DL-Art-School/codes/models/classifiers/cifar_resnet_branched.py
2021-06-05 21:34:07 -06:00

186 lines
6.0 KiB
Python

"""resnet in pytorch
[1] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun.
Deep Residual Learning for Image Recognition
https://arxiv.org/abs/1512.03385v1
"""
import torch
import torch.nn as nn
from trainer.networks import register_model
class BasicBlock(nn.Module):
"""Basic Block for resnet 18 and resnet 34
"""
#BasicBlock and BottleNeck block
#have different output size
#we use class attribute expansion
#to distinct
expansion = 1
def __init__(self, in_channels, out_channels, stride=1):
super().__init__()
#residual function
self.residual_function = nn.Sequential(
nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=stride, padding=1, bias=False),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True),
nn.Conv2d(out_channels, out_channels * BasicBlock.expansion, kernel_size=3, padding=1, bias=False),
nn.BatchNorm2d(out_channels * BasicBlock.expansion)
)
#shortcut
self.shortcut = nn.Sequential()
#the shortcut output dimension is not the same with residual function
#use 1*1 convolution to match the dimension
if stride != 1 or in_channels != BasicBlock.expansion * out_channels:
self.shortcut = nn.Sequential(
nn.Conv2d(in_channels, out_channels * BasicBlock.expansion, kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(out_channels * BasicBlock.expansion)
)
def forward(self, x):
return nn.ReLU(inplace=True)(self.residual_function(x) + self.shortcut(x))
class BottleNeck(nn.Module):
"""Residual block for resnet over 50 layers
"""
expansion = 4
def __init__(self, in_channels, out_channels, stride=1):
super().__init__()
self.residual_function = nn.Sequential(
nn.Conv2d(in_channels, out_channels, kernel_size=1, bias=False),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True),
nn.Conv2d(out_channels, out_channels, stride=stride, kernel_size=3, padding=1, bias=False),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True),
nn.Conv2d(out_channels, out_channels * BottleNeck.expansion, kernel_size=1, bias=False),
nn.BatchNorm2d(out_channels * BottleNeck.expansion),
)
self.shortcut = nn.Sequential()
if stride != 1 or in_channels != out_channels * BottleNeck.expansion:
self.shortcut = nn.Sequential(
nn.Conv2d(in_channels, out_channels * BottleNeck.expansion, stride=stride, kernel_size=1, bias=False),
nn.BatchNorm2d(out_channels * BottleNeck.expansion)
)
def forward(self, x):
return nn.ReLU(inplace=True)(self.residual_function(x) + self.shortcut(x))
class ResNetTail(nn.Module):
def __init__(self, block, num_block, num_classes=100):
super().__init__()
self.in_channels = 128
self.conv4_x = self._make_layer(block, 256, num_block[2], 2)
self.conv5_x = self._make_layer(block, 512, num_block[3], 2)
self.avg_pool = nn.AdaptiveAvgPool2d((1, 1))
self.fc = nn.Linear(512 * block.expansion, num_classes)
def _make_layer(self, block, out_channels, num_blocks, stride):
strides = [stride] + [1] * (num_blocks - 1)
layers = []
for stride in strides:
layers.append(block(self.in_channels, out_channels, stride))
self.in_channels = out_channels * block.expansion
return nn.Sequential(*layers)
def forward(self, x):
output = self.conv4_x(x)
output = self.conv5_x(output)
output = self.avg_pool(output)
output = output.view(output.size(0), -1)
output = self.fc(output)
return output
class ResNet(nn.Module):
def __init__(self, block, num_block, num_classes=100, num_tails=20):
super().__init__()
self.in_channels = 64
self.conv1 = nn.Sequential(
nn.Conv2d(3, 64, kernel_size=3, padding=1, bias=False),
nn.BatchNorm2d(64),
nn.ReLU(inplace=True))
self.conv2_x = self._make_layer(block, 64, num_block[0], 1)
self.conv3_x = self._make_layer(block, 128, num_block[1], 2)
self.tails = nn.ModuleList([ResNetTail(block, num_block, 256) for _ in range(num_tails)])
self.selector = ResNetTail(block, num_block, num_tails)
self.final_linear = nn.Linear(256, 100)
def _make_layer(self, block, out_channels, num_blocks, stride):
strides = [stride] + [1] * (num_blocks - 1)
layers = []
for stride in strides:
layers.append(block(self.in_channels, out_channels, stride))
self.in_channels = out_channels * block.expansion
return nn.Sequential(*layers)
def forward(self, x, coarse_label):
output = self.conv1(x)
output = self.conv2_x(output)
output = self.conv3_x(output)
keys = []
for t in self.tails:
keys.append(t(output))
keys = torch.stack(keys, dim=1)
query = self.selector(output).unsqueeze(2)
attn = torch.nn.functional.softmax(query * keys, dim=1)
values = self.final_linear(attn * keys)
return values.sum(dim=1)
#bs = output.shape[0]
#return (tailouts[coarse_label] * torch.eye(n=bs, device=x.device).view(bs,bs,1)).sum(dim=1)
@register_model
def register_cifar_resnet18_branched(opt_net, opt):
""" return a ResNet 18 object
"""
return ResNet(BasicBlock, [2, 2, 2, 2])
def resnet34():
""" return a ResNet 34 object
"""
return ResNet(BasicBlock, [3, 4, 6, 3])
def resnet50():
""" return a ResNet 50 object
"""
return ResNet(BottleNeck, [3, 4, 6, 3])
def resnet101():
""" return a ResNet 101 object
"""
return ResNet(BottleNeck, [3, 4, 23, 3])
def resnet152():
""" return a ResNet 152 object
"""
return ResNet(BottleNeck, [3, 8, 36, 3])
if __name__ == '__main__':
model = ResNet(BasicBlock, [2,2,2,2])
print(model(torch.randn(2,3,32,32), torch.LongTensor([4,19])).shape)