From cc9e0663353ccfc1ce1b436d43f644452147dbd4 Mon Sep 17 00:00:00 2001
From: Nikolai Hartmann <Nikolai.Hartmann@physik.uni-muenchen.de>
Date: Tue, 12 Jun 2018 17:54:14 +0200
Subject: [PATCH] Normalise training and test sample correctly for significance
scan
---
toolkit.py | 15 ++++++++++-----
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/toolkit.py b/toolkit.py
index b8ae993..df89ef3 100755
--- a/toolkit.py
+++ b/toolkit.py
@@ -921,15 +921,18 @@ class ClassificationProject(object):
significances_train = []
significances_test = []
- for hist_sig, hist_bkg, rel_errors_sig, rel_errors_bkg, significances in [
- (hist_sig_train, hist_bkg_train, rel_errors_bkg_train, rel_errors_sig_train, significances_train),
- (hist_sig_test, hist_bkg_test, rel_errors_bkg_test, rel_errors_sig_test, significances_test),
+ for hist_sig, hist_bkg, rel_errors_sig, rel_errors_bkg, significances, w, y in [
+ (hist_sig_train, hist_bkg_train, rel_errors_bkg_train, rel_errors_sig_train, significances_train, self.w_train, self.y_train),
+ (hist_sig_test, hist_bkg_test, rel_errors_bkg_test, rel_errors_sig_test, significances_test, self.w_test, self.y_test),
]:
+ # factor to rescale due to using only a fraction of events (training and test samples)
+ normfactor_sig = (np.sum(self.w_train[self.y_train==1])+np.sum(self.w_test[self.y_test==1]))/np.sum(w[y==1])
+ normfactor_bkg = (np.sum(self.w_train[self.y_train==0])+np.sum(self.w_test[self.y_test==0]))/np.sum(w[y==0])
# first set nan values to 0 and multiply by lumi
for arr in hist_sig, hist_bkg, rel_errors_bkg:
arr[np.isnan(arr)] = 0
- hist_sig *= lumifactor
- hist_bkg *= lumifactor
+ hist_sig *= lumifactor*normfactor_sig
+ hist_bkg *= lumifactor*normfactor_bkg
for i in range(len(hist_sig)):
s = sum(hist_sig[i:])
b = sum(hist_bkg[i:])
@@ -941,6 +944,8 @@ class ClassificationProject(object):
z = 0
else:
z = significanceFunction(s, b, db)
+ if z == float('inf'):
+ z = 0
logger.debug("s, b, db, z = {}, {}, {}, {}".format(s, b, db, z))
significances.append(z)
--
GitLab