futility 0.3.4

various utilities for unpacking and analyzing .fits files

ok so here's the rundown on all this stuff:

unpack_folder.py: - can unpack and relocate an entire folder of files compressed with fpack - deals with all the necessary renaming as well as deletes the compressed versions - usage: python unpack_folder.py infolderpath outfolderpath - if no outfolderpath is given, it will unpack them in infolder - paths can be with or without slash at the end

noise_info.py - analyzes background noise in chunks of arbitrary size(default=15 pixels) - returns a median/mean pixel value, matrix of each chunk's mean/median and a matrix of their standard deviations - can make plot of means/medians vs stds, and 3d plot of each chunk's mean/median - has a rather extensive ui including multiple, gradually increasing in computational overkill, methods of file searching for the lazy - will also automatically display when run in X window Usage: python noise_info path/to/file

fim.py - does everything noise_info can - everything you need to operate/change it is in default.cfg, just change the values and rerun -

paths.py - stabilizes pathing accross accounts and machines - please add and use it as much as possible - but also dont change anything already there - gives you access to all path names pre-stored under variables

difference.py - this was a bid of a pipedream longshot from the start - just finds difference between fits files from SExtractor's perspective - good use of k-d tree if you wanna use the code for similar projects

flats_noise.py - i barely remember making this one - from when i was going insane trying to figure out how to denoise and renoise images - statistical nonsense

get_sources.py - SExtractor script, hardcoded to draw from my(Benny) own settings

gaussian.py - adds a pixel offset but radially symmetric, gaussian "star" at a randomized point around a specific point(typically galaxies) - the offset adds surprisingly accurate radial irregularities - uniformly randomizes sigma of gaussian to be between siga and sigb - uniformly randomizes amp of gaussian - also randomizes placement around specified point (done in radial coordinate and transformed back to cartesian) - used by gym_teacher.py

gym_teacher.py - named gym teacher because it comes up with "games" to train the ai - takes in a folder of unpacked fits files and injects randomized new stars into them around both the most galaxylike galaxies - if there arent enough "galaxy-like" galaxies, it starts to pick random spots to fill its ranks - for each file it records name, star placements, star sigma, star amp, and star array size for each file in folder under starcoords.xml