read_batch

lightautoml.reader.tabular_batch_generator.read_batch(data, features_names=None, n_jobs=1, batch_size=None, read_csv_params=None)

Read data for inference by batches for simple tabular data.

Note

Supported now data formats:

• Path to .csv, .parquet, .feather files.

• ndarray, or dict of ndarray. For example, {'data': X...}. In this case, roles are optional, but train_features and valid_features required.

• pandas.DataFrame.

Parameters:

• data (Union[str, ndarray, DataFrame, Dict[str, ndarray], Batch]) – Readable to DataFrame data.

• features_names (Optional[Sequence[str]]) – Optional features names if numpy.ndarray.

• n_jobs (int) – Number of processes to read file.

• batch_size (Optional[int]) – Batch size.

• read_csv_params (Optional[dict]) – Params to read csv file.

Return type:

Iterable[BatchGenerator]

Returns:

Generator of batches.