Untitled

import tensorflow as tf
from tensorflow.keras.applications import ResNet50
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.callbacks import ReduceLROnPlateau

# Load pre-trained ResNet50 model
base_model = ResNet50(weights='imagenet', include_top=False, input_shape=(img_height, img_width, 3))

# Freeze all layers except the last 5
for layer in base_model.layers[:-5]:
    layer.trainable = False

# Add custom layers
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(1024, activation='relu')(x)
predictions = Dense(len(lb.classes_), activation='softmax')(x)

# Create the model
model = Model(inputs=base_model.input, outputs=predictions)

# Compile the model
model.compile(optimizer=Adam(learning_rate=1e-4), loss='categorical_crossentropy', metrics=['accuracy'])

# Train the model
history = model.fit(
    train_generator,
    epochs=8,
    validation_data=val_generator,
    callbacks=[ReduceLROnPlateau(monitor='val_loss', factor=0.2, patience=2, min_lr=1e-6)]
)


for layer in model.layers:
    layer.trainable = True

model.compile(optimizer=Adam(learning_rate=1e-4), loss='categorical_crossentropy', metrics=['accuracy'])

history = model.fit(
    train_generator,
    epochs=3,
    validation_data=val_generator,
    callbacks=[ReduceLROnPlateau(monitor='val_loss', factor=0.2, patience=2, min_lr=1e-6)]
)

# Evaluate the model
loss, accuracy = model.evaluate(val_generator)
print(f'Test loss: {loss:.3f}')
print(f'Test accuracy: {accuracy:.3f}')

# Save the model
model.save('resnet50_fine_tuned.h5')