Skip to content

dataset

CustomDataset

Bases: Dataset

A custom dataset class that extends from datasets.Dataset and is used to generate an Argilla FeedbackDataset instance from the pre-defined configuration within the task provided to Pipeline.generate.

Source code in src/distilabel/dataset.py 
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
class CustomDataset(Dataset):
    """A custom dataset class that extends from `datasets.Dataset` and is used to generate
    an Argilla `FeedbackDataset` instance from the pre-defined configuration within the task
    provided to `Pipeline.generate`.
    """

    task: Union["Task", None] = None

    def to_argilla(
        self,
        dataset_columns: List[str] = None,
        vector_strategy: Union[bool, "SentenceTransformersExtractor"] = True,
        metric_strategy: Union[bool, "TextDescriptivesExtractor"] = True,
    ) -> "FeedbackDataset":
        """Converts the dataset to an Argilla `FeedbackDataset` instance, based on the
        task defined in the dataset as part of `Pipeline.generate`.

        Args:
            dataset_columns (List[str]): the dataset columns or fields to be used for the Argilla `FeedbackDataset` instance.
                By default, the first 5 columns or fields will be used.
            vector_strategy (Union[bool, SentenceTransformersExtractor]): the strategy to be used for
                adding vectors to the dataset. If `True`, the default `SentenceTransformersExtractor`
                will be used with the `TaylorAI/bge-micro-2` model. If `False`, no vectors will be added to the dataset.
            metrics_strategy (Union[bool, TextDescriptivesExtractor]): the strategy to be used for
                adding metrics to the dataset. If `True`, the default `TextDescriptivesExtractor`
                will be used. If `False`, no metrics will be added to the dataset.

        Raises:
            ImportError: if the argilla library is not installed.
            ValueError: if the task is not set.

        Returns:
            FeedbackDataset: the Argilla `FeedbackDataset` instance.
        """
        if not _ARGILLA_AVAILABLE:
            raise ImportError(
                "To use `to_argilla` method is required to have `argilla` installed. "
                "Please install it with `pip install argilla`."
            )

        if self.task is None:
            raise ValueError(
                "The task is not set. Please set it with `dataset.task = <task>`."
            )

        try:
            rg_dataset = self.task.to_argilla_dataset(dataset_row=self[0])  # type: ignore
        except Exception as e:
            raise ValueError(
                f"Error while converting the dataset to an Argilla `FeedbackDataset` instance: {e}"
            ) from e

        for dataset_row in self:
            if any(
                dataset_row[input_arg_name] is None  # type: ignore
                for input_arg_name in self.task.input_args_names
            ):
                continue
            try:
                rg_dataset.add_records(
                    self.task._to_argilla_record(dataset_row=dataset_row)  # type: ignore
                )  # type: ignore
            except Exception as e:
                warnings.warn(
                    f"Error while converting a row into an Argilla `FeedbackRecord` instance: {e}",
                    UserWarning,
                    stacklevel=2,
                )

        selected_fields = infer_field_from_dataset_columns(
            dataset_columns=dataset_columns, dataset=rg_dataset, task=self.task
        )
        rg_dataset = self.add_vectors_to_argilla_dataset(
            dataset=rg_dataset, vector_strategy=vector_strategy, fields=selected_fields
        )
        rg_dataset = self.add_metrics_to_argilla_dataset(
            dataset=rg_dataset, metric_strategy=metric_strategy, fields=selected_fields
        )
        return rg_dataset

    def add_vectors_to_argilla_dataset(
        self,
        dataset: Union["FeedbackRecord", List["FeedbackRecord"], "FeedbackDataset"],
        vector_strategy: Union[bool, "SentenceTransformersExtractor"],
        fields: List[str] = None,
    ) -> Union["FeedbackRecord", List["FeedbackRecord"], "FeedbackDataset"]:
        if _ARGILLA_AVAILABLE and vector_strategy:
            if isinstance(vector_strategy, SentenceTransformersExtractor):
                ste: SentenceTransformersExtractor = vector_strategy
            elif vector_strategy:
                ste = SentenceTransformersExtractor()
            dataset = ste.update_dataset(dataset=dataset, fields=fields)

