Dzień Społeczności ML jest 9 listopada! Dołącz do nas na aktualizacje z TensorFlow Jax i więcej Dowiedz się więcej

Przycinanie w celu wnioskowania na urządzeniu z XNNPACK

Zobacz na TensorFlow.org Uruchom w Google Colab Wyświetl źródło na GitHub Pobierz notatnik

Witamy w przewodniku o ciężarach Keras przycinanie poprawy latencji wnioskowania na urządzeniu poprzez XNNPACK .

Przewodnik omawia wykorzystanie nowo wprowadzonej tfmot.sparsity.keras.PruningPolicy API i pokazuje, w jaki sposób może on być wykorzystany do przyspieszenia modele głównie splotowych na nowoczesnych procesorach wykorzystujących XNNPACK Sparse wnioskowania .

Poradnik obejmuje następujące etapy procesu tworzenia modelu:

  • Buduj i trenuj gęstą linię bazową
  • Dostrój model z przycinaniem
  • Konwertuj na TFlite
  • Test porównawczy na urządzeniu

Przewodnik nie obejmuje najlepszych praktyk dotyczących dostrajania za pomocą przycinania. Aby uzyskać więcej szczegółowych informacji na ten temat, proszę sprawdzić nasz obszerny przewodnik .

Ustawiać

 pip install -q tensorflow
 pip install -q tensorflow-model-optimization
import tempfile

import tensorflow as tf
import numpy as np

from tensorflow import keras
import tensorflow_datasets as tfds
import tensorflow_model_optimization as tfmot

%load_ext tensorboard

Zbuduj i wytrenuj gęsty model

Budujemy i trenować prosty bazowego CNN za zadanie klasyfikacji na CIFAR10 zbiorze.

# Load CIFAR10 dataset.
(ds_train, ds_val, ds_test), ds_info = tfds.load(
    'cifar10',
    split=['train[:90%]', 'train[90%:]', 'test'],
    as_supervised=True,
    with_info=True,
)

# Normalize the input image so that each pixel value is between 0 and 1.
def normalize_img(image, label):
  """Normalizes images: `uint8` -> `float32`."""
  return tf.image.convert_image_dtype(image, tf.float32), label

# Load the data in batches of 128 images.
batch_size = 128
def prepare_dataset(ds, buffer_size=None):
  ds = ds.map(normalize_img, num_parallel_calls=tf.data.experimental.AUTOTUNE)
  ds = ds.cache()
  if buffer_size:
    ds = ds.shuffle(buffer_size)
  ds = ds.batch(batch_size)
  ds = ds.prefetch(tf.data.experimental.AUTOTUNE)
  return ds

ds_train = prepare_dataset(ds_train,
                           buffer_size=ds_info.splits['train'].num_examples)
ds_val = prepare_dataset(ds_val)
ds_test = prepare_dataset(ds_test)

# Build the dense baseline model.
dense_model = keras.Sequential([
    keras.layers.InputLayer(input_shape=(32, 32, 3)),
    keras.layers.ZeroPadding2D(padding=1),
    keras.layers.Conv2D(
        filters=8,
        kernel_size=(3, 3),
        strides=(2, 2),
        padding='valid'),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.DepthwiseConv2D(kernel_size=(3, 3), padding='same'),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.Conv2D(filters=16, kernel_size=(1, 1)),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.ZeroPadding2D(padding=1),
    keras.layers.DepthwiseConv2D(
        kernel_size=(3, 3), strides=(2, 2), padding='valid'),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.Conv2D(filters=32, kernel_size=(1, 1)),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.GlobalAveragePooling2D(),
    keras.layers.Flatten(),
    keras.layers.Dense(10)
])

# Compile and train the dense model for 10 epochs.
dense_model.compile(
    loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
    optimizer='adam',
    metrics=['accuracy'])

dense_model.fit(
  ds_train,
  epochs=10,
  validation_data=ds_val)

