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

16 votes246 views6 answers
Pip install pandas
Python
6 votes172 views2 answers
Python file to array
Python
1 vote181 views1 answer
Python make button do more than one command
Python
9 votes176 views1 answer
Area of triangle
Python
5 votes180 views3 answers
Convert list to dataframe
Python
1 vote158 views1 answer
Pytorch summary model
Python
0 vote196 views1 answer
Django clodinarystorage
Python
0 vote170 views1 answer
YouCompleteMe unavailable: requires Vim compiled with Python (3.6.0+) support.
Python
0 vote180 views1 answer
Histogram for categorical data with plotly
Python
9 votes204 views4 answers
Pandas print full dataframe
Python