forked from mbandrews/MLAnalyzer
-
Notifications
You must be signed in to change notification settings - Fork 2
/
convert_Tree2Dask_EB+EE+HBHEupsample.py
161 lines (140 loc) · 6.57 KB
/
convert_Tree2Dask_EB+EE+HBHEupsample.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
import numpy as np
import ROOT
from root_numpy import tree2array
from dask.delayed import delayed
import dask.array as da
eosDir='/eos/uscms/store/user/mba2012/IMGs'
#decays = ["H125GGgluonfusion_Pt25_Eta23_13TeV_TuneCUETP8M1_HighLumiPileUpv2", "PromptDiPhoton_MGG80toInf_Pt25_Eta23_13TeV_TuneCUETP8M1_HighLumiPileUp"]
decays = ['H125GGgluonfusion_Pt25_Eta14_13TeV_TuneCUETP8M1_HighLumiPileUpv4','GJet_DoubleEMEnriched_PtHat20_MGG80toInf_Pt25_Eta14_13TeV_TuneCUETP8M1_HighLumiPileUp']
#decays = ['dummy','GJet_DoubleEMEnriched_PtHat20_MGG80toInf_Pt25_Eta23_13TeV_TuneCUETP8M1_HighLumiPileUp']
#chunk_size = 250
chunk_size = 100
#scale = [100., 150.]
scale = [100., 100.]
@delayed
def load_X(tree, start_, stop_, branches_, readouts, scale):
X = tree2array(tree, start=start_, stop=stop_, branches=branches_)
# Convert the object array X to a multidim array:
# 1: for each event x in X, concatenate the object columns (branches) into a flat array of shape (readouts*branches)
# 2: reshape the flat array into a stacked array: (branches, readouts)
# 3: embed each stacked array as a single row entry in a list via list comprehension
# 4: convert this list into an array with shape (events, branches, readouts)
X = np.array([np.concatenate(x).reshape(len(branches_),readouts[0]*readouts[1]) for x in X])
#print "X.shape:",X.shape
X = X.reshape((-1,len(branches_),readouts[0],readouts[1]))
X = np.transpose(X, [0,2,3,1])
# Rescale
X /= scale
return X
@delayed
def load_X_upsampled(tree, start_, stop_, branches_, readouts, scale, upscale):
X = tree2array(tree, start=start_, stop=stop_, branches=branches_)
# Convert the object array X to a multidim array:
# 1: for each event x in X, concatenate the object columns (branches) into a flat array of shape (readouts*branches)
# 2: reshape the flat array into a stacked array: (branches, readouts)
# 3: embed each stacked array as a single row entry in a list via list comprehension
# 4: convert this list into an array with shape (events, branches, readouts)
X = np.array([np.concatenate(x).reshape(len(branches_),readouts[0]*readouts[1]) for x in X])
#print "X.shape:",X.shape
X = X.reshape((-1,len(branches_),readouts[0],readouts[1]))
#print "unsampled.shape",X.shape
X = np.stack([tile_stacked_array(x, upscale) for x in X])
#print "upsampled.shape",X.shape
X = np.transpose(X, [0,2,3,1])
# Rescale
X /= scale
return X
from numpy.lib.stride_tricks import as_strided
def tile_stacked_array(X, upscale):
#print "un-tile_stacked.shape",X.shape
X = np.stack([tile_array(x, upscale, upscale) for x in X])
#print "tile_stacked.shape",X.shape
return X
def tile_array(x, b0, b1):
r, c = x.shape # number of rows/columns
rs, cs = x.strides # row/column strides
