Help protect the Great Barrier Reef with TensorFlow on Kaggle Join Challenge

Cluster preserving quantization aware training (CQAT) Keras example

View on TensorFlow.org Run in Google Colab View on GitHub Download notebook

Overview

This is an end to end example showing the usage of the cluster preserving quantization aware training (CQAT) API, part of the TensorFlow Model Optimization Toolkit's collaborative optimization pipeline.

Other pages

For an introduction to the pipeline and other available techniques, see the collaborative optimization overview page.

Contents

In the tutorial, you will:

  1. Train a tf.keras model for the MNIST dataset from scratch.
  2. Fine-tune the model with clustering and see the accuracy.
  3. Apply QAT and observe the loss of clusters.
  4. Apply CQAT and observe that the clustering applied earlier has been preserved.
  5. Generate a TFLite model and observe the effects of applying CQAT on it.
  6. Compare the achieved CQAT model accuracy with a model quantized using post-training quantization.

Setup

You can run this Jupyter Notebook in your local virtualenv or colab. For details of setting up dependencies, please refer to the installation guide.

 pip install -q tensorflow-model-optimization
import tensorflow as tf

import numpy as np
import tempfile
import zipfile
import os

Train a tf.keras model for MNIST without clustering

# Load MNIST dataset
mnist = tf.keras.datasets.mnist
(train_images, train_labels), (test_images, test_labels) = mnist.load_data()

# Normalize the input image so that each pixel value is between 0 to 1.
train_images = train_images / 255.0
test_images  = test_images / 255.0

model = tf.keras.Sequential([
  tf.keras.layers.InputLayer(input_shape=(28, 28)),
  tf.keras.layers.Reshape(target_shape=(28, 28, 1)),
  tf.keras.layers.Conv2D(filters=12, kernel_size=(3, 3),
                         activation=tf.nn.relu),
  tf.keras.layers.MaxPooling2D(pool_size=(2, 2)),
  tf.keras.layers.Flatten(),
  tf.keras.layers.Dense(10)
])

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

model.fit(
    train_images,
    train_labels,
    validation_split=0.1,
    epochs=10
)
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz
11493376/11490434 [==============================] - 0s 0us/step
Epoch 1/10
1688/1688 [==============================] - 8s 5ms/step - loss: 0.2827 - accuracy: 0.9211 - val_loss: 0.1057 - val_accuracy: 0.9728
Epoch 2/10
1688/1688 [==============================] - 8s 5ms/step - loss: 0.1070 - accuracy: 0.9696 - val_loss: 0.0768 - val_accuracy: 0.9793
Epoch 3/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0805 - accuracy: 0.9765 - val_loss: 0.0663 - val_accuracy: 0.9823
Epoch 4/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0671 - accuracy: 0.9804 - val_loss: 0.0614 - val_accuracy: 0.9840
Epoch 5/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0592 - accuracy: 0.9827 - val_loss: 0.0594 - val_accuracy: 0.9837
Epoch 6/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0537 - accuracy: 0.9842 - val_loss: 0.0610 - val_accuracy: 0.9838
Epoch 7/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0487 - accuracy: 0.9851 - val_loss: 0.0644 - val_accuracy: 0.9827
Epoch 8/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0437 - accuracy: 0.9862 - val_loss: 0.0624 - val_accuracy: 0.9847
Epoch 9/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0399 - accuracy: 0.9882 - val_loss: 0.0595 - val_accuracy: 0.9847
Epoch 10/10
1688/1688 [==============================] - 7s 4ms/step - loss: 0.0362 - accuracy: 0.9885 - val_loss: 0.0613 - val_accuracy: 0.9835
<tensorflow.python.keras.callbacks.History at 0x7f24a4cc7990>

Evaluate the baseline model and save it for later usage

_, baseline_model_accuracy = model.evaluate(
    test_images, test_labels, verbose=0)

print('Baseline test accuracy:', baseline_model_accuracy)

_, keras_file = tempfile.mkstemp('.h5')
print('Saving model to: ', keras_file)
tf.keras.models.save_model(model, keras_file, include_optimizer=False)
Baseline test accuracy: 0.9814000129699707
Saving model to:  /tmp/tmppd3opqzk.h5

Cluster and fine-tune the model with 8 clusters

Apply the cluster_weights() API to cluster the whole pre-trained model to demonstrate and observe its effectiveness in reducing the model size when applying zip, while maintaining accuracy. For how best to use the API to achieve the best compression rate while maintaining your target accuracy, refer to the clustering comprehensive guide.

