Footprint functionality

Here we show how to use the footprint functionality. It is not used directly in the matching, but can be applied on the recovery rates computation

%load_ext autoreload
%autoreload 2

import numpy as np
import pylab as plt
import healpy as hp

Random data example

For display purposes, we will generate some quick random data to be used with the footprints.

# For reproducibility
np.random.seed(1)

from support import gen_cluster
input1, input2 = gen_cluster(ra_min=0, ra_max=30, dec_min=9, dec_max=30)

Initial number of clusters (logM>12.48): 2,740
Clusters in catalog1: 835
Clusters in catalog2: 928

from clevar import ClCatalog
tags = dict(ra='RA', dec='DEC', z='Z', mass='MASS',
            mass_err='MASS_ERR', z_err='Z_ERR',
            radius='RADIUS_ARCMIN')

c1 = ClCatalog('Cat1', data=input1, tags=tags, radius_unit='arcmin')
c2 = ClCatalog('Cat2', data=input2, tags=tags, radius_unit='arcmin')

# Format for nice display
for c in ('ra', 'dec', 'z', 'z_err', 'radius'):
    c1[c].info.format = '.2f'
    c2[c].info.format = '.2f'
for c in ('mass', 'mass_err'):
    c1[c].info.format = '.2e'
    c2[c].info.format = '.2e'

/home/aguena/.local/lib/python3.9/site-packages/clevar-0.13.2-py3.9.egg/clevar/catalog.py:267: UserWarning: id column missing, additional one is being created.
  warnings.warn(

Check position in healpix

Check what are the pixels that contain the clusters

nside = 32
pixels1 = hp.ang2pix(nside, c1['ra'], c1['dec'], lonlat=True)
pixels2 = hp.ang2pix(nside, c2['ra'], c2['dec'], lonlat=True)

Plot to confirm selected pixels are correct

from matplotlib import cm
import copy
cmap = copy.copy(cm.jet)
cmap.set_under('.1')
gcol = lambda cmap, level: '#{:02x}{:02x}{:02x}{:02x}'.format(*cmap(level,bytes=True))

# Map with pixels of each catalog
map_ = np.zeros(hp.nside2npix(nside))
map_[pixels1] += 1
map_[pixels2] += 2
map_[map_==0] = np.nan

f = plt.figure(figsize=(10, 10))
hp.cartview(map_, hold=True, latra=[5, 35], lonra=[-5, 40], cmap=cmap, cbar=False, flip='geo')
ax = f.axes[0]
ax.axis('on')
ax.scatter(c1['ra'], c1['dec'], s=5, label='Cat 1 clusters')
ax.scatter(c2['ra'], c2['dec'], s=5, label='Cat 2 clusters')

ax.plot(0, 0, zorder=0, color=gcol(cmap, 0.0), label='Footptint - Cat1 only')
ax.plot(0, 0, zorder=0, color=gcol(cmap, 0.5), label='Footptint - Cat2 only')
ax.plot(0, 0, zorder=0, color=gcol(cmap, 1.0), label='Footptint - BOTH')
ax.legend(loc=3)

/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:202: MatplotlibDeprecationWarning: Passing parameters norm and vmin/vmax simultaneously is deprecated since 3.3 and will become an error two minor releases later. Please pass vmin/vmax directly to the norm when creating it.
  aximg = self.imshow(

<matplotlib.legend.Legend at 0x7f33ef5896a0>
../_images/footprint_10_2.png

The Footprint object

ClEvaR uses the Footprint object to handle operations related to spatial masking. It has the following internal attributes: - data: Table with main footprint data (ex: pixel, detfrac, zmax) - tags: Dictionary that tells which are the default columns to be used - size: Number of pixels in the catalog - pixel_dict: Dictionary of indicies given the object pixel

from clevar.footprint import Footprint

Adding external data to footprint

The input data for the footprint is the following:

  • nside (int): Heapix NSIDE

  • nest (bool): If ordering is nested (default=False)

  • pixel (array): Pixels inside the footprint

  • detfrac_vals (array, None): Detection fraction, if None is set to 1

  • zmax_vals (array, None): Zmax, if None is set to 99

Just like the ClCatalog object, data in Footorint can be added from columns, a dictionary or a table. First, let’s create some input data:

