ContinualAI / ContinualAI/avalanche
Support initialization of strategy plugins from precomputed state (e.g. Fisher Importances in EWC, Teacher in LwF)
Open
Feature - Low Priority
Training
- Dominant language
- Python
- Stars
- 2.1k
- Forks
- 321
- PR merge metrics
- No merged PRs in 30d
Description
Hi There.
If for example I have a pretrained model and I want to apply EWC plugin, and to begin training with its regulization I can't as its coefficients are not computed yet.
I think that the feature of pretrained model is a must to have a fair baseline and to save computation.
Contributor guide
Assessment
This issue has not been assessed yet.