How to reset the GroupChat after one inference? #529
Unanswered
paulilioaica
asked this question in
Q&A
Replies: 2 comments
-
Beta Was this translation helpful? Give feedback.
0 replies
-
I hope you are all well. I'd like to share an additional job search resource with you all. It's called Pathrise. Below are two links that explain how it works along with the link for enrollment. Reach out if you have any questions. Continued good luck with your job search. https://www.pathrise.com/apply?utm_source=ramone&utm_medium=affiliate&utm_campaign=apply_linkedin |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello everyone.
I am trying to integrate AutoGen into a inference pipeline with GroupChat.
My main workflow is this:
Finally, i want to run 1,2,3 on another data point.
Whenever I run this in a loop, for some reason, the inputs merge in some way and I reach the token limit, and I get messages like the gpt4 is only 8k but I tried to feed 129k into it.
This is somewhat related to reseting the groupchat since
increases with every call with precisely number of tokens of current data point.
My code is this
In my main code, I call
reset()
after eachextract_dependency()
of failure in a try catchTL;DR
How can I completley reset the enviroment between inferences? groupchat.reset() doesn't do it.
Beta Was this translation helpful? Give feedback.
All reactions