fix: Simplify Docker configuration for CLI tool usage

- Remove unnecessary port exposure and health checks from Dockerfile
- Simplify docker-compose.yml for CLI tool (remove Redis, ports, health checks)
- Update DOCKER.md with CLI-focused usage examples
- Fix README.md Docker example to show proper CLI usage
- Add more API key environment variables (Cohere, Mistral, Google)

Co-authored-by: openhands <[email protected]>

Author: h-albert-lee

DOCKER.md

@@ -20,33 +20,44 @@ This guide explains how to build, run, and deploy the Haerae Evaluation Toolkit
 ### 2. Run the Container
 
 ```bash
-# Basic run
+# Basic run (shows help)
 docker run -it --rm haerae-evaluation-toolkit:latest
 
-# Run with environment variables
+# Run with environment variables and volume mounts
 docker run -it --rm \
   -e OPENAI_API_KEY=your_key_here \
   -e WANDB_API_KEY=your_wandb_key \
+  -v $(pwd)/data:/app/data \
+  -v $(pwd)/configs:/app/configs \
+  -v $(pwd)/results:/app/results \
   haerae-evaluation-toolkit:latest
 
-# Run with volume mounts
+# Run specific evaluation
 docker run -it --rm \
+  -e OPENAI_API_KEY=your_key_here \
   -v $(pwd)/data:/app/data \
   -v $(pwd)/configs:/app/configs \
-  -p 8000:8000 \
-  haerae-evaluation-toolkit:latest
+  haerae-evaluation-toolkit:latest \
+  python run_eval.py --config configs/your_config.yaml
+
+# Interactive shell
+docker run -it --rm \
+  -v $(pwd):/workspace \
+  haerae-evaluation-toolkit:latest \
+  bash
 ```
 
 ### 3. Using Docker Compose
 
 ```bash
-# Start all services
-docker-compose up -d
+# Start container (interactive mode)
+docker-compose up
 
-# View logs
-docker-compose logs -f
+# Run in background and execute commands
+docker-compose up -d
+docker-compose exec hret python run_eval.py --help
 
-# Stop services
+# Stop container
 docker-compose down
 ```
 
@@ -88,6 +99,9 @@ The container supports the following environment variables:
 |----------|-------------|---------|
 | `OPENAI_API_KEY` | OpenAI API key | - |
 | `ANTHROPIC_API_KEY` | Anthropic API key | - |
+| `COHERE_API_KEY` | Cohere API key | - |
+| `MISTRAL_API_KEY` | Mistral API key | - |
+| `GOOGLE_API_KEY` | Google API key | - |
 | `WANDB_API_KEY` | Weights & Biases API key | - |
 | `HRET_HOME` | Application home directory | `/app` |
 | `PYTHONPATH` | Python path | `/app` |
@@ -96,20 +110,18 @@ The container supports the following environment variables:
 
 Recommended volume mounts:
 
-- `/app/data` - For datasets and evaluation results
+- `/app/data` - For datasets and input data
 - `/app/logs` - For application logs
 - `/app/configs` - For configuration files
-
-## Ports
-
-- `8000` - FastAPI web server (if running web interface)
+- `/app/results` - For evaluation results and outputs
 
 ## Docker Compose Configuration
 
 The `docker-compose.yml` includes:
 
-- **hret**: Main application container
-- **redis**: Optional Redis cache (commented out by default)
+- **hret**: Main application container with interactive mode
+- **Volume Mounts**: For persistent data and configuration
+- **Environment Variables**: For API keys and configuration
 
 ### Environment File
 

Dockerfile

@@ -64,12 +64,5 @@ RUN mkdir -p /app/data /app/logs /app/configs && \
 # Switch to non-root user
 USER hret
 
-# Expose port for FastAPI
-EXPOSE 8000
-
-# Health check
-HEALTHCHECK --interval=30s --timeout=30s --start-period=5s --retries=3 \
-    CMD curl -f http://localhost:8000/health || exit 1
-
 # Default command - can be overridden
 CMD ["python", "run_eval.py", "--help"]

README.md

@@ -85,7 +85,9 @@ cd haerae-evaluation-toolkit
 docker run -it --rm \
   -e OPENAI_API_KEY=your_key_here \
   -v $(pwd)/data:/app/data \
-  haerae-evaluation-toolkit:latest
+  -v $(pwd)/configs:/app/configs \
+  haerae-evaluation-toolkit:latest \
+  python run_eval.py --help
 ```
 
 For detailed Docker usage, see [DOCKER.md](DOCKER.md).

docker-compose.yml

@@ -7,44 +7,21 @@ services:
       dockerfile: Dockerfile
     image: haerae-evaluation-toolkit:latest
     container_name: hret-container
-    ports:
-      - "8000:8000"
     environment:
       - PYTHONPATH=/app
       - HRET_HOME=/app
       - WANDB_API_KEY=${WANDB_API_KEY:-}
       - OPENAI_API_KEY=${OPENAI_API_KEY:-}
       - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
+      - COHERE_API_KEY=${COHERE_API_KEY:-}
+      - MISTRAL_API_KEY=${MISTRAL_API_KEY:-}
+      - GOOGLE_API_KEY=${GOOGLE_API_KEY:-}
     volumes:
       - ./data:/app/data
       - ./logs:/app/logs
       - ./configs:/app/configs
-    networks:
-      - hret-network
-    restart: unless-stopped
-    healthcheck:
-      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 40s
-
-  # Optional: Redis for caching (if needed)
-  redis:
-    image: redis:7-alpine
-    container_name: hret-redis
-    ports:
-      - "6379:6379"
-    networks:
-      - hret-network
-    restart: unless-stopped
-    command: redis-server --appendonly yes
-    volumes:
-      - redis-data:/data
-
-networks:
-  hret-network:
-    driver: bridge
-
-volumes:
-  redis-data:
+      - ./results:/app/results
+    working_dir: /app
+    # Keep container running for interactive use
+    tty: true
+    stdin_open: true