[AI] 간단한 모델 이용한 성능 항샹 테크닉 실습

간단한 모델

ResNet
AlexNet
vgg
SqueezeNet
DenseNet
Inception
모델을 실습해봄.

ResNet

        model_ft = models.resnet18(pretrained=use_pretrained)
        num_ftrs = model_ft.fc.in_features
        model_ft.fc = nn.Linear(num_ftrs, num_classes)
        input_size = 224

AlexNet

        model_ft = models.alexnet(pretrained=use_pretrained)
        num_ftrs = model_ft.classifier[6].in_features
        model_ft.classifier[6] = nn.Linear(num_ftrs, num_classes)
        input_size = 224

vgg

        model_ft = models.vgg11_bn(pretrained=use_pretrained)
        num_ftrs = model_ft.classifier[6].in_features
        model_ft.classifier[6] = nn.Linear(num_ftrs, num_classes)
        input_size = 224

SqueezeNet

        model_ft.classifier[1] = nn.Conv2d(
            512, num_classes, kernel_size=(1, 1), stride=(1, 1))
        model_ft.num_classes = num_classes
        input_size = 224

DenseNet

        model_ft = models.densenet121(pretrained=use_pretrained)
        num_ftrs = model_ft.classifier.in_features
        model_ft.classifier = nn.Linear(num_ftrs, num_classes)
        input_size = 224

Inception

        model_ft = models.inception_v3(pretrained=use_pretrained)
        # Handle the auxilary net
        num_ftrs = model_ft.AuxLogits.fc.in_features
        model_ft.AuxLogits.fc = nn.Linear(num_ftrs, num_classes)
        # Handle the primary net
        num_ftrs = model_ft.fc.in_features
        model_ft.fc = nn.Linear(num_ftrs, num_classes)
        input_size = 299