Pytorch

• References

  • http://pytorch.org/about/

• tensorflow와 다르게 기본 데이터 구조인 tensor가 cpu tensor(torch.Tensor)와 gput tensor(torch.cuda.Tensor)로 나뉜다.

  • WHY???
  • torch VS Others
    • https://www.reddit.com/r/MachineLearning/comments/5w3q74/d_so_pytorch_vs_tensorflow_whats_the_verdict_on/

• Dynamic Computational Graph???

  • https://medium.com/intuitionmachine/pytorch-dynamic-computational-graphs-and-modular-deep-learning-7e7f89f18d1

  dynamic computational graphs are valuable for situations where you cannot determine the computation. One clear example of this are recursive computations that are based on variable data.

load/save

• Model Architecture/ Only the model parameters

#============================ Save and load the model ============================#
# Save and load the entire model.
torch.save(resnet, 'model.pkl')
model = torch.load('model.pkl')

# Save and load only the model parameters(recommended).
torch.save(resnet.state_dict(), 'params.pkl')
resnet.load_state_dict(torch.load('params.pkl'))

multigpus

• Reference
  • http://pytorch.org/tutorials/beginner/former_torchies/parallelism_tutorial.html