Getting Started

These tutorials walk through complete, working examples that demonstrate TinyMind’s capabilities on real problems. Each tutorial includes source code, build instructions, and size analysis.

Tutorial	What You’ll Learn	Final Size
Neural Network in Under 4KB	Feed-forward NN with fixed-point, XOR prediction	3,892 bytes
Q-Learning in Under 1KB	Tabular Q-learning, maze solving	869 bytes
DQN Maze Solver	Deep Q-Network with neural network function approximation	~16 KB
Keyword Spotting CNN on a Cortex-M	Depthwise-separable 2D CNN, bench harness, MCU porting	~19 KB static
Predictive Maintenance on AI4I 2020	Q16.16 MLP, imbalanced binary classification, confusion matrix	~35 KB static

---

Table of contents

• Neural Network in Under 4KB
• Q-Learning in Under 1KB
• DQN Maze Solver
• Keyword Spotting CNN on a Cortex-M
• Predictive Maintenance on the AI4I 2020 Dataset