Tensorflow函数说明(4)—— variable_scope/name_scope

来源:互联网 发布:企业it数据基础设施 编辑:程序博客网 时间:2024/05/25 19:59

主要针对 tf.get_variable 来介绍共享变量的用法。

tf.get_variable 与 tf.variable 的用法不同。前者在创建变量时会查名字,如果给的名字在之前已经被别的变量占用,则会报错,不会创建相应变量。而后者并不进行检查,如果有重复,则自动的修改名字,加上数字来进行区别。所以从这来看要想共享变量并不能通过使用相同的名字来调用多次 tf.get_variable 和 tf.variable 做到。

比如下面这样的代码:

def my_image_filter(input_images):    conv1_weights = tf.Variable(tf.random_normal([5, 5, 32, 32]),        name="conv1_weights")    conv1_biases = tf.Variable(tf.zeros([32]), name="conv1_biases")    conv1 = tf.nn.conv2d(input_images, conv1_weights,        strides=[1, 1, 1, 1], padding='SAME')    relu1 = tf.nn.relu(conv1 + conv1_biases)    conv2_weights = tf.Variable(tf.random_normal([5, 5, 32, 32]),        name="conv2_weights")    conv2_biases = tf.Variable(tf.zeros([32]), name="conv2_biases")    conv2 = tf.nn.conv2d(relu1, conv2_weights,        strides=[1, 1, 1, 1], padding='SAME')    return tf.nn.relu(conv2 + conv2_biases)
在这个函数中,我们有 'conv1_weights','conv1_biases','conv2_weights','conv2_biases' 4个变量。如果我们重用这个函数,则会产生多组变量,并不会使用相同的变量,如下面调用:

# First call creates one set of variables.result1 = my_image_filter(image1)# Another set is created in the second call.result2 = my_image_filter(image2)
上面实际上用两个不同的滤波器对 image1 和 image2 进行滤波,虽然用的是相同的函数。所以呢,这就产生了问题,下面介绍如何进行变量共享。

我们使用 with tf.variable_scope 来进行共享。比如有下面的代码:

def conv_relu(input, kernel_shape, bias_shape):    # Create variable named "weights".    weights = tf.get_variable("weights", kernel_shape,        initializer=tf.random_normal_initializer())    # Create variable named "biases".    biases = tf.get_variable("biases", bias_shape,        initializer=tf.constant_intializer(0.0))    conv = tf.nn.conv2d(input, weights,        strides=[1, 1, 1, 1], padding='SAME')    return tf.nn.relu(conv + biases)def my_image_filter(input_images):    with tf.variable_scope("conv1"):        # Variables created here will be named "conv1/weights", "conv1/biases".        relu1 = conv_relu(input_images, [5, 5, 32, 32], [32])    with tf.variable_scope("conv2"):        # Variables created here will be named "conv2/weights", "conv2/biases".        return conv_relu(relu1, [5, 5, 32, 32], [32])
若要调用两次 my_image_filter 并且使用相同的变量,则如下所示:

with tf.variable_scope("image_filters") as scope:    result1 = my_image_filter(image1)    scope.reuse_variables()    result2 = my_image_filter(image2)
利用 reuse_variables() 来使变量重用。值得注意的是下面的代码解释了tf.get_variable 工作原理:

with tf.variable_scope("foo"):    v = tf.get_variable("v", [1])with tf.variable_scope("foo", reuse=True):    v1 = tf.get_variable("v", [1])assert v1 == v
如果 reuse 开启,当检查到有相同的名字时,直接返回那个有相同名字的变量而不是重新定义一个再复制值。


下面是使用时需要注意的地方

1. 在 variable_scope 里面的 variable_scope 会继承上面的 reuse 值,即上面一层开启了 reuse ,则下面的也跟着开启。但是不能人为的设置 reuse 为 false ,只有退出 variable_scope 才能让 reuse 变为 false:

with tf.variable_scope("root"):    # At start, the scope is not reusing.    assert tf.get_variable_scope().reuse == False    with tf.variable_scope("foo"):        # Opened a sub-scope, still not reusing.        assert tf.get_variable_scope().reuse == False    with tf.variable_scope("foo", reuse=True):        # Explicitly opened a reusing scope.        assert tf.get_variable_scope().reuse == True        with tf.variable_scope("bar"):            # Now sub-scope inherits the reuse flag.            assert tf.get_variable_scope().reuse == True    # Exited the reusing scope, back to a non-reusing one.    assert tf.get_variable_scope().reuse == False


2. 当在某一 variable_scope 内使用别的 scope 的名字时,此时不再受这里的等级关系束缚,直接与使用的 scope 的名字一样:

with tf.variable_scope("foo") as foo_scope:    assert foo_scope.name == "foo"with tf.variable_scope("bar")    with tf.variable_scope("baz") as other_scope:        assert other_scope.name == "bar/baz"        with tf.variable_scope(foo_scope) as foo_scope2:            assert foo_scope2.name == "foo"  # Not changed.

3. name_scope 与 variable_scope 稍有不同。name_scope 只会影响 ops 的名字,而并不会影响 variables 的名字。

with tf.variable_scope("foo"):    with tf.name_scope("bar"):        v = tf.get_variable("v", [1])        x = 1.0 + vassert v.name == "foo/v:0"assert x.op.name == "foo/bar/add"

1 0
原创粉丝点击