unable to train ; dependency error

rui-yan / SSL-FL

Self-supervised federated learning for medical imaging - IEEE TMI

63 stars 9 forks source link

unable to train ; dependency error #5

Closed chillum-codeX closed 1 year ago

chillum-codeX commented 1 year ago

actually the code was written on torch 1.7.1 and it required cuda 11.0 now i have RTX 3090 which has compute capability of sm86 so unable to train and ampere architecture needds cuda 11.3 or greater and torch 1.7.1 doesnt have cuda 11.3 support

can you please make support of pytorch 1.10 oor greater

Thanks

rui-yan commented 1 year ago

Please update the version of Torch accordingly based on your CUDA version.