        elif not _ARGILLA_AVAILABLE and vector_strategy:
            warnings.warn(
                "An error occurred while adding vectors to the dataset: "
                "The `argilla`/`sentence-transformers` packages are not installed or the installed version is not compatible with the"
                " required version. If you want to add vectors to your dataset, please run `pip install 'distilabel[argilla]'`.",
                stacklevel=2,
            )
        return dataset

    def add_metrics_to_argilla_dataset(
        self,
        dataset: Union["FeedbackRecord", List["FeedbackRecord"], "FeedbackDataset"],
        metric_strategy: Union[bool, "TextDescriptivesExtractor"],
        fields: List[str] = None,
    ) -> Union["FeedbackRecord", List["FeedbackRecord"], "FeedbackDataset"]:
        if _ARGILLA_AVAILABLE and metric_strategy:
            if isinstance(metric_strategy, TextDescriptivesExtractor):
                tde: TextDescriptivesExtractor = metric_strategy
            elif metric_strategy:
                tde = TextDescriptivesExtractor()

            dataset = tde.update_dataset(dataset=dataset, fields=fields)
        elif not _ARGILLA_AVAILABLE and metric_strategy:
            warnings.warn(
                "An error occurred while adding metrics to the dataset: "
                "The `argilla`/`text-descriptives` packages are not installed or the installed version is not compatible with the"
                " required version. If you want to add metrics to your dataset, please run `pip install 'distilabel[argilla]'`.",
                stacklevel=2,
            )
        return dataset

    def save_to_disk(self, dataset_path: PathLike, **kwargs: Any) -> None:
        """Saves the datataset to disk, also saving the task.

        Args:
            dataset_path: Path to the dataset.
            **kwargs: Additional arguments to be passed to `datasets.Dataset.save_to_disk`.
        """
        super().save_to_disk(dataset_path, **kwargs)
        if self.task is not None:
            self.task.save(Path(dataset_path))

    @classmethod
    def load_from_disk(cls, dataset_path: PathLike, **kwargs: Any):
        """Load a CustomDataset from disk, also reading the task.

        Args:
            dataset_path: Path to the dataset, as you would do with a standard Dataset.

        Returns:
            The loaded dataset.
        """
        ds = super().load_from_disk(dataset_path, *kwargs)
        # Dynamically remaps the `datasets.Dataset` to be a `CustomDataset` instance
        ds.__class__ = cls
        task = load_task_from_disk(dataset_path)
        ds.task = task
        return ds

    def push_to_hub(
        self,
        repo_id: str,
        *args: Optional[Any],
        push_task: bool = True,
        **kwargs: Optional[Any],
    ) -> None:
        """Same method as `datasets.Dataset.push_to_hub`, but also pushes the task to simplify
        creating a CustomDataset from HuggingFace hub.

        Args:
            repo_id (str):
                The ID of the repository to push to in the following format: `<user>/<dataset_name>` or
                `<org>/<dataset_name>`. Also accepts `<dataset_name>`, which will default to the namespace
                of the logged-in user.
            args (Any): Additional arguments to be passed to `datasets.Dataset.push_to_hub`.
            push_task (bool, optional): _description_. Defaults to True.
            kwargs (Any): Additional arguments to be passed to `datasets.Dataset.push_to_hub`.

        Examples:
            >>> from distilabel.dataset import CustomDataset
            >>> dataset = CustomDataset(...)
            >>> dataset.push_to_hub("path/to/dataset")
        """
        super().push_to_hub(repo_id, *args, **kwargs)
        if self.task is not None and push_task:
            try:
                logger.info("Pushing task to the hub...")
                with tempfile.NamedTemporaryFile(mode="w", delete=False) as f:
                    f.write(json.dumps(self.task.dump(), indent=2))
                    f.flush()

                    HfApi().upload_file(
                        path_or_fileobj=f.name,
                        path_in_repo=TASK_FILE_NAME,
                        repo_id=repo_id,
                        repo_type="dataset",
                        token=kwargs.get("token"),
                    )
            except Exception as e:
                logger.info(f"Error while pushing the task to the hub: {e}.")

load_from_disk(dataset_path, **kwargs) classmethod

Load a CustomDataset from disk, also reading the task.