Define the model and apply the clustering API

The model needs to be pre-trained before using the clustering API.

import tensorflow_model_optimization as tfmot

cluster_weights = tfmot.clustering.keras.cluster_weights
CentroidInitialization = tfmot.clustering.keras.CentroidInitialization

clustering_params = {
  'number_of_clusters': 8,
  'cluster_centroids_init': CentroidInitialization.KMEANS_PLUS_PLUS
}

clustered_model = cluster_weights(model, **clustering_params)

# Use smaller learning rate for fine-tuning
opt = tf.keras.optimizers.Adam(learning_rate=1e-5)

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

clustered_model.summary()
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.7/site-packages/tensorflow/python/ops/array_ops.py:5049: calling gather (from tensorflow.python.ops.array_ops) with validate_indices is deprecated and will be removed in a future version.
Instructions for updating:
The `validate_indices` argument has no effect. Indices are always validated on CPU and never validated on GPU.
Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
cluster_reshape (ClusterWeig (None, 28, 28, 1)         0         
_________________________________________________________________
cluster_conv2d (ClusterWeigh (None, 26, 26, 12)        236       
_________________________________________________________________
cluster_max_pooling2d (Clust (None, 13, 13, 12)        0         
_________________________________________________________________
cluster_flatten (ClusterWeig (None, 2028)              0         
_________________________________________________________________
cluster_dense (ClusterWeight (None, 10)                40578     
=================================================================
Total params: 40,814
Trainable params: 20,426
Non-trainable params: 20,388
_________________________________________________________________

Fine-tune the model and evaluate the accuracy against baseline

Fine-tune the model with clustering for 3 epochs.

# Fine-tune model
clustered_model.fit(
  train_images,
  train_labels,
  epochs=3,
  validation_split=0.1)
Epoch 1/3
1688/1688 [==============================] - 8s 5ms/step - loss: 0.0363 - accuracy: 0.9887 - val_loss: 0.0610 - val_accuracy: 0.9830
Epoch 2/3
1688/1688 [==============================] - 8s 5ms/step - loss: 0.0333 - accuracy: 0.9898 - val_loss: 0.0599 - val_accuracy: 0.9830
Epoch 3/3
1688/1688 [==============================] - 8s 5ms/step - loss: 0.0319 - accuracy: 0.9905 - val_loss: 0.0580 - val_accuracy: 0.9837
<tensorflow.python.keras.callbacks.History at 0x7f240d271990>

Define helper functions to calculate and print the number of clustering in each kernel of the model.

def print_model_weight_clusters(model):

    for layer in model.layers:
        if isinstance(layer, tf.keras.layers.Wrapper):
            weights = layer.trainable_weights
        else:
            weights = layer.weights
        for weight in weights:
            # ignore auxiliary quantization weights
            if "quantize_layer" in weight.name:
                continue
            if "kernel" in weight.name:
                unique_count = len(np.unique(weight))
                print(
                    f"{layer.name}/{weight.name}: {unique_count} clusters "
                )

Check that the model kernels were correctly clustered. We need to strip the clustering wrapper first.

stripped_clustered_model = tfmot.clustering.keras.strip_clustering(clustered_model)

print_model_weight_clusters(stripped_clustered_model)
conv2d/kernel:0: 8 clusters 
dense/kernel:0: 8 clusters

For this example, there is minimal loss in test accuracy after clustering, compared to the baseline.

_, clustered_model_accuracy = clustered_model.evaluate(
  test_images, test_labels, verbose=0)

print('Baseline test accuracy:', baseline_model_accuracy)
print('Clustered test accuracy:', clustered_model_accuracy)
Baseline test accuracy: 0.9814000129699707
Clustered test accuracy: 0.9800999760627747

Apply QAT and CQAT and check effect on model clusters in both cases

Next, we apply both QAT and cluster preserving QAT (CQAT) on the clustered model and observe that CQAT preserves weight clusters in your clustered model. Note that we stripped clustering wrappers from your model with tfmot.clustering.keras.strip_clustering before applying CQAT API.

