Update model.py

model.py

@@ -102,16 +102,22 @@ norm_layer.adapt(data=train_spectrogram_ds.map(map_func=lambda spec, label: spec
 model = models.Sequential([
     layers.Input(shape=input_shape),
     # Downsample the input.
+    #This layer resizes the input to a target size of 32x32. It is commonly used to ensure a consistent input shape for subsequent layers.
     layers.Resizing(32, 32),
     # Normalize.
+    #This is the normalization layer that you previously adapted with the training data.
     norm_layer,
+    #This is a convolutional layer with 32 filters, a kernel size of 3x3, and ReLU activation. It performs convolutional operations on the input.
     layers.Conv2D(32, 3, activation='relu'),
     layers.Conv2D(64, 3, activation='relu'),
+    #This layer performs max pooling, which reduces the spatial dimensions of the input data.
     layers.MaxPooling2D(),
     layers.Dropout(0.25),
+    #This layer flattens the input data into a 1-dimensional vector, preparing it for the fully connected layers.
     layers.Flatten(),
     layers.Dense(128, activation='relu'),
     layers.Dropout(0.5),
+    #This is the output layer with num_labels units, representing the number of output classes in your classification problem.
     layers.Dense(num_labels),
 ])