search

allenai / XNOR-Net

ImageNet classification using binary Convolutional Neural Networks
https://xnor.ai/
Other
856 stars 239 forks source link

Bug util.lua #5

Closed albicasty closed 7 years ago

albicasty commented 7 years ago

Hi, I noticed you modified the util.lua file, adding m:add(1/(n)):mul(1-1/s[2]):mul(n) in the updateBinaryGradWeight function. However, I don't get which is the origin of this modification and where in your paper you talk about it. Moreover, I am training a binarized version of the resnet: the problem is the training diverges with the mul(n).

In addition, may you also upload the model you used for training the binarized Resnet? Thank you.

mrastegari commented 7 years ago

please see the discussion at https://github.com/mrastegari/XNOR-Net/issues/3You may use the provided function in the discussion for ResNet. Since I have made changes in the code I need to re train the ResNet again. I will upload it once I am done.

albicasty commented 7 years ago

Thank you!

I am trying to reproduce you results with the ResNet 18. However, for the first epochs the Top5 and Top1 values are much lower that the ones you declared. For this reason, I think I have not modified properly their model.