I faced the same error and tried to test everything with no value, but I heard that you have to make the number of folders in the dataset the SAME as the one in Dense
.
I don’t know if this will solve your specific bug or not but try this with your code:
vgg_model.add(tf.keras.layers.Dense(10, activation = 'softmax'))
Replace 10
with the number of training dataset folders or can call ‘classes’.