py-power-profile 0.1.0

Profile and visualize energy consumption of Python code

py-power-profile 🔋 | Python Energy Profiling Tool

Profile and visualize energy consumption of Python code on laptops, desktops, and Raspberry Pi devices. No external services or paid APIs required.

🚀 Quick Start

# Install py-power-profile
pip install py-power-profile

# Profile your Python script
py-power profile your_script.py --output results.json

# Generate energy badge
py-power badge results.json --target 100

✨ Key Features

• 🔋 Real-time Energy Profiling: Measure CPU energy consumption at function and line level
• 🖥️ Multi-Platform Support: Works on Intel/AMD (RAPL), ARM (hwmon), and universal fallback
• 📊 Rich Visual Reports: Beautiful tables with energy breakdowns and visual progress bars
• 🔄 Performance Comparison: Diff two runs to detect energy regressions and improvements
• 🏷️ CI/CD Integration: Generate Shields.io-compatible badges for GitHub/GitLab
• ⚡ Low Overhead: <5% CPU overhead, <150MB memory footprint
• 🔧 Zero Configuration: Auto-detects best energy measurement backend

📦 Installation

Basic Installation

pip install py-power-profile

With RAPL Support (Intel/AMD Processors)

pip install py-power-profile[rapl]

Development Installation

git clone https://github.com/Sherin-SEF-AI/py-power-profile.git
cd py-power-profile
pip install -e .[dev]

🛠️ Usage Examples

Profile Energy Consumption

# Basic profiling
py-power profile my_script.py

# Save results to JSON
py-power profile my_script.py --output energy_results.json

# Use specific backend
py-power profile my_script.py --backend rapl

# Line-level profiling (higher accuracy)
py-power profile my_script.py --line

Compare Performance Changes

# Compare two profiling runs
py-power compare old_results.json new_results.json

Generate Energy Badges

# Generate badge for CI/CD
py-power badge results.json --target 100 --output badge.svg

# Status-only badge
py-power badge results.json --target 100 --status-only

🔧 Supported Energy Measurement Backends

🖥️ Intel/AMD RAPL (Recommended)

• Accuracy: High (hardware-level measurement)
• Requirements: Intel/AMD processor with RAPL support
• Installation: pip install py-power-profile[rapl]

📱 ARM/Raspberry Pi HWMON

• Accuracy: High (hardware sensors)
• Requirements: ARM device with power sensors
• Availability: Raspberry Pi, ARM-based systems

💻 Universal PSUTIL Estimation

• Accuracy: Medium (CPU usage estimation)
• Requirements: None (fallback option)
• Availability: All systems

🧪 Mock Backend (Testing)

• Accuracy: Deterministic (for testing)
• Use Case: Unit tests, CI/CD
• Availability: All systems

📊 Output Formats

Rich Console Tables

Energy Profile Results (Backend: rapl)
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━┳━━━━━━━┳━━━━━━┳━━━━━━━┳━━━━━━┓
┃ Function                                           ┃ Calls┃ Energy┃ Avg   ┃ Time   ┃ %     ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━╇━━━━━━━╇━━━━━━╇━━━━━━━╇━━━━━━┩
│ my_script.py:heavy_computation                     │  100 │ 1500mJ│ 15.0mJ│ 50.0ms │ 75.0% │
│ my_script.py:light_operation                       │   10 │  100mJ│ 10.0mJ│  5.0ms │  5.0% │
└────────────────────────────────────────────────────┴──────┴───────┴──────┴───────┴──────┘

JSON Output

{
  "metadata": {
    "backend": "rapl",
    "line_level": false,
    "timestamp": 1640995200.0
  },
  "functions": {
    "my_script.py:heavy_computation": {
      "calls": 100,
      "total_energy_mj": 1500.0,
      "avg_energy_mj": 15.0,
      "total_time_ms": 50.0
    }
  },
  "summary": {
    "total_energy_mj": 2000.0,
    "total_time_ms": 100.0,
    "function_count": 5
  }
}

SVG Badges

⚙️ Configuration

Environment Variables

export PY_POWER_BACKEND="rapl"
export PY_POWER_TDP_WATTS="15"
export PY_POWER_ENERGY_BUDGET_MJ="1000"

pyproject.toml Configuration

[tool.py-power-profile]
backend = "auto"
tdp_watts = 15          # CPU TDP for estimation
energy_budget_mj = 1000 # CI threshold
ignore = ["tests/*"]    # glob patterns

🔄 GitHub Actions Integration

name: Energy Profile
on: [push, pull_request]

jobs:
  energy-profile:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v4
    
    - name: Set up Python
      uses: actions/setup-python@v4
      with:
        python-version: '3.9'
    
    - name: Install py-power-profile
      run: pip install py-power-profile[rapl]
    
    - name: Run energy profile
      run: py-power profile tests/test_script.py --output results.json
    
    - name: Generate badge
      run: py-power badge results.json --target 100 --output badge.svg
    
    - name: Upload results
      uses: actions/upload-artifact@v3
      with:
        name: energy-results
        path: [results.json, badge.svg]

🧪 Testing

# Run all tests
pytest

# Run with coverage
pytest --cov=py_power_profile

# Test specific backend
py-power profile samples/quick.py --backend mock

📈 Performance Benchmarks

Metric	Value
Profiling Overhead	<5% CPU
Memory Footprint	<150MB
Supported Python	3.9+
Supported OS	Linux, macOS, Windows

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

Development Setup

git clone https://github.com/Sherin-SEF-AI/py-power-profile.git
cd py-power-profile
pip install -e .[dev]
pre-commit install

📚 Documentation

• Installation Guide
• API Reference
• Accuracy Guide
• Troubleshooting

🔍 Use Cases

Software Development

• Performance Optimization: Identify energy-intensive functions
• Code Review: Energy impact analysis in pull requests
• CI/CD: Automated energy regression detection

Research & Academia

• Algorithm Analysis: Compare energy efficiency of algorithms
• System Research: Energy consumption studies
• Green Computing: Sustainable software development

IoT & Embedded Systems

• Battery Life: Optimize Python applications for battery-powered devices
• Raspberry Pi: Energy profiling on ARM devices
• Edge Computing: Resource-constrained environments

🏆 Why py-power-profile?

• 🔬 Scientific Accuracy: Hardware-level energy measurement
• 🚀 Easy Integration: Simple CLI with rich output
• 🔧 Flexible Configuration: Multiple backends and options
• 📊 Professional Reports: Beautiful, informative output
• 🔄 CI/CD Ready: GitHub Actions and badge integration
• 📱 Cross-Platform: Works on laptops, desktops, and SBCs

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• pyRAPL for Intel/AMD RAPL support
• Rich for beautiful terminal output
• Typer for CLI framework

📞 Support

• GitHub Issues: Report bugs
• Discussions: Community support
• Email: [e-mail]

---

Made with ❤️ by sherin joseph roy

Empowering developers to build energy-efficient Python applications