Could not reach the server when using Macbook air M2.

[antoineross]

I am using AutoGen and trying to implement it with chainlit, however it seems like the UI won't update after the agents send a message. Is this a mac issue or is my code a problem? I am using chainlit ver. 0.7.3 and here is my code:

import openai import json, ast import os import chainlit as cl import autogen

MAX_ITER = 5

@cl.on_chat_start async def setup_agent(): config_list = autogen.config_list_from_json( "OAI_CONFIG_LIST") llm_config={ "request_timeout": 600, "seed": 42, "config_list": config_list, "temperature": 0, }

create an AssistantAgent named "assistant"

assistant = autogen.AssistantAgent(
                name="assistant",
                is_termination_msg=lambda x: x.get("content", "").rstrip().endswith("TERMINATE") or x.get("content", "").strip() == "",
                llm_config={
                    "seed": 42,  # seed for caching and reproducibility
                    "config_list": config_list,  # a list of OpenAI API configurations
                    "temperature": 0,  # temperature for sampling
                },  # configuration for autogen's enhanced inference API which is compatible with OpenAI API
                )
cl.user_session.set('assistant', assistant)
# create a UserProxyAgent instance named "user_proxy"
user_proxy = autogen.UserProxyAgent(
name="user_proxy",
human_input_mode="TERMINATE",
max_consecutive_auto_reply=10,
is_termination_msg=lambda x: x.get("content", "").rstrip().endswith("TERMINATE"),
code_execution_config={"work_dir": "web"},
llm_config=llm_config,
system_message="""Reply TERMINATE if the task has been solved at full satisfaction.
Otherwise, reply CONTINUE, or the reason why the task is not solved yet."""
)
cl.user_session.set('user_proxy', user_proxy)
await cl.Message(content=f"Start chatting with code interpreter").send()

@cl.on_message async def run_conversation(msg: cl.Message):

check if user message changed

user_message = msg.content
if user_message == cl.user_session.get('user_message'):
    return
assistant = cl.user_session.get('assistant')
user_proxy = cl.user_session.get('user_proxy')
cur_iter = 0
if msg.elements:
    for element in msg.elements:
        file_name = element.name
        content = element.content
        # If want to show content Content: {content.decode('utf-8')}\n\n
        await cl.Message(content=f"Uploaded file: {file_name}\n").send()

        # Save the file locally
        with open(os.path.join("coding", file_name), "wb") as file:
            file.write(content)
        user_proxy.send(
            recipient=assistant,
            message=f"User uploaded file: {file_name}",
        )
print('CONVERSATION')

while cur_iter < MAX_ITER:
    if len(assistant.chat_messages[user_proxy]) == 0 :
        print('initiating chat')
        user_proxy.initiate_chat(
            assistant,
            message=user_message,
        )
    else:
        print('FOLLOW up message')
        # followup of the previous question
        user_proxy.send(
            recipient=assistant,
            message=user_message,
        )
    message_history = assistant.chat_messages[user_proxy]
    last_seen_message_index = cl.user_session.get('last_seen_message_index', 0)
    print("message hirsotry", message_history)
    for message in message_history[last_seen_message_index+1:0]:
        await cl.Message(author=message["role"], content=message["content"]).send()
    cl.user_session.set('last_seen_message_index', len(message_history))
    cur_iter += 1
    return

[willydouhard]

Is this fixed based on the cookbook example?

[ruifengma]

for me, I run on the windows laptop and the cookbook example only run the initial once and after that, there is no reaction no matter what I did. What if I only want to type the task through the UI instead of using the initial task, for example, if I wish to deploy the service, what should I do?

[antoineross]

Is this fixed based on the cookbook example?

Yes it's fixed. Thanks!