rise_timeseries
===============

.. py:module:: rise_timeseries


Classes
-------

.. autoapisummary::

   rise_timeseries.RISETimeseries


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

.. py:class:: RISETimeseries(n_masks: int = 1000, feature_res: int = 8, p_keep: float = 0.5, preprocess_function: Optional[callable] = None, keep_masks: bool = False, keep_masked_data: bool = False, keep_predictions: bool = False)

   RISE implementation for timeseries adapted from the image version of RISE.


   .. py:attribute:: n_masks
      :value: 1000



   .. py:attribute:: feature_res
      :value: 8



   .. py:attribute:: p_keep
      :value: 0.5



   .. py:attribute:: preprocess_function
      :value: None



   .. py:attribute:: masks
      :value: None



   .. py:attribute:: masked
      :value: None



   .. py:attribute:: predictions
      :value: None



   .. py:attribute:: keep_masks
      :value: False



   .. py:attribute:: keep_masked_data
      :value: False



   .. py:attribute:: keep_predictions
      :value: False



   .. py:method:: explain(model_or_function, input_timeseries, labels, batch_size=100, mask_type='mean')

      Runs the RISE explainer on images.

         The model will be called with masked timeseries,
         with a shape defined by `batch_size` and the shape of `input_data`.

      :param model_or_function: The function that runs the model to be explained _or_
                                the path to a ONNX model on disk.
      :type model_or_function: callable or str
      :param input_timeseries: Input timeseries data to be explained
      :type input_timeseries: np.ndarray
      :param batch_size: Batch size to use for running the model.
      :type batch_size: int
      :param labels: Labels to be explained
      :type labels: Iterable(int)
      :param mask_type: Masking strategy for masked values. Choose from 'mean' or a callable(input_timeseries)

      :returns: Explanation heatmap for each class (np.ndarray).