# QAT
qat_model = tfmot.quantization.keras.quantize_model(stripped_clustered_model)

qat_model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])
print('Train qat model:')
qat_model.fit(train_images, train_labels, batch_size=128, epochs=1, validation_split=0.1)

# CQAT
quant_aware_annotate_model = tfmot.quantization.keras.quantize_annotate_model(
              stripped_clustered_model)
cqat_model = tfmot.quantization.keras.quantize_apply(
              quant_aware_annotate_model,
              tfmot.experimental.combine.Default8BitClusterPreserveQuantizeScheme())

cqat_model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])
print('Train cqat model:')
cqat_model.fit(train_images, train_labels, batch_size=128, epochs=1, validation_split=0.1)
Train qat model:
422/422 [==============================] - 4s 8ms/step - loss: 0.0328 - accuracy: 0.9901 - val_loss: 0.0578 - val_accuracy: 0.9840
WARNING:root:Input layer does not contain zero weights, so apply CQAT instead.
WARNING:root:Input layer does not contain zero weights, so apply CQAT instead.
Train cqat model:
WARNING:tensorflow:Gradients do not exist for variables ['conv2d/kernel:0', 'dense/kernel:0'] when minimizing the loss.
WARNING:tensorflow:Gradients do not exist for variables ['conv2d/kernel:0', 'dense/kernel:0'] when minimizing the loss.
WARNING:tensorflow:Gradients do not exist for variables ['conv2d/kernel:0', 'dense/kernel:0'] when minimizing the loss.
WARNING:tensorflow:Gradients do not exist for variables ['conv2d/kernel:0', 'dense/kernel:0'] when minimizing the loss.
422/422 [==============================] - 4s 8ms/step - loss: 0.0310 - accuracy: 0.9908 - val_loss: 0.0592 - val_accuracy: 0.9838
<tensorflow.python.keras.callbacks.History at 0x7f240c60e6d0>
print("QAT Model clusters:")
print_model_weight_clusters(qat_model)
print("CQAT Model clusters:")
print_model_weight_clusters(cqat_model)
QAT Model clusters:
quant_conv2d/conv2d/kernel:0: 108 clusters 
quant_dense/dense/kernel:0: 19931 clusters 
CQAT Model clusters:
quant_conv2d/conv2d/kernel:0: 8 clusters 
quant_dense/dense/kernel:0: 8 clusters

See compression benefits of CQAT model

Define helper function to get zipped model file.

def get_gzipped_model_size(file):
  # It returns the size of the gzipped model in kilobytes.

  _, zipped_file = tempfile.mkstemp('.zip')
  with zipfile.ZipFile(zipped_file, 'w', compression=zipfile.ZIP_DEFLATED) as f:
    f.write(file)

  return os.path.getsize(zipped_file)/1000

Note that this is a small model. Applying clustering and CQAT to a bigger production model would yield a more significant compression.

# QAT model
converter = tf.lite.TFLiteConverter.from_keras_model(qat_model)
converter.optimizations = [tf.lite.Optimize.DEFAULT]
qat_tflite_model = converter.convert()
qat_model_file = 'qat_model.tflite'
# Save the model.
with open(qat_model_file, 'wb') as f:
    f.write(qat_tflite_model)

# CQAT model
converter = tf.lite.TFLiteConverter.from_keras_model(cqat_model)
converter.optimizations = [tf.lite.Optimize.DEFAULT]
cqat_tflite_model = converter.convert()
cqat_model_file = 'cqat_model.tflite'
# Save the model.
with open(cqat_model_file, 'wb') as f:
    f.write(cqat_tflite_model)

print("QAT model size: ", get_gzipped_model_size(qat_model_file), ' KB')
print("CQAT model size: ", get_gzipped_model_size(cqat_model_file), ' KB')
WARNING:absl:Found untraced functions such as reshape_layer_call_and_return_conditional_losses, reshape_layer_call_fn, conv2d_layer_call_and_return_conditional_losses, conv2d_layer_call_fn, flatten_layer_call_and_return_conditional_losses while saving (showing 5 of 20). These functions will not be directly callable after loading.
INFO:tensorflow:Assets written to: /tmp/tmpy_b1e0tx/assets
INFO:tensorflow:Assets written to: /tmp/tmpy_b1e0tx/assets
WARNING:absl:Found untraced functions such as reshape_layer_call_and_return_conditional_losses, reshape_layer_call_fn, conv2d_layer_call_and_return_conditional_losses, conv2d_layer_call_fn, flatten_layer_call_and_return_conditional_losses while saving (showing 5 of 20). These functions will not be directly callable after loading.
INFO:tensorflow:Assets written to: /tmp/tmp282pfq0n/assets
INFO:tensorflow:Assets written to: /tmp/tmp282pfq0n/assets
QAT model size:  16.685  KB
CQAT model size:  10.121  KB

See the persistence of accuracy from TF to TFLite

Define a helper function to evaluate the TFLite model on the test dataset.