Parameters:

Name Type Description Default
dataset_path PathLike

Path to the dataset, as you would do with a standard Dataset.

 required

Returns:

Type Description

The loaded dataset.
Source code in src/distilabel/dataset.py 
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
@classmethod
def load_from_disk(cls, dataset_path: PathLike, **kwargs: Any):
    """Load a CustomDataset from disk, also reading the task.

    Args:
        dataset_path: Path to the dataset, as you would do with a standard Dataset.

    Returns:
        The loaded dataset.
    """
    ds = super().load_from_disk(dataset_path, *kwargs)
    # Dynamically remaps the `datasets.Dataset` to be a `CustomDataset` instance
    ds.__class__ = cls
    task = load_task_from_disk(dataset_path)
    ds.task = task
    return ds

push_to_hub(repo_id, *args, push_task=True, **kwargs)

Same method as datasets.Dataset.push_to_hub, but also pushes the task to simplify creating a CustomDataset from HuggingFace hub.

Parameters:

Name Type Description Default
repo_id str

The ID of the repository to push to in the following format: <user>/<dataset_name> or <org>/<dataset_name>. Also accepts <dataset_name>, which will default to the namespace of the logged-in user.

 required
args Any

Additional arguments to be passed to datasets.Dataset.push_to_hub.

 ()
push_task bool

description. Defaults to True.

 True
kwargs Any

Additional arguments to be passed to datasets.Dataset.push_to_hub.

 {}

Examples:

>>> from distilabel.dataset import CustomDataset
>>> dataset = CustomDataset(...)
>>> dataset.push_to_hub("path/to/dataset")
Source code in src/distilabel/dataset.py 
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
def push_to_hub(
    self,
    repo_id: str,
    *args: Optional[Any],
    push_task: bool = True,
    **kwargs: Optional[Any],
) -> None:
    """Same method as `datasets.Dataset.push_to_hub`, but also pushes the task to simplify
    creating a CustomDataset from HuggingFace hub.

    Args:
        repo_id (str):
            The ID of the repository to push to in the following format: `<user>/<dataset_name>` or
            `<org>/<dataset_name>`. Also accepts `<dataset_name>`, which will default to the namespace
            of the logged-in user.
        args (Any): Additional arguments to be passed to `datasets.Dataset.push_to_hub`.
        push_task (bool, optional): _description_. Defaults to True.
        kwargs (Any): Additional arguments to be passed to `datasets.Dataset.push_to_hub`.

    Examples:
        >>> from distilabel.dataset import CustomDataset
        >>> dataset = CustomDataset(...)
        >>> dataset.push_to_hub("path/to/dataset")
    """
    super().push_to_hub(repo_id, *args, **kwargs)
    if self.task is not None and push_task:
        try:
            logger.info("Pushing task to the hub...")
            with tempfile.NamedTemporaryFile(mode="w", delete=False) as f:
                f.write(json.dumps(self.task.dump(), indent=2))
                f.flush()

                HfApi().upload_file(
                    path_or_fileobj=f.name,
                    path_in_repo=TASK_FILE_NAME,
                    repo_id=repo_id,
                    repo_type="dataset",
                    token=kwargs.get("token"),
                )
        except Exception as e:
            logger.info(f"Error while pushing the task to the hub: {e}.")

save_to_disk(dataset_path, **kwargs)

Saves the datataset to disk, also saving the task.

Parameters:

Name Type Description Default
dataset_path PathLike

Path to the dataset.

 required
**kwargs Any

Additional arguments to be passed to datasets.Dataset.save_to_disk.

 {}
Source code in src/distilabel/dataset.py 
183
184
185
186
187
188
189
190
191
192
def save_to_disk(self, dataset_path: PathLike, **kwargs: Any) -> None:
    """Saves the datataset to disk, also saving the task.

    Args:
        dataset_path: Path to the dataset.
        **kwargs: Additional arguments to be passed to `datasets.Dataset.save_to_disk`.
    """
    super().save_to_disk(dataset_path, **kwargs)
    if self.task is not None:
        self.task.save(Path(dataset_path))

to_argilla(dataset_columns=None, vector_strategy=True, metric_strategy=True)

Converts the dataset to an Argilla FeedbackDataset instance, based on the task defined in the dataset as part of Pipeline.generate.

