TensorFlow Lite

import tensorflow as tf

# Train or load a Keras model
model = tf.keras.applications.MobileNetV2(
    weights="imagenet",
    input_shape=(224, 224, 3)
)

# Convert to TFLite (no optimization)
converter = tf.lite.TFLiteConverter.from_keras_model(model)
tflite_model = converter.convert()

# Save the model
with open("mobilenet_v2.tflite", "wb") as f:
    f.write(tflite_model)

print(f"Model size: {len(tflite_model) / 1024 / 1024:.1f} MB")

import tensorflow as tf
import numpy as np

# Representative dataset for calibration
def representative_dataset():
    for _ in range(100):
        data = np.random.rand(1, 224, 224, 3).astype(np.float32)
        yield [data]

converter = tf.lite.TFLiteConverter.from_keras_model(model)

# Full integer quantization (INT8)
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.representative_dataset = representative_dataset
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
converter.inference_input_type = tf.int8
converter.inference_output_type = tf.int8

quantized_model = converter.convert()

with open("mobilenet_v2_int8.tflite", "wb") as f:
    f.write(quantized_model)

print(f"Quantized size: {len(quantized_model) / 1024 / 1024:.1f} MB")
# ~3.4 MB vs ~14 MB original (4x smaller)

import numpy as np
import tflite_runtime.interpreter as tflite

# Load TFLite model
interpreter = tflite.Interpreter(model_path="mobilenet_v2_int8.tflite")
interpreter.allocate_tensors()

# Get input/output details
input_details = interpreter.get_input_details()
output_details = interpreter.get_output_details()

# Prepare input image
image = preprocess_image("cat.jpg")  # Resize to 224x224, normalize
interpreter.set_tensor(input_details[0]['index'], image)

# Run inference
interpreter.invoke()

# Get predictions
output = interpreter.get_tensor(output_details[0]['index'])
predicted_class = np.argmax(output)
print(f"Predicted class: {predicted_class}")