New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

The model is not trainable #71

Open

kingman1980 opened this issue Jul 11, 2019 · 0 comments

kingman1980 commented Jul 11, 2019

Keras: 2.2.4
Tensorflow: 1.14.0

During training, the loss became smaller.
But when I use the best model to test, there is nothing in the predicted image.

Epoch 00001: val_loss improved from inf to 0.32606, saving model to
Epoch 2/150

49s - loss: 0.3457 - acc: 0.8952 - val_loss: 0.3242 - val_acc: 0.9042

Epoch 00002: val_loss improved from 0.32606 to 0.32418, saving model to
Epoch 3/150

54s - loss: 0.3440 - acc: 0.8952 - val_loss: 0.3230 - val_acc: 0.9042

Epoch 00003: val_loss improved from 0.32418 to 0.32298, saving model to
Epoch 4/150

56s - loss: 0.3429 - acc: 0.8952 - val_loss: 0.3221 - val_acc: 0.9042

Epoch 00004: val_loss improved from 0.32298 to 0.32206, saving model to
Epoch 5/150

55s - loss: 0.3420 - acc: 0.8952 - val_loss: 0.3212 - val_acc: 0.9042

Epoch 00005: val_loss improved from 0.32206 to 0.32117, saving model to

someone can tell me why?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment