resnet18.py

resnet18.py

'''ResNet in PyTorch.

For Pre-activation ResNet, see 'preact_resnet.py'.

Reference:
[1] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
    Deep Residual Learning for Image Recognition. arXiv:1512.03385
'''
import torch
import torch.nn as nn
import torch.nn.functional as F
from inplace_abn import ABN, InPlaceABN

Norm = InPlaceABN
act = 'leaky_relu'
act_param = 0.01 

class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, in_planes, planes, stride=1):
        super(BasicBlock, self).__init__()
        self.conv1 = nn.Conv2d(in_planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
        #self.bn1 = nn.BatchNorm2d(planes)
        self.bn1 = Norm(planes, activation=act, activation_param=act_param)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, padding=1, bias=False)
        #self.bn2 = nn.BatchNorm2d(planes)
        self.bn2 = Norm(planes, activation=act, activation_param=act_param)

        self.shortcut = nn.Sequential()
        if stride != 1 or in_planes != self.expansion*planes:
            self.shortcut = nn.Sequential(
                nn.Conv2d(in_planes, self.expansion*planes, kernel_size=1, stride=stride, bias=False),
                #nn.BatchNorm2d(self.expansion*planes)
                Norm(self.expansion*planes, activation=act, activation_param=act_param)
            )

    def forward(self, x):
        out = self.bn1(self.conv1(x))
        out = self.bn2(self.conv2(out))
        out = out + self.shortcut(x)
        out = F.relu(out, inplace=True)
        return out


class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, in_planes, planes, stride=1):
        super(Bottleneck, self).__init__()
        self.conv1 = nn.Conv2d(in_planes, planes, kernel_size=1, bias=False)
        #self.bn1 = nn.BatchNorm2d(planes)
        self.bn1 = Norm(planes, activation=act, activation_param=act_param)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
        #self.bn2 = nn.BatchNorm2d(planes)
        self.bn2 = Norm(planes, activation=act, activation_param=act_param)
        self.conv3 = nn.Conv2d(planes, self.expansion*planes, kernel_size=1, bias=False)
        #self.bn3 = nn.BatchNorm2d(self.expansion*planes)
        self.bn3 = Norm(self.expansion*planes, activation=act, activation_param=act_param)

        self.shortcut = nn.Sequential()
        if stride != 1 or in_planes != self.expansion*planes:
            self.shortcut = nn.Sequential(
                nn.Conv2d(in_planes, self.expansion*planes, kernel_size=1, stride=stride, bias=False),
                #nn.BatchNorm2d(self.expansion*planes)
                Norm(self.expansion*planes, activation=act, activation_param=act_param)
            )

    def forward(self, x):
        out = self.bn1(self.conv1(x))
        out = self.bn2(self.conv2(out))
        out = self.bn3(self.conv3(out))
        out = out + self.shortcut(x)
        out = F.relu(out, inplace=True)
        return out


class ResNet(nn.Module):
    def __init__(self, block, num_blocks, num_classes=10):
        super(ResNet, self).__init__()
        self.in_planes = 64

        self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1, bias=False)
        #self.bn1 = nn.BatchNorm2d(64)
        self.bn1 = Norm(64, activation=act, activation_param=act_param)
        self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
        self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2)
        self.layer3 = self._make_layer(block, 256, num_blocks[2], stride=2)
        self.layer4 = self._make_layer(block, 512, num_blocks[3], stride=2)
        self.linear = nn.Linear(512*block.expansion, num_classes)
        '''
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
            elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm, InPlaceABN)):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)

        # Zero-initialize the last BN in each residual branch,
        # so that the residual branch starts with zeros, and each residual block behaves like an identity.
        # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677
        for m in self.modules():
            if isinstance(m, Bottleneck):
                nn.init.constant_(m.bn3.weight, 0)
            elif isinstance(m, BasicBlock):
                nn.init.constant_(m.bn2.weight, 0)
        '''

    def _make_layer(self, block, planes, num_blocks, stride):
        strides = [stride] + [1]*(num_blocks-1)
        layers = []
        for stride in strides:
            layers.append(block(self.in_planes, planes, stride))
            self.in_planes = planes * block.expansion
        return nn.Sequential(*layers)

    def forward(self, x):
        out = self.bn1(self.conv1(x))
        out = self.layer1(out)
        out = self.layer2(out)
        out = self.layer3(out)
        out = self.layer4(out)
        out = F.avg_pool2d(out, 4)
        out = out.view(out.size(0), -1)
        out = self.linear(out)
        return out


def ResNet18():
    return ResNet(BasicBlock, [2,2,2,2])