mirror of
https://github.com/Richard-Sti/csiborgtools_public.git
synced 2025-06-08 09:51:12 +00:00
Add 2PCF calculation (#42)
* Add 2PCF calculation * Add 2PCF reader * Add tpcf scripts * Add random state * Fix marked typo * Fix marks and more randoms * Stop calculating projected * Add mean 2PCF * Remove pimax * Edit submit script * Update nb * Add corrfunc check
This commit is contained in:
Richard Stiskalek
GitHub
parent
0b743756ef
commit
1344fa40b6
9 changed files with 635 additions and 2179 deletions
|
|
@ -1,87 +0,0 @@
|
|||
# Copyright (C) 2023 Richard Stiskalek
|
||||
# This program is free software; you can redistribute it and/or modify it
|
||||
# under the terms of the GNU General Public License as published by the
|
||||
# Free Software Foundation; either version 3 of the License, or (at your
|
||||
# option) any later version.
|
||||
#
|
||||
# This program is distributed in the hope that it will be useful, but
|
||||
# WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
|
||||
# Public License for more details.
|
||||
#
|
||||
# You should have received a copy of the GNU General Public License along
|
||||
# with this program; if not, write to the Free Software Foundation, Inc.,
|
||||
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
||||
"""
|
||||
2PCF calculation.
|
||||
|
||||
NOTE: This is an old script that needs to be updated.
|
||||
"""
|
||||
import numpy
|
||||
from Corrfunc.mocks import DDtheta_mocks
|
||||
from Corrfunc.utils import convert_3d_counts_to_cf
|
||||
from .utils import (rvs_on_sphere, wrapRA)
|
||||
|
||||
|
||||
def sphere_angular_tpcf(bins, RA1, DEC1, RA2=None, DEC2=None, nthreads=1,
|
||||
Nmult=5, seed1=42, seed2=666):
|
||||
"""
|
||||
Calculate the angular two-point correlation function. The coordinates must
|
||||
be provided in degrees. With the right ascension and degrees being
|
||||
in range of :math:`[-180, 180]` and :math:`[-90, 90]` degrees.
|
||||
If `RA2` and `DEC2` are provided cross-correlates the first data set with
|
||||
the second. Creates a uniformly sampled randoms on the surface of a sphere
|
||||
of size `Nmult` times the corresponding number of data points. Uses the
|
||||
Landy-Szalay estimator.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
bins : 1-dimensional array
|
||||
Angular bins to calculate the angular twop-point correlation function.
|
||||
RA1 : 1-dimensional array
|
||||
Right ascension of the 1st data set, in degrees.
|
||||
DEC1 : 1-dimensional array
|
||||
Declination of the 1st data set, in degrees.
|
||||
RA2 : 1-dimensional array, optional
|
||||
Right ascension of the 2nd data set, in degrees.
|
||||
DEC2 : 1-dimensional array, optional
|
||||
Declination of the 2nd data set, in degrees.
|
||||
nthreads : int, optional
|
||||
Number of threads, by default 1.
|
||||
Nmult : int, optional
|
||||
Relative randoms size with respect to the data set. By default 5.
|
||||
seed1 : int, optional
|
||||
Seed to generate the first set of randoms.
|
||||
seed2 : int, optional
|
||||
Seed to generate the second set of randoms.
|
||||
|
||||
Returns
|
||||
-------
|
||||
cf : 1-dimensional array
|
||||
The angular 2-point correlation function.
|
||||
"""
|
||||
# If not provided calculate autocorrelation
|
||||
if RA2 is None:
