Support new Avalanche version
Closed this issue · 3 comments
This new version (ContinualAI/avalanche@4380ebc) includes a refactoring of the RLScenario and RLExperience using the newly defined benchmark stack components.
In order to support it, we must also modify our RLStrategy
to inherit from the "BaseTemplate" (https://github.com/ContinualAI/avalanche/tree/master/avalanche/training/templates), this is the newest "hierarchy in town" that just joined the Avalanche training module.
This integration should represent a significant step toward an eventual merge of Avalanche RL into Avalanche.
Other significant changes in avl since version 0.0.1 include Plugins (TODO: BasePlugin->RLPlugin) and loggers
also we should be installing it from the master branch, the beta version is the one you get when you run pip
A solution to this is currently being implemented here https://github.com/ContinualAI/avalanche-rl/tree/new-avalanche-version.
It involves a workaround to keep the old avalanche BaseStrategy
around before fully switching to the Templates model.