Parameters:

Name Type Description Default
dataset_columns List[str]

the dataset columns or fields to be used for the Argilla FeedbackDataset instance. By default, the first 5 columns or fields will be used.

 None
vector_strategy Union[bool, SentenceTransformersExtractor]

the strategy to be used for adding vectors to the dataset. If True, the default SentenceTransformersExtractor will be used with the TaylorAI/bge-micro-2 model. If False, no vectors will be added to the dataset.

 True
metrics_strategy Union[bool, TextDescriptivesExtractor]

the strategy to be used for adding metrics to the dataset. If True, the default TextDescriptivesExtractor will be used. If False, no metrics will be added to the dataset.

 required

Raises:

Type Description
ImportError

if the argilla library is not installed.
ValueError

if the task is not set.

Returns:

Name Type Description
FeedbackDataset FeedbackDataset

the Argilla FeedbackDataset instance.
Source code in src/distilabel/dataset.py 
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
def to_argilla(
    self,
    dataset_columns: List[str] = None,
    vector_strategy: Union[bool, "SentenceTransformersExtractor"] = True,
    metric_strategy: Union[bool, "TextDescriptivesExtractor"] = True,
) -> "FeedbackDataset":
    """Converts the dataset to an Argilla `FeedbackDataset` instance, based on the
    task defined in the dataset as part of `Pipeline.generate`.

    Args:
        dataset_columns (List[str]): the dataset columns or fields to be used for the Argilla `FeedbackDataset` instance.
            By default, the first 5 columns or fields will be used.
        vector_strategy (Union[bool, SentenceTransformersExtractor]): the strategy to be used for
            adding vectors to the dataset. If `True`, the default `SentenceTransformersExtractor`
            will be used with the `TaylorAI/bge-micro-2` model. If `False`, no vectors will be added to the dataset.
        metrics_strategy (Union[bool, TextDescriptivesExtractor]): the strategy to be used for
            adding metrics to the dataset. If `True`, the default `TextDescriptivesExtractor`
            will be used. If `False`, no metrics will be added to the dataset.

    Raises:
        ImportError: if the argilla library is not installed.
        ValueError: if the task is not set.

    Returns:
        FeedbackDataset: the Argilla `FeedbackDataset` instance.
    """
    if not _ARGILLA_AVAILABLE:
        raise ImportError(
            "To use `to_argilla` method is required to have `argilla` installed. "
            "Please install it with `pip install argilla`."
        )

    if self.task is None:
        raise ValueError(
            "The task is not set. Please set it with `dataset.task = <task>`."
        )

    try:
        rg_dataset = self.task.to_argilla_dataset(dataset_row=self[0])  # type: ignore
    except Exception as e:
        raise ValueError(
            f"Error while converting the dataset to an Argilla `FeedbackDataset` instance: {e}"
        ) from e

    for dataset_row in self:
        if any(
            dataset_row[input_arg_name] is None  # type: ignore
            for input_arg_name in self.task.input_args_names
        ):
            continue
        try:
            rg_dataset.add_records(
                self.task._to_argilla_record(dataset_row=dataset_row)  # type: ignore
            )  # type: ignore
        except Exception as e:
            warnings.warn(
                f"Error while converting a row into an Argilla `FeedbackRecord` instance: {e}",
                UserWarning,
                stacklevel=2,
            )

    selected_fields = infer_field_from_dataset_columns(
        dataset_columns=dataset_columns, dataset=rg_dataset, task=self.task
    )
    rg_dataset = self.add_vectors_to_argilla_dataset(
        dataset=rg_dataset, vector_strategy=vector_strategy, fields=selected_fields
    )
    rg_dataset = self.add_metrics_to_argilla_dataset(
        dataset=rg_dataset, metric_strategy=metric_strategy, fields=selected_fields
    )
    return rg_dataset

DatasetCheckpoint dataclass

A checkpoint class that contains the information of a checkpoint.

Parameters:

Name Type Description Default
path Path

The path to the checkpoint.

 cwd() / 'ckpt'
save_frequency int

The frequency at which the checkpoint should be saved By default is set to -1 (no checkpoint is saved to disk, but the dataset is returned upon failure).

