def getFinalTruthDictSA(self):
truthdict = self.getFinalTruthDictNumpy()
out={}
for k in truthdict.keys():
a,rs = truthdict[k]
out[k] = SimpleArray(a,rs,name=k)
return out
def convertFromSourceFile(self, filename, weighterobjects, istraining):
global raggedtester
import hashlib
from DeepJetCore import SimpleArray
seed = int(hashlib.sha1(filename.encode('utf-8')).hexdigest(),
16) % (10**8)
np.random.seed(seed)
nsamples = np.random.randint(12, 101, size=1)
data, rs = raggedtester.createData(nsamples)
farr = SimpleArray(data, rs)
true_arr = SimpleArray(data, rs)
farrint = SimpleArray(np.array(data, dtype='int32'), rs)
#farr.createFromNumpy()
return [farr, farrint], [true_arr], []
def convertFromSourceFile(self, filename, weighterobjects, istraining):
from DeepJetCore.preprocessing import MeanNormApply, MeanNormZeroPad, MeanNormZeroPadParticles
import numpy
from DeepJetCore.stopwatch import stopwatch
sw = stopwatch()
swall = stopwatch()
import ROOT
fileTimeOut(filename, 120) #give eos a minute to recover
rfile = ROOT.TFile(filename)
tree = rfile.Get("deepntuplizer/tree")
self.nsamples = tree.GetEntries()
print('took ', sw.getAndReset(), ' seconds for getting tree entries')
# split for convolutional network
x_global = MeanNormZeroPad(filename, None, ['x'], [1], self.nsamples)
print('took ', sw.getAndReset(),
' seconds for mean norm and zero padding (C module)')
Tuple = self.readTreeFromRootToTuple(
filename, branches=['class1', 'class2', 'x'])
truthtuple = Tuple[self.truthclasses]
alltruth = self.reduceTruth(truthtuple)
#print(x_global.shape,x_global[0:10])
#print(alltruth.shape,alltruth[0:10])
#print(alltruth.flags)
newnsamp = x_global.shape[0]
self.nsamples = newnsamp
print(x_global.shape, alltruth.shape, self.nsamples)
truth = SimpleArray(alltruth, name="truth")
feat = SimpleArray(x_global, name="features0")
return [feat], [truth], []
def convertFromSourceFile(self, filename, weighterobjects, istraining, treename="SLCIOConverted"):
fileTimeOut(filename, 10)#10 seconds for eos to recover
tree = uproot.open(filename)[treename]
nevents = tree.numentries
selection=None
hit_energy , rs = self.branchToFlatArray(tree["energy"], True,selection)
hit_x = self.branchToFlatArray(tree["positionX"], False,selection)
hit_y = self.branchToFlatArray(tree["positionY"], False,selection)
hit_z = self.branchToFlatArray(tree["positionZ"], False,selection)
hit_ass_truth_idx = self.branchToFlatArray(tree["maxE_particle_index"], False,selection)
hit_ass_truth_energy = self.branchToFlatArray(tree["maxE_particle_energy"], False,selection)
#not used right now
hit_ass_truth_pX = self.branchToFlatArray(tree["maxE_particle_pX"], False,selection)
hit_ass_truth_pY = self.branchToFlatArray(tree["maxE_particle_pY"], False,selection)
hit_ass_truth_pZ = self.branchToFlatArray(tree["maxE_particle_pZ"], False,selection)
features = np.concatenate([
hit_energy,
hit_x ,
hit_y,
hit_z
], axis=-1)
farr = SimpleArray(features,rs,name="features")
t_idxarr = SimpleArray(hit_ass_truth_idx,rs,name="t_idx")
t_energyarr = SimpleArray(hit_ass_truth_energy,rs,name="t_energy")
zeros = np.zeros_like(hit_ass_truth_energy)
#just for compatibility
t_posarr = SimpleArray(zeros,rs,name="t_pos")
t_time = SimpleArray(zeros,rs,name="t_time")
t_pid = SimpleArray(zeros,rs,name="t_pid") #this would need some massaging so we can't use the PID directly
t_spectator = SimpleArray(zeros,rs,name="t_spectator")
t_fully_contained = SimpleArray(zeros,rs,name="t_fully_contained")
t_rest = SimpleArray(zeros,rs,name="t_rest") #breaks with old plotting but needs to be done at some point
return [farr, t_idxarr, t_energyarr, t_posarr, t_time, t_pid, t_spectator, t_fully_contained],[t_rest], []
def createFromCsvs(self, filename_truth, filename_hits, filename_cells,
filename_particles, outfilename):
df_hits = pd.read_csv(os.path.join(
'/Users/shahrukhqasim/Downloads/kaggle_trackml/train_100_events/',
filename_hits),
sep=',')
df_truth = pd.read_csv(os.path.join(
'/Users/shahrukhqasim/Downloads/kaggle_trackml/train_100_events/',
filename_truth),
sep=',')
df_particles = pd.read_csv(os.path.join(
'/Users/shahrukhqasim/Downloads/kaggle_trackml/train_100_events/',
filename_particles),
sep=',')
df_cells = pd.read_csv(os.path.join(
'/Users/shahrukhqasim/Downloads/kaggle_trackml/train_100_events/',
filename_cells),
sep=',')
pt = np.sqrt(df_particles['px']**2 + df_particles['py']**2)
df_particles = df_particles[(pt > 1.5)]
# print("Particles", df_particles.shape)
# print("Truths before", df_truth['particle_id'].shape)
df_truth = df_truth[np.isin(df_truth['particle_id'],
df_particles['particle_id'])]
# print("Truth after",df_truth.shape)
# print("Hits before", df_hits.shape)
df_hits = df_hits[np.isin(df_hits['hit_id'], df_truth['hit_id'])]
# print("Hits after", df_hits.shape)
particle_id = df_truth['particle_id']
particle_id2 = particle_id.copy()
unique_pids = np.unique(particle_id)
for i, u in enumerate(unique_pids):
particle_id2[particle_id == u] = i
# print(np.unique(particle_id2))
particle_id2 = particle_id2.astype(np.int32)
farr = np.stack(
(df_hits['x'], df_hits['y'], df_hits['z'], df_hits['volume_id'],
df_hits['layer_id'], df_hits['module_id']),
axis=-1)
farr = farr.astype(np.float32)
tarr = np.stack((particle_id2, df_truth['hit_id'], df_truth['tx'],
df_truth['ty'], df_truth['tz'], df_truth['weight']),
axis=-1)
tarr = tarr.astype(np.float32)
#
rs = np.array([0, len(farr)], np.int64)
print(rs, len(farr), len(tarr))
# rs[1] = 100
# print(rs, rs.dtype)
#
# 0/0
farr_ = SimpleArray(name="feats")
farr_.createFromNumpy(farr, rs)
tarr_ = SimpleArray(name="truths")
tarr_.createFromNumpy(tarr, rs)
tarr_2 = SimpleArray(name="truths")
