Updated Variational AutoEncoder example for Keras 3

[sitamgithub-MSIT]

This PR updates the Variational AutoEncoder Keras 3.0 example [TF Only Backend]. All TF ops are replaced with corresponding Keras ops.

For example, here is the notebook link provided:
https://colab.research.google.com/drive/15VMQIEUK8jhqI8nYyFSfWDh9fsMMs116?usp=sharing

cc: @fchollet

The following describes the Git difference for the changed files:

Changes:

diff --git a/examples/generative/vae.py b/examples/generative/vae.py
index d1d195c9..3396f4f7 100644
--- a/examples/generative/vae.py
+++ b/examples/generative/vae.py
@@ -18,6 +18,7 @@ os.environ["KERAS_BACKEND"] = "tensorflow"
 import numpy as np
 import tensorflow as tf
 import keras
+from keras import ops
 from keras import layers
 
 """
@@ -30,10 +31,10 @@ class Sampling(layers.Layer):
 
     def call(self, inputs):
         z_mean, z_log_var = inputs
-        batch = tf.shape(z_mean)[0]
-        dim = tf.shape(z_mean)[1]
-        epsilon = tf.random.normal(shape=(batch, dim))
-        return z_mean + tf.exp(0.5 * z_log_var) * epsilon
+        batch = ops.shape(z_mean)[0]
+        dim = ops.shape(z_mean)[1]
+        epsilon = keras.random.normal(shape=(batch, dim))
+        return z_mean + ops.exp(0.5 * z_log_var) * epsilon
 
 
 """
@@ -94,14 +95,14 @@ class VAE(keras.Model):
         with tf.GradientTape() as tape:
             z_mean, z_log_var, z = self.encoder(data)
             reconstruction = self.decoder(z)
-            reconstruction_loss = tf.reduce_mean(
-                tf.reduce_sum(
+            reconstruction_loss = ops.mean(
+                ops.sum(
                     keras.losses.binary_crossentropy(data, reconstruction),
                     axis=(1, 2),
                 )
             )
-            kl_loss = -0.5 * (1 + z_log_var - tf.square(z_mean) - tf.exp(z_log_var))
-            kl_loss = tf.reduce_mean(tf.reduce_sum(kl_loss, axis=1))
+            kl_loss = -0.5 * (1 + z_log_var - ops.square(z_mean) - ops.exp(z_log_var))
+            kl_loss = ops.mean(ops.sum(kl_loss, axis=1))
             total_loss = reconstruction_loss + kl_loss
         grads = tape.gradient(total_loss, self.trainable_weights)
         self.optimizer.apply_gradients(zip(grads, self.trainable_weights))
(END)

[fchollet]

Thanks for the PR!

[fchollet]

This needs to be seeded via seed=self.seed_generator (to be created in the constructor, self.seed_generator = keras.random.SeedGenerator())

[sitamgithub-MSIT]

I apologize! I overlooked the RNG layers' statelessness. Updated in the most recent commit.

[fchollet]

LGTM, thank you -- please add the generated files.

[sitamgithub-MSIT]

LGTM, thank you -- please add the generated files.

Absolutely! The generated files have been added.