当前位置: 首页 > 编程笔记 >

dpn网络的pytorch实现方式

潘佐
2023-03-14
本文向大家介绍dpn网络的pytorch实现方式,包括了dpn网络的pytorch实现方式的使用技巧和注意事项,需要的朋友参考一下

我就废话不多说了,直接上代码吧!

import torch
import torch.nn as nn
import torch.nn.functional as F



class CatBnAct(nn.Module):
 def __init__(self, in_chs, activation_fn=nn.ReLU(inplace=True)):
  super(CatBnAct, self).__init__()
  self.bn = nn.BatchNorm2d(in_chs, eps=0.001)
  self.act = activation_fn

 def forward(self, x):
  x = torch.cat(x, dim=1) if isinstance(x, tuple) else x
  return self.act(self.bn(x))


class BnActConv2d(nn.Module):
 def __init__(self, s, out_chs, kernel_size, stride,
     padding=0, groups=1, activation_fn=nn.ReLU(inplace=True)):
  super(BnActConv2d, self).__init__()
  self.bn = nn.BatchNorm2d(in_chs, eps=0.001)
  self.act = activation_fn
  self.conv = nn.Conv2d(in_chs, out_chs, kernel_size, stride, padding, groups=groups, bias=False)

 def forward(self, x):
  return self.conv(self.act(self.bn(x)))


class InputBlock(nn.Module):
 def __init__(self, num_init_features, kernel_size=7,
     padding=3, activation_fn=nn.ReLU(inplace=True)):
  super(InputBlock, self).__init__()
  self.conv = nn.Conv2d(
   3, num_init_features, kernel_size=kernel_size, stride=2, padding=padding, bias=False)
  self.bn = nn.BatchNorm2d(num_init_features, eps=0.001)
  self.act = activation_fn
  self.pool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)

 def forward(self, x):
  x = self.conv(x)
  x = self.bn(x)
  x = self.act(x)
  x = self.pool(x)
  return x


class DualPathBlock(nn.Module):
 def __init__(
   self, in_chs, num_1x1_a, num_3x3_b, num_1x1_c, inc, groups, block_type='normal', b=False):
  super(DualPathBlock, self).__init__()
  self.num_1x1_c = num_1x1_c
  self.inc = inc
  self.b = b
  if block_type is 'proj':
   self.key_stride = 1
   self.has_proj = True
  elif block_type is 'down':
   self.key_stride = 2
   self.has_proj = True
  else:
   assert block_type is 'normal'
   self.key_stride = 1
   self.has_proj = False

  if self.has_proj:
   # Using different member names here to allow easier parameter key matching for conversion
   if self.key_stride == 2:
    self.c1x1_w_s2 = BnActConv2d(
     in_chs=in_chs, out_chs=num_1x1_c + 2 * inc, kernel_size=1, stride=2)
   else:
    self.c1x1_w_s1 = BnActConv2d(
     in_chs=in_chs, out_chs=num_1x1_c + 2 * inc, kernel_size=1, stride=1)
  self.c1x1_a = BnActConv2d(in_chs=in_chs, out_chs=num_1x1_a, kernel_size=1, stride=1)
  self.c3x3_b = BnActConv2d(
   in_chs=num_1x1_a, out_chs=num_3x3_b, kernel_size=3,
   stride=self.key_stride, padding=1, groups=groups)
  if b:
   self.c1x1_c = CatBnAct(in_chs=num_3x3_b)
   self.c1x1_c1 = nn.Conv2d(num_3x3_b, num_1x1_c, kernel_size=1, bias=False)
   self.c1x1_c2 = nn.Conv2d(num_3x3_b, inc, kernel_size=1, bias=False)
  else:
   self.c1x1_c = BnActConv2d(in_chs=num_3x3_b, out_chs=num_1x1_c + inc, kernel_size=1, stride=1)

 def forward(self, x):
  x_in = torch.cat(x, dim=1) if isinstance(x, tuple) else x
  if self.has_proj:
   if self.key_stride == 2:
    x_s = self.c1x1_w_s2(x_in)
   else:
    x_s = self.c1x1_w_s1(x_in)
   x_s1 = x_s[:, :self.num_1x1_c, :, :]
   x_s2 = x_s[:, self.num_1x1_c:, :, :]
  else:
   x_s1 = x[0]
   x_s2 = x[1]
  x_in = self.c1x1_a(x_in)
  x_in = self.c3x3_b(x_in)
  if self.b:
   x_in = self.c1x1_c(x_in)
   out1 = self.c1x1_c1(x_in)
   out2 = self.c1x1_c2(x_in)
  else:
   x_in = self.c1x1_c(x_in)
   out1 = x_in[:, :self.num_1x1_c, :, :]
   out2 = x_in[:, self.num_1x1_c:, :, :]
  resid = x_s1 + out1
  dense = torch.cat([x_s2, out2], dim=1)
  return resid, dense


class DPN(nn.Module):
 def __init__(self, small=False, num_init_features=64, k_r=96, groups=32,
     b=False, k_sec=(3, 4, 20, 3), inc_sec=(16, 32, 24, 128),
     num_classes=1000, test_time_pool=False):
  super(DPN, self).__init__()
  self.test_time_pool = test_time_pool
  self.b = b
  bw_factor = 1 if small else 4

  blocks = OrderedDict()

