djl

Model Zoo

Deep Java Library’s (DJL) Model Zoo is more than a collection of pre-trained models. It’s a bridge between a model vendor and a consumer. It provides a framework for developers to create and publish their own models.

A ZooModel has the following characteristics:

Basic model zoo

We provide engine agnostic ZooModels in basic model zoo package. They can be used on any DJL backend engine.

MXNet symbolic model zoo

Apache MXNet has a large number of existing pre-trained models. We created an Apache MXNet model zoo to make it easy for users to consume them.

PyTorch model zoo

We created a PyTorch model zoo to make it easy for users to consume them.

TensorFlow model zoo

We created an TensorFlow model zoo to make it easy for users to consume them.

Publish your own model to the model zoo

You can create your own model in the model zoo so customers can easily consume it. For more information, see Add a new Model to the model zoo .

Load models from ModelZoo

See: How to load model