-
Notifications
You must be signed in to change notification settings - Fork 2
/
scalesmear.py
192 lines (156 loc) · 6.4 KB
/
scalesmear.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
import os
import sys
import shutil
import warnings
import re
import numpy as np
with warnings.catch_warnings():
warnings.simplefilter("ignore")
from uproot3_methods.classes.TH1 import Methods as TH1Methods
import uproot3
from scipy.interpolate import interp1d
from mplhep.error_estimation import poisson_interval
class AffineMorphTemplate(object):
def __init__(self, hist):
'''
hist: a numpy-histogram-like tuple of (sumw, edges)
'''
self.sumw, self.edges = hist
self.centers = self.edges[:-1] + np.diff(self.edges)/2
self.norm = self.sumw.sum()
self.mean = (self.sumw*self.centers).sum() / self.norm
self.cdf = interp1d(x=self.edges,
y=np.r_[0, np.cumsum(self.sumw / self.norm)],
kind='linear',
assume_sorted=True,
bounds_error=False,
fill_value=(0, 1),
)
def get(self, shift=0., scale=1.):
'''
Return a shifted and scaled histogram
i.e. new edges = edges * scale + shift
'''
if not np.isclose(scale, 1.):
shift += self.mean * (1 - scale)
scaled_edges = (self.edges - shift) / scale
return np.diff(self.cdf(scaled_edges)) * self.norm, self.edges
class MorphHistW2(object):
def __init__(self, hist):
'''
hist: uproot/UHI histogram or a tuple (values, edges, variances)
'''
try:
self.sumw = hist.values
self.edges = hist.edges
self.variances = hist.variances
except:
self.sumw, self.edges, self.variances = hist
from mplhep.error_estimation import poisson_interval
down, up = np.nan_to_num(np.abs(poisson_interval(self.sumw, self.variances)), 0.)
self.nominal = AffineMorphTemplate((self.sumw, self.edges))
self.w2s = AffineMorphTemplate((self.variances, self.edges))
def get(self, shift=0., scale=1.):
nom, edges = self.nominal.get(shift, scale)
w2s, edges = self.w2s.get(shift, scale)
return nom, edges, w2s
class TH1(TH1Methods, list):
pass
class TAxis(object):
def __init__(self, fNbins, fXmin, fXmax):
self._fNbins = fNbins
self._fXmin = fXmin
self._fXmax = fXmax
def export1d(hist, name='x', label='x', histtype=b"TH1F"):
"""Export a 1-dimensional `Hist` object to uproot
"""
try:
sumw, edges, sumw2 = hist
except:
sumw, edges = hist
sumw2 = sumw
sumw = np.r_[0, sumw, 0]
sumw2 = np.r_[0, sumw, 0]
out = TH1.__new__(TH1)
out._fXaxis = TAxis(len(edges) - 1, edges[0], edges[-1])
out._fXaxis._fName = name
out._fXaxis._fTitle = label
out._fXaxis._fXbins = edges.astype(">f8")
centers = (edges[:-1] + edges[1:]) / 2.0
out._fEntries = out._fTsumw = out._fTsumw2 = sumw[1:-1].sum()
out._fTsumwx = (sumw[1:-1] * centers).sum()
out._fTsumwx2 = (sumw[1:-1] * centers**2).sum()
out._fName = "histogram"
out._fTitle = label
out._classname = histtype.encode()
out.extend(sumw.astype(">f8"))
out._fSumw2 = sumw2.astype(">f8")
return out
def mdev(hist):
w, edges = hist
N = np.sum(w)
centers = edges[:-1] + 0.5*np.diff(edges)
mean = 1/N * np.sum(w * centers)
stdev2 = 1/N * np.sum(w * (centers-mean)**2)
return np.array([mean, np.sqrt(stdev2)])
if __name__ == "__main__":
import argparse
parser = argparse.ArgumentParser()
def str2bool(v):
if isinstance(v, bool):
return v
if v.lower() in ('yes', 'true', 't', 'y', '1'):
return True
elif v.lower() in ('no', 'false', 'f', 'n', '0'):
return False
else:
raise argparse.ArgumentTypeError('Boolean value expected.')
parser.add_argument('-i', '--in', dest='in_file', required=True, help="Source file")
parser.add_argument('-o', '--out', dest='out_file', default=None, help="Out file")
parser.add_argument("--scale", default='1', type=float, help="Scale value.")
parser.add_argument("--smear", default='0.5', type=float, help="Smear value.")
parser.add_argument('--plot', action='store_true', help="Make control plots")
parser.add_argument('--type', dest='hist_type', type=str, choices=["TH1F", "TH1D"], default="TH1D", help="TH1 type. Should be consistent with input.")
args = parser.parse_args()
if args.out_file is None:
args.out_file = args.in_file.replace("_pass", "_var_pass").replace("_fail", "_var_fail")
print("Running with the following options:")
print(args)
source_file = uproot3.open(args.in_file)
work_dir = os.path.dirname(args.in_file)
morph_base = MorphHistW2(source_file['catp2'])
scale_up = morph_base.get(shift=args.scale)
scale_down = morph_base.get(shift=-args.scale)
smear_up = morph_base.get(scale=1 + args.smear)
smear_down = morph_base.get(scale=1 - args.smear)
if args.plot:
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import mplhep as hep
hep.set_style("CMS")
fig, ax = plt.subplots()
hep.histplot(morph_base.get()[:2], c='black' , ls=':', label='Nominal')
hep.histplot(scale_up[:2], c='blue' , ls='--', label='Up')
hep.histplot(scale_down[:2], c='red' , ls='--', label='Down')
ax.set_xlabel('jet $m_{SD}$')
ax.legend()
fig.savefig(f'{work_dir}/scale.png')
fig, ax = plt.subplots()
hep.histplot(morph_base.get()[:2], c='black' , ls=':', label='Nominal')
hep.histplot(smear_up[:2], c='blue' , ls='--', label='Up')
hep.histplot(smear_down[:2], c='red' , ls='--', label='Down')
ax.set_xlabel('jet $m_{SD}$')
ax.legend()
fig.savefig(f'{work_dir}/smear.png')
if os.path.exists(args.out_file):
os.remove(args.out_file)
fout = uproot3.create(args.out_file)
fout['data_obs'] = source_file['data_obs']
fout['catp1'] = source_file['catp1']
fout['catp2'] = source_file['catp2']
fout['catp2_central'] = source_file['catp2']
fout['catp2_smearDown'] = export1d(smear_down, histtype=args.hist_type)
fout['catp2_smearUp'] = export1d(smear_up, histtype=args.hist_type)
fout['catp2_scaleDown'] = export1d(scale_down, histtype=args.hist_type)
fout['catp2_scaleUp'] = export1d(scale_up, histtype=args.hist_type)