wDAE_GNN_FewShot
wDAE_GNN_FewShot copied to clipboard
gidariss
Generating Classification Weights with GNN Denoising Autoencoders for Few-Shot Learning
I ran the code with dataloader num_workers=4, the training process seems to be lagging due to the dataloader because num_workers=4 causes the GPU to progress 4 batches at a time,...
Code for running the MiniImageNet-based experiments
Could you share some details on how to generate the miniImageNet pickle files, I want to test on my own datasets, many thanks.
It is a really nice job! We tried to use resnet as a backbone just like what you did in our own experiment. However, it didn't improve performance as we...
When I run lowshot_train_stage1.py using mini imagenet, I got the following error Traceback (most recent call last): File "scripts/lowshot_train_stage1.py", line 64, in epoch_size=data_train_opt['epoch_size'], # num of batches per epoch File...
Hi there, Thanks for the code releasing. I am wondering if you consider release the model of stage1 training, for example the Resnet10 or WRN-28-10 classification model trained with miniImagenet...
Hi, I have just downloaded your extracted features and found that they are in the json format, which is different from the h5 files you use in the code. Do...
i can find some setting in wide_resnet.py. opt = {} opt['depth'] = 28 opt['widen_Factor'] = 10 opt['dropRate'] = 0.0 opt['extra_block'] = False opt['pool'] = 'none'
