人工智能——作业5：卷积-池化-激活

最新推荐文章于 2022-12-29 23:46:30 发布

孟小起

最新推荐文章于 2022-12-29 23:46:30 发布

阅读量137

点赞数 1

文章标签：人工智能深度学习 cnn

本文链接：https://blog.csdn.net/mqj010321/article/details/125015742

版权

1.For循环版本：手工实现卷积-池化-激活代码如下：
代码来源
没有直接调用库函数来计算，而是运用for循环来完成卷积-池化-激活效果
代码如下：

import numpy as np
 
x = np.array([[-1, -1, -1, -1, -1, -1, -1, -1, -1],
              [-1, 1, -1, -1, -1, -1, -1, 1, -1],
              [-1, -1, 1, -1, -1, -1, 1, -1, -1],
              [-1, -1, -1, 1, -1, 1, -1, -1, -1],
              [-1, -1, -1, -1, 1, -1, -1, -1, -1],
              [-1, -1, -1, 1, -1, 1, -1, -1, -1],
              [-1, -1, 1, -1, -1, -1, 1, -1, -1],
              [-1, 1, -1, -1, -1, -1, -1, 1, -1],
              [-1, -1, -1, -1, -1, -1, -1, -1, -1]])
print("x=\n", x)
# 初始化 三个 卷积核
Kernel = [[0 for i in range(0, 3)] for j in range(0, 3)]
Kernel[0] = np.array([[1, -1, -1],
                      [-1, 1, -1],
                      [-1, -1, 1]])
Kernel[1] = np.array([[1, -1, 1],
                      [-1, 1, -1],
                      [1, -1, 1]])
Kernel[2] = np.array([[-1, -1, 1],
                      [-1, 1, -1],
                      [1, -1, -1]])
 
# --------------- 卷积  ---------------
stride = 1  # 步长
feature_map_h = 7  # 特征图的高
feature_map_w = 7  # 特征图的宽
feature_map = [0 for i in range(0, 3)]  # 初始化3个特征图
for i in range(0, 3):
    feature_map[i] = np.zeros((feature_map_h, feature_map_w))  # 初始化特征图
for h in range(feature_map_h):  # 向下滑动，得到卷积后的固定行
    for w in range(feature_map_w):  # 向右滑动，得到卷积后的固定行的列
        v_start = h * stride  # 滑动窗口的起始行（高）
        v_end = v_start + 3  # 滑动窗口的结束行（高）
        h_start = w * stride  # 滑动窗口的起始列（宽）
        h_end = h_start + 3  # 滑动窗口的结束列（宽）
        window = x[v_start:v_end, h_start:h_end]  # 从图切出一个滑动窗口
        for i in range(0, 3):
            feature_map[i][h, w] = np.divide(np.sum(np.multiply(window, Kernel[i][:, :])), 9)
print("feature_map:\n", np.around(feature_map, decimals=2))
 
# --------------- 池化  ---------------
pooling_stride = 2  # 步长
pooling_h = 4  # 特征图的高
pooling_w = 4  # 特征图的宽
feature_map_pad_0 = [[0 for i in range(0, 8)] for j in range(0, 8)]
for i in range(0, 3):  # 特征图 补 0 ，行 列 都要加 1 (因为上一层是奇数，池化窗口用的偶数)
    feature_map_pad_0[i] = np.pad(feature_map[i], ((0, 1), (0, 1)), 'constant', constant_values=(0, 0))
# print("feature_map_pad_0 0:\n", np.around(feature_map_pad_0[0], decimals=2))
 
pooling = [0 for i in range(0, 3)]
for i in range(0, 3):
    pooling[i] = np.zeros((pooling_h, pooling_w))  # 初始化特征图
for h in range(pooling_h):  # 向下滑动，得到卷积后的固定行
    for w in range(pooling_w):  # 向右滑动，得到卷积后的固定行的列
        v_start = h * pooling_stride  # 滑动窗口的起始行（高）
        v_end = v_start + 2  # 滑动窗口的结束行（高）
        h_start = w * pooling_stride  # 滑动窗口的起始列（宽）
        h_end = h_start + 2  # 滑动窗口的结束列（宽）
        for i in range(0, 3):
            pooling[i][h, w] = np.max(feature_map_pad_0[i][v_start:v_end, h_start:h_end])
print("pooling:\n", np.around(pooling[0], decimals=2))
print("pooling:\n", np.around(pooling[1], decimals=2))
print("pooling:\n", np.around(pooling[2], decimals=2))
 
