Warning
This fork does not follow the structure or versioning of the original project. Changes have been made to the code to fix certain problems and add new features, but it is not guaranteed to be stable or to work as intended from the original project, nor will the original documentation apply to this fork.
The main features added are the generation of pure c-code for models instead of the Cpp-Arduino templates/namespaces; in addition to corrections to the original code.
Please note that as this fork may have diverged from the original project for my own use, I won't be making a pull request to the original project.
MicroML is an attempt to bring Machine Learning algorithms to microcontrollers. Please refer to this blog post to an introduction to the topic.
pip install git+https://github.com/PYBrulin/micromlgen.git
micromlgen can port to plain C many types of classifiers:
- DecisionTree
- RandomForest
- XGBoost
- GaussianNB
- Support Vector Machines (SVC and OneClassSVM)
- Relevant Vector Machines (from
skbayes.rvm_ard_modelspackage) - SEFR
- PCA
from micromlgen import port from sklearn.svm import SVC from sklearn.datasets import load_iris if __name__ == '__main__': iris = load_iris() X = iris.data y = iris.target clf = SVC(kernel='linear').fit(X, y) print(port(clf))You may pass a classmap to get readable class names in the ported code
from micromlgen import port from sklearn.svm import SVC from sklearn.datasets import load_iris if __name__ == '__main__': iris = load_iris() X = iris.data y = iris.target clf = SVC(kernel='linear').fit(X, y) print(port(clf, classmap={ 0: 'setosa', 1: 'virginica', 2: 'versicolor' }))It can export a PCA transformer.
from sklearn.decomposition import PCA from sklearn.datasets import load_iris from micromlgen import port if __name__ == '__main__': X = load_iris().data pca = PCA(n_components=2, whiten=False).fit(X) print(port(pca))Read the post about SEFR.
pip install sefrfrom sefr import SEFR from micromlgen import port clf = SEFR() clf.fit(X, y) print(port(clf))pip install micromlgen>=1.1.26from sklearn.datasets import load_boston from sklearn.tree import DecisionTreeRegressor from sklearn.ensemble import RandomForestRegressor from micromlgen import port if __name__ == '__main__': X, y = load_boston(return_X_y=True) regr = DecisionTreeRegressor(max_depth=10, min_samples_leaf=5).fit(X, y) regr = RandomForestRegressor(n_estimators=10, max_depth=10, min_samples_leaf=5).fit(X, y) with open('RandomForestRegressor.h', 'w') as file: file.write(port(regr))// Arduino sketch #include "RandomForestRegressor.h" Eloquent::ML::Port::RandomForestRegressor regressor; float X[] = {...}; void setup() { } void loop() { float y_pred = regressor.predict(X); }