TensorFlow第一发

# 导包
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt


# 1.创造训练数据，定义运算过程
def create_data():
    train_x = np.linspace(-1, 1, 100)
    # print(*train_x)
    train_y = 2 * train_x + np.random.randn(*train_x.shape) * 0.3
    # plt.plot(train_x,train_y,'ro',label="Orginal data")
    # plt.legend()
    # plt.show()
    # 2.占位X Y
    X = tf.placeholder("float")
    Y = tf.placeholder("float")
    w = tf.Variable(tf.random_normal([1]), name="weight")
    b = tf.Variable(tf.zeros([1]), name="bias")
    z = tf.multiply(X, w) + b
    return train_x, train_y, X, Y, w, b, z


# 反向优化
def reverse_trsf(Y, z):
    cost = tf.reduce_mean(tf.square(Y - z))
    learning_rate = 0.01
    optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)
    return cost, optimizer


def start_nerve(train_x, train_y, X, Y, w, b, optimizer, cost):
    # 初始化所有变量
    init = tf.global_variables_initializer()
    training_epochs = 20
    display_step = 2
    # 启动session
    with tf.Session() as sess:
        sess.run(init)
        plotdata = {"batchsize": [], "loss": []}
        for epoch in range(training_epochs):
            # 向模型输入数据进行训练
            for (x, y) in zip(train_x, train_y):
                sess.run(optimizer, feed_dict={X: x, Y: y})

            if epoch % display_step == 0:
                loss = sess.run(cost, feed_dict={X: train_x, Y: train_y})
                print("Epoch:", epoch + 1, 'cost', loss, 'w', sess
                      .run(w), 'b', sess.run(b))
                if not (loss == "NA"):
                    plotdata["batchsize"].append(epoch)
                    plotdata["loss"].append(loss)

        print("finished")
        print("cost=", sess.run(cost, feed_dict={X: train_x, Y: train_y}),
              'w=', sess.run(w), 'b=', sess.run(b))


if __name__ == "__main__":
    train_x, train_y,X, Y, w, b, z = create_data()
    cost, optimizer = reverse_trsf(Y, z)
    start_nerve(train_x, train_y, X, Y, w, b, optimizer, cost)