卷积神经网络的实现

最新推荐文章于 2022-08-10 13:48:47 发布

一辈子的拥抱

最新推荐文章于 2022-08-10 13:48:47 发布

阅读量1k

点赞数

分类专栏： python 文章标签： cnn 深度学习 python

本文链接：https://blog.csdn.net/weixin_50719427/article/details/124605275

版权

python 专栏收录该内容

4 篇文章 0 订阅

订阅专栏

第1关：实现卷积层的前向传播

import numpy as np
from utils import im2col


class Convolution:
    def __init__(self, W, b, stride=1, pad=0):
        r'''
        卷积层的初始化

        Parameter:
        - W: numpy.array, (C_out, C_in, K_h, K_w)
        - b: numpy.array, (C_out)
        - stride: int
        - pad: int
        '''
        self.W = W
        self.b = b
        self.stride = stride
        self.pad = pad

    def forward(self, x):
        r'''
        卷积层的前向传播

        Parameter:
        - x: numpy.array, (B, C, H, W)

        Return:
        - y: numpy.array, (B, C', H', W')
             H' = (H - Kh + 2P) / S + 1
             W' = (W - Kw + 2P) / S + 1
        '''
        ########## Begin ##########
        FN, C, FH, FW = self.W.shape
        N, C, H, W = x.shape
        out_h = int(1 + (H + 2 * self.pad - FH) / self.stride)
        out_w = int(1 + (W + 2 * self.pad - FW) / self.stride)

        col = im2col(x, FH, FW, self.stride, self.pad)
        col_w = self.W.reshape(FN, -1).T
        out = np.dot(col, col_w) + self.b

        out = out.reshape(N, out_h, out_w, -1).transpose(0, 3, 1, 2)
        return out

        ########## End ##########

第2关：实现池化层的前向传播

import numpy as np
from utils import im2col


class MaxPool:
    def __init__(self, pool_h, pool_w, stride=1, pad=0):
        r'''
        池化层的初始化

        Parameter:
        - pool_h: int
        - pool_h: int
        - stride: int
        - pad: int
        '''
        self.pool_h = pool_h
        self.pool_w = pool_w
        self.stride = stride
        self.pad = pad

    def forward(self, x):
        r'''
        池化层的前向传播

        Parameter:
        - x: numpy.array, (B, C, H, W)

        Return:
        - y: numpy.array, (B, C, H', W')
             H' = (H - Kh + 2P) / S + 1
             W' = (W - Kw + 2P) / S + 1
        '''
        ########## Begin ##########
        N, C, H, W = x.shape
        out_h = int(1 + (H - self.pool_h) / self.stride)
        out_w = int(1 + (W - self.pool_w) / self.stride)

        col = im2col(x, self.pool_h, self.pool_w, self.stride, self.pad)
        col = col.reshape(-1, self.pool_h * self.pool_w)

        out = np.max(col, axis=1)
        out = out.reshape(N, out_h, out_w, C).transpose(0, 3, 1, 2)
        return out

        ########## End ##########