LFS bandwidth exceeded - request for alternative model weights download

[stevetodman]

Congratulations on winning the 2024 PhysioNet Challenge! I'm trying to use ECG-Digitiser for an educational ECG teaching application.

Problem

When cloning the repository, the model weights fail to download:

Error downloading object: models/M1/nnUNet_results/.../checkpoint_final.pth
batch response: This repository exceeded its LFS budget.

Environment

• macOS (Apple Silicon M-series)
• All Python dependencies install successfully
• PyTorch with MPS acceleration works
• Only the LFS-hosted weights are inaccessible

Request

Could you provide an alternative download location for the pretrained weights? Options:

• Google Drive / Dropbox link
• Hugging Face Hub (free LFS hosting for ML models)
• GitHub Release attachment (up to 2GB per file)
• Zenodo (free for research artifacts)

Happy to help migrate if useful. Thanks for open-sourcing this work!

[malahmadi1]

@stevetodman thank you for asking this, any luck on finding the pretrained weights?

[stevetodman]

No luck, unfortunately

…

________________________________ From: malahmadi1 ***@***.***> Sent: Saturday, January 31, 2026 2:42 PM To: felixkrones/ECG-Digitiser ***@***.***> Cc: stevetodman ***@***.***>; Mention ***@***.***> Subject: Re: [felixkrones/ECG-Digitiser] LFS bandwidth exceeded - request for alternative model weights download (Issue #19) [https://avatars.githubusercontent.com/u/18080404?s=20&v=4]malahmadi1 left a comment (felixkrones/ECG-Digitiser#19)<#19 (comment)> @stevetodman<https://github.com/stevetodman> thank you for asking this, any luck on finding the pretrained weights? — Reply to this email directly, view it on GitHub<#19 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BMOH6OCROHDN7FNCUDSQU4L4JUHR7AVCNFSM6AAAAACO6V6IGGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTQMRZGMYDGOJZG4>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[malahmadi1]

I'm trying to use this repo, maybe it is helpful for you as well.
https://github.com/Ahus-AIM/Open-ECG-Digitizer