def f(x, p, aim):
'''x : variable de contrôle
p : paramètre du réseau
renvoie le temps de trajet total pour (x,p) calculé avec Aimsun'''
Aimsun_simul(x, p, aim)
TTT = data.getTotalTravelTime(file)
return TTT
for path_data_set in [path_train_set, path_cv_set, path_test_set]:
print("On", path_data_set.split('_')[-1], 'set')
data_set = h5py.File(path_data_set, 'r')
pos_topleft_coord = np.array(data_set['positive_example'])
neg_topleft_coord = np.array(data_set['negative_example'])
raw_image = np.array(data_set['raw_image'])
road_mask = np.array(data_set['road_mask'])
data_set.close()
if classifier.classifier_type == 'LR':
data_extractor = Data_Extractor(raw_image,
road_mask,
size,
pos_topleft_coord=pos_topleft_coord,
neg_topleft_coord=neg_topleft_coord,
normalization=norm)
else:
assert classifier.classifier_type == 'FCN'
data_extractor = FCN_Data_Extractor(
raw_image,
road_mask,
size,
pos_topleft_coord=pos_topleft_coord,
neg_topleft_coord=neg_topleft_coord,
normalization=norm)
gc.collect()
train_neg_topleft_coord = np.array(train_set['negative_example'])
train_raw_image = np.array(train_set['raw_image'])
train_road_mask = np.array(train_set['road_mask'])
train_set.close()
# Load cross-validation set
CV_set = h5py.File(path_cv_set, 'r')
CV_pos_topleft_coord = np.array(CV_set['positive_example'])
CV_neg_topleft_coord = np.array(CV_set['negative_example'])
CV_raw_image = np.array(CV_set['raw_image'])
CV_road_mask = np.array(CV_set['road_mask'])
CV_set.close()
Train_Data = Data_Extractor(train_raw_image,
train_road_mask,
size,
pos_topleft_coord=train_pos_topleft_coord,
neg_topleft_coord=train_neg_topleft_coord,
normalization=norm)
# run garbage collector
gc.collect()
CV_Data = Data_Extractor(CV_raw_image,
CV_road_mask,
size,
pos_topleft_coord=CV_pos_topleft_coord,
neg_topleft_coord=CV_neg_topleft_coord,
normalization=norm)
# run garbage collector
gc.collect()
print("train data:")
"""
Main Setup file to run the whole process for s given source data
"""
import os
import time
from sys import argv, path
path.append(
"/home/ahmedr/Documents/Backend-Retailstreets/venv/lib/python3.6/site-packages"
)
import Data_Extractor
import Patterns_Recognition
input_file_path = ""
try:
if len(argv[1]) > 0 and os.stat("data/source_file/" +
str(argv[1])).st_size > 0:
input_file_path = "data/source_file/" + str(argv[1])
except Exception:
exit("file not found: " + str(argv[1]))
os.system("python3 Reset_files.py")
texted_file = Data_Extractor.main(input_file_path)
Patterns_Recognition.main(texted_file)
time.sleep(10)
os.system("python3 PotentialMatch_Cleanser.py")
time.sleep(5)
os.system("python3 Address_Validator.py")
time.sleep(5)
os.system("python3 Information_Fetcher.py")
def Aimsun_q():
'''renvoie le paramètre exogène q du queuing model
sortie : array (p,(k,gamma)) taille N²,2*N'''
q = data.getQueuingModelParameters(file2)
return q
def Aimsun_p():
'''renvoie le paramètre exogène p d'Aimsun
sortie : [id_CP, id_n, cyc. dur., avail. cyc. ratio, nb of phases, start index]'''
p = data.getNetworkParameters(file2)[1]
return p
def Aimsun_M():
'''renvoie le nombre de noeuds du réseau'''
M = data.getNetworkParameters(file2)[0]
return M
def Aimsun_N():
'''renvoie le nombre de files dans le métamodèle'''
N = data.getLanesNumber(file2)
return N