# Random values for detfrac and zmax for ftpt1
set_pixels1 = list(set(pixels1))
set_pixels2 = list(set(pixels2))
detfrac_rand = 0.9+.1*np.random.rand(len(set_pixels1))
z_rand = 0.5+.5*np.random.rand(len(set_pixels1))

From columns

To create a footprint fom columns, you have to pass the name as the initial argument and the data columns for the table as keyword arguments:

ftpt1 = Footprint(nside=nside, pixel=set_pixels1, detfrac=detfrac_rand, zmax=z_rand)
ftpt1[:5]
tags: pixel(pixel), detfrac(detfrac), zmax(zmax)
ClData length=5
pixeldetfraczmax
int64float64float64
46810.90633453300784980.5603701608159029
51900.9516470911191410.7147891228605607
46820.90205477755607740.9995254135867374
51920.93888154885962420.5046096565358194
51940.96735798334309960.7466031661083729

You can also pass a tags dictionary as input if you want your catalog to have names that are not default for ClEvaR use:

ftpt1 = Footprint(nside=nside, hpix=set_pixels1, df=detfrac_rand, zm=z_rand,
                 tags={'pixel': 'hpix', 'detfrac':'df', 'zmax': 'zm'})
ftpt1[:5]
tags: hpix(pixel), df(detfrac), zm(zmax)
ClData length=5
hpix (pixel)df (detfrac)zm (zmax)
int64float64float64
46810.90633453300784980.5603701608159029
51900.9516470911191410.7147891228605607
46820.90205477755607740.9995254135867374
51920.93888154885962420.5046096565358194
51940.96735798334309960.7466031661083729

From data table

You can also create a ClCatalog passing directly a full data table:

from astropy.table import Table
ap_table = Table([set_pixels1, detfrac_rand], names=['hpix', 'df'])
ftpt1 = Footprint(nside=32, data=ap_table, tags={'pixel': 'hpix', 'detfrac':'df'})
ftpt1[:3]
tags: hpix(pixel), df(detfrac), zmax(zmax)
ClData length=3
hpix (pixel)df (detfrac)zmax
int64float64float64
46810.906334533007849899.0
51900.95164709111914199.0
46820.902054777556077499.0

The data table can also be a dictionary or a numpy array with names:

# default colnames
print('default colnames:')
ftpt1 = Footprint(nside=32, data={'pixel': set_pixels1, 'detfrac': detfrac_rand})
display(ftpt1[:3])


print('different colnames:')
# different colnames
ftpt1 = Footprint(nside=32, data={'hpix': set_pixels1, 'df': detfrac_rand},
                  tags={'pixel': 'hpix', 'detfrac':'df'})
display(ftpt1[:3])

default colnames:
tags: pixel(pixel), detfrac(detfrac), zmax(zmax)
ClData length=3
pixeldetfraczmax
int64float64float64
46810.906334533007849899.0
51900.95164709111914199.0
46820.902054777556077499.0
different colnames:
tags: hpix(pixel), df(detfrac), zmax(zmax)
ClData length=3
hpix (pixel)df (detfrac)zmax
int64float64float64
46810.906334533007849899.0
51900.95164709111914199.0
46820.902054777556077499.0
# default colnames
print('default colnames:')
np_table = np.array(list(zip(set_pixels1, detfrac_rand)),
                    dtype=[('pixel', 'f4'), ('detfrac', 'f4')])
ftpt1 = Footprint(nside=32, data=np_table)
display(ftpt1[:3])


print('different colnames:')
# different colnames
np_table = np.array(list(zip(set_pixels1, detfrac_rand)),
                    dtype=[('hpix', 'f4'), ('df', 'f4')])
ftpt1 = Footprint(nside=32, data=np_table,
                  tags={'pixel': 'hpix', 'detfrac':'df'})
display(ftpt1[:3])

default colnames:
tags: pixel(pixel), detfrac(detfrac), zmax(zmax)
ClData length=3
pixeldetfraczmax
int64float32float64
46810.906334599.0
51900.951647199.0
46820.902054899.0
different colnames:
tags: hpix(pixel), df(detfrac), zmax(zmax)
ClData length=3
hpix (pixel)df (detfrac)zmax
int64float32float64
46810.906334599.0
51900.951647199.0
46820.902054899.0

Read the footprint from fits file

To read the footprint, the arguments are:

