def test_save_predictions(self):
tr_path = os.path.join(DATA_DIR, 'test_tr')
t_path = os.path.join(DATA_DIR, 'test_t')
rfc = RFC(n_estimators=100, criterion="entropy", n_jobs=-1)
sampling_settings = {
'bin_count': 16,
'neg_samples': 7,
'bin_samples': 20,
'seed': 0,
'nan_value': -1000000
}
loading_tool = LoadingTool(sampling_settings)
clas_tool = ClassificationTool(rfc)
tr_data = loading_tool.load_training_data(tr_path)
tr_data = loading_tool.quantize_data(tr_data)
clas_tool.train_classifier(tr_data)
tr_data = None
output_file = os.path.join(ROOT_DIR, 'outputs/rfc.test')
for t_data in loading_tool.load_testing_data(t_path):
t_data = loading_tool.quantize_data(t_data)
clas_tool.save_predictions(t_data, output_file)
t_data = None
assert os.path.isfile(output_file)
os.remove(output_file)
def test_train_classifier(self):
tr_path = os.path.join(DATA_DIR, 'test_tr')
rfc = RFC(n_estimators=100, criterion="entropy", n_jobs=-1)
sampling_settings = {
'bin_count': 16,
'neg_samples': 7,
'bin_samples': 20,
'seed': 0,
'nan_value': -1000000
}
loading_tool = LoadingTool(sampling_settings)
clas_tool = ClassificationTool(rfc)
tr_data = loading_tool.load_training_data(tr_path)
tr_data = loading_tool.quantize_data(tr_data)
clas_tool.train_classifier(tr_data)
tr_data = None
assert list(clas_tool.classifier.classes_) == [0, 1, 2, 3]