ARML issues

Question about the performance of ARML

1

I tried to run the ARML training and inference code, and my result on plainmulti dataset seems not match with the result on paper. In my experiment, the acc of...

peterdarkdarkgogo

Cannot parse tensor from proto: dtype: DT_UINT8

1

Running `plainmulti_1shot.sh`, I ran into an error in the first python invocation: `python main.py --datasource=plainmulti --datadir=xxx --metatrain_iterations=50000 --meta_batch_size=4 --update_batch_size=1 --update_lr=0.01 --num_updates=5 --num_classes=5 --logdir=xxx --num_filters=32 --hidden_dim=128 --emb_loss_weight=0.01` [Possibly related to this...

jabowery

why reverse the order of decoder output?

2

Hi, @huaxiuyao , thanks for your implementation, I notice that when you generate the decoder output in the task_embedding.py, you reverse the order of the prototypes, what is the meaning...

d12306