I'm excited to run the code, thanks for uploading.
I wonder how the pretrained model was selected for downstream task. Did you use the the pretrained model after 800 epochs? Or, pretrained model was selected among all epochs that work best for each downstream tasks?
Also, pretrained dataset code15 is not prepared in WFDB format, so I can't run /data/process_data.ecg.
I'm excited to run the code, thanks for uploading.
I wonder how the pretrained model was selected for downstream task. Did you use the the pretrained model after 800 epochs? Or, pretrained model was selected among all epochs that work best for each downstream tasks?
Also, pretrained dataset code15 is not prepared in WFDB format, so I can't run /data/process_data.ecg.
How can I prepare a dataset for downstream task?