The basic case is really fast and needs no parallelism, but large-dimensional and conditional cases could benefit from parallelization. Consider extending the parallelism heuristic with a "performance score" tailored for each of these cases. Related to #22.
Closing this by making the estimation fast enough even without parallelism :smile: As entropy is a secondary feature, the feature is unnecessary. It should be quite easy to implement later if necessary.
The basic case is really fast and needs no parallelism, but large-dimensional and conditional cases could benefit from parallelization. Consider extending the parallelism heuristic with a "performance score" tailored for each of these cases. Related to #22.