编写代码：
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
# 初始化数据
data = np.array([
[137.97, 145.00],
[104.50, 110.00],
[100.00, 93.00],
[124.32, 116.00],
[79.20, 65.32],
[99.00, 104.00],
[124.00, 118.00],
[114.00, 91.00],
[106.69, 62.00],
[138.05, 133.00],
[53.75, 51.00],
[46.91, 45.00],
[68.00, 78.50],
[63.02, 69.65],
[81.26, 75.69],
[86.21, 95.30],
], dtype=np.float64)
X = data[:, 0:1] # 输入变量
y = data[:, 1] # 目标变量
# 归一化处理
X_mean = X.mean()
X_std = X.std()
X_normalized = (X - X_mean) / X_std
def linear_regression_model(X, y):
W = tf.Variable(tf.random.normal([1, 1], dtype=tf.float64), name='weight')
b = tf.Variable(tf.random.normal([1], dtype=tf.float64), name='bias')
optimizer = tf.optimizers.SGD(learning_rate=0.01)
for epoch in range(1000):
with tf.GradientTape() as tape:
predictions = tf.matmul(X, W) + b
loss = tf.reduce_mean(tf.square(predictions - y))
gradients = tape.gradient(loss, [W, b])
资源评论
