https://github.com/onnx/sklearn-onnx

Convert scikit-learn models and pipelines to ONNX
https://github.com/onnx/sklearn-onnx

onnx scikit-learn

Last synced: 30 days ago
JSON representation

Convert scikit-learn models and pipelines to ONNX

• Host: GitHub
• URL: https://github.com/onnx/sklearn-onnx
• Owner: onnx
• License: apache-2.0
• Created: 2018-12-18T20:18:48.000Z (almost 6 years ago)
• Default Branch: main
• Last Pushed: 2024-05-22T11:26:23.000Z (6 months ago)
• Last Synced: 2024-05-22T11:26:47.051Z (6 months ago)
• Topics: onnx, scikit-learn
• Language: Python
• Homepage:
• Size: 84.7 MB
• Stars: 510
• Watchers: 16
• Forks: 95
• Open Issues: 92
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOGS.md
  • License: LICENSE
  • Support: docs/supported.rst

Awesome Lists containing this project

README

        




[![Linux](https://github.com/onnx/sklearn-onnx/actions/workflows/linux-ci.yml/badge.svg)](https://github.com/onnx/sklearn-onnx/actions/workflows/linux-ci.yml)

[![Windows/Macos](https://github.com/onnx/sklearn-onnx/actions/workflows/windows-macos-ci.yml/badge.svg)](https://github.com/onnx/sklearn-onnx/actions/workflows/windows-macos-ci.yml)

[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)

## Introduction

*sklearn-onnx* converts [scikit-learn](https://scikit-learn.org/stable/) models to [ONNX](https://github.com/onnx/onnx).

Once in the ONNX format, you can use tools like [ONNX Runtime](https://github.com/Microsoft/onnxruntime) for high performance scoring.

All converters are tested with [onnxruntime](https://onnxruntime.ai/).

Any external converter can be registered to convert scikit-learn pipeline

including models or transformers coming from external libraries.

## Documentation

Full documentation including tutorials is available at [https://onnx.ai/sklearn-onnx/](https://onnx.ai/sklearn-onnx/).

[Supported scikit-learn Models](https://onnx.ai/sklearn-onnx/supported.html)

Last supported opset is 21.

You may also find answers in [existing issues](https://github.com/onnx/sklearn-onnx/issues?utf8=%E2%9C%93&q=is%3Aissue)

or submit a new one.

## Installation

You can install from [PyPi](https://pypi.org/project/skl2onnx/):

```

pip install skl2onnx

```

Or you can install from the source with the latest changes.

```

pip install git+https://github.com/onnx/sklearn-onnx.git

```

## Getting started

```python

# Train a model.

import numpy as np

from sklearn.datasets import load_iris

from sklearn.model_selection import train_test_split

from sklearn.ensemble import RandomForestClassifier

iris = load_iris()

X, y = iris.data, iris.target

X = X.astype(np.float32)

X_train, X_test, y_train, y_test = train_test_split(X, y)

clr = RandomForestClassifier()

clr.fit(X_train, y_train)

# Convert into ONNX format.

from skl2onnx import to_onnx

onx = to_onnx(clr, X[:1])

with open("rf_iris.onnx", "wb") as f:

    f.write(onx.SerializeToString())

# Compute the prediction with onnxruntime.

import onnxruntime as rt

sess = rt.InferenceSession("rf_iris.onnx", providers=["CPUExecutionProvider"])

input_name = sess.get_inputs()[0].name

label_name = sess.get_outputs()[0].name

pred_onx = sess.run([label_name], {input_name: X_test.astype(np.float32)})[0]

```

## Contribute

We welcome contributions in the form of feedback, ideas, or code.

## License

[Apache License v2.0](LICENSE)