[TEST] Run 8x8x8 to check the whole pipeline with non 1D files

[garikoitz]

No description provided.

[garikoitz]

@noahbenson , right now if the synthetic json file is not passed the system fails. I started checking popeye but I guess the problem is in all tools. The problem I see with this is that some of the params for solving are coming from the json file, so if we use a real file, we'll need to pass those variables in the prfanalyze-tool-config.json file. I am guessing that we can assume this, but checking with you before doing anything:

• If there is no json file, it should assume that it is external data, and continue
• There will be missing configs, so if there is no json file, we'll nee to check that all the minimally required values are being passed to the function or fail (some defaults are not valid)
• Can external data be any size? (I mean 1D or volume) I think so, but let's discuss.
• The output files with the results will be the same as the input format, 1D or volume, but following the same naming convention

[noahbenson]

Hmm, I'm a bit confused:

• "synthetic json file" is the file output from prfSynth?
• when is that file not passed? (only when a default config file should be written out?)
• can you give me some examples of the variables we'd need to add to one of the prfanalyze-tool-config.json file?

[garikoitz]

On Mon, Jan 27, 2020 at 11:59 AM Noah C. Benson ***@***.***> wrote: Hmm, I'm a bit confused: - "synthetic json file" is the file output from prfSynth? >>>> Yes, the system should work even if this file it is not being passed,

so then the system should assume that it is real human participant BOLD time series is being passed right?

- when is that file not passed? (only when a default config file should be written out?) >>>> No, when prfsynth is not being used at all. We create manually a BIDS

directory with the Stimuli and the sub-subjectname/ses-sessionname and put there our own dataset (HCP?), and if we run prfanalyze, it will check the derivatives/prfsynth directory always looking for the json file, but it does not exists

- can you give me some examples of the variables we'd need to add to one of the prfanalyze-tool-config.json file? >>>> Most of them we can calculate (TR, time points, stim size in

pixels...), some they must be passed: Stimulus.ResizedHorz Stimulus.ResizedVert Stimulus.fieldofviewHorz Stimulus.fieldofviewVert Stimulus.spatialSampleHorz Stimulus.spatialSampleVert stimradius = pm1.Stimulus.fieldofviewHorz/2; Basically, if we know the fov (os stimDiameter) in degrees, could we calculate the rest? I need to consider many other things to generate the files, but for solving, not so much right? Would you add a flag in prfanalyze-config-defaults like syntheticDataset = true/false, so that we can ignore the prfsynth output?

…

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#44?email_source=notifications&email_token=ABCZAVYB4X2FEPXLQEFAVITQ744JVA5CNFSM4KKZMOSKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKA3FAI#issuecomment-578925185>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABCZAV6EUMSOLU2QQZX2BS3Q744JVANCNFSM4KKZMOSA> .