tensorflow学习笔记（一）——《案例：自实现线性回归》代码校正调整

# 1) 准备数据
X = tf.random.normal(shape=[100, 1])
# X = tf.random_normal(shape=[100, 1])
y_true = tf.matmul(X, [[0.8]])+0.7

# 2) 构造模型
# 定义模型参数 用变量
weights = tf.Variable(initial_value=tf.random.normal(shape=[1, 1]))
bias = tf.Variable(initial_value=tf.random.normal(shape=[1, 1]))
y_predict = tf.matmul(X, weights) + bias

 #报错： 
 AttributeError: module 'tensorflow_core._api.v2.train' has no attribute 'GradientDescentOptimizer'
 #原代码：
 tf.train.GradientDescentOptimizer(learning_rate=0.01)
 #改为：
 tf.compat.v1.train.GradientDescentOptimizer(learn_rate=0.01)

#AttributeError: module 'tensorflow' has no attribute 'Session'
# 开启会话
with tf.compat.v1.Session() as sess:
	 # 初始化向量
	 sess.run(init)
	 # 查看初始化模型参数之后的值
	 print("训练前模型参数为：权重 %f，偏置%f" % (weights.eval(), bias.eval()))

#报错
RuntimeError: `loss` passed to Optimizer.compute_gradients should be a function when eager execution
#在下句代码之前 
#optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)
#添加tf.compat.v1.disable_eager_execution()

#更改后代码

tf.compat.v1.disable_eager_execution()
optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)

import tensorflow as tf

def linear_regression():
    """
    自实现一个线性回归
    :return:
    """
    # 1) 准备数据
    X = tf.random.normal(shape=[100, 1])
    # X = tf.random_normal(shape=[100, 1])
    y_true = tf.matmul(X, [[0.8]])+0.7

    # 2) 构造模型
    # 定义模型参数 用变量
    weights = tf.Variable(initial_value=tf.random.normal(shape=[1, 1]))
    bias = tf.Variable(initial_value=tf.random.normal(shape=[1, 1]))
    y_predict = tf.matmul(X, weights) + bias

    # 3) 构造损失函数
    error = tf.reduce_mean(tf.square(y_predict - y_true))

    # 4) 优化损失
    tf.compat.v1.disable_eager_execution()
    optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)

    # 显示初始化变量
    init = tf.compat.v1.global_variables_initializer()

    # 开启会话
    with tf.compat.v1.Session() as sess:
        # 初始化向量
        sess.run(init)

        # 查看初始化模型参数之后的值
        print("训练前模型参数为：权重 %f，偏置%f" % (weights.eval(), bias.eval()))
    return None


if __name__ == "__main__":
    linear_regression()