§ ÇPƒi· ãó—ddlmZmZmZmZddlmZddlmZddl m Z ddddddd œd e d e d eee e fd e dee deede dee efdefd„Z dS)é)ÚAnyÚCallableÚDictÚOptional)ÚChosenRejectedToMessages)ÚPreferenceDataset)ÚModelTokenizerz RLHFlow/HH-RLHF-Helpful-standardNFÚtrain)ÚsourceÚ column_mapÚtrain_on_inputÚnew_system_promptÚ filter_fnÚsplitÚ tokenizerr r r rrrÚload_dataset_kwargsÚreturnc óJ—t|||¬¦«}td|||||dœ|¤ŽS)að Constructs preference datasets similar to `Anthropic's helpful/harmless RLHF data `_. This is the processed helpful subset of the original dataset in a standardized format. Args: tokenizer (ModelTokenizer): Tokenizer used by the model that implements the ``tokenize_messages`` method. source (str): path to dataset repository on Hugging Face. For local datasets, define source as the data file type (e.g. "json", "csv", "text") and pass in the filepath in ``data_files``. See Hugging Face's ``load_dataset`` (https://huggingface.co/docs/datasets/en/package_reference/loading_methods#datasets.load_dataset.path) for more details. Default is ``RLHFlow/HH-RLHF-Helpful-standard``. column_map (Optional[Dict[str, str]]): a mapping from the expected columns "chosen" and "rejected" in the message transform :class:`~torchtune.data.ChosenRejectedToMessages` to the new column names in the dataset. Keys should be "chosen" and "rejected" and values should be the actual column names. If None, keep the default columns "chosen" and "rejected". train_on_input (bool): Whether the model is trained on the prompt or not. Default is False. new_system_prompt (Optional[str]): if specified, prepend a system message to every sample for both chosen and rejected. This can serve as instructions to guide the model response. Setting this will OVERRIDE any system messages already present in the dataset. Default is None. filter_fn (Optional[Callable]): callable used to filter the dataset prior to any pre-processing. See the Hugging Face `docs `_ for more details. split (str): ``split`` argument for ``datasets.load_dataset``. You can use this argument to load a subset of a given split, e.g. ``split="train[:10%]"``. Default is "train". **load_dataset_kwargs (Dict[str, Any]): additional keyword arguments to pass to ``load_dataset``. Returns: PreferenceDataset: The preference dataset built from source paired data. )r r r)r Úmessage_transformrrr©)rr) rr r r rrrrrs úw/home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/torchtune/datasets/_hh_rlhf_helpful.pyÚhh_rlhf_helpful_datasetrsZ€õV1Ø%ØØ+ðñôÐõ ð ØØ+ØØØð  ð ð ð  ð ðó)ÚtypingrrrrÚtorchtune.datarÚtorchtune.datasets._preferencerÚ'torchtune.modules.transforms.tokenizersr ÚstrÚboolrrrrúr sðð1Ð0Ð0Ð0Ð0Ð0Ð0Ð0Ð0Ð0Ð0Ð0à3Ð3Ð3Ð3Ð3Ð3Ø<Ð<Ð<Ð<Ð<Ð<ØBÐBÐBÐBÐBÐBð 5Ø+/Ø Ø'+Ø$(Øð8ð8ð8Øð8ð ð8ð˜˜c 3˜hœÔ(ð 8ð ð 8ð   ”}ð 8ð˜Ô!ð8ð ð8ð   S œ>ð8ðð8ð8ð8ð8ð8ð8r