search

Sha-Lab / FEAT

The code repository for "Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions"
MIT License
418 stars 84 forks source link

Can these codes work on Pytorch v1.1.0? #28

Closed SnowyJune973 closed 4 years ago

SnowyJune973 commented 4 years ago

Hi! The codes are really neat! Thanks for sharing! I notice that it requires a Pytorch>=1.4. However because of the limitation of my working environment, the CUDA requirement of Pytorch v1.4 can't be realized, so I wonder if the codes will work on lower versions of Pytorch that work with CUDA 9.0 (pytorch v1.1.0 for example)?

Han-Jia commented 4 years ago

Yes, the code works with lower versions of PyTorch, but you may need to change some functions to make them compatible (e.g., change flatten() to view(-1) in protonet.py).