def infert_ds(label_index, label_name='Label'):
file_schema = 'sep=, col=id:TX:0 col=education:TX:1 col={}:R4:{} ' \
'col=Features:R4:{}-8 header=+'.format(
label_name, label_index, label_index + 1)
data = FileDataStream(infert_file, schema=file_schema)
if label_name != 'Label':
data._set_role(Role.Label, label_name)
return data
def data_wt_rename(self, label_name, group_id, features):
simpleinput_file = get_dataset("gen_tickettrain").as_filepath()
file_schema = 'sep=, col={label}:R4:0 col={group_id}:TX:1 ' \
'col={features}:R4:3-5'.format(
label=label_name, group_id=group_id, features=features)
data = FileDataStream(simpleinput_file, schema=file_schema)
if label_name != 'Label':
data._set_role(Role.Label, label_name)
return data
def test_linear_file_role(self):
pipeline = Pipeline([OneHotVectorizer() << categorical_columns,
FastLinearBinaryClassifier(train_threads=1,
shuffle=False)])
train_stream = FileDataStream(train_file, schema=file_schema)
train_stream._set_role('Label', label_column)
pipeline.fit(train_stream)
test_stream = FileDataStream(test_file, schema=file_schema)
out_data = pipeline.predict(test_stream)
check_accuracy(test_file, label_column, out_data, 0.65)