def eval_model(interpreter):
  input_index = interpreter.get_input_details()[0]["index"]
  output_index = interpreter.get_output_details()[0]["index"]

  # Run predictions on every image in the "test" dataset.
  prediction_digits = []
  for i, test_image in enumerate(test_images):
    if i % 1000 == 0:
      print(f"Evaluated on {i} results so far.")
    # Pre-processing: add batch dimension and convert to float32 to match with
    # the model's input data format.
    test_image = np.expand_dims(test_image, axis=0).astype(np.float32)
    interpreter.set_tensor(input_index, test_image)

    # Run inference.
    interpreter.invoke()

    # Post-processing: remove batch dimension and find the digit with highest
    # probability.
    output = interpreter.tensor(output_index)
    digit = np.argmax(output()[0])
    prediction_digits.append(digit)

  print('\n')
  # Compare prediction results with ground truth labels to calculate accuracy.
  prediction_digits = np.array(prediction_digits)
  accuracy = (prediction_digits == test_labels).mean()
  return accuracy

You evaluate the model, which has been clustered and quantized, and then see the accuracy from TensorFlow persists in the TFLite backend.

interpreter = tf.lite.Interpreter(cqat_model_file)
interpreter.allocate_tensors()

cqat_test_accuracy = eval_model(interpreter)

print('Clustered and quantized TFLite test_accuracy:', cqat_test_accuracy)
print('Clustered TF test accuracy:', clustered_model_accuracy)
Evaluated on 0 results so far.
Evaluated on 1000 results so far.
Evaluated on 2000 results so far.
Evaluated on 3000 results so far.
Evaluated on 4000 results so far.
Evaluated on 5000 results so far.
Evaluated on 6000 results so far.
Evaluated on 7000 results so far.
Evaluated on 8000 results so far.
Evaluated on 9000 results so far.


Clustered and quantized TFLite test_accuracy: 0.9795
Clustered TF test accuracy: 0.9800999760627747

Apply post-training quantization and compare to CQAT model

Next, we use post-training quantization (no fine-tuning) on the clustered model and check its accuracy against the CQAT model. This demonstrates why you would need to use CQAT to improve the quantized model's accuracy.

First, define a generator for the callibration dataset from the first 1000 training images.

def mnist_representative_data_gen():
  for image in train_images[:1000]:  
    image = np.expand_dims(image, axis=0).astype(np.float32)
    yield [image]

Quantize the model and compare accuracy to the previously acquired CQAT model. Note that the model quantized with fine-tuning achieves higher accuracy.

converter = tf.lite.TFLiteConverter.from_keras_model(stripped_clustered_model)
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.representative_dataset = mnist_representative_data_gen
post_training_tflite_model = converter.convert()
post_training_model_file = 'post_training_model.tflite'
# Save the model.
with open(post_training_model_file, 'wb') as f:
    f.write(post_training_tflite_model)

# Compare accuracy
interpreter = tf.lite.Interpreter(post_training_model_file)
interpreter.allocate_tensors()

post_training_test_accuracy = eval_model(interpreter)

print('CQAT TFLite test_accuracy:', cqat_test_accuracy)
print('Post-training (no fine-tuning) TF test accuracy:', post_training_test_accuracy)
WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.
WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.
INFO:tensorflow:Assets written to: /tmp/tmp6vw7f4l2/assets
INFO:tensorflow:Assets written to: /tmp/tmp6vw7f4l2/assets
Evaluated on 0 results so far.
Evaluated on 1000 results so far.
Evaluated on 2000 results so far.
Evaluated on 3000 results so far.
Evaluated on 4000 results so far.
Evaluated on 5000 results so far.
Evaluated on 6000 results so far.
Evaluated on 7000 results so far.
Evaluated on 8000 results so far.
Evaluated on 9000 results so far.


CQAT TFLite test_accuracy: 0.9795
Post-training (no fine-tuning) TF test accuracy: 0.9804

Conclusion

In this tutorial, you learned how to create a model, cluster it using the cluster_weights() API, and apply the cluster preserving quantization aware training (CQAT) to preserve clusters while using QAT. The final CQAT model was compared to the QAT one to show that the clusters are preserved in the former and lost in the latter. Next, the models were converted to TFLite to show the compression benefits of chaining clustering and CQAT model optimization techniques and the TFLite model was evaluated to ensure that the accuracy persists in the TFLite backend. Finally, the CQAT model was compared to a quantized clustered model achieved using the post-training quantization API to demonstrate the advantage of CQAT in recovering the accuracy loss from normal quantization.