This is an old revision of the document!
Installation
As FAST has been developed in Matlab, you can download its unencrypted .m and .fig files and run it directly from Matlab's command line. This option allows you to make direct modifications to FAST's code, necessary for some advanced capabilities. It also allows use of FAST's batch processing system. To use this version, please follow the instructions under Option 1.
If you do not have access to a Matlab license however, FAST is also available as a compiled, stand-alone application. Details for installing this version can be found under Option 2.
Option 1: Installing the Matlab-based version
Step 1: Downloading the source code
To download the latest version of FAST, please click here. You should now extract the zipped FAST folder you just downloaded to a known location. The resulting directory is referred to as your FAST code directory.
The latest release of FAST's source code is also available at Github. Please note however that this version does not contain the full version of Bio-Formats, and as such cannot be used to import datasets.
Step 2: Adding FAST to your path
You are now ready to lauch Matlab. To use FAST, you will need to ensure that Matlab can find your FAST code directory. If you only wish to use FAST for a single session, navigate to the FAST code directory in the 'Current folder' panel of the main Matlab window, right-click on the FAST code directory, and select 'Add to Path → Selected Folders and Subfolders':
For a more permanent addition of FAST to your path, simply type
savepath
into your command line after you have temporarily added it to your path as describe above.
Step 3: Adding toolboxes to Matlab
FAST currently relies, in part, on functions available in several toolboxes. If you do not already have these installed, you can find guidance on how to install them here.
Toolboxes FAST requires to run:
- Image processing toolbox
- Signal processing toolbox
- Curve fitting toolbox
- Statistics and machine learning toolbox
Work to remove these dependencies is currently underway, and should be completed in a future release. If you do not have access to licenses for these toolboxes, currently you will need to use the stand-alone version of FAST (see below).
Option 2: Installing the stand-alone version
The stand-alone version of FAST can be installed using the platform-specific installers available here. After this has finished downloading, open the installer to begin the installation process. Please follow the directions inside the installer to complete the process.
Please note that if you choose this option, FAST's output will need to be converted from .mat files to structures appropriate for the platform of your choice. Functions for performing this conversion can be found in the post-processing toolbox.
Platform specific instructions
MacOS
When trying to start the installation process, MacOS will warn you that the app was developed by an unidentified developer and block its installation. To override this, please follow the instructions provided here:
Linux
Once you have completed the installation process, you will have generated two new folders, one corresponding to your installation of FAST (the location of which will be denoted as \Path\To\FAST\) and one to your installation of the Matlab runtime (\Path\To\Matlab_runtime\). To run FAST, simply type the following into the terminal:
\Path\To\FAST\application\run_FAST.sh \Path\To\Matlab_runtime\vXX\
where vXX denotes the version of the Matlab runtime installed in your runtime directory (e.g. 'v95').
List of FAST features not compatible with stand-alone versions
Unfortunately, not all parts of FAST are currently supported in the stand-alone versions of the software. To make use of these capabilities, you will instead need to use FAST directly in Matlab (option 1):