AutoGen AGI: Advancing AI agents using AutoGen towards AGI capabilities. Explore cutting-edge enhancements in group chat dynamics, decision-making, and complex task proficiency. Join our journey in shaping AI's future!
Home Page: https://www.metamindsolutions.ai/
License: MIT License
I already install ollama (From https://github.com/jmorganca/ollama) in Ubuntu Server 18.04 LTS.
ollama list
NAME ID SIZE MODIFIED
orca-mini:latest 2dbd9f439647 2.0 GB 2 hours ago
Test with curl, It's OK.
curl http://127.0.0.1:11434/api/generate -d '{
> "model": "orca-mini",
> "prompt": "Why is the sky blue?"
> }'
{"model":"orca-mini","created_at":"2024-01-05T16:46:54.937363861Z","response":" The","done":false}
{"model":"orca-mini","created_at":"2024-01-05T16:46:54.963548218Z","response":" sky","done":false}
{"model":"orca-mini","created_at":"2024-01-05T16:46:54.98967223Z","response":" appears","done":false}
{"model":"orca-mini","created_at":"2024-01-05T16:46:55.015511833Z","response":" blue","done":false}
..........
{"model":"orca-mini","created_at":"2024-01-05T16:46:57.631165519Z","response":"","done":true,"context":[31822,13,8458,31922,3244,31871,13,3838,397,363,7421,8825,342,5243,10389,5164,828,31843,9530,362,988,362,365,473,31843,13,13,8458,31922,9779,31871,13,12056,322,266,7661,4842,31902,13,13,8458,31922,13166,31871,13,347,7661,4725,4842,1177,266,1124,906,287,260,1249,1676,6697,27554,27289,31843,1408,21062,16858,266,4556,31876,31829,7965,31844,357,19322,8634,12285,859,362,11944,291,22329,16450,31843,1872,16450,640,3304,266,1954,288,484,11468,31844,504,266,13830,4842,23893,31829,685,18752,541,4083,661,266,3002,2729,23893,31829,31843,672,1901,342,662,382,871,550,389,266,7661,31844,382,820,541,287,266,4842,23893,31829,661,266,2729,3688,31844,540,1988,266,7661,2024,4842,289,459,31843],"total_duration":4769137102,"load_duration":1973124460,"prompt_eval_count":46,"prompt_eval_duration":126935000,"eval_count":96,"eval_duration":2667461000}
Then I install autogen-agi in python 3.11 environment via Anaconda
conda create --name autogen-agi python=3.11
conda activate autogen-agi
python --version
Python 3.11.5
pip --version
pip 23.3.1 from /home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/pip (python 3.11)
Test with Autogen-AGI
In .env file
OPENAI_API_KEY=openai-api-key
GOOGLE_SEARCH_API_KEY=google-search-api-key
GOOGLE_CUSTOM_SEARCH_ENGINE_ID=google-custom-search-engine-id
GITHUB_PERSONAL_ACCESS_TOKEN=github-personal-access-token
SERP_API_KEY=serp-api-key
# Recommended engine: google or serpapi
SEARCH_ENGINE=ddg
# Uncomment below if you want to use ollama on a remote host like google collab
# See: https://www.youtube.com/watch?v=Qa1h7ygwQq8&t=329s&ab_channel=TechwithMarco
OLLAMA_HOST=http://127.0.0.1:11434
In file OAI_CONFIG_LIST.json
[
{
"model": "orca-mini",
"base_url": "http://127.0.0.1:11434/api/generate"
}
]
The I run python autogen_test.py
It has error.
llm_config_user_proxy: {'config_list': [{'model': 'orca-mini', 'base_url': 'http://127.0.0.1:11434/api/generate'}]}
llm_config_assistant: {'config_list': [{'model': 'orca-mini', 'base_url': 'http://127.0.0.1:11434/api/generate'}]}
user_proxy (to assistant):
Please execute a python script that prints 10 dad jokes.
