search

vrtompki / pd2se_net_project

0 stars 0 forks source link

Update model.py #3

Open jiafw opened 4 years ago

jiafw commented 4 years ago

For model.py: If you use the nn.CrossEntropyLoss() to compute the loss, you shouldn't use softmax before computing loss. Because nn.CrossEntropyLoss() include the softmax step. It is clearly that it would be wrong to use the softmax steps twice.

For dataset.py: I find that the number of test image and label is not equal. So I checked out the code and found the bug.

vrtompki commented 4 years ago

I was trying not to deviate as much as possible from the paper's model. I wanted to implement what they stated literally but found some questionable details, and the authors would not answer these questions I had about the model and their sample selection criteria, and would only tell me where their data came from.