Tensorflow报错: Shape must be rank 2 but is rank 1 for 'MatMul' (op: 'MatMul') with input shapes

import tensorflow as tf

# 创建两个常量Tensor
const1 = tf.constant([2, 2])
const2 = tf.constant([4, 4])

multiple = tf.matmul(const1, const2)

# 尝试用print输出multiple的值
print(multiple)

初学TF时矩阵乘法报错：

ValueError: Shape must be rank 2 but is rank 1 for 'MatMul' (op: 'MatMul') with input shapes: [2], [2].

 

解决方法：

矩阵相乘是对应的行和列相乘的结果，所以需要第一个矩阵的列数与第二个矩阵的行数相等。

只需令

const1 = tf.constant([[2, 2]])
const2 = tf.constant([[4],[4]])