cbasspy.jackknives package

Submodules

cbasspy.jackknives.jkexamples module

Module to generate example dictionary file and options file for use with jackknife code. You will likely want to call these each time you are starting a new set of jackknives.

generateDict(dstName='./jkdict_eg.py')

Script to generate an example dictionary file.

Parameters:

dstName : str, optional {‘./jkdict_eg.py’}

Name to save example file as

generateOptions(dstName='./jkoptions_eg.ini')

Script to generate an example ini file.

Parameters:

dstName : str, optional {‘./jkoptions_eg.ini’}

Name to save example file as

cbasspy.jackknives.jkgeneral module

This module does most of the work when running jackknife tests. Typically you will call cbasspy.jackknives.jkgeneral.jackknife() or cbasspy.jackknives.jkgeneral.run_jackknife_glam().

calc_chisq(cov1, cov2, map1, map2, extraMask=None, lotsOfOutputs=False)

Given two covariance matrices and two maps return the reduced chi squared value and dof.

Parameters:

cov1 : array_like

cov2 : array_like

map1 : array_like

map2 : array_like

extraMask : array_like, optional {None}

Extra mask of points to ignore. Pixels set to 1 in mask will be used.

lotsOfOutputs : optional {False,True}

if True outputs chi^2_red(I) and chi^2_red(P) separately

Returns:

reduced_chisq : float

dof : float

chisqGainFactor(whichPol, map1, map2, cov1, cov2, reverse=False, brightFrac=0.01)

jackknife(maps, JKtitle, whichJK, jk_options, jkrun)

Do a set of jackknives

This is the main jackknife function.

Parameters:

maps : array_like

list of jkm.CBASSmap objects

JKtitle : str

name of jackknife

whichJK : array_like

list of pairs of maps to compare, eg [(0,1),(0,2),(1,2)]

jk_options : ‘str’

.ini file with jackknife options. This determines the outputs

See also

run_jackknife_glam

Run on Glamdring queue.

map_stats(map_masked)

Calculate mean and rms of IQU masked map

Parameters:

map_masked : array_like

Input I,Q,U map

Returns:

meanI : float

meanQ : float

meanU : float

rmsI : float

rmsQ : float

rmsU : float

plotAlmTriangle(wdiff, fig, ax, doJustNoise=False, doCbar=False)

Plot a triangle of alms.

Parameters:

wdiff : array_like

HealPix weighted difference map

fig : matplotlib figure

ax : matplotlib ax

doJustNoise : bool, False

Whether to do a noise simulation

doCbar : bool, False

Whether to plot a colorbar

Returns:

image : array_like

Image of alms

plotAlmZoom(image_wdiff, titleStr='')

Plot a zoom in on ell,m=0,0

Parameters:

image_wdiff : array_like

Image of weighted difference alms

titleStr : str

Title for plot

plotDataAndSimNoise(wdiff, titleStr='')

Generate a figure containing a pair of traingle plots of alms. The left hand plot is the data. The right hand plot is a realisation of noise with the same std.

Parameters:

wdiff : array_like

HealPix weighted difference map

titleStr : str

Title of plot

plotPowerSpec(maskName, skyClName, weightName, sigma_0, numWhiteSims, outDir, diffMapName, fig_title, outName, spiceParams, ubin=20, pngDpi=300)

plotPowerSpecFactorOf2(maskName, numWhiteSims, outDir, diff2MapName, sum2MapName, noiseOver2, outName, spiceParams, ubin=20, pngDpi=300)

Parameters:

maskName: name of sky mask file

nuWhiteSims: Number of realization of white noise to create

outDir: Output directory

diff2MapName: Name of difference map

sum2MapName: Name of sum map

outName: Name of output figure

spiceParams: Spice param file

ubin: width of ell bins

noiseOver2: Noise map

plotPowerSpecFactorOf2PlaneMask(maskNames, numWhiteSims, outDir, diff2MapName, sum2MapName, noiseOver2, outName, spiceParams, ubin=20, pngDpi=300)

Parameters:

maskNames: namea of sky mask files, [mask10,mask20,mask30,mask40,mask50,mask60]]

nuWhiteSims: Number of realization of white noise to create

outDir: Output directory

diff2MapName: Name of difference map

sum2MapName: Name of sum map

outName: Name of output figure

spiceParams: Spice param file

ubin: width of ell bins

noiseOver2: Noise map

plotPte(wdiff, mMin=5, ellMin=50, titleStr='')

Plot histograms of the PTE values for the a weighted difference map alms. The PTE values are calcualted by comaring the distribution to the equivalent Gaussian.

