Implementation:Microsoft Onnxruntime Sklearn Model Training

Metadata

Field	Value
Implementation Name	Sklearn_Model_Training
Repository	Microsoft_Onnxruntime
Source Repository	https://github.com/microsoft/onnxruntime
Type	External Tool Doc
External Tool	scikit-learn
Language	Python
Domain	ML_Inference, Model_Conversion
Last Updated	2026-02-10
Workflow	Train_Convert_Predict
Pair	1 of 5

Overview

External tool documentation for training scikit-learn machine learning models as the first step in the ONNX conversion pipeline. The trained model will subsequently be converted to ONNX format for optimized inference.

API Signature

sklearn.linear_model.LogisticRegression().fit(X_train, y_train)

Or for ensemble models:

sklearn.ensemble.RandomForestClassifier(n_estimators=10).fit(X_train, y_train)

Import

from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split

Code Reference

Reference	Location
LogisticRegression training	docs/python/examples/plot_train_convert_predict.py:L22-34
RandomForest training	docs/python/examples/plot_train_convert_predict.py:L178-181

I/O Contract

Inputs

Parameter	Type	Required	Description
X_train	numpy.ndarray	Yes	Training feature matrix with shape (n_samples, n_features).
y_train	numpy.ndarray	Yes	Training label vector with shape (n_samples,).

Outputs

Output	Type	Description
Trained model	scikit-learn estimator object	A fitted model object containing learned parameters, ready for conversion to ONNX format.

Usage Example

LogisticRegression Training

from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split

iris = load_iris()
X, y = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y)
clr = LogisticRegression()
clr.fit(X_train, y_train)

From the source at docs/python/examples/plot_train_convert_predict.py:L22-34:

from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split

iris = load_iris()
X, y = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y)

clr = LogisticRegression()
clr.fit(X_train, y_train)

RandomForestClassifier Training

From the source at docs/python/examples/plot_train_convert_predict.py:L178-181:

from sklearn.ensemble import RandomForestClassifier

rf = RandomForestClassifier(n_estimators=10)
rf.fit(X_train, y_train)

Related Pages

• Principle:Microsoft_Onnxruntime_Source_Framework_Training
• Implementation:Microsoft_Onnxruntime_FloatTensorType_Init
• Environment:Microsoft_Onnxruntime_Sklearn_Conversion_Environment