bowlderstudio
commented
4 years ago I debug the problem. It seems the input of your network is a status which the dimension is only (4, 16, 8, 4) which only presents status, however the status return by the game and suppose to input into your model is (8, 4, 84, 84), which looks like image. So I guess maybe the game package I installed in different from yours.
Could you tell me the version of your pytorch. I got runtime error: python main.py --algo=ppo --mode=train pygame 1.9.6 Hello from the pygame community. https://www.pygame.org/contribute.html /workdata/git/drl-experiments/ppo.py:54: UserWarning: nn.init.uniform is now deprecated in favor of nn.init.uniform_. torch.nn.init.uniform(m.weight, -0.01, 0.01) libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: cHRM chunk does not match sRGB libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile Traceback (most recent call last): File "main.py", line 133, in
agent.train()
File "/workdata/git/drl-experiments/ppo.py", line 241, in train
values, actions, action_log_probs = self.net.act(states)
File "/workdata/git/drl-experiments/ppo.py", line 87, in act
values, action_logits = self.forward(x)
File "/workdata/git/drl-experiments/ppo.py", line 72, in forward
x = self.relu1(self.conv1(x))
File "/home/minshi/.virtualenvs/drl-experiments-0ZyJdSJ6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, **kwargs)
File "/home/minshi/.virtualenvs/drl-experiments-0ZyJdSJ6/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 345, in forward
return self.conv2d_forward(input, self.weight)
File "/home/minshi/.virtualenvs/drl-experiments-0ZyJdSJ6/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 342, in conv2d_forward
self.padding, self.dilation, self.groups)
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same