8000 [Bug Report] Missing model checkpoint files in `/ckpt` directory · Issue #4 · ASLP-lab/SongEval · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

[Bug Report] Missing model checkpoint files in /ckpt directory #4

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
DearCora opened this issue Jun 8, 2025 · 1 comment
Open

[Bug Report] Missing model checkpoint files in /ckpt directory #4

DearCora opened this issue Jun 8, 2025 · 1 comment

Comments

@DearCora
Copy link
DearCora commented Jun 8, 2025

The repository contains an empty /ckpt directory but no actual model checkpoint files are provided. This prevents users from running the evaluation toolkit as the pretrained models mentioned in the README are not accessible.

Steps to Reproduce

  1. Clone the repository: git clone https://github.com/ASLP-lab/SongEval.git
  2. Navigate to the project directory: cd SongEval
  3. Install dependencies: pip install -r requirements.txt
  4. Try to run evaluation: python eval.py -i /path/to/audio.mp3 -o /path/to/output

Expected Behavior

The toolkit should load pretrained models and successfully evaluate audio files across the five aesthetic dimensions:

  • Overall Coherence
  • Memorability
  • Naturalness of Vocal Breathing and Phrasing
  • Clarity of Song Structure
  • Overall Musicality

Actual Behavior

The evaluation fails because no model checkpoint files are found in the /ckpt directory.

Environment

  • Python version: [Please specify]
  • Operating System: [Please specify]
  • Hardware: [CPU/GPU details]

Possible Solutions

  1. Add the missing checkpoint files to the /ckpt directory
  2. Provide download links or instructions for obtaining the pretrained models
  3. Update the README with clear instructions on how to acquire the model weights
  4. Consider hosting the models on Hugging Face Hub (similar to the dataset at huggingface.co/datasets/ASLP-lab/SongEval)

Additional Context

The repository README mentions "pretrained neural models" but doesn't provide information on how to obtain or download these models. This is a critical blocker for users wanting to use the aesthetic evaluation toolkit.

Suggested Documentation Updates

Please consider adding a section to the README that includes:

  • Model download instructions
  • Model file requirements and expected locations
  • File size and storage requirements
  • Alternative hosting locations if models are too large for Git LFS

Thank you for developing this valuable tool for song aesthetic evaluation!
@ASLP-lab
Copy link
Owner
ASLP-lab commented Jun 9, 2025

Thank you for your feedback!

To resolve the issue you're encountering, please follow these steps:

  1. Make sure that you're in the SongEval directory when executing the eval.py script. If you're not in the correct directory, the script won't be able to locate the necessary files.

  2. Ensure that the SongEval directory contains a subdirectory named ckpt, and that it includes the pretrained model file named model.safetensors.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@DearCora @ASLP-lab
0