Thank you for your high quality work,could you plase tell me how to train the baseline model based ResNet12 on Teried_Imagenet?
I follow the “$ python train.py --eval_all --unsupervised --batch_size 64 --augment 'AMDIM' --num_tasks 256 --max_epoch 200 --model_class ProtoNet --backbone_class Res12 --dataset TieredImageNet --way 5 --shot 1 --query 5 --eval_query 15 --temperature 1 --temperature2 1 --lr 0.03 --lr_scheduler cosine --gpu 0 --eval_interval 2 --similarity sns” setting, but only get 53.75 % on the performance of the last epoch.
Thank you for your high quality work,could you plase tell me how to train the baseline model based ResNet12 on Teried_Imagenet? I follow the “$ python train.py --eval_all --unsupervised --batch_size 64 --augment 'AMDIM' --num_tasks 256 --max_epoch 200 --model_class ProtoNet --backbone_class Res12 --dataset TieredImageNet --way 5 --shot 1 --query 5 --eval_query 15 --temperature 1 --temperature2 1 --lr 0.03 --lr_scheduler cosine --gpu 0 --eval_interval 2 --similarity sns” setting, but only get 53.75 % on the performance of the last epoch.