# Evaluate the dense model.
_, dense_model_accuracy = dense_model.evaluate(ds_test, verbose=0)
2021-08-13 11:13:35.517009: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
2021-08-13 11:13:35.517068: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (kokoro-gcp-ubuntu-prod-1682665100): /proc/driver/nvidia/version does not exist
2021-08-13 11:13:35.517823: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Epoch 1/10
2021-08-13 11:13:36.392179: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)
352/352 [==============================] - 12s 21ms/step - loss: 1.9929 - accuracy: 0.2651 - val_loss: 2.5594 - val_accuracy: 0.1466
Epoch 2/10
352/352 [==============================] - 7s 19ms/step - loss: 1.7293 - accuracy: 0.3582 - val_loss: 1.7533 - val_accuracy: 0.3414
Epoch 3/10
352/352 [==============================] - 7s 19ms/step - loss: 1.6531 - accuracy: 0.3849 - val_loss: 1.6463 - val_accuracy: 0.3886
Epoch 4/10
352/352 [==============================] - 7s 19ms/step - loss: 1.6073 - accuracy: 0.4024 - val_loss: 1.6127 - val_accuracy: 0.3980
Epoch 5/10
352/352 [==============================] - 7s 19ms/step - loss: 1.5692 - accuracy: 0.4200 - val_loss: 1.5552 - val_accuracy: 0.4228
Epoch 6/10
352/352 [==============================] - 7s 19ms/step - loss: 1.5358 - accuracy: 0.4344 - val_loss: 1.6375 - val_accuracy: 0.4030
Epoch 7/10
352/352 [==============================] - 7s 19ms/step - loss: 1.5074 - accuracy: 0.4475 - val_loss: 1.5514 - val_accuracy: 0.4258
Epoch 8/10
352/352 [==============================] - 7s 19ms/step - loss: 1.4810 - accuracy: 0.4598 - val_loss: 1.7087 - val_accuracy: 0.3866
Epoch 9/10
352/352 [==============================] - 7s 19ms/step - loss: 1.4610 - accuracy: 0.4669 - val_loss: 1.5219 - val_accuracy: 0.4492
Epoch 10/10
352/352 [==============================] - 7s 19ms/step - loss: 1.4445 - accuracy: 0.4748 - val_loss: 1.5329 - val_accuracy: 0.4302

Zbuduj rzadki model

Korzystanie z instrukcjami z kompleksowym przewodnikiem stosujemy tfmot.sparsity.keras.prune_low_magnitude funkcji z parametrami, które cel przyspieszenie na urządzeniu poprzez przycinanie tj tfmot.sparsity.keras.PruneForLatencyOnXNNPack polityki.

prune_low_magnitude = tfmot.sparsity.keras.prune_low_magnitude

# Compute end step to finish pruning after after 5 epochs.
end_epoch = 5

num_iterations_per_epoch = len(ds_train)
end_step =  num_iterations_per_epoch * end_epoch

# Define parameters for pruning.
pruning_params = {
      'pruning_schedule': tfmot.sparsity.keras.PolynomialDecay(initial_sparsity=0.25,
                                                               final_sparsity=0.75,
                                                               begin_step=0,
                                                               end_step=end_step),
      'pruning_policy': tfmot.sparsity.keras.PruneForLatencyOnXNNPack()
}

# Try to apply pruning wrapper with pruning policy parameter.
try:
  model_for_pruning = prune_low_magnitude(dense_model, **pruning_params)
except ValueError as e:
  print(e)
Could not find a `GlobalAveragePooling2D` layer with `keepdims = True` in all output branches

Wezwanie prune_low_magnitude skutkuje ValueError z komunikatem Could not find a GlobalAveragePooling2D layer with keepdims = True in all output branches . Komunikat wskazuje, że model nie jest obsługiwany przez przycinanie z polityki tfmot.sparsity.keras.PruneForLatencyOnXNNPack a konkretnie warstwę GlobalAveragePooling2D wymaga się parametrów keepdims = True . Powiedzmy, że poprawka i ponownie prune_low_magnitude funkcja.

