yaoyao-liu/meta-transfer-learning

Issues

• Formula（4） for updating SS parameters

  #63 opened 10 months ago by iwei666
  1

• the implementation for the hard-meta batch

  #64 opened 4 months ago by zzltly
  1

• the code for choosing hard class-m？

  #56 opened 3 years ago by gyr0104
  1

• Where is the code for choosing hard class-m ?

  #48 opened 4 years ago by nobel861017
  11

• how to get the code about Hard Task Meta-Batch?

  #36 opened 4 years ago by Wiikk
  9

• how to fix feature encoder weights during SS process

  #52 opened 4 years ago by mengruwg
  2

• About the correspondence between the phases proposed in the paper and the phases in the code

  #62 opened 2 years ago by luhexin
  2

• ValueError: max() arg is an empty sequence

  #61 opened 2 years ago by Lcance
  2

• About SS operation.

  #60 opened 2 years ago by biwanqing
  2

• How do you update the base_leaner's parameters

  #59 opened 2 years ago by xugy16
  3

• “MAX_ITER=20000” during meta-test in Tensorflow code

  #58 opened 3 years ago by licj1
  3

• in conv2d_mtl.py, the mtl_weight has 4-dimensional and the weight has 3-dimensional, new_mtl_weight = self.mtl_weight.expand(self.weight.shape) has error,is there some problem?

  #57 opened 3 years ago by huifangchen1993
  1

• Where the fine-tuning in the code

  #55 opened 3 years ago by JianqunZhang
  4

• Very low performance during pre-training phase on FC100

  #54 opened 3 years ago by TristanGomez44
  1

• question about dataset

  #53 opened 4 years ago by changykang
  10

• About the computer operating system

  #51 opened 4 years ago by jiayi1999
  2

• FC-100 data loader

  #47 opened 4 years ago by YoungBigBird1985
  4

• Data Loader for CIFAR-FS

  #50 opened 4 years ago by YoungBigBird1985
  3

• In meta_train phase how to change batch_size

  #49 opened 4 years ago by lishuaijun1997
  6

• a question about meta-training strategy

  #45 opened 4 years ago by Sword-keeper
  15

• Performance Pytorch Version

  #46 opened 4 years ago by mikehuisman
  3

• I have some questions about test...

  #44 opened 4 years ago by Sword-keeper
  2

• meta batch size (number of episodes) always equal to 1

  #43 opened 4 years ago by DanqingZ
  4

• multi-gpu training & maml baseline

  #42 opened 4 years ago by DanqingZ
  13

• Questions about hard batch

  #27 opened 5 years ago by hortune
  3

• GPU can not working

  #29 opened 5 years ago by yangt1013
  10

• How to use a custom dataset with this model?

  #41 opened 4 years ago by Vaibhav-ML
  5

• Question Regarding Meta-train of PyTorch Implementation

  #40 opened 4 years ago by khaghanijavad
  2

• what was it to =: in formula (1) ？

  #38 opened 4 years ago by eeric
  1

• Question in meta phase: Why Generate the labels of support and query ?

  #39 opened 4 years ago by im-wll
  5

• Some question about tensorflow results

  #37 opened 4 years ago by Yitian-Li
  4

• some question about resnet-12

  #35 opened 4 years ago by Sword-keeper
  3

• validation acc in the pretrain phase in pytorch

  #34 opened 5 years ago by Sword-keeper
  10

• Results on backbone ResNet-12 without pre-training.

  #33 opened 5 years ago by LouieYang
  2

• resnet model not consistant between pytorch and tensorflow

  #32 opened 5 years ago by leon-liangwu
  4

• the choice of optimizer

  #31 opened 5 years ago by Sword-keeper
  3

• a question about task num in meta-training phase

  #30 opened 5 years ago by Sword-keeper
  4

• Question in pre-train phase: Why use meta-val set to select pre-train model?

  #28 opened 5 years ago by Errorfinder
  11

• Questions regarding reporting the test performance

  #26 opened 5 years ago by JKDomoguen
  1

• some questions about meta-training and meta-testing phase

  #25 opened 5 years ago by Sword-keeper
  4

• Question Regarding the Finetuning Ablation Study

  #24 opened 5 years ago by JKDomoguen
  2

• Original Bias is not added with shifting parameter?

  #23 opened 5 years ago by JKDomoguen
  2

• Resnet Clarification

  #22 opened 5 years ago by JKDomoguen
  2

• a question about 'trlog'

  #21 opened 5 years ago by Sword-keeper
  5

• some questions about performs

  #20 opened 5 years ago by ljjcoder
  1

• Question about the pytorch implementation

  #18 opened 5 years ago by bellos1203
  5

• Some questions about ResNet.

  #19 opened 5 years ago by vaynelau
  1

• Queaten about SS for pytorch

  #17 opened 5 years ago by xiaobeikankan
  1

• Questian about the choice of base-learner

  #16 opened 5 years ago by Sebastian-X
  4

• YaoShenNB

  #15 opened 5 years ago by xu282934741
  1