overview.ipynb

Kernel: Python 3

Copyright 2021 The TensorFlow Authors.

#@title Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

Conversión del modelo Jax para TFLite

Visión general

Nota: Esta API es nueva y sólo está disponible a través de pip install tf-nightly. Estará disponible en la versión 2.7 de TensorFlow. Además, la API es aún experimental y está sujeta a cambios.

Este CodeLab demuestra cómo construir un modelo para el reconocimiento MNIST usando Jax, y cómo convertirlo a TensorFlow Lite. Este codelab también demostrará cómo optimizar el modelo TFLite convertido a Jax con cuantización post-entrenamiento.

Ver en TensorFlow.org

Ejecutar en Google Colab

Ver código fuente en GitHub

Descargar el bloc de notas

Requisitos previos

Se recomienda probar esta característica con la más reciente compilación de TensorFlow nightly pip.

!pip install tf-nightly --upgrade
!pip install jax --upgrade
!pip install jaxlib --upgrade

Preparación de datos

Descargue los datos MNIST con el conjunto de datos Keras y preprocese.

import numpy as np
import tensorflow as tf
import functools

import time
import itertools

import numpy.random as npr

import jax.numpy as jnp
from jax import jit, grad, random
from jax.example_libraries import optimizers
from jax.example_libraries import stax

def _one_hot(x, k, dtype=np.float32):
  """Create a one-hot encoding of x of size k."""
  return np.array(x[:, None] == np.arange(k), dtype)

(train_images, train_labels), (test_images, test_labels) = tf.keras.datasets.mnist.load_data()
train_images, test_images = train_images / 255.0, test_images / 255.0
train_images = train_images.astype(np.float32)
test_images = test_images.astype(np.float32)

train_labels = _one_hot(train_labels, 10)
test_labels = _one_hot(test_labels, 10)

Genere el modelo MNIST con Jax

def loss(params, batch):
  inputs, targets = batch
  preds = predict(params, inputs)
  return -jnp.mean(jnp.sum(preds * targets, axis=1))

def accuracy(params, batch):
  inputs, targets = batch
  target_class = jnp.argmax(targets, axis=1)
  predicted_class = jnp.argmax(predict(params, inputs), axis=1)
  return jnp.mean(predicted_class == target_class)

init_random_params, predict = stax.serial(
    stax.Flatten,
    stax.Dense(1024), stax.Relu,
    stax.Dense(1024), stax.Relu,
    stax.Dense(10), stax.LogSoftmax)

rng = random.PRNGKey(0)

Entrene y evalúe el modelo

step_size = 0.001
num_epochs = 10
batch_size = 128
momentum_mass = 0.9


num_train = train_images.shape[0]
num_complete_batches, leftover = divmod(num_train, batch_size)
num_batches = num_complete_batches + bool(leftover)

def data_stream():
  rng = npr.RandomState(0)
  while True:
    perm = rng.permutation(num_train)
    for i in range(num_batches):
      batch_idx = perm[i * batch_size:(i + 1) * batch_size]
      yield train_images[batch_idx], train_labels[batch_idx]
batches = data_stream()

opt_init, opt_update, get_params = optimizers.momentum(step_size, mass=momentum_mass)

@jit
def update(i, opt_state, batch):
  params = get_params(opt_state)
  return opt_update(i, grad(loss)(params, batch), opt_state)

_, init_params = init_random_params(rng, (-1, 28 * 28))
opt_state = opt_init(init_params)
itercount = itertools.count()

print("\nStarting training...")
for epoch in range(num_epochs):
  start_time = time.time()
  for _ in range(num_batches):
    opt_state = update(next(itercount), opt_state, next(batches))
  epoch_time = time.time() - start_time

  params = get_params(opt_state)
  train_acc = accuracy(params, (train_images, train_labels))
  test_acc = accuracy(params, (test_images, test_labels))
  print("Epoch {} in {:0.2f} sec".format(epoch, epoch_time))
  print("Training set accuracy {}".format(train_acc))
  print("Test set accuracy {}".format(test_acc))

Convertir a modelo TFLite.

Tenga en cuenta que

1. Aplicamos los parámetros en línea a la func predict de Jax con functools.partial.

2. Generamos un jnp.zeros, se trata de un tensor "marcador de posición" usado para que Jax trace el modelo.

3. Llamamos a experimental_from_jax:

• La serving_func se encapsula en una lista.

• La entrada se asocia a un nombre determinado y se pasa como un arreglo encapsulado en una lista.

serving_func = functools.partial(predict, params)
x_input = jnp.zeros((1, 28, 28))
converter = tf.lite.TFLiteConverter.experimental_from_jax(
    [serving_func], [[('input1', x_input)]])
tflite_model = converter.convert()
with open('jax_mnist.tflite', 'wb') as f:
  f.write(tflite_model)

Compruebe el modelo TFLite convertido

Compare los resultados del modelo convertido con el modelo Jax.

expected = serving_func(train_images[0:1])

# Run the model with TensorFlow Lite
interpreter = tf.lite.Interpreter(model_content=tflite_model)
interpreter.allocate_tensors()
input_details = interpreter.get_input_details()
output_details = interpreter.get_output_details()
interpreter.set_tensor(input_details[0]["index"], train_images[0:1, :, :])
interpreter.invoke()
result = interpreter.get_tensor(output_details[0]["index"])

# Assert if the result of TFLite model is consistent with the JAX model.
np.testing.assert_almost_equal(expected, result, 1e-5)

Optimizar el modelo

Daremos un representative_dataset para hacer la cuantización postentrenamiento para optimizar el modelo.

def representative_dataset():
  for i in range(1000):
    x = train_images[i:i+1]
    yield [x]

converter = tf.lite.TFLiteConverter.experimental_from_jax(
    [serving_func], [[('x', x_input)]])
tflite_model = converter.convert()
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.representative_dataset = representative_dataset
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
tflite_quant_model = converter.convert()
with open('jax_mnist_quant.tflite', 'wb') as f:
  f.write(tflite_quant_model)

Evaluar el modelo optimizado

expected = serving_func(train_images[0:1])

# Run the model with TensorFlow Lite
interpreter = tf.lite.Interpreter(model_content=tflite_quant_model)
interpreter.allocate_tensors()
input_details = interpreter.get_input_details()
output_details = interpreter.get_output_details()
interpreter.set_tensor(input_details[0]["index"], train_images[0:1, :, :])
interpreter.invoke()
result = interpreter.get_tensor(output_details[0]["index"])

# Assert if the result of TFLite model is consistent with the Jax model.
np.testing.assert_almost_equal(expected, result, 1e-5)

Compare el tamaño del modelo cuantizado

Deberíamos poder ver que el modelo cuantizado es cuatro veces más pequeño que el modelo original.

!du -h jax_mnist.tflite
!du -h jax_mnist_quant.tflite