Installation

This repository is tested on Python 3.6+, and Linux/Debian systems. It is recommended to install in a virtual environment to keep your system in order. Currently supported ML Frameworks are the following: torch>=1.1.0,<=1.8.0, tensorflow>=1.8.0,<=2.0.0, tensorflow.keras >= 2.2.0.

Install with pip using:

pip install sparseml

Supported Framework Versions

The currently supported framework versions are:

  • PyTorch-supported versions: >= 1.1.0, < 1.8.0

  • Keras-supported versions: 2.3.0-tf (through the TensorFlow 2.2 package; as of Feb 1st, 2021, keras2onnx has not been tested for TensorFlow >= 2.3).

  • TensorFlow V1-supported versions: >= 1.8.0 (TensorFlow >= 2.X is not currently supported)

Optional Dependencies

Additionally, optional dependencies can be installed based on the framework you are using.

PyTorch:

pip install sparseml[torch]

Keras:

pip install sparseml[tf_keras]

TensorFlow V1:

pip install sparseml[tf_v1]

TensorFlow V1 with GPU operations enabled:

pip install sparseml[tf_v1_gpu]

Depending on your device and CUDA version, you may need to install additional dependencies for using TensorFlow V1 with GPU operations.
You can find these steps here.

Note, TensorFlow V1 is no longer being built for newer operating systems such as Ubuntu 20.04. Therefore, SparseML with TensorFlow V1 is unsupported on these operating systems as well.