TensorFlow——TensorBoard应用进阶

最新推荐文章于 2021-06-06 11:32:44 发布

Sam Chou

最新推荐文章于 2021-06-06 11:32:44 发布

阅读量185

点赞数

分类专栏： TensorFlow

本文链接：https://blog.csdn.net/weixin_45602979/article/details/108870336

版权

TensorFlow 专栏收录该内容

15 篇文章 1 订阅

订阅专栏

1、TensorBoard1

import tensorflow as tf
import tensorflow.examples.tutorials.mnist.input_data as input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

x = tf.placeholder(tf.float32, [None, 784])
y_actual = tf.placeholder(tf.float32, [None, 10])

y_ = tf.layers.dense(x, 500, activation='relu')
output = tf.layers.dense(y_, 10)
# 求交叉熵
loss = tf.losses.softmax_cross_entropy(onehot_labels=y_actual, logits=output)
# 用梯度下降法使得残差最小
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(loss)

# 在测试阶段，测试准确度计算
correct_prediction = tf.equal(tf.argmax(output, 1), tf.argmax(y_actual, 1))
# 多个批次的准确度均值
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

with tf.Session() as sess:
    init = tf.global_variables_initializer()
    sess.run(init)
    writer = tf.summary.FileWriter('./logs/1')
    writer.add_graph(sess.graph)
    for i in range(100000):
        batch_xs, batch_ys = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={x: batch_xs, y_actual: batch_ys})
        if i % 100 == 0:
            print("test_accuracy:", sess.run(accuracy, feed_dict={x: mnist.test.images, y_actual: mnist.test.labels}))

2、TensorBoard2

import tensorflow as tf
import tensorflow.examples.tutorials.mnist.input_data as input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

x = tf.placeholder(tf.float32, [None, 784], name='input_x')
y_actual = tf.placeholder(tf.float32, [None, 10], name='input_y')

y_ = tf.layers.dense(x, 500, activation='relu', name='dense')
output = tf.layers.dense(y_, 10, name='output')
# 求交叉熵
loss = tf.losses.softmax_cross_entropy(onehot_labels=y_actual, logits=output)
# 用梯度下降法使得残差最小
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(loss)

# 在测试阶段，测试准确度计算
correct_prediction = tf.equal(tf.argmax(output, 1), tf.argmax(y_actual, 1), name='test_precision')
# 多个批次的准确度均值
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

with tf.Session() as sess:
    init = tf.global_variables_initializer()
    sess.run(init)
    tf.summary.FileWriter('./logs/2', sess.graph)

    for i in range(100000):
        batch_xs, batch_ys = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={x: batch_xs, y_actual: batch_ys})
        if i % 100 == 0:
            print("test_accuracy:", sess.run(accuracy, feed_dict={x: mnist.test.images, y_actual: mnist.test.labels}))

3、TensorBoard3

import tensorflow as tf
import tensorflow.examples.tutorials.mnist.input_data as input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

x = tf.placeholder(tf.float32, [None, 784], name='input_x')
y_actual = tf.placeholder(tf.float32, [None, 10], name='input_y')

y_ = tf.layers.dense(x, 500, activation='relu', name='dense')
output = tf.layers.dense(y_, 10, name='output')
# 求交叉熵
loss = tf.losses.softmax_cross_entropy(onehot_labels=y_actual, logits=output)
# 用梯度下降法使得残差最小
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(loss)
# 在测试阶段，测试准确度计算
correct_prediction = tf.equal(tf.argmax(output, 1), tf.argmax(y_actual, 1), name='test_precision')
# 多个批次的准确度均值
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

tf.summary.scalar('loss', loss)
tf.summary.scalar('accuracy', accuracy)
merged_summary = tf.summary.merge_all()

with tf.Session() as sess:
    init = tf.global_variables_initializer()
    sess.run(init)
    writer = tf.summary.FileWriter('./logs/3')
    writer.add_graph(sess.graph)
    for i in range(100000):
        batch_xs, batch_ys = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={x: batch_xs, y_actual: batch_ys})
        if i % 100 == 0:
            test_accuracy, ss = sess.run([accuracy, merged_summary], feed_dict={x: mnist.test.images,
                                                                                y_actual: mnist.test.labels})
            print("test_accuracy:", test_accuracy)
            writer.add_summary(ss, i)

