pdstools.utils.datasets
=======================

.. py:module:: pdstools.utils.datasets


Attributes
----------

.. autoapisummary::

   pdstools.utils.datasets.ADMTreesModel


Functions
---------

.. autoapisummary::

   pdstools.utils.datasets.cdh_sample
   pdstools.utils.datasets.sample_trees
   pdstools.utils.datasets.sample_value_finder
   pdstools.utils.datasets.sample_explanations
   pdstools.utils.datasets.dq_sample


Module Contents
---------------

.. py:data:: ADMTreesModel
   :value: None


.. py:function:: cdh_sample(query: pdstools.utils.types.QUERY | None = None) -> pdstools.adm.ADMDatamart.ADMDatamart

   Import a sample dataset from the CDH Sample application

   :param query: An optional query to apply to the data, by default None
   :type query: QUERY | None, optional

   :returns: The ADM Datamart class populated with CDH Sample data
   :rtype: ADMDatamart


.. py:function:: sample_trees()

   Load the anonymized AGB sample model (100 trees, with sampleCount).

   :returns: An :class:`~pdstools.adm.trees.ADMTreesModel` loaded from the
             bundled ``data/agb/ModelExportWithSampleCount.json`` file (dev
             environment) or from the canonical GitHub raw URL (installed
             package).
   :rtype: ADMTreesModel


.. py:function:: sample_value_finder(threshold: float | None = None) -> pdstools.valuefinder.ValueFinder.ValueFinder

   Import a sample dataset of a Value Finder simulation

   This simulation was ran on a stock CDH Sample system.

   :param threshold: Optional override of the propensity threshold in the system, by default None
   :type threshold: float | None, optional

   :returns: The Value Finder class populated with the Value Finder simulation data
   :rtype: ValueFinder


.. py:function:: sample_explanations(*, model_name: str = 'AdaptiveBoostCT', from_date: datetime.datetime | None = None, to_date: datetime.datetime | None = None) -> pdstools.explanations.Explanations

   Import a sample set of pre-aggregated AGB global explanations.

   These were generated from a stock CDH Sample system. The aggregates are
   fetched from the pdstools repository over HTTPS, so this works from an
   installed package as well as from a repository checkout.

   :param model_name: Name of the model rule. Used for report metadata only.
   :type model_name: str, keyword-only, default "AdaptiveBoostCT"
   :param from_date: Start of the reporting window. Defaults to a week before ``to_date``.
   :type from_date: datetime | None, keyword-only, optional
   :param to_date: End of the reporting window. Defaults to today.
   :type to_date: datetime | None, keyword-only, optional

   :returns: The Explanations class populated with the sample aggregates.
   :rtype: Explanations


.. py:function:: dq_sample(*, similarity_threshold: float = 0.8) -> pdstools.data_quality.TopicDataQuality

   Load the built-in smalltalk sample dataset for Topic Data Quality.

   Returns a ready-to-use ``TopicDataQuality`` instance with embeddings,
   UMAP, and similarity already computed.

   :param similarity_threshold: Topic pairs above this TF-IDF cosine similarity are flagged.
   :type similarity_threshold: float, default 0.8

   :returns: A fully-initialized instance with precomputed results.
   :rtype: TopicDataQuality