forked from mbandrews/MLAnalyzer
-
Notifications
You must be signed in to change notification settings - Fork 1
/
crop_preprocess_EBcrops.py
250 lines (217 loc) · 8.45 KB
/
crop_preprocess_EBcrops.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
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
import numpy as np
from os.path import splitext
import matplotlib.pyplot as plt
import ROOT
from scipy.ndimage import maximum_position
from scipy.sparse import csr_matrix
#import root_numpy
from matplotlib.colors import LogNorm
import argparse
# Register command line options
parser = argparse.ArgumentParser(description='Run STEALTH selection.')
parser.add_argument('-d','--decay', required=True, help='Decay:Single*Pt50_FEVTDEBUG_n250k_IMG',type=str)
parser.add_argument('-n','--nevts', default=10, help='Number of events to process.',type=int)
args = parser.parse_args()
def crop_around_max(b,row0_,col0_):
global n_rows, n_cols, w
return np.array(b, dtype=np.float32).reshape(n_rows,n_cols)[row0_-w:row0_+w,col0_-w:col0_+w].flatten()
def process_en(b,row0_,col0_):
b_csr = csr_matrix(crop_around_max(b,row0_,col0_))
b_csr.data = (np.log10(b_csr.data)+1.3)/4.
#b_csr.data = (np.log10(b_csr.data)+6.)/9.
return b_csr.toarray().flatten()
def process_t(b,row0_,col0_):
return crop_around_max(b,row0_,col0_)/50.
def log_noise(lin_presample):
b_csr = csr_matrix(lin_presample)
b_csr.data = np.log10(b_csr.data)
return b_csr.toarray().flatten()
'''
# Case 1: flat subtraction on log
def process_digi(b,row0_,col0_,_):
b = crop_around_max(b,row0_,col0_)
b_csr = csr_matrix(b)
b_csr.data = np.log10(b_csr.data)-2.3
return b_csr.toarray().flatten()
# Case 2: lin presample subtract, then log
def process_digi(b,row0_,col0_,lin_presample):
b = crop_around_max(b,row0_,col0_) - lin_presample + 10.
b_csr = csr_matrix(b.clip(0.,b.max()))
b_csr.data = np.log10(b_csr.data)-1.
return b_csr.toarray().flatten()/2.
'''
# Case 3: log presample subtraction
def process_digi(b,row0_,col0_,log_presample):
b = crop_around_max(b,row0_,col0_)
b_csr = csr_matrix(b)
b_csr.data = np.log10(b_csr.data)
return b_csr.toarray().flatten()-log_presample
'''
# Case 5: lin presample subtract only
def process_digi(b,row0_,col0_,lin_presample):
b = crop_around_max(b,row0_,col0_) - lin_presample
return b/1000.
'''
s = 32
crop_size = int(s*s)
w = s//2
n_rows = 170 # n_phi
n_cols = 360 # n_eta
##### I/O #####
eos_dir = '/eos/cms/store/user/mandrews/ML'
#decay = 'SinglePhotonPt50_FEVTDEBUG_n250k_IMG'
#decay = 'SingleElectronPt50_FEVTDEBUG_n250k_IMG'
decay = args.decay
file_in_str = '%s/IMGs/%s.root'%(eos_dir,decay)
tree_in = ROOT.TChain('fevt/RHTree')
tree_in.Add(file_in_str)
n_events = tree_in.GetEntries()
branch_list = [br.GetName() for br in tree_in.GetListOfBranches()]
print " >> Read input file:", file_in_str
print " >> N of events:", n_events
print " >> Input branch list:",branch_list
#file_out_str = 'test.root'
file_out_str = '%s/IMGs/%s_CROPS32_c3.root'%(eos_dir,decay)
file_out = ROOT.TFile(file_out_str, 'RECREATE')
RHTree = ROOT.TTree("RHTree", "RecHit tree")
EBenergy = np.zeros(crop_size, dtype=np.float32)
EBtime = np.zeros(crop_size, dtype=np.float32)
EBenergyRed = np.zeros(crop_size, dtype=np.float32)
EBtimeRed = np.zeros(crop_size, dtype=np.float32)
EB_adc0 = np.zeros(crop_size, dtype=np.float32)
EB_adc1 = np.zeros(crop_size, dtype=np.float32)
EB_adc2 = np.zeros(crop_size, dtype=np.float32)
EB_adc3 = np.zeros(crop_size, dtype=np.float32)
EB_adc4 = np.zeros(crop_size, dtype=np.float32)
EB_adc5 = np.zeros(crop_size, dtype=np.float32)
EB_adc6 = np.zeros(crop_size, dtype=np.float32)
EB_adc7 = np.zeros(crop_size, dtype=np.float32)
EB_adc8 = np.zeros(crop_size, dtype=np.float32)
EB_adc9 = np.zeros(crop_size, dtype=np.float32)
RHTree.Branch('EBenergy' ,EBenergy , 'EBenergy[%d]/F'%crop_size )
RHTree.Branch('EBtime' ,EBtime , 'EBtime[%d]/F'%crop_size )
RHTree.Branch('EBenergyRed',EBenergyRed , 'EBenergyRed[%d]/F'%crop_size)
RHTree.Branch('EBtimeRed' ,EBtimeRed , 'EBtimeRed[%d]/F'%crop_size )
RHTree.Branch('EB_adc0' ,EB_adc0 , 'EB_adc0[%d]/F'%crop_size )
