I don’t why, I’ve tried many and many times, and it doesn’t work at that line:

history = model.fit(train_generator, epochs=200, steps_per_epoch=167,
                             callbacks=[modelcheckpoint],
                             validation_data=validation_generator, validation_steps=50)

Gosh, I can’t get enough. I don’t know why. I’ve done all shapes correctly, and it appears that mistake:

ValueError: `logits` and `labels` must have the same shape, received ((None, None, None, 1) vs (None,)).

https://drive.google.com/file/d/1r-TznZVZI9HcPGjfS_i-DfpEJ7mBUw0x/view?usp=sharing

I’ve changed a lot of things on X_train and scaled. And nothing happens!

I manipulated image reading, and it doesn’t work at all. I cannot even find out the root cause