Model optimization (quantization, pruning) in TensorFlow

Introduction

Model optimization helps make machine learning models smaller and faster. This is useful to run models on devices like phones or tablets.

You want your model to run quickly on a smartphone.

You need to save storage space by making the model smaller.

You want to reduce the power consumption of your model on edge devices.

You want to deploy a model on hardware with limited memory.

You want faster predictions without losing much accuracy.

Syntax

TensorFlow

import tensorflow_model_optimization as tfmot

# For pruning
pruning_params = {
    'pruning_schedule': tfmot.sparsity.keras.PolynomialDecay(
        initial_sparsity=0.0,
        final_sparsity=0.5,
        begin_step=0,
        end_step=end_step
    )
}
model = tfmot.sparsity.keras.prune_low_magnitude(model, **pruning_params)

# For quantization aware training
quantize_model = tfmot.quantization.keras.quantize_model
quantized_model = quantize_model(model)

Pruning removes less important connections to make the model sparse.

Quantization reduces the precision of numbers to make the model smaller and faster.

Examples

This example prunes 30% of the model weights gradually during training.

TensorFlow

import tensorflow_model_optimization as tfmot

# Pruning example
pruning_params = {
    'pruning_schedule': tfmot.sparsity.keras.PolynomialDecay(
        initial_sparsity=0.0,
        final_sparsity=0.3,
        begin_step=0,
        end_step=1000
    )
}
pruned_model = tfmot.sparsity.keras.prune_low_magnitude(model, **pruning_params)

This prepares the model for quantization aware training to keep accuracy high.

TensorFlow

import tensorflow_model_optimization as tfmot

# Quantization aware training example
quantize_model = tfmot.quantization.keras.quantize_model
quantized_model = quantize_model(model)

Sample Model

This program trains a simple model on MNIST data, then applies pruning to reduce model size. It shows accuracy before and after pruning.

TensorFlow

import tensorflow as tf
import tensorflow_model_optimization as tfmot

# Load example data
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

# Simple model
model = tf.keras.Sequential([
    tf.keras.layers.Flatten(input_shape=(28, 28)),
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dense(10)
])

model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

# Train original model
model.fit(x_train, y_train, epochs=1, verbose=0)

# Evaluate original model
loss, acc = model.evaluate(x_test, y_test, verbose=0)

# Pruning setup
batch_size = 128
epochs = 2
validation_split = 0.1
num_images = x_train.shape[0] * (1 - validation_split)
end_step = int(num_images / batch_size) * epochs

pruning_params = {
    'pruning_schedule': tfmot.sparsity.keras.PolynomialDecay(
        initial_sparsity=0.0,
        final_sparsity=0.5,
        begin_step=0,
        end_step=end_step
    )
}

# Apply pruning
pruned_model = tfmot.sparsity.keras.prune_low_magnitude(model, **pruning_params)

pruned_model.compile(optimizer='adam',
                     loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
                     metrics=['accuracy'])

callbacks = [tfmot.sparsity.keras.UpdatePruningStep()]

# Train pruned model
pruned_model.fit(x_train, y_train,
                 batch_size=batch_size,
                 epochs=epochs,
                 validation_split=validation_split,
                 callbacks=callbacks,
                 verbose=0)

# Evaluate pruned model
loss_pruned, acc_pruned = pruned_model.evaluate(x_test, y_test, verbose=0)

print(f"Original model accuracy: {acc:.4f}")
print(f"Pruned model accuracy: {acc_pruned:.4f}")

Important Notes

Pruning usually requires retraining the model to keep accuracy.

Quantization can be done after training or during training (quantization aware training).

Always test your optimized model to check if accuracy is still good enough.

Summary

Model optimization makes models smaller and faster for devices with limited resources.

Pruning removes less important parts of the model to reduce size.

Quantization reduces number precision to speed up the model.

Practice

1. What is the main goal of model pruning in TensorFlow model optimization?

easy

A. Remove less important weights to reduce model size

B. Convert model weights to lower precision

C. Add more layers to improve accuracy

D. Increase the batch size during training

Solution

1. Step 1: Understand pruning concept

   Pruning means removing parts of the model that contribute less, usually weights with small values.

2. Step 2: Identify pruning effect

   This reduces model size and speeds up inference by making the model simpler.

