sima_qat.sima_quantizer
=======================

.. py:module:: sima_qat.sima_quantizer


Classes
-------

.. autoapisummary::

   sima_qat.sima_quantizer.SimaQuantizer


Functions
---------

.. autoapisummary::

   sima_qat.sima_quantizer.get_sima_quantization_config


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

.. py:function:: get_sima_quantization_config(is_qat: bool = False)

.. py:class:: SimaQuantizer



   This quantizer definition uses XNNPACK implementation for the majority of ops. This is because
   the XNNPACK code simply looks for appropriate patterns, and applies the QuantizationAnnotation
   attributes accordingly. The quantization rules are defined separately, and specified here using
   Sima properties.


   .. py:attribute:: supported_config_and_operators


   .. py:attribute:: STATIC_QAT_ONLY_OPS
      :value: ['sima_conv_bn_hardtanh', 'conv_bn_relu', 'conv_bn']



   .. py:attribute:: STATIC_OPS
      :value: ['linear_relu', 'linear', 'sima_conv_add_or_mul_const', 'sima_conv_hardtanh', 'conv_relu',...



   .. py:attribute:: global_config
      :type:  Optional[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig]
      :value: None



   .. py:attribute:: operator_type_config
      :type:  Dict[torch._ops.OpOverloadPacket, Optional[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig]]


   .. py:attribute:: module_type_config
      :type:  Dict[Callable, Optional[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig]]


   .. py:attribute:: module_name_config
      :type:  Dict[str, Optional[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig]]


   .. py:method:: get_supported_quantization_configs() -> List[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig]
      :classmethod:



   .. py:method:: get_supported_operator_for_quantization_config(quantization_config: Optional[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig]) -> List[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.OperatorPatternType]
      :classmethod:



   .. py:method:: set_global(quantization_config: torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig) -> SimaQuantizer


   .. py:method:: set_operator_type(operator_type: torch._ops.OpOverloadPacket, quantization_config: torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig) -> SimaQuantizer


   .. py:method:: set_module_type(module_type: Callable, quantization_config: torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig)

      Set quantization_config for a submodule with type: `module_type`, for example:
      quantizer.set_module_name(Sub) or quantizer.set_module_name(nn.Linear), it will quantize all supported operator/operator
      patterns in the submodule with this module type with the given `quantization_config`



   .. py:method:: set_module_name(module_name: str, quantization_config: Optional[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.QuantizationConfig])

      Set quantization_config for a submodule with name: `module_name`, for example:
      quantizer.set_module_name("blocks.sub"), it will quantize all supported operator/operator
      patterns in the submodule with this module name with the given `quantization_config`



   .. py:method:: transform_for_annotation(model: torch.fx.GraphModule) -> torch.fx.GraphModule

      Transforms scalar values to tensor attributes



   .. py:method:: annotate(model: torch.fx.GraphModule) -> torch.fx.GraphModule

      just handling global spec for now



   .. py:method:: validate(model: torch.fx.GraphModule) -> None


   .. py:method:: get_supported_operators() -> List[torch.ao.quantization.quantizer.xnnpack_quantizer_utils.OperatorConfig]
      :classmethod: