Issues
- 1
- 0
Why is the preprocessed Stanford cars dataset only consisting of training set and testing set without projecting set, nor does it have a train_crop?
#13 opened by xymyys - 0
connect to swin transformer
#12 opened by robchenchen - 0
Model inference question
#11 opened by robchenchen - 1
is finetuning avaliable?
#10 opened by robchenchen - 1
Accuracy of Prototree ensemble 5
#9 opened by Millielele - 1
Model can not be loaded
#8 opened by lx913 - 1
- 2
Accuracy of non-iNat Networks
#6 opened by EoinKenny - 1
Projection of prototypes
#5 opened by EoinKenny - 4
Acc lower than paper claims
#4 opened by huygens12 - 2
My final acc is only 76%
#3 opened by mobulan - 3
Training time
#2 opened by pedrosbmartins - 1
Upsample issue multiple bounding boxes
#1 opened by GuidoHeijden