def test_fit_transform(self): X = _data() ct, ct_categorical_feature = make_lightgbm_column_transformer(X.dtypes, missing_value_aware = False) dfm, dfm_categorical_feature = make_lightgbm_dataframe_mapper(X.dtypes, missing_value_aware = False) self.assertEqual(ct.fit_transform(X).tolist(), dfm.fit_transform(X).tolist()) self.assertEqual([0, 1, 3], ct_categorical_feature) self.assertEqual([0, 1, 3], dfm_categorical_feature) ct, ct_categorical_feature = make_lightgbm_column_transformer(X.dtypes, missing_value_aware = True) dfm, dfm_categorical_feature = make_lightgbm_dataframe_mapper(X.dtypes, missing_value_aware = True) self.assertEqual(ct.fit_transform(X).tolist(), dfm.fit_transform(X).tolist()) self.assertEqual([0, 1, 3], ct_categorical_feature) self.assertEqual([0, 1, 3], dfm_categorical_feature)
cat_columns = ["Education", "Employment", "Marital", "Occupation"]
cont_columns = ["Age", "Hours", "Income"]
df_X = df[cat_columns + cont_columns]
df_y = df["Adjusted"]
dtypes = df_X.dtypes
mapper = ColumnTransformer([(cat_column, CategoricalDomain(), [cat_column])
for cat_column in cat_columns] +
[(cont_column, ContinuousDomain(), [cont_column])
for cont_column in cont_columns])
dtypes = Series(dtypes.values, index=[0, 1, 2, 3, 4, 5, 6])
lightgbm_mapper, lightgbm_categorical_feature = make_lightgbm_column_transformer(
dtypes, missing_value_aware=True)
lightgbm_pipeline = Pipeline([
("mapper", lightgbm_mapper),
("classifier",
LGBMClassifier(n_estimators=31,
max_depth=3,
random_state=13,
categorical_feature=lightgbm_categorical_feature))
])
xgboost_mapper = make_xgboost_column_transformer(dtypes,
missing_value_aware=True)
xgboost_pipeline = Pipeline([("mapper", xgboost_mapper),
("classifier",
XGBClassifier(n_estimators=31,
learning_rate=0.1,