[JOSS] Installation instructions

ritagnunes / PulseqDiffusion

GNU Affero General Public License v3.0

9 stars 3 forks source link

[JOSS] Installation instructions #4

Open mathieuboudreau opened 4 years ago

mathieuboudreau commented 4 years ago

Hi,

The installation instructions on the README are somewhate lacking for me to be able to get up and running to use the software. https://github.com/ritagnunes/PulseqDiffusion/blob/master/README.md

I have MATLAB installed.
I created a venv with python 3.6. I downloaded this github repository (even though those aren't in the installation instructions). I installed pip install pypulseq in it after activating the venv.
I installed the NIfTI reader you recommended in Ref 7 of your README and placed it in a folder called 'nifti_tools'.
I have FSL installed.

Now what am I supposed to do? Is it installed correctly now that I've done those steps?

ritagnunes commented 4 years ago

Yes, hopefully the Usage instructions that have been added to README.md in a recent commit will help to make this clearer.

grlee77 commented 4 years ago

Yes, it seems this is not a traditional software library with an installation process, but is instead a series of Python and Matlab scripts.

The Python scripts generate the sequence files that would be used on the scanner and the Matlab scripts are for reconstruction, generating gradient tables, etc.

The Python code can be run as long as the user has installed pypulseq as indicated in the README. Then there is no actual installation step for PulseqDiff itself, one just navigates to the folder with the scripts and can run them directly from the shell via python write_se_dwi_epi_bValue.py, etc.

ritagnunes commented 4 years ago

At present we are using Matlab for reconstructing the images as this code spans from previous work, developed over time. In any case we would like to point out that as these scripts don't rely on specific toolboxes, users who do not have access to Matlab should be able to use Octave instead. We are planning to provide a Python version of the reconstruction code in a future release.