Data preparation | Model Operations using td_lightgbm | teradataml OpenSourceML - Data preparation - Teradata Package for Python

Use the relevant single or multi model case statement to prepare your data, then validate the dataset and get the pandas DataFrame of data.

Single model case

>>> obj_s = td_lightgbm.Dataset(df_x_classif, df_y_classif, silent=True, free_raw_data=False)

Multi model case

>>> obj_m = td_lightgbm.Dataset(df_x_classif, df_y_classif, free_raw_data=False,
                                partition_columns=["partition_column_1", "partition_column_2"])

Validation dataset

>>> obj_m_v = td_lightgbm.Dataset(df_x_classif, df_y_classif, free_raw_data=False,
                                  partition_columns=["partition_column_1", "partition_column_2"])

Get pandas DataFrame of data for training locally

>>> pdf_x = df_x_classif.to_pandas().reset_index()
>>> pdf_y = df_y_classif.to_pandas()