#_openai.py

an interface to give DEFAULT_CREATE_COMPLETION_URL value

DEFAULT_CREATE_COMPLETION_URL = "http://0.0.0.0:4000"

OpenAI(
api_key=self.openai_api_key,
base_url=DEFAULT_CREATE_COMPLETION_URL
)

and disable model model availability check in this mode.

The teacher_runtimes will call OpenAIRuntime(model='gpt-4') all the time when we new a Agent instance like the example in Quickstart:

The problem is that when the user doesn't have access to GPT4, they can not create any Agent for now. The raised errors like:

Requested model "gpt-4" is not available in your OpenAI account. Available models are:
...

Try to change the runtime settings for OpenAIRuntime, for example:

OpenAIRuntime(..., model="gpt-3.5-turbo")
...

---> 51     raise ValidationError(f'Requested model {self.gpt_model_name} is not available in your OpenAI account.')

TypeError: No constructor defined

Currently, Adala requires openai==0.28.1
The latest version available is 1.9.0.
Is there a plan to update openai depedencies?

Adala's and latest's openai libs version are not compatible e.g. below error occurs:
"AttributeError: module 'openai' has no attribute 'api_base'. Did you mean: 'api_type'?"

Is it possible add parameter to use claude3 and Gemini
or is possible to support litellm as proxy

I meeting this error when to run examples/classification_skill_with_CoT.ipynb

import pandas as pd
df = pd.DataFrame([
{"text": "Apple product with a sleek design.", "category": "Electronics"},
{"text": "Laptop stand for the kitchen.", "category": "Furniture/Home Decor"},
{"text": "Chocolate leather boots.", "category": "Footwear/Clothing"},
{"text": "Wooden cream for surfaces.", "category": "Furniture/Home Decor"},
{"text": "Natural finish for your lips.", "category": "Beauty/Personal Care"}
])
df

from adala.agents import Agent
from adala.skills import ClassificationSkill

agent = Agent(
skills=ClassificationSkill(
name='product_category_classification',
instructions='Categorize text into different product categories.',
input_template='Text: {text}',
output_template='Thoughts: {thoughts}\n\nClassification: {predicted_class}',
labels={'predicted_class': [
"Footwear/Clothing",
"Electronics",
"Food/Beverages",
"Furniture/Home Decor",
"Beauty/Personal Care"
]},
)
)

agent.run(df)

Applying skill: product_category_classification
60%|████████████████████████████████████ | 3/5 [00:05<00:04, 2.04s/it]
Traceback (most recent call last):
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 109, in run
await self.visit(self.parse_tree, VariableStack([self.program._variables], self))
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 559, in visit
visited_children.append(await self.visit(child, variable_stack, inner_next_node, inner_next_next_node, inner_prev_node, node, parent_node))
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 186, in visit
out = await self.visit(tree, variable_stack)
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 559, in visit
visited_children.append(await self.visit(child, variable_stack, inner_next_node, inner_next_next_node, inner_prev_node, node, parent_node))
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 266, in visit
visited_children = [await self.visit(child, variable_stack, next_node, next_next_node, prev_node, node, parent_node) for child in node]
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 266, in
visited_children = [await self.visit(child, variable_stack, next_node, next_next_node, prev_node, node, parent_node) for child in node]
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 379, in visit
command_output = await command_function(*positional_args, **named_args)
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/library/_select.py", line 177, in select
raise ValueError("No valid option generated in #select! Please post a GitHub issue since this should not happen :)")
ValueError: No valid option generated in #select! Please post a GitHub issue since this should not happen :)

Error in program: No valid option generated in #select! Please post a GitHub issue since this should not happen :)
100%|████████████████████████████████████████████████████████████| 5/5 [00:16<00:00, 4.03s/it]
Traceback (most recent call last):
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 109, in run
await self.visit(self.parse_tree, VariableStack([self.program._variables], self))
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 559, in visit
visited_children.append(await self.visit(child, variable_stack, inner_next_node, inner_next_next_node, inner_prev_node, node, parent_node))
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 186, in visit
out = await self.visit(tree, variable_stack)
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 559, in visit
visited_children.append(await self.visit(child, variable_stack, inner_next_node, inner_next_next_node, inner_prev_node, node, parent_node))
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 266, in visit
visited_children = [await self.visit(child, variable_stack, next_node, next_next_node, prev_node, node, parent_node) for child in node]
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 266, in
visited_children = [await self.visit(child, variable_stack, next_node, next_next_node, prev_node, node, parent_node) for child in node]
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/_program_executor.py", line 379, in visit
command_output = await command_function(*positional_args, **named_args)
File "/Users/zhaoxuefeng/opt/miniconda3/envs/adala/lib/python3.10/site-packages/guidance/library/_select.py", line 177, in select
raise ValueError("No valid option generated in #select! Please post a GitHub issue since this should not happen :)")
ValueError: No valid option generated in #select! Please post a GitHub issue since this should not happen :)

Error in program: No valid option generated in #select! Please post a GitHub issue since this should not happen :)

🌐 Hello, amazing community!

I'm exploring the integration of two powerful libraries: MiniAutoGen and Adala, and I would greatly appreciate your help and insights!

MiniAutoGen is an innovative open-source library designed to take applications with Large Language Models (LLMs) to the next level. Its differentiators are its lightweight and flexible approach, which allows for a high degree of customization.

Here are some notable features of MiniAutoGen:

• Multi-Agent Dialogues: The ability to create complex and nuanced interactions with multiple intelligent agents operating together.
• Agent Coordination: A mechanism that ensures harmony and efficient management among the agents.
• Customizable Agents: Total freedom to shape agent behaviors according to project needs.
• Action Pipeline: Simplifies and automates agent operations, facilitating scalability and maintenance.
• Integration with +100 LLMs: Expanding conversational capabilities with over 100 LLMs for intelligent and contextualized responses.

My Challenge: I'm seeking help from the community to develop new integrations and modules.

I Seek Your Help: Do you have examples, tips, or guidance on how I can accomplish this integration? Any insight or shared experience would be extremely valuable!

Check out MiniAutoGen on Google Colab: MiniAutoGen on Google Colab
And here is the GitHub repository for more information: GitHub - brunocapelao/miniAutoGen

I'm looking forward to your ideas and suggestions. Let's shape the future of AI conversations together! 🌟

I noticed that you're making async calls to OpenAI API manually in async def async_create_completion(). Is there any reason for this instead of using the AsyncOpenAI from the official library?

The reason I found this is because I was instrumenting our app with OpenTelemetry and OpenLLMetry just to find that async OpenAI calls from Adala are not instrumented - because OpenLLMetry expects standard openai package calls. It would be great to stick to the official package to make this (and other) useful tricks working.

Error in executing your sample code

Very urgently needed skills in information extraction。Can this be done first?

openai always timeout or raise exception, is there a plan to support openai request timeout and retry?

I think we should be able to pass in more params when initializing an openAI LLM. For example I want to change the endpoint to point to an internal openai instance but would need to write a basically identical OpenAI Runtime class with a minor change. If we could just pass through kwargs to the guidance LLM so we support all their args that would be great. Id be happy to add the PR as well.

This is very important. Can we do this first?