關于tensorflow中卷積池化運算的相關參數

import tensorflow as tf
import numpy as np

X = tf.constant([[1, 2, 3, 4], [4, 5, 6, 7], [7, 8, 9, 10], [10, 11, 12, 13]])
X = tf.reshape(X, [1, 4, 4, 1])
Y = tf.nn.max_pool(X,ksize=[1,2,2,1],strides=[1,4,4,1],padding="SAME")
with tf.Session() as sess:
    X = sess.run(X)
    print(X.shape)
    Y = sess.run(Y)
    print(Y)
           

value：池化的輸入，shape為[batch, height, width, channels]這

ksize ：池化視窗的大小，一個四維向量[1, height, width, 1]，batch和channels不做池化設定為1

strides： 視窗在每一個次元上滑動的步長，一般也是[1, stride,stride, 1]

padding：VALID容易了解 ；SAME和步長：經過嘗試，padding設定為SAME時隻在輸入矩陣右邊添加一列，下邊添加一行。卷積層是添加一圈0