  • filename (str): Name of .fits catalog

  • nside (int): Heapix NSIDE

  • tags (dict): Dictionary with the tags and column names. It must contain

    • pixel (str): Name of pixels column inside the footprint

    • detfrac (str): Name of detection fraction column, if None is set to 1

    • zmax (str): Name of Zmax column, if None is set to 99

  • nest (bool): If ordering is nested (default=False)

  • full (bool): read all columns of file

Let’s first create a file with the footprint info:

Footprint(nside=nside, pixel=set_pixels1, detfrac=detfrac_rand, zmax=z_rand).write('ftpt_temp.fits')

and then read it:

Footprint.read('ftpt_temp.fits', nside=nside,
               tags={'pixel': 'pixel',
                     'detfrac': 'detfrac',
                     'zmax': 'zmax'})[:3]
tags: pixel(pixel), detfrac(detfrac), zmax(zmax)
ClData length=3
pixeldetfraczmax
int64float64float64
46810.90633453300784980.5603701608159029
51900.9516470911191410.7147891228605607
46820.90205477755607740.9995254135867374
Footprint.read('ftpt_temp.fits', nside=nside,
               tags={'pixel': 'pixel'})[:3]
tags: pixel(pixel), detfrac(detfrac), zmax(zmax)
ClData length=3
pixeldetfraczmax
int64float64float64
46811.099.0
51901.099.0
46821.099.0
Footprint.read('ftpt_temp.fits', nside=nside,
               tags={'pixel': 'pixel'}, full=True)[:3]
tags: pixel(pixel), detfrac(detfrac), zmax(zmax)
ClData length=3
pixeldetfraczmax
int64float64float64
46810.90633453300784980.5603701608159029
51900.9516470911191410.7147891228605607
46820.90205477755607740.9995254135867374

Plotting the footprint

The footprints have an inbuilt function to plot their values

f = ftpt1.plot('detfrac', bad_val=np.nan, auto_lim=True)
f = ftpt1.plot('zmax', bad_val=np.nan, auto_lim=True)

/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:920: MatplotlibDeprecationWarning: You are modifying the state of a globally registered colormap. In future versions, you will not be able to modify a registered colormap in-place. To remove this warning, you can make a copy of the colormap first. cmap = copy.copy(mpl.cm.get_cmap("viridis"))
  newcm.set_over(newcm(1.0))
/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:921: MatplotlibDeprecationWarning: You are modifying the state of a globally registered colormap. In future versions, you will not be able to modify a registered colormap in-place. To remove this warning, you can make a copy of the colormap first. cmap = copy.copy(mpl.cm.get_cmap("viridis"))
  newcm.set_under(bgcolor)
/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:922: MatplotlibDeprecationWarning: You are modifying the state of a globally registered colormap. In future versions, you will not be able to modify a registered colormap in-place. To remove this warning, you can make a copy of the colormap first. cmap = copy.copy(mpl.cm.get_cmap("viridis"))
  newcm.set_bad(badcolor)
/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:202: MatplotlibDeprecationWarning: Passing parameters norm and vmin/vmax simultaneously is deprecated since 3.3 and will become an error two minor releases later. Please pass vmin/vmax directly to the norm when creating it.
  aximg = self.imshow(
../_images/footprint_33_1.png ../_images/footprint_33_2.png

Clusters can also be added to the plot with their actual angular size:

# increase cluster radius for display
c1['radius'] *= 10
f = ftpt1.plot('detfrac', bad_val=np.nan,
               ra_lim=[3, 8], dec_lim=[10, 15],
               cluster=c1)
# return original value
c1['radius'] /= 10
../_images/footprint_35_0.png

Use ClEvaR functions to create a footprint

Import create_footprint functions to create a footprint based on a cluster catalog. It can create a footprint based on cluster positions with a given NSIDE, or compute the best NSIDE based on a cluster density per pixel. It also can fill holes in the footprint.

from clevar.footprint import create_artificial_footprint

Fixed NSIDE:

ftpt1 = create_artificial_footprint(c1['ra'], c1['dec'], nside=64)

Footprint NSIDE: 64
Pixels in footprint: 506
Pixels in footprint: 506

NSIDE from density:

ftpt1 = create_artificial_footprint(c1['ra'], c1['dec'], nside=None, min_density=4)

