def main():
# Parameters
data_directory = '../../data/generated-data-r-10-n-6-4/'
features_path = '../../data/features-generated-data-r-10-n-6-4'
booking_file = '../../data/booking.csv'
users_file = '../../data/user.csv'
rating_thresholds = []
true_objects_indexes = [0, 1, 2, 3, 4, 5]
false_objects_indexes = [6, 7, 8, 9]
file_names = os.listdir(data_directory)
img_ids_vector = [int(name.split('-')[0]) for name in file_names]
ratings_vector = [int(name.split('-')[-2]) for name in file_names]
name_vector = [data_directory + name for name in file_names]
images_indexes = [name.split('-')[3].split('.')[0] for name in file_names]
ratings_matrix, images_indexes_for_id, ids_indexes, users_matrix = load_data(
data_directory, booking_file, users_file, rating_thresholds)
features = get_features(features_path, name_vector)
fa = FeatureAgglomeration(n_clusters=50)
fa.fit(features)
features = fa.transform(features)
scores_auc = []
scores_rmse = []
for i in range(10):
cv_results_file = '../results/cv-generated-data-r-10-n-6-4-rf-fa-' + str(
i) + '.csv'
selection = ObjectSelection(show_selection_results=False,
selection_algorithm='rf')
selection.transform(ids=img_ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=images_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector,
z_score=False)
selection.evaluate(evaluation_metric='auc')
selection.evaluate(evaluation_metric='rmse')
print('\n\n-----\n\n')
score_auc, score_rmse = selection.evaluate(evaluation_metric='auc')
scores_auc.append(score_auc)
scores_rmse.append(score_rmse)
results_file = '../scores/generated-data-r-10-n-6-4-rf-fa-auc.csv'
save_scores(scores_auc, results_file)
results_file = '../scores/generated-data-r-10-n-6-4-rf-fa-rmse.csv'
save_scores(scores_rmse, results_file)
ratings_matrix, images_indexes_for_id, ids_indexes, users_matrix = load_data(
data_directory, booking_file, users_file, rating_thresholds)
features, new_ratings_vector, new_categories_vector, new_ids_vector, new_paths_vector, text_indexes = divide_texts(
name_vector, ratings_vector, categories_vector, ids_vector, n=10)
ratings_vector = new_ratings_vector
ids_vector = new_ids_vector
scores_auc = []
scores_rmse = []
for i in range(10):
cv_results_file = '../results/cv-generated-data-r-10-n-02-random-' + str(
i) + '.csv'
selection = ObjectSelection(show_selection_results=False,
selection_algorithm='random')
selection.transform(ids=ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=text_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector,
z_score=False)
score_auc, score_rmse = selection.evaluate(evaluation_metric='auc')
scores_auc.append(score_auc)
scores_rmse.append(score_rmse)
def main():
# Parameters
data_directory = '../data/generated-data-r-10-n-8-2/'
features_path = '../data/features-generated-data-r-10-n-8-2'
booking_file = '../data/booking.csv'
users_file = '../data/user.csv'
rating_thresholds = []
true_objects_indexes = [0, 1, 2, 3, 4, 5, 6, 7]
false_objects_indexes = [8, 9]
file_names = os.listdir(data_directory)
img_ids_vector = [int(name.split('-')[0]) for name in file_names]
ratings_vector = [int(name.split('-')[-2]) for name in file_names]
name_vector = [data_directory + name for name in file_names]
images_indexes = [name.split('-')[3].split('.')[0] for name in file_names]
ratings_matrix, images_indexes_for_id, ids_indexes, users_matrix = load_data(
data_directory, booking_file, users_file, rating_thresholds)
features = get_features(features_path, name_vector)
fa = FeatureAgglomeration(n_clusters=50)
fa.fit(features)
features = fa.transform(features)
scores = []
cv_results_file = './results/bf_real.csv'
#ratings_matrix = ratings_matrix[:30, :30]
#selection = BasicFactorization(show_selection_results=False, selection_algorithm='random')
#selection.transform(ids=img_ids_vector, features=features, ratings=ratings_vector, users_ratings=ratings_matrix,
