Closed gkiar closed 9 years ago
@ebridge2 report the distribution of normalized errors, within scans (but across runs), vs. across scans (but within subjects). assuming it is good, we are done.
believe he gave us this in gitter - pipeline was deterministic across runs.
do this first.