How to use multiple GPUs to accelerate training?

cure-lab / DeciWatch

[ECCV 2022] Official implementation of the paper "DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation"

Apache License 2.0

177 stars 15 forks source link

How to use multiple GPUs to accelerate training? #9

Closed haofeng0705 closed 2 years ago

haofeng0705 commented 2 years ago

Tried to add following code: model = nn.DataParallel(model) model = model.cuda() not working :(

juxuan27 commented 2 years ago

Hi, @Elliot-hao ! Thank you for your focus! Our code does not support multiple GPU training and testing. We are welcome for PR if you can help us add it!