# users=users_matrix, cv_results_file=cv_results_file, images_indexes=images_indexes,
# true_objects_indexes=true_objects_indexes, false_objects_indexes=false_objects_indexes,
# paths=name_vector, z_score=True)
#score, score_rmse = selection.evaluate(evaluation_metric='auc')
#scores.append(score)
#exit()
# K Nearest Neighbors
#cv_results_file = './results/cv-generated-data-nr-2-n-02-l-100-knn.csv'
scores_auc = []
scores_rmse = []
for i in range(1):
cv_results_file = './results/xxp1-cv-generated-data-r-10-n-8-2-random-' + str(
i) + '.csv'
selection = ObjectSelection(show_selection_results=False,
selection_algorithm='random')
selection.transform(ids=img_ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=images_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector,
z_score=False)
selection.evaluate(evaluation_metric='auc')
selection.evaluate(evaluation_metric='rmse')
print('\n\n-----\n\n')
score_auc, score_rmse = selection.evaluate(evaluation_metric='auc')
scores_auc.append(score_auc)
scores_rmse.append(score_rmse)
results_file = './scores/v-generated-data-r-10-n-8-2-random-fa-auc.csv'
save_scores(scores_auc, results_file)
results_file = './scores/v-generated-data-r-10-n-8-2-random-fa-rmse.csv'
save_scores(scores_rmse, results_file)
exit()
for i in range(10):
print()
for _ in range(0):
selection = ObjectSelection(show_selection_results=False,
selection_algorithm='random')
# selection.transform(ids=img_ids_vector, features=features, ratings=ratings_vector, users_ratings=ratings_matrix, users=users_matrix, cv_results_file=cv_results_file)
selection.transform(ids=img_ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=images_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector,
z_score=True)
print('\n\n-----\n\n')
score_auc, score_rmse = selection.evaluate(evaluation_metric='auc')
scores.append(score_auc)
for i in range(10):
print()
for _ in range(10):
selection = BasicFactorization(show_selection_results=False,
selection_algorithm='random')
selection.transform(ids=img_ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=images_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector)
score = selection.evaluate(evaluation_metric='auc')
scores.append(score)
exit()
# Parameters
#data_directory = '../data/experience-6/'
#features_path = '../data/features-experience-6'
data_directory = '../data/generated-data-r-2-n-8-2/'
features_path = '../data/features-generated-data-r-2-n-8-2'
booking_file = '../data/booking.csv'
users_file = '../data/user.csv'
cv_results_file = 'results/cv-generated-data-r-2-n-8-2-x.csv'
true_objects_indexes = [0, 1, 2, 3, 4, 5, 6, 7]
false_objects_indexes = [8, 9]
#file_to_delete = data_directory + '.DS_Store'
#os.remove(file_to_delete)
file_names = os.listdir(data_directory)
img_ids_vector = [int(name.split('-')[0]) for name in file_names]
ratings_vector = [int(name.split('-')[-2]) for name in file_names]
name_vector = [data_directory + name for name in file_names]
images_indexes = [name.split('-')[3].split('.')[0] for name in file_names]
rating_thresholds = [1, 2]
#rating_thresholds = []
ratings_matrix, images_indexes_for_id, ids_indexes, users_matrix = load_data(
data_directory,
booking_file,
users_file,
rating_thresholds,
binary=True)
features = get_features(features_path, name_vector)
cv_results_file = './results/cv-generated-data-r-2-n-8-2-knn-y.csv'
selection = ObjectSelection(show_selection_results=False,
selection_algorithm='random')
selection.transform(ids=img_ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=images_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector,
use_user_data=True)
selection.evaluate(evaluation_metric='auc')
exit()
selection = BasicFactorizationNmf(show_selection_results=True,
selection_algorithm='random')
selection.transform(ids=img_ids_vector,
features=features,
ratings=ratings_vector,
users_ratings=ratings_matrix,
