one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [8, 1024, 13, 13]], which is output 0 of CudnnBatchNormBackward, is at version 1; expected version 0 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).
zbyuan
commented
4 years ago
one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [8, 1024, 13, 13]], which is output 0 of CudnnBatchNormBackward, is at version 1; expected version 0 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).