tensorboard —— 利用mnist展示tensorboard的基本功能

如下文件，使用 tensorflow中最基础的入门学习例子 mnist，以最直观最简单的方式展示了tensorboard的使用方法

其中包括tensorboard的：scalar、image、histogram、以及特征空间降为展示的代码。

tensorboard_test.py

#coding:utf-8
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import time
import os
import PROJECTOR_visual

"""
权重初始化
初始化为一个接近0的很小的正数
"""
def weight_variable(shape):
    initial = tf.truncated_normal(shape, stddev = 0.1)
    return tf.Variable(initial)

def bias_variable(shape):
    initial = tf.constant(0.1, shape = shape)
    return tf.Variable(initial)

"""
卷积和池化，使用卷积步长为1（stride size）,0边距（padding size）
池化用简单传统的2x2大小的模板做max pooling
"""
def conv2d(x, W):
    return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding = 'SAME')
    # tf.nn.conv2d(input, filter, strides, padding, use_cudnn_on_gpu=None, data_format=None, name=None)
    # x(input)  : [batch, in_height, in_width, in_channels]
    # W(filter) : [filter_height, filter_width, in_channels, out_channels]
    # strides   : The stride of the sliding window for each dimension of input.
    #             For the most common case of the same horizontal and vertices strides, strides = [1, stride, stride, 1]

def max_pool_2x2(x):
    return tf.nn.max_pool(x, ksize = [1, 2, 2, 1],
                          strides = [1, 2, 2, 1], padding = 'SAME')
    # tf.nn.max_pool(value, ksize, strides, padding, data_format='NHWC', name=None)
    # x(value)              : [batch, height, width, channels]
    # ksize(pool大小)        : A list of ints that has length >= 4. The size of the window for each dimension of the input tensor.
    # strides(pool滑动大小)   : A list of ints that has length >= 4. The stride of the sliding window for each dimension of the input tensor.


start = time.clock() #计算开始时间
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True) #MNIST数据输入

os.system("python create_sprite.py")


"""
第一层 卷积层

x_image(batch, 28, 28, 1) -> h_pool1(batch, 14, 14, 32)
"""
x = tf.placeholder(tf.float32,[None, 784])
x_image = tf.reshape(x, [-1, 28, 28, 1]) #最后一维代表通道数目，如果是rgb则为3 

tf.summary.image('input_image', x_image, 10)

W_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])

tf.summary.histogram('W_conv1',W_conv1)
tf.summary.histogram('b_conv1',b_conv1)

conv1 = conv2d(x_image, W_conv1) + b_conv1

h_conv1 = tf.nn.relu(conv1)
# x_image -> [batch, in_height, in_width, in_channels]
#            [batch, 28, 28, 1]
# W_conv1 -> [filter_height, filter_width, in_channels, out_channels]
#            [5, 5, 1, 32]
# output  -> [batch, out_height, out_width, out_channels]
#            [batch, 28, 28, 32]
h_pool1 = max_pool_2x2(h_conv1)
# h_conv1 -> [batch, in_height, in_weight, in_channels]
#            [batch, 28, 28, 32]
# output  -> [batch, out_height, out_weight, out_channels]
#            [batch, 14, 14, 32]

"""
第二层 卷积层

h_pool1(batch, 14, 14, 32) -> h_pool2(batch, 7, 7, 64)
"""
W_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])

tf.summary.histogram('W_conv2',W_conv2)
tf.summary.histogram('b_conv2',b_conv2)

conv2 = conv2d(h_pool1, W_conv2) + b_conv2

h_conv2 = tf.nn.relu(conv2)
# h_pool1 -> [batch, 14, 14, 32]
# W_conv2 -> [5, 5, 32, 64]
# output  -> [batch, 14, 14, 64]
h_pool2 = max_pool_2x2(h_conv2)
# h_conv2 -> [batch, 14, 14, 64]
# output  -> [batch, 7, 7, 64]

"""
反卷积层，为了输出图像而加入的
"""
reverse_weight1 = weight_variable([5,5,32,64])
reverse_conv1 = tf.nn.conv2d_transpose(conv2,reverse_weight1,[50,14,14,32],strides=[1,1,1,1],padding="SAME")
reverse_weight2 = weight_variable([5,5,1,32])
reverse_conv2 = tf.nn.conv2d_transpose(reverse_conv1,reverse_weight2,[50,28,28,1],strides=[1,2,2,1],padding="SAME")

reverse_weight3 = weight_variable([5,5,1,32])
reverse_conv3 = tf.nn.conv2d_transpose(conv1,reverse_weight3,[50,28,28,1],strides=[1,1,1,1],padding="SAME")
tf.summary.image("reverse_conv2",reverse_conv2,10)
tf.summary.image("reverse_conv1",reverse_conv3,10)


"""
第三层 全连接层

h_pool2(batch, 7, 7, 64) -> h_fc1(1, 1024)
"""
W_fc1 = weight_variable([7 * 7 * 64, 1024])
b_fc1 = bias_variable([1024])

tf.summary.histogram('W_fc1',W_fc1)
tf.summary.histogram('b_fc1',b_fc1)

h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)

