 View on TensorFlow.org
|
 Run in Google Colab
|
 View source on GitHub
|
 Download notebook
|
Overview
Welcome to the end-to-end example for weight clustering, part of the TensorFlow Model Optimization Toolkit.
Other pages
For an introduction to what weight clustering is and to determine if you should use it (including what's supported), see the overview page.
To quickly find the APIs you need for your use case (beyond fully clustering a model with 16 clusters), see the comprehensive guide.
Contents
In the tutorial, you will:
- Train a
tf.kerasmodel for the MNIST dataset from scratch. - Fine-tune the model by applying the weight clustering API and see the accuracy.
- Create a 6x smaller TF and TFLite models from clustering.
- Create a 8x smaller TFLite model from combining weight clustering and post-training quantization.
- See the persistence of accuracy from TF to TFLite.
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
from tensorflow import keras
import numpy as np
import tempfile
import zipfile
import os
2022-12-14 12:04:58.016740: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory 2022-12-14 12:04:58.016842: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory 2022-12-14 12:04:58.016856: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Train a tf.keras model for MNIST without clustering
# Load MNIST dataset
mnist = 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
# Define the model architecture.
model = keras.Sequential([
keras.layers.InputLayer(input_shape=(28, 28)),
keras.layers.Reshape(target_shape=(28, 28, 1)),
keras.layers.Conv2D(filters=12, kernel_size=(3, 3), activation=tf.nn.relu),
keras.layers.MaxPooling2D(pool_size=(2, 2)),
keras.layers.Flatten(),
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 11490434/11490434 [==============================] - 0s 0us/step 2022-12-14 12:04:59.791159: E tensorflow/compiler/xla/stream_executor/cuda/cuda_driver.cc:267] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected Epoch 1/10 1688/1688 [==============================] - 8s 5ms/step - loss: 0.2977 - accuracy: 0.9160 - val_loss: 0.1267 - val_accuracy: 0.9637 Epoch 2/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.1198 - accuracy: 0.9650 - val_loss: 0.0895 - val_accuracy: 0.9758 Epoch 3/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0876 - accuracy: 0.9741 - val_loss: 0.0783 - val_accuracy: 0.9793 Epoch 4/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0719 - accuracy: 0.9788 - val_loss: 0.0675 - val_accuracy: 0.9807 Epoch 5/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0621 - accuracy: 0.9821 - val_loss: 0.0629 - val_accuracy: 0.9822 Epoch 6/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0543 - accuracy: 0.9839 - val_loss: 0.0654 - val_accuracy: 0.9823 Epoch 7/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0490 - accuracy: 0.9855 - val_loss: 0.0597 - val_accuracy: 0.9845 Epoch 8/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0443 - accuracy: 0.9867 - val_loss: 0.0582 - val_accuracy: 0.9862 Epoch 9/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0401 - accuracy: 0.9881 - val_loss: 0.0642 - val_accuracy: 0.9827 Epoch 10/10 1688/1688 [==============================] - 7s 4ms/step - loss: 0.0365 - accuracy: 0.9894 - val_loss: 0.0592 - val_accuracy: 0.9840 <keras.callbacks.History at 0x7f984f91f850>
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: 0.9803000092506409 Saving model to: /tmpfs/tmp/tmp3acdvm6d.h5
Fine-tune the pre-trained model with clustering
Apply the cluster_weights() API to a whole pre-trained model to demonstrate its effectiveness in reducing the model size after applying zip while keeping decent accuracy. For how best to balance the accuracy and compression rate for your use case, please refer to the per layer example in the comprehensive guide.
Define the model and apply the clustering API
Before you pass the model to the clustering API, make sure it is trained and shows some acceptable accuracy.
import tensorflow_model_optimization as tfmot
cluster_weights = tfmot.clustering.keras.cluster_weights
CentroidInitialization = tfmot.clustering.keras.CentroidInitialization
clustering_params = {
'number_of_clusters': 16,
'cluster_centroids_init': CentroidInitialization.LINEAR
}
# Cluster a whole model
clustered_model = cluster_weights(model, **clustering_params)
# Use smaller learning rate for fine-tuning clustered model
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()
Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
cluster_reshape (ClusterWei (None, 28, 28, 1) 0
ghts)
cluster_conv2d (ClusterWeig (None, 26, 26, 12) 244
hts)
cluster_max_pooling2d (Clus (None, 13, 13, 12) 0
terWeights)
cluster_flatten (ClusterWei (None, 2028) 0
ghts)
cluster_dense (ClusterWeigh (None, 10) 40586
ts)
=================================================================
Total params: 40,830
Trainable params: 20,442
Non-trainable params: 20,388
_________________________________________________________________
Fine-tune the model and evaluate the accuracy against baseline
Fine-tune the model with clustering for 1 epoch.