|
||||
RA2 = RA1
|
||||
DEC2 = DEC1
|
||||
# Get the array sizes
|
||||
ND1 = RA1.size
|
||||
ND2 = RA2.size
|
||||
NR1 = ND1 * Nmult
|
||||
NR2 = ND2 * Nmult
|
||||
# Generate randoms. Note that these are over the sphere!
|
||||
randRA1, randDEC1 = rvs_on_sphere(NR1, indeg=True, random_state=seed1)
|
||||
randRA2, randDEC2 = rvs_on_sphere(NR2, indeg=True, random_state=seed2)
|
||||
# Wrap RA
|
||||
RA1 = wrapRA(numpy.copy(RA1), indeg=True)
|
||||
RA2 = wrapRA(numpy.copy(RA2), indeg=True)
|
||||
# Calculate pairs
|
||||
D1D2 = DDtheta_mocks(0, nthreads, bins, RA1, DEC1, RA2=RA2, DEC2=DEC2)
|
||||
D1R2 = DDtheta_mocks(0, nthreads, bins, RA1, DEC1,
|
||||
RA2=randRA2, DEC2=randDEC2)
|
||||
D2R1 = DDtheta_mocks(0, nthreads, bins, RA2, DEC2,
|
||||
RA2=randRA1, DEC2=randDEC1)
|
||||
R1R2 = DDtheta_mocks(0, nthreads, bins, randRA1, randDEC1,
|
||||
RA2=randRA2, DEC2=randDEC2)
|
||||
# Convert to the CF
|
||||
return convert_3d_counts_to_cf(ND1, ND2, NR1, NR2, D1D2, D1R2, D2R1, R1R2)
|
||||
|
|
@ -12,5 +12,11 @@
|
|||
# You should have received a copy of the GNU General Public License along
|
||||
# with this program; if not, write to the Free Software Foundation, Inc.,
|
||||
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
||||
from warnings import warn
|
||||
from .knn import kNN_CDF # noqa
|
||||
from .utils import (RVSinsphere, RVSinbox, RVSonsphere, BaseRVS, normalised_marks) # noqa
|
||||
try:
|
||||
import Corrfunc
|
||||
from .tpcf import Mock2PCF # noqa
|
||||
except ImportError:
|
||||
warn("`Corrfunc` not installed. 2PCF modules will not be available (`Mock2PCF`).") # noqa
|
||||
|
|
|
|||
68
csiborgtools/clustering/tpcf.py
Normal file
68
csiborgtools/clustering/tpcf.py
Normal file
|
|
@ -0,0 +1,68 @@
|
|||
# Copyright (C) 2023 Richard Stiskalek
|
||||
# This program is free software; you can redistribute it and/or modify it
|
||||
# under the terms of the GNU General Public License as published by the
|
||||
# Free Software Foundation; either version 3 of the License, or (at your
|
||||
# option) any later version.
|
||||
#
|
||||
# This program is distributed in the hope that it will be useful, but
|
||||
# WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
|
||||
# Public License for more details.
|
||||
#
|
||||
# You should have received a copy of the GNU General Public License along
|
||||
# with this program; if not, write to the Free Software Foundation, Inc.,
|
||||
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
||||
"""2PCF calculation."""
|
||||
import numpy
|
||||
from Corrfunc.theory.DD import DD
|
||||
from Corrfunc.utils import convert_3d_counts_to_cf
|
||||
from .utils import BaseRVS
|
||||
|
||||
|
||||
class Mock2PCF:
|
||||
"""
|
||||
Tool to calculate the 2PCF of a catalogue.
|
||||
"""
|
||||
def __call__(self, pos, rvs_gen, nrandom, bins, random_state=42):
|
||||
"""
|
||||
Calculate the 2PCF from 3D pair counts.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
pos : 2-dimensional array of shape `(ndata, 3)`
|
||||
Positions of the data.
|
||||
rvs_gen : :py:class:`csiborgtools.clustering.BaseRVS`
|
||||
Uniform RVS generator.
|
||||
nrandom : int
|
||||
Number of random points to generate.
|
||||
bins : 1-dimensional array of shape `(nbins,)`
|
||||
Separation bins.
|
||||
random_state : int, optional
|
||||
Random state for the RVS generator.
|
||||
|
||||
Returns
|
||||
-------
|
||||
rp : 1-dimensional array of shape `(nbins - 1,)`
|
||||
Projected separation where the auto-2PCF is evaluated.
|
||||
xi : 1-dimensional array of shape `(nbins - 1,)`
|
||||
The auto-2PCF.
