[RTM] ENH: Improve confound header names

[effigies]

See #360 (comment) and nipy/nipype#2136

Closes #360.

[effigies]

While that fixes it, there is an issue for datasets too short to produce confounds, but with non-steady-state outliers, such as ds005/sub-01/func/*run-01*.bold. Not sure how I missed this. There's a fix over at nipy/nipype#2138, which I've tested on that dataset locally.

I'll update the pins here, so that Circle will build it...

[oesteban]

Should I merge this already?

[effigies]

No. Need to merge into nipype and niworkflows first.

[oesteban]

Understood!

…

On Aug 1, 2017 16:00, "Chris Markiewicz" ***@***.***> wrote: No. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#634 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAkhxhHfxSe7bw2fV2byB2cD18ZyGqfyks5sT64pgaJpZM4OqAwA> .

[effigies]

Excerpted confounds.tsv:

tCompCor00	tCompCor05	aCompCor00	aCompCor05	NonSteadyStateOutlier00
0	0	0	0	1
-0.4371572005	0.0557365655	-0.3931056217	0.2711029978	0
0.0644078063	-0.5244772443	0.13139585	-0.0393245026	0
0.3763848735	0.2411832222	0.2838979808	-0.3508611558	0
0.3745648673	-0.0766074055	0.2799187512	-0.0620435332	0
0.1518852892	0.2507682694	0.3989339177	0.3213743293	0
-0.1056200975	0.1460221678	0.3054514479	0.2945456173	0
-0.0022437328	-0.1199075195	0.2638761357	-0.2511155202	0
0.3316998401	-0.2322500403	0.1232900596	-0.2069637969	0
0.1620694153	0.3295860188	-0.0552770858	0.1250341463	0
-0.1316354992	-0.1373579747	-0.1013527616	0.5480407703	0
-0.5190575626	0.0931088749	-0.3375762751	-0.1050603748	0
-0.2409135032	-0.0395117923	-0.2651946021	-0.1856786497	0
0.0516509609	0.3878416506	-0.2130329332	-0.0009426788	0
-0.0070883494	-0.4582866366	-0.1992427774	0.0164638213	0
-0.0689471074	0.084151844	-0.221982086	-0.3745714704	0

@oesteban @chrisfilo Do you want to have a look at the whole file to see if there are any other things to fix up while we're playing around here?

[effigies]

Just a pin update. Taking nipype merge as go-ahead to merge.