import json import torch import torch.nn as nn import torch.utils.model_zoo as model_zoo def conv3x3(in_planes, out_planes, stride=1): """3x3 convolution with padding""" return nn.Conv2d(in_planes, out_planes, kernel_size=(3, 3), stride=(stride, stride), padding=1, bias=False) def conv1x1(in_planes, out_planes, stride=1): """1x1 convolution""" return nn.Conv2d(in_planes, out_planes, kernel_size=(1, 1), stride=(stride, stride), bias=False) class BasicBlock(nn.Module): expansion = 1 def __init__(self, inplanes, planes, stride=1, downsample=None): super(BasicBlock, self).__init__() self.conv1 = conv3x3(inplanes, planes, stride) self.bn1 = nn.BatchNorm2d(planes) self.relu = nn.ReLU(inplace=True) self.conv2 = conv3x3(planes, planes) self.bn2 = nn.BatchNorm2d(planes) self.downsample = downsample self.stride = stride def forward(self, x): identity = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) if self.downsample is not None: identity = self.downsample(x) out += identity out = self.relu(out) return out class Bottleneck(nn.Module): expansion = 4 def __init__(self, inplanes, planes, stride=1, downsample=None): super(Bottleneck, self).__init__() self.conv1 = conv1x1(inplanes, planes) self.bn1 = nn.BatchNorm2d(planes) self.conv2 = conv3x3(planes, planes, stride) self.bn2 = nn.BatchNorm2d(planes) self.conv3 = conv1x1(planes, planes * self.expansion) self.bn3 = nn.BatchNorm2d(planes * self.expansion) self.relu = nn.ReLU(inplace=True) self.downsample = downsample self.stride = stride def forward(self, x): identity = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) out = self.relu(out) out = self.conv3(out) out = self.bn3(out) if self.downsample is not None: identity = self.downsample(x) out += identity out = self.relu(out) return out class fc_block(nn.Module): def __init__(self, inplanes, planes, drop_rate=0.15): super(fc_block, self).__init__() self.fc = nn.Linear(inplanes, planes) self.bn = nn.BatchNorm1d(planes) if drop_rate > 0: self.dropout = nn.Dropout(drop_rate) self.relu = nn.ReLU(inplace=True) self.drop_rate = drop_rate def forward(self, x): x = self.fc(x) x = self.bn(x) if self.drop_rate > 0: x = self.dropout(x) x = self.relu(x) return x class ResNet(nn.Module): def __init__(self, block, layers, attr_file, zero_init_residual=False, dropout_rate=0): super(ResNet, self).__init__() self.inplanes = 64 self.conv1 = nn.Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=3, bias=False) self.bn1 = nn.BatchNorm2d(64) self.relu = nn.ReLU(inplace=True) self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) self.layer1 = self._make_layer(block, 64, layers[0]) self.layer2 = self._make_layer(block, 128, layers[1], stride=2) self.layer3 = self._make_layer(block, 256, layers[2], stride=2) self.layer4 = self._make_layer(block, 512, layers[3], stride=2) self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) self.stem = fc_block(512 * block.expansion, 512, dropout_rate) # Construct classifier heads according to the number of values of each attribute self.attr_file = attr_file with open(self.attr_file, 'r') as f: attr_f = json.load(f) self.attr_info = attr_f['attr_info'] for idx, (key, val) in enumerate(self.attr_info.items()): num_val = int(len(val["value"])) setattr(self, 'classifier' + str(key).zfill(2) + val["name"], nn.Sequential(fc_block(512, 256, dropout_rate), nn.Linear(256, num_val))) 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.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 if zero_init_residual: 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, blocks, stride=1): downsample = None if stride != 1 or self.inplanes != planes * block.expansion: downsample = nn.Sequential(conv1x1(self.inplanes, planes * block.expansion, stride), nn.BatchNorm2d(planes * block.expansion)) layers = [block(self.inplanes, planes, stride, downsample)] self.inplanes = planes * block.expansion for _ in range(1, blocks): layers.append(block(self.inplanes, planes)) return nn.Sequential(*layers) def forward(self, x): x = self.conv1(x) x = self.bn1(x) x = self.relu(x) x = self.maxpool(x) x = self.layer1(x) x = self.layer2(x) x = self.layer3(x) x = self.layer4(x) x = self.avgpool(x) x = x.view(x.size(0), -1) x = self.stem(x) predictions = {} for idx, (key, val) in enumerate(self.attr_info.items()): classifier = getattr(self, 'classifier' + str(key).zfill(2) + val["name"]) predictions.update({val["name"]: classifier(x)}) return predictions def celeba_attr_predictor(attr_file, pretrained='models/pretrained/celeba_attributes/predictor_1024.pth.tar'): model = ResNet(Bottleneck, [3, 4, 6, 3], attr_file=attr_file) init_pretrained_weights(model, 'https://download.pytorch.org/models/resnet50-19c8e357.pth') model.load_state_dict(torch.load(pretrained)['state_dict'], strict=True) return model def init_pretrained_weights(model, model_url): """Initialize model with pretrained weights. Layers that don't match with pretrained layers in name or size are kept unchanged. """ pretrain_dict = model_zoo.load_url(model_url) model_dict = model.state_dict() pretrain_dict = { k: v for k, v in pretrain_dict.items() if k in model_dict and model_dict[k].size() == v.size() } model_dict.update(pretrain_dict) model.load_state_dict(model_dict)