4、TensorBoard4

import tensorflow as tf
import tensorflow.examples.tutorials.mnist.input_data as input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

x = tf.placeholder(tf.float32, [None, 784])
y_actual = tf.placeholder(tf.float32, [None, 10])

# 初始化权值W
W = tf.Variable(tf.random_uniform([784, 500], -1., 1.))
W2 = tf.Variable(tf.random_uniform([500, 10], -1., 1.))
# 初始化偏置项b
b = tf.Variable(tf.zeros([500]))
b2 = tf.Variable(tf.zeros([10]))
# 加权变换,添加ReLU非线性激励函数
y_ = tf.nn.relu((tf.matmul(x, W) + b))
output = tf.matmul(y_, W2) + b2
# 求交叉熵
loss = tf.losses.softmax_cross_entropy(onehot_labels=y_actual, logits=output)
# 用梯度下降法使得残差最小
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(loss)
# 在测试阶段，测试准确度计算
correct_prediction = tf.equal(tf.argmax(output, 1), tf.argmax(y_actual, 1))
# 多个批次的准确度均值
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

tf.summary.histogram('weights_1', W)
tf.summary.histogram('biases_1', b)
tf.summary.histogram('weights_2', W2)
tf.summary.histogram('biases_2', b2)
tf.summary.histogram('activations', y_)
tf.summary.scalar('loss', loss)
tf.summary.scalar('accuracy', accuracy)
merged_summary = tf.summary.merge_all()

with tf.Session() as sess:
    init = tf.global_variables_initializer()
    sess.run(init)
    # tf.summary.FileWriter('./logs', sess.graph)
    writer = tf.summary.FileWriter('./logs/4')
    writer.add_graph(sess.graph)
    for i in range(100000):
        batch_xs, batch_ys = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={x: batch_xs, y_actual: batch_ys})
        if i % 100 == 0:
            test_accuracy, ss = sess.run([accuracy, merged_summary], feed_dict={x: mnist.test.images,
                                                                                y_actual: mnist.test.labels})
            print("test_accuracy:", test_accuracy)
            writer.add_summary(ss, i)

5、TensorBoard5

import tensorflow as tf
import tensorflow.examples.tutorials.mnist.input_data as input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

x = tf.placeholder(tf.float32, [None, 784], name='input_x')
image = tf.reshape(x, [-1, 28, 28, 1])
y_actual = tf.placeholder(tf.float32, [None, 10], name='input_y')

y_ = tf.layers.dense(x, 500, activation='relu', name='dense')
output = tf.layers.dense(y_, 10, name='output')
# 求交叉熵
loss = tf.losses.softmax_cross_entropy(onehot_labels=y_actual, logits=output)
# 用梯度下降法使得残差最小
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(loss)
# 在测试阶段，测试准确度计算
correct_prediction = tf.equal(tf.argmax(output, 1), tf.argmax(y_actual, 1), name='test_precision')
# 多个批次的准确度均值
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

tf.summary.image('images', image, 3)
tf.summary.scalar('loss', loss)
tf.summary.scalar('accuracy', accuracy)
merged_summary = tf.summary.merge_all()

with tf.Session() as sess:
    init = tf.global_variables_initializer()
    sess.run(init)
    writer = tf.summary.FileWriter('./logs/5')
    writer.add_graph(sess.graph)
    for i in range(100000):
        batch_xs, batch_ys = mnist.train.next_batch(100)
        sess.run(train_step, feed_dict={x: batch_xs, y_actual: batch_ys})
        if i % 100 == 0:
            test_accuracy, ss = sess.run([accuracy, merged_summary], feed_dict={x: mnist.test.images,
                                                                                y_actual: mnist.test.labels})
            print("test_accuracy:", test_accuracy)
            writer.add_summary(ss, i)