- class DataRecord(data_source, data_label, storage_type, file_format='parquet', copy_from_source=False)[source]
Bases:
objectClass for managing data records in different storage types and formats.
- Attributes:
VALID_STORAGE_TYPES (tuple): Allowed storage types. VALID_FILE_FORMATS (tuple): Allowed file formats.
- Attributes:
backing_sourcedatadata_labelformatnum_rows
Get the backing file for the data record.
Get the data associated with the data record.
Get the label for the data record.
Get the file format of the data record.
Get the number of rows in the data record.
Methods
load()Load a cuDF DataFrame from the DataRecord.
- property backing_source: str
Get the backing file for the data record.
- Returns:
str: Backing file for the data record.
- property data: Union[cudf.DataFrame, pandas.DataFrame]
Get the data associated with the data record.
- Returns:
Any: Data associated with the data record.
- property data_label: str
Get the label for the data record.
- Returns:
str: Label for the data record.
- property format: str
Get the file format of the data record. Returns: str: File format of the data record.
- load()[source]
Load a cuDF DataFrame from the DataRecord.
- Returns:
cudf.DataFrame: Loaded cuDF DataFrame.
- property num_rows: int
Get the number of rows in the data record.
Returns: int: Number of rows in the data record.
