def predict(path, num_predictions, predicted):
if path not in predicted:
if os.path.isfile(os.path.dirname(os.path.dirname(path))+'/data_obj.pickle'):
with open(os.path.dirname(os.path.dirname(path))+'/data_obj.pickle', 'rb') as f:
d = pickle.load(f)
else:
edges_dir = os.path.dirname(os.path.dirname(os.path.dirname(path))) + '/edges'
edges = pd.read_csv(edges_dir, header = None)
event_log_dir = os.path.dirname(os.path.dirname(path))
event_log = pd.read_csv(event_log_dir+ '/event_log', header=None)
d = Data()
d.load_data_data_frame(event_log, edges)
with open(event_log_dir+'/data_obj.pickle', 'wb') as f:
pickle.dump(d, f)
# print(event_log_dir)
with open(os.path.dirname(os.path.dirname(path)) + '/contagion.pickle', 'rb') as file:
cc = pickle.load(file)
with open(os.path.dirname(os.path.dirname(path)) + '/adjacency.pickle', 'rb') as file:
a = pickle.load(file)
with open(path + '/threshold.pickle', 'rb') as f:
t = pickle.load(f)
m = MCDOI()
m.assign_contagions_correlation_matrix(cc)
m.assign_adjacency_matrix(a)
m.assign_thresholds_matrix(t)
m.fill_state_matrix(d)
result = m.predict(num_predictions)
new_path = path.split('/')
new_path[4] = 'negative-random-activation'
save_results(result, '/'+os.path.join(*new_path), num_predictions)
with open(directory+'predicted_7days', 'a+', encoding='utf-8') as handle:
handle.write(path + '\n')
def estimate_and_predict(d, dir, batch_type, batch_size, num_predictions):
try:
m = MCDOI()
m.fit(d, batch_type = batch_type, batch_size = batch_size)
result = m.predict(num_predictions)
save_results(result, dir + batch_type + '/size_' + str(batch_size) + '/', num_predictions)
save_parameters(m, dir + batch_type + '/size_' + str(batch_size) + '/')
except Exception as err:
write_to_logger(err.args)
print(err.args)
exit(1)
finally:
send_email()
def main():
directory = '/datasets/mcdoi/louvain/louvain_46_720/'
try:
d = Data()
d.load_data(directory)
m = MCDOI()
m.fit(d, batch_type='time', batch_size=86400)
m.predict(3) #predict(3) zwraca 63 aktywacje
except Exception as err:
writeToLogger(err.args)
print(err.args)
exit(1)
finally:
send_email()
def estimate_t_and_predict(path_dataset_history, batch_type, batch_size,
num_predictions):
print('Start')
edges = pd.read_csv(os.path.dirname(path_dataset_history) + '/edges',
header=None)
event_log = pd.read_csv(path_dataset_history + '/event_log', header=None)
print('Data frames loaded')
with open(path_dataset_history + '/contagion.pickle', 'rb') as file:
cc = pickle.load(file)
with open(path_dataset_history + '/adjacency.pickle', 'rb') as file:
a = pickle.load(file)
print('Pickles loaded')
d = Data()
d.load_data_data_frame(event_log, edges)
print('Data object initialised')
m = MCDOI()
m.assign_contagions_correlation_matrix(cc)
m.assign_adjacency_matrix(a)
print('Matrices assigned')
m.fit_only_thresholds_states(d,
batch_type=batch_type,
batch_size=batch_size)
# file_name = path_dataset_history + '/' + batch_type + '/size_' + str(batch_size) + '/threshold.pickle'
# os.makedirs(os.path.dirname(file_name), exist_ok=True)
# with open(file_name, 'wb') as threshold_file:
# pickle.dump(m.thresholds.matrix, threshold_file)
result = m.predict(num_predictions)
def estimate_t_and_predict(path_dataset_history, batch_type, batch_sizes,
num_predictions, estimated):
flag = False
for batch_size in batch_sizes:
if path_dataset_history + '/' + batch_type + '/size_' + str(
batch_size) not in estimated:
flag = True
if flag:
edges = pd.read_csv(os.path.dirname(path_dataset_history) + '/edges',
header=None)