fixed_dense_model = keras.Sequential([
    keras.layers.InputLayer(input_shape=(32, 32, 3)),
    keras.layers.ZeroPadding2D(padding=1),
    keras.layers.Conv2D(
        filters=8,
        kernel_size=(3, 3),
        strides=(2, 2),
        padding='valid'),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.DepthwiseConv2D(kernel_size=(3, 3), padding='same'),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.Conv2D(filters=16, kernel_size=(1, 1)),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.ZeroPadding2D(padding=1),
    keras.layers.DepthwiseConv2D(
        kernel_size=(3, 3), strides=(2, 2), padding='valid'),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.Conv2D(filters=32, kernel_size=(1, 1)),
    keras.layers.BatchNormalization(),
    keras.layers.ReLU(),
    keras.layers.GlobalAveragePooling2D(keepdims=True),
    keras.layers.Flatten(),
    keras.layers.Dense(10)
])

# Use the pretrained model for pruning instead of training from scratch.
fixed_dense_model.set_weights(dense_model.get_weights())

# Try to reapply pruning wrapper.
model_for_pruning = prune_low_magnitude(fixed_dense_model, **pruning_params)
/tmpfs/src/tf_docs_env/lib/python3.7/site-packages/keras/engine/base_layer.py:2223: UserWarning: `layer.add_variable` is deprecated and will be removed in a future version. Please use `layer.add_weight` method instead.
  warnings.warn('`layer.add_variable` is deprecated and '

Wywołanie prune_low_magnitude zakończyła bez żadnych błędów, co oznacza, że model jest w pełni obsługiwany przez tfmot.sparsity.keras.PruneForLatencyOnXNNPack polityki i można przyspieszyć stosując XNNPACK Sparse wnioskowania .

Dostrój rzadki model

Po przykład przycinanie , możemy dostroić rzadki model wykorzystujący wagi modelu gęsta. Dostrajanie modelu zaczynamy od 25% rzadkości (25% wag jest ustawione na zero), a kończymy na 75% rzadkości.

logdir = tempfile.mkdtemp()

callbacks = [
  tfmot.sparsity.keras.UpdatePruningStep(),
  tfmot.sparsity.keras.PruningSummaries(log_dir=logdir),
]

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

model_for_pruning.fit(
  ds_train,
  epochs=15,
  validation_data=ds_val,
  callbacks=callbacks)

# Evaluate the dense model.
_, pruned_model_accuracy = model_for_pruning.evaluate(ds_test, verbose=0)

print('Dense model test accuracy:', dense_model_accuracy)
print('Pruned model test accuracy:', pruned_model_accuracy)
2021-08-13 11:14:50.266658: I tensorflow/core/profiler/lib/profiler_session.cc:131] Profiler session initializing.
2021-08-13 11:14:50.266694: I tensorflow/core/profiler/lib/profiler_session.cc:146] Profiler session started.
2021-08-13 11:14:50.833248: I tensorflow/core/profiler/lib/profiler_session.cc:164] Profiler session tear down.
2021-08-13 11:14:50.851018: W tensorflow/python/util/util.cc:348] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.
Epoch 1/15
 10/352 [..............................] - ETA: 8s - loss: 1.4245 - accuracy: 0.5016
2021-08-13 11:14:52.593103: I tensorflow/core/profiler/lib/profiler_session.cc:131] Profiler session initializing.
2021-08-13 11:14:52.593147: I tensorflow/core/profiler/lib/profiler_session.cc:146] Profiler session started.
2021-08-13 11:14:52.617240: I tensorflow/core/profiler/lib/profiler_session.cc:66] Profiler session collecting data.
2021-08-13 11:14:52.619415: I tensorflow/core/profiler/lib/profiler_session.cc:164] Profiler session tear down.
2021-08-13 11:14:52.623098: I tensorflow/core/profiler/rpc/client/save_profile.cc:136] Creating directory: /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52

2021-08-13 11:14:52.625016: I tensorflow/core/profiler/rpc/client/save_profile.cc:142] Dumped gzipped tool data for trace.json.gz to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.trace.json.gz
2021-08-13 11:14:52.628674: I tensorflow/core/profiler/rpc/client/save_profile.cc:136] Creating directory: /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52

