SqueezeNet 1.0 and 1.1 models

torchvision/models/__init__.py

@@ -4,6 +4,7 @@
 -  `AlexNet`_
 -  `VGG`_
 -  `ResNet`_
+-  `SqueezeNet`_
 
 You can construct a model with random weights by calling its constructor:
 
@@ -12,6 +13,7 @@
     import torchvision.models as models
     resnet18 = models.resnet18()
     alexnet = models.alexnet()
+    squeezenet = models.squeezenet1_0()
 
 We provide pre-trained models for the ResNet variants and AlexNet, using the
 PyTorch :mod:`torch.utils.model_zoo`. These can  constructed by passing
@@ -26,8 +28,10 @@
 .. _AlexNet: https://arxiv.org/abs/1404.5997
 .. _VGG: https://arxiv.org/abs/1409.1556
 .. _ResNet: https://arxiv.org/abs/1512.03385
+.. _SqueezeNet: https://arxiv.org/abs/1602.07360
 """
 
 from .alexnet import *
 from .resnet import *
 from .vgg import *
+from .squeezenet import *

torchvision/models/squeezenet.py

@@ -0,0 +1,130 @@
+import math
+import torch
+import torch.nn as nn
+import torch.utils.model_zoo as model_zoo
+
+
+__all__ = ['SqueezeNet', 'squeezenet1_0', 'squeezenet1_1']
+
+
+model_urls = {
+    'squeezenet1_0': 'https://s3.amazonaws.com/pytorch/models/squeezenet1_0-a815701f.pth',
+    'squeezenet1_1': 'https://s3.amazonaws.com/pytorch/models/squeezenet1_1-f364aa15.pth',
+}
+
+
+class Fire(nn.Module):
+    def __init__(self, inplanes, squeeze_planes,
+            expand1x1_planes, expand3x3_planes):
+        super(Fire, self).__init__()
+        self.inplanes = inplanes
+        self.squeeze = nn.Conv2d(inplanes, squeeze_planes, kernel_size=1)
+        self.squeeze_activation = nn.ReLU(inplace=True)
+        self.expand1x1 = nn.Conv2d(squeeze_planes, expand1x1_planes,
+                                   kernel_size=1)
+        self.expand1x1_activation = nn.ReLU(inplace=True)
+        self.expand3x3 = nn.Conv2d(squeeze_planes, expand3x3_planes,
+                                   kernel_size=3, padding=1)
+        self.expand3x3_activation = nn.ReLU(inplace=True)
+
+    def forward(self, x):
+        x = self.squeeze_activation(self.squeeze(x))
+        return torch.cat([
+            self.expand1x1_activation(self.expand1x1(x)),
+            self.expand3x3_activation(self.expand3x3(x))
+        ], 1)
+
+
+class SqueezeNet(nn.Module):
+    def __init__(self, version=1.0, num_classes=1000):
+        super(SqueezeNet, self).__init__()
+        if version not in [1.0, 1.1]:
+            raise ValueError("Unsupported SqueezeNet version {version}:"
+                             "1.0 or 1.1 expected".format(version=version))
+        self.num_classes = num_classes
+        if version == 1.0:
+            self.features = nn.Sequential(
+                nn.Conv2d(3, 96, kernel_size=7, stride=2),
+                nn.ReLU(inplace=True),
+                nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True),
+                Fire(96, 16, 64, 64),
+                Fire(128, 16, 64, 64),
+                Fire(128, 32, 128, 128),
+                nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True),
+                Fire(256, 32, 128, 128),
+                Fire(256, 48, 192, 192),
+                Fire(384, 48, 192, 192),
+                Fire(384, 64, 256, 256),
+                nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True),
+                Fire(512, 64, 256, 256),
+            )
+        else:
+            self.features = nn.Sequential(
+                nn.Conv2d(3, 64, kernel_size=3, stride=2),
+                nn.ReLU(inplace=True),
+                nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True),
+                Fire(64, 16, 64, 64),
+                Fire(128, 16, 64, 64),
+                nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True),
+                Fire(128, 32, 128, 128),
+                Fire(256, 32, 128, 128),
+                nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True),
+                Fire(256, 48, 192, 192),
+                Fire(384, 48, 192, 192),
+                Fire(384, 64, 256, 256),
+                Fire(512, 64, 256, 256),
+            )
+        # Final convolution is initialized differently form the rest
+        final_conv = nn.Conv2d(512, num_classes, kernel_size=1)
+        self.classifier = nn.Sequential(
+            nn.Dropout(p=0.5),
+            final_conv,
+            nn.ReLU(inplace=True),
+            nn.AvgPool2d(13)
+        )
+
+        for m in self.modules():
+            if isinstance(m, nn.Conv2d):
+                gain = 2.0
+                if m is final_conv:
+                    m.weight.data.normal_(0, 0.01)
+                else:
+                    fan_in = m.kernel_size[0] * m.kernel_size[1] * m.in_channels
+                    u = math.sqrt(3.0 * gain / fan_in)
+                    m.weight.data.uniform_(-u, u)
+                if m.bias is not None:
+                    m.bias.data.zero_()
+
+    def forward(self, x):
+        x = self.features(x)
+        x = self.classifier(x)
+        return x.view(x.size(0), self.num_classes)
+
+
+def squeezenet1_0(pretrained=False):
+    r"""SqueezeNet model architecture from the `"SqueezeNet: AlexNet-level
+    accuracy with 50x fewer parameters and <0.5MB model size"
+    <https://arxiv.org/abs/1602.07360>`_ paper.
+
+    Args:
+        pretrained (bool): If True, returns a model pre-trained on ImageNet
+    """
+    model = SqueezeNet(version=1.0)
+    if pretrained:
+        model.load_state_dict(model_zoo.load_url(model_urls['squeezenet1_0']))
+    return model
+
+
+def squeezenet1_1(pretrained=False):
+    r"""SqueezeNet 1.1 model from the `official SqueezeNet repo
+    <https://github.com/DeepScale/SqueezeNet/tree/master/SqueezeNet_v1.1>`_.
+    SqueezeNet 1.1 has 2.4x less computation and slightly fewer parameters
+    than SqueezeNet 1.0, without sacrificing accuracy.
+
+    Args:
+        pretrained (bool): If True, returns a model pre-trained on ImageNet
+    """
+    model = SqueezeNet(version=1.1)
+    if pretrained:
+        model.load_state_dict(model_zoo.load_url(model_urls['squeezenet1_1']))
+    return model