Add dynamic max_timeIDX_offset controlled by CSV and update harness templates

[wshlavacek]

Yoke training harness to allow dynamic control of the max_timeIDX_offset parameter using a CSV file. Previously, the parameter was hard-coded in the training script. See README.md for full details.

[galegozi]

Hi @wshlavacek ,
I did not do a full review, but I have a few things to say regarding the PR:

1. I like what you're doing here (making a parameter available to the user instead of having it fixed in file)!!!
2. I made a PR (Automating the slurm file generation #12 ) to automate the SLURM file generation, would it be possible to interface with that? (instead of having the SLURM files as part of the harness)? I would be more than happy to help you with this! What I mean by this is can we use the config.json instead of the SLURM files (the training_slurm.tmpl and the training_START.slurm)?
3. Can we wait until [WIP] START_study.py upgrade #34 is figured out? bc I have some concerns there that would effect this
   Thanks, and good work!

[wshlavacek]

@galegozi
#12 and #34 seem like improvements to me. Fewer files in a harness is better IMHO.

[hickmank]

@galegozi #12 and #34 seem like improvements to me. Fewer files in a harness is better IMHO.

@wshlavacek @galegozi it would be good to use the slurm json config functionality in PR#12. The other PR, modifying START_study.py, has too many details to work out for us to hold this work up for.

[hickmank]

A couple of minor changes requested in comments. It looks good. Let's meet to discuss the final study before we push "go".

[galegozi]

Hi @wshlavacek ,
I looked over your PR. I noticed some of your files were included in the harness. I think some of the files should be moved to core Yoke (not in the harness).
What do you think?

[hickmank]

The max_timeIDX_offset change looks fine along with the study that's been set up. I agree with @galegozi comments that some of the scripts you've included provide nice functionality generally so should live somewhere else. I think those should be included in a separate PR and removed from this PR. Open to dissenting opinions though :-)

[hickmank]

@wshlavacek where are we at on this? Let me know when I should re-review.

[wshlavacek]

I created a separate PR for the four helper scripts. I'd like to keep copies of the helper scripts here, for now. I can remove later.

[hickmank]

@wshlavacek can this PR be closed in favor of PR #62 ?