 -1
extra_kwargs dict[str, Any]

Additional kwargs to be passed to the save_to_disk method of the Dataset.

 field(default_factory=dict)

Examples:

>>> from distilabel.dataset import DatasetCheckpoint
>>> # Save the dataset every 10% of the records generated.
>>> checkpoint = DatasetCheckpoint(save_frequency=len(dataset) // 10)
>>> # Afterwards, we can access the checkpoint's checkpoint.path.
Source code in src/distilabel/dataset.py 
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
@dataclass
class DatasetCheckpoint:
    """A checkpoint class that contains the information of a checkpoint.

    Args:
        path (Path): The path to the checkpoint.
        save_frequency (int): The frequency at which the checkpoint should be saved
            By default is set to -1 (no checkpoint is saved to disk, but the dataset
            is returned upon failure).
        extra_kwargs (dict[str, Any]): Additional kwargs to be passed to the `save_to_disk` method of the Dataset.

    Examples:
        >>> from distilabel.dataset import DatasetCheckpoint
        >>> # Save the dataset every 10% of the records generated.
        >>> checkpoint = DatasetCheckpoint(save_frequency=len(dataset) // 10)
        >>> # Afterwards, we can access the checkpoint's checkpoint.path.
    """

    path: Path = Path.cwd() / "ckpt"
    save_frequency: int = -1
    extra_kwargs: Dict[str, Any] = field(default_factory=dict)

    # Internal fields to keep track of the number of records generated and when to check.
    _total_checks: int = field(repr=False, default=0)

    def do_checkpoint(self, step: int) -> bool:
        """Determines if a checkpoint should be done.

        Args:
            step (int): The number of records generated.

        Returns:
            bool: Whether a checkpoint should be done.
        """
        if self.save_frequency == -1:
            return False

        if (step - self._total_checks * self.save_frequency) // self.save_frequency:
            self._total_checks += 1
            return True
        return False

do_checkpoint(step)

Determines if a checkpoint should be done.

Parameters:

Name Type Description Default
step int

The number of records generated.

 required

Returns:

Name Type Description
bool bool

Whether a checkpoint should be done.
Source code in src/distilabel/dataset.py 
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
def do_checkpoint(self, step: int) -> bool:
    """Determines if a checkpoint should be done.

    Args:
        step (int): The number of records generated.

    Returns:
        bool: Whether a checkpoint should be done.
    """
    if self.save_frequency == -1:
        return False

    if (step - self._total_checks * self.save_frequency) // self.save_frequency:
        self._total_checks += 1
        return True
    return False

load_dataset(path, *args, **kwargs)

Load a dataset from HuggingFace hub.

Overloads the datasets.load_dataset method to return a CustomDataset instance, downloading the Task from the hub (if any).

Parameters:

Name Type Description Default
path str

Path to the dataset in the hub.

 required
args, kwargs

and any other arguments used by datasets.load_dataset

 required

Returns:

Name Type Description
dataset Union[Dataset, CustomDataset]

CustomDataset instance, with the Task stored if available.

Examples:

>>> from distilabel.dataset import load_dataset
>>> dataset: CustomDataset = load_dataset("argilla/distilabel-sample-evol-instruct", split="train")
Source code in src/distilabel/dataset.py 
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
def load_dataset(path: str, *args: Any, **kwargs: Any) -> Union[Dataset, CustomDataset]:
    """Load a dataset from HuggingFace hub.

    Overloads the `datasets.load_dataset` method to return a `CustomDataset` instance,
    downloading the `Task` from the hub (if any).

    Args:
        path (str): Path to the dataset in the hub.
        args, kwargs: and any other arguments used by `datasets.load_dataset`

    Returns:
        dataset: CustomDataset instance, with the `Task` stored if available.

    Examples:
        >>> from distilabel.dataset import load_dataset
        >>> dataset: CustomDataset = load_dataset("argilla/distilabel-sample-evol-instruct", split="train")
    """
    from datasets import load_dataset as _load_dataset

    ds = _load_dataset(path, *args, **kwargs)
    cds = CustomDataset(ds.data.table)
    # download the task
    try:
        task_path = hf_hub_download(
            repo_id=path, filename=TASK_FILE_NAME, repo_type="dataset"
        )
        task = load_from_dict(read_json(task_path))
        cds.task = task
    except Exception as e:
        logger.info(f"Error while downloading the task from the hub: {e}.")
    return cds