when i run main.py ,i met some problem:
Traceback (most recent call last):
File "main.py", line 186, in
main()
File "main.py", line 159, in main
expert_acc, learner_acc = train_vdb(vdb, memory, vdb_optim, demonstrations, 0, args)
File "/home/parallels/workspace/lets-do-irl-master/mujoco/vail/train_model.py", line 37, in train_vdb
vdb_loss.backward(retain_graph=True)
File "/home/parallels/anaconda3/envs/mujoco/lib/python3.8/site-packages/torch/_tensor.py", line 487, in backward
torch.autograd.backward(
File "/home/parallels/anaconda3/envs/mujoco/lib/python3.8/site-packages/torch/autograd/init.py", line 197, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.FloatTensor [100, 4]], which is output 0 of AsStridedBackward0, is at version 35; expected version 34 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!
when i run main.py ,i met some problem:
Traceback (most recent call last):
File "main.py", line 186, in
main()
File "main.py", line 159, in main
expert_acc, learner_acc = train_vdb(vdb, memory, vdb_optim, demonstrations, 0, args)
File "/home/parallels/workspace/lets-do-irl-master/mujoco/vail/train_model.py", line 37, in train_vdb
vdb_loss.backward(retain_graph=True)
File "/home/parallels/anaconda3/envs/mujoco/lib/python3.8/site-packages/torch/_tensor.py", line 487, in backward
torch.autograd.backward(
File "/home/parallels/anaconda3/envs/mujoco/lib/python3.8/site-packages/torch/autograd/init.py", line 197, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.FloatTensor [100, 4]], which is output 0 of AsStridedBackward0, is at version 35; expected version 34 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!