Update app.py

app.py

@@ -6,7 +6,6 @@ import tensorflow as tf
 from tensorflow import keras
 from tensorflow.keras import layers
 from tensorflow.keras.models import Sequential
-from PIL import Image
 import matplotlib.pyplot as plt
 import gradio as gr
 import numpy as np
@@ -105,37 +104,42 @@ for images, _ in train_ds.take(1):
         plt.imshow(augmented_images[0].numpy().astype("uint8"))
         plt.axis("off")
 
-# Define a deeper CNN model with softmax activation in the final layer
+# Define a deeper CNN model with more regularization techniques
 num_classes = len(class_names)
-model = Sequential([
-    data_augmentation,
-    layers.Rescaling(1./255),
-    
-    layers.Conv2D(32, 3, padding='same', activation='relu'),
-    layers.MaxPooling2D(),
-    
-    layers.Conv2D(64, 3, padding='same', activation='relu'),
-    layers.MaxPooling2D(),
-    
-    layers.Conv2D(128, 3, padding='same', activation='relu'),
-    layers.MaxPooling2D(),
-    
-    layers.Conv2D(256, 3, padding='same', activation='relu'),
-    layers.MaxPooling2D(),
-    
-    layers.Conv2D(512, 3, padding='same', activation='relu'),
-    layers.MaxPooling2D(),
-    
-    layers.Dropout(0.5),
-    layers.Flatten(),
-    
-    layers.Dense(256, activation='relu'),
-    layers.Dropout(0.5),
-    
-    layers.Dense(num_classes, activation='softmax', name="outputs")
-])
-
-model.compile(optimizer='adam',
+model = Sequential()
+
+model.add(data_augmentation)
+model.add(layers.Rescaling(1./255))
+
+model.add(layers.Conv2D(32, 3, padding='same', activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D())
+
+model.add(layers.Conv2D(64, 3, padding='same', activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D())
+
+model.add(layers.Conv2D(128, 3, padding='same', activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D())
+
+model.add(layers.Conv2D(256, 3, padding='same', activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D())
+
+model.add(layers.Conv2D(512, 3, padding='same', activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D())
+
+model.add(layers.Dropout(0.5))
+model.add(layers.Flatten())
+
+model.add(layers.Dense(256, activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)))
+model.add(layers.Dropout(0.5))
+
+model.add(layers.Dense(num_classes, activation='softmax', name="outputs"))
+
+model.compile(optimizer=keras.optimizers.Adam(learning_rate=1e-4),
               loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
               metrics=['accuracy'])
 
@@ -197,7 +201,6 @@ body {
 }
 """
 
-
 gr.Interface(
     fn=predict_image,
     inputs=image,