Andrea Cossu
Andrea Cossu
1. Exactly, the `nc_benchmark` needs to be explicitly said the kind of targets you want. If you do not specify anything, it will use the entire set of targets. With...
Sure thing :smile:
@digantamisra98 can you work on this?
Hi @VerwimpEli , feel free to open a PR about this :smile:
Hi @tachyonicClock ! In this [repository](https://github.com/ContinualAI/reproducible-continual-learning) we are reproducing results from the continual learning literature with avalanche. We already worked on Synaptic Intelligence and it seems we are able to...
@digantamisra98 any news on this?
Do you mean something like a running average and standard deviation of the input patterns features?
Thanks @EdenBelouadah ! Just a couple of comments, then I'll leave to @AntonioCarta the final check: 1. It would be better to let the user specify how to access the...
Hi @ashok-arjun , we do not currently have an implementation for the task confusion matrix, even if it is definitely possible to implement it. We will try our best to...
I agree, this is something long overdue.