tarr_2.createFromNumpy(tarr, rs)
# print(farr.shape, tarr.shape)
self._store([farr_, tarr_], [tarr_2], [])
# print(outfilename)
self.writeToFile(outfilename)
def createFromCsvsIntoStandard(self, filename_truth, filename_hits,
filename_cells, filename_particles,
outfilename):
df_hits = pd.read_csv(filename_hits, sep=',')
df_truth = pd.read_csv(filename_truth, sep=',')
df_particles = pd.read_csv(filename_particles, sep=',')
df_cells = pd.read_csv(filename_cells, sep=',')
cells_hit_ids = df_cells['hit_id'].to_numpy(copy=True)
cells_hit_weights = df_cells['value'].to_numpy(copy=True)
hit_ids = df_hits['hit_id']
last_hit_id = -1
last_sum = -1
hit_ids_c = []
rechitEnergy = []
# recHitEnergy,recHitEta,zeroFeature,recHitTheta,recHitR,recHitX,recHitY,recHitZ,recHitTime,
for id, weight in zip(cells_hit_ids, cells_hit_weights):
if id == last_hit_id:
rechitEnergy[-1] += weight
else:
hit_ids_c.append(id)
rechitEnergy.append(weight)
last_hit_id = id
assert np.sum(hit_ids - hit_ids_c) == 0
df_hits['energy'] = rechitEnergy
ptx = np.sqrt(df_particles['px']**2 + df_particles['py']**2)
p = np.sqrt(df_particles['px']**2 + df_particles['py']**2 +
df_particles['pz']**2)
# print("Particles", df_particles.shape)
# print("Truths before", df_truth['particle_id'].shape)
# print(len(np.unique(df_particles['particle_id'])), len(np.unique(df_truth['particle_id'])))
df_particles = df_particles[(ptx > 1.5)]
df_truth = df_truth[np.isin(df_truth['particle_id'],
df_particles['particle_id'])]
# print(len(np.unique(df_particles['particle_id'])), len(np.unique(df_truth['particle_id'])))
# 0/0
# df_truth['pt'] = pt
# print("Truth after",df_truth.shape)
# print("Hits before", df_hits.shape)
df_hits = df_hits[np.isin(df_hits['hit_id'], df_truth['hit_id'])]
df_cells = df_cells[np.isin(df_cells['hit_id'], df_truth['hit_id'])]
rechHitEnergy = df_hits['energy'].to_numpy(dtype=np.float32)
recHitX = df_hits['x'].to_numpy(dtype=np.float32)
recHitY = df_hits['y'].to_numpy(dtype=np.float32)
recHitZ = df_hits['z'].to_numpy(dtype=np.float32)
recHitR = np.sqrt(recHitX**2. + recHitY**2. + recHitZ**2.)
recHitTheta = np.arccos(recHitZ / recHitR)
recHitEta = -np.log(np.tan(recHitTheta / 2))
zeroFeature = recHitEta * 0
# print("Hits after", df_hits.shape)
particle_id = df_truth['particle_id'].to_numpy()
particle_id2 = particle_id.copy()
# unique_pids = np.unique(particle_id)
# x = df_particles['particle_id']
# print(unique_pids.shape, x.shape)
# print(np.sum(x-unique_pids))
# 0/0
recHitTruthEnergy = zeroFeature.copy()
recHitTruthDepEnergy = zeroFeature.copy()
unique_pids = df_particles['particle_id']
px = df_particles['px'].to_numpy()
py = df_particles['py'].to_numpy()
pz = df_particles['pz'].to_numpy()
for i, u in enumerate(unique_pids):
particle_id2[particle_id == u] = i
recHitTruthEnergy[particle_id == u] = np.sqrt(px[i]**2 + py[i]**2 +
0 * pz[i]**2)
recHitTruthDepEnergy[particle_id == u] = np.sum(
rechHitEnergy[particle_id == u])
df_truth['particle_id'] = particle_id2
df_truth['p'] = recHitTruthEnergy
recHitSimClusIdx = df_truth['particle_id'].to_numpy().astype(np.int32)
# recHitTruthEnergy = df_truth['pt'].to_numpy(dtype=np.float32)
recHitTruthX = df_truth['tx'].to_numpy().astype(np.float32)
recHitTruthY = df_truth['ty'].to_numpy().astype(np.float32)
recHitTruthZ = df_truth['tz'].to_numpy().astype(np.float32)
recHitTruthR = np.sqrt(recHitTruthX**2. + recHitTruthY**2. +
recHitTruthZ**2.).astype(np.float32)
recHitTruthTheta = np.arccos(recHitTruthZ / recHitTruthR).astype(
np.float32)
recHitTruthEta = -np.log(np.tan(recHitTruthTheta / 2)).astype(
np.float32)
recHitTruthPhi = np.arctan(recHitTruthY / recHitTruthX).astype(
np.float32)
recHitTruthTime = zeroFeature
recHitTruthDepEnergy = recHitTruthDepEnergy.astype(np.float32)
recHitTruthPID = zeroFeature
truth = np.stack(
[
np.array(recHitSimClusIdx, dtype='float32'), # 0
recHitTruthEnergy,
recHitTruthX,
recHitTruthY,
recHitTruthZ, # 4
zeroFeature, # truthHitAssignedDirX,
zeroFeature, # 6
zeroFeature,
recHitTruthEta,
recHitTruthPhi,
recHitTruthTime, # 10
zeroFeature,
zeroFeature,
recHitTruthDepEnergy, # 13
zeroFeature, # 14
zeroFeature, # 15
recHitTruthPID, # 16 - 16+n_classes #won't be used anymore
zeroFeature,
zeroFeature
],
axis=1)
truth = truth.astype(np.float32)
features = np.stack(
[
rechHitEnergy,
recHitEta,
zeroFeature, # indicator if it is track or not
recHitTheta,
recHitR,
recHitX,
recHitY,
recHitZ,
zeroFeature,
zeroFeature,
],
axis=1)
features = features.astype(np.float32)
rs = np.array([0, len(features)], np.int64)
farr = SimpleArray(name="recHitFeatures")
farr.createFromNumpy(features, rs)
t_rest = SimpleArray(name="recHitTruth")
t_rest.createFromNumpy(truth, rs)
# rs[1] = 100
# print(rs, rs.dtype)
#
# 0/0
t_idxarr = SimpleArray(name="recHitTruthClusterIdx")
t_idxarr.createFromNumpy(recHitSimClusIdx[..., np.newaxis], rs)
t_energyarr = SimpleArray(name="recHitTruthEnergy")
t_energyarr.createFromNumpy(recHitTruthEnergy[..., np.newaxis], rs)
t_posarr = SimpleArray(name="recHitTruthPosition")
t_posarr.createFromNumpy(
np.concatenate(
[recHitTruthX[..., np.newaxis], recHitTruthY[..., np.newaxis]],
axis=-1), rs)
# print(np.concatenate([recHitTruthX[..., np.newaxis], recHitTruthY[..., np.newaxis]], axis=-1).shape)
# 0/0
t_time = SimpleArray(name="recHitTruthTime")
t_time.createFromNumpy(recHitTruthTime[..., np.newaxis], rs)
t_pid = SimpleArray(name="recHitTruthID")
t_pid.createFromNumpy(recHitTruthPID[..., np.newaxis], rs)
t_spectator = SimpleArray(
name="recHitSpectatorFlag"
) # why do we have inconsistent namings, where is it needed? wrt. to truth array
t_spectator.createFromNumpy(zeroFeature[..., np.newaxis], rs)