 
# --------------- 激活  ---------------
def relu(x):
    return (abs(x) + x) / 2
 
 
relu_map_h = 7  # 特征图的高
relu_map_w = 7  # 特征图的宽
relu_map = [0 for i in range(0, 3)]  # 初始化3个特征图
for i in range(0, 3):
    relu_map[i] = np.zeros((relu_map_h, relu_map_w))  # 初始化特征图
 
for i in range(0, 3):
    relu_map[i] = relu(feature_map[i])
 
print("relu map :\n",np.around(relu_map[0], decimals=2))
print("relu map :\n",np.around(relu_map[1], decimals=2))
print("relu map :\n",np.around(relu_map[2], decimals=2))

结果：

x=
[[-1 -1 -1 -1 -1 -1 -1 -1 -1]
[-1 1 -1 -1 -1 -1 -1 1 -1]
[-1 -1 1 -1 -1 -1 1 -1 -1]
[-1 -1 -1 1 -1 1 -1 -1 -1]
[-1 -1 -1 -1 1 -1 -1 -1 -1]
[-1 -1 -1 1 -1 1 -1 -1 -1]
[-1 -1 1 -1 -1 -1 1 -1 -1]
[-1 1 -1 -1 -1 -1 -1 1 -1]
[-1 -1 -1 -1 -1 -1 -1 -1 -1]]
feature_map:
[[[ 0.78 -0.11 0.11 0.33 0.56 -0.11 0.33]
[-0.11 1. -0.11 0.33 -0.11 0.11 -0.11]
[ 0.11 -0.11 1. -0.33 0.11 -0.11 0.56]
[ 0.33 0.33 -0.33 0.56 -0.33 0.33 0.33]
[ 0.56 -0.11 0.11 -0.33 1. -0.11 0.11]
[-0.11 0.11 -0.11 0.33 -0.11 1. -0.11]
[ 0.33 -0.11 0.56 0.33 0.11 -0.11 0.78]]
[[ 0.33 -0.56 0.11 -0.11 0.11 -0.56 0.33]
[-0.56 0.56 -0.56 0.33 -0.56 0.56 -0.56]
[ 0.11 -0.56 0.56 -0.78 0.56 -0.56 0.11]
[-0.11 0.33 -0.78 1. -0.78 0.33 -0.11]
[ 0.11 -0.56 0.56 -0.78 0.56 -0.56 0.11]
[-0.56 0.56 -0.56 0.33 -0.56 0.56 -0.56]
[ 0.33 -0.56 0.11 -0.11 0.11 -0.56 0.33]]
[[ 0.33 -0.11 0.56 0.33 0.11 -0.11 0.78]
[-0.11 0.11 -0.11 0.33 -0.11 1. -0.11]
[ 0.56 -0.11 0.11 -0.33 1. -0.11 0.11]
[ 0.33 0.33 -0.33 0.56 -0.33 0.33 0.33]
[ 0.11 -0.11 1. -0.33 0.11 -0.11 0.56]
[-0.11 1. -0.11 0.33 -0.11 0.11 -0.11]
[ 0.78 -0.11 0.11 0.33 0.56 -0.11 0.33]]]
pooling:
[[1. 0.33 0.56 0.33]
[0.33 1. 0.33 0.56]
[0.56 0.33 1. 0.11]
[0.33 0.56 0.11 0.78]]
pooling:
[[0.56 0.33 0.56 0.33]
[0.33 1. 0.56 0.11]
[0.56 0.56 0.56 0.11]
[0.33 0.11 0.11 0.33]]
pooling:
[[0.33 0.56 1. 0.78]
[0.56 0.56 1. 0.33]
[1. 1. 0.11 0.56]
[0.78 0.33 0.56 0.33]]
relu map :
[[0.78 0. 0.11 0.33 0.56 0. 0.33]
[0. 1. 0. 0.33 0. 0.11 0. ]
[0.11 0. 1. 0. 0.11 0. 0.56]
[0.33 0.33 0. 0.56 0. 0.33 0.33]
[0.56 0. 0.11 0. 1. 0. 0.11]
[0. 0.11 0. 0.33 0. 1. 0. ]
[0.33 0. 0.56 0.33 0.11 0. 0.78]]
relu map :
[[0.33 0. 0.11 0. 0.11 0. 0.33]
[0. 0.56 0. 0.33 0. 0.56 0. ]
[0.11 0. 0.56 0. 0.56 0. 0.11]
[0. 0.33 0. 1. 0. 0.33 0. ]
[0.11 0. 0.56 0. 0.56 0. 0.11]
[0. 0.56 0. 0.33 0. 0.56 0. ]
[0.33 0. 0.11 0. 0.11 0. 0.33]]
relu map :
[[0.33 0. 0.56 0.33 0.11 0. 0.78]
[0. 0.11 0. 0.33 0. 1. 0. ]
[0.56 0. 0.11 0. 1. 0. 0.11]
[0.33 0.33 0. 0.56 0. 0.33 0.33]
[0.11 0. 1. 0. 0.11 0. 0.56]
[0. 1. 0. 0.33 0. 0.11 0. ]
[0.78 0. 0.11 0.33 0.56 0. 0.33]]

