此页面由 Cloud Translation API 翻译。
Switch to English

权聚类综合指南

查看上TensorFlow.org 在谷歌Colab运行 GitHub上查看源代码 下载笔记本

欢迎对权聚类的TensorFlow模型优化工具包的一部分的综合指南。

本页介绍各种使用情况,并展示了如何使用API​​的每一个。一旦你知道你需要哪些API,发现在参数和低层次的细节的API文档

  • 如果你想看到权聚类什么好处支持的功能,检查概述
  • 对于单端至端例如,参照权聚类例子

在本指南中,以下用例涵盖:

  • 定义一个集群模型。
  • 检查点和反序列化集群模式。
  • 提高集群模型的准确性。
  • 仅用于部署,必须采取措施,以压缩的优势。

建立

 ! pip install -q tensorflow-model-optimization

import tensorflow as tf
import numpy as np
import tempfile
import os
import tensorflow_model_optimization as tfmot

input_dim = 20
output_dim = 20
x_train = np.random.randn(1, input_dim).astype(np.float32)
y_train = tf.keras.utils.to_categorical(np.random.randn(1), num_classes=output_dim)

def setup_model():
  model = tf.keras.Sequential([
      tf.keras.layers.Dense(input_dim, input_shape=[input_dim]),
      tf.keras.layers.Flatten()
  ])
  return model

def train_model(model):
  model.compile(
      loss=tf.keras.losses.categorical_crossentropy,
      optimizer='adam',
      metrics=['accuracy']
  )
  model.summary()
  model.fit(x_train, y_train)
  return model

def save_model_weights(model):
  _, pretrained_weights = tempfile.mkstemp('.h5')
  model.save_weights(pretrained_weights)
  return pretrained_weights

def setup_pretrained_weights():
  model= setup_model()
  model = train_model(model)
  pretrained_weights = save_model_weights(model)
  return pretrained_weights

def setup_pretrained_model():
  model = setup_model()
  pretrained_weights = setup_pretrained_weights()
  model.load_weights(pretrained_weights)
  return model

def save_model_file(model):
  _, keras_file = tempfile.mkstemp('.h5') 
  model.save(keras_file, include_optimizer=False)
  return keras_file

def get_gzipped_model_size(model):
  # It returns the size of the gzipped model in bytes.
  import os
  import zipfile

  keras_file = save_model_file(model)

  _, zipped_file = tempfile.mkstemp('.zip')
  with zipfile.ZipFile(zipped_file, 'w', compression=zipfile.ZIP_DEFLATED) as f:
    f.write(keras_file)
  return os.path.getsize(zipped_file)

setup_model()
pretrained_weights = setup_pretrained_weights()
 
Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_1 (Dense)              (None, 20)                420       
_________________________________________________________________
flatten_1 (Flatten)          (None, 20)                0         
=================================================================
Total params: 420
Trainable params: 420
Non-trainable params: 0
_________________________________________________________________
1/1 [==============================] - 0s 1ms/step - loss: 1.3192 - accuracy: 0.0000e+00

定义一个集群模型

簇整体模型(顺序的和功能的)

为更好的模型精度提示

  • 你必须通过与可接受的精度预训练模式,这个API。从头聚类的准确性欠佳结果的培训模式。
  • 在某些情况下,汇聚某些层对模型的准确性产生不利影响。检查“集群一些层”来看看如何跳过聚类影响精度的大部分层。

要集群中的所有层,应用tfmot.clustering.keras.cluster_weights到模型。

 import tensorflow_model_optimization as tfmot

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

clustering_params = {
  'number_of_clusters': 3,
  'cluster_centroids_init': CentroidInitialization.DENSITY_BASED
}

model = setup_model()
model.load_weights(pretrained_weights)

clustered_model = cluster_weights(model, **clustering_params)

clustered_model.summary()
 
Model: "sequential_2"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
cluster_dense_2 (ClusterWeig (None, 20)                423       
_________________________________________________________________
cluster_flatten_2 (ClusterWe (None, 20)                0         
=================================================================
Total params: 423
Trainable params: 23
Non-trainable params: 400
_________________________________________________________________

簇一些层(顺序和功能模型)