t_fully_contained = SimpleArray(name="recHitFullyContainedFlag")
t_fully_contained.createFromNumpy(
(zeroFeature[..., np.newaxis] + 1).astype(np.int32), rs)
# remaining truth is mostly for consistency in the plotting tools
t_rest = SimpleArray(name="recHitTruth")
t_rest.createFromNumpy(truth, rs)
x, y, z = [
farr, t_idxarr, t_energyarr, t_posarr, t_time, t_pid, t_spectator,
t_fully_contained
], [], []
self._store(x, y, z)
self.writeToFile(outfilename)
print("Storing in new format")
def base_convertFromSourceFile(self,
filename,
weighterobjects,
istraining,
treename="Events",
removeTracks=True):
fileTimeOut(filename, 10) #10 seconds for eos to recover
tree = uproot.open(filename)[treename]
hits = "RecHitHGC"
front_face_z = 323 #this needs to be more precise
recHitZUnsplit = self.hitObservable(tree,
hits,
"z",
split=False,
flatten=False)
self.setSplitIdx(recHitZUnsplit < 0)
recHitZ = self.splitJaggedArray(recHitZUnsplit)
offsets = recHitZ.offsets
recHitX = self.hitObservable(tree,
hits,
"x",
split=True,
flatten=False)
recHitY = self.hitObservable(tree,
hits,
"y",
split=True,
flatten=False)
recHitSimClusIdx = self.hitObservable(tree,
hits,
"BestMergedSimClusterIdx",
split=True,
flatten=False)
#Define spectators
recHit_df_events = [
pd.DataFrame({
"recHitX": recHitX[i],
"recHitY": recHitY[i],
"recHitZ": recHitZ[i],
"recHitSimClusIdx": recHitSimClusIdx[i]
}) for i in range(recHitX.shape[0])
]
for ievent in range(len(recHit_df_events)):
df_event = recHit_df_events[ievent]
unique_shower_idx = np.unique(df_event['recHitSimClusIdx'])
df_event['spectator_distance'] = 0 #
df_event['recHitSimClus_nHits'] = df_event.groupby(
'recHitSimClusIdx'
).recHitX.transform(
len
) #adding number of rec hits that are associated to this truth cluster
for idx in unique_shower_idx:
df_shower = df_event[df_event['recHitSimClusIdx'] == idx]
x_to_fit = df_shower[['recHitX', 'recHitY',
'recHitZ']].to_numpy()
spectators_shower_dist = find_pcas(x_to_fit,
PCA_n=2,
min_hits=10)
if (spectators_shower_dist is not None):
spectators_idx = (df_shower.index.tolist())
df_event.loc[spectators_idx,
'spectator_distance'] = spectators_shower_dist
del df_shower
del df_event
#Expand back
recHitX = np.expand_dims(recHitX.content, axis=1)
recHitY = np.expand_dims(recHitY.content, axis=1)
recHitZ = np.expand_dims(recHitZ.content, axis=1)
recHitSpectatorFlag = np.concatenate(
np.array([
recHit_df_events[i]['spectator_distance'].to_numpy()
for i in range(len(recHit_df_events))
],
dtype=object)).reshape(-1, 1)
recHitSimClusterNumHits = np.concatenate(
np.array([
recHit_df_events[i]['recHitSimClus_nHits'].to_numpy()
for i in range(len(recHit_df_events))
],
dtype=object)).reshape(-1, 1) #number of rec hits
del recHit_df_events
recHitEnergy = self.hitObservable(tree, hits, "energy")
recHitDetaId = self.hitObservable(tree, hits, "detId")
recHitTime = self.hitObservable(tree, hits, "time")
recHitR = np.sqrt(recHitX * recHitX + recHitY * recHitY +
recHitZ * recHitZ)
recHitTheta = np.arccos(recHitZ / recHitR)
recHitEta = -np.log(np.tan(recHitTheta / 2))
# Don't split this until the end, so it can be used to index the truth arrays
recHitSimClusIdx = self.hitObservable(tree,
hits,
"BestMergedSimClusterIdx",
split=False,
flatten=False)
simClusterDepEnergy = tree["MergedSimCluster_recEnergy"].array()
simClusterEnergy = tree["MergedSimCluster_boundaryEnergy"].array()
simClusterEnergyNoMu = tree[
"MergedSimCluster_boundaryEnergyNoMu"].array()
simClusterNumHits = tree["MergedSimCluster_nHits"].array(
) #numebr of sim hits
# Remove muon energy, add back muon deposited energy
unmergedId = tree["SimCluster_pdgId"].array()
unmergedDepE = tree["SimCluster_recEnergy"].array()
unmergedMatchIdx = tree["MergedSimCluster_SimCluster_MatchIdx"].array()
unmergedMatches = tree["MergedSimCluster_SimClusterNumMatch"].array()
unmergedDepEMuOnly = unmergedDepE
unmergedDepEMuOnly[np.abs(unmergedId) != 13] = 0
# Add another layer of nesting, then sum over all unmerged associated to merged
unmergedDepEMuOnly = ak.JaggedArray.fromcounts(
unmergedMatches.counts,
ak.JaggedArray.fromcounts(
unmergedMatches.content,
unmergedDepEMuOnly[unmergedMatchIdx].flatten()))
depEMuOnly = unmergedDepEMuOnly.sum()
#why wasn't it possible to just do instead of all of the above : simClusterEnergy[simClusterPdgId == 13] = simClusterDepEnergy ?
simClusterEnergyMuCorr = simClusterEnergyNoMu + depEMuOnly
simClusterX = tree["MergedSimCluster_impactPoint_x"].array()
simClusterY = tree["MergedSimCluster_impactPoint_y"].array()
simClusterZ = tree["MergedSimCluster_impactPoint_z"].array()
simClusterTime = tree["MergedSimCluster_impactPoint_t"].array()
simClusterEta = tree["MergedSimCluster_impactPoint_eta"].array()
simClusterPhi = tree["MergedSimCluster_impactPoint_phi"].array()
simClusterPdgId = tree["MergedSimCluster_pdgId"].array()
# Mark simclusters outside of volume or with very few hits as noise
# Maybe not a good idea if the merged SC pdgId is screwed up
# Probably removing neutrons is a good idea though
#noNeutrons = simClusterPdgId[recHitSimClusIdx] == 2112
#filter non-boundary positions. Hopefully working?
goodSimClus = tree["MergedSimCluster_isTrainable"].array()
# Don't split by index here to keep same dimensions as SimClusIdx
markNoise = self.truthObjects(~goodSimClus,
recHitSimClusIdx,
False,
split=False,
flatten=False).astype(np.bool_)
nbefore = (recHitSimClusIdx < 0).sum().sum()
recHitSimClusIdx[markNoise] = -1
nafter = (recHitSimClusIdx < 0).sum().sum()
print("Number of noise hits before", nbefore, "after", nafter)
print('removed another factor of', nafter / nbefore,
' bad simclusters')
recHitTruthPID = self.truthObjects(simClusterPdgId, recHitSimClusIdx,
0.)
