Skip to content

Transform ML models into a native code (Java, C, Python, etc.) with zero dependencies

License

Notifications You must be signed in to change notification settings

JiafeiPan/m2cgen

 
 

Repository files navigation

m2cgen

Build Status Coverage Status License: MIT

m2cgen (Model 2 Code Generator) - is a lightweight library which provides an easy way to transpile trained statistical models into a native code (Python, C, Java).

Installation

Supported Python version is >= 3.4.

pip install m2cgen

Supported Languages

  • Python
  • Java
  • C

Supported Models

Classification Regression
Linear LogisticRegression, LogisticRegressionCV, RidgeClassifier, RidgeClassifierCV, SGDClassifier, PassiveAggressiveClassifier LinearRegression, HuberRegressor, ElasticNet, ElasticNetCV, TheilSenRegressor, Lars, LarsCV, Lasso, LassoCV, LassoLars, LassoLarsIC, OrthogonalMatchingPursuit, OrthogonalMatchingPursuitCV, Ridge, RidgeCV, BayesianRidge, ARDRegression, SGDRegressor, PassiveAggressiveRegressor
SVM LinearSVC LinearSVR
Tree DecisionTreeClassifier, ExtraTreeClassifier DecisionTreeRegressor, ExtraTreeRegressor
Random Forest RandomForestClassifier, ExtraTreesClassifier RandomForestRegressor, ExtraTreesRegressor
Boosting XGBClassifier(gbtree/dart booster only), LGBMClassifier(gbdt/dart booster only) XGBRegressor(gbtree/dart booster only), LGBMRegressor(gbdt/dart booster only)

Classification Output

Binary Multiclass Comment
Linear Scalar value; signed distance of the sample to the hyperplane for the second class Vector value; signed distance of the sample to the hyperplane per each class The output is consistent with the output of LinearClassifierMixin.decision_function
Tree/Random Forest/XGBoost/LightGBM Vector value; class probabilities Vector value; class probabilities The output is consistent with the output of the predict_proba method of DecisionTreeClassifier/ForestClassifier/XGBClassifier/LGBMClassifier

Usage

Here's a simple example of how a trained linear model can be represented in Java code:

from sklearn.datasets import load_boston
from sklearn import linear_model
import m2cgen as m2c

boston = load_boston()
X, y = boston.data, boston.target

estimator = linear_model.LinearRegression()
estimator.fit(X, y)

code = m2c.export_to_java(estimator)

The example of the generated code:

public class Model {

    public static double score(double[] input) {
        return (((((((((((((36.45948838508965) + ((input[0]) * (-0.10801135783679647))) + ((input[1]) * (0.04642045836688297))) + ((input[2]) * (0.020558626367073608))) + ((input[3]) * (2.6867338193449406))) + ((input[4]) * (-17.76661122830004))) + ((input[5]) * (3.8098652068092163))) + ((input[6]) * (0.0006922246403454562))) + ((input[7]) * (-1.475566845600257))) + ((input[8]) * (0.30604947898516943))) + ((input[9]) * (-0.012334593916574394))) + ((input[10]) * (-0.9527472317072884))) + ((input[11]) * (0.009311683273794044))) + ((input[12]) * (-0.5247583778554867));
    }
}

You can find more examples of generated code for different models/languages here

CLI

m2cgen can be used as a CLI tool to generate code using serialized model objects (pickle protocol):

$ m2cgen <pickle_file> --language <language> [--indent <indent>]
         [--class_name <class_name>] [--package_name <package_name>]
         [--recursion-limit <recursion_limit>]

Piping is also supported:

$ cat <pickle_file> | m2cgen --language <language>

FAQ

Q: Generation fails with RuntimeError: maximum recursion depth exceeded error.

A: If this error occurs while generating code using an ensemble model, try to reduce the number of trained estimators within that model. Alternatively you can increase the maximum recursion depth with sys.setrecursionlimit(<new_depth>).

About

Transform ML models into a native code (Java, C, Python, etc.) with zero dependencies

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 82.2%
  • Java 10.1%
  • C 7.7%