They are also plotted excluding the region below mMin and ellMin

Parameters:

wdiff : array_like

Weighted difference map

mMin : int

Minimum m value for restricted region

ellMin : int

Minimum ell value for restricted region

titleStr : str

Title of figure

powerSpecResids(jkParams, PS_jk, PS_sky, PS_noise)

jkParams[0] = multiplicative factor for sky jkParams[1] = multiplicative factor for noise

powerSpecResids2(jkParams, PS_diff2, PS_whiteNoise_mu, ell)

jkParams[0] : intercept jkParams[1] : slope

produce_plane_masks(NSIDE, outDir, rotTo, cutOff)

Produce a mask of the galactic plane

eg, NSIDE=512, outDir=outDir, rotTo=’c’, cutOff=10. i.e. cutOff in degrees from galactic plane.

returnPowerSpecFactorOf2(maskName, numWhiteSims, outDir, diff2MapName, sum2MapName, noiseOver2, outName, spiceParams, ubin=20, pngDpi=300, apodizesigma=90, thetamax=90)

run_jackknife_glam(maps, JKtitle, whichJK, jk_options, jkrun)

Do a set of jackknives on glamdring.

Inputs are strings, eg:

• maps = ‘[allGood_EL37_S30,allGood_EL47_S30,allGood_EL67_S30,allGood_EL77_S30]’
• JKtitle = ‘elevation’
• whichJK = ‘list(itertools.combinations(range(0,len(maps)), 2)) # Every combination’
• jk_options = ‘jk_options.ini’
• jk_run = ‘jk_run23.py’ (for example)

The example above has set whichJK to every combination of maps Could instead specify just the few you want to compare, eg: > whichJK = ‘([0,1],[2,3])’

See also

jackknife

If you don’t want to run on Glamdring

save_diff_map(map, mapTitle, mapName, doGnomic=False, doCart=False, unitFontSize=15, kwargs={}, pngDpi=300)

Save a map as either gnomic or mollview image

Parameters:

map : array_like

map to save

mapTitle : str

Title in image

mapName : str

File name and extension

doGnomic : optional, {False,True}

Whether to do a gnomic projection

doCart : optional, {False,True}

Whether to do a cartesian projection

kwargs : *

Dictionary to pass to moll/gnom view

subtract_masked_maps(map1_masked, map2_masked, weights=[], divide_by_two=False)

Subtracts two maps from each other.

If weights are provided then calculate weighted diff

Parameters:

map1_masked : masked array_like

I,Q,U map

map2_masked : masked array_like

I,Q,U map

weights : array_like, optional

weights map

divide_by_two : boolean, optional {False}

If True, divide the difference map by two

Returns:

diffMap : array_like

(weighted) difference map

sum_masked_maps(map1_masked, map2_masked, divide_by_two=False)

Sums two maps

Parameters:

map1_masked : masked array_like

I,Q,U map

map2_masked : masked array_like

I,Q,U map

divide_by_two : boolean, optional {False}

If True, divide the sum map by two

cbasspy.jackknives.jkmaps module

Map objects, used by jackknife codes.

class CBASSmap(cutName, fileListName, paramName, dir, coord, jkd)

Bases: object

Maps objects, used by the jackknife code

When creating a CBASSmap object, you need to define the following attributes,

• cutName
• fileListName
• paramName
• dir
• coord
• jkd

The rest are calculated by various methods.

Attributes

Name	(str) map name has format [cutName]_[fileListName]_[paramName]_[coord]
cutName	(str) name of cut on the data
cut	(str) cbass.split_data command to obtain the cut
fileListName	(str) short name of .txt file list
fileList	(str) full file name
fileListDir	(str) directory to find .txt file
paramName	(str) short name of .ini parameter file
param	(str) full parameter file name
paramDir	(str) directory to find .ini file
coord	(str) coordinate system of map, eg ‘c’, ‘g’
jkd	(dictionary) dictionary containing all the information, see cbasspy.jackknives.jkexamples.generateDict() and how to run jackknives doc.
mapDir	(str) where to find the map and cov .fits files
todDir	(str) where to find the tod .fits files

Methods

make_map()

Tell glamdring to make the map

mv_map()

Tell glamdring to move the map to mapDir

read_cov()

Read in covariance matrix

Returns:

cov : array_like

II,QQ,UU,IQ,IU,QU covariance maps.

read_hits()

Read in the hits map

Returns:

map : array_like

H map

read_map()

Read in a map

Returns:

map : array_like

I, Q, U map.

split_data()

Tell Glamdring to perform the cut on the data on files in fileList.

DESCART = '~/CBASS/cbass_analysis/descart/map_making/descart_cbass'

str: Descart function

cbasspy.jackknives.scale_two_maps module

cbasspy.jackknives.web_publish module