recHitTruthDepEnergy = self.truthObjects(simClusterDepEnergy,
recHitSimClusIdx, 0)
recHitTruthEnergy = self.truthObjects(simClusterEnergy,
recHitSimClusIdx, 0)
recHitTruthEnergyCorrMu = self.truthObjects(simClusterEnergyMuCorr,
recHitSimClusIdx, 0)
low_energy_shower_cutoff = 3
# Uncorrected currently not used
recHitTruthEnergy = np.where(
recHitTruthEnergy > low_energy_shower_cutoff, recHitTruthEnergy,
recHitTruthDepEnergy)
recHitTruthEnergy = np.where(
recHitTruthEnergyCorrMu > low_energy_shower_cutoff,
recHitTruthEnergyCorrMu, recHitTruthDepEnergy)
#very bad names because these quatities are associated to Merged Clusters and not hits
recHitTruthX = self.truthObjects(simClusterX, recHitSimClusIdx, 0)
recHitTruthY = self.truthObjects(simClusterY, recHitSimClusIdx, 0)
recHitTruthZ = self.truthObjects(simClusterZ, recHitSimClusIdx, 0)
recHitTruthTime = self.truthObjects(simClusterTime, recHitSimClusIdx,
0)
recHitTruthR = np.sqrt(recHitTruthX * recHitTruthX +
recHitTruthY * recHitTruthY +
recHitTruthZ * recHitTruthZ)
recHitTruthTheta = np.arccos(
np.divide(recHitTruthZ,
recHitTruthR,
out=np.zeros_like(recHitTruthZ),
where=recHitTruthR != 0))
recHitTruthPhi = self.truthObjects(simClusterPhi, recHitSimClusIdx, 0)
recHitTruthEta = self.truthObjects(simClusterEta, recHitSimClusIdx, 0)
#recHitAverageEnergy = self.truthObjects(simClusterDepEnergy/simClusterNumHits, recHitSimClusIdx, 0) #this is not technically very good because simClusterNumHits is number of sim clusters, not reco
recHitAverageEnergy = recHitTruthDepEnergy / recHitSimClusterNumHits
#print(recHitTruthPhi)
#print(np.max(recHitTruthPhi))
#print(np.min(recHitTruthPhi))
# Placeholder
zeroFeature = np.zeros(shape=(len(recHitEnergy), 1), dtype='float32')
features = np.concatenate(
[
recHitEnergy,
recHitEta,
zeroFeature, #indicator if it is track or not
recHitTheta,
recHitR,
recHitX,
recHitY,
recHitZ,
recHitTime,
],
axis=1)
farr = SimpleArray(name="recHitFeatures")
farr.createFromNumpy(features, offsets)
del features
recHitSimClusIdx = np.expand_dims(
self.splitJaggedArray(recHitSimClusIdx).content.astype(np.int32),
axis=1)
print('noise', (100 * np.count_nonzero(recHitSimClusIdx < 0)) //
recHitSimClusIdx.shape[0], '% of hits')
print('truth eta min max',
np.min(np.abs(recHitTruthEta[recHitSimClusIdx >= 0])),
np.max(np.abs(recHitTruthEta[recHitSimClusIdx >= 0])))
print(
'non-boundary truth positions',
np.count_nonzero(
np.abs(np.abs(recHitTruthZ[recHitSimClusIdx >= 0]) - 320) > 5)
/ recHitTruthZ[recHitSimClusIdx >= 0].shape[0])
#now all numpy
#Why do we want noise (-1) sim hits to be equal rec?
recHitTruthX[recHitSimClusIdx < 0] = recHitX[recHitSimClusIdx < 0]
recHitTruthY[recHitSimClusIdx < 0] = recHitY[recHitSimClusIdx < 0]
recHitTruthZ[recHitSimClusIdx < 0] = recHitZ[recHitSimClusIdx < 0]
recHitTruthEnergyCorrMu[recHitSimClusIdx < 0] = recHitEnergy[
recHitSimClusIdx < 0]
recHitTruthTime[recHitSimClusIdx < 0] = recHitTime[
recHitSimClusIdx < 0]
#import matplotlib.pyplot as plt
#plt.hist(np.abs(recHitTruthEnergyCorrMu[recHitSimClusIdx>=0]/recHitTruthDepEnergy[recHitSimClusIdx>=0]))
#plt.yscale('log')
#plt.savefig("scat.pdf")
truth = np.concatenate(
[
np.array(recHitSimClusIdx, dtype='float32'), # 0
recHitTruthEnergyCorrMu,
recHitTruthX,
recHitTruthY,
recHitTruthZ, #4
zeroFeature, #truthHitAssignedDirX,
zeroFeature, #6
zeroFeature,
recHitTruthEta,
recHitTruthPhi,
recHitTruthTime, #10
zeroFeature,
zeroFeature,
recHitTruthDepEnergy, #13
zeroFeature, #14
zeroFeature, #15
recHitTruthPID, #16 - 16+n_classes #won't be used anymore
np.array(recHitSpectatorFlag, dtype='float32'),
np.where(recHitTruthZ < front_face_z, 1., 0.).astype('float32')
],
axis=1)
t_idxarr = SimpleArray(recHitSimClusIdx,
offsets,
name="recHitTruthClusterIdx")
t_energyarr = SimpleArray(name="recHitTruthEnergy")
t_energyarr.createFromNumpy(recHitTruthEnergyCorrMu, offsets)
t_posarr = SimpleArray(name="recHitTruthPosition")
t_posarr.createFromNumpy(
np.concatenate([recHitTruthX, recHitTruthY], axis=-1), offsets)
t_time = SimpleArray(name="recHitTruthTime")
t_time.createFromNumpy(recHitTruthTime, offsets)
t_pid = SimpleArray(name="recHitTruthID")
t_pid.createFromNumpy(recHitTruthPID, offsets)
t_spectator = SimpleArray(
name="recHitSpectatorFlag"
) #why do we have inconsistent namings, where is it needed? wrt. to truth array
t_spectator.createFromNumpy(recHitSpectatorFlag.astype('float32'),
offsets)
t_fully_contained = SimpleArray(name="recHitFullyContainedFlag")
t_fully_contained.createFromNumpy(
np.where(recHitTruthZ < front_face_z, 1., 0.).astype('float32'),
offsets)
#remaining truth is mostly for consistency in the plotting tools
t_rest = SimpleArray(name="recHitTruth")
t_rest.createFromNumpy(truth, offsets)
return [
farr, t_idxarr, t_energyarr, t_posarr, t_time, t_pid, t_spectator,
t_fully_contained
], [t_rest], []
def base_convertFromSourceFile(self,
filename,
weighterobjects,
istraining,
treename="WindowNTupler/tree",
removeTracks=True):
fileTimeOut(filename, 10) #10 seconds for eos to recover
tree = uproot.open(filename)[treename]
nevents = tree.numentries
print("n entries: ", nevents)
selection = (tree["recHitEnergy"]).array() > 0
if removeTracks:
selection = np.logical_and(selection,
(tree["recHitID"]).array() > -0.5)
recHitEnergy, rs = self.branchToFlatArray(tree["recHitEnergy"], True,
selection)
recHitEta = self.branchToFlatArray(tree["recHitEta"], False, selection)
#recHitRelPhi = self.branchToFlatArray(tree["recHitRelPhi"], False,selection)
recHitTheta = self.branchToFlatArray(tree["recHitTheta"], False,
selection)
recHitR = self.branchToFlatArray(tree["recHitR"], False, selection)
recHitX = self.branchToFlatArray(tree["recHitX"], False, selection)
recHitY = self.branchToFlatArray(tree["recHitY"], False, selection)
recHitZ = self.branchToFlatArray(tree["recHitZ"], False, selection)
recHitDetID = self.branchToFlatArray(tree["recHitDetID"], False,
selection)
recHitTime = self.branchToFlatArray(tree["recHitTime"], False,
selection)
recHitID = self.branchToFlatArray(tree["recHitID"], False, selection)
recHitPad = self.branchToFlatArray(tree["recHitPad"], False, selection)
## weird shape for this truthHitFractions = self.branchToFlatArray(tree["truthHitFractions"], False)
truthHitAssignementIdx = self.branchToFlatArray(
