Reimbursement_RAG/docker-compose.yml at main · AmanSikarwar/Reimbursement_RAG · 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
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
services:
  # Qdrant Vector Database
  qdrant:
    image: qdrant/qdrant:latest
    container_name: invoice-qdrant
    ports:
      - "6333:6333"
      - "6334:6334"
    volumes:
      - qdrant_data:/qdrant/storage
    environment:
      - QDRANT__SERVICE__HTTP_PORT=6333
      - QDRANT__SERVICE__GRPC_PORT=6334
    healthcheck:
      test: ["CMD-SHELL", "find /proc -maxdepth 2 -name comm -exec grep -l qdrant {} + | head -1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    networks:
      - invoice-network

  # FastAPI Backend
  backend:
    build:
      context: .
      dockerfile: Dockerfile
      target: production
    container_name: invoice-backend
    ports:
      - "8000:8000"
    environment:
      # Application settings
      - APP_NAME=Invoice Reimbursement System
      - APP_VERSION=1.0.0
      - DEBUG=false
      - LOG_LEVEL=INFO
      - ENVIRONMENT=development

      # Security settings
      - ALLOWED_HOSTS=localhost,127.0.0.1,backend,invoice-backend

      # Qdrant configuration
      - QDRANT_URL=http://qdrant:6333
      - QDRANT_API_KEY=${QDRANT_API_KEY:-}

      # Gemini API configuration
      - GOOGLE_API_KEY=${GOOGLE_API_KEY}

      # Vector store configuration
      - COLLECTION_NAME=invoice_reimbursements
      - EMBEDDING_MODEL=all-MiniLM-L6-v2
      - VECTOR_SIZE=384

      # File upload configuration
      - MAX_FILE_SIZE=50
      - ALLOWED_EXTENSIONS=pdf,zip
      - UPLOAD_DIRECTORY=uploads

      # LLM configuration
      - LLM_MODEL=gemini-2.5-flash
      - LLM_TEMPERATURE=0.1
      - MAX_TOKENS=4096

      # Chat configuration
      - MAX_CONVERSATION_HISTORY=10

      # Cache directories for models
      - HF_HOME=/app/.cache/huggingface
      - TRANSFORMERS_CACHE=/app/.cache/transformers
      - SENTENCE_TRANSFORMERS_HOME=/app/.cache/sentence-transformers
    volumes:
      - ./uploads:/app/uploads
      - ./logs:/app/logs
      - model_cache:/app/.cache
    depends_on:
      qdrant:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/api/v1/health/quick"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    networks:
      - invoice-network
    restart: unless-stopped

  # Streamlit Frontend
  frontend:
    build:
      context: .
      dockerfile: Dockerfile
      target: production
    container_name: invoice-frontend
    ports:
      - "8501:8501"
    environment:
      - STREAMLIT_API_BASE_URL=http://backend:8000
      - STREAMLIT_API_TIMEOUT=300
      - STREAMLIT_DEFAULT_CURRENCY=INR
    command: ["python", "-m", "streamlit", "run", "streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0", "--server.headless=true"]
    volumes:
      - ./uploads:/app/uploads
    depends_on:
      backend:
        condition: service_healthy
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8501/_stcore/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 40s
    networks:
      - invoice-network
    restart: unless-stopped

volumes:
  qdrant_data:
    driver: local
  model_cache:
    driver: local

networks:
  invoice-network:
    driver: bridge