:orphan:

:py:mod:`neural_compressor.torch.algorithms.static_quant.utility`
=================================================================

.. py:module:: neural_compressor.torch.algorithms.static_quant.utility


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

Classes
~~~~~~~

.. autoapisummary::

   neural_compressor.torch.algorithms.static_quant.utility.Statistics
   neural_compressor.torch.algorithms.static_quant.utility.TransformerBasedModelBlockPatternDetector



Functions
~~~~~~~~~

.. autoapisummary::

   neural_compressor.torch.algorithms.static_quant.utility.check_cfg_and_qconfig
   neural_compressor.torch.algorithms.static_quant.utility.generate_activation_observer
   neural_compressor.torch.algorithms.static_quant.utility.get_quantizable_ops_recursively
   neural_compressor.torch.algorithms.static_quant.utility.simple_inference
   neural_compressor.torch.algorithms.static_quant.utility.dump_model_op_stats
   neural_compressor.torch.algorithms.static_quant.utility.get_depth
   neural_compressor.torch.algorithms.static_quant.utility.get_dict_at_depth
   neural_compressor.torch.algorithms.static_quant.utility.get_element_under_depth
   neural_compressor.torch.algorithms.static_quant.utility.parse_cfgs
   neural_compressor.torch.algorithms.static_quant.utility.get_quantizable_ops_from_cfgs



.. py:function:: check_cfg_and_qconfig(user_cfg, cfgs, op_infos_from_cfgs, output_tensor_ids_op_name)

   Check configs and quantization configs.

   :param user_cfg: quantization configuration for ops.
   :type user_cfg: dict
   :param cfgs: configs loaded from ipex config path.
   :type cfgs: dict
   :param op_infos_from_cfgs: dict containing configs that have been parsed for each op.
   :type op_infos_from_cfgs: dict
   :param output_tensor_ids_op_name: dict containing op names corresponding to 'op_infos_from_cfgs'.
   :type output_tensor_ids_op_name: dict

   :returns: updated configs.
   :rtype: cfgs (dict)


.. py:function:: generate_activation_observer(scheme, algorithm, smooth_quant=False, smooth_quant_enable=False)

   This is a helper method to generate a dict containing activation observer info.

   :param scheme: Quantization scheme to be used.
   :type scheme: str
   :param algorithm: What algorithm for computing the quantization parameters based on.
   :type algorithm: str

   :returns: A dict containing observer info.zs


.. py:function:: get_quantizable_ops_recursively(model, example_inputs)

   Get all quantizable ops from model.

   :param model: input model
   :type model: object
   :param example_inputs: used to trace torch model.
   :type example_inputs: dict|list|tuple|torch.Tensor

   :returns: list of tuples of op_name and op_type.
             cfgs (dict): dict of configuration
   :rtype: quantizable_ops (list)


.. py:function:: simple_inference(q_model, example_inputs, iterations=1)

   The function is used for ipex warm-up inference.


.. py:function:: dump_model_op_stats(user_cfg)

   This is a function to dump quantizable ops of model to user.

   :param user_cfg: quantization config
   :type user_cfg: dict

   :returns: None


.. py:function:: get_depth(d) -> int

   Query the depth of the dict.


.. py:function:: get_dict_at_depth(d, target_depth, result, depth=0)

   Get all sub-dicts that are at a specified depth in a nested dict.


.. py:function:: get_element_under_depth(d, ops_lst)

   Get all values in a nested dict.


.. py:function:: parse_cfgs(cfgs)

   Parse configs.

   :param cfgs: the input configs.
   :type cfgs: dict

   :returns: list of op names.
             tune_cfg (dict): dictionary of quantization configuration.
             op_infos_from_cfgs (dict): op infos from configs.
             output_tensor_ids_op_name (dict): dictionary of output tensor op names.
   :rtype: ops_name (list)


.. py:function:: get_quantizable_ops_from_cfgs(ops_name, op_infos_from_cfgs, input_tensor_ids_op_name)

   Get quantizable ops from configs, combine fused ops as one op.

   :param ops_name: list of op names.
   :type ops_name: list
   :param op_infos_from_cfgs: op infos from configs.
   :type op_infos_from_cfgs: dict
   :param input_tensor_ids_op_name: dictionary of input tensor op names.
   :type input_tensor_ids_op_name: dict

   :returns: cfgs (dict).


.. py:class:: Statistics(data, header, field_names, output_handle=logger.info)


   The statistics printer.


.. py:class:: TransformerBasedModelBlockPatternDetector(model: torch.nn.Module, pattern_lst: List[List[Union[str, int]]] = BLOCK_PATTERNS)


   Detect the attention block and FFN block in transformer-based model.