inceptionv4_sa.py

import torch
import torch.nn as nn
import torch.utils.model_zoo as model_zoo
import os
import sys
from torch.nn.parameter import Parameter
import torch.nn.functional as F

__all__ = ['InceptionV4', 'inceptionv4']

pretrained_settings = {
    'inceptionv4': {
        'imagenet': {
            'url': 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth',
            'input_space': 'RGB',
            'input_size': [3, 299, 299],
            'input_range': [0, 1],
            'mean': [0.5, 0.5, 0.5],
            'std': [0.5, 0.5, 0.5],
            'num_classes': 1000
        },
        'imagenet+background': {
            'url': 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth',
            'input_space': 'RGB',
            'input_size': [3, 299, 299],
            'input_range': [0, 1],
            'mean': [0.5, 0.5, 0.5],
            'std': [0.5, 0.5, 0.5],
            'num_classes': 1001
        }
    }
}



class ParamFree(nn.Module):
    def __init__(self, channel):
        super(ParamFree, self).__init__()
        self.gap = nn.AdaptiveAvgPool2d(1)
        self.gap2 = nn.AdaptiveAvgPool2d(2)
        self.gap4 = nn.AdaptiveAvgPool2d(4)
        self.sig = nn.Sigmoid()
        self.learn = Parameter(torch.zeros(3))
        self.batch_norm = nn.BatchNorm2d(channel)

    def forward(self, x):
        b, c, h, w = x.size()
        gap = self.gap(x)
        gap2 = self.gap2(x)
        gap4 = self.gap4(x)
        gapALL = (gap*self.learn[0]+gap4*self.learn[2]+F.interpolate(gap2, scale_factor=2)*self.learn[1])/3
        gapALL = F.interpolate(gapALL, size=(h, w))
        gapALL = self.batch_norm(gapALL)
        gapALL = self.sig(gapALL)

        return x*gapALL



class BasicConv2d(nn.Module):

    def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0):
        super(BasicConv2d, self).__init__()
        self.conv = nn.Conv2d(in_planes, out_planes,
                              kernel_size=kernel_size, stride=stride,
                              padding=padding, bias=False)  # verify bias false
        self.bn = nn.BatchNorm2d(out_planes,
                                 eps=0.001,  # value found in tensorflow
                                 momentum=0.1,  # default pytorch value
                                 affine=True)
        self.relu = nn.ReLU(inplace=True)

    def forward(self, x):
        x = self.conv(x)
        x = self.bn(x)
        x = self.relu(x)
        return x


class Mixed_3a(nn.Module):

    def __init__(self):
        super(Mixed_3a, self).__init__()
        self.maxpool = nn.MaxPool2d(3, stride=2)
        self.conv = BasicConv2d(64, 96, kernel_size=3, stride=2)
        self.sa = ParamFree(160)

    def forward(self, x):
        x0 = self.maxpool(x)
        x1 = self.conv(x)
        out = torch.cat((x0, x1), 1)
        out = self.sa(out)

        return out


class Mixed_4a(nn.Module):

    def __init__(self):
        super(Mixed_4a, self).__init__()

        self.branch0 = nn.Sequential(
            BasicConv2d(160, 64, kernel_size=1, stride=1),
            BasicConv2d(64, 96, kernel_size=3, stride=1)
        )

        self.branch1 = nn.Sequential(
            BasicConv2d(160, 64, kernel_size=1, stride=1),
            BasicConv2d(64, 64, kernel_size=(1, 7), stride=1, padding=(0, 3)),
            BasicConv2d(64, 64, kernel_size=(7, 1), stride=1, padding=(3, 0)),
            BasicConv2d(64, 96, kernel_size=(3, 3), stride=1)
        )
        self.sa = ParamFree(192)

    def forward(self, x):
        x0 = self.branch0(x)
        x1 = self.branch1(x)
        out = torch.cat((x0, x1), 1)
        out = self.sa(out)

        return out


class Mixed_5a(nn.Module):

    def __init__(self):
        super(Mixed_5a, self).__init__()
        self.conv = BasicConv2d(192, 192, kernel_size=3, stride=2)
        self.maxpool = nn.MaxPool2d(3, stride=2)
        self.sa = ParamFree(384)

    def forward(self, x):
        x0 = self.conv(x)
        x1 = self.maxpool(x)
        out = torch.cat((x0, x1), 1)
        out = self.sa(out)

        return out


class Inception_A(nn.Module):

    def __init__(self):
        super(Inception_A, self).__init__()
        self.branch0 = BasicConv2d(384, 96, kernel_size=1, stride=1)

        self.branch1 = nn.Sequential(
            BasicConv2d(384, 64, kernel_size=1, stride=1),
            BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1)
        )

        self.branch2 = nn.Sequential(
            BasicConv2d(384, 64, kernel_size=1, stride=1),
            BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1),
            BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1)
        )

