search

MadcowD / ell

A language model programming library.
http://docs.ell.so/
MIT License
5.29k stars 310 forks source link

I try to run the get start example i got this error #261

Closed ParseDark closed 1 month ago

ParseDark commented 1 month ago 
>>> greeting = hello("Sam Altman")
╔═════════════════════════════════════════════════════════════════════════════════╗
║ hello(Sam Altm..)
╠═════════════════════════════════════════════════════════════════════════════════╣
║ Prompt:
╟─────────────────────────────────────────────────────────────────────────────────╢
│      system: You are a helpful assistant.
│
│        user: Say hello to Sam Altman!
╟─────────────────────────────────────────────────────────────────────────────────╢
║ Output:
╟─────────────────────────────────────────────────────────────────────────────────╢
│   assistant: Got it! How can I assist you today?Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/lmp/_track.py", line 64, in tracked_func
    return func_to_track(*fn_args, **fn_kwargs, _invocation_origin=invocation_id)[0]
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/lmp/complex.py", line 68, in model_call
    (result, final_api_params, metadata) = provider.call(ell_call, origin_id=_invocation_origin, logger=_logger if should_log else None)
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/provider.py", line 127, in call
    messages, metadata = self.translate_from_provider(
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/providers/openai.py", line 126, in translate_from_provider
    Message(role=role,
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/types/message.py", line 319, in __init__
    super().__init__(content=content, role=role)
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/pydantic/main.py", line 212, in __init__
    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for Message
role
  Input should be a valid string [type=string_type, input_value=None, input_type=NoneType]
    For further information visit https://errors.pydantic.dev/2.9/v/string_type
ParseDark commented 1 month ago

Code: 

import ell

@ell.simple(model="gpt-4o")
def hello(name: str):
    """You are a helpful assistant.""" # System prompt
    return f"Say hello to {name}!" # User prompt

greeting = hello("Sam Altman")
print(greeting)
MadcowD commented 1 month ago

What version of ell are you on? Can you pip install --upgrade ell.

ParseDark commented 1 month ago

What version of ell are you on? Can you pip install --upgrade ell.

thanks, ell dev team. Currently my ell-ai version is 0.0.12

# pip list | grep ell
ell-ai             0.0.12

So I try to re-install 

pip install ell-ai
# Successfully installed ell-ai-0.0.12

I re-run the Hello World example, but I still get this error.:

   1   │ import ell
   2   │ import logging
   3   │ from typing import List
   4   │ 
   5   │ # Configure logging
   6   │ logging.basicConfig(level=logging.DEBUG)
   7   │ 
   8   │ @ell.simple(model="gpt-4o")
   9   │ def hello(name: str) -> str:
  10   │     """You are a helpful assistant."""  # System prompt
  11   │     try:
  12   │         greeting = f"Say hello to {name}!"  # User prompt
  13   │         logging.debug(f"Greeting generated for {name}: {greeting}")
  14   │         return greeting
  15   │     except Exception as e:
  16   │         logging.error(f"Error generating greeting for {name}: {e}")
  17   │         return "Error generating greeting."
  18   │ 
  19   │ def greet_multiple(names: List[str]) -> List[str]:
  20   │     """Greet multiple people."""
  21   │     greetings = []
  22   │     for name in names:
  23   │         greeting = hello(name)
  24   │         greetings.append(greeting)
  25   │     logging.debug(f"All greetings generated: {greetings}")
  26   │     return greetings
  27   │ 
  28   │ # Example usage of the new function
  29   │ if __name__ == "__main__":
  30   │     greeting = hello("Sam Altman")
  31   │     print(greeting)  # Debug information
  32   │     greetings_list = greet_multiple(["Alice", "Bob", "Charlie"])
  33   │     print(greetings_list)  # Debug information