RHTree.Branch('EB_adc1' ,EB_adc1 , 'EB_adc1[%d]/F'%crop_size )
RHTree.Branch('EB_adc2' ,EB_adc2 , 'EB_adc2[%d]/F'%crop_size )
RHTree.Branch('EB_adc3' ,EB_adc3 , 'EB_adc3[%d]/F'%crop_size )
RHTree.Branch('EB_adc4' ,EB_adc4 , 'EB_adc4[%d]/F'%crop_size )
RHTree.Branch('EB_adc5' ,EB_adc5 , 'EB_adc5[%d]/F'%crop_size )
RHTree.Branch('EB_adc6' ,EB_adc6 , 'EB_adc6[%d]/F'%crop_size )
RHTree.Branch('EB_adc7' ,EB_adc7 , 'EB_adc7[%d]/F'%crop_size )
RHTree.Branch('EB_adc8' ,EB_adc8 , 'EB_adc8[%d]/F'%crop_size )
RHTree.Branch('EB_adc9' ,EB_adc9 , 'EB_adc9[%d]/F'%crop_size )
branch_list = [br.GetName() for br in RHTree.GetListOfBranches()]
print " >> Output file:",file_out_str
print " >> Output branch list:",branch_list
# Temp arrays to calculate presample
EB_adc0_ = np.zeros(crop_size, dtype=np.float32)
EB_adc1_ = np.zeros(crop_size, dtype=np.float32)
EB_adc2_ = np.zeros(crop_size, dtype=np.float32)
##### IMAGE SELECTION #####
istart, istop = 0, n_events
if istop < args.nevts:
istop = n_evts
else:
istop = args.nevts
row0, col0 = -1, -1
print " >> Processing entries: [",istart,"->",istop,")"
for ievt in range(istart,istop):
# Initialize event
if ievt > istop:
break
treeStatus = tree_in.LoadTree(ievt)
if treeStatus < 0:
break
evtStatus = tree_in.GetEntry(ievt)
if evtStatus <= 0:
continue
if ievt % 1000 == 0:
print " .. Processing entry",ievt
### Crop around shower max ###
# Get position of max adc
row0, col0 = -1, -1
#print len(maximum_position(np.array(tree_in.EB_adc6, dtype=np.float32).reshape(n_rows,n_cols)))
row0, col0 = maximum_position(np.array(tree_in.EB_adc6, dtype=np.float32).reshape(n_rows,n_cols))
if col0 < w or col0 >= n_cols-w or row0 < w or row0 >= n_rows-w:
continue
#print row0,col0
### Energy ###
b = process_en(tree_in.EBenergy,row0,col0)
for i,val in enumerate(b):
EBenergy[i] = val
b = process_en(tree_in.EBenergyRed,row0,col0)
for i,val in enumerate(b):
EBenergyRed[i] = val
### Timing ###
b = process_t(tree_in.EBtime,row0,col0)
for i,val in enumerate(b):
EBtime[i] = val
b = process_t(tree_in.EBtimeRed,row0,col0)
for i,val in enumerate(b):
EBtimeRed[i] = val
### Digis ###
b = crop_around_max(tree_in.EB_adc0,row0,col0)
for i,val in enumerate(b):
EB_adc0[i] = val
b = crop_around_max(tree_in.EB_adc1,row0,col0)
for i,val in enumerate(b):
EB_adc1[i] = val
b = crop_around_max(tree_in.EB_adc2,row0,col0)
for i,val in enumerate(b):
EB_adc2[i] = val
EB_adc0_ = crop_around_max(tree_in.EB_adc0,row0,col0)
EB_adc1_ = crop_around_max(tree_in.EB_adc1,row0,col0)
EB_adc2_ = crop_around_max(tree_in.EB_adc2,row0,col0)
presample = np.mean([EB_adc0_, EB_adc1_, EB_adc2_], axis=0)
presample = log_noise(presample)
#presample = 0.
b = process_digi(tree_in.EB_adc0,row0,col0,presample)
for i,val in enumerate(b):
EB_adc0[i] = val
b = process_digi(tree_in.EB_adc1,row0,col0,presample)
for i,val in enumerate(b):
EB_adc1[i] = val
b = process_digi(tree_in.EB_adc2,row0,col0,presample)
for i,val in enumerate(b):
EB_adc2[i] = val
b = process_digi(tree_in.EB_adc3,row0,col0,presample)
for i,val in enumerate(b):
EB_adc3[i] = val
b = process_digi(tree_in.EB_adc4,row0,col0,presample)
for i,val in enumerate(b):
EB_adc4[i] = val
b = process_digi(tree_in.EB_adc5,row0,col0,presample)
for i,val in enumerate(b):
EB_adc5[i] = val
b = process_digi(tree_in.EB_adc6,row0,col0,presample)
for i,val in enumerate(b):
EB_adc6[i] = val
b = process_digi(tree_in.EB_adc7,row0,col0,presample)
for i,val in enumerate(b):
EB_adc7[i] = val
b = process_digi(tree_in.EB_adc8,row0,col0,presample)
for i,val in enumerate(b):
EB_adc8[i] = val
b = process_digi(tree_in.EB_adc9,row0,col0,presample)
for i,val in enumerate(b):
EB_adc9[i] = val
RHTree.Fill()
'''
# Check plots
img = b['EB_adc9'].reshape(32,32)
plt.imshow(img, interpolation="None", cmap='seismic', vmin=-0.1, vmax=0.8) # Blues, seismic
plt.colorbar()
plt.show()
hist = ROOT.TH1F("h","h",100,-1.*(img.ravel().max()+2.),img.ravel().max()+2.)
print img.ravel().max()
for i in img.ravel():
hist.Fill(i)
c = ROOT.TCanvas("c")
ROOT.gPad.SetLogy()
hist.Draw()
c.Draw()
'''
file_out.Write()
file_out.Close()