pytorch-maml

pytorch-maml copied to clipboard

can you provide the instructions to prepare the datasets?

3

xwhan

Thanks for your nice implementation. But I meet this mistake when I sh train-omniglot-20way-1shot.sh ,is there any dimension dismatch? I don't know how to fix it. It's much appreciated if...

PanPanZou

About the Model parameters updating in OmniglotNet Class

1

Thanks for your good implementation of MAML, however, I think that maybe use state_dict() and load_stat_dict() is much faster than modifying the weights (in omniglot_net.py 43), can I first deepcopy...

zhaoyu-li

Net_helper file missing?

jiwoongim

batch_cutoff not working for split = 'train'

1

Hi Kate, Under split='train' in inner loop, the data loader always output the same data no matter under what batch_cutoff. I know that this is the setting in the MAML...

pppplin

get_data_loader function gets same train samples

3

Hi kelly, I read the code and find that the get_data_loader function get the same batch samples every iteration, is that you want? (sorry, I'm not familiar with the maml...

whcjb

Parallelize inner loop updates with multiple processes.

ajabri