The authors are not responsible for any implications that stem from the use of this software!

We recommend to use virtualenv to install the package and all dependencies. In order to create and activate a new virtual environment, please execute:

$ virtualenv -p python3 myenv
$ source myenv/bin/activate

The package is available on PyPI, but can also be installed directly from the source code. Note that OpenCL (version 1.2 or higher) must be installed on the machine.

Installation via PyPI

To install the package via PyPI, type:

$ pip install bfast

Installation From Sources

To install the package from the sources, first get the current stable release via:

$ git clone

Subsequently, install the package locally via:

$ cd bfast
$ python install --user

or, globally for all users, via:

$ sudo python install

In case you would like to extend the package, type:

$ python develop

Google Colab Installation

A simple way to run the massively-parallel implementation using powerful GPUs is to resort to Google Colab. To install and run the code, you can proceed as follows: Start a new notebook and change the runtime to GPU (Runtime->Change runtime type). Afterwards, type in:

$ !pip install bfast

Execute the cell (hit Shift+Return). You might have to restart the runtime by clicking on the button that appeared at the bottom for the cell (Restart Runtime). Afterwards, you can follow the instructions provided in Getting Started.


The bfast package requires Python 3.*. Furthermore, OpenCL (version 1.2 or higher) has to be installed on the system. The installation of OpenCL depends on the particular system, see, e.g.,

We refer to Andreas Klöckner’s wiki page for an excellent description of the OpenCL installation process on Linux-based systems. OpenCL is installed on macOS. For Windows, we refer to this blog post.

The bfast package depends on the following Python packages:

  • numpy==1.16.3

  • pandas==0.24.2

  • pyopencl==2018.2.5

  • scikit-learn==0.20.3

  • scipy==1.2.1

  • matplotlib==2.2.2

  • wget==3.2

  • Sphinx==2.2.0

  • sphinx-bootstrap-theme==0.7.1