NSIDE(2) -> 208.75 clusters per pixel
NSIDE(4) -> 139.16666666666666 clusters per pixel
NSIDE(8) -> 46.388888888888886 clusters per pixel
NSIDE(16) -> 14.910714285714286 clusters per pixel
NSIDE(32) -> 4.417989417989418 clusters per pixel
NSIDE(64) -> 1.650197628458498 clusters per pixel
Footprint NSIDE: 64
Pixels in footprint: 506
Pixels in footprint: 506

  • there is also an option to fill holes in this artificial footprint

ftpt1 = create_artificial_footprint(c1['ra'], c1['dec'], nside=64, neighbor_fill=5)

Footprint NSIDE: 64
Pixels in footprint: 506
### filling holes ###
 - filling
   size: 506 -> 673 (+167)
 - filling
   size: 673 -> 697 (+24)
 - filling
   size: 697 -> 699 (+2)
 - filling
   size: 699 -> 700 (+1)
 - filling
   size: 700 -> 700 (+0)
 Total Change:
   size: 506 -> 700 (+194)
### filled! ###
Pixels in footprint: 700

ftpt1['detfrac'] = np.random.rand(ftpt1.size)
f = ftpt1.plot('detfrac', np.nan,  latra=[5, 35], lonra=[-5, 40])

/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:920: MatplotlibDeprecationWarning: You are modifying the state of a globally registered colormap. In future versions, you will not be able to modify a registered colormap in-place. To remove this warning, you can make a copy of the colormap first. cmap = copy.copy(mpl.cm.get_cmap("viridis"))
  newcm.set_over(newcm(1.0))
/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:921: MatplotlibDeprecationWarning: You are modifying the state of a globally registered colormap. In future versions, you will not be able to modify a registered colormap in-place. To remove this warning, you can make a copy of the colormap first. cmap = copy.copy(mpl.cm.get_cmap("viridis"))
  newcm.set_under(bgcolor)
/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:922: MatplotlibDeprecationWarning: You are modifying the state of a globally registered colormap. In future versions, you will not be able to modify a registered colormap in-place. To remove this warning, you can make a copy of the colormap first. cmap = copy.copy(mpl.cm.get_cmap("viridis"))
  newcm.set_bad(badcolor)
/home/aguena/miniconda3/envs/clmmenv/lib/python3.9/site-packages/healpy/projaxes.py:202: MatplotlibDeprecationWarning: Passing parameters norm and vmin/vmax simultaneously is deprecated since 3.3 and will become an error two minor releases later. Please pass vmin/vmax directly to the norm when creating it.
  aximg = self.imshow(
../_images/footprint_44_1.png

Footprint masks

Add masks to clusters regarding the footprint. The ClCatalog object has has 3 functions related to the footprint: - add_ftpt_masks: info for cluster in footprint - add_ftpt_coverfrac: computes cover fraction - add_ftpt_coverfrac_nfw2D: computes cover fraction weighted by a project NFW profile

ftpt1 = Footprint(nside=nside, pixel=set_pixels1, detfrac=detfrac_rand, zmax=z_rand)
ftpt2 = Footprint(nside=nside, pixel=set_pixels2)
ftpt2['detfrac'][::2] = .5 # add effects to this footprint

%%time
c1.add_ftpt_masks(ftpt1, ftpt2)
c2.add_ftpt_masks(ftpt2, ftpt1)

## creating visibility mask ##
## creating visibility mask ##
## creating visibility mask ##
## creating visibility mask ##
CPU times: user 15.5 ms, sys: 31.8 ms, total: 47.2 ms
Wall time: 15.7 ms

