afe.load.importers.caffe2
=========================

.. py:module:: afe.load.importers.caffe2


Functions
---------

.. autoapisummary::

   afe.load.importers.caffe2.get_caffe2_predictor
   afe.load.importers.caffe2.validate_input_parameters
   afe.load.importers.caffe2.import_caffe2_to_tvm


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

.. py:function:: get_caffe2_predictor(init_net_file_path: str, predict_net_file_path: str)

   Returns a predictor object we can use to execute the caffe2 model
   :param init_net_file_path: filepath to the caffe2 .pb init_net file
   :param predict_net_file_path: filepath to the caffe2 .pb predict_net file


.. py:function:: validate_input_parameters(init_net_file_path: str, predict_net_file_path: str, shape_dict: Dict[str, Tuple[int, Ellipsis]], dtype_dict: Dict[str, afe.ir.tensor_type.ScalarType]) -> None

   Validates the user supplied input.
   :param init_net_file_path: filepath to the caffe2 .pb init_net file
   :param predict_net_file_path: filepath to the caffe2 .pb predict_net file
   :param shape_dict: dictionary of input names to input shapes (eg. (1,224,224,3))
   :param dtype_dict: dictionary of input names to input types (eg. float32 or int64)


.. py:function:: import_caffe2_to_tvm(init_net_file_path: str, predict_net_file_path: str, shape_dict: Dict[str, Tuple[int, Ellipsis]], dtype_dict: Dict[str, afe.ir.tensor_type.ScalarType]) -> afe._tvm._defines.TVMIRModule

   Use TVM frontend to import a caffe2 pb model into TVM Relay IR
   :param init_net_file_path: filepath to the caffe2 .pb init_net file
   :param predict_net_file_path: filepath to the caffe2 .pb predict_net file
   :param shape_dict: dictionary of input names to input shapes (eg. (1,224,224,3))
   :param dtype_dict: dictionary of input names to input types (eg. float32 or int64)
   :return: Imported TVM IR module.