model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])

history = model.fit(
      train_generator,
      epochs=6,
      validation_data=validation_generator,
      verbose=2)

Train for 63 steps, validate for 32 steps
Epoch 1/6
63/63 - 444s - loss: 0.2388 - accuracy: 0.9155 - val_loss: 0.0793 - val_accuracy: 0.9810
Epoch 2/6
63/63 - 212s - loss: 0.0728 - accuracy: 0.9805 - val_loss: 0.0563 - val_accuracy: 0.9860
Epoch 3/6
63/63 - 88s - loss: 0.0552 - accuracy: 0.9840 - val_loss: 0.0487 - val_accuracy: 0.9870
Epoch 4/6
63/63 - 89s - loss: 0.0443 - accuracy: 0.9875 - val_loss: 0.0443 - val_accuracy: 0.9850
Epoch 5/6
63/63 - 77s - loss: 0.0375 - accuracy: 0.9895 - val_loss: 0.0417 - val_accuracy: 0.9860
Epoch 6/6
63/63 - 79s - loss: 0.0326 - accuracy: 0.9920 - val_loss: 0.0403 - val_accuracy: 0.9870