Can I use this function (temperature_scaling) for other classification network??

[edwardcho]

Hello Sir.

I tested your code (temperature scaling) in my code.
my code is based on inception_v3.

But I met some error.

# Code
    # This function probably should live outside of this class, but whatever
    #def set_temperature(self, valid_loader):
    def set_temperature(self, valid_loader):
        """
        Tune the tempearature of the model (using the validation set).
        We're going to set it to optimize NLL.
        valid_loader (DataLoader): validation set loader
        """
        self.cuda()
        nll_criterion = nn.CrossEntropyLoss().cuda()
        ece_criterion = _ECELoss().cuda()

        # First: collect all the logits and labels for the validation set
        logits_list = []
        labels_list = []
        with torch.no_grad():
            for input, label in valid_loader:
                input = input.cuda()
                logits = self.model(input)
                logits_list.append(logits)
                labels_list.append(label)
            logits = torch.cat(logits_list).cuda()
            labels = torch.cat(labels_list).cuda()

        # Calculate NLL and ECE before temperature scaling
        before_temperature_nll = nll_criterion(logits, labels).item()
        before_temperature_ece = ece_criterion(logits, labels).item()
        print('Before temperature - NLL: %.3f, ECE: %.3f' % (before_temperature_nll, before_temperature_ece))

        # Next: optimize the temperature w.r.t. NLL
        optimizer = optim.LBFGS([self.temperature], lr=0.01, max_iter=50)
        ...
# Code
'''
RuntimeError: Calculated padded input size per channel: (4 x 4). Kernel size: (5 x 5). Kernel size can't be greater than actual input size
'''

I checked error code in my code.
The part of error code is
logits = self.model(input)

So I think that this code is for DenseNet only. ??

Thanks,

[edwardcho]

Hello Sir,

I add error message...

ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 768, 1, 1])

This error was occurred in this part.

logits = self.model(input)

What can I do for solving this problem??

Thanks..

[shaktikshri]

hi there, this is an issue in your model. Try debugging your classification network without the set temperature function first.