为更好的模型精度提示

  • 你必须通过与可接受的精度预训练模式,这个API。从头聚类的准确性欠佳结果的培训模式。
  • 群集后来层用更多的冗余参数(例如tf.keras.layers.Densetf.keras.layers.Conv2D ),相对于早期的层。
  • 冻结之前微调期间群集层初层。把冷冻层为超参数的数量。根据经验,冻结最早层是非常适合当前集群API。
  • 避免聚类关键层(例如注意机制)。

以上tfmot.clustering.keras.cluster_weights API文档提供关于如何改变每一层的聚类配置细节。

 # Create a base model
base_model = setup_model()
base_model.load_weights(pretrained_weights)

# Helper function uses `cluster_weights` to make only 
# the Dense layers train with clustering
def apply_clustering_to_dense(layer):
  if isinstance(layer, tf.keras.layers.Dense):
    return cluster_weights(layer, **clustering_params)
  return layer

# Use `tf.keras.models.clone_model` to apply `apply_clustering_to_dense` 
# to the layers of the model.
clustered_model = tf.keras.models.clone_model(
    base_model,
    clone_function=apply_clustering_to_dense,
)

clustered_model.summary()
 
Model: "sequential_3"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
cluster_dense_3 (ClusterWeig (None, 20)                423       
_________________________________________________________________
flatten_3 (Flatten)          (None, 20)                0         
=================================================================
Total params: 423
Trainable params: 23
Non-trainable params: 400
_________________________________________________________________

检查点和反序列化集群模型

你的使用情况:只需要在HDF5模型格式(不HDF5权或其他格式)的代码。

 # Define the model.
base_model = setup_model()
base_model.load_weights(pretrained_weights)
clustered_model = cluster_weights(base_model, **clustering_params)

# Save or checkpoint the model.
_, keras_model_file = tempfile.mkstemp('.h5')
clustered_model.save(keras_model_file, include_optimizer=True)

# `cluster_scope` is needed for deserializing HDF5 models.
with tfmot.clustering.keras.cluster_scope():
  loaded_model = tf.keras.models.load_model(keras_model_file)

loaded_model.summary()
 
WARNING:tensorflow:No training configuration found in the save file, so the model was *not* compiled. Compile it manually.
Model: "sequential_4"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
cluster_dense_4 (ClusterWeig (None, 20)                423       
_________________________________________________________________
cluster_flatten_4 (ClusterWe (None, 20)                0         
=================================================================
Total params: 423
Trainable params: 23
Non-trainable params: 400
_________________________________________________________________

提高集群模型的准确性

为了您的具体使用情况下,您可以考虑建议:

  • 质心初始化起着最终的优化模型精度的关键作用。在一般情况下,线性初始化性能优于密度和随机初始化,因为它不容易错过大的权重。然而,密度初始化已观察到提供更好的准确性使用与双峰分布权重很少集群的情况。

  • 设置一个学习速度比在训练时使用的微调群集模式一个下。

  • 对于一般的想法下提高模型的准确性,看看为你的使用情况(S)提示“定义集群模式”。

部署

与尺寸压缩的出口模式

常见错误 :既strip_clustering和应用标准的压缩算法(例如,通过gzip的)是必要的,看到聚集的压缩优点。

 model = setup_model()
clustered_model = cluster_weights(model, **clustering_params)

clustered_model.compile(
    loss=tf.keras.losses.categorical_crossentropy,
    optimizer='adam',
    metrics=['accuracy']
)

clustered_model.fit(
    x_train,
    y_train
)

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

print("final model")
final_model.summary()

print("\n")
print("Size of gzipped clustered model without stripping: %.2f bytes" 
      % (get_gzipped_model_size(clustered_model)))
print("Size of gzipped clustered model with stripping: %.2f bytes" 
      % (get_gzipped_model_size(final_model)))
 
1/1 [==============================] - 0s 1ms/step - loss: 2.2136 - accuracy: 0.0000e+00
final model
Model: "sequential_5"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_5 (Dense)              (None, 20)                420       
_________________________________________________________________
flatten_5 (Flatten)          (None, 20)                0         
=================================================================
Total params: 420
Trainable params: 420
Non-trainable params: 0
_________________________________________________________________


Size of gzipped clustered model without stripping: 1822.00 bytes
Size of gzipped clustered model with stripping: 1408.00 bytes