DEBUG:httpcore.connection:connect_tcp.started host='127.0.0.1' port=8083 local_address=None timeout=5.0 socket_options=None
DEBUG:httpcore.connection:connect_tcp.complete return_value=<httpcore._backends.sync.SyncStream object at 0x7f7df8010460>
DEBUG:httpcore.http11:send_request_headers.started request=<Request [b'CONNECT']>
DEBUG:httpcore.http11:send_request_headers.complete
DEBUG:httpcore.http11:send_request_body.started request=<Request [b'CONNECT']>
DEBUG:httpcore.http11:send_request_body.complete
DEBUG:httpcore.http11:receive_response_headers.started request=<Request [b'CONNECT']>
DEBUG:httpcore.http11:receive_response_headers.complete return_value=(b'HTTP/1.1', 200, b'Connection established', [])
DEBUG:httpcore.proxy:start_tls.started ssl_context=<ssl.SSLContext object at 0x7f7e0b042cc0> server_hostname='api.openai.com' timeout=5.0
DEBUG:httpcore.proxy:start_tls.complete return_value=<httpcore._backends.sync.SyncStream object at 0x7f7df80106a0>
DEBUG:httpcore.http11:send_request_headers.started request=<Request [b'POST']>
DEBUG:httpcore.http11:send_request_headers.complete
DEBUG:httpcore.http11:send_request_body.started request=<Request [b'POST']>
DEBUG:httpcore.http11:send_request_body.complete
DEBUG:httpcore.http11:receive_response_headers.started request=<Request [b'POST']>
DEBUG:httpcore.http11:receive_response_headers.complete return_value=(b'HTTP/1.1', 200, b'OK', [(b'Date', b'Sun, 29 Sep 2024 05:13:19 GMT'), (b'Content-Type', b'text/event-stream'), (b'Transfer-Encoding', b'chunked'), (b'Connection', b'keep-alive'), (b'Cache-Control', b'no-cache'), (b'Cache-Control', b'no-cache'), (b'X-Oneapi-Request-Id', b'202409291313166292105581082218'), (b'CF-Cache-Status', b'DYNAMIC'), (b'Report-To', b'{"endpoints":[{"url":"https:\\/\\/a.nel.cloudflare.com\\/report\\/v4?s=rLzSbiz%2BexC5ZYfdJM2s8%2FsyMxXpKjPV6dKqSsMGnv%2BGWQOJt83nGqbwI9v3cTEzINjRvtYGwzsjblXUaO5grClnOeoQbJMtHXLZLV679e50HXSYkZ08UV7vOOrl2%2B%2F8"}],"group":"cf-nel","max_age":604800}'), (b'NEL', b'{"success_fraction":0,"report_to":"cf-nel","max_age":604800}'), (b'Strict-Transport-Security', b'max-age=31536000; includeSubDomains; preload'), (b'X-Content-Type-Options', b'nosniff'), (b'Server', b'cloudflare'), (b'CF-RAY', b'8ca97ba2bf16a343-TPE')])
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
DEBUG:openai._base_client:HTTP Response: POST https://api.openai.com/v1/chat/completions "200 OK" Headers([('date', 'Sun, 29 Sep 2024 05:13:19 GMT'), ('content-type', 'text/event-stream'), ('transfer-encoding', 'chunked'), ('connection', 'keep-alive'), ('cache-control', 'no-cache'), ('cache-control', 'no-cache'), ('x-oneapi-request-id', '202409291313166292105581082218'), ('cf-cache-status', 'DYNAMIC'), ('report-to', '{"endpoints":[{"url":"https:\\/\\/a.nel.cloudflare.com\\/report\\/v4?s=rLzSbiz%2BexC5ZYfdJM2s8%2FsyMxXpKjPV6dKqSsMGnv%2BGWQOJt83nGqbwI9v3cTEzINjRvtYGwzsjblXUaO5grClnOeoQbJMtHXLZLV679e50HXSYkZ08UV7vOOrl2%2B%2F8"}],"group":"cf-nel","max_age":604800}'), ('nel', '{"success_fraction":0,"report_to":"cf-nel","max_age":604800}'), ('strict-transport-security', 'max-age=31536000; includeSubDomains; preload'), ('x-content-type-options', 'nosniff'), ('server', 'cloudflare'), ('cf-ray', '8ca97ba2bf16a343-TPE')])
DEBUG:openai._base_client:request_id: None
DEBUG:httpcore.http11:receive_response_body.started request=<Request [b'POST']>
DEBUG:httpcore.http11:receive_response_body.complete
DEBUG:httpcore.http11:response_closed.started
DEBUG:httpcore.http11:response_closed.complete
Traceback (most recent call last):
  File "/Volumes/workspace/SWorkspace/ell_dir/main.py", line 30, in <module>
    greeting = hello("Sam Altman")
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/lmp/_track.py", line 64, in tracked_func
    return func_to_track(*fn_args, **fn_kwargs, _invocation_origin=invocation_id)[0]
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/lmp/complex.py", line 68, in model_call
    (result, final_api_params, metadata) = provider.call(ell_call, origin_id=_invocation_origin, logger=_logger if should_log else None)
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/provider.py", line 127, in call
    messages, metadata = self.translate_from_provider(
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/providers/openai.py", line 126, in translate_from_provider
    Message(role=role,
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/ell/types/message.py", line 319, in __init__
    super().__init__(content=content, role=role)
  File "/Users/hawei/miniconda3/envs/ell/lib/python3.10/site-packages/pydantic/main.py", line 212, in __init__
    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for Message
role
  Input should be a valid string [type=string_type, input_value=None, input_type=NoneType]
    For further information visit https://errors.pydantic.dev/2.9/v/string_type

response:

image
MadcowD commented 1 month ago

OH! What is "gptapi.us"?

ParseDark commented 1 month ago

OH! What is "gptapi.us"?

Thanks, @MadcowD . well. you know i am from China. if follow the policy of the openai. we can't use the openai service. But my country also has a third-party company that does the proxy things. basically is creating an outset server to do the proxy. I will keep trying another tool. Thanks for your help.

MadcowD commented 1 month ago

I think o can still fix this for you! It just the proxy forgot to send you a role :)

ParseDark commented 1 month ago

I think o can still fix this for you! It just the proxy forgot to send you a role :)

Oooooooi!!! thanks I also think proxy server didn't send the role field is the root cause. Here is what i change on the source code.

image

openai client call on here

image

So my mind is if the calling part have one condition for the None. Maybe in the Message type part we should be handle this condition. Should i create a PR for this condition? I don't want to offend, I just want to discuss whether this situation needs to be handled, because the proxy is not part of openai, but I really want to use this library haha