深度学习：TensorFlow 基础操作

tensorflow是当前最流行的深度学习框架，在这里记录和分享一些tf入门的基本姿势。

国际惯例，代码的第一行输出献给“Hello,World!”。

"""
HelloWorld example using TensorFlow.
"""

from __future__ import print_function

import datetime
import tensorflow as tf

print('Start at: ', datetime.datetime.now())  # 开始时间

hello = tf.constant('Hello, World!')  # 定义一个常量

sess = tf.Session()  # 开始执行tensorflow的会话

# 打印输出
print(sess.run(hello))  # b'Hello, World!'
print('End at: ', datetime.datetime.now())  # 结束时间

基本运算操作

"""
Basic constant operations example using TensorFlow.
"""

from __future__ import print_function

import tensorflow as tf

#  Method1：定义常量后执行
#  定义常量
a = tf.constant(3)
b = tf.constant(5)

# 执行默认图操作
with tf.Session() as sess:
    print("a=3, b=5")
    print("常量相加: %i" % sess.run(a + b))  # 8
    print("常量相乘: %i" % sess.run(a * b))  # 15

# Method2：定义类型、输入值后再执行
# 定义输入类型
a = tf.placeholder(tf.int16)
b = tf.placeholder(tf.int16)

# 定义运算规则
add = tf.add(a, b)
mul = tf.multiply(a, b)

# 执行默认图操作
with tf.Session() as sess:
    # 使用输入变量执行运算
    print("变量相加: %i" % sess.run(add, feed_dict={a: 3, b: 5}))  # 8
    print("变量相乘: %i" % sess.run(mul, feed_dict={a: 3, b: 5}))  # 15

# 矩阵乘法
# 定义一个1x2矩阵
matrix1 = tf.constant([[1., 4.]])

# 定义一个2x1矩阵
matrix2 = tf.constant([[3.], [2.]])

# 定义运算
product = tf.matmul(matrix1, matrix2)

# 执行
with tf.Session() as sess:
    result = sess.run(product)
    print(result)  # [[ 11.]]

最最基础的线性回归模型

from __future__ import print_function

import matplotlib.pyplot as plt
import numpy
import tensorflow as tf

rng = numpy.random

# 参数
learning_rate = 0.01
training_epochs = 1000
display_step = 50

# 训练数据
train_X = numpy.asarray([3.3, 4.4, 5.5, 6.71, 6.93, 4.168, 9.779, 6.182, 7.59, 2.167,
                         7.042, 10.791, 5.313, 7.997, 5.654, 9.27, 3.1])
train_Y = numpy.asarray([1.7, 2.76, 2.09, 3.19, 1.694, 1.573, 3.366, 2.596, 2.53, 1.221,
                         2.827, 3.465, 1.65, 2.904, 2.42, 2.94, 1.3])
n_samples = train_X.shape[0]

# tf图输入
X = tf.placeholder("float")
Y = tf.placeholder("float")

# 设置模型权重
W = tf.Variable(rng.randn(), name="weight")
b = tf.Variable(rng.randn(), name="bias")

# 构造线性模型
pred = tf.add(tf.multiply(X, W), b)

# 均方差（Mean squared error）
cost = tf.reduce_sum(tf.pow(pred - Y, 2)) / (2 * n_samples)
# 梯度下降
# Variable对象的trainable属性值默认为True，minimize()函数会调整W和b
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

# 初始化全局的变量
init = tf.global_variables_initializer()

# 开始训练
with tf.Session() as sess:
    # 执行初始化
    sess.run(init)

    # 拟合训练数据
    for epoch in range(training_epochs):
        for (x, y) in zip(train_X, train_Y):
            sess.run(optimizer, feed_dict={X: x, Y: y})

        # 每轮训练输出日志
        if (epoch + 1) % display_step == 0:
            c = sess.run(cost, feed_dict={X: train_X, Y: train_Y})
            print("Epoch:", '%04d' % (epoch + 1), "cost=", "{:.9f}".format(c),
                  "W=", sess.run(W), "b=", sess.run(b))

    print("Optimization Finished!")
    training_cost = sess.run(cost, feed_dict={X: train_X, Y: train_Y})
    print("Training cost=", training_cost, "W=", sess.run(W), "b=", sess.run(b), '\n')

    # 展示
    plt.plot(train_X, train_Y, 'ro', label='Original data')
    plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fitted line')
    plt.legend()
    plt.show()

    # 测试样本
    test_X = numpy.asarray([6.83, 4.668, 8.9, 7.91, 5.7, 8.7, 3.1, 2.1])
    test_Y = numpy.asarray([1.84, 2.273, 3.2, 2.831, 2.92, 3.24, 1.35, 1.03])

    print("Testing... (Mean square loss Comparison)")
    testing_cost = sess.run(
        tf.reduce_sum(tf.pow(pred - Y, 2)) / (2 * test_X.shape[0]),
        feed_dict={X: test_X, Y: test_Y})  # 同训练损失函数
    print("Testing cost=", testing_cost)
    print("Absolute mean square loss difference:", abs(training_cost - testing_cost))

    plt.plot(test_X, test_Y, 'bo', label='Testing data')
    plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fitted line')
    plt.legend()
    plt.show()

基础模型中的逻辑回归模型

from __future__ import print_function

import tensorflow as tf
# 导入MNIST数据
from tensorflow.examples.tutorials.mnist import input_data

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

# 参数
learning_rate = 0.01
training_epochs = 25
batch_size = 100
display_step = 1

# 图输入
x = tf.placeholder(tf.float32, [None, 784])  # mnist图像数据（28*28=784）
y = tf.placeholder(tf.float32, [None, 10])  # 0-9 数字识别（10类）

# 设置模型权重
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))

# 构造模型
pred = tf.nn.softmax(tf.matmul(x, W) + b)  # Softmax

# 使用交叉熵（cross entropy）最小化误差
cost = tf.reduce_mean(-tf.reduce_sum(y * tf.log(pred), reduction_indices=1))
# 梯度下降
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

# 初始化全局的变量
init = tf.global_variables_initializer()

# 开始训练
with tf.Session() as sess:
    # 执行初始化
    sess.run(init)

    # 周期训练
    for epoch in range(training_epochs):
        avg_cost = 0.
        total_batch = int(mnist.train.num_examples / batch_size)
        # 全部批次循环
        for i in range(total_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            # 执行优化与计算损失率
            _, c = sess.run([optimizer, cost], feed_dict={x: batch_xs,
                                                          y: batch_ys})
            # 计算平均损失率
            avg_cost += c / total_batch
        # 间隔打印日志
        if (epoch + 1) % display_step == 0:
            print("Epoch:", '%04d' % (epoch + 1), "cost=", "{:.9f}".format(avg_cost))

    print("Optimization Finished!")

    # 测试模型
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # 计算准确率
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
    print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))

今天整理先到这里，如果还有老铁想要学习人工智能，领取学习资料的可以来加我！！