2020-06-08 02:50:22,317 ERROR worker.py:1559 -- Possible unhandled error from worker: ray_worker (pid=20930, host=gra972)
File "/project/6005065/avrech/distributedRL/common/abstract/learner.py", line 84, in run
step_info, idxes, priorities = self.learning_step(replay_data)
File "/project/6005065/avrech/distributedRL/apex_dqn/dqn_learner.py", line 55, in learning_step
loss1.backward()
File "/project/6005065/avrech/venv/lib/python3.7/site-packages/torch/tensor.py", line 195, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "/project/6005065/avrech/venv/lib/python3.7/site-packages/torch/autograd/__init__.py", line 99, in backward
allow_unreachable=True) # allow_unreachable flag
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn
I am restricted to use torch 1.4.0 on the server, and this is the torch version I tested on my desktop.