2martens/masterthesis
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Added names and types to loss metrics

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-02-07 18:06:33 +01:00
parent cce0975a13
commit c52a3cdfb9
1 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/twomartens/masterthesis/aae/train.py
View File

@ -110,11 +110,11 @@ def train_mnist(folding_id: int, inlier_classes: Sequence[int], total_classes: i
for epoch in range(train_epoch):
# define loss variables
encoder_loss_avg = tfe.metrics.Mean()
decoder_loss_avg = tfe.metrics.Mean()
enc_dec_loss_avg = tfe.metrics.Mean()
zd_loss_avg = tfe.metrics.Mean()
xd_loss_avg = tfe.metrics.Mean()
encoder_loss_avg = tfe.metrics.Mean(name='encoder_loss', dtype=tf.float32)
decoder_loss_avg = tfe.metrics.Mean(name='decoder_loss', dtype=tf.float32)
enc_dec_loss_avg = tfe.metrics.Mean(name='enc_dec_loss', dtype=tf.float32)
zd_loss_avg = tfe.metrics.Mean(name='zd_loss', dtype=tf.float32)
xd_loss_avg = tfe.metrics.Mean(name='xd_loss', dtype=tf.float32)
epoch_start_time = time.time()