mrleo1nidandClaude Opus 5.5
58bd14ef38
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>
125 lines
4.6 KiB
Python
125 lines
4.6 KiB
Python
"""Сервис «Моей волны»: собирает пачку из каталога, профиля, сессии и контекста."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import logging
|
|
from typing import Literal
|
|
|
|
import numpy as np
|
|
import psycopg
|
|
from pydantic import BaseModel, Field
|
|
|
|
from pipu_rec.train.ranker import RankerHolder
|
|
from pipu_rec.wave.contexts import Contexts, Rule
|
|
from pipu_rec.wave.engine import Context, Pick, build_batch, model_version, taste_centroids, update_session
|
|
from pipu_rec.wave.store import IndexHolder, SessionStore, load_profile, seed_tracks
|
|
|
|
MAX_SERVED_MEMORY = 300
|
|
|
|
log = logging.getLogger(__name__)
|
|
|
|
Outcome = Literal["complete", "skip", "skip_early", "like", "dislike"]
|
|
|
|
|
|
class Feedback(BaseModel):
|
|
track_id: str
|
|
outcome: Outcome
|
|
|
|
|
|
class Seed(BaseModel):
|
|
type: Literal["track", "album", "artist"]
|
|
id: str
|
|
|
|
|
|
class BatchRequest(BaseModel):
|
|
user_id: str
|
|
session_id: str
|
|
activity: str | None = None
|
|
mood: str | None = None
|
|
seed: Seed | None = None
|
|
exclude_ids: list[str] = Field(default_factory=list, max_length=1000)
|
|
feedback: list[Feedback] = Field(default_factory=list, max_length=200)
|
|
n: int = Field(default=5, ge=1, le=20)
|
|
|
|
|
|
class BatchTrack(BaseModel):
|
|
id: str
|
|
source: str
|
|
score: float
|
|
|
|
|
|
class BatchResponse(BaseModel):
|
|
tracks: list[BatchTrack]
|
|
model_ver: str
|
|
|
|
|
|
class WaveService:
|
|
def __init__(self, dsn: str, redis_url: str, contexts: Contexts) -> None:
|
|
self.dsn = dsn
|
|
self.index = IndexHolder(dsn)
|
|
self.sessions = SessionStore(redis_url)
|
|
self.contexts = contexts
|
|
self.ranker = RankerHolder(dsn)
|
|
|
|
def reload(self) -> None:
|
|
"""После ночного обучения: перечитать каталог (векторы ALS) и активный ранжировщик."""
|
|
self.index.refresh()
|
|
self.ranker.reload()
|
|
|
|
def rules(self, activity: str | None, mood: str | None) -> list[Rule]:
|
|
"""Занятие и настроение комбинируются — пересечение профилей; неизвестный ключ игнорируется."""
|
|
rules: list[Rule] = []
|
|
if activity and activity in self.contexts.activities:
|
|
rules += self.contexts.activities[activity].rules
|
|
if mood and mood in self.contexts.moods:
|
|
rules += self.contexts.moods[mood].rules
|
|
return rules
|
|
|
|
def batch(self, req: BatchRequest) -> BatchResponse:
|
|
index = self.index.get()
|
|
with psycopg.connect(self.dsn) as conn:
|
|
profile = load_profile(conn, req.user_id)
|
|
seeds = seed_tracks(conn, req.seed.type, req.seed.id) if req.seed else None
|
|
session = self.sessions.get(req.session_id)
|
|
update_session(
|
|
index, session, [(f.track_id, f.outcome) for f in req.feedback], taste_centroids(index, profile)
|
|
)
|
|
ranker = self.ranker.get()
|
|
picks: list[Pick] = build_batch(
|
|
index,
|
|
profile,
|
|
session,
|
|
self.rules(req.activity, req.mood),
|
|
n=req.n,
|
|
exclude=set(req.exclude_ids),
|
|
seed_ids=seeds,
|
|
rng=np.random.default_rng(),
|
|
ranker=ranker,
|
|
ctx=Context(has_activity=bool(req.activity), has_mood=bool(req.mood)),
|
|
)
|
|
self.log_features(req.session_id, picks)
|
|
session.served = (session.served + [p.id for p in picks])[-MAX_SERVED_MEMORY:]
|
|
artists = [index.artist[index.pos[p.id]] for p in picks]
|
|
session.recent_artists = (session.recent_artists + artists)[-10:]
|
|
self.sessions.put(req.session_id, session)
|
|
return BatchResponse(
|
|
tracks=[BatchTrack(id=p.id, source=p.source, score=p.score) for p in picks],
|
|
model_ver=model_version(index, profile, ranker),
|
|
)
|
|
|
|
def log_features(self, session_id: str, picks: list[Pick]) -> None:
|
|
"""Признаки показанных треков — обучающие данные ранжировщика v2. Сбой записи не ломает волну."""
|
|
rows = [(session_id, p.id, p.features) for p in picks if p.features is not None]
|
|
if not rows:
|
|
return
|
|
try:
|
|
with psycopg.connect(self.dsn, autocommit=True) as conn, conn.cursor() as cur:
|
|
cur.executemany(
|
|
"INSERT INTO rec_impression_features (wave_session_id, track_id, features) "
|
|
"VALUES (%s, %s, %s) "
|
|
"ON CONFLICT DO NOTHING",
|
|
rows,
|
|
)
|
|
except psycopg.Error as e:
|
|
log.warning("impression features not saved: %s", e)
|