        self.branch3 = nn.Sequential(
            nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False),
            BasicConv2d(384, 96, kernel_size=1, stride=1)
        )
        self.sa = ParamFree(384)

    def forward(self, x):
        x0 = self.branch0(x)
        x1 = self.branch1(x)
        x2 = self.branch2(x)
        x3 = self.branch3(x)
        out = torch.cat((x0, x1, x2, x3), 1)
        out = self.sa(out)

        return out


class Reduction_A(nn.Module):

    def __init__(self):
        super(Reduction_A, self).__init__()
        self.branch0 = BasicConv2d(384, 384, kernel_size=3, stride=2)

        self.branch1 = nn.Sequential(
            BasicConv2d(384, 192, kernel_size=1, stride=1),
            BasicConv2d(192, 224, kernel_size=3, stride=1, padding=1),
            BasicConv2d(224, 256, kernel_size=3, stride=2)
        )

        self.branch2 = nn.MaxPool2d(3, stride=2)

    def forward(self, x):
        x0 = self.branch0(x)
        x1 = self.branch1(x)
        x2 = self.branch2(x)
        out = torch.cat((x0, x1, x2), 1)
        return out


class Inception_B(nn.Module):

    def __init__(self):
        super(Inception_B, self).__init__()
        self.branch0 = BasicConv2d(1024, 384, kernel_size=1, stride=1)

        self.branch1 = nn.Sequential(
            BasicConv2d(1024, 192, kernel_size=1, stride=1),
            BasicConv2d(192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3)),
            BasicConv2d(224, 256, kernel_size=(7, 1), stride=1, padding=(3, 0))
        )

        self.branch2 = nn.Sequential(
            BasicConv2d(1024, 192, kernel_size=1, stride=1),
            BasicConv2d(192, 192, kernel_size=(7, 1), stride=1, padding=(3, 0)),
            BasicConv2d(192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3)),
            BasicConv2d(224, 224, kernel_size=(7, 1), stride=1, padding=(3, 0)),
            BasicConv2d(224, 256, kernel_size=(1, 7), stride=1, padding=(0, 3))
        )

        self.branch3 = nn.Sequential(
            nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False),
            BasicConv2d(1024, 128, kernel_size=1, stride=1)
        )
        self.sa = ParamFree(1024)

    def forward(self, x):
        x0 = self.branch0(x)
        x1 = self.branch1(x)
        x2 = self.branch2(x)
        x3 = self.branch3(x)
        out = torch.cat((x0, x1, x2, x3), 1)
        out = self.sa(out)

        return out


class Reduction_B(nn.Module):

    def __init__(self):
        super(Reduction_B, self).__init__()

        self.branch0 = nn.Sequential(
            BasicConv2d(1024, 192, kernel_size=1, stride=1),
            BasicConv2d(192, 192, kernel_size=3, stride=2)
        )

        self.branch1 = nn.Sequential(
            BasicConv2d(1024, 256, kernel_size=1, stride=1),
            BasicConv2d(256, 256, kernel_size=(1, 7), stride=1, padding=(0, 3)),
            BasicConv2d(256, 320, kernel_size=(7, 1), stride=1, padding=(3, 0)),
            BasicConv2d(320, 320, kernel_size=3, stride=2)
        )

        self.branch2 = nn.MaxPool2d(3, stride=2)

    def forward(self, x):
        x0 = self.branch0(x)
        x1 = self.branch1(x)
        x2 = self.branch2(x)
        out = torch.cat((x0, x1, x2), 1)
        return out


class Inception_C(nn.Module):

    def __init__(self):
        super(Inception_C, self).__init__()

        self.branch0 = BasicConv2d(1536, 256, kernel_size=1, stride=1)

        self.branch1_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1)
        self.branch1_1a = BasicConv2d(384, 256, kernel_size=(1, 3), stride=1, padding=(0, 1))
        self.branch1_1b = BasicConv2d(384, 256, kernel_size=(3, 1), stride=1, padding=(1, 0))

        self.branch2_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1)
        self.branch2_1 = BasicConv2d(384, 448, kernel_size=(3, 1), stride=1, padding=(1, 0))
        self.branch2_2 = BasicConv2d(448, 512, kernel_size=(1, 3), stride=1, padding=(0, 1))
        self.branch2_3a = BasicConv2d(512, 256, kernel_size=(1, 3), stride=1, padding=(0, 1))
        self.branch2_3b = BasicConv2d(512, 256, kernel_size=(3, 1), stride=1, padding=(1, 0))

        self.branch3 = nn.Sequential(
            nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False),
            BasicConv2d(1536, 256, kernel_size=1, stride=1)
        )
        self.sa = ParamFree(1536)

    def forward(self, x):
        x0 = self.branch0(x)

