it reported an error in loss.backward()

[CCJing14]

Error detected in SigmoidBackward0. Traceback of forward call that caused the error: File "test.py", line 636, in test(opt.data, File "test.py", line 235, in test adapt_single(image, optimizer, batch_size, model, device) File "test.py", line 62, in adapt_single outputs,train_output = model(inputs_t) # [batch_size, num_boxes, 85] File "/home/miniconda3/envs/yolov7/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/miniconda3/envs/yolov7/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl return forward_call(*args, **kwargs) File "/home/yolov7/models/yolo.py", line 599, in forward return self.forward_once(x, profile) # single-scale inference, train File "/home/yolov7/models/yolo.py", line 625, in forward_once x = m(x) # run File "/home/miniconda3/envs/yolov7/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/envs/yolov7/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl return forward_call(*args, **kwargs) File "/home/yolov7/models/yolo.py", line 153, in fuseforward y = x[i].sigmoid() (Triggered internally at ../torch/csrc/autograd/python_anomaly_mode.cpp:114.)

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [32, 3, 1, 1, 85]], which is output 0 of SigmoidBackward0, is at version 2; expected version 0 instead. Hint: the backtrace further above shows the operation that failed to compute its gradient. The variable in question was changed in there or anywhere later. Good luck!