We have received multiple feedbacks about the current tutorial being somehow hard to undersand.
In addition it has some minor mistakes (k_to_flux and flux_to_k are swapped), it currently fails to execute (see #286) and defaults to standard likelihood instead of standard_matrix without any discussion on what is appropriate where (see #14).
I have since try to address these points and I am ready to update the tutorial.
There are however a several situations where we need to over-explain things just because the way things are in flarestack is counter intuitive or still far from optimal:
[ ] the name FLARESTACK_SCRATCH_DIR seems to imply that the path will be exclusively used by flarestack, but then everything is put under a flarestack__data subdirectory meaning FLARESTACK_SCRATCH_DIR should rather be a generic/shared scratch space;
[ ] there can be confusion between flarestack__data and FLARESTACK_DATASET_DIR;
[ ] importing a flarestack module will trigger the creation of a flarestack__data directory structure, I'd like my imports to be read-only operations;
[ ] there is some ambiguity in relation to what is a name, sometimes name implies a path (such in the case of analyses, catalogues...)
[ ] flarestack.shared.plot_output_dir is a function rather than a variable, so it should better named get_plot_output_dir();
[ ] flarestack.utils.prepare_catalogue.ps_catalogue_name has a rather obscure naming, something like create_catalogue would be more understandable.
We have received multiple feedbacks about the current tutorial being somehow hard to undersand.
In addition it has some minor mistakes (
k_to_fluxandflux_to_kare swapped), it currently fails to execute (see #286) and defaults tostandardlikelihood instead ofstandard_matrixwithout any discussion on what is appropriate where (see #14).I have since try to address these points and I am ready to update the tutorial.
There are however a several situations where we need to over-explain things just because the way things are in
flarestackis counter intuitive or still far from optimal:FLARESTACK_SCRATCH_DIRseems to imply that the path will be exclusively used byflarestack, but then everything is put under aflarestack__datasubdirectory meaningFLARESTACK_SCRATCH_DIRshould rather be a generic/shared scratch space;flarestack__dataandFLARESTACK_DATASET_DIR;flarestackmodule will trigger the creation of aflarestack__datadirectory structure, I'd like myimportsto be read-only operations;name, sometimesnameimplies a path (such in the case of analyses, catalogues...)flarestack.shared.plot_output_diris a function rather than a variable, so it should better namedget_plot_output_dir();flarestack.utils.prepare_catalogue.ps_catalogue_namehas a rather obscure naming, something likecreate_cataloguewould be more understandable.