users=users_matrix,
cv_results_file=cv_results_file,
images_indexes=images_indexes,
true_objects_indexes=true_objects_indexes,
false_objects_indexes=false_objects_indexes,
paths=name_vector)
selection.evaluate(evaluation_metric='auc')
def make_scribblers():
# JSON scribbler. Produces:
# - inCertifiedLumiSections [bool]
json_path = os.path.join(
os.getcwd(),
"data/Cert_271036-284044_13TeV_23Sep2016ReReco_Collisions16_JSON.txt")
json_scribbler = in_certified_lumi_sections(json_path)
# PUWeight. Produces:
# - puWeight [float]
puweight_scribbler = NanoaodtoolsModuleWrapper(
"postprocessing.modules.common.puWeightProducer",
"puWeight",
mc_only=True,
)
# lepton SF. Produces:
# - Muon_effSF
# - Electron_effSF
leptonsf_scribbler = NanoaodtoolsModuleWrapper(
"postprocessing.modules.common.lepSFProducer",
"lepSF",
mc_only=True,
)
# JEC Uncs. Produces:
# - Jet_jecUncert*
jecunc_scribbler = NanoaodtoolsModuleWrapper(
"postprocessing.modules.jme.jecUncertainties",
"jecUncert",
mc_only=True,
)
# JetMET Uncs. Produces:
# - Jet_pt_smeared
# - MET_{pt,phi}_smeared
# - Jet_pt_{jer,jes*,unclustEn}{Up,Down}
# - MET_{pt,phi}_{jer,jes*,unclustEn}{Up,Down}
jetmetunc_scribbler = NanoaodtoolsModuleWrapper(
"postprocessing.modules.jme.jetmetUncertainties",
"jetmetUncertainties",
mc_only=True,
)
# BTagSF. Produces:
# - Jet_btagSF
# - Jet_btagSF_{up,down}
btagsf_scribbler = NanoaodtoolsModuleWrapper(
"postprocessing.modules.btv.btagSFProducer",
"btagSF",
mc_only=True,
)
# Object vetoes
jet_veto = dict(All=(
'j: j.pt>40.',
'j: abs(j.eta)<5.',
'j: j.puId>=1', # loose
'j: j.jetId>=1', # loose
))
muon_veto = dict(All=(
'u: u.pt>10.',
'u: abs(u.eta)<2.4',
'u: abs(u.dxy)<0.118',
'u: abs(u.dz)<0.882',
))
electron_veto = dict(All=(
'e: e.pt>10.',
'e: abs(e.eta)<2.1',
'e: abs(e.dxy)<0.5',
'e: abs(e.dz)<1.0',
'e: e.miniPFRelIso_all<0.1',
'e: e.lostHits<=1',
'e: e.cutBased>=1', # Veto or higher
))
photon_veto = dict(All=(
'y: y.pt>25.',
'y: abs(y.eta)<2.5',
'y: y.cutBased>=1', # Loose of higher
))
# Object selections
jet_selection = dict(All=('j: abs(j.eta)<2.4', ))
muon_selection = dict(All=(
'u: u.pt>30.',
'u: abs(u.eta)<2.1',
'u: u.tightId',
))
electron_selection = dict(All=(
'e: e.pt>30.',
'e: abs(e.eta)<2.1',
'e: e.cutBased>=4', # Tight or higher
))
photon_selection = dict(All=(
'y: y.pt>165.',
'y: abs(y.eta)<1.45',
'y: y.cutBased>=3', # Tight or higher
))
obj_attrs = ['pt', 'eta', 'phi']
return [
Collection("Jet", attrs=obj_attrs + ["puId", "jetId"]),
Collection("Muon",
attrs=obj_attrs +
["dxy", "dz", "miniPFRelIso_all", "tightId", "jetIdx"]),
Collection("Electron",
attrs=obj_attrs + [
"dxy", "dz", "miniPFRelIso_all", "lostHits", "cutBased",
"jetIdx"
]),
Collection("Photon", attrs=obj_attrs + ["cutBased", "jetIdx"]),
Collection("Tau", attrs=obj_attrs),
ObjectSelection(in_obj_name="Muon",
out_obj_name="MuonVeto",
path_cfg=muon_veto),
ObjectSelection(in_obj_name="MuonVeto",
out_obj_name="MuonSelection",
path_cfg=muon_selection),
ObjectSelection(in_obj_name="Electron",
out_obj_name="ElectronVeto",
path_cfg=electron_veto),
ObjectSelection(in_obj_name="ElectronVeto",
out_obj_name="ElectronSelection",
path_cfg=electron_selection),
ObjectSelection(in_obj_name="Photon",
out_obj_name="PhotonVeto",
path_cfg=photon_veto),
ObjectSelection(in_obj_name="PhotonVeto",
out_obj_name="PhotonSelection",
path_cfg=photon_selection),
ObjectSelection(in_obj_name="Jet",
out_obj_name="JetVeto",
path_cfg=jet_veto),
OverlapRemoval(
collection_name="JetVeto",
ref_collection="MuonVeto",
),
OverlapRemoval(
collection_name="JetVeto",
ref_collection="ElectronVeto",
),
OverlapRemoval(
collection_name="JetVeto",
ref_collection="PhotonVeto",
),
ObjectSelection(in_obj_name="JetVeto",
out_obj_name="JetSelection",
path_cfg=jet_selection),
DatasetInfo(),
json_scribbler,
cutflowId(),
MetNoX(),
#leptonsf_scribbler,
#jecunc_scribbler,
#btagsf_scribbler,
]