tf.slim使用方法

首先要导入slim：

from tensorflow.contrib import slim

TF-slim主要由下面几个组成：

arg_scope
data
evluation
layers
learning
losses
metrics
nets
queues
regularizers
variables

Layer

最常用的是 slim 的 Layers ， 创建 Layer 十分方便：

input = ...
net = slim.conv2d(input, 128, [3, 3], scope='conv1_1')
net = slim.max_pool2d(net, kernel_size=[2,2], stride=2, scope='pool1')
# 一般为 (inputs=?, kernel_size=?, stride=?, padding=?, ....)
net = slim.repeat(net, 3, slim.conv2d, 256, [3, 3], scope='conv3')
# repeat操作即为重复创建某个layer

Layer	TF-Slim
BiasAdd	slim.bias_add
BatchNorm	slim.batch_norm
Conv2d	slim.conv2d
Conv2dInPlane	slim.conv2d_in_plane
Conv2dTranspose (Deconv)	slim.conv2d_transpose
FullyConnected	slim.fully_connected
AvgPool2D	slim.avg_pool2d
Dropout	slim.dropout
Flatten	slim.flatten
MaxPool2D	slim.max_pool2d
OneHotEncoding	slim.one_hot_encoding
SeparableConv2	slim.separable_conv2d
UnitNorm	slim.unit_norm

# 所有的参数如下
@add_arg_scope
def convolution2d_in_plane(
    inputs,
    kernel_size,
    stride=1,
    padding='SAME',
    activation_fn=nn.relu,
    normalizer_fn=None,
    normalizer_params=None,
    weights_initializer=initializers.xavier_initializer(),
    weights_regularizer=None,
    biases_initializer=init_ops.zeros_initializer(),
    biases_regularizer=None,
    reuse=None,
    variables_collections=None,
    outputs_collections=None,
    trainable=True,
    scope=None):

Scopes:

使用scope也十分方便
使用arg_scope后可以避免在每个slim.conv2d中添加重复的参数（如weight_initializer=xxx等）

# 为里面的每个slim.conv2d中添加同样的参数
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')

Variables:

可像如下创建：

weights = slim.variable('weights',
                             shape=[10, 10, 3 , 3],
                             initializer=tf.truncated_normal_initializer(stddev=0.1),
                             regularizer=slim.l2_regularizer(0.05),
                             device='/CPU:0')

# Model Variables
weights = slim.model_variable('weights',
                              shape=[10, 10, 3 , 3],
                              initializer=tf.truncated_normal_initializer(stddev=0.1),
                              regularizer=slim.l2_regularizer(0.05),
                              device='/CPU:0')
model_variables = slim.get_model_variables()

# Regular variables
my_var = slim.variable('my_var',
                       shape=[20, 1],
                       initializer=tf.zeros_initializer())
regular_variables_and_model_variables = slim.get_variables()

Training Loop

简易的训练方式，调用 slim.learning.create_train_op 和 slim.learning.train 来实现

# 调用 slim.learning.create_train_op 和 slim.learning.train 来实现
g = tf.Graph()

# Create the model and specify the losses...
...

total_loss = slim.losses.get_total_loss()
optimizer = tf.train.GradientDescentOptimizer(learning_rate)

# create_train_op ensures that each time we ask for the loss, the update_ops
# are run and the gradients being computed are applied too.
train_op = slim.learning.create_train_op(total_loss, optimizer)
logdir = ... # Where checkpoints are stored.

slim.learning.train(
    train_op,
    logdir,
    number_of_steps=1000,
    save_summaries_secs=300,
    save_interval_secs=600):

其他

还有其他的方法本文没写，只把作者（小白）使用的较多的列出
其他的可以参考
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/slim

Reference：

TensorFlow-Slim：
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/slim