2021-08-13 11:14:52.628785: I tensorflow/core/profiler/rpc/client/save_profile.cc:142] Dumped gzipped tool data for memory_profile.json.gz to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.memory_profile.json.gz
2021-08-13 11:14:52.629073: I tensorflow/core/profiler/rpc/client/capture_profile.cc:251] Creating directory: /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52
Dumped tool data for xplane.pb to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.xplane.pb
Dumped tool data for overview_page.pb to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.overview_page.pb
Dumped tool data for input_pipeline.pb to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.input_pipeline.pb
Dumped tool data for tensorflow_stats.pb to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.tensorflow_stats.pb
Dumped tool data for kernel_stats.pb to /tmp/tmpkwu32h8j/train/plugins/profile/2021_08_13_11_14_52/kokoro-gcp-ubuntu-prod-1682665100.kernel_stats.pb
352/352 [==============================] - 9s 20ms/step - loss: 1.4474 - accuracy: 0.4732 - val_loss: 1.5224 - val_accuracy: 0.4368
Epoch 2/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4763 - accuracy: 0.4601 - val_loss: 1.9179 - val_accuracy: 0.3514
Epoch 3/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4861 - accuracy: 0.4602 - val_loss: 1.5849 - val_accuracy: 0.4100
Epoch 4/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4838 - accuracy: 0.4614 - val_loss: 1.5123 - val_accuracy: 0.4412
Epoch 5/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4669 - accuracy: 0.4696 - val_loss: 1.7005 - val_accuracy: 0.3620
Epoch 6/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4497 - accuracy: 0.4772 - val_loss: 1.4644 - val_accuracy: 0.4576
Epoch 7/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4397 - accuracy: 0.4799 - val_loss: 1.4532 - val_accuracy: 0.4710
Epoch 8/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4307 - accuracy: 0.4844 - val_loss: 2.0308 - val_accuracy: 0.3674
Epoch 9/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4254 - accuracy: 0.4849 - val_loss: 1.6031 - val_accuracy: 0.4180
Epoch 10/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4200 - accuracy: 0.4834 - val_loss: 1.8140 - val_accuracy: 0.3768
Epoch 11/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4132 - accuracy: 0.4892 - val_loss: 1.4289 - val_accuracy: 0.4810
Epoch 12/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4075 - accuracy: 0.4915 - val_loss: 1.4257 - val_accuracy: 0.4734
Epoch 13/15
352/352 [==============================] - 7s 19ms/step - loss: 1.4032 - accuracy: 0.4922 - val_loss: 1.4693 - val_accuracy: 0.4620
Epoch 14/15
352/352 [==============================] - 7s 19ms/step - loss: 1.3992 - accuracy: 0.4950 - val_loss: 1.3901 - val_accuracy: 0.4860
Epoch 15/15
352/352 [==============================] - 7s 19ms/step - loss: 1.3957 - accuracy: 0.4952 - val_loss: 1.4754 - val_accuracy: 0.4620
Dense model test accuracy: 0.43209999799728394
Pruned model test accuracy: 0.4596000015735626

Dzienniki pokazują postęp rzadkości w poszczególnych warstwach.

%tensorboard --logdir={logdir}

Po dostrojeniu za pomocą przycinania dokładność testu wykazuje niewielką poprawę (od 43% do 44%) w porównaniu z modelem gęstym. Porównajmy na urządzeniu opóźnieniach przy użyciu TFLite odniesienia .