3. Final Answer:

   Remove less important weights to reduce model size -> Option A

4. Quick Check:

   Pruning = Remove less important weights [OK]

Hint: Pruning cuts small weights to shrink model size [OK]

Common Mistakes:

• Confusing pruning with quantization
• Thinking pruning adds layers
• Believing pruning changes batch size

2. Which of the following is the correct way to apply post-training quantization in TensorFlow Lite?

easy

A. converter.optimizations = [tf.lite.Optimize.DEFAULT]

B. model.compile(optimizer='adam', loss='mse')

C. tf.keras.layers.Dense(10, activation='relu')

D. model.fit(x_train, y_train, epochs=10)

Solution

1. Step 1: Recognize quantization syntax

   Post-training quantization in TensorFlow Lite uses the converter.optimizations property set to [tf.lite.Optimize.DEFAULT].

2. Step 2: Eliminate unrelated options

   Other options relate to training or model layers, not quantization.

3. Final Answer:

   converter.optimizations = [tf.lite.Optimize.DEFAULT] -> Option A

4. Quick Check:

   Quantization uses converter.optimizations = [tf.lite.Optimize.DEFAULT] [OK]

Hint: Quantization uses converter.optimizations setting [OK]

Common Mistakes:

• Confusing training code with quantization setup
• Using model.compile for quantization
• Applying quantization inside model.fit

3. What will be the output shape of the pruned model after applying TensorFlow Model Optimization pruning API on a model with input shape (None, 28, 28, 1) and a Dense layer with 10 units?

medium

A. (None, 28, 28, 10)

B. (None, 10)

C. (None, 28, 28, 1)

D. (10,)

Solution

1. Step 1: Understand pruning effect on shape

   Pruning removes weights but does not change the model architecture or output shape.

2. Step 2: Identify output shape of Dense layer

   A Dense layer with 10 units outputs shape (None, 10) regardless of pruning.

3. Final Answer:

   (None, 10) -> Option B

4. Quick Check:

   Pruning keeps output shape same = (None, 10) [OK]

Hint: Pruning doesn't change output shape, only weights [OK]

Common Mistakes:

• Thinking pruning changes output dimensions
• Confusing input shape with output shape
• Assuming pruning adds extra dimensions

4. You applied pruning to your TensorFlow model but noticed no size reduction after saving. What is the likely cause?

medium

A. You used quantization instead of pruning

B. Pruning increases model size by adding masks

C. Model was not stripped of pruning wrappers before saving

D. Model was saved in HDF5 format

Solution

1. Step 1: Understand pruning wrappers

   Pruning adds wrappers around layers that must be stripped before saving to reduce size.

2. Step 2: Identify saving step

   If wrappers are not stripped, the saved model keeps extra pruning info, so size doesn't reduce.

3. Final Answer:

   Model was not stripped of pruning wrappers before saving -> Option C

4. Quick Check:

   Strip pruning wrappers before saving = smaller model [OK]

Hint: Strip pruning wrappers before saving model [OK]

Common Mistakes:

• Assuming pruning always reduces size without stripping
• Confusing pruning with quantization effects
• Ignoring model saving format impact

5. You want to optimize a TensorFlow model for deployment on a mobile device with limited memory and CPU. Which combination of techniques is best to reduce model size and speed up inference without much accuracy loss?

hard

A. Convert model to float64 precision

B. Only increase training epochs without optimization

C. Use dropout layers and batch normalization

D. Apply pruning during training and then post-training quantization

Solution

1. Step 1: Identify optimization goals

   For mobile deployment, reducing size and speeding inference are key, while keeping accuracy.

2. Step 2: Choose suitable techniques

   Pruning removes unnecessary weights, quantization reduces number precision, both reduce size and speed up inference.

3. Step 3: Eliminate wrong options

   Increasing epochs doesn't optimize size; dropout and batch norm are for training regularization; float64 increases size.

4. Final Answer:

   Apply pruning during training and then post-training quantization -> Option D

5. Quick Check:

   Pruning + quantization = smaller, faster model [OK]

Hint: Combine pruning and quantization for best optimization [OK]

Common Mistakes:

• Ignoring quantization after pruning
• Confusing training tricks with optimization
• Using higher precision increases size