tree["truthHitAssignementIdx"], False, selection) #0
truthHitAssignedEnergies = self.branchToFlatArray(
tree["truthHitAssignedEnergies"], False, selection) #1
truthHitAssignedX = self.branchToFlatArray(tree["truthHitAssignedX"],
False, selection) #2
truthHitAssignedY = self.branchToFlatArray(tree["truthHitAssignedY"],
False, selection) #3
truthHitAssignedZ = self.branchToFlatArray(tree["truthHitAssignedZ"],
False, selection) #3
truthHitAssignedDirX = self.branchToFlatArray(
tree["truthHitAssignedDirX"], False, selection) #4
truthHitAssignedDirY = self.branchToFlatArray(
tree["truthHitAssignedDirY"], False, selection) #4
truthHitAssignedDirZ = self.branchToFlatArray(
tree["truthHitAssignedDirZ"], False, selection) #4
truthHitAssignedT = self.branchToFlatArray(tree["truthHitAssignedT"],
False, selection)
truthHitAssignedEta = self.branchToFlatArray(
tree["truthHitAssignedEta"], False, selection) #2
truthHitAssignedPhi = self.branchToFlatArray(
tree["truthHitAssignedPhi"], False, selection) #3
truthHitAssignedDirEta = self.branchToFlatArray(
tree["truthHitAssignedDirEta"], False, selection) #4
truthHitAssignedDepEnergies = self.branchToFlatArray(
tree["truthHitAssignedDepEnergies"], False, selection) #4
truthHitAssignedDirR = self.branchToFlatArray(
tree["truthHitAssignedDirR"], False, selection) #4
## weird shape for this truthHitAssignedPIDs = self.branchToFlatArray(tree["truthHitAssignedPIDs"], False)
truthHitAssignedPIDs = self.branchToFlatArray(
tree["truthHitAssignedPIDs"], False, selection, is3d=True)
truthHitAssignedPIDs = np.expand_dims(np.argmax(truthHitAssignedPIDs,
axis=-1),
axis=1) #no one-hot encoding
truthHitAssignedPIDs = np.array(truthHitAssignedPIDs, dtype='float32')
ticlHitAssignementIdx = self.branchToFlatArray(
tree["ticlHitAssignementIdx"], False, selection) #4
ticlHitAssignedEnergies = self.branchToFlatArray(
tree["ticlHitAssignedEnergies"], False, selection) #4
#for now!
truthHitAssignedEnergies = truthHitAssignedDepEnergies #for now rechitsSum
#object weighted energy (1.0 for highest energy hit per object)
features = np.concatenate(
[
recHitEnergy,
recHitEta,
recHitID, #indicator if it is track or not
recHitTheta,
recHitR,
recHitX,
recHitY,
recHitZ,
recHitTime
],
axis=-1)
farr = SimpleArray()
farr.createFromNumpy(features, rs)
del features
truth = np.concatenate(
[
truthHitAssignementIdx, # 0
truthHitAssignedEnergies,
truthHitAssignedX,
truthHitAssignedY,
truthHitAssignedZ, #4
truthHitAssignedDirX,
truthHitAssignedDirY, #6
truthHitAssignedDirZ,
truthHitAssignedEta,
truthHitAssignedPhi,
truthHitAssignedT, #10
truthHitAssignedDirEta,
truthHitAssignedDirR,
truthHitAssignedDepEnergies, #13
ticlHitAssignementIdx, #14
ticlHitAssignedEnergies, #15
truthHitAssignedPIDs #16 - 16+n_classes #won't be used anymore
],
axis=-1)
t_idxarr = SimpleArray()
t_idxarr.createFromNumpy(truthHitAssignementIdx, rs)
t_energyarr = SimpleArray()
t_energyarr.createFromNumpy(truthHitAssignedEnergies, rs)
t_posarr = SimpleArray()
t_posarr.createFromNumpy(
np.concatenate([truthHitAssignedX, truthHitAssignedY], axis=-1),
rs)
t_time = SimpleArray()
t_time.createFromNumpy(truthHitAssignedT, rs)
t_pid = SimpleArray()
t_pid.createFromNumpy(truthHitAssignedPIDs, rs)
#remaining truth is mostly for consistency in the plotting tools
t_rest = SimpleArray()
t_rest.createFromNumpy(truth, rs)
return [farr, t_idxarr, t_energyarr, t_posarr, t_time,
t_pid], [t_rest], []
def base_convertFromSourceFile(self,
filename,
weighterobjects,
istraining,
treename="Events",
removeTracks=True):
fileTimeOut(filename, 10) #10 seconds for eos to recover
tree = uproot.open(filename)[treename]
hits = "RecHitHGC"
recHitZUnsplit = self.hitObservable(tree,
hits,
"z",
split=False,
flatten=False)
self.setSplitIdx(recHitZUnsplit < 0)
recHitZ = self.splitJaggedArray(recHitZUnsplit)
offsets = recHitZ.offsets
recHitZ = np.expand_dims(recHitZ.content, axis=1)
recHitX = self.hitObservable(tree, hits, "x")
recHitY = self.hitObservable(tree, hits, "y")
recHitEnergy = self.hitObservable(tree, hits, "energy")
recHitDetaId = self.hitObservable(tree, hits, "detId")
recHitTime = self.hitObservable(tree, hits, "time")
recHitR = np.sqrt(recHitX * recHitX + recHitY * recHitY +
recHitZ * recHitZ)
recHitTheta = np.arccos(recHitZ / recHitR)
recHitEta = -np.log(np.tan(recHitTheta / 2))
# Don't split this until the end, so it can be used to index the truth arrays
recHitSimClusIdx = self.hitObservable(tree,
hits,
"BestMergedSimClusterIdx",
split=False,
flatten=False)
simClusterDepEnergy = tree["MergedSimCluster_recEnergy"].array()
simClusterEnergy = tree["MergedSimCluster_boundaryEnergy"].array()
simClusterEnergyNoMu = tree[
"MergedSimCluster_boundaryEnergyNoMu"].array()
# Remove muon energy, add back muon deposited energy
unmergedId = tree["SimCluster_pdgId"].array()
unmergedDepE = tree["SimCluster_recEnergy"].array()
unmergedMatchIdx = tree["MergedSimCluster_SimCluster_MatchIdx"].array()
unmergedMatches = tree["MergedSimCluster_SimClusterNumMatch"].array()
unmergedDepEMuOnly = unmergedDepE
unmergedDepEMuOnly[np.abs(unmergedId) != 13] = 0
# Add another layer of nesting, then sum over all unmerged associated to merged
unmergedDepEMuOnly = ak.JaggedArray.fromcounts(
unmergedMatches.counts,
ak.JaggedArray.fromcounts(
unmergedMatches.content,
unmergedDepEMuOnly[unmergedMatchIdx].flatten()))
depEMuOnly = unmergedDepEMuOnly.sum()
simClusterEnergyMuCorr = simClusterEnergyNoMu + depEMuOnly
simClusterX = tree["MergedSimCluster_impactPoint_x"].array()
simClusterY = tree["MergedSimCluster_impactPoint_y"].array()
simClusterZ = tree["MergedSimCluster_impactPoint_z"].array()
simClusterTime = tree["MergedSimCluster_impactPoint_t"].array()
simClusterEta = tree["MergedSimCluster_impactPoint_eta"].array()
simClusterPhi = tree["MergedSimCluster_impactPoint_phi"].array()
simClusterPdgId = tree["MergedSimCluster_pdgId"].array()
# Mark simclusters outside of volume or with very few hits as noise
# Maybe not a good idea if the merged SC pdgId is screwed up
# Probably removing neutrons is a good idea though
#noNeutrons = simClusterPdgId[recHitSimClusIdx] == 2112
#filter non-boundary positions. Hopefully working?
