Tensorflow TypeError:variables_initializer（）缺少1个必需的位置参数：'；变量列表'；

代码

import tensorflow as tf
tf.compat.v1.disable_eager_execution()

x_train = [1.0, 2.0, 3.0, 4.0]
y_train = [-1.0, -2.0, -3.0, -4.0]

W = tf.Variable(initial_value=[1.0], dtype=tf.float32)
b = tf.Variable(initial_value=[1.0], dtype=tf.float32)

x = tf.compat.v1.placeholder(dtype=tf.float32)
y_input = tf.compat.v1.placeholder(dtype=tf.float32)

y_output = W * x + b

loss = tf.reduce_sum(input_tensor=tf.square(x=y_output - y_input))
optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01)
train_step = optimizer.minimize(loss=loss)

session = tf.compat.v1.Session()
session.run(tf.compat.v1.variables_initializer())

print(session.run(fetches=loss, feed_dict={x: x_train, y_input: y_train}))

for _ in range(1000):
    session.run(fetches=train_step, feed_dict={x: x_train, y_input: y_train})

print(session.run(fetches=[loss, W, b], feed_dict={x: x_train, y_input: y_train}))

print(session.run(fetches=y_output, feed_dict={x: [5.0, 10.0, 15.0]}))

错误

Traceback (most recent call last):
  File "D:/phycarm projects/linearRegression.py", line 20, in <module>
session.run(tf.compat.v1.variables_initializer())
  TypeError: variables_initializer() missing 1 required positional argument: 'var_list'

Process finished with exit code 1

回溯（最近一次呼叫最后一次）：
文件“D:/phycarm projects/linearRegression.py”，第20行，在
session.run（tf.compat.v1.variables\u initializer（））
TypeError:variables\u initializer（）缺少1个必需的位置参数：“var\u list”
进程已完成，退出代码为1

确定只需在变量初始化器中添加变量列表，如下所示

您只需要键入您在var列表中定义的yor变量

run（tf.compat.v1.variables\u初始值设定项（var\u list=[W，b]））

以下是正确的代码：-

import tensorflow as tf

tf.compat.v1.disable_eager_execution()

x_train = [1.0, 2.0, 3.0, 4.0]
y_train = [-1.0, -2.0, -3.0, -4.0]

W = tf.Variable(initial_value=[1.0], dtype=tf.float32)
b = tf.Variable(initial_value=[1.0], dtype=tf.float32)

x = tf.compat.v1.placeholder(dtype=tf.float32)
y_input = tf.compat.v1.placeholder(dtype=tf.float32)

y_output = W * x + b

loss = tf.reduce_sum(input_tensor=tf.square(x=y_output - y_input))
optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01)
train_step = optimizer.minimize(loss=loss)

session = tf.compat.v1.Session()
session.run(tf.compat.v1.variables_initializer(var_list=[W, b]))

print(session.run(fetches=loss, feed_dict={x: x_train, y_input: y_train}))

for _ in range(1000):
    session.run(fetches=train_step, feed_dict={x: x_train, y_input: y_train})

print(session.run(fetches=[loss, W, b], feed_dict={x: x_train, y_input: y_train}))

print(session.run(fetches=y_output, feed_dict={x: [5.0, 10.0, 15.0]}))