net.py

# Code reused from: https://github.com/kuangliu/pytorch-cifar
import math
import torch
import torch.nn as nn
import torch.nn.functional as F
import losses


class BasicBlock(nn.Module):
    def __init__(self, in_planes, out_planes, dropRate=0.0):
        super(BasicBlock, self).__init__()
        self.bn1 = nn.BatchNorm2d(in_planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv1 = nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=1, padding=1, bias=False)
        self.droprate = dropRate

    def forward(self, x):
        out = self.conv1(self.relu(self.bn1(x)))
        if self.droprate > 0:
            out = F.dropout(out, p=self.droprate, training=self.training)
        return torch.cat([x, out], 1)


class BottleneckBlock(nn.Module):
    def __init__(self, in_planes, out_planes, dropRate=0.0):
        super(BottleneckBlock, self).__init__()
        inter_planes = out_planes * 4
        self.bn1 = nn.BatchNorm2d(in_planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv1 = nn.Conv2d(in_planes, inter_planes, kernel_size=1, stride=1, padding=0, bias=False)
        self.bn2 = nn.BatchNorm2d(inter_planes)
        self.conv2 = nn.Conv2d(inter_planes, out_planes, kernel_size=3, stride=1, padding=1, bias=False)
        self.droprate = dropRate

    def forward(self, x):
        out = self.conv1(self.relu(self.bn1(x)))
        if self.droprate > 0:
            out = F.dropout(out, p=self.droprate, inplace=False, training=self.training)
        out = self.conv2(self.relu(self.bn2(out)))
        if self.droprate > 0:
            out = F.dropout(out, p=self.droprate, inplace=False, training=self.training)
        return torch.cat([x, out], 1)


class TransitionBlock(nn.Module):
    def __init__(self, in_planes, out_planes, dropRate=0.0):
        super(TransitionBlock, self).__init__()
        self.bn1 = nn.BatchNorm2d(in_planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv1 = nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=1, padding=0, bias=False)
        self.droprate = dropRate

    def forward(self, x):
        out = self.conv1(self.relu(self.bn1(x)))
        if self.droprate > 0:
            out = F.dropout(out, p=self.droprate, inplace=False, training=self.training)
        return F.avg_pool2d(out, 2)


class DenseBlock(nn.Module):
    def __init__(self, nb_layers, in_planes, growth_rate, block, dropRate=0.0):
        super(DenseBlock, self).__init__()
        self.layer = self._make_layer(block, in_planes, growth_rate, nb_layers, dropRate)

    def _make_layer(self, block, in_planes, growth_rate, nb_layers, dropRate):
        layers = []
        for i in range(int(nb_layers)):
            layers.append(block(in_planes + i * growth_rate, growth_rate, dropRate))
        return nn.Sequential(*layers)

    def forward(self, x):
        return self.layer(x)


class DenseNet3(nn.Module):
    def __init__(self, depth, num_classes, growth_rate=12, reduction=0.5, bottleneck=True, dropRate=0.0):
        super(DenseNet3, self).__init__()
        in_planes = 2 * growth_rate
        n = (depth - 4) / 3
        if bottleneck == True:
            n = n / 2
            block = BottleneckBlock
        else:
            block = BasicBlock
        # 1st conv before any dense block
        self.conv1 = nn.Conv2d(3, in_planes, kernel_size=3, stride=1, padding=1, bias=False)
        # 1st block
        self.block1 = DenseBlock(n, in_planes, growth_rate, block, dropRate)
        in_planes = int(in_planes + n * growth_rate)
        self.trans1 = TransitionBlock(in_planes, int(math.floor(in_planes * reduction)), dropRate=dropRate)
        in_planes = int(math.floor(in_planes * reduction))
        # 2nd block
        self.block2 = DenseBlock(n, in_planes, growth_rate, block, dropRate)
        in_planes = int(in_planes + n * growth_rate)
        self.trans2 = TransitionBlock(in_planes, int(math.floor(in_planes * reduction)), dropRate=dropRate)
        in_planes = int(math.floor(in_planes * reduction))
        # 3rd block
        self.block3 = DenseBlock(n, in_planes, growth_rate, block, dropRate)
        in_planes = int(in_planes + n * growth_rate)
        # global average pooling and classifier
        self.bn1 = nn.BatchNorm2d(in_planes)
        self.relu = nn.ReLU(inplace=True)
        self.in_planes = in_planes

        #############################################################################
        #self.classifier = nn.Linear(in_planes, num_classes)
        self.classifier = losses.IsoMaxPlusLossFirstPart(in_planes, num_classes)
        #############################################################################
      
        # Official init from torch repo.
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                #nn.init.kaiming_normal_(m.weight)
                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
            elif isinstance(m, nn.BatchNorm2d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)
            #elif isinstance(m, nn.Linear):
            #    nn.init.constant_(m.bias, 0)

    def forward(self, x):
        out = self.conv1(x)
        out = self.trans1(self.block1(out))
        out = self.trans2(self.block2(out))
        out = self.block3(out)
        out = self.relu(self.bn1(out))
        out = F.avg_pool2d(out, 8)
        out = out.view(-1, self.in_planes)
        return self.classifier(out)