Skip to content

Speed optimization for unbinned NaivePoisson background fit #109

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Speed optimization for unbinned NaivePoisson background fit #109

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
28 changes: 17 additions & 11 deletions src/gdt/core/background/unbinned.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,6 +92,21 @@ def fit(self, window_width=100.0, fast=True):
self._actual_widths = actual_widths
self._rates = rates

rates_interp = []
width_interp = []
for i in range(self._numchans):
if self._times[i].size-1 == rates[i].size:
idx = 0
else:
idx = 1
rates_interp.append(
interp1d(self._times[i][idx:-1], rates[i], fill_value='extrapolate'))
width_interp.append(
interp1d(self._times[i][idx:-1], actual_widths[i], fill_value='extrapolate'))

self._rates_interp = rates_interp
self._width_interp = width_interp

def interpolate(self, tstart, tstop):
"""Interpolate the background at the given times

Expand All @@ -107,18 +122,9 @@ def interpolate(self, tstart, tstop):
rates = []
uncert = []
for i in range(self._numchans):
if self._times[i].size-1 == self._rates[i].size:
idx = 0
else:
idx = 1
rates_interp = interp1d(self._times[i][idx:-1], self._rates[i],
fill_value='extrapolate')
width_interp = interp1d(self._times[i][idx:-1],
self._actual_widths[i],
fill_value='extrapolate')
r = rates_interp(times)
r = self._rates_interp[i](times)
r[r < 0.0] = 0.0
widths = width_interp(times)
widths = self._width_interp[i](times)
mask = (widths > 0.0)
u = np.zeros_like(widths)
u[mask] = np.sqrt(r[mask]/widths[mask])
Expand Down