2.Pytorch版本：调用函数完成卷积-池化-激活代码如下：
直接调用库函数，只需要给出初值

代码：

# https://blog.csdn.net/qq_26369907/article/details/88366147
# https://zhuanlan.zhihu.com/p/405242579
import numpy as np
import torch
import torch.nn as nn
 
x = torch.tensor([[[[-1, -1, -1, -1, -1, -1, -1, -1, -1],
                    [-1, 1, -1, -1, -1, -1, -1, 1, -1],
                    [-1, -1, 1, -1, -1, -1, 1, -1, -1],
                    [-1, -1, -1, 1, -1, 1, -1, -1, -1],
                    [-1, -1, -1, -1, 1, -1, -1, -1, -1],
                    [-1, -1, -1, 1, -1, 1, -1, -1, -1],
                    [-1, -1, 1, -1, -1, -1, 1, -1, -1],
                    [-1, 1, -1, -1, -1, -1, -1, 1, -1],
                    [-1, -1, -1, -1, -1, -1, -1, -1, -1]]]], dtype=torch.float)
print(x.shape)
print(x)
 
print("--------------- 卷积  ---------------")
conv1 = nn.Conv2d(1, 1, (3, 3), 1)  # in_channel , out_channel , kennel_size , stride
conv1.weight.data = torch.Tensor([[[[1, -1, -1],
                                    [-1, 1, -1],
                                    [-1, -1, 1]]
                                   ]])
conv2 = nn.Conv2d(1, 1, (3, 3), 1)  # in_channel , out_channel , kennel_size , stride
conv2.weight.data = torch.Tensor([[[[1, -1, 1],
                                    [-1, 1, -1],
                                    [1, -1, 1]]
                                   ]])
conv3 = nn.Conv2d(1, 1, (3, 3), 1)  # in_channel , out_channel , kennel_size , stride
conv3.weight.data = torch.Tensor([[[[-1, -1, 1],
                                    [-1, 1, -1],
                                    [1, -1, -1]]
                                   ]])
 
feature_map1 = conv1(x)
feature_map2 = conv2(x)
feature_map3 = conv3(x)
 
print(feature_map1 / 9)
print(feature_map2 / 9)
print(feature_map3 / 9)
 
print("--------------- 池化  ---------------")
max_pool = nn.MaxPool2d(2, padding=0, stride=2)  # Pooling
zeroPad = nn.ZeroPad2d(padding=(0, 1, 0, 1))  # pad 0 , Left Right Up Down
 
