ZFTurbo/Music-Source-Separation-Training — GitHub trend

◂ BACK TO LEADERBOARD

FINDING #7053 · UNIT ID 715096001

ZFTurbo/Music-Source-Separation-Training

Repository for training models for music source separation.

[ PYTHON ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.000848

ENGAGEMENT0.31

FRESHNESS1.14

SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE

SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)

3% OF STARS IN ARCHIVE

Growth Telemetry

VELOCITY /D

1.50

ACCEL

-0.04

RETENTION

36.5%

PEAK 2026-08-04 · FORK-RETENTION 0.0% · 45 STARS / WINDOW

Author Audience

AUDIENCE

1,729

FOLLOWERS

1,031

OWNER ★

6,984

Engagement Signals

FORKS

219

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 45 / 45 (DIVERSITY 1.00)

Why This Is A Finding

ZFTurbo/Music-Source-Separation-Training собрал 45 звёзд за окно, тогда как у автора всего 1,031 подписчиков — эффективная аудитория ≈ 1,729. Это даёт surprise-индекс 0.000848 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 0.0% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.

Related Findings

RANKS ABOVE 26% OF 9593 FINDINGS

#02guillaumemeyer/watermarks-remover[ PYTHON ]0.44 #07awesome-dsh-plugin/awesome-dsh-plugin[ PYTHON ]0.38 #14wang2122/sprix-sage-router[ PYTHON ]0.36 #23ayghri/i-have-adhd[ PYTHON ]0.26 #24sowarma/wp2shell-PoC[ PYTHON ]0.26

METRICS IN CONTEXT

MEDIAN ACROSS ALL 9593 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW

METRICVALUEMEDIANΔ MEDPERCENTILE

SCORE0.000.00-0.00ABOVE 26%

VELOCITY1.503.63-2.13ABOVE 0%

RETENTION36.5%25.0%+11.5 PPABOVE 72%

FORKS219117+102ABOVE 65%

SURPRISE0.000.01-0.01ABOVE 14%