        x1_0 = self.branch1_0(x)
        x1_1a = self.branch1_1a(x1_0)
        x1_1b = self.branch1_1b(x1_0)
        x1 = torch.cat((x1_1a, x1_1b), 1)

        x2_0 = self.branch2_0(x)
        x2_1 = self.branch2_1(x2_0)
        x2_2 = self.branch2_2(x2_1)
        x2_3a = self.branch2_3a(x2_2)
        x2_3b = self.branch2_3b(x2_2)
        x2 = torch.cat((x2_3a, x2_3b), 1)

        x3 = self.branch3(x)

        out = torch.cat((x0, x1, x2, x3), 1)
        out = self.sa(out)

        return out


class InceptionV4(nn.Module):

    def __init__(self, num_classes=1001):
        super(InceptionV4, self).__init__()
        # Special attributs
        self.input_space = None
        self.input_size = (299, 299, 3)
        self.mean = None
        self.std = None
        # Modules
        self.features = nn.Sequential(
            BasicConv2d(3, 32, kernel_size=3, stride=2),
            BasicConv2d(32, 32, kernel_size=3, stride=1),
            BasicConv2d(32, 64, kernel_size=3, stride=1, padding=1),
            Mixed_3a(),
            Mixed_4a(),
            Mixed_5a(),
            Inception_A(),
            Inception_A(),
            Inception_A(),
            Inception_A(),
            Reduction_A(),  # Mixed_6a
            Inception_B(),
            Inception_B(),
            Inception_B(),
            Inception_B(),
            Inception_B(),
            Inception_B(),
            Inception_B(),
            Reduction_B(),  # Mixed_7a
            Inception_C(),
            Inception_C(),
            Inception_C()
        )
        self.avg_pool = nn.AvgPool2d(8, count_include_pad=False)
        self.last_linear = nn.Linear(1536, num_classes)

    def logits(self, features):
        x = self.avg_pool(features)
        x = x.view(x.size(0), -1)
        x = self.last_linear(x)
        return x

    def forward(self, input):
        x = self.features(input)
        x = self.logits(x)
        return x


def inceptionv4(num_classes=365, pretrained=None):
    if pretrained:
        print('pretrained')
        settings = pretrained_settings['inceptionv4'][pretrained]
        assert num_classes == settings['num_classes'], \
            "num_classes should be {}, but is {}".format(settings['num_classes'], num_classes)

        # both 'imagenet'&'imagenet+background' are loaded from same parameters
        model = InceptionV4(num_classes=1001)
        model.load_state_dict(model_zoo.load_url(settings['url']))

        if pretrained == 'imagenet':
            new_last_linear = nn.Linear(1536, 365)
            new_last_linear.weight.data = model.last_linear.weight.data[1:]
            new_last_linear.bias.data = model.last_linear.bias.data[1:]
            model.last_linear = new_last_linear

        model.input_space = settings['input_space']
        model.input_size = settings['input_size']
        model.input_range = settings['input_range']
        model.mean = settings['mean']
        model.std = settings['std']
    else:
        model = InceptionV4(num_classes=num_classes)
    return model

def demo():
    from torchstat import stat

    net = inceptionv4(pretrained=None)
    y = net(torch.randn(1, 3, 299, 299))
    print(y.size())
    stat(net, (3, 299, 299))

# demo()


'''
TEST
Run this code with:
```
cd $HOME/pretrained-models.pytorch
python -m pretrainedmodels.inceptionv4
```
'''
# if __name__ == '__main__':
#     model_conv = inceptionv4(num_classes=1000, pretrained='imagenet')
#     ## Lets freeze the first few layers. This is done in two stages
#     # Stage-1 Freezing all the layers
#     for i, param in model_conv.named_parameters():
#         param.requires_grad = False
#
#     # Since imagenet as 1000 classes , We need to change our last layer according to the number of classes we have,
#     num_ftrs = model_conv.last_linear.in_features
#     model_conv.last_linear = nn.Linear(num_ftrs, 5)
#
#     # Stage-2 , Freeze all the layers till "Conv2d_4a_3*3"
#     ct = []
#     for name, child in model_conv.features.named_children():
#         if "4" in ct:
#             for params in child.parameters():
#                 params.requires_grad = True
#         ct.append(name)
#     # To view which layers are freeze and which layers are not freezed:
#     for name, child in model_conv.features.named_children():
#
#         for name, params in child.named_parameters():
#             print(name, params.requires_grad)
#     #
#     assert inceptionv4(num_classes=10, pretrained=None)
#     print('success')
#     assert inceptionv4(num_classes=1000, pretrained='imagenet')
#     print('success')
#     assert inceptionv4(num_classes=1001, pretrained='imagenet+background')
#     print('success')
#
#     # fail
#     assert inceptionv4(num_classes=1001, pretrained='imagenet')