Tensorflow2 tf.nn.maxpool2d()进行池化运算及其可视化

Posted 2021-09-09 空中旋转篮球

1.tf.nn.maxpool2d()函数介绍

tf.nn.max_pool2d(input, ksize, strides, padding, data_format='NHWC', name=None)

参数说明：

Args
input	A 4-D Tensor of the format specified by data_format.
ksize	An int or list of ints that has length 1, 2 or 4. The size of the window for each dimension of the input tensor.
strides	An int or list of ints that has length 1, 2 or 4. The stride of the sliding window for each dimension of the input tensor.
padding	Either the string "SAME" or "VALID" indicating the type of padding algorithm to use, or a list indicating the explicit paddings at the start and end of each dimension. When explicit padding is used and data_format is "NHWC", this should be in the form [[0, 0], [pad_top,pad_bottom], [pad_left, pad_right], [0, 0]]. When explicit padding used and data_format is "NCHW", this should be in the form [[0, 0], [0, 0],[pad_top, pad_bottom], [pad_left, pad_right]]. When using explicit padding, the size of the paddings cannot be greater than the sliding window size.
data_format	A string. 'NHWC', 'NCHW' and 'NCHW_VECT_C' are supported.
name	Optional name for the operation.

2.使用数据

flower_photos数据集中选一张玫瑰花图片

3.池化运算代码

import numpy as np
import matplotlib.pyplot as plt
from PIL import Image
import tensorflow as tf

data = Image.open("roses_4483444865_65962cea07_m.jpg")  # 返回一个PIL图像对象
plt.imshow(data)
plt.show()

x = np.array(data)
x = x / 255
x = x.reshape(1, 240, 180, 3)

image_tensor = tf.convert_to_tensor(x)
x_input = tf.cast(image_tensor, tf.float32)

print("x_in{}", x_input.shape)

kernel_in = np.array(
    [[[-1, 1]], [[1, 1]],])
print(kernel_in.shape)
kernel = tf.constant(kernel_in, dtype=tf.float32)

Z1=tf.nn.max_pool2d(x_input, [1,2,2,1], strides=[1, 2, 2, 1], padding='SAME')

x_max_pool2d = np.array(Z1)
print(x_max_pool2d.shape)
x_max_pool2d = x_max_pool2d.reshape(120, 90, 3)
plt.imshow(x_max_pool2d)
plt.show()

4.计算结果