Introduction

Core ML is an Apple framework to integrate machine learning models into your app.

Use the coremltools Python package to convert models from third-party training libraries such as TensorFlow and PyTorch to Core ML. You can then use Core ML to integrate the models into your app.

Core ML provides a unified representation for all models. Your app uses Core ML APIs and user data to make predictions, and to fine-tune models, all on the user’s device. Running a model strictly on the user’s device removes any need for a network connection, which helps keep the user’s data private and your app responsive.

Core ML optimizes on-device performance by leveraging the CPU, GPU, and Apple Neural Engine (ANE) while minimizing its memory footprint and power consumption.

Additional resources

The Machine Learning page provides educational material, tutorials, guides, and documentation for Apple developers.
The WWDC session videos are a great place to start if you are new to machine learning technology and Core ML.
The Core ML documentation walks you through the first steps in developing an app with a machine learning model.
Try out coremltools in your browser with Binder:

What is coremltools?

The coremltools Python package is the primary way to convert third-party models to Core ML.

With coremltools, you can do the following:

Convert trained models from libraries and frameworks such as TensorFlow and PyTorch to the Core ML Model Package.
Read, write, and optimize Core ML models.
Verify conversion/creation in macOS by making predictions using Core ML.

Supported libraries and frameworks

You can convert trained models from the following libraries and frameworks to Core ML:

Model Family	Supported Packages
Neural Networks	TensorFlow 1 (1.14.0+) TensorFlow 2 (2.1.0+) PyTorch (1.4.0+)
Tree Ensembles	XGboost (1.1.0) scikit-learn (0.18.1)
Generalized Linear Models	scikit-learn (0.18.1)
Support Vector Machines	LIBSVM (3.22) scikit-learn (0.18.1)
Pipelines (pre- and post-processing)	scikit-learn (0.18.1)