goodSimClus = tree["MergedSimCluster_isTrainable"].array()
# Don't split by index here to keep same dimensions as SimClusIdx
markNoise = self.truthObjects(~goodSimClus,
recHitSimClusIdx,
False,
split=False,
flatten=False).astype(np.bool_)
nbefore = (recHitSimClusIdx < 0).sum().sum()
recHitSimClusIdx[markNoise] = -1
nafter = (recHitSimClusIdx < 0).sum().sum()
print("Number of noise hits before", nbefore, "after", nafter)
print('removed another factor of', nafter / nbefore,
' bad simclusters')
recHitTruthPID = self.truthObjects(simClusterPdgId, recHitSimClusIdx,
0.)
recHitTruthEnergy = self.truthObjects(simClusterEnergy,
recHitSimClusIdx, 0)
recHitTruthDepEnergy = self.truthObjects(simClusterDepEnergy,
recHitSimClusIdx, 0)
recHitTruthEnergyCorrMu = self.truthObjects(simClusterEnergyMuCorr,
recHitSimClusIdx, 0)
recHitTruthX = self.truthObjects(simClusterX, recHitSimClusIdx, 0)
recHitTruthY = self.truthObjects(simClusterY, recHitSimClusIdx, 0)
recHitTruthZ = self.truthObjects(simClusterZ, recHitSimClusIdx, 0)
recHitTruthTime = self.truthObjects(simClusterTime, recHitSimClusIdx,
0)
recHitTruthR = np.sqrt(recHitTruthX * recHitTruthX +
recHitTruthY * recHitTruthY +
recHitTruthZ * recHitTruthZ)
recHitTruthTheta = np.arccos(
np.divide(recHitTruthZ,
recHitTruthR,
out=np.zeros_like(recHitTruthZ),
where=recHitTruthR != 0))
recHitTruthPhi = self.truthObjects(simClusterPhi, recHitSimClusIdx, 0)
recHitTruthEta = self.truthObjects(simClusterEta, recHitSimClusIdx, 0)
#print(recHitTruthPhi)
#print(np.max(recHitTruthPhi))
#print(np.min(recHitTruthPhi))
# Placeholder
zeroFeature = np.zeros(shape=(len(recHitEnergy), 1), dtype='float32')
features = np.concatenate(
[
recHitEnergy,
recHitEta,
zeroFeature, #indicator if it is track or not
recHitTheta,
recHitR,
recHitX,
recHitY,
recHitZ,
recHitTime,
],
axis=1)
farr = SimpleArray(name="recHitFeatures")
farr.createFromNumpy(features, offsets)
del features
recHitSimClusIdx = np.expand_dims(
self.splitJaggedArray(recHitSimClusIdx).content.astype(np.int32),
axis=1)
print('noise', (100 * np.count_nonzero(recHitSimClusIdx < 0)) //
recHitSimClusIdx.shape[0], '% of hits')
print('truth eta min max',
np.min(np.abs(recHitTruthEta[recHitSimClusIdx >= 0])),
np.max(np.abs(recHitTruthEta[recHitSimClusIdx >= 0])))
print(
'non-boundary truth positions',
np.count_nonzero(
np.abs(np.abs(recHitTruthZ[recHitSimClusIdx >= 0]) - 320) > 5)
/ recHitTruthZ[recHitSimClusIdx >= 0].shape[0])
#now all numpy
recHitTruthX[recHitSimClusIdx < 0] = recHitX[recHitSimClusIdx < 0]
recHitTruthY[recHitSimClusIdx < 0] = recHitY[recHitSimClusIdx < 0]
recHitTruthZ[recHitSimClusIdx < 0] = recHitZ[recHitSimClusIdx < 0]
recHitTruthEnergyCorrMu[recHitSimClusIdx < 0] = recHitEnergy[
recHitSimClusIdx < 0]
recHitTruthTime[recHitSimClusIdx < 0] = recHitTime[
recHitSimClusIdx < 0]
#import matplotlib.pyplot as plt
#plt.hist(np.abs(recHitTruthEnergyCorrMu[recHitSimClusIdx>=0]/recHitTruthDepEnergy[recHitSimClusIdx>=0]))
#plt.yscale('log')
#plt.savefig("scat.pdf")
truth = np.concatenate(
[
np.array(recHitSimClusIdx, dtype='float32'), # 0
recHitTruthEnergyCorrMu,
recHitTruthX,
recHitTruthY,
recHitTruthZ, #4
zeroFeature, #truthHitAssignedDirX,
zeroFeature, #6
zeroFeature,
recHitTruthEta,
recHitTruthPhi,
recHitTruthTime, #10
zeroFeature,
zeroFeature,
recHitTruthDepEnergy, #13
zeroFeature, #14
zeroFeature, #15
recHitTruthPID #16 - 16+n_classes #won't be used anymore
],
axis=1)
t_idxarr = SimpleArray(recHitSimClusIdx,
offsets,
name="recHitTruthClusterIdx")
t_energyarr = SimpleArray(name="recHitTruthEnergy")
t_energyarr.createFromNumpy(recHitTruthEnergy, offsets)
t_posarr = SimpleArray(name="recHitTruthPosition")
t_posarr.createFromNumpy(
np.concatenate([recHitTruthX, recHitTruthY], axis=-1), offsets)
t_time = SimpleArray(name="recHitTruthTime")
t_time.createFromNumpy(recHitTruthTime, offsets)
t_pid = SimpleArray(name="recHitTruthID")
t_pid.createFromNumpy(recHitTruthPID, offsets)
#remaining truth is mostly for consistency in the plotting tools
t_rest = SimpleArray(name="recHitTruth")
t_rest.createFromNumpy(truth, offsets)
return [farr, t_idxarr, t_energyarr, t_posarr, t_time,
t_pid], [t_rest], []
def getFinalFeaturesSA(self):
a,rs = self.getFinalFeaturesNumpy()
sa = SimpleArray(a,rs,name="recHitFeatures")
#sa.setFeatureNames(self.featurenames) #not yet
return sa
def base_convertFromSourceFile(self,
filename,
weighterobjects,