Konwersja modeli i benchmarking

Aby przekonwertować wyczyszczony modelu do TFLite, potrzebujemy wymienić PruneLowMagnitude owijarki z oryginalnymi warstwami poprzez strip_pruning funkcji. Ponadto, ponieważ ciężary usuwane modelu ( model_for_pruning ) są głównie zera, możemy zastosować optymalizacji tf.lite.Optimize.EXPERIMENTAL_SPARSITY efektywnie przechowywać spowodowało modelu TFLite. Ta flaga optymalizacji nie jest wymagana dla gęstego modelu.

converter = tf.lite.TFLiteConverter.from_keras_model(dense_model)
dense_tflite_model = converter.convert()

_, dense_tflite_file = tempfile.mkstemp('.tflite')
with open(dense_tflite_file, 'wb') as f:
  f.write(dense_tflite_model)

model_for_export = tfmot.sparsity.keras.strip_pruning(model_for_pruning)

converter = tf.lite.TFLiteConverter.from_keras_model(model_for_export)
converter.optimizations = [tf.lite.Optimize.EXPERIMENTAL_SPARSITY]
pruned_tflite_model = converter.convert()

_, pruned_tflite_file = tempfile.mkstemp('.tflite')
with open(pruned_tflite_file, 'wb') as f:
  f.write(pruned_tflite_model)
INFO:tensorflow:Assets written to: /tmp/tmp0yx5e3fy/assets
INFO:tensorflow:Assets written to: /tmp/tmp0yx5e3fy/assets
2021-08-13 11:16:36.564681: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0
2021-08-13 11:16:36.564926: I tensorflow/core/grappler/clusters/single_machine.cc:357] Starting new session
2021-08-13 11:16:36.568512: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:1137] Optimization results for grappler item: graph_to_optimize
  function_optimizer: function_optimizer did nothing. time = 0.008ms.
  function_optimizer: function_optimizer did nothing. time = 0.001ms.
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.
2021-08-13 11:16:36.664551: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:351] Ignored output_format.
2021-08-13 11:16:36.664597: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:354] Ignored drop_control_dependency.
2021-08-13 11:16:36.668981: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:210] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.
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/tmpenn8hns6/assets
INFO:tensorflow:Assets written to: /tmp/tmpenn8hns6/assets
2021-08-13 11:16:39.184787: I tensorflow/core/grappler/devices.cc:66] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0
2021-08-13 11:16:39.185019: I tensorflow/core/grappler/clusters/single_machine.cc:357] Starting new session
2021-08-13 11:16:39.188948: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:1137] Optimization results for grappler item: graph_to_optimize
  function_optimizer: function_optimizer did nothing. time = 0.01ms.
  function_optimizer: function_optimizer did nothing. time = 0.002ms.

2021-08-13 11:16:39.294765: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:351] Ignored output_format.
2021-08-13 11:16:39.294816: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:354] Ignored drop_control_dependency.

Postępując zgodnie z instrukcjami TFLite modelu Benchmarking Tool i budujemy narzędzia, przesłać go do urządzenia Android wraz z gęstej i przycinać modeli TFLite i odniesienia obu modeli na urządzeniu.

! adb shell /data/local/tmp/benchmark_model \
    --graph=/data/local/tmp/dense_model.tflite \
    --use_xnnpack=true \
    --num_runs=100 \
    --num_threads=1
/bin/bash: adb: command not found
! adb shell /data/local/tmp/benchmark_model \
    --graph=/data/local/tmp/pruned_model.tflite \
    --use_xnnpack=true \
    --num_runs=100 \
    --num_threads=1
/bin/bash: adb: command not found

Benchmarki na piksel 4 spowodowało średni czas wnioskowania o 17us dla gęstej modelu i 12us dla usuwane modelu. Benchmarki na urządzeniu wykazywać wyraźne 5US lub 30% poprawę latencji nawet dla takich małych modeli. Z naszego doświadczenia wynika, większe modele oparte na MobileNetV3 lub EfficientNet-lite występów podobnej wydajności. Przyspieszenie różni się w zależności od względnego udziału zwojów 1x1 w całym modelu.

Wniosek

W tym samouczku pokazujemy, w jaki sposób można tworzyć rzadkie modele dla szybszej wydajności na urządzeniu, korzystając z nowej funkcjonalności wprowadzonej przez API TF MOT i XNNPack. Te rzadkie modele są mniejsze i szybsze niż ich gęste odpowiedniki, zachowując, a nawet przewyższając ich jakość.

Zachęcamy do wypróbowania tej nowej funkcji, która może być szczególnie ważna przy wdrażaniu modeli na urządzeniu.