Python torchvision.models.ResNet() Examples
The following are 30
code examples of torchvision.models.ResNet().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
torchvision.models
, or try the search function
.
Example #1
Source File: resnet.py From tracking_wo_bnw with GNU General Public License v3.0 | 6 votes |
def __init__(self, block, layers, output_dim): super(ResNet, self).__init__(block, layers) self.name = "ResNet" self.avgpool = nn.AvgPool2d((8,4), stride=1) self.fc = nn.Linear(512 * block.expansion, 1024) self.bn_fc = nn.BatchNorm1d(1024) self.relu_fc = nn.ReLU(inplace=True) self.fc_out = nn.Linear(1024, output_dim) for m in self.modules(): if isinstance(m, nn.BatchNorm1d): m.weight.data.fill_(1) m.bias.data.zero_() self.fc_compare = nn.Linear(output_dim, 1)
Example #2
Source File: resnet_encoder.py From DF-VO with MIT License | 6 votes |
def resnet_multiimage_input(num_layers, pretrained=False, num_input_images=1): """Constructs a ResNet model. Args: num_layers (int): Number of resnet layers. Must be 18 or 50 pretrained (bool): If True, returns a model pre-trained on ImageNet num_input_images (int): Number of frames stacked as input """ assert num_layers in [18, 50], "Can only run with 18 or 50 layer resnet" blocks = {18: [2, 2, 2, 2], 50: [3, 4, 6, 3]}[num_layers] block_type = {18: models.resnet.BasicBlock, 50: models.resnet.Bottleneck}[num_layers] model = ResNetMultiImageInput(block_type, blocks, num_input_images=num_input_images) if pretrained: loaded = model_zoo.load_url(models.resnet.model_urls['resnet{}'.format(num_layers)]) loaded['conv1.weight'] = torch.cat( [loaded['conv1.weight']] * num_input_images, 1) / num_input_images model.load_state_dict(loaded) return model
Example #3
Source File: resnet_encoder.py From SC-SfMLearner-Release with GNU General Public License v3.0 | 6 votes |
def resnet_multiimage_input(num_layers, pretrained=False, num_input_images=1): """Constructs a ResNet model. Args: num_layers (int): Number of resnet layers. Must be 18 or 50 pretrained (bool): If True, returns a model pre-trained on ImageNet num_input_images (int): Number of frames stacked as input """ assert num_layers in [18, 50], "Can only run with 18 or 50 layer resnet" blocks = {18: [2, 2, 2, 2], 50: [3, 4, 6, 3]}[num_layers] block_type = {18: models.resnet.BasicBlock, 50: models.resnet.Bottleneck}[num_layers] model = ResNetMultiImageInput(block_type, blocks, num_input_images=num_input_images) if pretrained: loaded = model_zoo.load_url(models.resnet.model_urls['resnet{}'.format(num_layers)]) loaded['conv1.weight'] = torch.cat( [loaded['conv1.weight']] * num_input_images, 1) / num_input_images model.load_state_dict(loaded) return model
Example #4
Source File: resnet_encoder.py From packnet-sfm with MIT License | 6 votes |
def resnet_multiimage_input(num_layers, pretrained=False, num_input_images=1): """Constructs a ResNet model. Args: num_layers (int): Number of resnet layers. Must be 18 or 50 pretrained (bool): If True, returns a model pre-trained on ImageNet num_input_images (int): Number of frames stacked as input """ assert num_layers in [18, 50], "Can only run with 18 or 50 layer resnet" blocks = {18: [2, 2, 2, 2], 50: [3, 4, 6, 3]}[num_layers] block_type = {18: models.resnet.BasicBlock, 50: models.resnet.Bottleneck}[num_layers] model = ResNetMultiImageInput(block_type, blocks, num_input_images=num_input_images) if pretrained: loaded = model_zoo.load_url(models.resnet.model_urls['resnet{}'.format(num_layers)]) loaded['conv1.weight'] = torch.cat( [loaded['conv1.weight']] * num_input_images, 1) / num_input_images model.load_state_dict(loaded) return model
Example #5
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet34(num_classes=1000): """Constructs a ResNet-34 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBasicBlock, [3, 4, 6, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #6
Source File: resnet_earlyexit.py From dnn-quant-ocs with Apache License 2.0 | 5 votes |
def resnet50_earlyexit(**kwargs): """Constructs a ResNet-50 model. """ model = ResNetEarlyExit(Bottleneck, [3, 4, 6, 3], **kwargs) return model
Example #7
Source File: resnet_earlyexit.py From dnn-quant-ocs with Apache License 2.0 | 5 votes |
def resnet101_earlyexit(**kwargs): """Constructs a ResNet-101 model. """ model = ResNetEarlyExit(Bottleneck, [3, 4, 23, 3], **kwargs) return model
Example #8
Source File: resnet_earlyexit.py From dnn-quant-ocs with Apache License 2.0 | 5 votes |
def resnet152_earlyexit(**kwargs): """Constructs a ResNet-152 model. """ model = ResNetEarlyExit(Bottleneck, [3, 8, 36, 3], **kwargs) return model
Example #9
Source File: resnet.py From tracking_wo_bnw with GNU General Public License v3.0 | 5 votes |
def resnet50(pretrained=False, **kwargs): """Constructs a ResNet-50 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(Bottleneck, [3, 4, 6, 3], **kwargs) if pretrained: model.load_pretrained_dict(model_zoo.load_url(model_urls['resnet50'])) return model
