emmi.datasets.ahmedml
=====================

.. py:module:: emmi.datasets.ahmedml


Classes
-------

.. autoapisummary::

   emmi.datasets.ahmedml.AhmedMLStats
   emmi.datasets.ahmedml.AhmedMLDefaultSplitIDs
   emmi.datasets.ahmedml.AhmedMLDataset


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

.. py:class:: AhmedMLStats

   .. py:attribute:: raw_pos_min
      :type:  tuple[float]


   .. py:attribute:: raw_pos_max
      :type:  tuple[float]
      :value: (6.0,)



   .. py:attribute:: surface_pressure_mean
      :type:  tuple[float]


   .. py:attribute:: surface_pressure_std
      :type:  tuple[float]
      :value: (0.188242,)



   .. py:attribute:: surface_wallshearstress_mean
      :type:  tuple[float, float, float]


   .. py:attribute:: surface_wallshearstress_std
      :type:  tuple[float, float, float]
      :value: (0.00117512, 0.000652266, 0.000713125)



   .. py:attribute:: volume_totalpcoeff_mean
      :type:  tuple[float]
      :value: (0.812013,)



   .. py:attribute:: volume_totalpcoeff_std
      :type:  tuple[float]
      :value: (0.367992,)



   .. py:attribute:: volume_velocity_mean
      :type:  tuple[float, float, float]
      :value: (0.8746, 1.42877e-05, 0.00776145)



   .. py:attribute:: volume_velocity_std
      :type:  tuple[float, float, float]
      :value: (0.300305, 0.114927, 0.124698)



   .. py:attribute:: volume_vorticity_mean
      :type:  tuple[float, float, float]


   .. py:attribute:: volume_vorticity_std
      :type:  tuple[float, float, float]
      :value: (19.2822, 142.25, 129.528)



   .. py:attribute:: volume_vorticity_logscale_mean
      :type:  tuple[float, float, float]


   .. py:attribute:: volume_vorticity_logscale_std
      :type:  tuple[float, float, float]
      :value: (1.11017, 1.9653, 1.77495)



   .. py:attribute:: volume_vorticity_magnitude_mean
      :type:  float
      :value: 55.2088



   .. py:attribute:: volume_vorticity_magnitude_std
      :type:  float
      :value: 185.313



   .. py:attribute:: raw_pos_z_min
      :type:  float
      :value: 0.0



.. py:class:: AhmedMLDefaultSplitIDs

   .. py:attribute:: train


   .. py:attribute:: val


   .. py:attribute:: test


   .. py:method:: create_split(seed=42)
      :staticmethod:


      A helper function to create a random split of the dataset. The default indices were created with seed=42.



.. py:class:: AhmedMLDataset(split, root = None, ais_bucket_name = None, ais_endpoint = 'http://compute-permanent-node-179:8080', **kwargs)

   Bases: :py:obj:`ksuit.data.Dataset`


   Dataset implementation for AhmedML that supports both local and AISTORE storage.

   Currently, supports the following properties:
   - position of car surface points and corresponding values for pressure and wall shear stress.
   - position of the volumetric cells with corresponding values for pressure, velocity and vorticity.

   :param split: Which split to use. "debug" will use all available "run_X" folders. "train", "valid", "test" will load
                 the splits as specified in the "train_val_test_splits/{split}_design_ids.txt" file.
   :param root: path to the processed dataset, e.g. /nfs-gpu/simulation/data/ahmedml_processed/subsampled_10x.
   :param ais_bucket_name: The name of the ais bucket on oracle object storage, e.g. 'emmi-ahmedml-subsampled-10x'.
   :param ais_endpoint: The endpoint (host) running the AISTORE instance.

   Args:
   dataset_config_provider: Optional provider for dataset configuration.
   dataset_normalizers: A dictionary that contains normalization ComposePreProcess(ers) for each data type. The key for each normalizer can be used in the with_normalizers decorator.


   .. py:attribute:: is_using_ais


   .. py:attribute:: split


   .. py:property:: default_split_ids
      :type: dict[str, list[int]]


      Property to get the sorted split IDs for train, validation, and test sets, based on DEFAULT_SPLIT_*_IDS.


   .. py:method:: get_normalization_stats()
      :staticmethod:


      Returns the normalization stats for the dataset.
      Exists for backwards compatibility with downstream repo, where collators use this.



   .. py:method:: getitem_surface_position_vtp(idx)

      Retrieves surface positions (num_surface_points, 3)



   .. py:method:: getitem_surface_position_stl(idx)

      Retrieves surface positions (num_surface_points, 3)



   .. py:method:: getitem_surface_position_stl_resampled100k(idx)

      Retrieves surface positions (num_surface_points, 3)



   .. py:method:: getitem_surface_pressure(idx)

      Retrieves surface pressures (num_surface_points, 1)



   .. py:method:: getitem_surface_wallshearstress(idx)

      Retrieves surface wallshearstress (num_surface_points, 3)



   .. py:method:: getitem_volume_position(idx)

      Retrieves volume position (num_volume_points, 3)



   .. py:method:: getitem_volume_totalpcoeff(idx)

      Retrieves volume pressures (num_volume_points, 1)



   .. py:method:: getitem_volume_velocity(idx)

      Retrieves volume velocity (num_volume_points, 3)



   .. py:method:: getitem_volume_vorticity(idx)

      Retrieves volume vorticity (num_volume_points, 3)



   .. py:method:: getitem_volume_distance_to_surface(idx)

      Retrieves distance to surface of volume points (num_volume_points, 1)