Documentation
¶
Overview ¶
Package scoring provides importance scoring and decay algorithms for memory retrieval ranking.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type DecayScorer ¶
type DecayScorer struct {
// HalfLife is the duration after which importance decays to 50%.
// Default: 7 days.
HalfLife time.Duration
// DecayWeight controls how much importance affects the final score.
// Range: 0-1. Default: 0.3.
DecayWeight float64
// AccessBoostFactor is the boost per access. Default: 0.01.
AccessBoostFactor float64
// MaxAccessBoost caps the total access boost. Default: 0.2.
MaxAccessBoost float64
}
DecayScorer combines similarity with time-based importance decay and access frequency. It balances relevance (similarity) with recency (decay) and usage (access count).
Final score = similarity * (1 - decayWeight) + decayedImportance * decayWeight + accessBoost
func NewDecayScorer ¶
func NewDecayScorer() *DecayScorer
NewDecayScorer creates a scorer with sensible defaults.
func (*DecayScorer) Score ¶
func (s *DecayScorer) Score(result memoryrails.SearchResult, now time.Time) float64
Score computes the final retrieval score.
Source Files
Click to show internal directories.
Click to hide internal directories.