event_log = pd.read_csv(path_dataset_history + '/event_log',
header=None)
with open(path_dataset_history + '/contagion.pickle', 'rb') as file:
cc = pickle.load(file)
with open(path_dataset_history + '/adjacency.pickle', 'rb') as file:
a = pickle.load(file)
for batch_size in batch_sizes:
if path_dataset_history + '/' + batch_type + '/size_' + str(
batch_size) not in estimated:
if os.path.isfile(path_dataset_history + '/data_obj.pickle'):
with open(path_dataset_history + '/data_obj.pickle',
'rb') as f:
d = pickle.load(f)
else:
d = Data()
d.load_data_data_frame(event_log, edges)
with open(path_dataset_history + '/data_obj.pickle',
'wb') as f:
pickle.dump(d, f)
m = MCDOI()
m.assign_contagions_correlation_matrix(cc)
m.assign_adjacency_matrix(a)
m.fit_only_thresholds_states(d,
batch_type=batch_type,
batch_size=batch_size)
file_name = path_dataset_history + '/' + batch_type + '/size_' + str(
batch_size) + '/threshold.pickle'
os.makedirs(os.path.dirname(file_name), exist_ok=True)
with open(file_name, 'wb') as threshold_file:
pickle.dump(m.thresholds.matrix, threshold_file)
result = m.predict(num_predictions)
save_results(
result, path_dataset_history + '/' + batch_type +
'/size_' + str(batch_size), num_predictions)
with open(directory + 'estimated_t+predict',
'a+',
encoding='utf-8') as handle:
handle.write(path_dataset_history + '/' + batch_type +
'/size_' + str(batch_size) + '\n')
with open(directory + 'predicted_7days',
'a+',
encoding='utf-8') as handle:
handle.write(path_dataset_history + '/' + batch_type +
'/size_' + str(batch_size) + '\n')
def predict(path_dataset_history, batch_type, batch_sizes, num_predictions):
if os.path.isfile(path_dataset_history + '/data_obj.pickle'):
with open(path_dataset_history + '/data_obj.pickle', 'rb') as f:
d = pickle.load(f)
else:
edges = pd.read_csv(os.path.dirname(path_dataset_history) + '/edges',
header=None)
event_log = pd.read_csv(path_dataset_history + '/event_log',
header=None)
d = Data()
d.load_data_data_frame(event_log, edges)
with open(path_dataset_history + '/data_obj.pickle', 'wb') as f:
pickle.dump(d, f)
with open(path_dataset_history + '/contagion.pickle', 'rb') as file:
cc = pickle.load(file)
with open(path_dataset_history + '/adjacency.pickle', 'rb') as file:
a = pickle.load(file)
new_path_dataset_history = path_dataset_history.split('/')
new_path_dataset_history[4] = 'paper/' + model
new_path_dataset_history = '/' + os.path.join(*new_path_dataset_history)
for batch_size in batch_sizes:
m = MCDOI()
m.assign_contagions_correlation_matrix(cc)
m.assign_adjacency_matrix(a)
with open(
new_path_dataset_history + '/' + batch_type + '/size_' +
str(batch_size) + '/threshold.pickle', 'rb') as f:
t = pickle.load(f)
m.assign_thresholds_matrix(t)
m.fill_state_matrix(d)
# m.fit_only_thresholds_states(d, batch_type = batch_type, batch_size = batch_size)
# file_name = new_path_dataset_history + '/' + batch_type + '/size_' + str(batch_size) + '/threshold.pickle'
# os.makedirs(os.path.dirname(file_name), exist_ok=True)
# with open(file_name, 'wb') as threshold_file:
# pickle.dump(m.thresholds.matrix, threshold_file)
# with open(directory+'estimated_thresholds', 'a+', encoding='utf-8') as handle:
# handle.write(path_dataset_history + '/' + batch_type + '/size_' + str(batch_size) + '\n')
result = m.predict(num_predictions)
# print(new_path_dataset_history + '/' + batch_type + '/size_' + str(batch_size))
save_results(
result, new_path_dataset_history + '/' + batch_type + '/size_' +
str(batch_size), num_predictions)