Documentation

FsrsParameters
in package

FinalYes

FSRS-6 tuning parameters.

The 21 weights are the algorithm's fitted constants (w0..w20). The defaults are FSRS-6's published values — identical to DEFAULT_PARAMETERS in open-spaced-repetition/py-fsrs v6 — and are what every user gets until per-user optimisation exists (that needs accumulated review_log history, deliberately out of scope for phase 2a).

w20 is the learnable decay of the forgetting curve, new in FSRS-6.

Table of Contents

Constants

DEFAULT_DECAY  : mixed = 0.1542
FSRS-6 default decay (w20).
DEFAULT_WEIGHTS  : array<int, float> = [0.212, 1.2931, 2.3065, 8.2956, 6.4133, 0.8334,...
Published FSRS-6 defaults, w0..w20.
MAX_DIFFICULTY  : mixed = 10.0
MIN_DIFFICULTY  : mixed = 1.0
PARAMETER_COUNT  : mixed = 21
STABILITY_MIN  : mixed = 0.001
FSRS clamps stability to this floor to keep the power terms finite.

Properties

$desiredRetention  : float
$maximumInterval  : int
$weights  : array<int, float>

Methods

__construct()  : mixed
decay()  : float
Curve decay, i.e. -w20. Negative by construction.
factor()  : float
The constant that makes the forgetting curve pass through R = 0.9 at t = stability.
w()  : float
Weight w{$index}.

Constants

DEFAULT_DECAY

FSRS-6 default decay (w20).

public mixed DEFAULT_DECAY = 0.1542

DEFAULT_WEIGHTS

Published FSRS-6 defaults, w0..w20.

public array<int, float> DEFAULT_WEIGHTS = [0.212, 1.2931, 2.3065, 8.2956, 6.4133, 0.8334, 3.0194, 0.001, 1.8722, 0.1666, 0.796, 1.4835, 0.0614, 0.2629, 1.6483, 0.6014, 1.8729, 0.5425, 0.0912, 0.0658, self::DEFAULT_DECAY]

STABILITY_MIN

FSRS clamps stability to this floor to keep the power terms finite.

public mixed STABILITY_MIN = 0.001

Properties

$desiredRetention read-only

public float $desiredRetention = 0.9

Methods

__construct()

public __construct([array<int, float>|null $weights = null ][, float $desiredRetention = 0.9 ][, int $maximumInterval = 36500 ]) : mixed
Parameters
$weights : array<int, float>|null = null

21 FSRS weights; null = defaults.

$desiredRetention : float = 0.9

Target recall probability at the moment a term comes due.

$maximumInterval : int = 36500

Hard cap on scheduled days.

decay()

Curve decay, i.e. -w20. Negative by construction.

public decay() : float
Return values
float

factor()

The constant that makes the forgetting curve pass through R = 0.9 at t = stability.

public factor() : float
Return values
float

w()

Weight w{$index}.

public w(int $index) : float
Parameters
$index : int
Return values
float
On this page

Search results