与二进制分类器实施InceptionV3的问题 - 与Pytorch进行转移学习

Question

我有一个问题，使Inception v3与Pytorch的二进制分类器一起作为功能提取器工作。我将启动的主要和辅助网更新为二进制类（如）

但是我遇到了一个错误

#Parameters for Inception V3
num_classes= 2 
model_ft = models.inception_v3(pretrained=True)
# set_parameter_requires_grad(model_ft, feature_extract)
#handle auxilliary net 
num_ftrs = model_ft.AuxLogits.fc.in_features
model_ft.AuxLogits.fc = nn.Linear(num_ftrs, num_classes)
#handle primary net
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs,num_classes)
# input_size = 299

#simulate data input 
x = torch.rand([64, 3, 299, 299])

#create model with inception backbone
backbone = model_ft
num_filters = backbone.fc.in_features
layers = list(backbone.children())[:-1]
feature_extractor = nn.Sequential(*layers)

# use the pretrained model to classify damage 2 classes
num_target_classes = 2
classifier = nn.Linear(num_filters, num_target_classes)

feature_extractor.eval()
with torch.no_grad():
    representations = feature_extractor(x).flatten(1)
x = classifier(representations)

错误

RuntimeError                              Traceback (most recent call last)
<ipython-input-54-c2be64b8a99e> in <module>()
     11 feature_extractor.eval()
     12 with torch.no_grad():
---> 13     representations = feature_extractor(x)
     14 x = classifier(representations)

9 frames
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/conv.py in _conv_forward(self, input, weight, bias)
    442                             _pair(0), self.dilation, self.groups)
    443         return F.conv2d(input, weight, bias, self.stride,
--> 444                         self.padding, self.dilation, self.groups)
    445 
    446     def forward(self, input: Tensor) -> Tensor:

RuntimeError: Expected 3D (unbatched) or 4D (batched) input to conv2d, but got input of size: [64, 2]

，但是我会在将类更新为2（当它是1000时）之前遇到 64，1000]。这种创建骨干和添加分类器的方法可用于Resnet，但在这里不适合。我认为这是因为辅助网络结构，但不确定如何更新它来处理双重输出？谢谢

Answer 1

继承 feature_extracture by 儿童 line layers = layers = List（Backbone.Children（））[： - 1] 将从 Backbone to feature_extressure ，而不是 forward> forward 函数中的操作。

让我们看一下下面的代码：

class Example(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.avg = torch.nn.AdaptiveAvgPool2d((1,1))
        self.linear = torch.nn.Linear(10, 1)
    def forward(self, x):
        out = self.avg(x)
        out = out.squeeze()
        out = self.linear(out)
        return out

x = torch.randn(5, 10, 12, 12)

model = Example()
y = model(x) # work well

new_model = torch.nn.Sequential(*list(model.children()))
y = new_model(x) # error

模块模型和 new_model 具有相同的块，但工作方式不相同。在 new_module 中，池中的输出尚未挤压，因此线性输入的形状违反了其假设，从而导致错误。

在您的情况下，最后两个注释是多余的，这就是为什么它返回错误的原因，您确实在line mode_ft.fc = nn.linear（num_ftrs，num_ftrs，num_ftrs， num_classes）。因此，将最后一个替换为下面的代码应正常工作：

with torch.no_grad():
    x = model_ft(x)