", line 314, in iter
return fut.result()
File "/usr/local/Cellar/[email protected]/3.9.12/Frameworks/Python.framework/Versions/3.9/lib/python3.9/concurrent/futures/_base.py", line 439, in result
return self.__get_result()
File "/usr/local/Cellar/[email protected]/3.9.12/Frameworks/Python.framework/Versions/3.9/lib/python3.9/concurrent/futures/_base.py", line 391, in __get_result
raise self._exception
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/tenacity/init.py", line 382, in call
result = fn(*args, **kwargs)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/langchain/chat_models/openai.py", line 226, in _completion_with_retry
return self.client.create(**kwargs)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/openai/api_resources/chat_completion.py", line 25, in create
return super().create(*args, **kwargs)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/openai/api_resources/abstract/engine_api_resource.py", line 153, in create
response, _, api_key = requestor.request(
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/openai/api_requestor.py", line 226, in request
resp, got_stream = self._interpret_response(result, stream)
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/openai/api_requestor.py", line 619, in _interpret_response
self._interpret_response_line(
File "/Users/helen/code/xiaomi/auto-evaluator/venv/lib/python3.9/site-packages/openai/api_requestor.py", line 679, in _interpret_response_line
raise self.handle_error_response(
openai.error.InvalidRequestError: This model's maximum context length is 4097 tokens. However, your messages resulted in 4280 tokens. Please reduce the length of the messages.

I would really love to see how the open source embedding instruct models compare to the top from openai:

• facebook/tart-full-flan-t5-xl · Hugging Face
• hkunlp/instructor-xl · Hugging Face

Any chance you could add support?

ValidationError: 1 validation error for ChatOpenAI root Did not find openai_api_key, please add an environment variable OPENAI_API_KEY which contains it, or pass openai_api_key as a named parameter. (type=value_error)

Traceback:
File "/Users/yujiedou/Desktop/auto-evaluator-main/venv/lib/python3.9/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 565, in _run_script
exec(code, module.dict)
File "/Users/yujiedou/Desktop/auto-evaluator-main/auto-evaluator.py", line 302, in
eval_set = generate_eval(text, num_eval_questions, 3000)
File "/Users/yujiedou/Desktop/auto-evaluator-main/venv/lib/python3.9/site-packages/streamlit/runtime/caching/cache_utils.py", line 194, in wrapper
return cached_func(*args, **kwargs)
File "/Users/yujiedou/Desktop/auto-evaluator-main/venv/lib/python3.9/site-packages/streamlit/runtime/caching/cache_utils.py", line 223, in call
return self._get_or_create_cached_value(args, kwargs)
File "/Users/yujiedou/Desktop/auto-evaluator-main/venv/lib/python3.9/site-packages/streamlit/runtime/caching/cache_utils.py", line 248, in _get_or_create_cached_value
return self._handle_cache_miss(cache, value_key, func_args, func_kwargs)
File "/Users/yujiedou/Desktop/auto-evaluator-main/venv/lib/python3.9/site-packages/streamlit/runtime/caching/cache_utils.py", line 302, in _handle_cache_miss
computed_value = self._info.func(*func_args, **func_kwargs)
File "/Users/yujiedou/Desktop/auto-evaluator-main/auto-evaluator.py", line 81, in generate_eval
chain = QAGenerationChain.from_llm(ChatOpenAI(temperature=0, openai_api_key=OPENAI_API_KEY))
File "pydantic/main.py", line 341, in pydantic.main.BaseModel.init