sima_utils.transformer.model.vision_model
=========================================

.. py:module:: sima_utils.transformer.model.vision_model


Classes
-------

.. autoapisummary::

   sima_utils.transformer.model.vision_model.VisionModel
   sima_utils.transformer.model.vision_model.VisionLayerModel


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

.. py:class:: VisionModel



   Vision model implementation.


   .. py:attribute:: is_single_vision_model
      :type:  bool
      :value: True



   .. py:attribute:: actual_num_hidden_layers
      :type:  int


   .. py:method:: gen_files(gen_mode: sima_utils.transformer.model.base.FileGenMode, *, precision: sima_utils.transformer.model.base.FileGenPrecision | dict[str, sima_utils.transformer.model.base.FileGenPrecision] | None = None, log_level: int = logging.NOTSET, num_processes: int = 1, part: str | None = None, part_idx: int | None = None, resume: bool = False)


.. py:class:: VisionLayerModel



   Vision model for each transformer layer with embedding or multimodal projection.


   .. py:attribute:: layer_idx
      :type:  int


   .. py:attribute:: num_layers
      :type:  int


   .. py:attribute:: include_embeddings
      :type:  bool


   .. py:attribute:: include_mm_proj
      :type:  bool


   .. py:method:: gen_onnx_files()