gaoyiyeah / emtrees

Tree-based machine learning classifiers for embedded systems

emtrees

Tree-based machine learning classifiers for embedded systems. Train in Python, deploy on microcontroller.

Want Naive Bayes instead? Go to embayes

Key features

Embedded-friendly Classifier

• Portable C99 code
• No stdlib required
• No dynamic allocations
• Integer/fixed-point math only
• Single header file include
• Fast, sub-millisecond classification

Convenient Training

• API-compatible with scikit-learn
• Implemented in Python 3
• C classifier accessible in Python using pybind11

MIT licensed

Status

Minimally useful

• Random Forests and ExtraTrees classifiers implemented
• Tested running on AVR, ESP8266 and Linux.
• On ESP8266, 8x8 digits classify in under 0.3ms with 95%+ accuracy

Installing

Install from PyPI

pip install emtrees --user

Usage

1. Train your model in Python

import emtrees
estimator = emtrees.RandomForest(n_estimators=10, max_depth=10)
estimator.fit(X_train, Y_train)
...

2. Generate C code

code = estimator.output_c('sonar')
with open('sonar.h', 'w') as f:
   f.write(code)

3. Use the C code

#include <emtrees.h>
#include "sonar.h"

const int32_t length = 60;
int32_t values[length] = { ... };
const int32_t predicted_class = sonar_predict(values, length):

For full example code, see examples/digits.py and arduinobench.ino

TODO

0.2

• Standalone example application on microcontroller

1.0

• Support serializing/deserializing trees
• Support multi-target classification
• Maybe implement a Very Fast Decision Tree (VFDT) learning algorithm