microsoft/qlib · error · NotImplementedError

This type of input is not supported

Error message

This type of input is not supported

What it means

MetaDataset.prepare (qlib/model/meta/dataset.py:66) accepts segments either as a single segment name (str) or a list/tuple of segment names, dispatching to _prepare_seg. Any other type (None, dict, int, a segments object) raises NotImplementedError('This type of input is not supported'). This enforces that meta-task preparation is keyed by named segments defined in the task.

Source

Thrown at qlib/model/meta/dataset.py:66

                train_meta_tasks, test_meta_tasks = meta_dataset.prepare_tasks(["train", "test"])

        Parameters
        ----------
        segments: Union[List[Text], Tuple[Text], Text]
            the info to select data

        Returns
        -------
        list:
            A list of the prepared data of each meta-task for training the meta-model. For multiple segments [seg1, seg2, ... , segN], the returned list will be [[tasks in seg1], [tasks in seg2], ... , [tasks in segN]].
            Each task is a meta task
        """
        if isinstance(segments, (list, tuple)):
            return [self._prepare_seg(seg) for seg in segments]
        elif isinstance(segments, str):
            return self._prepare_seg(segments)
        else:
            raise NotImplementedError(f"This type of input is not supported")

    @abc.abstractmethod
    def _prepare_seg(self, segment: Text):
        """
        prepare a single segment of data for training data

        Parameters
        ----------
        seg : Text
            the name of the segment
        """

View on GitHub (pinned to 79633dd950)

Solutions

1. Pass a segment name string or list of segment names, e.g. prepare('train') or prepare(['train', 'valid'])
2. Confirm the segment names exist in the task template's dataset segments definition
3. If you need custom segment handling, subclass MetaDataset and implement _prepare_seg for your type

Example fix

# before
meta_dataset.prepare({'train': ('2008-01-01', '2014-12-31')})  # NotImplementedError

# after
meta_dataset.prepare(['train', 'valid'])

if not isinstance(segments, (str, list, tuple)):
    raise TypeError(f'segments must be str or list/tuple of str, got {type(segments)}')
segments = [segments] if isinstance(segments, str) else list(segments)

def is_valid_segments(segments) -> bool:
    return isinstance(segments, str) or (isinstance(segments, (list, tuple)) and all(isinstance(s, str) for s in segments))

try:
    data = meta_dataset.prepare(segments)
except NotImplementedError as e:
    raise ValueError("pass segment names like 'train' or ['train','valid']") from e

When it happens

Trigger: Calling meta_dataset.prepare(segments) with None, a dict like {'train': (...), 'valid': (...)}, or a qlib segments tuple-of-tuples format instead of segment name strings; passing the full task object instead of segment names.

Common situations: Confusing the meta-dataset segments convention (segment name strings like 'train') with qlib's normal expression-based segments; migrating regular workflow code into the meta-learning workflow; passing None expecting all segments to be prepared.

Related errors

• nfs-common is not found, please install it by execute: sudo
• Mount failed: requires sudo or permission denied
• mount {provider_uri} on {mount_path} error! Command error
• Mount failed: {e.stderr}
• We can't find the project path

AI-assisted analysis of microsoft/qlib@79633dd950 (2026-08-15). Data as JSON: /api/errors/76cdef538365dfd3. Report an issue: GitHub.