display(c1[:4])
display(c2[:4])
Cat1
tags: id(id), RA(ra), DEC(dec), Z(z), MASS(mass), MASS_ERR(mass_err), Z_ERR(z_err), RADIUS_ARCMIN(radius)
Radius unit: arcmin
ClData length=4
idRA (ra)DEC (dec)Z (z)RADIUS_ARCMIN (radius)MASS (mass)MASS_ERR (mass_err)Z_ERR (z_err)mt_selfmt_othermt_multi_selfmt_multi_otherft_selfft_other
str3float64float64float64float64float64float64float64objectobjectobjectobjectboolbool
012.5823.601.530.941.00e+133.69e+120.08NoneNone[][]FalseTrue
16.1323.170.330.481.57e+134.29e+120.04NoneNone[][]TrueTrue
212.5210.820.780.841.49e+136.11e+120.07NoneNone[][]FalseTrue
34.2114.781.010.421.13e+134.71e+120.06NoneNone[][]FalseTrue
Cat2
tags: id(id), RA(ra), DEC(dec), Z(z), MASS(mass), MASS_ERR(mass_err), Z_ERR(z_err), RADIUS_ARCMIN(radius)
Radius unit: arcmin
ClData length=4
idRA (ra)DEC (dec)Z (z)RADIUS_ARCMIN (radius)MASS (mass)MASS_ERR (mass_err)Z_ERR (z_err)mt_selfmt_othermt_multi_selfmt_multi_otherft_selfft_other
str3float64float64float64float64float64float64float64objectobjectobjectobjectboolbool
012.5823.601.510.941.12e+133.69e+120.08NoneNone[][]TrueFalse
16.1323.170.330.482.35e+134.29e+120.04NoneNone[][]TrueTrue
212.5210.820.800.841.36e+136.11e+120.07NoneNone[][]TrueFalse
34.2114.780.990.421.71e+134.71e+120.06NoneNone[][]TrueFalse

Add coverfraction values based on the footprint. It needs a cosmology object.

from clevar.cosmology import AstroPyCosmology
cosmo = AstroPyCosmology()

%%time
c1.add_ftpt_coverfrac(ftpt2, 1, 'mpc', cosmo=cosmo, window='flat')
c1.add_ftpt_coverfrac(ftpt2, 1, 'mpc', cosmo=cosmo, window='nfw2D')
c2.add_ftpt_coverfrac(ftpt1, 1, 'mpc', cosmo=cosmo, window='nfw2D')

CPU times: user 8.31 s, sys: 19.5 ms, total: 8.33 s
Wall time: 8.29 s

display(c1[:4])
display(c2[:4])
Cat1
tags: id(id), RA(ra), DEC(dec), Z(z), MASS(mass), MASS_ERR(mass_err), Z_ERR(z_err), RADIUS_ARCMIN(radius)
Radius unit: arcmin
ClData length=4
idRA (ra)DEC (dec)Z (z)RADIUS_ARCMIN (radius)MASS (mass)MASS_ERR (mass_err)Z_ERR (z_err)mt_selfmt_othermt_multi_selfmt_multi_otherft_selfft_othercf_1_mpccf_nfw_1_mpc
str3float64float64float64float64float64float64float64objectobjectobjectobjectboolboolfloat64float64
012.5823.601.530.941.00e+133.69e+120.08NoneNone[][]FalseTrue0.83333333333333340.7741572794115797
16.1323.170.330.481.57e+134.29e+120.04NoneNone[][]TrueTrue0.50.5
212.5210.820.780.841.49e+136.11e+120.07NoneNone[][]FalseTrue0.750.7260951172194341
34.2114.781.010.421.13e+134.71e+120.06NoneNone[][]FalseTrue1.01.0
Cat2
tags: id(id), RA(ra), DEC(dec), Z(z), MASS(mass), MASS_ERR(mass_err), Z_ERR(z_err), RADIUS_ARCMIN(radius)
Radius unit: arcmin
ClData length=4
idRA (ra)DEC (dec)Z (z)RADIUS_ARCMIN (radius)MASS (mass)MASS_ERR (mass_err)Z_ERR (z_err)mt_selfmt_othermt_multi_selfmt_multi_otherft_selfft_othercf_nfw_1_mpc
str3float64float64float64float64float64float64float64objectobjectobjectobjectboolboolfloat64
012.5823.601.510.941.12e+133.69e+120.08NoneNone[][]TrueFalse0.0
16.1323.170.330.482.35e+134.29e+120.04NoneNone[][]TrueTrue0.9664374259818176
212.5210.820.800.841.36e+136.11e+120.07NoneNone[][]TrueFalse0.4350822741933005
34.2114.780.990.421.71e+134.71e+120.06NoneNone[][]TrueFalse0.0

Saving and loading footprint quantities

ClEvaR has internal functions to save and load these quantities into the catalog so you don’t have to compute them again:

c1.save_footprint_quantities('cat1_ft_quantities.fits', overwrite=True)
c1.load_footprint_quantities('cat1_ft_quantities.fits')

Application of footprint flags on recovery rate

The recovery rate of clusters should take into account the footprint of the catalogs. Regions without overlaps should not be taken into consideration. Here we show how this can be done with ClEvaR.

