using python 3.7 and torch 1.3.1, begin train kitti dataset and met this error.
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [960]], which is output 0 of SelectBackward, is at version 4; expected version 3 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).
using python 3.7 and torch 1.3.1, begin train kitti dataset and met this error.
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [960]], which is output 0 of SelectBackward, is at version 4; expected version 3 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).