Static Methods#

Summary#

staticmethod

class parent

truncated documentation

__class_getitem__

NDArray

Overwrites this method.

__class_getitem__

OnnxOperator

Enables expression cls[opset]. It returns the appropriate class cls_opset. Parameter op_version should be …

__class_getitem__

OnnxOperatorFunction

Enables expression cls[opset]. It returns the appropriate class cls_opset. Parameter op_version should be …

_add_node

XGBClassifierConverter

_add_node

XGBConverter

_add_node

XGBRegressorConverter

_create_model

GPT2TokenizerTransformer

_create_model

SentencePieceTokenizerTransformer

_fill_node_attributes

XGBClassifierConverter

_fill_node_attributes

XGBConverter

_fill_node_attributes

XGBRegressorConverter

_find_static_inputs

OnnxInferenceNode

Determines the loop inputs. It is any defined inputs by the subgraphs + any results used as a constant in …

_fmatmul00

FusedMatMul

_fmatmul01

FusedMatMul

_fmatmul10

FusedMatMul

_fmatmul11

FusedMatMul

_gemm00

Gemm

_gemm01

Gemm

_gemm10

Gemm

_gemm11

Gemm

_generate_classes

WrappedLightGbmBooster

_get_default_tree_attribute_pairs

XGBClassifierConverter

_get_default_tree_attribute_pairs

XGBConverter

_get_default_tree_attribute_pairs

XGBRegressorConverter

_get_shape

OnnxShapeInference

_get_type_property

OnnxInference

_infer_merged_type

ShapeObject

_infer_merged_type

ShapeObjectFct

_load

OnnxBackendTest

_make_tuple

NumpyCode

_merge_op_version

OnnxOperator

_merge_op_version

OnnxOperatorFunction

_node_to_graph_get_type

OnnxOperator

_node_to_graph_get_type

OnnxOperatorFunction

_node_to_graph_preprocess_list

OnnxOperator

_node_to_graph_preprocess_list

OnnxOperatorFunction

_node_to_graph_process_input

OnnxOperator

_node_to_graph_process_input

OnnxOperatorFunction

_node_to_graph_reorder_by_name

OnnxOperator

_node_to_graph_reorder_by_name

OnnxOperatorFunction

_norm_L1_inplace

Normalizer

_norm_max_inplace

Normalizer

_onnx2bigraph_basic

BiGraph

Implements graph type ‘basic’ for function onnx2bigraph().

_onnx2bigraph_simplified

BiGraph

Implements graph type ‘simplified’ for function onnx2bigraph().

_process_type

NDArraySameType

Nicknames such as floats, int, ints, all can be used to describe multiple inputs for a signature. …

_process_type

NDArraySameTypeSameShape

Nicknames such as floats, int, ints, all can be used to describe multiple inputs for a signature. …

_process_type

NDArrayType

Nicknames such as floats, int, ints, all can be used to describe multiple inputs for a signature. …

_process_type

NDArrayTypeSameShape

Nicknames such as floats, int, ints, all can be used to describe multiple inputs for a signature. …

_process_type

_NDArrayAlias

Nicknames such as floats, int, ints, all can be used to describe multiple inputs for a signature. …

_read_proto_from_file

OnnxBackendTest

_remap_nodeid

XGBClassifierConverter

_remap_nodeid

XGBConverter

_remap_nodeid

XGBRegressorConverter

_same_

DimensionObject

Returns obj if obj is DimensionObject otherwise converts it.

_sort

OnnxBackendTest

_to_onnx

OnnxSubEstimator

Converts a model into ONNX and inserts it into an ONNX graph.

_to_onnx_sklearn

OnnxSubEstimator

Converts a scikit-learn model into ONNX and inserts it into an ONNX graph. The library relies on …

attribute_to_value

OnnxOperatorFunction

Converts an attribute into a value using python structures.

broadcast

ShapeResult

Broadcasts dimensions for an element wise operator.

build_einsum

CachedEinsum

Creates an instance of CachedEinsum.

build_rev_keys

ZipMapDictionary

cache

AutoAction

Caches the result of a function.

common_members

XGBClassifierConverter

common to regresssor and classifier

common_members

XGBConverter

common to regresssor and classifier

common_members

XGBRegressorConverter

common to regresssor and classifier

convert

XGBClassifierConverter

convert method

convert

XGBRegressorConverter

converter method

create_inference_session

OnnxInferenceBackend

Instantiates an instance of class @see cl OnnxInference. This method should be overwritten to change the runtime …

create_inference_session

OnnxInferenceBackendMicro

create_inference_session

OnnxInferenceBackendOrt

create_inference_session

OnnxInferenceBackendPyC

create_inference_session

OnnxInferenceBackendPyEval

create_inference_session

OnnxInferenceBackendShape

einsum_shape

ShapeObject

Computes einsum shapes. Not the most efficient one as it creates variables of the given shapes.

einsum_shape

ShapeObjectFct

Computes einsum shapes. Not the most efficient one as it creates variables of the given shapes.