# Fine-tune model
clustered_model.fit(
train_images,
train_labels,
batch_size=500,
epochs=1,
validation_split=0.1)
108/108 [==============================] - 2s 15ms/step - loss: 0.0359 - accuracy: 0.9896 - val_loss: 0.0650 - val_accuracy: 0.9837 <keras.callbacks.History at 0x7f983528a0a0>
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: 0.9803000092506409 Clustered test accuracy: 0.978600025177002
Create 6x smaller models from clustering
Both strip_clustering and applying a standard compression algorithm (e.g. via gzip) are necessary to see the compression benefits of clustering.
First, create a compressible model for TensorFlow. Here, strip_clustering removes all variables (e.g. tf.Variable for storing the cluster centroids and the indices) that clustering only needs during training, which would otherwise add to model size during inference.
final_model = tfmot.clustering.keras.strip_clustering(clustered_model)
_, clustered_keras_file = tempfile.mkstemp('.h5')
print('Saving clustered model to: ', clustered_keras_file)
tf.keras.models.save_model(final_model, clustered_keras_file,
include_optimizer=False)
Saving clustered model to: /tmpfs/tmp/tmpb61tr4z3.h5 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.
Then, create compressible models for TFLite. You can convert the clustered model to a format that's runnable on your targeted backend. TensorFlow Lite is an example you can use to deploy to mobile devices.
clustered_tflite_file = '/tmp/clustered_mnist.tflite'
converter = tf.lite.TFLiteConverter.from_keras_model(final_model)
tflite_clustered_model = converter.convert()
with open(clustered_tflite_file, 'wb') as f:
f.write(tflite_clustered_model)
print('Saved clustered TFLite model to:', clustered_tflite_file)
WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op while saving (showing 1 of 1). These functions will not be directly callable after loading. INFO:tensorflow:Assets written to: /tmpfs/tmp/tmphyio29iz/assets INFO:tensorflow:Assets written to: /tmpfs/tmp/tmphyio29iz/assets Saved clustered TFLite model to: /tmp/clustered_mnist.tflite 2022-12-14 12:06:20.355264: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format. 2022-12-14 12:06:20.355311: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.
Define a helper function to actually compress the models via gzip and measure the zipped size.
def get_gzipped_model_size(file):
# It returns the size of the gzipped model in bytes.
import os
import zipfile
_, 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)
Compare and see that the models are 6x smaller from clustering
print("Size of gzipped baseline Keras model: %.2f bytes" % (get_gzipped_model_size(keras_file)))
print("Size of gzipped clustered Keras model: %.2f bytes" % (get_gzipped_model_size(clustered_keras_file)))
print("Size of gzipped clustered TFlite model: %.2f bytes" % (get_gzipped_model_size(clustered_tflite_file)))
Size of gzipped baseline Keras model: 78148.00 bytes Size of gzipped clustered Keras model: 12248.00 bytes Size of gzipped clustered TFlite model: 11750.00 bytes
Create an 8x smaller TFLite model from combining weight clustering and post-training quantization
You can apply post-training quantization to the clustered model for additional benefits.
converter = tf.lite.TFLiteConverter.from_keras_model(final_model)
converter.optimizations = [tf.lite.Optimize.DEFAULT]
tflite_quant_model = converter.convert()
_, quantized_and_clustered_tflite_file = tempfile.mkstemp('.tflite')
with open(quantized_and_clustered_tflite_file, 'wb') as f:
f.write(tflite_quant_model)
print('Saved quantized and clustered TFLite model to:', quantized_and_clustered_tflite_file)
print("Size of gzipped baseline Keras model: %.2f bytes" % (get_gzipped_model_size(keras_file)))
print("Size of gzipped clustered and quantized TFlite model: %.2f bytes" % (get_gzipped_model_size(quantized_and_clustered_tflite_file)))
WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op while saving (showing 1 of 1). These functions will not be directly callable after loading. INFO:tensorflow:Assets written to: /tmpfs/tmp/tmpvieo83o5/assets INFO:tensorflow:Assets written to: /tmpfs/tmp/tmpvieo83o5/assets Saved quantized and clustered TFLite model to: /tmpfs/tmp/tmpitd7fd5s.tflite Size of gzipped baseline Keras model: 78148.00 bytes Size of gzipped clustered and quantized TFlite model: 9095.00 bytes 2022-12-14 12:06:21.169890: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format. 2022-12-14 12:06:21.169929: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.
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('Evaluated on {n} results so far.'.format(n=i))
# 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 to the TFLite backend.
interpreter = tf.lite.Interpreter(model_content=tflite_quant_model)
interpreter.allocate_tensors()
test_accuracy = eval_model(interpreter)
print('Clustered and quantized TFLite test_accuracy:', test_accuracy)
print('Clustered TF test accuracy:', clustered_model_accuracy)
INFO: Created TensorFlow Lite XNNPACK delegate for CPU. 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: 0.9786 Clustered TF test accuracy: 0.978600025177002
Conclusion
In this tutorial, you saw how to create clustered models with the TensorFlow Model Optimization Toolkit API. More specifically, you've been through an end-to-end example for creating an 8x smaller model for MNIST with minimal accuracy difference. We encourage you to try this new capability, which can be particularly important for deployment in resource-constrained environments.