DPN网络学习

科普知识

[1].深度学习中卷积层和pooling层的输出计算公式
[2].理解分组卷积和深度可分离卷积如何降低参数量
[2].DPN（Dual Path Network）算法详解
[3].Pytorch实现版

DPN架构

import torch
import torch.nn as nn
import torch.nn.functional as F
from collections import OrderedDict

__all__ = ['DPN', 'dpn92', 'dpn98', 'dpn131', 'dpn107', 'dpns']


def dpn92(num_classes=1000):
    return DPN(num_init_features=64, k_R=96, G=32, k_sec=(3,4,20,3), inc_sec=(16,32,24,128), num_classes=num_classes)


def dpn98(num_classes=1000):
    return DPN(num_init_features=96, k_R=160, G=40, k_sec=(3,6,20,3), inc_sec=(16,32,32,128), num_classes=num_classes)


def dpn131(num_classes=1000):
    ret