  # conv1
  if small:
   blocks['conv1_1'] = InputBlock(num_init_features, kernel_size=3, padding=1)
  else:
   blocks['conv1_1'] = InputBlock(num_init_features, kernel_size=7, padding=3)

  # conv2
  bw = 64 * bw_factor
  inc = inc_sec[0]
  r = (k_r * bw) // (64 * bw_factor)
  blocks['conv2_1'] = DualPathBlock(num_init_features, r, r, bw, inc, groups, 'proj', b)
  in_chs = bw + 3 * inc
  for i in range(2, k_sec[0] + 1):
   blocks['conv2_' + str(i)] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'normal', b)
   in_chs += inc

  # conv3
  bw = 128 * bw_factor
  inc = inc_sec[1]
  r = (k_r * bw) // (64 * bw_factor)
  blocks['conv3_1'] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'down', b)
  in_chs = bw + 3 * inc
  for i in range(2, k_sec[1] + 1):
   blocks['conv3_' + str(i)] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'normal', b)
   in_chs += inc

  # conv4
  bw = 256 * bw_factor
  inc = inc_sec[2]
  r = (k_r * bw) // (64 * bw_factor)
  blocks['conv4_1'] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'down', b)
  in_chs = bw + 3 * inc
  for i in range(2, k_sec[2] + 1):
   blocks['conv4_' + str(i)] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'normal', b)
   in_chs += inc

  # conv5
  bw = 512 * bw_factor
  inc = inc_sec[3]
  r = (k_r * bw) // (64 * bw_factor)
  blocks['conv5_1'] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'down', b)
  in_chs = bw + 3 * inc
  for i in range(2, k_sec[3] + 1):
   blocks['conv5_' + str(i)] = DualPathBlock(in_chs, r, r, bw, inc, groups, 'normal', b)
   in_chs += inc
  blocks['conv5_bn_ac'] = CatBnAct(in_chs)

  self.features = nn.Sequential(blocks)

  # Using 1x1 conv for the FC layer to allow the extra pooling scheme
  self.last_linear = nn.Conv2d(in_chs, num_classes, kernel_size=1, bias=True)

 def logits(self, features):
  if not self.training and self.test_time_pool:
   x = F.avg_pool2d(features, kernel_size=7, stride=1)
   out = self.last_linear(x)
   # The extra test time pool should be pooling an img_size//32 - 6 size patch
   out = adaptive_avgmax_pool2d(out, pool_type='avgmax')
  else:
   x = adaptive_avgmax_pool2d(features, pool_type='avg')
   out = self.last_linear(x)
  return out.view(out.size(0), -1)

 def forward(self, input):
  x = self.features(input)
  x = self.logits(x)
  return x

""" PyTorch selectable adaptive pooling
Adaptive pooling with the ability to select the type of pooling from:
 * 'avg' - Average pooling
 * 'max' - Max pooling
 * 'avgmax' - Sum of average and max pooling re-scaled by 0.5
 * 'avgmaxc' - Concatenation of average and max pooling along feature dim, doubles feature dim

Both a functional and a nn.Module version of the pooling is provided.

"""

def pooling_factor(pool_type='avg'):
 return 2 if pool_type == 'avgmaxc' else 1


def adaptive_avgmax_pool2d(x, pool_type='avg', padding=0, count_include_pad=False):
 """Selectable global pooling function with dynamic input kernel size
 """
 if pool_type == 'avgmaxc':
  x = torch.cat([
   F.avg_pool2d(
    x, kernel_size=(x.size(2), x.size(3)), padding=padding, count_include_pad=count_include_pad),
   F.max_pool2d(x, kernel_size=(x.size(2), x.size(3)), padding=padding)
  ], dim=1)
 elif pool_type == 'avgmax':
  x_avg = F.avg_pool2d(
    x, kernel_size=(x.size(2), x.size(3)), padding=padding, count_include_pad=count_include_pad)
  x_max = F.max_pool2d(x, kernel_size=(x.size(2), x.size(3)), padding=padding)
  x = 0.5 * (x_avg + x_max)
 elif pool_type == 'max':
  x = F.max_pool2d(x, kernel_size=(x.size(2), x.size(3)), padding=padding)
 else:
  if pool_type != 'avg':
   print('Invalid pool type %s specified. Defaulting to average pooling.' % pool_type)
  x = F.avg_pool2d(
   x, kernel_size=(x.size(2), x.size(3)), padding=padding, count_include_pad=count_include_pad)
 return x


