Accelerate inference without tears - View it on GitHub