AI4HealthUOL / SSSD-ECG

Repository for the paper: 'Diffusion-based Conditional ECG Generation with Structured State Space Models'
MIT License
36 stars 4 forks source link

Data Loader #2

Closed elizczimm closed 1 year ago

elizczimm commented 1 year ago

https://github.com/AI4HealthUOL/SSSD-ECG/blob/d82aa2e4a207cf2c3211dd4c41c34a5aec7f61db/src/sssd/train.py#L81

How should the data be formatted?

juanlopezcode commented 1 year ago

hello, the numpy file should be [samples,channels,length], and you can get the data from here: https://mega.nz/folder/kT91jYpI#97GyTkVVUk97fzs1Oy4nBQ/folder/sW8UiBpT

elizczimm commented 1 year ago

Thanks for your help! I didn't see files for labels referenced on line 82 -- are they generated separately? And what should their format be?

https://github.com/AI4HealthUOL/SSSD-ECG/blob/d82aa2e4a207cf2c3211dd4c41c34a5aec7f61db/src/sssd/train.py#L82

juanlopezcode commented 1 year ago

Hello @elizczimm, the format of the label set would be [samples,labels], I have uploaded under src/ptb_xl/ instructions of how to get the data