TensorFlow 学习2（方便日后查阅用法）

最新推荐文章于 2022-09-20 08:58:14 发布

aiqiu_gogogo

最新推荐文章于 2022-09-20 08:58:14 发布

阅读量684

点赞数

分类专栏： tensorflow学习文章标签： tensorflow deep learning

tensorflow学习专栏收录该内容

4 篇文章 0 订阅

订阅专栏

TensorFlow——slim模块

转自：https://www.2cto.com/kf/201706/649266.html

作用：代码瘦身，类似于Keras，TensorLayer，tfLearn，目的是消除原生tensorflow里面很多重复的模板性的代码，让代码更紧凑，更具备可读性，使模型的构建，训练，测试都变得更加简单；

导入方法：import tensorflow.contrib.slim as slim

什么是tensorflow.contrib？
官方说明：此目录中的任何代码未经官方支持，可能会随时更改或删除。每个目录下都有指定的所有者。它旨在包含额外功能和贡献，最终会合并到核心TensorFlow中，但其接口可能仍然会发生变化。所以slim依然不属于原生tensorflow。

slim中变量分为两类：模型变量和局部变量，局部变量是不作为模型参数保存的，而模型变量会再保存的时候保存下来；

定义模型变量并获取模型变量：

weights = slim.model_variable('weights', shape=[20,20,20,20],
                              initializer=tf.truncated_normal_initializer(stddev=0.1),
                              regularizer=slim.l2_regularizer(0.05))
model_variables = slim.get_model_variables()

定义局部变量并获取所有变量：

local_var = slim.variable('local_var ', shape=[20,20,20,20],
                       initializer=tf.zeros_initializer())
local_and_model_variables = slim.get_variables()

简单的例子

构建一个卷积层：

input = ...
net = slim.conv2d(input, 128, [3, 3], scope='conv1_1')

构建三个一样的卷积层：

net = ...
net = slim.conv2d(net, 256, [3, 3], scope='conv3_1')
net = slim.conv2d(net, 256, [3, 3], scope='conv3_2')
net = slim.conv2d(net, 256, [3, 3], scope='conv3_3')
net = slim.max_pool2d(net, [2, 2], scope='pool2')

使用repeat简化上述操作：

net = slim.repeat(net, 3, slim.conv2d, 256, [3, 3], scope='conv3')
net = slim.max_pool2d(net, [2, 2], scope='pool2')

构建三个不一样的卷积层：

x = slim.conv2d(x, 32, [3, 3], scope='core/core_1')
x = slim.conv2d(x, 32, [1, 1], scope='core/core_2')
x = slim.conv2d(x, 64, [3, 3], scope='core/core_3')
x = slim.conv2d(x, 64, [1, 1], scope='core/core_4')

使用stack简化上述操作：

slim.stack(x, slim.conv2d, [(32, [3, 3]), (32, [1, 1]), (64, [3, 3]), (64, [1, 1])], scope='core')

构建三个全连接层：

x = slim.fully_connected(x, 32, scope='fc/fc_1')
x = slim.fully_connected(x, 64, scope='fc/fc_2')
x = slim.fully_connected(x, 128, scope='fc/fc_3')

使用stack简化上述操作：

slim.stack(x, slim.fully_connected, [32, 64, 128], scope='fc')

再复杂点儿（arg_scope的使用）

用于简化需要定义的网络有很多参数；

定义三个卷积层（每个卷积层参数都比较多）：

net = slim.conv2d(inputs, 64, [11, 11], 4, padding='SAME',
                  weights_initializer=tf.truncated_normal_initializer(stddev=0.01),
                  weights_regularizer=slim.l2_regularizer(0.0005), scope='conv1')
net = slim.conv2d(net, 128, [11, 11], padding='VALID',
                  weights_initializer=tf.truncated_normal_initializer(stddev=0.01),
                  weights_regularizer=slim.l2_regularizer(0.0005), scope='conv2')
net = slim.conv2d(net, 256, [11, 11], padding='SAME',
                  weights_initializer=tf.truncated_normal_initializer(stddev=0.01),
                  weights_regularizer=slim.l2_regularizer(0.0005), scope='conv3')

使用arg_scope简化上述操作：

with slim.arg_scope([slim.conv2d], padding='SAME',
                      weights_initializer=tf.truncated_normal_initializer(stddev=0.01)
                      weights_regularizer=slim.l2_regularizer(0.0005)):
    net = slim.conv2d(inputs, 64, [11, 11], scope='conv1')
    net = slim.conv2d(net, 128, [11, 11], padding='VALID', scope='conv2')
    net = slim.conv2d(net, 256, [11, 11], scope='conv3')

说明：arg_scope的作用范围内，是定义了指定层的默认参数，若想特别指定某些层的参数，可以重新赋值；

写个VGG16呢

def vgg16(inputs):
  with slim.arg_scope([slim.conv2d, slim.fully_connected],
                      activation_fn=tf.nn.relu,
                      weights_initializer=tf.truncated_normal_initializer(0.0, 0.01),
                      weights_regularizer=slim.l2_regularizer(0.0005)):
    net = slim.repeat(inputs, 2, slim.conv2d, 64, [3, 3], scope='conv1')
    net = slim.max_pool2d(net, [2, 2], scope='pool1')
    net = slim.repeat(net, 2, slim.conv2d, 128, [3, 3], scope='conv2')
    net = slim.max_pool2d(net, [2, 2], scope='pool2')
    net = slim.repeat(net, 3, slim.conv2d, 256, [3, 3], scope='conv3')
    net = slim.max_pool2d(net, [2, 2], scope='pool3')
    net = slim.repeat(net, 3, slim.conv2d, 512, [3, 3], scope='conv4')
    net = slim.max_pool2d(net, [2, 2], scope='pool4')
    net = slim.repeat(net, 3, slim.conv2d, 512, [3, 3], scope='conv5')
    net = slim.max_pool2d(net, [2, 2], scope='pool5')
    net = slim.fully_connected(net, 4096, scope='fc6')
    net = slim.dropout(net, 0.5, scope='dropout6')
    net = slim.fully_connected(net, 4096, scope='fc7')
    net = slim.dropout(net, 0.5, scope='dropout7')
    net = slim.fully_connected(net, 1000, activation_fn=None, scope='fc8')
  return net

训练一下，OK？

# Load the images and labels.
images, labels = ...
# Create the model.
predictions, _ = vgg16(images)
# Define the loss functions and get the total loss.
loss = slim.losses.softmax_cross_entropy(predictions, labels)