WHQ1111

1. the output of function xent in utils.py is not a scalar but a vector? Is that right? 2. why divide it by FLAGS.update_batch_size?

what does oracle mean?

1

If oracle means ground-truth, then why we train it?

If oracle means ground-truth, then why we train it?

Why the second derivative is the gradient of the pooling operation.

1

When I read 'special_grads.py', I wondering why the second derivative is the gradient of the pooling operation.

when it runs in linux system, I get this error? What's this mean?

where is m.infer

1

In train.py,there is y_hat=m.infer(xs,ys),so where is m.infer

Some questions about the code.

3

As we can see, the line 280 and the line 283 of the file 'methods/backbone.py' mean that we use the Feature Wise Transformation module in MAML not the metric-based models....

Can‘t solve QP problem when dimension is too high.

4

When I try this: > G = torch.eye(2000).expand(8,2000,2000).cuda() > e = torch.randint(0,1,[8,2000]).cuda() > C = torch.eye(2000).expand(8,2000,2000).cuda() > h = torch.randint(0,1,[8,2000]).cuda() > A = torch.randn(8,100,2000).cuda() > b = torch.zeros(8,100).cuda() > qp_sol...

Can it use ResNet12 as backbone?

1

I try to change the backbone from Conv64-4 to ResNet12, but it is harmful for the accuracy. Is this model not suitable for large networks like ResNet or there is...

How to implement Concentrator, Projector and Reshaper when use ResNet18 as backbone

2

As I can see in the code, when I use ResNet18 as backbone, the Concentrator, Projector and Reshaper are same which is ``` self.main_component = nn.Sequential(self._make_layer(Bottleneck, out_size*2, 3, stride=1), self._make_layer(Bottleneck,...