feature_map_pad_0_1 = zeroPad(feature_map1)
feature_pool_1 = max_pool(feature_map_pad_0_1)
feature_map_pad_0_2 = zeroPad(feature_map2)
feature_pool_2 = max_pool(feature_map_pad_0_2)
feature_map_pad_0_3 = zeroPad(feature_map3)
feature_pool_3 = max_pool(feature_map_pad_0_3)
 
print(feature_pool_1.size())
print(feature_pool_1 / 9)
print(feature_pool_2 / 9)
print(feature_pool_3 / 9)
 
print("--------------- 激活  ---------------")
activation_function = nn.ReLU()
 
feature_relu1 = activation_function(feature_map1)
feature_relu2 = activation_function(feature_map2)
feature_relu3 = activation_function(feature_map3)
print(feature_relu1 / 9)
print(feature_relu2 / 9)
print(feature_relu3 / 9)

结果：

torch.Size([1, 1, 9, 9])
tensor([[[[-1., -1., -1., -1., -1., -1., -1., -1., -1.],
[-1., 1., -1., -1., -1., -1., -1., 1., -1.],
[-1., -1., 1., -1., -1., -1., 1., -1., -1.],
[-1., -1., -1., 1., -1., 1., -1., -1., -1.],
[-1., -1., -1., -1., 1., -1., -1., -1., -1.],
[-1., -1., -1., 1., -1., 1., -1., -1., -1.],
[-1., -1., 1., -1., -1., -1., 1., -1., -1.],
[-1., 1., -1., -1., -1., -1., -1., 1., -1.],
[-1., -1., -1., -1., -1., -1., -1., -1., -1.]]]])
--------------- 卷积 ---------------
tensor([[[[ 0.7606, -0.1283, 0.0939, 0.3161, 0.5384, -0.1283, 0.3161],
[-0.1283, 0.9828, -0.1283, 0.3161, -0.1283, 0.0939, -0.1283],
[ 0.0939, -0.1283, 0.9828, -0.3505, 0.0939, -0.1283, 0.5384],
[ 0.3161, 0.3161, -0.3505, 0.5384, -0.3505, 0.3161, 0.3161],
[ 0.5384, -0.1283, 0.0939, -0.3505, 0.9828, -0.1283, 0.0939],
[-0.1283, 0.0939, -0.1283, 0.3161, -0.1283, 0.9828, -0.1283],
[ 0.3161, -0.1283, 0.5384, 0.3161, 0.0939, -0.1283, 0.7606]]]],
grad_fn=)
tensor([[[[ 0.3295, -0.5594, 0.1073, -0.1150, 0.1073, -0.5594, 0.3295],
[-0.5594, 0.5517, -0.5594, 0.3295, -0.5594, 0.5517, -0.5594],
[ 0.1073, -0.5594, 0.5517, -0.7816, 0.5517, -0.5594, 0.1073],
[-0.1150, 0.3295, -0.7816, 0.9962, -0.7816, 0.3295, -0.1150],
[ 0.1073, -0.5594, 0.5517, -0.7816, 0.5517, -0.5594, 0.1073],
[-0.5594, 0.5517, -0.5594, 0.3295, -0.5594, 0.5517, -0.5594],
[ 0.3295, -0.5594, 0.1073, -0.1150, 0.1073, -0.5594, 0.3295]]]],
grad_fn=)
tensor([[[[ 0.3483, -0.0961, 0.5705, 0.3483, 0.1261, -0.0961, 0.7928],
[-0.0961, 0.1261, -0.0961, 0.3483, -0.0961, 1.0150, -0.0961],
