Closed dnth closed 9 years ago
I've changed the training interface to generate a (train, valid) monitor pair after every iteration. The validation monitors might not be updated every training iteration, but whatever is most recent will be returned.
for train, valid in exp.itertrain(dataset):
print('training loss:', train['loss'])
print('most recent validation loss:', valid['loss'])
It would be really nice to have the itertrain method to return the validation set errors in order for a plot of learning curves.