Install TensorFlow MRI#
Requirements#
TensorFlow MRI should work in most Linux systems that meet the requirements for TensorFlow.
Warning
TensorFlow MRI is not yet available for Windows or macOS. Help us support them!.
TensorFlow compatibility#
Each TensorFlow MRI release is compiled against a specific version of TensorFlow. To ensure compatibility, it is recommended to install matching versions of TensorFlow and TensorFlow MRI according to the TensorFlow compatibility table.
Warning
Each TensorFlow MRI version aims to target and support the latest TensorFlow version only. A new version of TensorFlow MRI will be released shortly after each TensorFlow release. TensorFlow MRI versions that target older versions of TensorFlow will not generally receive any updates.
Set up your system#
You will need a working TensorFlow installation. Follow the TensorFlow installation instructions if you do not have one already.
Use a GPU#
If you need GPU support, we suggest that you use one of the TensorFlow Docker images. These come with a GPU-enabled TensorFlow installation and are the easiest way to run TensorFlow and TensorFlow MRI on your system.
$ docker pull tensorflow/tensorflow:latest-gpu
Alternatively, make sure you follow these instructions when setting up your system.
Download from PyPI#
TensorFlow MRI is available on the Python package index (PyPI) and can be
installed using the pip package manager:
$ pip install tensorflow-mri
Run in Google Colab#
To get started without installing anything on your system, you can use
Google Colab.
Simply create a new notebook and use pip to install TensorFlow MRI.
!pip install tensorflow-mri
The Colab environment is already configured to run TensorFlow and has GPU support.