Closed Jupiter1995 closed 10 months ago
Greetings! Thanks for your first issue!
I don't run into this error in my env. Seems like a version mismatch issue. What pydantic version are you using?
If you create a new env and pip install gentopia
, you should get pydantic == 1.9.0
I don't run into this error in my env. Seems like a version mismatch issue. What pydantic version are you using?
If you create a new env and
pip install gentopia
, you should get pydantic == 1.9.0
I did install the required packages via requirement.txt under docs. But I think you're right. Mine is 2.4.2. Let me uninstall it and check again. I'll keep you posted.
I don't run into this error in my env. Seems like a version mismatch issue. What pydantic version are you using? If you create a new env and
pip install gentopia
, you should get pydantic == 1.9.0I did install the required packages via requirement.txt under docs. But I think you're right. Mine is 2.4.2. Let me uninstall it and check again. I'll keep you posted.
I created a new env and ran pip install gentopia
, and it solved the old problem! However, I'm receiving a new error from typing package lol:
Any idea or suggestions?
what's your agent config file
what's your agent config file
The agent.yaml
file is attached below. Please ignore the environment
as the implementation hasn't finished, so it will ignored in the agent_assembler.py
.
name: mark
type: react_conv
version: 0.0.1
description: mark is a professional personal trainer. He is able to create customized plans for different people depends on their expectations and body conditions.
target_tasks:
- create training plans and strategies.
- collaborate with nutritionist to adjust training plans.
prompt_template: !prompt multi_agent_test.agents.personal_trainer.prompt.PromptOfTrianer
llm:
model_name: baize
params:
temperature: 0.5
top_p: 0.9
repetition_penalty: 0.0
max_tokens: 1024
environment:
name: training_plan_test01
plugins:
- name: google_search
- name: arxiv_search
param:
top_k: 3
- name: web_page
I tried to use the original Gentopia without any changes and also the example agent elon
but still received the same error within the new environment I created using pip install gentopia
. The agent.yaml
attached below.
name: elon
type: react
version: 0.0.1
description: elon is an experienced and visionary entrepreneur. He is able to create a startup from scratch and get a strong team to support his ideas.
target_tasks:
- create business plans and strategies.
- arrange a bunch of tools and agents to collaborate on his ideas.
prompt_template: !prompt gentpool.pool.elon.prompt.PromptOfElon
llm:
model_name: baize
params:
temperature: 0.0
top_p: 0.9
repetition_penalty: 0.0
max_tokens: 1024
plugins:
- name: google_search
- name: arxiv_search
param:
top_k: 3
- name: web_page
- name: bash_shell
- !tool gentpool.pool.elon.tool.ElonDrawing
- !include ../dr_science/agent.yaml
Screenshot of the error:
I tried to use the original Gentopia without any changes and also the example agent
elon
but still received the same error within the new environment I created usingpip install gentopia
. Theagent.yaml
attached below.name: elon type: react version: 0.0.1 description: elon is an experienced and visionary entrepreneur. He is able to create a startup from scratch and get a strong team to support his ideas. target_tasks: - create business plans and strategies. - arrange a bunch of tools and agents to collaborate on his ideas. prompt_template: !prompt gentpool.pool.elon.prompt.PromptOfElon llm: model_name: baize params: temperature: 0.0 top_p: 0.9 repetition_penalty: 0.0 max_tokens: 1024 plugins: - name: google_search - name: arxiv_search param: top_k: 3 - name: web_page - name: bash_shell - !tool gentpool.pool.elon.tool.ElonDrawing - !include ../dr_science/agent.yaml
Screenshot of the error:
Problem solved! I'll create a pull request
When running a customized
assembler.py
, I received a pydnatic error saying, "Fieldname
doesn't have the annotation." The detailed error screenshot is attached below and some printouts to help understand the problem.error:
printout code and printouts:
# test block codes inside basetool.py under ToolMetaclass.__new__():
print("----test block-----")
print(f"tool name: {name} \n")
print(f"namespace dict annotation list: {dct.get('__annotations__', [])} \n")
print(f"namespace items: {[k for k,v in dct.items()]} \n")
print(f"namespace vals: {[v for k,v in dct.items()]} \n")
print("----test block end----")