| | 21 | * Dual convolution (branches/pap) |
| | 22 | + Merged into trunk (rr25999,26000) |
| | 23 | + Ran on 4 diffs from ThreePi_SouthernRegion with DUAL=T, |
| | 24 | SPATIAL.ORDER=0. Some strange results for a small subset of skycells; |
| | 25 | otherwise good. |
| | 26 | + Some bad subtractions, clearly identifiable with chi^2^ > 100! One |
| | 27 | has 25453.6! Cannot fix merely by increasing ITER. |
| | 28 | + Softening errors produces nice behaviour --- fixed! |
| | 29 | + Some sources show up in the subtraction that should be masked --- |
| | 30 | fixed. |
| | 31 | + Happy with dual convolution again, pending testing on a few |
| | 32 | exposures. |
| | 33 | * Stacking (branches/pap): investigating rejection behaviour |
| | 34 | + Completed rework of combinePixels |
| | 35 | + On each rejection pass, throw out only the most variant pixel; |
| | 36 | change iterations parameter to iterations per input. |
| | 37 | + Happy with behaviour of stacking (r26011): |
| | 38 | - 4 inputs from stack_id = 14006 (the Draper stack) |
| | 39 | - stack_id = 25532 (full Draper stack from more recent warps) |
| | 40 | + Would benefit from threading the creation of the fake images |
| | 41 | * Moved TEMP.DIR from ppStack recipe to site configuration; updated |
| | 42 | production version |
| | 43 | |
