Pytorch use multiple gpu

#easiest solution is to wrap you model in DataParallel like so:

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = Model(input_size, output_size)
if torch.cuda.device_count() > 1:
  print("Let's use", torch.cuda.device_count(), "GPUs!")
  model = nn.DataParallel(model)

model.to(device)

Source: pytorch.org

Similar codes

31 votes177 views5 answers
Python loop through files in directory
Python
29 votes184 views6 answers
Python empty dataframe
Python
1 vote164 views1 answer
Array of dic
Python
1 vote182 views1 answer
Convert number to time python
Python
0 vote169 views1 answer
What are the mouseX/mouseY variebles in pycharm
Python
23 votes169 views6 answers
Flask post
Python
3 votes166 views1 answer
Python3 return a list of indexes of a specific character in a string
Python
23 votes177 views3 answers
Pandas read csv without index
Python
4 votes168 views1 answer
Standardscaler into df data frame pandas
Python
0 vote176 views1 answer
Np.ma.filled
Python