jannerm / ddpo

Code for the paper "Training Diffusion Models with Reinforcement Learning"

Home Page:https://rl-diffusion.github.io

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Unable to run the LLaVa bertscore

bhattg opened this issue · comments

Hi,
I am trying to run the code for LLaVa BERTscore, using

python pipeline/policy_gradient.py --dataset llava_bertscore

However, I am getting the following timeout error. Is the readme missing things on the hosting server?
I am not too familiar with it.

WARNING:urllib3.connectionpool:Retrying (Retry(total=998, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f3570062e20>: Failed to establish a new connection: [Errno 111] Connection refused')': /

I use a GPU machine, is there a way to directly call the LLaVa model in case there is a script?

I just figured it out on own from the other repo's app.py and test.py code. Thanks