MAML-Pytorch

MAML-Pytorch copied to clipboard

accuracy format

1

Hello, First of all, I would like to thank you for your work. I have a question concerning accuracy format. I do not understand why there is more than one...

eghouti

Only use Conv and Linear

1

Hi, has anyone tried only to use Conv, and Linear function without BN? I commented BN in the config, and the result gives me random guess ... For 5-ways, the...

CongWeilin

for step, (x_spt, y_spt, x_qry, y_qry) in enumerate(db):

1

y_spt and y_qry are not the true labels of corrsponding images, are they just the indexes?

Peterdingpeng

bug in 2nd order?

4

I see in your code, just using `self.net(x_spt[i], fast_weights, bn_training=True)` however the torch.autograd.grad() method contain the following parameter: > create_graph (bool, optional) – If True, graph of the derivative will...

Interesting6

RuntimeError: Expected object of scalar type Long but got scalar type Int for argument #2 'target'

1

Traceback (most recent call last): File "C:/Users/Dingd/Documents/GitHub/MAML-Pytorch-master/omniglot_train.py", line 95, in main(args) File "C:/Users/Dingd/Documents/GitHub/MAML-Pytorch-master/omniglot_train.py", line 55, in main accs = maml(x_spt, y_spt, x_qry, y_qry) File "C:\Users\Dingd\AppData\Local\Programs\Python\Python36\lib\site-packages\torch\nn\modules\module.py", line 489, in __call__ result...

Peterdingpeng

In the code, running_mean/running_var requires_grad are set False. And in the forward code of Learner, training argument in F.batch_norm is set to True. So why?

arxrean

Conv-ReLU-BN issue

1

I just found this code use conv-relu-bn. However it should be conv-bn-relu. Could you please fix it.

mileyan

why not set bn_training=False when test net on x_qry in finetunning phase

2

in finetunning phase why you set bn_training=True when test net on x_qry ? line 178 、188 and 204 in meta,py

boy-be-ambitious

1. About validation accuracy of MiniImageNet in MAML original code, the evaluation is performed on the validation set instead of the test set, but in this code, evaluation is performed...

HongduanTian

OmniglotNShot

1

I used omniglot to train this model, but I found that train_data and test_data used in fine-tune are same classes. Such as 5-way, train_data are [0,1, 2, 3 ,4] and...

li829