df_test_collated = pd.read_csv(path + 'df_test_collated.csv')
# specify output folder to save plots in
output_folder = '../old_output/shallow_tctc/%s/train_collated_test_collated/' % KEY
lm = LearningModel(df_train_collated,
target_variable='demand',
split_ratio=0.2,
output_folder=output_folder,
scale=True,
scale_output=False,
output_zscore=False,
output_minmax=False,
output_box=False,
output_log=False,
input_zscore=None,
input_minmax=scaling[KEY],
input_box=None,
input_log=None,
cols_drop=None,
grid=True,
random_grid=False,
nb_folds_grid=10,
nb_repeats_grid=10,
testing_data=df_test_collated,
save_errors_xlsx=True,
save_validation=False)
for model in models_to_test:
model_name = models_to_test[model]
print('\n********** Results for %s **********' % model_name)
% (service, mohafaza))
mylist = list(df.columns.values)
mylist.remove('demand')
lm = LearningModel(df,
target_variable='demand',
split_ratio=0.2,
output_folder=output_folder + '%s_%s/' %
(service, mohafaza),
scale=True,
scale_output=False,
output_zscore=False,
output_minmax=False,
output_box=False,
output_log=False,
input_zscore=None,
input_minmax=(0, 11) if 'w_{t-5}' in list(
df.columns.values) else (0, 10),
input_box=None,
input_log=None,
cols_drop=None,
grid=True,
random_grid=False,
nb_folds_grid=10,
nb_repeats_grid=10,
save_errors_xlsx=True,
save_validation=False)
for model in models_to_test:
model_name = models_to_test[model]
print('\n********** Results for %s **********' %