Feat/power stabilization curve [WIP]

open-spaced-repetition / srs-benchmark

A benchmark for spaced repetition schedulers/algorithms

62 stars 9 forks source link

Closed L-M-Sherlock closed 3 months ago

L-M-Sherlock commented 5 months ago

Weighted average by reviews:

Algorithm	Log Loss	RMSE (bins)	Parameters
FSRS-4.5	0.3252±0.1514	0.0533±0.0334	17
FSRS-4.5 + power	0.3250±0.1516	0.0532±0.0334	17

Weighted average by log(reviews):

Algorithm	Log Loss	RMSE (bins)	Parameters
FSRS-4.5	0.3485±0.1701	0.0733±0.0474	17
FSRS-4.5 + power	0.3483±0.1706	0.0729±0.0476	17

improved ~0.2% and ~0.5%.

Expertium commented 4 months ago

Do you think it's worth it? This is such a minor improvement that, personally, I wouldn't bother.

L-M-Sherlock commented 4 months ago

Absolutely not. It's just used to record the experiment.