Match catalogs

Let’s match the catalogs to compute the recovery rate

from clevar.match import ProximityMatch

match_config = {
    'type': 'cross', # options are cross, cat1, cat2
    'which_radius': 'max', # Case of radius to be used, can be: cat1, cat2, min, max
    'preference': 'angular_proximity', # options are more_massive, angular_proximity or redshift_proximity
    'catalog1': {'delta_z':.2,
                'match_radius': '1 mpc'},
    'catalog2': {'delta_z':.2,
                'match_radius': '10 arcsec'}
}

%%time
mt = ProximityMatch()
mt.match_from_config(c1, c2, match_config, cosmo=cosmo)

## ClCatalog 1
## Prep mt_cols
* zmin|zmax from config value
* ang radius from set scale

## ClCatalog 2
## Prep mt_cols
* zmin|zmax from config value
* ang radius from set scale

## Multiple match (catalog 1)
Finding candidates (Cat1)
* 719/835 objects matched.

## Multiple match (catalog 2)
Finding candidates (Cat2)
* 721/928 objects matched.

## Finding unique matches of catalog 1
Unique Matches (Cat1)
* 719/835 objects matched.

## Finding unique matches of catalog 2
Unique Matches (Cat2)
* 720/928 objects matched.
Cross Matches (Cat1)
* 719/835 objects matched.
Cross Matches (Cat2)
* 719/928 objects matched.
CPU times: user 18.5 s, sys: 209 ms, total: 18.7 s
Wall time: 18.5 s

Recovery rate

Use pass the parameters mask (masks all clusters) or mask_unmatched (masks only unmatched clusters) to consider only specific clusters on the recovery rate. This way, you can exclude clusters outside the common regions from the consideration.

from clevar.match_metrics import recovery

zbins = np.linspace(0, 2, 11)
mbins = np.logspace(13, 14, 5)

Mask based on footprint overlap

f, axes = plt.subplots(1, 3, figsize=(15, 5))
recovery.plot(c1, 'cross', zbins, mbins, ax=axes[0], add_legend=False, shape='line')
recovery.plot(c1, 'cross', zbins, mbins, ax=axes[1], add_legend=False,
              mask=c1['ft_other'], shape='line')
recovery.plot(c1, 'cross', zbins, mbins, ax=axes[2],
              mask_unmatched=~c1['ft_other'], shape='line')
for ax in axes:
    ax.set_ylim(-.01, 1.05)
axes[0].text(1, 1.1, 'no mask')
axes[1].text(1, 1.1, 'mask all')
axes[2].text(1, 1.1, 'mask unmatched')
plt.show()
../_images/footprint_64_0.png

Mask based on coverfraction

f, axes = plt.subplots(1, 3, figsize=(15, 5))
recovery.plot(c1, 'cross', zbins, mbins, ax=axes[0], add_legend=False, shape='line')
recovery.plot(c1, 'cross', zbins, mbins, ax=axes[1], add_legend=False,
              mask=c1['cf_nfw_1_mpc']>.8, shape='line')
recovery.plot(c1, 'cross', zbins, mbins, ax=axes[2],
              mask_unmatched=c1['cf_nfw_1_mpc']<.8, shape='line')
for ax in axes:
    ax.set_ylim(-.01, 1.05)
axes[0].text(1, 1.1, 'no mask')
axes[1].text(1, 1.1, 'mask all')
axes[2].text(1, 1.1, 'mask unmatched')
plt.show()
../_images/footprint_66_0.png

You can check the exact numbers used on the 2D plots

f, axes = plt.subplots(1, 3, figsize=(20, 5))

recovery.plot2D(c1, 'cross', zbins, mbins, ax=axes[0],
                add_num=True, num_kwargs={'fontsize':12
                                         })
recovery.plot2D(c1, 'cross', zbins, mbins, ax=axes[1],
                add_num=True, num_kwargs={'fontsize':12},
               mask=c1['cf_nfw_1_mpc']>.8)
recovery.plot2D(c1, 'cross', zbins, mbins, ax=axes[2],
                add_num=True, num_kwargs={'fontsize':12},
               mask_unmatched=c1['cf_nfw_1_mpc']<.8)
axes[0].text(1, mbins[-1]*1.1,'no mask')
axes[1].text(1, mbins[-1]*1.1,'mask all')
axes[2].text(1, mbins[-1]*1.1,'mask unmatched')

plt.show()
../_images/footprint_68_0.png