Example #10
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet18(num_classes=1000): """Constructs a ResNet-18 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBasicBlock, [2, 2, 2, 2], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #11
Source File: resnet_earlyexit.py From dnn-quant-ocs with Apache License 2.0 | 5 votes |
def resnet34_earlyexit(**kwargs): """Constructs a ResNet-34 model. """ model = ResNetEarlyExit(BasicBlock, [3, 4, 6, 3], **kwargs) return model
Example #12
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet50(num_classes=1000, pretrained=False): """Constructs a ResNet-50 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBottleneck, [3, 4, 6, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) if pretrained: model.load_state_dict(load_state_dict_from_url( "https://github.com/moskomule/senet.pytorch/releases/download/archive/seresnet50-60a8950a85b2b.pkl")) return model
Example #13
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet101(num_classes=1000): """Constructs a ResNet-101 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBottleneck, [3, 4, 23, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #14
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet152(num_classes=1000): """Constructs a ResNet-152 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBottleneck, [3, 8, 36, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #15
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet32(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEResNet(CifarSEBasicBlock, 5, **kwargs) return model
Example #16
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_resnet56(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEResNet(CifarSEBasicBlock, 9, **kwargs) return model
Example #17
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_preactresnet20(**kwargs): """Constructs a ResNet-18 model. """ model = CifarSEPreActResNet(CifarSEBasicBlock, 3, **kwargs) return model
Example #18
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_preactresnet32(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEPreActResNet(CifarSEBasicBlock, 5, **kwargs) return model
Example #19
Source File: se_resnet.py From netharn with Apache License 2.0 | 5 votes |
def se_preactresnet56(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEPreActResNet(CifarSEBasicBlock, 9, **kwargs) return model
Example #20
Source File: resnet_earlyexit.py From dnn-quant-ocs with Apache License 2.0 | 5 votes |
def resnet18_earlyexit(**kwargs): """Constructs a ResNet-18 model. """ model = ResNetEarlyExit(BasicBlock, [2, 2, 2, 2], **kwargs) return model
Example #21
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_resnet34(num_classes=1_000): """Constructs a ResNet-34 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBasicBlock, [3, 4, 6, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #22
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_resnet101(num_classes=1_000): """Constructs a ResNet-101 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBottleneck, [3, 4, 23, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #23
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_resnet152(num_classes=1_000): """Constructs a ResNet-152 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBottleneck, [3, 8, 36, 3], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model
Example #24
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_resnet20(**kwargs): """Constructs a ResNet-18 model. """ model = CifarSEResNet(CifarSEBasicBlock, 3, **kwargs) return model
Example #25
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_resnet56(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEResNet(CifarSEBasicBlock, 9, **kwargs) return model
Example #26
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_preactresnet20(**kwargs): """Constructs a ResNet-18 model. """ model = CifarSEPreActResNet(CifarSEBasicBlock, 3, **kwargs) return model
Example #27
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_preactresnet32(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEPreActResNet(CifarSEBasicBlock, 5, **kwargs) return model
Example #28
Source File: se_resnet.py From senet.pytorch with MIT License | 5 votes |
def se_preactresnet56(**kwargs): """Constructs a ResNet-34 model. """ model = CifarSEPreActResNet(CifarSEBasicBlock, 9, **kwargs) return model
Example #29
Source File: se_nets.py From DualResidualNetworks with MIT License | 5 votes |
def conv3x3(in_planes, out_planes, stride=1): return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, padding=1, bias=False) # SE-ResNet Module
Example #30
Source File: senet.py From pytorch2keras with MIT License | 5 votes |
def se_resnet18(num_classes): """Constructs a ResNet-18 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet """ model = ResNet(SEBasicBlock, [2, 2, 2, 2], num_classes=num_classes) model.avgpool = nn.AdaptiveAvgPool2d(1) return model