search

GATECH-EIC / ShiftAddNet

[NeurIPS 2020] ShiftAddNet: A Hardware-Inspired Deep Network
MIT License
66 stars 18 forks source link

have you reproduce the resnet20-cifar10 results(91.8) in the original addernet paper, using your cuda version code? #8

Open brisker opened 3 years ago

brisker commented 3 years ago

have you reproduce the resnet20-cifar10 results(91.8) in the original addernet paper, using your cuda version code?

ranery commented 3 years ago

Hi, thanks for asking!

We have a check.py file to guarantee both the forward and backward outputs from our CUDA codes are the same with the PyTorch version implementation, while largely reducing the GPU memory costs.

So if you adopt the same training receipts with AdderNet, it should perform similarly.