Skip to content

[Bug] Pre-formatted data not working #555

Description

@veritas9872

Checklist

  • 1. I have searched related issues but cannot get the expected help.
  • 2. The bug has not been fixed in the latest version.
  • 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
  • 4. If the issue you raised is not a bug but a question, please raise a discussion at https://github.com/sgl-project/SpecForge/discussions/new/choose Otherwise, it will be closed.
  • 5. Please use English, otherwise it will be closed.

Describe the bug

When using pre-formatted data, I found that most of the data was skipped during training, leading to 1024 times smaller batch sizes. It turned out that the culprit was the different tools being passed during the construction of the eagle dataset.

Reproduction

specforge/data/preprocessing.py Line 378

Add: tools=[[] for _ in range(len(examples["text"]))],

Environment

This is a logical error and should be reproduced in the main branch with pre-formatted data that uses the build_eagle3_dataset function.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions