start = time.time()
with open(results_file, 'w') as fout:
csv_writer = csv.writer(fout,
delimiter=';',
quotechar='"',
quoting=csv.QUOTE_MINIMAL)
csv_writer.writerow(
["dataset", "cls", "params", "nr_events", "metric", "score"])
total = 0
total_acc = 0
total_mae = 0
for nr_events in range(2, max_len):
# encode only prefixes of this length
X, y_a, y_t = dataset_manager.generate_3d_data_for_prefix_length_no_padding(
dt_train, nr_events, mode="train")
X_test, y_a_test, y_t_test = dataset_manager.generate_3d_data_for_prefix_length_no_padding(
dt_test, nr_events, mode="test")
if X.shape[0] == 0 or X_test.shape[0] == 0:
break
y_t_test = y_t_test * dataset_manager.divisors["timesincelastevent"]
print(nr_events, X.shape, X_test.shape, y_a_test.shape, y_t_test.shape)
# train models
cls_a = RandomForestClassifier(n_estimators=n_estimators,
max_features=max_features)
cls_a.fit(X, y_a)
cls_t = RandomForestRegressor(n_estimators=n_estimators,
