Skip to content
PyTorch implementation Large-Margin Softmax (L-Softmax) loss
Branch: master
Clone or download
Latest commit 5c7ea75 Mar 16, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore Baseline implementation Dec 11, 2017
README.md Update README.md Mar 16, 2018
data.py Baseline implementation Dec 11, 2017
lsoftmax.py Remove unnecessary codes Dec 15, 2017
lsoftmax_test.py Implement L-Softmax Dec 15, 2017
model.py More BN Dec 15, 2017
train_mnist.py Implement L-Softmax Dec 15, 2017

README.md

Large-Margin Softmax (L-Softmax) Loss

PyTorch implementation of L-Softmax loss.

Caution

This quick and dirty implementation is getting some stars. Please note that I've not heavily tested the code, so if you want to use this in your research, business, or whatever, please re-check the correctness of the implementation.

You can’t perform that action at this time.