"""
Dropout

h_fc1 -> h_fc1_drop, 训练中启用，测试中关闭
"""
keep_prob = tf.placeholder("float")
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)

"""
第四层 Softmax输出层
"""
W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])

y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)

"""
训练和评估模型

ADAM优化器来做梯度最速下降,feed_dict中加入参数keep_prob控制dropout比例
"""
y_ = tf.placeholder("float", [None, 10])
cross_entropy = -tf.reduce_sum(y_ * tf.log(y_conv)) #计算交叉熵

train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy) #使用adam优化器来以0.0001的学习率来进行微调
correct_prediction = tf.equal(tf.argmax(y_conv,1), tf.argmax(y_,1)) #判断预测标签和实际标签是否匹配
accuracy = tf.reduce_mean(tf.cast(correct_prediction,"float"))

tf.summary.scalar('cross_entropy', cross_entropy)
tf.summary.scalar('accuracy', accuracy)

merged = tf.summary.merge_all()

sess = tf.Session() #启动创建的模型

writer = tf.summary.FileWriter("logs/", sess.graph)

#sess.run(tf.initialize_all_variables()) #旧版本
sess.run(tf.global_variables_initializer()) #初始化变量

for i in range(PROJECTOR_visual.TRAINING_STEPS): #开始训练模型，循环训练5000次
    batch = mnist.train.next_batch(50) #batch大小设置为50
    if i % 100 == 0:
        train_accuracy = accuracy.eval(session = sess,
                                       feed_dict = {x:batch[0], y_:batch[1], keep_prob:1.0})
        print("step %d, train_accuracy %g" %(i, train_accuracy))
    sess.run(train_step, feed_dict = {x:batch[0], y_:batch[1],
                   keep_prob:0.5}) #神经元输出保持不变的概率 keep_prob 为0.5
    rs, _=sess.run([merged, train_step], feed_dict={x:batch[0], y_:batch[1], keep_prob:1.0})
    writer.add_summary(rs, i)

final_result = sess.run(h_fc1, feed_dict={x:mnist.test.images})

print("test accuracy %g" %accuracy.eval(session = sess,
      feed_dict = {x:mnist.test.images, y_:mnist.test.labels,
                   keep_prob:1.0})) #神经元输出保持不变的概率 keep_prob 为 1，即不变，一直保持输出

PROJECTOR_visual.visualisation(final_result)

　

PPROJECTOR_visual.py

import tensorflow as tf
import os
import tqdm

from tensorflow.contrib.tensorboard.plugins import projector

TRAINING_STEPS = 1000

LOG_DIR = 'logs'
SPRITE_FILE = 'mnist_sprite.jpg'
META_FIEL = "mnist_meta.tsv"
TENSOR_NAME = "FINAL_LOGITS"


# 生成可视化最终输出层向量所需要的日志文件
def visualisation(final_result):
    # 使用一个新的变量来保存最终输出层向量的结果，因为embedding是通过Tensorflow中变量完成的，所以PROJECTOR可视化的都是TensorFlow中的变哇。
    # 所以这里需要新定义一个变量来保存输出层向量的取值
    y_visual = tf.Variable(final_result, name=TENSOR_NAME)
    summary_writer = tf.summary.FileWriter(LOG_DIR)

    # 通过project.ProjectorConfig类来帮助生成日志文件
    config = projector.ProjectorConfig()
    # 增加一个需要可视化的bedding结果
    embedding = config.embeddings.add()
    # 指定这个embedding结果所对应的Tensorflow变量名称
    embedding.tensor_name = y_visual.name

    # Specify where you find the metadata
    # 指定embedding结果所对应的原始数据信息。比如这里指定的就是每一张MNIST测试图片对应的真实类别。在单词向量中可以是单词ID对应的单词。
    # 这个文件是可选的，如果没有指定那么向量就没有标签。
    embedding.metadata_path = META_FIEL

    # Specify where you find the sprite (we will create this later)
    # 指定sprite 图像。这个也是可选的，如果没有提供sprite 图像，那么可视化的结果
    # 每一个点就是一个小困点，而不是具体的图片。
    embedding.sprite.image_path = SPRITE_FILE
    # 在提供sprite图像时，通过single_image_dim可以指定单张图片的大小。
    # 这将用于从sprite图像中截取正确的原始图片。
    embedding.sprite.single_image_dim.extend([28, 28])

    # Say that you want to visualise the embeddings
    # 将PROJECTOR所需要的内容写入日志文件。
    projector.visualize_embeddings(summary_writer, config)

    # 生成会话，初始化新声明的变量并将需要的日志信息写入文件。
    sess = tf.InteractiveSession()
    sess.run(tf.global_variables_initializer())
    saver = tf.train.Saver()
    saver.save(sess, os.path.join(LOG_DIR, "model"), TRAINING_STEPS)

    summary_writer.close()

将两个 .py文件放在同一个文件夹下，然后运行的时候，直接使用cmd，执行 python tensorboard_test.py，便可以启动。

然后在浏览器上输入：http://localhost:8080  便可以打开 tensorboard画面