this is a code demo!

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
from sklearn import datasets

# iris.data = [(Sepal Length, Sepal Width, Petal Length, Petal Width)]
# x: Petal Width, y: Sepal Length
iris = datasets.load_iris()
x_train = np.array([x[3] for x in iris.data])
y_train = np.array([y[0] for y in iris.data])


x = tf.placeholder(shape=[None, 1], dtype=np.float32)
y = tf.placeholder(shape=[None, 1], dtype=np.float32)
W = tf.Variable(tf.random_normal(shape=[1, 1]))
b = tf.Variable(tf.zeros(shape=[1,1]))
y_ = tf.add(tf.matmul(x, W), b)
loss = tf.reduce_mean(tf.square(y - y_))
optimizer = tf.train.GradientDescentOptimizer(0.05).minimize(loss)
# optimizer = tf.train.AdamOptimizer(learning_rate= 0.05).minimize(loss)

loss_vec = []
batch_size = 25
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for i in range(100):
        index = np.random.choice(len(x_train), size=batch_size)
        x_batch = np.transpose([x_train[index]])
        y_batch = np.transpose([y_train[index]])
        sess.run(optimizer, feed_dict={x:x_batch, y:y_batch})
        loss_batch = sess.run(loss, feed_dict={x:x_batch, y:y_batch})
        loss_vec.append(loss_batch)
        if (i+1) % 20 == 0:
            print("Step: "+str(i+1)+", W: "+str(sess.run(W))+", b: "+str(sess.run(b)))
            print("Loss: "+str(loss_batch))

# Plot loss over time
# 迭代100次的L2正则损失函数
plt.plot(loss_vec, 'k-')
plt.title('L2 Loss per Generation')
plt.xlabel('Generation')
plt.ylabel('L2 Loss')
plt.show()

GAN – github