mrleo1nidandClaude Opus 5.5 58bd14ef38 feat(recommender): LightGBM lambdarank ranker with impression features (M6) ...

The wave logs each served candidate's feature vector to rec_impression_features.
Nightly training joins them with impression outcomes, validates on the last week
and activates the model only when it beats the v1 order and the active model.
The service loads the active ranker and uses its order in place of the v1 score.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-20 22:41:26 +03:00

..

__init__.py

feat(recommender): nightly ALS with offline metrics and CF candidates in the wave (M6)

2026-10-20 17:54:25 +03:00

als.py

feat(recommender): nightly ALS with offline metrics and CF candidates in the wave (M6)

2026-10-20 17:54:25 +03:00

bandit.py

feat(recommender): Thompson-sampling bandit over candidate sources (M6)

2026-10-20 22:31:24 +03:00

metrics.py

feat(recommender): nightly ALS with offline metrics and CF candidates in the wave (M6)

2026-10-20 17:54:25 +03:00

pipeline.py

feat(recommender): LightGBM lambdarank ranker with impression features (M6)

2026-10-20 22:41:26 +03:00

ranker.py

feat(recommender): LightGBM lambdarank ranker with impression features (M6)

2026-10-20 22:41:26 +03:00

signals.py

feat(recommender): nightly ALS with offline metrics and CF candidates in the wave (M6)

2026-10-20 17:54:25 +03:00