zjunlp / autoact Goto Github PK

View Code? Open in Web Editor NEW

112.0 112.0 8.0 6.47 MB

AUTOACT: Automatic Agent Learning from Scratch for QA via Self-Planning

Home Page: https://zjunlp.github.io/project/AutoAct

License: Apache License 2.0

Python 98.79% Shell 1.21%

agent agent-learning agents autoact large-language-models natural-language-processing nlp self-planning

autoact's People

Contributors

Stargazers

Watchers

Forkers

mivanovitch rolnand zxlzr qixucen taofeifei0215 bipolar713 fellowtraveler konisberg

autoact's Issues

How to get the final metric after Group Planning？

There is no output when performing Self_Planning/Traj_Syn/run_task.py

Hi,
When I execute Self_Planning/Traj_Syn/run_task.py, there is no output from Thought and Action, and there is a warning message, I don't know if you have a solution

experimental trajectory

Thank you very much for your work. Can you please share your experimental trajectory? I would be very grateful if you could.

How do I run this project and I'm having some problems

When I run data_generation directly, the top_k parameter is not displayed in the openai model. I have changed several openai and langchian versions, but there is still no way to solve the default parameter problem

KeyError: 'conversations'

Hi,
During training, there was no keyword "conversations" in the list raw_data
I printed raw_data and found that there were only two keywords, "input" and "output"
is there something I did wrong?