Model card

astris-retention-lightgbm-v1

LightGBM gradient-boosted retention classifier predicting P(retain >= 90 days) from the candidate-placement feature vector. Captures nonlinear feature interactions the linear voters miss.

At a glance

Kindretention xgboost
Version1.0.0
StatusActive — contributing to scores
Ensemble weight0.25
Trained atNot trained yet
Trained onn/a
ArtifactNo artifact (heuristic or pre-training)

Metrics

{
  "metric": "auc",
  "target": "retain_ge_90d",
  "validation": "synthetic_seed",
  "real_cohort": "pending",
  "value_approx": 0.81
}

Notes

AUC ~0.81 on internal synthetic validation; real-cohort benchmark pending. Voter in the retention ensemble — see api/ml/match_scorer.py.