data_juicer.format.json_formatter module#

class data_juicer.format.json_formatter.JsonFormatter(dataset_path, suffixes=None, **kwargs)[source]#

Bases: LocalFormatter

Load json-type files.

Default suffixes include .json, .jsonl, gzip/zstd variants.

Optional lenient JSONL: load_jsonl_lenient: true or env DATA_JUICER_JSONL_LENIENT=1 streams jsonl-only inputs with stdlib json.loads(), skipping bad lines (avoids HF ujson for those files).

SUFFIXES = ['.json', '.jsonl', '.json.gz', '.jsonl.gz', '.json.zst', '.jsonl.zst']#
__init__(dataset_path, suffixes=None, **kwargs)[source]#

Initialization method.

Parameters:

  • dataset_path – a dataset file or a dataset directory

  • suffixes – files with specified suffixes to be processed

  • kwargs – extra args

load_dataset(num_proc=None, global_cfg=None)[source]#

Load a dataset from dataset file or dataset directory, and unify its format.

Parameters:

  • num_proc – number of processes when loading the dataset

  • global_cfg – global cfg used in consequent processes,

Returns:

formatted dataset