【TensorFlow】TensorBoard的使用（二）

在《【TensorFlow】TensorBoard的使用（一）》中讲述了如何简单的使用TensorBoard，但是生成的图比较乱，本文旨在对生成的图进行优化。

优化主要分为如下步骤：

1. 定义操作名字的作用域：with tf.name_scope(name):
2. 定义张量的name值：w = tf.Variable(tf.truncated_normal([5, 5, size_in, size_out], stddev=0.1), name='W')

本文在《【TensorFlow】TensorBoard的使用（一）》示例的基础上对TensorBoard进行优化。

示例

实现以上两步优化，代码如下：

mnist_board_2.py：

import os
import tensorflow as tf

LOGDIR = './mnist'

mnist = tf.contrib.learn.datasets.mnist.read_data_sets(train_dir=LOGDIR + 'data', one_hot=True)

# 加上name值，方便在tensorboard里面查看
def conv_layer(input, size_in, size_out, name='conv'):
    # 定义名字作用域
    with tf.name_scope(name):
        w = tf.Variable(tf.truncated_normal([5, 5, size_in, size_out], stddev=0.1), name='W')
        b = tf.Variable(tf.constant(0.1, shape=[size_out]), name='B')
        conv = tf.nn.conv2d(input, w, strides=[1, 1, 1, 1], padding='SAME')
        act = tf.nn.relu(conv + b)

        return tf.nn.max_pool(act, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')


def fc_layer(input, size_in, size_out, name='fc'):
    with tf.name_scope(name):
        w = tf.Variable(tf.truncated_normal([size_in, size_out], stddev=0.1), name='W')
        b = tf.Variable(tf.constant(0.1, shape=[size_out]), name='B')
        act = tf.nn.relu(tf.matmul(input, w) + b)

        return act


def mnist_model(learning_rate, use_two_conv, use_two_fc, hparam):
    tf.reset_default_graph()
    sess = tf.Session()

    # setup placeholders, and reshape the data
    x = tf.placeholder(tf.float32, shape=[None, 784], name='x')
    x_image = tf.reshape(x, [-1, 28, 28, 1])

    y = tf.placeholder(tf.float32, shape=[None, 10], name='labels')

    if use_two_conv:
        conv1 = conv_layer(x_image, 1, 32, 'conv1')
        conv_out = conv_layer(conv1, 32, 64, 'conv2')

    else:
        conv1 = conv_layer(x_image, 1, 64, 'conv')
        conv_out = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')

    flattened = tf.reshape(conv_out, [-1, 7 * 7 * 64])

    if use_two_fc:
        fc1 = fc_layer(flattened, 7 * 7 * 64, 1024, 'fc1')
        embedding_input = fc1
        embedding_size = 1024
        logits = fc_layer(fc1, 1024, 10, 'fc2')

    else:
        embedding_input = flattened
        embedding_size = 7 * 7 * 64
        logits = fc_layer(flattened, 7 * 7 * 64, 10, 'fc')

    with tf.name_scope('loss'):
        xent = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=y), name='loss')
        # tf.summary.scalar('loss', xent)

    with tf.name_scope('train'):
        train_step = tf.train.AdamOptimizer(learning_rate).minimize(xent)

    with tf.name_scope('accuracy'):
        correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(y, 1))
        accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
        # tf.summary.scalar('accuracy', accuracy)

    # summ = tf.summary.merge_all()

    emdedding = tf.Variable(tf.zeros([1024, embedding_size]))
    # assignment = emdedding.assign(embedding_input)

    sess.run(tf.global_variables_initializer())
    # 保存路径
    tenboard_dir = './tensorboard/test2/'

    # 指定一个文件用来保存图
    writer = tf.summary.FileWriter(tenboard_dir + hparam)
    # 把图add进去
    writer.add_graph(sess.graph)

    for i in range(2001):
        batch = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={x: batch[0], y: batch[1]})


def make_hparam_string(learning_rate, use_two_fc, use_two_conv):
    conv_param = 'conv=2' if use_two_conv else 'conv=1'
    fc_param = 'fc=2' if use_two_fc else 'fc=1'
    return 'lr_%.0E,%s,%s' % (learning_rate, conv_param, fc_param)


def main():
    # You can try adding some more learning rates
    for learning_rate in [1E-4]:

        # Include 'False' as a value to try different model architectures.
        for use_two_fc in [True]:
            for use_two_conv in [True]:
                # Construct a hyperparameter string for each one(example: 'lr_1E-3,fc=2,conv=2')
                hparam = make_hparam_string(learning_rate, use_two_fc, use_two_conv)
                print('Starting run for %s' % hparam)

                # Actually run with the new settings
                mnist_model(learning_rate, use_two_fc, use_two_conv, hparam)


if __name__ == '__main__':
    main()

展示

执行上述程序，生成test2文件夹，运行TensorBoard，显示如下：

由上图可以很明显的看出，输入的x经过reshape操作后，执行conv1-->conv2-->fc1-->fc2操作，最后经过和标签labels相比对，生成loss、accuracy值，图中也展示了train环节的各张量的维度。

上图虽然能清晰的展示出train的流程，但是其展示的结果较少，用户想了解权重w和偏置b的分布情况、损失loss和精度accuracy的变化情况等，都查看不了。

下篇文章讲述如何针对上述情况进行优化——丰富TensorBoard的展示结果。