Closed AndrewDDavis closed 5 years ago
The discrepancy can be attributed to non deterministic registration by ANTs. I wouldn't be worried about this since meanFA changed in the right direction and came closer to that of reference site.
ITKv5 and ANTs 2.3.0 allows deterministic registration. Utilizing that, we shall come up with concrete numbers on a future release.
OK, that's good to hear -- thanks for the explanataion.
I ran the
pipeline_test.sh
, and it did indeed finish in an hour with 16 cores and 8 GB of RAM. The readme says the result should have been this:However, my result values were different:
Is this difference expected?