speakora/docker-compose.gpu.yml at main · rennerdo30/speakora · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
version: '3.9'

# GPU-enabled docker-compose configuration
# Requires: nvidia-docker2 or Docker with GPU support
# Usage: docker-compose -f docker-compose.gpu.yml up

services:
  api:
    build:
      context: .
      args:
        BASE_IMAGE: nvidia/cuda:12.1-runtime-ubuntu22.04
    command: python tool/main.py gui --host 0.0.0.0 --port 5000
    ports:
      - "5000:5000"
    volumes:
      - ./input:/app/input
      - ./output:/app/output
      - ./config:/app/config
      - ~/.cache/huggingface:/root/.cache/huggingface
    environment:
      - SEAMLESS_DEVICE=cuda
      - LOG_LEVEL=${LOG_LEVEL:-INFO}
      - CUDA_VISIBLE_DEVICES=${CUDA_VISIBLE_DEVICES:-0}
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:5000/api/system/status"]
      interval: 30s
      timeout: 10s
      retries: 3

  worker:
    build:
      context: .
      args:
        BASE_IMAGE: nvidia/cuda:12.1-runtime-ubuntu22.04
    command: python tool/main.py worker --num-workers ${NUM_WORKERS:-1}
    volumes:
      - ./input:/app/input
      - ./output:/app/output
      - ./config:/app/config
      - ~/.cache/huggingface:/root/.cache/huggingface
    environment:
      - SEAMLESS_DEVICE=cuda
      - LOG_LEVEL=${LOG_LEVEL:-INFO}
      - NUM_WORKERS=${NUM_WORKERS:-1}
      - CUDA_VISIBLE_DEVICES=${CUDA_VISIBLE_DEVICES:-0}
    restart: unless-stopped
    depends_on:
      - api
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]