Dokploy

Argilla

Argilla is a robust platform designed to help engineers and data scientists streamline the management of machine learning data workflows. It simplifies tasks like data labeling, annotation, and quality control.

Argilla logo

Configuration

version: "3.8"
services:
  argilla-web:
    image: argilla/argilla-server:latest
    restart: unless-stopped
    ports:
      - 6900
    environment:
      - ARGILLA_HOME_PATH=/var/lib/argilla
      - ARGILLA_ELASTICSEARCH=http://argilla-elasticsearch:9200
      - ARGILLA_DATABASE_URL=postgresql+asyncpg://postgres:${DB_PASSWORD}@argilla-db:5432/argilla
      - ARGILLA_REDIS_URL=redis://:${REDIS_PASSWORD}@argilla-redis:6379/0
      - USERNAME=${LOGIN_USERNAME}
      - PASSWORD=${LOGIN_PASSWORD}
      - API_KEY=argilla.apikey
      - WORKSPACE=default
    volumes:
      - argilladata:/var/lib/argilla
    depends_on:
      - argilla-elasticsearch
      - argilla-db
      - argilla-redis

  argilla-worker:
    image: argilla/argilla-server:latest
    restart: unless-stopped
    environment:
      - BACKGROUND_NUM_WORKERS=2
      - ARGILLA_HOME_PATH=/var/lib/argilla
      - ARGILLA_ELASTICSEARCH=http://argilla-elasticsearch:9200
      - ARGILLA_DATABASE_URL=postgresql+asyncpg://postgres:${DB_PASSWORD}@argilla-db:5432/argilla
      - ARGILLA_REDIS_URL=redis://:${REDIS_PASSWORD}@argilla-redis:6379/0
    volumes:
      - argilladata:/var/lib/argilla
    command: python -m argilla_server worker --num-workers ${BACKGROUND_NUM_WORKERS}
    depends_on:
      - argilla-elasticsearch
      - argilla-db
      - argilla-redis

  argilla-elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch:8.12.2
    restart: unless-stopped
    environment:
      - node.name=elasticsearch
      - cluster.name=es-argilla-local
      - discovery.type=single-node
      - ES_JAVA_OPTS=-Xms512m -Xmx512m
      - cluster.routing.allocation.disk.threshold_enabled=false
      - xpack.security.enabled=false
    volumes:
      - elasticdata:/usr/share/elasticsearch/data

  argilla-db:
    image: postgres:15-alpine
    restart: unless-stopped
    environment:
      - POSTGRES_USER=postgres
      - POSTGRES_PASSWORD=${DB_PASSWORD}
      - POSTGRES_DB=argilla
    volumes:
      - dbdata:/var/lib/postgresql/data

  argilla-redis:
    image: redis:7-alpine
    restart: unless-stopped
    environment:
      - REDIS_PASSWORD=${REDIS_PASSWORD}
    command: redis-server --requirepass ${REDIS_PASSWORD}
    volumes:
      - redisdata:/data

volumes:
  argilladata: {}
  elasticdata: {}
  dbdata: {}
  redisdata: {}
[variables]
main_domain = "${domain}"
login_username = "${username}"
login_password = "${password:8}"
db_password = "${password:16}"
redis_password = "${password:16}"

[config]
[[config.domains]]
serviceName = "argilla-web"
port = 6900
host = "${main_domain}"

[config.env]
LOGIN_USERNAME = "${login_username}"
LOGIN_PASSWORD = "${login_password}"
DB_PASSWORD = "${db_password}"
REDIS_PASSWORD = "${redis_password}"
BACKGROUND_NUM_WORKERS = "2"

Base64

To import this template in Dokploy: create a Compose service → Advanced → Base64 import and paste the content below:

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

Tags

machine-learning, data-labeling, ai

Version: latest

ArangoDB

ArangoDB is a native multi-model database with flexible data models for documents, graphs, and key-values. Build high performance applications using a convenient SQL-like query language or JavaScript extensions.

Audiobookshelf

Audiobookshelf is a self-hosted server designed to manage and play your audiobooks and podcasts. It works best when you have an organized directory structure.

On this page

ConfigurationBase64LinksTags