Support CombineDataFlag of 0 when averaging rotated bvecs #202
Conversation
| @@ -114,7 +114,7 @@ else # Combining across diffusion directions with opposing phase-encoding polari | |||
|
|
|||
| # Average Eddy-Rotated bvecs. Get for each direction the two b matrices, average those and then eigendecompose the average b-matrix to get the new bvec and bval. | |||
| # Also outputs an index file (1-based) with the indices of the input (Pos/Neg) volumes that have been retained in the output | |||
| ${globalscriptsdir}/average_bvecs.py ${eddydir}/Pos.bval ${eddydir}/Pos_rotated.bvec ${eddydir}/Neg.bval ${eddydir}/Neg_rotated.bvec ${datadir}/avg_data ${eddydir}/Pos_SeriesVolNum.txt ${eddydir}/Neg_SeriesVolNum.txt | |||
| ${globalscriptsdir}/average_bvecs.py ${eddydir}/Pos.bval ${eddydir}/Pos_rotated.bvec ${eddydir}/Neg.bval ${eddydir}/Neg_rotated.bvec ${datadir}/avg_data ${CombineDataFlag} ${eddydir}/Pos_SeriesVolNum.txt ${eddydir}/Neg_SeriesVolNum.txt | |||
There was a problem hiding this comment.
Have you tested that this works? Because based on the definition of main() in average_bvecs.py, it doesn't appear to me that we've properly provided inputs for bvalsoutfile, bvecsoutfile, and indicesoutfile.
There was a problem hiding this comment.
Yeah, I've tested running this. The usage for the updated average_bvecs.py is:
usage: average_bvecs bvals1 bvecs1 bvals2 bvecs2 output_basename combine_data_flag [overlap1 overlap2]
This matches the way I run the script. bvalsoutfile, bvecsoutfile, and indicesoutfiles are defined based on the output_basename
| @@ -27,6 +27,7 @@ def main(bvals1file, | |||
| bvalsoutfile, | |||
| bvecsoutfile, | |||
| indicesoutfile, | |||
| only_combine, | |||
There was a problem hiding this comment.
I'm confused -- this code block here does not appear consistent with usage: average_bvecs bvals1 bvecs1 bvals2 bvecs2 output_basename combine_data_flag [overlap1 overlap2]
There was a problem hiding this comment.
That's because python scripts don't start by calling the main function. The first code to be run is at the very end of the script (after the if __name__ == "__main__": on line 293). The code after that actually parses the arguments from the command line and defines bvalsoutfile, bvecsoutfile, and indicesoutfiles. Finally it calls the main function to do the actual work.
There was a problem hiding this comment.
I see. Thanks for the tutorial.
|
So, all testing is done, and this is ready to merge into #199 and from there into the master? |
|
I'm just rerunning the post_eddy script for the complete dataset (i.e., where we have the same amount of data for both phase encoding). I'm not expecting these changes to affect that output, but it would be good to check. I should know by tomorrow. |
|
The testing is done; this can be merged. |
It seems like this was never actually supported