changing agenda for data processing module

ReproNim / sfn2018-training

Materials for SfN 2018 training event

http://repronim.org/sfn2018-training

2 stars 5 forks source link

changing agenda for data processing module #5

Closed djarecka closed 5 years ago

yarikoptic commented 5 years ago

Dear @djarecka could you provide description on what was actually done and what is the intention behind the changes? "small edit" (instead of "changed time allocation") gave me 0 bytes of information as a commit message. Then I see probably a big piece extracted as a separate document -- was it just extracted or also edited? (if both, ideally should have been two commits)

djarecka commented 5 years ago

the intention is to change the agenda as @satra asked. i will let him to read it first and commented. we might have to change more

On Tue, Oct 23, 2018, 18:27 Yaroslav Halchenko notifications@github.com wrote:

Dear @djarecka https://github.com/djarecka could you provide description on what was actually done and what is the intention behind the changes? "small edit" (instead of "changed time allocation") gave me 0 bytes of information as a commit message. Then I see probably a big piece extracted as a separate document -- was it just extracted or also edited? (if both, ideally should have been two commits)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ReproNim/sfn2018-training/pull/5#issuecomment-432444592, or mute the thread https://github.com/notifications/unsubscribe-auth/ABlfGnXL70DKgTSIVZ5w8s_Ksqbt5ka0ks5un5fLgaJpZM4X2vz5 .

djarecka commented 5 years ago

@yarikoptic - also I haven't edited Michael's lesson, just divide it into 2 parts.

yarikoptic commented 5 years ago

Let's decide on the split ASAP and commit it with proper commit messages (for our own sake later on to see what, why and how was changed without guessing from the diff). There is outstanding #7 already to be merged and it could cause conflicts, and I am about to add a "fix" for #11

yarikoptic commented 5 years ago

or whatever -- 2 parts it is, I will merge and adjust PRs and text accordingly - if it was just split, we need a note that the 2nd part follows the first and requires all the setup etc to be done as was done there - that extracted submodule is not "independent".

djarecka commented 5 years ago

I thought that we will wait for Satra's comments, but we can always revert

yarikoptic commented 5 years ago

also I would remove a pretty much blank _episodes/04-04-reprozip.md one for this

shortage of time
the datalad rerun at the end of the GLM does complete rerun - reproduces the analysis. Since containers are used all the way, nothing really to do more with reprozip. Reprozip is great to generate such containers, and could/should eventually be a part earlier when talking about environments IMHO.

yarikoptic commented 5 years ago

and you do have it there in http://www.reproducibleimaging.org/sfn2018-training/presentations/processing/#17 which is the first section of the workflows! so let's just stay linear, you @djarecka present it in that opening section of workflows. Everything stays kosher, no need for non-linear jumps at the end into that IMHO

djarecka commented 5 years ago

@yarikoptic - I'm not planning to present this, I believe @satra will, I suggested this order, because that how it was during OHBM if you see slide 15th from the presentation, i.e. first intro, later Michael, and reprozip at the end.

But whatever works better, I will let you and Satra decide