Skip to content

Add weighting control in fit_velocities #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jluastro merged 1 commit into from
Apr 3, 2024
Merged

Add weighting control in fit_velocities #1

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
74 changes: 57 additions & 17 deletions flystar/startables.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
from astropy.table import Table, Column
from astropy.stats import sigma_clipping
from scipy.optimize import curve_fit
from tqdm import tqdm
import numpy as np
import warnings
import collections
Expand Down Expand Up @@ -526,11 +527,14 @@ def detections(self):
return


def fit_velocities(self, bootstrap=0, fixed_t0=False, verbose=False,
def fit_velocities(self, weighting='var', bootstrap=0, fixed_t0=False, verbose=False,
mask_val=None, mask_lists=False):
"""
Fit velocities for all stars in the table.
"""
if weighting not in ['var', 'std']:
raise ValueError(f"Weighting must either be 'var' or 'std', not {weighting}!")

if ('t' not in self.colnames) and ('list_times' not in self.meta):
raise RuntimeError('fit_velocities: Failed to time values.')

Expand Down Expand Up @@ -591,9 +595,9 @@ def fit_velocities(self, bootstrap=0, fixed_t0=False, verbose=False,

# STARS LOOP through the stars and work on them 1 at a time.
# This is slow; but robust.
for ss in range(N_stars):
for ss in tqdm(range(N_stars)):
self.fit_velocity_for_star(ss, bootstrap=bootstrap, fixed_t0=fixed_t0,
mask_val=mask_val, mask_lists=mask_lists)
mask_val=mask_val, mask_lists=mask_lists, weighting=weighting)

if verbose:
stop_time = time.time()
Expand All @@ -602,7 +606,7 @@ def fit_velocities(self, bootstrap=0, fixed_t0=False, verbose=False,
return

def fit_velocity_for_star(self, ss, bootstrap=False, fixed_t0=False,
mask_val=None, mask_lists=False):
mask_val=None, mask_lists=False, weighting='var'):
def poly_model(time, *params):
pos = np.polynomial.polynomial.polyval(time, params)
return pos
Expand All @@ -613,6 +617,12 @@ def poly_model(time, *params):
if mask_val:
x = np.ma.masked_values(x, mask_val)
y = np.ma.masked_values(y, mask_val)
# If no mask, convert x.mask to list
if not np.ma.is_masked(x):
x.mask = np.zeros_like(x.data, dtype=bool)
if not np.ma.is_masked(y):
y.mask = np.zeros_like(y.data, dtype=bool)


if mask_lists is not False:
# Remove a list
Expand Down Expand Up @@ -655,6 +665,11 @@ def poly_model(time, *params):
if mask_val:
xe = np.ma.masked_values(xe, mask_val)
ye = np.ma.masked_values(ye, mask_val)
# If no mask, convert xe.mask to list
if not np.ma.is_masked(xe):
xe.mask = np.zeros_like(xe.data, dtype=bool)
if not np.ma.is_masked(ye):
ye.mask = np.zeros_like(ye.data, dtype=bool)

if mask_lists is not False:
# Remove a list
Expand All @@ -675,6 +690,8 @@ def poly_model(time, *params):

if mask_val:
t = np.ma.masked_values(t, mask_val)
if not np.ma.is_masked(t):
t.mask = np.zeros_like(t.data, dtype=bool)

if mask_lists is not False:
# Remove a list
Expand All @@ -685,11 +702,13 @@ def poly_model(time, *params):
# Throw a warning if mask_lists is not a list
if not isinstance(mask_lists, list):
raise RuntimeError('mask_lists needs to be a list.')


# For inconsistent masks, mask the star if any of the values are masked.
new_mask = np.logical_or.reduce((t.mask, x.mask, y.mask, xe.mask, ye.mask))
# Figure out where we have detections (as indicated by error columns)
good = np.where((xe != 0) & (ye != 0) &
np.isfinite(xe) & np.isfinite(ye) &
np.isfinite(x) & np.isfinite(y))[0]
np.isfinite(x) & np.isfinite(y) & ~new_mask)[0]

N_good = len(good)

Expand Down Expand Up @@ -721,8 +740,12 @@ def poly_model(time, *params):

# Catch the case where all the times are identical
if (dt == dt[0]).all():
wgt_x = (1.0/xe)**2
wgt_y = (1.0/ye)**2
if weighting == 'var':
wgt_x = (1.0/xe)**2
wgt_y = (1.0/ye)**2
elif weighting == 'std':
wgt_x = 1./np.abs(xe)
wgt_y = 1./np.abs(ye)

self['x0'][ss] = np.average(x, weights=wgt_x)
self['y0'][ss] = np.average(y, weights=wgt_y)
Expand All @@ -739,9 +762,16 @@ def poly_model(time, *params):
# Catch the case where we have enough measurements to actually
# fit a velocity!
if N_good > 2:
vx_opt, vx_cov = curve_fit(poly_model, dt.compressed(), x.compressed(), p0=p0x, sigma=xe.compressed(),
absolute_sigma=True)
vy_opt, vy_cov = curve_fit(poly_model, dt.compressed(), y.compressed(), p0=p0y, sigma=ye.compressed(),
if weighting == 'var':
sigma_x = xe
sigma_y = ye
elif weighting == 'std':
sigma_x = np.abs(xe)**0.5
sigma_y = np.abs(ye)**0.5

vx_opt, vx_cov = curve_fit(poly_model, dt, x, p0=p0x, sigma=sigma_x,
absolute_sigma=True)
vy_opt, vy_cov = curve_fit(poly_model, dt, y, p0=p0y, sigma=sigma_y,
absolute_sigma=True)

self['x0'][ss] = vx_opt[0]
Expand All @@ -760,10 +790,16 @@ def poly_model(time, *params):

for bb in range(bootstrap):
bdx = np.random.choice(edx, N_good)

vx_opt_b, vx_cov_b = curve_fit(poly_model, dt[bdx].compressed(), x[bdx].compressed(), p0=vx_opt, sigma=xe[bdx].compressed(),
if weighting == 'var':
sigma_x = xe[bdx]
sigma_y = ye[bdx]
elif weighting == 'std':
sigma_x = xe[bdx]**0.5
sigma_y = ye[bdx]**0.5

vx_opt_b, vx_cov_b = curve_fit(poly_model, dt[bdx], x[bdx], p0=vx_opt, sigma=sigma_x,
absolute_sigma=True)
vy_opt_b, vy_cov_b = curve_fit(poly_model, dt[bdx].compressed(), y[bdx].compressed(), p0=vy_opt, sigma=ye[bdx].compressed(),
vy_opt_b, vy_cov_b = curve_fit(poly_model, dt[bdx], y[bdx], p0=vy_opt, sigma=sigma_y,
absolute_sigma=True)

fit_x0_b[bb] = vx_opt_b[0]
Expand All @@ -786,9 +822,13 @@ def poly_model(time, *params):
self['vye'][ss] = vy_err[1]

elif N_good == 2:
# Note nough epochs to fit a velocity.
self['x0'][ss] = np.average(x, weights=1.0/xe**2)
self['y0'][ss] = np.average(y, weights=1.0/ye**2)
# Not nough epochs to fit a velocity.
if weighting == 'var':
self['x0'][ss] = np.average(x, weights=1.0/xe**2)
self['y0'][ss] = np.average(y, weights=1.0/ye**2)
elif weighting == 'std':
self['x0'][ss] = np.average(x, weights=1.0/np.abs(xe))
self['y0'][ss] = np.average(y, weights=1.0/np.abs(ye))

dx = np.diff(x)[0]
dy = np.diff(y)[0]
Expand Down