istraining,
onlytruth,
treename="WindowNTupler/tree"):
fileTimeOut(filename, 10) #10 seconds for eos to recover
tree = uproot.open(filename)[treename]
nevents = tree.numentries
print("n entries: ", nevents)
selection = None
if onlytruth:
selection = (tree["truthHitAssignementIdx"]).array() > -0.1 #0
#remove zero energy hits from removing of bad simclusters
if selection is None:
selection = (tree["recHitEnergy"]).array() > 0
else:
selection = np.logical_and(selection,
(tree["recHitEnergy"]).array() > 0)
recHitEnergy, rs = self.branchToFlatArray(tree["recHitEnergy"], True,
selection)
recHitEta = self.branchToFlatArray(tree["recHitEta"], False, selection)
#recHitRelPhi = self.branchToFlatArray(tree["recHitRelPhi"], False,selection)
recHitTheta = self.branchToFlatArray(tree["recHitTheta"], False,
selection)
recHitR = self.branchToFlatArray(tree["recHitR"], False, selection)
recHitX = self.branchToFlatArray(tree["recHitX"], False, selection)
recHitY = self.branchToFlatArray(tree["recHitY"], False, selection)
recHitZ = self.branchToFlatArray(tree["recHitZ"], False, selection)
recHitDetID = self.branchToFlatArray(tree["recHitDetID"], False,
selection)
recHitTime = self.branchToFlatArray(tree["recHitTime"], False,
selection)
recHitID = self.branchToFlatArray(tree["recHitID"], False, selection)
recHitPad = self.branchToFlatArray(tree["recHitPad"], False, selection)
## weird shape for this truthHitFractions = self.branchToFlatArray(tree["truthHitFractions"], False)
truthHitAssignementIdx = self.branchToFlatArray(
tree["truthHitAssignementIdx"], False, selection) #0
truthHitAssignedEnergies = self.branchToFlatArray(
tree["truthHitAssignedEnergies"], False, selection) #1
truthHitAssignedX = self.branchToFlatArray(tree["truthHitAssignedX"],
False, selection) #2
truthHitAssignedY = self.branchToFlatArray(tree["truthHitAssignedY"],
False, selection) #3
truthHitAssignedZ = self.branchToFlatArray(tree["truthHitAssignedZ"],
False, selection) #3
truthHitAssignedDirX = self.branchToFlatArray(
tree["truthHitAssignedDirX"], False, selection) #4
truthHitAssignedDirY = self.branchToFlatArray(
tree["truthHitAssignedDirY"], False, selection) #4
truthHitAssignedDirZ = self.branchToFlatArray(
tree["truthHitAssignedDirZ"], False, selection) #4
truthHitAssignedEta = self.branchToFlatArray(
tree["truthHitAssignedEta"], False, selection) #2
truthHitAssignedPhi = self.branchToFlatArray(
tree["truthHitAssignedPhi"], False, selection) #3
#truthHitAssignedR = self.branchToFlatArray(tree["truthHitAssignedR"], False,selection) #3
truthHitAssignedDirEta = self.branchToFlatArray(
tree["truthHitAssignedDirEta"], False, selection) #4
truthHitAssignedDirPhi = self.branchToFlatArray(
tree["truthHitAssignedDirPhi"], False, selection) #4
truthHitAssignedDirR = self.branchToFlatArray(
tree["truthHitAssignedDirR"], False, selection) #4
## weird shape for this truthHitAssignedPIDs = self.branchToFlatArray(tree["truthHitAssignedPIDs"], False)
truthHitAssignedPIDs = self.branchToFlatArray(
tree["truthHitAssignedPIDs"], False, selection, is3d=True)
#print('truthHitAssignedPIDs',truthHitAssignedPIDs.shape)
#print('truthHitAssignedEnergies',truthHitAssignedEnergies.shape)
#print(truthHitAssignedPIDs)
#truthHitAssignedPIDs = np.zeros_like(truthHitAssignedEnergies)
#truthHitAssignedPIDs = np.tile(truthHitAssignedPIDs, [1, n_classes])
#type_ambiguous,
#type_electron,
#type_photon,
#type_mip,
#type_charged_hadron,
#type_neutral_hadron,
truthHitAssignedT = self.branchToFlatArray(tree["truthHitAssignedT"],
False, selection)
ticlHitAssignementIdx = self.branchToFlatArray(
tree["ticlHitAssignementIdx"], False, selection) #4
ticlHitAssignedEnergies = self.branchToFlatArray(
tree["ticlHitAssignedEnergies"], False, selection) #4
# For calculating spectators
rechitsSum = findRechitsSum(truthHitAssignementIdx, recHitEnergy, rs)
spectator = np.where(recHitEnergy < 0.0005 * rechitsSum,
np.ones_like(recHitEnergy),
np.zeros_like(recHitEnergy))
# If truth shower energy < 5% of sum of rechits, assign rechits sum to it instead
truthShowerEnergies = truthHitAssignedEnergies.copy()
#take them as is
#truthShowerEnergies[rechitsSum<0.25*truthHitAssignedEnergies] = rechitsSum[rechitsSum<0.25*truthHitAssignedEnergies]
#for now!
truthShowerEnergies = rechitsSum
features = np.concatenate(
[
recHitEnergy,
recHitEta,
truthHitAssignementIdx, #no phi anymore!
