tensorfreitas/Temporal-Ensembling-for-Semi-Supervised-Learning

Question about the ensemble target

Closed this issue · 1 comments

HHHedo commented

Hi,
Thanks for your code.
In your implement, the ensemble targets are network output before softmax. However, I think the target is the ensemble of output after softmax. Since in table 5 in the paper, output is the softmax.

tensorfreitas commented

Hi @HHHedo sorry for the late response. Only now I came back from vacations.

The last layer is a dense layer with softmax activation:

wight_norm_layers.Dense.Dense(units=10, activation=tf.nn.softmax,