deep learning - Error with layers compatibility using vgg-net

Question

Good day. Attempt to compile main.py code from and get these errors:

1. screenshot1

2. screenshot2

3. screenshot3

    generator = create_gen(lr_ip)
    discriminator = create_disc(hr_ip)
    discriminator.compile(loss="binary_crossentropy", optimizer="adam", metrics=['accuracy'])
    
    vgg = build_vgg()
    vgg.trainable = False
    
    
    gan_model = create_comb(generator, discriminator, vgg, lr_ip, hr_ip)
    gan_model.compile(loss=["binary_crossentropy","mse"], loss_weights=[1e-3, 1], optimizer="adam")

Error:

ValueError: Input 0 of layer fc1 is incompatible with the layer: expected axis -1 of input shape to have value 25088 but received input with shape (None, 4608)

If you know how to get through it, please hint.

question from:https://stackoverflow.com/questions/65883883/error-with-layers-compatibility-using-vgg-net

Answer 1

Waitting for answers