ContinualAI / ContinualAI/avalanche
Task confusion matrix
Open
Evaluation
- Dominant language
- Python
- Stars
- 2.1k
- Forks
- 321
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
Similar to the class confusion matrix available, is there a **task confusion matrix also available**? This would be useful to actually assess BWT and other metrics, since every task's accuracy after every other task will be in the same table.
Thanks!
Contributor guide
Assessment
This issue has not been assessed yet.