Closed henrikef closed 5 years ago
Merging #11 into master will decrease coverage by
0.56%
. The diff coverage is81.15%
.
@@ Coverage Diff @@
## master #11 +/- ##
==========================================
- Coverage 93.6% 93.03% -0.57%
==========================================
Files 38 39 +1
Lines 1454 1522 +68
==========================================
+ Hits 1361 1416 +55
- Misses 93 106 +13
This is awesome, thanks!
Please take care of at least the major problems found by Scrutinizer though. They are easy fixes (names of variables to change, and so on).
Done (I think).
I added a new
HealpixMapROI
class which supports arbitrarily shaped ROIs via healpix maps. The maps can be read in from a fits file or created in the analysis script and passed as a list.The user still has to supply a center position and radius for the model map, as for
HealpixConeROI
. A warning is printed if the ROI is not fully contained in the model 'cone'.In case the user-specified healpix map has a different NSIDE than the data maps,
hp.pixelfunc.ud_grade
is used to re-sample the ROI map to the same NSIDE as the data map.Some tests have been added to verify that the same pixels are selected by a cone ROI and by a map ROI with the same circular active region.
Non-circular ROIs are useful for example to cut out bright nearby sources or galactic diffuse emission.