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

Turned-off implicit summary saving at end of epoch

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-02-08 10:31:51 +01:00
parent 9d0f93689e
commit 3f08c54679
1 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
src/twomartens/masterthesis/aae/train.py
View File

@ -203,10 +203,11 @@ def train_mnist(folding_id: int, inlier_classes: Sequence[int], total_classes: i
print((
f"[{epoch + 1:d}/{train_epoch:d}] - "
f"train time: {per_epoch_time:.2f}, "
f"Decoder loss: {decoder_loss_avg.result()}, X Discriminator loss: {xd_loss_avg.result():.3f}, "
f"Z Discriminator loss: {zd_loss_avg.result():.3f}, "
f"Encoder + Decoder loss: {enc_dec_loss_avg.result():.3f}, "
f"Encoder loss: {encoder_loss_avg.result():.3f}"
f"Decoder loss: {decoder_loss_avg.result(False)}, "
f"X Discriminator loss: {xd_loss_avg.result(False):.3f}, "
f"Z Discriminator loss: {zd_loss_avg.result(False):.3f}, "
f"Encoder + Decoder loss: {enc_dec_loss_avg.result(False):.3f}, "
f"Encoder loss: {encoder_loss_avg.result(False):.3f}"
))
# save sample image