Sha-Lab/FEAT

Issues

• cannot reproduce the performance of FEAT

  #82 opened a year ago by Duuhiy
  0

• About the accuracy of Protonet.

  #81 opened 2 years ago by Glutaredoxin
  4

• question about pre training　and feat model

  #80 opened 2 years ago by dota2mhxy
  2

• fewshot model

  #79 opened 2 years ago by Catofblue
  2

• Can you upload data/tieredimagenet/split please?

  #78 opened 2 years ago by CandiceHan123
  1

• Accuracy of pre-trained model for Tiered Imagenet Dataset

  #77 opened 2 years ago by mayug
  1

• Pretraining hyperparameters for CUB dataset

  #76 opened 2 years ago by mayug
  1

• the pre-training hyperparameters for ResNet12 on CUB

  #74 opened 2 years ago by DongYY127
  1

• Some question about CTM

  #75 opened 2 years ago by WoshiBoluo
  1

• hyperparameters that you used for ResNet12 pre-training on CUB

  #69 opened 2 years ago by gyr0104
  0

• DDP available?

  #72 opened 2 years ago by brando90
  1

• Question about proto

  #71 opened 2 years ago by LoveMiki
  1

• Download the pretrained weights

  #70 opened 2 years ago by OverVan
  1

• 有关transductive FSL 的设置

  #68 opened 2 years ago by DongYY127
  1

• 关于位置编码positional encoding

  #65 opened 2 years ago by YiX98
  1

• We can reproduce all numbers except for FEAT

  #73 opened 3 years ago by hieulem
  1

• The ProtoNet achieves the best Acc just after 1 epoch based on the given pre-trained ResNet12 weights.

  #45 opened 4 years ago by JimZAI
  2

• Key and value intuition of transformer in FEAT

  #66 opened 3 years ago by kumam92
  3

• what is the difference between the model and para_model in prepare_model?

  #42 opened 4 years ago by seyeeet
  2

• Question about the cosine distance.

  #62 opened 3 years ago by z1358
  3

• Question about lr and gamma.

  #67 opened 3 years ago by zhanyuanyang
  4

• Confusion about the contrastive learning.

  #63 opened 3 years ago by z1358
  1

• 有关Pre-Avg和Post_Avg

  #61 opened 3 years ago by sherrijj
  1

• reason behind good performance of Convnet over ResNet for noisy image

  #64 opened 3 years ago by kumam92
  1

• 请问关于提高准确率的建议

  #57 opened 3 years ago by YiX98
  3

• Pretraining hyperparameters for Convnet and Mini Imagenet

  #53 opened 3 years ago by mayug
  3

• Curious about classification results without pre-training

  #60 opened 3 years ago by LoveMiki
  4

• dataset label in batches are not the actual labels

  #59 opened 3 years ago by kumam92
  5

• Confusion matrix plot for testing episodes

  #58 opened 3 years ago by kumarmanas
  0

• 用自己的数据集训练FEAT报错

  #56 opened 3 years ago by YiX98
  0

• Hyperparameter training command for protonet with WRN backbone on miniImagenet dataset

  #54 opened 3 years ago by techNN
  2

• Could you give the hyperparameters for the CUB ?

  #55 opened 3 years ago by xue19890510
  1

• About the 5-shot results on miniimagenet

  #50 opened 3 years ago by zhanyuanyang
  4

• About the results of MAML on mini-imageNet with ConvNet

  #52 opened 3 years ago by JimZAI
  0

• 关于feat.py模型内部关于输入数据shape的问题

  #51 opened 3 years ago by shuowang666
  2

• 请问可否提供WRN作backbone时fine-tune阶段训练的相关参数

  #49 opened 3 years ago by SnowyJune973
  1

• 预训练的结果没有达到您公布模型的准确率

  #48 opened 3 years ago by YuanMLer
  1

• pretraining hyper-parameters for tiered imagenet

  #47 opened 3 years ago by Tsingularity
  1

• hyper-parameters to reproduce protonet results on resnet-12?

  #43 opened 3 years ago by Tsingularity
  5

• Can anyone reproduce the results for tiered-imagenet?

  #37 opened 3 years ago by smallbox120
  3

• About the results you reported in the paper

  #33 opened 3 years ago by molybdenumyz
  5

• label problem for meta-training stage

  #46 opened 3 years ago by freeman-1995
  0

• no avg_pooling in resnet-12?

  #44 opened 4 years ago by Tsingularity
  2

• Question the weight of pre-training

  #39 opened 4 years ago by blue-blue272
  1

• understanding the basics of dataloader and procedure

  #40 opened 4 years ago by seyeeet
  4

• should I use the cache?

  #41 opened 4 years ago by seyeeet
  1

• a question about resnet-18/12 kernel size

  #34 opened 4 years ago by Sword-keeper
  1

• some confusion about how to get "aux_center" in feat.py

  #35 opened 4 years ago by tian0810
  1

• a question about trainning error

  #36 opened 4 years ago by wongsihan
  2

• how can I test the model using pretrained models?

  #38 opened 4 years ago by seyeeet
  0