Initial commit: mBART Translation API with Docker support

- FastAPI 기반 다국어 번역 REST API 서비스
- mBART-50 모델을 사용한 18개 언어 지원
- Docker 및 Docker Compose 설정 포함
- GPU/CPU 지원
- 헬스 체크 및 API 문서 자동 생성
- 외부 접속 지원 (172.30.1.2:8000)

주요 파일:
- main.py: FastAPI 애플리케이션
- translator.py: mBART 번역 서비스
- models.py: Pydantic 데이터 모델
- config.py: 환경 설정
- Dockerfile: 최적화된 Docker 이미지
- docker-compose.yml: 간편한 배포 설정

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

docker-compose.yml

@@ -0,0 +1,55 @@
+version: '3.8'
+
+services:
+  mbart-api:
+    build:
+      context: .
+      dockerfile: Dockerfile
+    container_name: mbart-translation-api
+    ports:
+      - "8000:8000"
+    environment:
+      - HOST=0.0.0.0
+      - PORT=8000
+      - MODEL_NAME=facebook/mbart-large-50-many-to-many-mmt
+      - MAX_LENGTH=512
+      - DEVICE=cpu  # Change to 'cuda' for GPU support
+    volumes:
+      # Cache HuggingFace models to avoid re-downloading
+      - huggingface-cache:/home/appuser/.cache/huggingface
+    restart: unless-stopped
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 60s
+
+  # GPU support (uncomment if you have NVIDIA GPU)
+  # mbart-api-gpu:
+  #   build:
+  #     context: .
+  #     dockerfile: Dockerfile
+  #   container_name: mbart-translation-api-gpu
+  #   ports:
+  #     - "8000:8000"
+  #   environment:
+  #     - HOST=0.0.0.0
+  #     - PORT=8000
+  #     - MODEL_NAME=facebook/mbart-large-50-many-to-many-mmt
+  #     - MAX_LENGTH=512
+  #     - DEVICE=cuda
+  #   volumes:
+  #     - huggingface-cache:/home/appuser/.cache/huggingface
+  #   deploy:
+  #     resources:
+  #       reservations:
+  #         devices:
+  #           - driver: nvidia
+  #             count: 1
+  #             capabilities: [gpu]
+  #   restart: unless-stopped
+
+volumes:
+  huggingface-cache:
+    driver: local