A simple tool to embed scikit-learn models into microcontrollers
Project description
Machine Learning for Embbedded Devices
sklearn2c is a tool that converts scikit-learn library classification algorithms to C code. It can be used to generate C code from trained models, which can then be used in microcontrollers or other embedded systems. The generated code can be used for real-time classification tasks, where the computational resources are limited.
Supported Models
Classification
-
Bayes Classifier*
-
Decision Trees
-
KNN Classifier
-
C-SVC**
*: sklearn2c does not use scikit-learn
GaussianNB()
, instead it uses the following cases to compute decision function.**:
linear
,poly
andrbf
kernels are supported.
Regression
- Linear Regression
- Polynomial Regression
- KNN
- Decision Trees
Clustering
- kmeans
- DBSCAN
Installation
You can install the library via pip either using:
pip install sklearn2c
or
pip install git+git@github.com:EmbeddedML/sklearn2c.git
Alternatively, you can install conda package:
conda install sklearn2c
or mamba install sklearn2c
Usage
Contributing
TBD
License
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for sklearn2c-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | f0bf5c9d61d26767399f01abe7ad87f166b0683b503c9543f061de5a1b8db6c6 |
|
MD5 | 5511706b047c60f6b4afed9ade840d9c |
|
BLAKE2b-256 | 3cc769dc7a82ad7a0448617f496e9abee8a209caae9b43cadc6b26c6f4efdd83 |