search

wy1iu / LargeMargin_Softmax_Loss

Implementation for <Large-Margin Softmax Loss for Convolutional Neural Networks> in ICML'16.
Other
350 stars 115 forks source link

train mnist，loss is nan #28

Open xiaosuzhang opened 6 years ago

xiaosuzhang commented 6 years ago

when i use your mnist_train_test.prototxt to train the mnist, the loss is nan? How should i do to overcome this issues??