truthShowerEnergies,
recHitR,
recHitX,
recHitY,
recHitZ,
recHitTime
],
axis=-1)
farr = SimpleArray()
farr.createFromNumpy(features, rs)
#farr.cout()
print("features", features.shape)
del features
truth = np.concatenate(
[
truthHitAssignementIdx, # 0
truthShowerEnergies,
truthHitAssignedX,
truthHitAssignedY,
truthHitAssignedZ, #4
truthHitAssignedDirX,
truthHitAssignedDirY, #6
truthHitAssignedDirZ,
truthHitAssignedEta,
truthHitAssignedPhi,
truthHitAssignedT, #10
truthHitAssignedDirEta,
truthHitAssignedDirPhi, #12
truthHitAssignedDirR,
spectator, #14
truthHitAssignedEnergies, #15
rechitsSum, #16
ticlHitAssignementIdx, #17
ticlHitAssignedEnergies, #18
truthHitAssignedPIDs #19 - 19+n_classes
],
axis=-1)
tarr = SimpleArray()
tarr.createFromNumpy(truth, rs)
print("truth", truth.shape)
return [farr], [tarr], []
def base_convertFromSourceFile(self,
filename,
weighterobjects,
istraining,
onlytruth,
treename="WindowNTupler/tree",
removeTracks=True):
fileTimeOut(filename, 10) #10 seconds for eos to recover
tree = uproot.open(filename)[treename]
nevents = tree.numentries
print("n entries: ", nevents)
selection = None
if onlytruth:
selection = (tree["truthHitAssignementIdx"]).array() > -0.1 #0
#remove zero energy hits from removing of bad simclusters
if selection is None:
selection = (tree["recHitEnergy"]).array() > 0
else:
selection = np.logical_and(selection,
(tree["recHitEnergy"]).array() > 0)
if removeTracks:
selection = np.logical_and(selection,
(tree["recHitID"]).array() > -0.5)
recHitEnergy, rs = self.branchToFlatArray(tree["recHitEnergy"], True,
selection)
recHitEta = self.branchToFlatArray(tree["recHitEta"], False, selection)
#recHitRelPhi = self.branchToFlatArray(tree["recHitRelPhi"], False,selection)
recHitTheta = self.branchToFlatArray(tree["recHitTheta"], False,
selection)
recHitR = self.branchToFlatArray(tree["recHitR"], False, selection)
recHitX = self.branchToFlatArray(tree["recHitX"], False, selection)
recHitY = self.branchToFlatArray(tree["recHitY"], False, selection)
recHitZ = self.branchToFlatArray(tree["recHitZ"], False, selection)
recHitDetID = self.branchToFlatArray(tree["recHitDetID"], False,
selection)
recHitTime = self.branchToFlatArray(tree["recHitTime"], False,
selection)
recHitID = self.branchToFlatArray(tree["recHitID"], False, selection)
recHitPad = self.branchToFlatArray(tree["recHitPad"], False, selection)
## weird shape for this truthHitFractions = self.branchToFlatArray(tree["truthHitFractions"], False)
truthHitAssignementIdx = self.branchToFlatArray(
tree["truthHitAssignementIdx"], False, selection) #0
truthHitAssignedEnergies = self.branchToFlatArray(
tree["truthHitAssignedEnergies"], False, selection) #1
truthHitAssignedX = self.branchToFlatArray(tree["truthHitAssignedX"],
False, selection) #2
truthHitAssignedY = self.branchToFlatArray(tree["truthHitAssignedY"],
False, selection) #3
truthHitAssignedZ = self.branchToFlatArray(tree["truthHitAssignedZ"],
False, selection) #3
truthHitAssignedDirX = self.branchToFlatArray(
tree["truthHitAssignedDirX"], False, selection) #4
truthHitAssignedDirY = self.branchToFlatArray(
tree["truthHitAssignedDirY"], False, selection) #4
truthHitAssignedDirZ = self.branchToFlatArray(
tree["truthHitAssignedDirZ"], False, selection) #4
truthHitAssignedEta = self.branchToFlatArray(
tree["truthHitAssignedEta"], False, selection) #2
truthHitAssignedPhi = self.branchToFlatArray(
tree["truthHitAssignedPhi"], False, selection) #3
#truthHitAssignedR = self.branchToFlatArray(tree["truthHitAssignedR"], False,selection) #3
truthHitAssignedDirEta = self.branchToFlatArray(
tree["truthHitAssignedDirEta"], False, selection) #4
truthHitAssignedDepEnergies = self.branchToFlatArray(
tree["truthHitAssignedDepEnergies"], False, selection) #4
truthHitAssignedDirR = self.branchToFlatArray(
tree["truthHitAssignedDirR"], False, selection) #4
## weird shape for this truthHitAssignedPIDs = self.branchToFlatArray(tree["truthHitAssignedPIDs"], False)
truthHitAssignedPIDs = self.branchToFlatArray(
tree["truthHitAssignedPIDs"], False, selection, is3d=True)
#print('truthHitAssignedPIDs',truthHitAssignedPIDs.shape)
#print('truthHitAssignedEnergies',truthHitAssignedEnergies.shape)
#print(truthHitAssignedPIDs)
#truthHitAssignedPIDs = np.zeros_like(truthHitAssignedEnergies)
#truthHitAssignedPIDs = np.tile(truthHitAssignedPIDs, [1, n_classes])
#type_ambiguous,
#type_electron,
#type_photon,
#type_mip,
#type_charged_hadron,
#type_neutral_hadron,
# For tracks
#
# *(data++) = t->obj->p(); *(data++) = recHit->hit->energy();
#### *(data++) = t->pos.eta(); *(data++) = recHit->pos.eta();
#### *(data++) = t->pos.phi(); *(data++) = recHit->pos.phi();
#### *(data++) = t->pos.theta(); *(data++) = recHit->pos.theta();
#### *(data++) = t->pos.mag(); *(data++) = recHit->pos.mag();
#### *(data++) = t->pos.x(); *(data++) = recHit->pos.x();
#### *(data++) = t->pos.y(); *(data++) = recHit->pos.y();
#### *(data++) = t->pos.z(); *(data++) = recHit->pos.z();
#### *(data++) = t->obj->charge(); *(data++) = (float)recHit->hit->detid();
#### *(data++) = t->obj->chi2(); *(data++) = recHit->hit->time();
#### *(data++) = -1.; //track ID bit *(data++) = 0.; //rechit ID bit
#### *(data++) = 0.; //pad *(data++) = 0.; //pad
#
#
#
#make these the only spectators, and set 'energy' to zero
truthHitAssignedT = self.branchToFlatArray(tree["truthHitAssignedT"],
False, selection)
ticlHitAssignementIdx = self.branchToFlatArray(
tree["ticlHitAssignementIdx"], False, selection) #4
ticlHitAssignedEnergies = self.branchToFlatArray(
tree["ticlHitAssignedEnergies"], False, selection) #4
# For calculating spectators
#rechitsSum = findRechitsSum(truthHitAssignementIdx, recHitEnergy, rs)
#spectator = np.where(recHitEnergy < 0.0005 * rechitsSum, np.ones_like(recHitEnergy), np.zeros_like(recHitEnergy))
############ special track treatment for now
#make tracks spectators
isTrack = recHitID < 0
spectator = np.where(isTrack, np.ones_like(isTrack),
np.zeros_like(isTrack))
recHitEnergy[isTrack] = 0. #don't use track momenta just use as seeds
##############
# If truth shower energy < 5% of sum of rechits, assign rechits sum to it instead
truthShowerEnergies = truthHitAssignedEnergies.copy()
#take them as is
#truthShowerEnergies[rechitsSum<0.25*truthHitAssignedEnergies] = rechitsSum[rechitsSum<0.25*truthHitAssignedEnergies]
#for now!
truthShowerEnergies = truthHitAssignedDepEnergies #for now rechitsSum
features = np.concatenate(
[
recHitEnergy,
recHitEta,
recHitID, #indicator if it is track or not
recHitTheta,
recHitR,
recHitX,
recHitY,
recHitZ,
recHitTime
],
axis=-1)
np.savetxt("textarr.txt", features[0:rs[1]])
farr = SimpleArray()
farr.createFromNumpy(features, rs)
#farr.cout()
print("features", features.shape)
del features
truth = np.concatenate(
[
truthHitAssignementIdx, # 0
truthShowerEnergies,
truthHitAssignedX,
truthHitAssignedY,
truthHitAssignedZ, #4
truthHitAssignedDirX,
truthHitAssignedDirY, #6
truthHitAssignedDirZ,
truthHitAssignedEta,
truthHitAssignedPhi,
truthHitAssignedT, #10
truthHitAssignedDirEta,
np.zeros_like(truthHitAssignedDirEta), #12
truthHitAssignedDirR,
spectator, #14
truthHitAssignedEnergies, #15
truthHitAssignedDepEnergies, #16
ticlHitAssignementIdx, #17
ticlHitAssignedEnergies, #18
truthHitAssignedPIDs #19 - 19+n_classes
],
axis=-1)
tarr = SimpleArray()
tarr.createFromNumpy(truth, rs)
print("truth", truth.shape)
return [farr], [tarr], []