--------------------------------------------------------------------------------
Traceback (most recent call last):
File "/ssd-disk3-data/devteam/autogen-agi/autogen_test.py", line 51, in <module>
user_proxy.initiate_chat(
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/agentchat/conversable_agent.py", line 544, in initiate_chat
self.send(self.generate_init_message(**context), recipient, silent=silent)
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/agentchat/conversable_agent.py", line 344, in send
recipient.receive(message, self, request_reply, silent)
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/agentchat/conversable_agent.py", line 475, in receive
reply = self.generate_reply(messages=self.chat_messages[sender], sender=sender)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/agentchat/conversable_agent.py", line 887, in generate_reply
final, reply = reply_func(self, messages=messages, sender=sender, config=reply_func_tuple["config"])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/agentchat/conversable_agent.py", line 619, in generate_oai_reply
response = client.create(
^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/oai/client.py", line 244, in create
response = self._completions_create(client, params)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/autogen/oai/client.py", line 314, in _completions_create
response = completions.create(**params)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/openai/_utils/_utils.py", line 299, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/openai/resources/chat/completions.py", line 556, in create
return self._post(
^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/openai/_base_client.py", line 1055, in post
return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/openai/_base_client.py", line 834, in request
return self._request(
^^^^^^^^^^^^^^
File "/home/devteam/anaconda3/envs/autogen-agi/lib/python3.11/site-packages/openai/_base_client.py", line 877, in _request
raise self._make_status_error_from_response(err.response) from None
openai.NotFoundError: 404 page not found
Please suggest how to fix this issue. Thank you.
openai.RateLimitError: Error code: 429 - {'error': {'message': 'Rate limit reached for gpt-4-1106-preview in organization org-BLAHBLAHBLAH on tokens_usage_based per day: Limit 1500000, Used 1497556, Requested 20136. Please try again in 16m59.059s. Visit https://platform.openai.com/account/rate-limits to learn more.', 'type': 'tokens_usage_based', 'param': None, 'code': 'rate_limit_exceeded'}}
Error above terminates execution. Ideal behaviour, query user on duration of sleep before resuming execution. No input means to default to 3x the time in the message, so here 45+ minutes
Hi there - first congrats on putting this together, looks like a very promising project.
I tried running example_rag with some of my own PDFs, and it seemed to run fine as far as ingesting the documents and sending the info to chatgpt - but an error occurred in whatever the next step is - specifically "OpenAIWrapper' object has no attribute 'extract_text_or_function_call':
DEBUG:openai._base_client:HTTP Request: POST https://api.openai.com/v1/chat/completions "200 OK"
ERROR:utils.rag_tools:Error in RAG fusion: 'OpenAIWrapper' object has no attribute 'extract_text_or_function_call'
Traceback (most recent call last):
File "C:\DEV\AutoGenProjects\autogen-agi\example_rag.py", line 55, in <module>
main()
File "C:\DEV\AutoGenProjects\autogen-agi\example_rag.py", line 39, in main
answer = get_informed_answer(
File "C:\DEV\AutoGenProjects\autogen-agi\utils\rag_tools.py", line 536, in get_informed_answer
nodes = get_retrieved_nodes(
File "C:\DEV\AutoGenProjects\autogen-agi\utils\rag_tools.py", line 339, in get_retrieved_nodes
query_variations = rag_fusion(query_str, query_context)
File "C:\DEV\AutoGenProjects\autogen-agi\utils\rag_tools.py", line 288, in rag_fusion
rag_fusion_response = light_gpt4_wrapper_autogen(
File "C:\DEV\AutoGenProjects\autogen-agi\utils\misc.py", line 142, in light_gpt4_wrapper_autogen
return light_gpt_wrapper_autogen(client, query, return_json, system_message)
File "C:\DEV\AutoGenProjects\autogen-agi\utils\misc.py", line 128, in light_gpt_wrapper_autogen
response = client.extract_text_or_function_call(response)
AttributeError: 'OpenAIWrapper' object has no attribute 'extract_text_or_function_call'
This could be my error as I noticed you were relying on autogen v0.2.0b4 - I actually installed v0.2.2, but there doesn't seem to be much difference in that function between 0.2.0b4 and 0.2.2. My config json file has gpt-4-1106-preview and gpt-3.5-turbo in it. I see in rag_tools you're defining LLM_CONFIGS and then referencing 'llm4' in the 'get_informed_answer' function.
Is my use of 0.2.2 possibly the issue or is it that I need to define an LLM config key/value pair named just 'gpt-4"?
Allow to specify URL to use to run it with local LLMs like Ooga Booga.
If the last call before a rate limit error or other error terminates execution was a call to the archive bot, the embeddings remain in the prompt when chat is resumed, but the agents are still waiting for the agent bot to return as completed. If the agent councel is instructed to run the bot again, the embeddings are duplicated.
Ideal behaviour: embeddings are evaluated for retention in context outside of chat history to ensure no duplication, or embedding de-duplication is performed on the chat history as part of the archive bots normal function
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.