Skip to content
#

continual-learning

Here are 125 public repositories matching this topic...

avalanche
AlbinSou
AlbinSou commented Sep 15, 2021

For the moment, the nc_benchmark generator function allows for a nc_first_task option, which is good for pre-training in the class-incremental learning scenario. However, the same kind of option is not available if one wants to pretrain in the task-incremental scenario. It would be nice to have an option that could be used together with task_labels=True and allows for pretraining on multiple tasks

A collection of online continual learning paper implementations and tricks for computer vision in PyTorch, including our ASER(AAAI-21), SCR(CVPR21-W) and a survey under review.

  • Updated Aug 16, 2021
  • Python

Continual Hyperparameter Selection Framework. Compares 11 state-of-the-art Lifelong Learning methods and 4 baselines. Official Codebase of "A continual learning survey: Defying forgetting in classification tasks." in IEEE TPAMI.

  • Updated Jun 3, 2021
  • Python

Improve this page

Add a description, image, and links to the continual-learning topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the continual-learning topic, visit your repo's landing page and select "manage topics."

Learn more