|
||||
"""
|
||||
assert isinstance(rvs_gen, BaseRVS)
|
||||
pos = pos.astype(numpy.float64)
|
||||
rand_pos = rvs_gen(nrandom, random_state=random_state,
|
||||
dtype=numpy.float64)
|
||||
|
||||
dd = DD(autocorr=1, nthreads=1, binfile=bins,
|
||||
X1=pos[:, 0], Y1=pos[:, 1], Z1=pos[:, 2], periodic=False)
|
||||
dr = DD(autocorr=0, nthreads=1, binfile=bins,
|
||||
X1=pos[:, 0], Y1=pos[:, 1], Z1=pos[:, 2],
|
||||
X2=rand_pos[:, 0], Y2=rand_pos[:, 1], Z2=rand_pos[:, 2],
|
||||
periodic=False)
|
||||
rr = DD(autocorr=1, nthreads=1, binfile=bins,
|
||||
X1=rand_pos[:, 0], Y1=rand_pos[:, 1], Z1=rand_pos[:, 2],
|
||||
periodic=False)
|
||||
|
||||
ndata = pos.shape[0]
|
||||
xi = convert_3d_counts_to_cf(ndata, ndata, nrandom, nrandom, dd, dr, dr, rr)
|
||||
rp = 0.5 * (bins[1:] + bins[:-1])
|
||||
return rp, xi
|
||||
|
|
@ -21,3 +21,4 @@ from .outsim import (dump_split, combine_splits) # noqa
|
|||
from .overlap_summary import (PairOverlap, NPairsOverlap, binned_resample_mean) # noqa
|
||||
from .knn_summary import kNNCDFReader # noqa
|
||||
from .pk_summary import PKReader # noqa
|
||||
from .tpcf_summary import TPCFReader # noqa
|
||||
|
|
|
|||
76
csiborgtools/read/tpcf_summary.py
Normal file
76
csiborgtools/read/tpcf_summary.py
Normal file
|
|
@ -0,0 +1,76 @@
|
|||
# Copyright (C) 2023 Richard Stiskalek
|
||||
# This program is free software; you can redistribute it and/or modify it
|
||||
# under the terms of the GNU General Public License as published by the
|
||||
# Free Software Foundation; either version 3 of the License, or (at your
|
||||
# option) any later version.
|
||||
#
|
||||
# This program is distributed in the hope that it will be useful, but
|
||||
# WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
|
||||
# Public License for more details.
|
||||
#
|
||||
# You should have received a copy of the GNU General Public License along
|
||||
# with this program; if not, write to the Free Software Foundation, Inc.,
|
||||
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
||||
"""2PCF reader."""
|
||||
from os.path import join
|
||||
from glob import glob
|
||||
import numpy
|
||||
import joblib
|
||||
|
||||
|
||||
class TPCFReader:
|
||||
"""
|
||||
Shortcut object to read in the 2PCF data.
|
||||
"""
|
||||
def read(self, run, folder):
|
||||
"""
|
||||
Read the auto- or cross-correlation kNN-CDF data. Infers the type from
|
||||
the data files.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
run : str
|
||||
Run ID to read in.
|
||||
folder : str
|
||||
Path to the folder where the auto-2PCF is stored.
|
||||
|
||||
Returns
|
||||
-------
|
||||
rp : 1-dimensional array of shape `(neval, )`
|
||||
Projected separations where the 2PCF is evaluated.
|
||||
out : 2-dimensional array of shape `(len(files), len(rp))`
|
||||
Array of 2PCFs.
|
||||
"""
|
||||
run += ".p"
|
||||
files = [f for f in glob(join(folder, "*")) if run in f]
|
||||
if len(files) == 0:
|
||||
raise RuntimeError("No files found for run `{}`.".format(run[:-2]))
|
||||
|
||||
for i, file in enumerate(files):
|
||||
data = joblib.load(file)
|
||||
if i == 0: # Initialise the array
|
||||
rp = data["rp"]
|
||||
out = numpy.full((len(files), rp.size), numpy.nan,
|
||||
dtype=numpy.float32)
|
||||
out[i, ...] = data["wp"]
|
||||
|
||||
return rp, out
|
||||
|
||||
def mean_wp(self, wp):
|
||||
r"""
|
||||
Calculate the mean 2PCF and its standard deviation averaged over the
|
||||
IC realisations.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
wp : 2-dimensional array of shape `(len(files), len(rp))`
|
||||
Array of CDFs
|
||||
Returns
|
||||
-------
|
||||
out : 2-dimensional array of shape `(len(rp), 2)`
|
||||
Mean 2PCF and its standard deviation, stored along the last
|
||||
dimension, respectively.
|
||||
"""
|
||||
return numpy.stack([numpy.mean(wp, axis=0), numpy.std(wp, axis=0)],
|
||||
axis=-1)
|
||||
Loading…
Add table
Add a link
Reference in a new issue