385d2cf386
- Add complete API documentation and architecture guides - Include quick start, installation, and deployment guides - Add troubleshooting and security documentation - Include CLI reference and configuration schema docs - Add production monitoring and operations guides - Implement MkDocs configuration with search functionality - Include comprehensive user and developer documentation Provides complete documentation for users and developers covering all aspects of the FetchML platform.
1.3 KiB
1.3 KiB
Simple Installation Guide
Quick Start (5 minutes)
# 1. Install
git clone https://github.com/jfraeys/fetch_ml.git
cd fetch_ml
make install
# 2. Setup (auto-configures)
./bin/ml setup
# 3. Run experiments
./bin/ml run my-experiment.py
That's it. Everything else is optional.
What If I Want More Control?
Manual Configuration (Optional)
# Edit settings if defaults don't work
nano ~/.ml/config.toml
Monitoring Dashboard (Optional)
# Real-time monitoring
./bin/tui
Senior Developer Feedback
"Keep it simple" - Most data scientists want:
- One installation command
- Sensible defaults
- Works without configuration
- Advanced features available when needed
Current plan is too complex because it asks users to decide between:
- CLI vs TUI vs Both
- Zig vs Go build tools
- Manual vs auto config
- Multiple environment variables
Better approach: Start simple, add complexity gradually.
Recommended Simplified Workflow
- Single Binary - Combine CLI + basic TUI functionality
- Auto-Discovery - Detect common ML environments automatically
- Progressive Disclosure - Show advanced options only when needed
- Zero Config - Work out-of-the-box with localhost defaults
The goal: "It just works" for 80% of use cases.