[BUG] Some Config Parameters dont get applied

[BenMajestix]

Describe the bug
Not all, but some Config Parameters dont get applied from the env file, to the Frontend/App/i am not sure.
I have got a extensive .env file, which mostly works, but at least these two Params dont get applied:

• MAX_DISTANCE
• MAX_SONGS_PER_ARTIST
  The MAX_DISTANCE Param, is easily checkable through the Webinterface, it always stays on the default value "0.5", no matter what i change it to.
  MAX_SONGS_PER_ARTIST is not so easily checkable, but in Playlists it creates, there is never more than 3 Songs per Artist, which I believe to be the default.

To Reproduce
Steps to reproduce the behavior:
0. Delete Volumes, so fresh Start

1. Input Values into the .env File for the two example Parameters
2. Start the Services
3. Observer that the default Values are still there

Expected behavior
My chosen Values are to be applied, not staying with the default.

Environment (please complete the following information):

• OS: Fedora Linux
• Deployment: Docker Compose
• AudioMuse-AI Version: latest (v0.8.2)
• Jellyfin/Navidrome Version: Jellyfin 10.11.3
• Browser: Firefox

Additional context
Most other Parameters get applied just fine, they always change when i change them in the .env file, for example:

NUM_CLUSTERS_MIN
SCORE_WEIGHT_DIVERSITY
SCORE_WEIGHT_PURITY
GEMINI_MODEL_NAME
MIN_SONGS_PER_GENRE_STRATIFICATION
etc.

.env File:

#--- Media Server Configuration ---

MEDIASERVER_TYPE=jellyfin
JELLYFIN_URL=http://192.168.178.57:8096
JELLYFIN_USER_ID=<redacted>
JELLYFIN_TOKEN=<redacted>

#--- Database Credentials ---

POSTGRES_USER=audiomuse
POSTGRES_PASSWORD=audiomusepassword
POSTGRES_DB=audiomusedb
POSTGRES_HOST=postgres
POSTGRES_PORT=5432

#--- Redis Configuration ---

REDIS_URL=redis://redis:6379

#--- AI Provider (Gemini) ---

AI_MODEL_PROVIDER=GEMINI
GEMINI_MODEL_NAME=gemini-2.5-flash
GEMINI_API_KEY=<redacted>

#--- Analysis Parameters ---

TEMP_DIR=/app/temp_audio
NUM_RECENT_ALBUMS=0
TOP_N_MOODS=5

#--- Clustering General ---

CLUSTER_ALGORITHM=kmeans
ENABLE_CLUSTERING_EMBEDDINGS=true
MAX_DISTANCE=0.8
MAX_SONGS_PER_CLUSTER=0
MAX_SONGS_PER_ARTIST=8

#--- K-Means Specific ---
NUM_CLUSTERS_MIN=10
NUM_CLUSTERS_MAX=20

#--- Playlist Generation ---
TOP_PLAYLIST_NUMBER=15
CLUSTERING_RUNS=5000

#--- Advanced / Scoring Weights ---
PCA_COMPONENTS_MIN=0
PCA_COMPONENTS_MAX=199
MIN_SONGS_PER_GENRE_STRATIFICATION=30
STRATIFIED_SAMPLING_TARGET_PERCENTILE=50
SCORE_WEIGHT_DIVERSITY=2.0
SCORE_WEIGHT_PURITY=1.0

Compose File:

services:
  # Redis service for RQ (task queue)
  redis:
    image: redis:7-alpine
    container_name: audiomuse-redis
    ports:
      - "6379:6379"
    volumes:
      - ./redisdata:/data
    restart: unless-stopped

  # PostgreSQL database service
  postgres:
    image: postgres:15-alpine
    container_name: audiomuse-postgres
    env_file:
      - .env
    ports:
      - "5432:5432"
    volumes:
      - ./pgdata:/var/lib/postgresql/data
    restart: unless-stopped

  # AudioMuse-AI Flask application service
  audiomuse-ai-flask:
    image: ghcr.io/neptunehub/audiomuse-ai:latest
    container_name: audiomuse-ai-flask-app
    ports:
      - "8000:8000"
    env_file:
      - .env
    environment:
      - SERVICE_TYPE=flask
    volumes:
      - ./tmp/audio-flask:/app/temp_audio
    depends_on:
      - redis
      - postgres
    restart: unless-stopped

  # AudioMuse-AI RQ Worker service
  audiomuse-ai-worker:
    image: ghcr.io/neptunehub/audiomuse-ai:latest
    container_name: audiomuse-ai-worker-instance
    env_file:
      - .env
    environment:
      - SERVICE_TYPE=worker
    volumes:
      - ./tmp/audio-worker:/app/temp_audio
    depends_on:
      - redis
      - postgres
    restart: unless-stopped

[NeptuneHub]

Hi and thanks for the feedback. I Just fixed this two parameter that was hardcoded:

MAX_DISTANCE = 0.5
MAX_SONGS_PER_CLUSTER = 0

the fix it's in the :devel image just now.

Instead MAX_SONGS_PER_ARTIST it works and it is set to 3 to avoid to have in cluster, song similarity more than 3 songs from the same artist.

[BenMajestix]

That worked, thanks.