avalanche
avalanche copied to clipboard
ContinualAI
Avalanche: an End-to-End Library for Continual Learning based on PyTorch.
Here are the differences between the last working environment and the new one that I tried to run: ``` 1c1 < name: avalanche-env --- > name: base 5d4 < -...
Here are the differences between the last working environment and the new one that I tried to run: ``` 1c1 < name: avalanche-env --- > name: base 5d4 < -...
Here are the differences between the last working environment and the new one that I tried to run: ``` 1c1 < name: avalanche-env --- > name: base 5d4 < -...
Here are the differences between the last working environment and the new one that I tried to run: ``` 1c1 < name: avalanche-env --- > name: base 5d4 < -...
*Still a work in progress* AvalancheDatasets are doing many different things together: - advanced indexing - management of class/task labels - management of transformations - efficient concat/subset operations needed for...
Hi, It is related to #https://github.com/ContinualAI/avalanche/issues/1116 I think there is still something wrong because the `dist_loss ` is zero. It seems that for the case where `isinstance(self.prev_model, MultiTaskModule)` is True,...
Hi, **Training** I initialize `net=MTSimpleCNN()` and train this multi-task model with multi-head classifier on two tasks (one task per experience). After training the model has two heads. I save the...
Hi, I think that in LwF plugin in the penalty computation the gradient doesn't flow correctly in some cases ``` with torch.no_grad(): if isinstance(self.prev_model, MultiTaskModule): # output from previous output...
### Discussed in https://github.com/ContinualAI/avalanche/discussions/1111 Originally posted by **SinHanYang** August 8, 2022 Hi, I tried to evaluate the model on all experiences after every epoch. I did experiment on `CIFAR10` dataset,...
I tried to implement the changes that would fix bugs described in Issue #1591 and also allow users to use Dynamic modules more easily outside of avalanche strategies. To do...
Metadata
Owner
Metadata
Avalanche: an End-to-End Library for Continual Learning based on PyTorch.