使用Tensorflow进行卷积操作模拟

---

Python版本:3.6.7
Tensorflow版本:1.12.3
使用jupyter Notebook操作

---

使用一张5x5 一通道的图像（对应的shape：[1，5，5，1]）
用一个3×3的卷积核（对应的shape：[3，3，1，1]）去做卷积

附上代码

import numpy as np
import tensorflow as tf
array = np.array([[1,2,3,4,5],
                  [2,3,4,5,6],
                  [3,4,5,6,7],
                  [4,5,6,7,8],
                  [5,6,7,8,9]], np.float32).reshape(1,5,5,1)
x = tf.convert_to_tensor(array, tf.float32)
array = np.array([[0,1,0],
                 [1,1,1],
                 [0,1,0]], np.float32).reshape(3,3,1,1)
y = tf.convert_to_tensor(array, tf.float32)
print(x)
print(y)

op = tf.nn.conv2d(x, y, strides=[1, 1, 1, 1], padding='SAME')

with tf.Session() as sess:
    print("矩阵 :\n",sess.run(x).reshape(5,5))
    print("卷积核 :\n",sess.run(y).reshape(3,3))
    print("结果 :\n",sess.run(op).reshape(5,5))