Model accuracy is not changing

Course Q&A TensorFlow Developer Professional Convolutional Neural Networks in TensorFlow
1

Hi, I am working right now on week3 assignment, and I facing a weird problem; the model accuracy is not changing at all after several epochs.

result:

Epoch 1/100
33/33 [==============================] - 93s 3s/step - loss: 0.8737 - accuracy: 0.5000 - val_loss: 0.6961 - val_accuracy: 0.5000
Epoch 2/100
33/33 [==============================] - 89s 3s/step - loss: 0.6958 - accuracy: 0.5000 - val_loss: 0.6938 - val_accuracy: 0.5000
Epoch 3/100
33/33 [==============================] - 87s 3s/step - loss: 0.6942 - accuracy: 0.5000 - val_loss: 0.6936 - val_accuracy: 0.5000
Epoch 4/100
33/33 [==============================] - 91s 3s/step - loss: 0.6937 - accuracy: 0.5000 - val_loss: 0.6934 - val_accuracy: 0.5000
2

Please click my name and message your notebook as an attachment. Do remove code from your topic since posting code in public is discouraged. It’s okay to share your stacktrace though.

3

okay, the notebook has been sent to you

4

Both training and validations image generators have the class mode set incorrectly. Class mode should not be set to ‘categorical’ when the output has a single neuron for classifying between 2 classes.

5

okay it now works, thank you