enumerate_create

OnnxTransformer

Creates multiple OnnxTransformer, one for each requested intermediate node. onnx_bytes : bytes …

fill_tree_attributes

XGBClassifierConverter

fills tree attributes

fill_tree_attributes

XGBConverter

fills tree attributes

fill_tree_attributes

XGBRegressorConverter

fills tree attributes

from_pb

Variable

Creates a Variable from a protobuf object.

from_skl2onnx

Variable

Converts var from sklearn-onnx into this class.

gather_shape

ShapeObject

Computes Gather shapes.

gather_shape

ShapeObjectFct

Computes Gather shapes.

get_xgb_params

XGBClassifierConverter

Retrieves parameters of a model.

get_xgb_params

XGBConverter

Retrieves parameters of a model.

get_xgb_params

XGBRegressorConverter

Retrieves parameters of a model.

guess_type

MLActionCst

Guesses a type given a value.

guess_type

MLActionVar

Guesses a type given a value.

is_compatible

OnnxInferenceBackend

Returns whether the model is compatible with the backend.

is_compatible

OnnxInferenceBackendMicro

Returns whether the model is compatible with the backend.

is_compatible

OnnxInferenceBackendOrt

Returns whether the model is compatible with the backend.

is_compatible

OnnxInferenceBackendPyC

Returns whether the model is compatible with the backend.

is_compatible

OnnxInferenceBackendPyEval

Returns whether the model is compatible with the backend.

is_compatible

OnnxInferenceBackendShape

Returns whether the model is compatible with the backend.

is_opset_supported

OnnxInferenceBackend

Returns whether the opset for the model is supported by the backend.

is_opset_supported

OnnxInferenceBackendMicro

Returns whether the opset for the model is supported by the backend.

is_opset_supported

OnnxInferenceBackendOrt

Returns whether the opset for the model is supported by the backend.

is_opset_supported

OnnxInferenceBackendPyC

Returns whether the opset for the model is supported by the backend.

is_opset_supported

OnnxInferenceBackendPyEval

Returns whether the opset for the model is supported by the backend.

is_opset_supported

OnnxInferenceBackendShape

Returns whether the opset for the model is supported by the backend.

norm_l1

Normalizer

L1 normalization

norm_l2

Normalizer

L2 normalization

norm_max

Normalizer

max normalization

number2alpha

_GraphBuilder

Converts a numbers into a string keeping the same alphabetical order.

onnx_graph_distance

BiGraph

Computes a distance between two ONNX graphs. They must not be too big otherwise this function might take for ever. …

prepare

OnnxInferenceBackend

Loads the model and creates @see cl OnnxInference.

prepare

OnnxInferenceBackendMicro

Loads the model and creates @see cl OnnxInference.

prepare

OnnxInferenceBackendOrt

Loads the model and creates @see cl OnnxInference.

prepare

OnnxInferenceBackendPyC

Loads the model and creates @see cl OnnxInference.

prepare

OnnxInferenceBackendPyEval

Loads the model and creates @see cl OnnxInference.

prepare

OnnxInferenceBackendShape

Loads the model and creates @see cl OnnxInference.

print_node

CodeNodeVisitor

Debugging purpose.

print_node

CodeNodeVisitor

Debugging purpose.

process_profiling

OnnxWholeSession

Flattens json returned by onnxruntime profiling.

run_model

OnnxInferenceBackend

Computes the prediction.

run_model

OnnxInferenceBackendMicro

Computes the prediction.

run_model

OnnxInferenceBackendOrt

Computes the prediction.

run_model

OnnxInferenceBackendPyC

Computes the prediction.

run_model

OnnxInferenceBackendPyEval

Computes the prediction.

run_model

OnnxInferenceBackendShape

Computes the prediction.

run_node

OnnxInferenceBackend

This method is not implemented as it is much more efficient to run a whole model than every node independently.

run_node

OnnxInferenceBackendMicro

This method is not implemented as it is much more efficient to run a whole model than every node independently.

run_node

OnnxInferenceBackendOrt

This method is not implemented as it is much more efficient to run a whole model than every node independently.

run_node

OnnxInferenceBackendPyC

This method is not implemented as it is much more efficient to run a whole model than every node independently.

run_node

OnnxInferenceBackendPyEval

This method is not implemented as it is much more efficient to run a whole model than every node independently.

run_node

OnnxInferenceBackendShape

This method is not implemented as it is much more efficient to run a whole model than every node independently.

supports_device

OnnxInferenceBackend

Checks whether the backend is compiled with particular device support.

supports_device

OnnxInferenceBackendMicro

Checks whether the backend is compiled with particular device support.

supports_device

OnnxInferenceBackendOrt

Checks whether the backend is compiled with particular device support.

supports_device

OnnxInferenceBackendPyC

Checks whether the backend is compiled with particular device support.

supports_device

OnnxInferenceBackendPyEval

Checks whether the backend is compiled with particular device support.

supports_device

OnnxInferenceBackendShape

Checks whether the backend is compiled with particular device support.

validate

XGBClassifierConverter

validate

XGBConverter

validates the model

validate

XGBRegressorConverter