When I ran DQN Server on CIFAR-10 dataset (In cifar-10.json.template, only changed server to:, "server": "dqn")
Complete Error as follows:
.
.
.
[INFO][21:32:06]: Training on client #98
[INFO][21:32:06]: Training on client #99
[INFO][21:36:37]: Reports recieved: 100
[INFO][21:36:37]: Training: 100 rounds or 99.0% accuracy
[INFO][21:36:37]: Round 1/100
Traceback (most recent call last):
File "run.py", line 50, in
main()
File "run.py", line 43, in main
fl_server.run()
File "/home/abc/server/dqn.py", line 108, in run
super().run()
File "/home/abc/server/server.py", line 151, in run
accuracy = self.round()
File "/home/abc/server/dqn.py", line 114, in round
sample_client_index = self.selection()
File "/home/abc/server/dqn.py", line 207, in selection
sample_client_index = self.dqn.choose_action(weights)
File "/home/abc/server/dqn.py", line 49, in choose_action
actions_value = self.eval_net.forward(x)
File "/home/abc/server/dqn.py", line 258, in forward
x = self.fc1(x)
File "/home/abc/.conda/envs/fl-py37/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/abc/.conda/envs/fl-py37/lib/python3.7/site-packages/torch/nn/modules/linear.py", line 92, in forward
return F.linear(input, self.weight, self.bias)
File "/home/abc/.conda/envs/fl-py37/lib/python3.7/site-packages/torch/nn/functional.py", line 1406, in linear
ret = torch.addmm(bias, input, weight.t())
RuntimeError: size mismatch, m1: [1 x 6200600], m2: [43108000 x 10] at /opt/conda/conda-bld/pytorch_1556653114079/work/aten/src/TH/generic/THTensorMath.cpp:961
wixidixi
commented
2 years ago
When I ran DQN Server on CIFAR-10 dataset (In cifar-10.json.template, only changed server to:, "server": "dqn")
Complete Error as follows: . . . [INFO][21:32:06]: Training on client #98 [INFO][21:32:06]: Training on client #99 [INFO][21:36:37]: Reports recieved: 100 [INFO][21:36:37]: Training: 100 rounds or 99.0% accuracy
[INFO][21:36:37]: Round 1/100 Traceback (most recent call last): File "run.py", line 50, in
main()
File "run.py", line 43, in main
fl_server.run()
File "/home/abc/server/dqn.py", line 108, in run
super().run()
File "/home/abc/server/server.py", line 151, in run
accuracy = self.round()
File "/home/abc/server/dqn.py", line 114, in round
sample_client_index = self.selection()
File "/home/abc/server/dqn.py", line 207, in selection
sample_client_index = self.dqn.choose_action(weights)
File "/home/abc/server/dqn.py", line 49, in choose_action
actions_value = self.eval_net.forward(x)
File "/home/abc/server/dqn.py", line 258, in forward
x = self.fc1(x)
File "/home/abc/.conda/envs/fl-py37/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/abc/.conda/envs/fl-py37/lib/python3.7/site-packages/torch/nn/modules/linear.py", line 92, in forward
return F.linear(input, self.weight, self.bias)
File "/home/abc/.conda/envs/fl-py37/lib/python3.7/site-packages/torch/nn/functional.py", line 1406, in linear
ret = torch.addmm(bias, input, weight.t())
RuntimeError: size mismatch, m1: [1 x 6200600], m2: [43108000 x 10] at /opt/conda/conda-bld/pytorch_1556653114079/work/aten/src/TH/generic/THTensorMath.cpp:961