modelscope / modelscope/easydistill

the way of loading teacher's logits may be wrong

Open
#21 1 comment 0 reactions 1 assignee View on GitHub

@Andrewyan123 is already working on this.

Since Aug 25, 2025.

Dominant language
Python
Stars
475
Forks
45
PR merge metrics
No merged PRs in 30d

Description

Image Image

When you use global_step, after you have trained for several epochs, you may get an all-zero teacher logits because the start_idx and end_idx may be out of range. (you can print the length of loaded_data to verify this issue)

I think you can include the idx of the training data and use the idx to obtain the logits of the teachers.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.