[ 0.5705, -0.0961, 0.1261, -0.3183, 1.0150, -0.0961, 0.1261],
[ 0.3483, 0.3483, -0.3183, 0.5705, -0.3183, 0.3483, 0.3483],
[ 0.1261, -0.0961, 1.0150, -0.3183, 0.1261, -0.0961, 0.5705],
[-0.0961, 1.0150, -0.0961, 0.3483, -0.0961, 0.1261, -0.0961],
[ 0.7928, -0.0961, 0.1261, 0.3483, 0.5705, -0.0961, 0.3483]]]],
grad_fn=)
--------------- 池化 ---------------
torch.Size([1, 1, 4, 4])
tensor([[[[0.9828, 0.3161, 0.5384, 0.3161],
[0.3161, 0.9828, 0.3161, 0.5384],
[0.5384, 0.3161, 0.9828, 0.0939],
[0.3161, 0.5384, 0.0939, 0.7606]]]], grad_fn=)
tensor([[[[0.5517, 0.3295, 0.5517, 0.3295],
[0.3295, 0.9962, 0.5517, 0.1073],
[0.5517, 0.5517, 0.5517, 0.1073],
[0.3295, 0.1073, 0.1073, 0.3295]]]], grad_fn=)
tensor([[[[0.3483, 0.5705, 1.0150, 0.7928],
[0.5705, 0.5705, 1.0150, 0.3483],
[1.0150, 1.0150, 0.1261, 0.5705],
[0.7928, 0.3483, 0.5705, 0.3483]]]], grad_fn=)
--------------- 激活 ---------------
tensor([[[[0.7606, 0.0000, 0.0939, 0.3161, 0.5384, 0.0000, 0.3161],
[0.0000, 0.9828, 0.0000, 0.3161, 0.0000, 0.0939, 0.0000],
[0.0939, 0.0000, 0.9828, 0.0000, 0.0939, 0.0000, 0.5384],
[0.3161, 0.3161, 0.0000, 0.5384, 0.0000, 0.3161, 0.3161],
[0.5384, 0.0000, 0.0939, 0.0000, 0.9828, 0.0000, 0.0939],
[0.0000, 0.0939, 0.0000, 0.3161, 0.0000, 0.9828, 0.0000],
[0.3161, 0.0000, 0.5384, 0.3161, 0.0939, 0.0000, 0.7606]]]],
grad_fn=)
tensor([[[[0.3295, 0.0000, 0.1073, 0.0000, 0.1073, 0.0000, 0.3295],
[0.0000, 0.5517, 0.0000, 0.3295, 0.0000, 0.5517, 0.0000],
[0.1073, 0.0000, 0.5517, 0.0000, 0.5517, 0.0000, 0.1073],
[0.0000, 0.3295, 0.0000, 0.9962, 0.0000, 0.3295, 0.0000],
[0.1073, 0.0000, 0.5517, 0.0000, 0.5517, 0.0000, 0.1073],
[0.0000, 0.5517, 0.0000, 0.3295, 0.0000, 0.5517, 0.0000],
[0.3295, 0.0000, 0.1073, 0.0000, 0.1073, 0.0000, 0.3295]]]],
grad_fn=)
tensor([[[[0.3483, 0.0000, 0.5705, 0.3483, 0.1261, 0.0000, 0.7928],
[0.0000, 0.1261, 0.0000, 0.3483, 0.0000, 1.0150, 0.0000],
[0.5705, 0.0000, 0.1261, 0.0000, 1.0150, 0.0000, 0.1261],
[0.3483, 0.3483, 0.0000, 0.5705, 0.0000, 0.3483, 0.3483],
[0.1261, 0.0000, 1.0150, 0.0000, 0.1261, 0.0000, 0.5705],
[0.0000, 1.0150, 0.0000, 0.3483, 0.0000, 0.1261, 0.0000],
[0.7928, 0.0000, 0.1261, 0.3483, 0.5705, 0.0000, 0.3483]]]],
grad_fn=)

