Expt/trainable grade

[L-M-Sherlock]

No description provided.

[Expertium]

I advise you to change the range of w[18], since the grades now go from 1 to 10 rather than from 1 to 4.
Also, rating - 3 and rating - 1 should be changed accordingly.

[L-M-Sherlock]

Also, rating - 3 and rating - 1 should be changed accordingly.

But how? What's the new formula?

[Expertium]

Just X_rating - self.w[21] and X_rating - self.w[19]

[L-M-Sherlock]

Here is the result:

Model: FSRS-5-dev
Total number of users: 9999
Total number of reviews: 349923850
Weighted average by reviews:
FSRS-5-dev LogLoss (mean±std): 0.3269±0.1522
FSRS-5-dev RMSE(bins) (mean±std): 0.0508±0.0330
FSRS-5-dev AUC (mean±std): 0.7027±0.0785

Weighted average by log(reviews):
FSRS-5-dev LogLoss (mean±std): 0.3530±0.1694
FSRS-5-dev RMSE(bins) (mean±std): 0.0707±0.0464
FSRS-5-dev AUC (mean±std): 0.7004±0.0889

Weighted average by users:
FSRS-5-dev LogLoss (mean±std): 0.3564±0.1721
FSRS-5-dev RMSE(bins) (mean±std): 0.0737±0.0484
FSRS-5-dev AUC (mean±std): 0.6994±0.0908

parameters: [0.4284, 1.1777, 3.2057, 15.8182, 7.1567, 0.5324, 1.8111, 0.0099, 1.5152, 0.1402, 1.0026, 1.9177, 0.0959, 0.2996, 2.3541, 0.2315, 3.0127, 0.4596, 0.6102, 1.0436, 2.0, 3.039, 4.1698]
Model: FSRS-5
Total number of users: 9999
Total number of reviews: 349923850
Weighted average by reviews:
FSRS-5 LogLoss (mean±std): 0.3276±0.1526
FSRS-5 RMSE(bins) (mean±std): 0.0518±0.0333
FSRS-5 AUC (mean±std): 0.7010±0.0786

Weighted average by log(reviews):
FSRS-5 LogLoss (mean±std): 0.3534±0.1696
FSRS-5 RMSE(bins) (mean±std): 0.0713±0.0462
FSRS-5 AUC (mean±std): 0.6995±0.0887

Weighted average by users:
FSRS-5 LogLoss (mean±std): 0.3568±0.1721
FSRS-5 RMSE(bins) (mean±std): 0.0742±0.0479
FSRS-5 AUC (mean±std): 0.6986±0.0908

parameters: [0.4299, 1.162, 3.1897, 15.8179, 7.1441, 0.5397, 1.7835, 0.0104, 1.5175, 0.1351, 1.0064, 1.9183, 0.1007, 0.3016, 2.3446, 0.2315, 3.0117, 0.4463, 0.635]

[Expertium]

I see. Well, that's unfortunate.
Hope you'll benchmark FSRS v1 and v2, just out of curiosity