input_data
==========

.. py:module:: input_data


Attributes
----------

.. autoapisummary::

   input_data.Setup


Classes
-------

.. autoapisummary::

   input_data.Data


Functions
---------

.. autoapisummary::

   input_data._guess_mode


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

.. py:data:: Setup

.. py:function:: _guess_mode(input)

.. py:class:: Data(input, model_shape, device = 'cpu', mode=None, process=False)

   .. py:attribute:: input


   .. py:attribute:: mode
      :type:  str | None
      :value: None



   .. py:attribute:: target
      :type:  rex_xai.responsibility.prediction.Prediction | List[rex_xai.responsibility.prediction.Prediction] | None
      :value: None



   .. py:attribute:: device
      :value: 'cpu'



   .. py:attribute:: setup
      :type:  Optional[Setup]
      :value: None



   .. py:attribute:: transposed
      :value: False



   .. py:attribute:: model_shape


   .. py:attribute:: model_height
      :type:  Optional[int]


   .. py:attribute:: model_width
      :type:  Optional[int]


   .. py:attribute:: model_depth
      :type:  Optional[int]
      :value: None



   .. py:attribute:: model_channels
      :type:  Optional[int]
      :value: 1



   .. py:attribute:: model_order
      :value: None



   .. py:attribute:: mask_value
      :value: None



   .. py:attribute:: background
      :value: None



   .. py:attribute:: context
      :value: None



   .. py:attribute:: context_noise
      :value: 0.4



   .. py:method:: set_height(h)


   .. py:method:: set_width(w)


   .. py:method:: set_channels(c=None)


   .. py:method:: __repr__()


   .. py:method:: set_classification(cl)


   .. py:method:: match_data_to_model_shape()

      a PIL image has the from H * W * C, so
      if the model takes C * H * W we need to transpose self.data to
      get it into the correct form for the model to consume
      This function does *not* add in the batch channel at the beginning



   .. py:method:: generic_tab_preprocess()


   .. py:method:: load_data(astype='float32')


   .. py:method:: _normalise_rgb_data(means, stds, norm)

      used for onnx input data only



   .. py:method:: try_unsqueeze()


   .. py:method:: generic_image_preprocess(means=None, stds=None, astype='float32', norm = 255.0)


   .. py:method:: __get_shape()

      returns height, width, channels, order, depth for the model



   .. py:method:: set_mask_value(m)