search

Mouse-Imaging-Centre / RMINC

Statistics for MINC volumes: A library to integrate voxel-based statistics for MINC volumes into the R environment. Supports getting and writing of MINC volumes, running voxel-wise linear models, correlations, etc.; correcting for multiple comparisons using the False Discovery Rate, and more. With contributions from Jason Lerch, Chris Hammill, Jim Nikelski and Matthijs van Eede. Some additional information can be found here:
https://mouse-imaging-centre.github.io/RMINC
Other
22 stars 17 forks source link

mincLogLikRatioParametricBootstrap doesn't handle vertex mask properly: #229

Closed gdevenyi closed 5 years ago

gdevenyi commented 5 years ago 
mincLogLikRatioParametricBootstrap(loglik1)

Error in minc.dimensions.sizes(filename) :
 Error opening input file: /opt/quarantine/resources/CIVET-CC-mask.txt.
/opt/quarantine/minc-toolkit/1.9.16/src/minc-toolkit-v2/libminc/libsrc/netcdf_convenience.c:573 (from miopen): Unable to open file '/opt/quarantine/resources/CIVET-CC-mask.txt'
/opt/quarantine/resources/CIVET-CC-mask.txt: No such file or directory
/opt/quarantine/minc-toolkit/1.9.16/src/minc-toolkit-v2/libminc/libsrc2/volume.c:1386 (from miopen): Unable to open file '/opt/quarantine/resources/CIVET-CC-mask.txt'
gdevenyi commented 5 years ago

Ideally, there probably should be a vertexLogLikRatioParametricBootstrap and anatLogLikRatioParametricBootstrap

cfhammill commented 5 years ago

I'm hesitant to invest time in this, although should be pretty quick. My sincere hope is that we won't be using that for model comparison in the future, instead using one of the approximate cross-validation techniques. Although I wonder if these bootstrap samples could be used to compute wAIC/PSIS-loo instead...

gdevenyi commented 5 years ago

Alternatively, compare_models could add support for *lmer objects...

Btw, mincFDR falls over as well on these objects, mostly because it thinks its a minc object when its not

cfhammill commented 5 years ago

Is that actually what you want? It is easier and more aligned with my aesthetics.

gdevenyi commented 5 years ago

Yes, I would gladly take that instead :)

gdevenyi commented 5 years ago

Only reason I'm using this is because its the only thing available for *lmer right now.

cfhammill commented 5 years ago

Ok, I'll try to do that this week.

gdevenyi commented 5 years ago

Closed in favour of #230

cfhammill commented 5 years ago

I'm going to reopen, it should be fixed at some point

gdevenyi commented 5 years ago

Ok, but rather than that, perhaps its time to start deprecating some functions in favour of others to reduce the maintenance load?