OSF refuses to host the 10m streamline tractogram. This means that, for instance, it is not possible to run the script that generates just the supplementary images, because they require that full tractogram as input. Currently I am adding a check to see if that file is present, and the supplementary material script won't run in its absence. However we may want to think about whether there's a better approach.
Eg. Maybe the main protocol script computes a reduced tractogram (& extracts the SIFT2 weights for those), and that then gets included in the derivatives download embedded in the Docker image, such that the supplementary material can be regenerated from only the contents of the image. Maybe there's a command-line option that triggers whether those supplementary calculations are performed using the full tractogram as recomputed locally, or only a reduced set.
OSF refuses to host the 10m streamline tractogram. This means that, for instance, it is not possible to run the script that generates just the supplementary images, because they require that full tractogram as input. Currently I am adding a check to see if that file is present, and the supplementary material script won't run in its absence. However we may want to think about whether there's a better approach.
Eg. Maybe the main protocol script computes a reduced tractogram (& extracts the SIFT2 weights for those), and that then gets included in the derivatives download embedded in the Docker image, such that the supplementary material can be regenerated from only the contents of the image. Maybe there's a command-line option that triggers whether those supplementary calculations are performed using the full tractogram as recomputed locally, or only a reduced set.
Not a huge fan of it, but it's an option.