3.可视化：了解数字与图像之间的关系
代码：

# https://blog.csdn.net/qq_26369907/article/details/88366147
# https://zhuanlan.zhihu.com/p/405242579
import torch
import torch.nn as nn
import matplotlib.pyplot as plt
plt.rcParams['font.sans-serif']=['SimHei'] #用来正常显示中文标签
plt.rcParams['axes.unicode_minus']=False #用来正常显示负号 #有中文出现的情况，需要u'内容
x = torch.tensor([[[[-1, -1, -1, -1, -1, -1, -1, -1, -1],
                    [-1, 1, -1, -1, -1, -1, -1, 1, -1],
                    [-1, -1, 1, -1, -1, -1, 1, -1, -1],
                    [-1, -1, -1, 1, -1, 1, -1, -1, -1],
                    [-1, -1, -1, -1, 1, -1, -1, -1, -1],
                    [-1, -1, -1, 1, -1, 1, -1, -1, -1],
                    [-1, -1, 1, -1, -1, -1, 1, -1, -1],
                    [-1, 1, -1, -1, -1, -1, -1, 1, -1],
                    [-1, -1, -1, -1, -1, -1, -1, -1, -1]]]], dtype=torch.float)
print(x.shape)
print(x)
img = x.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('原图')
plt.show()
 
print("--------------- 卷积  ---------------")
conv1 = nn.Conv2d(1, 1, (3, 3), 1)  # in_channel , out_channel , kennel_size , stride
conv1.weight.data = torch.Tensor([[[[1, -1, -1],
                                    [-1, 1, -1],
                                    [-1, -1, 1]]
                                   ]])
img = conv1.weight.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('Kernel 1')
plt.show()
conv2 = nn.Conv2d(1, 1, (3, 3), 1)  # in_channel , out_channel , kennel_size , stride
conv2.weight.data = torch.Tensor([[[[1, -1, 1],
                                    [-1, 1, -1],
                                    [1, -1, 1]]
                                   ]])
img = conv2.weight.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('Kernel 2')
plt.show()
conv3 = nn.Conv2d(1, 1, (3, 3), 1)  # in_channel , out_channel , kennel_size , stride
conv3.weight.data = torch.Tensor([[[[-1, -1, 1],
                                    [-1, 1, -1],
                                    [1, -1, -1]]
                                   ]])
img = conv3.weight.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('Kernel 3')
plt.show()
 
feature_map1 = conv1(x)
feature_map2 = conv2(x)
feature_map3 = conv3(x)
 
print(feature_map1 / 9)
print(feature_map2 / 9)
print(feature_map3 / 9)
 
img = feature_map1.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('卷积后的特征图1')
plt.show()
 
print("--------------- 池化  ---------------")
max_pool = nn.MaxPool2d(2, padding=0, stride=2)  # Pooling
zeroPad = nn.ZeroPad2d(padding=(0, 1, 0, 1))  # pad 0 , Left Right Up Down
 
feature_map_pad_0_1 = zeroPad(feature_map1)
feature_pool_1 = max_pool(feature_map_pad_0_1)
feature_map_pad_0_2 = zeroPad(feature_map2)
feature_pool_2 = max_pool(feature_map_pad_0_2)
feature_map_pad_0_3 = zeroPad(feature_map3)
feature_pool_3 = max_pool(feature_map_pad_0_3)
 
print(feature_pool_1.size())
print(feature_pool_1 / 9)
print(feature_pool_2 / 9)
print(feature_pool_3 / 9)
img = feature_pool_1.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('卷积池化后的特征图1')
plt.show()
 
print("--------------- 激活  ---------------")
activation_function = nn.ReLU()
 
feature_relu1 = activation_function(feature_map1)
feature_relu2 = activation_function(feature_map2)
feature_relu3 = activation_function(feature_map3)
print(feature_relu1 / 9)
print(feature_relu2 / 9)
print(feature_relu3 / 9)
img = feature_relu1.data.squeeze().numpy()  # 将输出转换为图片的格式
plt.imshow(img, cmap='gray')
plt.title('卷积 + relu 后的特征图1')
plt.show()

结果：
在这里插入图片描述

孟小起

关注

1
点赞
踩
0

收藏

觉得还不错? 一键收藏
0
评论
人工智能——作业5：卷积-池化-激活

1.For循环版本：手工实现卷积-池化-激活代码如下：代码来源没有直接调用库函数来计算，而是运用for循环来完成卷积-池化-激活效果代码如下：import numpy as np x = np.array([[-1, -1, -1, -1, -1, -1, -1, -1, -1], [-1, 1, -1, -1, -1, -1, -1, 1, -1], [-1, -1, 1, -1, -1, -1, 1, -1, -1],
复制链接

扫一扫