x = as_strided(x, (r, b0, c, b1), (rs, 0, cs, 0)) # view a as larger 4D array
return x.reshape(r*b0, c*b1) # create new 2D array
for j,decay in enumerate(decays):
if j == 0:
pass
#continue
#tfile_str = '%s/%s_IMG.root'%(eosDir,decay)
tfile_str = '%s/%s_FEVTDEBUG_IMG.root'%(eosDir,decay)
tfile = ROOT.TFile(tfile_str)
tree = tfile.Get('fevt/RHTree')
nevts = tree.GetEntries()
#neff = (nevts//1000)*1000
#neff = (nevts//100)*100
neff = 29900
#neff = 100
print " >> Doing decay:", decay
print " >> Input file:", tfile_str
print " >> Total events:", nevts
print " >> Effective events:", neff
# EB
readouts = [170,360]
branches = ["EB_energy"]
X_EB = da.concatenate([\
da.from_delayed(\
load_X(tree,i,i+chunk_size, branches, readouts, scale[0]),\
shape=(chunk_size, readouts[0], readouts[1], len(branches)),\
dtype=np.float32)\
for i in range(0,neff,chunk_size)])
print " >> Expected shape:", X_EB.shape
# EE-
readouts = [100,100]
branches = ["EEm_energy"]
X_EEm = da.concatenate([\
da.from_delayed(\
load_X(tree,i,i+chunk_size, branches, readouts, scale[1]),\
shape=(chunk_size, readouts[0], readouts[1], len(branches)),\
dtype=np.float32)\
for i in range(0,neff,chunk_size)])
print " >> Expected shape:", X_EEm.shape
# EE+
readouts = [100,100]
branches = ["EEp_energy"]
X_EEp = da.concatenate([\
da.from_delayed(\
load_X(tree,i,i+chunk_size, branches, readouts, scale[1]),\
shape=(chunk_size, readouts[0], readouts[1], len(branches)),\
dtype=np.float32)\
for i in range(0,neff,chunk_size)])
print " >> Expected shape:", X_EEp.shape
# HBHE
readouts = [34,72]
branches = ["HBHE_energy_EB"]
X_HBHE = da.concatenate([\
da.from_delayed(\
load_X(tree,i,i+chunk_size, branches, readouts, scale[1]),\
shape=(chunk_size, readouts[0], readouts[1], len(branches)),\
dtype=np.float32)\
for i in range(0,neff,chunk_size)])
print " >> Expected shape:", X_HBHE.shape
# HBHE upsample
readouts = [34,72]
branches = ["HBHE_energy_EB"]
upscale = 5
X_HBHE_up = da.concatenate([\
da.from_delayed(\
load_X_upsampled(tree,i,i+chunk_size, branches, readouts, scale[1], upscale),\
shape=(chunk_size, readouts[0]*upscale, readouts[1]*upscale, len(branches)),\
dtype=np.float32)\
for i in range(0,neff,chunk_size)])
print " >> Expected shape:", X_HBHE_up.shape
# Class label
label = j
#label = 1
print " >> Class label:",label
y = da.from_array(\
np.full(X_EB.shape[0], label, dtype=np.float32),\
chunks=(chunk_size,))
#file_out_str = "test.hdf5"
file_out_str = "%s/%s_IMG_EBEEHBup_RH%d_n%dk.hdf5"%(eosDir,decay,int(scale[0]),neff//1000.)
#file_out_str = "%s/%s_IMG_RH%d-%d_n%dk.hdf5"%(eosDir,decay,int(scale[0]),int(scale[1]),neff//1000.)
print " >> Writing to:", file_out_str
#da.to_hdf5(file_out_str, {'/X_EB': X_EB, 'X_EEm': X_EEm, 'X_EEp': X_EEp, 'X_HBHE': X_HBHE, '/y': y}, compression='lzf')
#da.to_hdf5(file_out_str, {'/X': X_EB, 'X_EEm': X_EEm, 'X_EEp': X_EEp, 'X_HBHE': X_HBHE, '/y': y}, compression='lzf')
da.to_hdf5(file_out_str, {'/X_EB': X_EB, 'X_EEm': X_EEm, 'X_EEp': X_EEp, 'X_HBHE': X_HBHE, 'X_HBHE_up': X_HBHE_up, '/y': y}, compression='lzf')
print " >> Done.\n"