GMvandeVen/continual-learning
PyTorch implementation of various methods for continual learning (XdG, EWC, SI, LwF, FROMP, DGR, BI-R, ER, A-GEM, iCaRL, Generative Classifier) in three different scenarios.
Jupyter NotebookMIT
Issues
- 0
Results for None ("lower target")
#31 opened by youtao1 - 0
About printing results of experimental output
#30 opened by youtao1 - 1
- 9
0 accuracy values for task-free setting
#28 opened by hiteshvaidya - 4
Final results of class incremental learning
#3 opened by Harry-Up - 1
Single head or multihead task incremental
#27 opened by COD1995 - 2
Task-IL evaluation
#26 opened by Young-Jo-Choi - 3
- 1
about kafc fisher infromation matrix
#22 opened by fmaaf - 2
How to create Resnet34
#23 opened by Peeaak - 9
Reproducing BI+SI method
#21 opened by valeriya-khan - 2
Link error
#20 opened by MingkangXiong - 3
Suspicious Precision
#19 opened by tarasrumezhak - 1
- 1
why batch_size has to be 1 when update fisher?
#16 opened by yuchenlin - 2
Lower/Upper Bound Experiments
#17 opened by kaustubholpadkar - 7
Performance
#10 opened by Johswald - 2
Wrong dataset?
#15 opened by GuillaumeLam - 4
Grad in SI
#14 opened by ssydasheng - 0
Just a request
#13 opened by jaytimbadia - 1
Datasets more complicated than MNIST
#12 opened by xu-ji - 3
Empirical Fisher Estimation
#11 opened by tangbinh - 2
Question about Online EWC
#9 opened by DavideA - 2
- 4
Why uses the prediction of all active classes?
#7 opened by suanrong - 6
average EWC pickle
#6 opened by Johswald - 2
Knowledge Distillation Loss
#5 opened by Johswald - 2
permutedMNIST accs
#4 opened by Johswald - 2
Thank you for your wonderful package!
#1 opened by JosephKJ - 2