Fix Spectral Normalization layer

[markub3327]

Hi,

spectral normalization layer is sensitive to initialized weights to zeroes. I fix it.

Thanks.

[bot-of-gabrieldemarmiesse]

@charlielito

You are owner of some files modified in this pull request. Would you kindly review the changes whenever you have the time to? Thank you very much.

[fsx950223]

Any reference about it?

[markub3327]

        self.discriminator = keras.Sequential(
            [
                keras.Input(shape=(64, 64, 3)),
                layers.GaussianNoise(0.1),

                layers.Conv2D(64, 7, padding="same", kernel_initializer=TruncatedNormal(stddev=0.02)),
                layers.LeakyReLU(alpha=0.2),

                layers.Conv2D(128, 4, strides=(2, 2), padding="same", kernel_initializer=TruncatedNormal(stddev=0.02)),
                layers.LeakyReLU(alpha=0.2),

                layers.Conv2D(256, 4, strides=(2, 2), padding="same", kernel_initializer=TruncatedNormal(stddev=0.02)),
                layers.LeakyReLU(alpha=0.2),

                layers.Conv2D(512, 4, strides=(2, 2), padding="same", kernel_initializer=TruncatedNormal(stddev=0.02)),
                layers.LeakyReLU(alpha=0.2),

                layers.Conv2D(512, 4, strides=(2, 2), padding="same", kernel_initializer=TruncatedNormal(stddev=0.02)),
                layers.LeakyReLU(alpha=0.2),

                layers.Flatten(),
                layers.Dropout(0.4),

                layers.Dense(1, kernel_initializer="zeros"),
            ],
            name="discriminator",
        )

[seanpmorgan]

Thank you for your contribution. We sincerely apologize for any delay in reviewing, but TensorFlow Addons is transitioning to a minimal maintenance and release mode. New features will not be added to this repository. For more information, please see our public messaging on this decision: TensorFlow Addons Wind Down

Please consider sending feature requests / contributions to other repositories in the TF community with a similar charters to TFA: Keras Keras-CV Keras-NLP