sima_utils.transformer.devkit.utils
===================================

.. py:module:: sima_utils.transformer.devkit.utils


Attributes
----------

.. autoapisummary::

   sima_utils.transformer.devkit.utils.VLM_EVAL_PRINT_INOUTS
   sima_utils.transformer.devkit.utils.VLM_EVAL_PROFILE
   sima_utils.transformer.devkit.utils.mla_rt_api


Classes
-------

.. autoapisummary::

   sima_utils.transformer.devkit.utils.MLAGpio
   sima_utils.transformer.devkit.utils.MLABuffer
   sima_utils.transformer.devkit.utils.MLAModel


Functions
---------

.. autoapisummary::

   sima_utils.transformer.devkit.utils.connect_mla_rt_api
   sima_utils.transformer.devkit.utils.disconnect_mla_rt_api


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

.. py:data:: VLM_EVAL_PRINT_INOUTS

.. py:data:: VLM_EVAL_PROFILE

.. py:data:: mla_rt_api
   :type:  mla_rt_service.devkit_python_api.Mla_rt_apis | None
   :value: None


.. py:function:: connect_mla_rt_api(lib_mla_rt_path: pathlib.Path | None)

.. py:function:: disconnect_mla_rt_api()

.. py:class:: MLAGpio(lib_mla_rt_path: pathlib.Path)

   .. py:attribute:: lib_mla_rt_path


   .. py:attribute:: mla_poke_path


   .. py:method:: mla_control_gpio(pin: int, value: int) -> None


   .. py:method:: mla_read_reg(reg_addr: str) -> int


   .. py:method:: mla_write_reg(reg_addr: str, value: str) -> None


.. py:class:: MLABuffer

   A wrapper around the Mla_buffer defined in devkit apis with additional buffer properties.


   .. py:attribute:: name
      :type:  str


   .. py:attribute:: shape
      :type:  tuple[int, Ellipsis]


   .. py:attribute:: dtype
      :type:  str


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



   .. py:attribute:: mla_buffer
      :type:  mla_rt_service.devkit_python_api.Mla_buffer | None
      :value: None



   .. py:property:: itemsize
      :type: int


      Returns the size of the buffer item in bytes.


   .. py:property:: size
      :type: int


      Returns the size of the buffer in bytes.


   .. py:method:: allocate()

      Creates and allocates the buffer.



   .. py:method:: free()

      Frees the buffer if it is allocated.



   .. py:method:: upload(data: numpy.ndarray)

      Uploads the data to the buffer.



   .. py:method:: download(begin: tuple[int, Ellipsis] | None = None, shape: tuple[int, Ellipsis] | None = None) -> numpy.ndarray

      Downloads the data from the buffer.



   .. py:method:: get_buf_addr_len(begin: tuple[int, Ellipsis] | None = None, shape: tuple[int, Ellipsis] | None = None) -> mla_rt_service.devkit_python_api.DADDR_LEN

      Creates a DADDR_LEN object for the slice of the buffer defined by begin and shape.



.. py:class:: MLAModel

   Model definition with buffers for ifms and ofms.


   .. py:attribute:: model_path
      :type:  pathlib.Path


   .. py:attribute:: ifms
      :type:  list[MLABuffer | tuple[MLABuffer, tuple[int, Ellipsis] | None, tuple[int, Ellipsis] | None]]


   .. py:attribute:: ofms
      :type:  list[MLABuffer | tuple[MLABuffer, tuple[int, Ellipsis] | None, tuple[int, Ellipsis] | None]]


   .. py:attribute:: ifm_buffer_addr_lens
      :type:  list[mla_rt_service.devkit_python_api.DADDR_LEN | None]
      :value: []



   .. py:attribute:: ofm_buffer_addr_lens
      :type:  list[mla_rt_service.devkit_python_api.DADDR_LEN]
      :value: []



   .. py:attribute:: unique_models
      :type:  ClassVar[dict[pathlib.Path, Any | None]]


   .. py:method:: load() -> Any

      Loads the model and caches the model object so that the model is only loaded once.



   .. py:method:: free()

      Frees the model if it is loaded.



   .. py:method:: run(update_ifm_addr_len_dict: dict[int, Any] | None = None) -> float

      Loads and runs the model. Updates the ifm addr_len if needed.
      Returns the runtime of the model.



   .. py:method:: load_all_models(do_parallel_load: bool = False, elf_dir: pathlib.Path | None = None)
      :classmethod: