alibaba / alibaba/EasyNLP

beautifulprompt PPO can't find pytorch_model.bin

Open
#368 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
2.2k
Forks
257
PR merge metrics
No merged PRs in 30d

Description

When i running the PPO code, is it necessary to look for the `pytorch_model.bin` file in the `output/sft` folder, but only a single model file `model.safetensors` is saved after `sft`, causing the PPO script to fail to run?

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the PPO code path that reads the checkpoint from output/sft, then compare its expectation of pytorch_model.bin with the model.safetensors file produced after SFT. Run the PPO flow against the generated output/sft checkpoint; done means it can load that checkpoint and proceed without the current missing-file failure.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.