class AdaptiveAvgMaxPool2d(torch.nn.Module):
 """Selectable global pooling layer with dynamic input kernel size
 """
 def __init__(self, output_size=1, pool_type='avg'):
  super(AdaptiveAvgMaxPool2d, self).__init__()
  self.output_size = output_size
  self.pool_type = pool_type
  if pool_type == 'avgmaxc' or pool_type == 'avgmax':
   self.pool = nn.ModuleList([nn.AdaptiveAvgPool2d(output_size), nn.AdaptiveMaxPool2d(output_size)])
  elif pool_type == 'max':
   self.pool = nn.AdaptiveMaxPool2d(output_size)
  else:
   if pool_type != 'avg':
    print('Invalid pool type %s specified. Defaulting to average pooling.' % pool_type)
   self.pool = nn.AdaptiveAvgPool2d(output_size)

 def forward(self, x):
  if self.pool_type == 'avgmaxc':
   x = torch.cat([p(x) for p in self.pool], dim=1)
  elif self.pool_type == 'avgmax':
   x = 0.5 * torch.sum(torch.stack([p(x) for p in self.pool]), 0).squeeze(dim=0)
  else:
   x = self.pool(x)
  return x

 def factor(self):
  return pooling_factor(self.pool_type)

 def __repr__(self):
  return self.__class__.__name__ + ' (' \
    + 'output_size=' + str(self.output_size) \
    + ', pool_type=' + self.pool_type + ')'

以上这篇dpn网络的pytorch实现方式就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持小牛知识库。

 类似资料:
  • 本文向大家介绍关于ResNeXt网络的pytorch实现,包括了关于ResNeXt网络的pytorch实现的使用技巧和注意事项,需要的朋友参考一下 此处需要pip install pretrainedmodels 以上这篇关于ResNeXt网络的pytorch实现就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持呐喊教程。

  • 本文向大家介绍PyTorch上实现卷积神经网络CNN的方法,包括了PyTorch上实现卷积神经网络CNN的方法的使用技巧和注意事项,需要的朋友参考一下 一、卷积神经网络 卷积神经网络(ConvolutionalNeuralNetwork,CNN)最初是为解决图像识别等问题设计的,CNN现在的应用已经不限于图像和视频,也可用于时间序列信号,比如音频信号和文本数据等。CNN作为一个深度学习架构被提出的

  • 本文向大家介绍pytorch打印网络结构的实例,包括了pytorch打印网络结构的实例的使用技巧和注意事项,需要的朋友参考一下 最简单的方法当然可以直接print(net),但是这样网络比较复杂的时候效果不太好,看着比较乱;以前使用caffe的时候有一个网站可以在线生成网络框图,tensorflow可以用tensor board,keras中可以用model.summary()、或者plot_mo

  • 本文向大家介绍Pytorch实现GoogLeNet的方法,包括了Pytorch实现GoogLeNet的方法的使用技巧和注意事项,需要的朋友参考一下 GoogLeNet也叫InceptionNet,在2014年被提出,如今已到V4版本。GoogleNet比VGGNet具有更深的网络结构,一共有22层,但是参数比AlexNet要少12倍,但是计算量是AlexNet的4倍,原因就是它采用很有效的Ince

  • 本文向大家介绍pytorch 求网络模型参数实例,包括了pytorch 求网络模型参数实例的使用技巧和注意事项,需要的朋友参考一下 用pytorch训练一个神经网络时,我们通常会很关心模型的参数总量。下面分别介绍来两种方法求模型参数 一 .求得每一层的模型参数,然后自然的可以计算出总的参数。 1.先初始化一个网络模型model 比如我这里是 model=cliqueNet(里面是些初始化的参数)

  • 本文向大家介绍pytorch构建网络模型的4种方法,包括了pytorch构建网络模型的4种方法的使用技巧和注意事项,需要的朋友参考一下 利用pytorch来构建网络模型有很多种方法,以下简单列出其中的四种。 假设构建一个网络模型如下: 卷积层--》Relu层--》池化层--》全连接层--》Relu层--》全连接层 首先导入几种方法用到的包: 第一种方法 这种方法比较常用,早期的教程通常就是使用这种