module `testing.test_utils.utils_backend_python`#

Short summary#

module mlprodict.testing.test_utils.utils_backend_python

Inspired from sklearn-onnx, handles two backends.

Classes#

class	truncated documentation
`MockVariableName`	A string.
`MockVariableNameShape`	A string and a shape.
`MockVariableNameShapeType`	A string and a shape and a type.
`OnnxInference2`	onnxruntime API

Functions#

function	truncated documentation
`compare_runtime`	The function compares the expected output (computed with the model before being converted to ONNX) and the ONNX output …

Properties#

property	truncated documentation
`input_names`	Returns the names of all inputs. It does not include the optional inputs.
`input_names_shapes`	Returns the names and shapes of all inputs. This method assumes all inputs are tensors. It does not include …
`input_names_shapes_types`	Returns the names, shapes, types of all inputs. This method assumes all inputs are tensors. It does not …
`optional_inputs`	Returns the list of optional inputs (the model has an initalizer of the same name as one input).
`output_names`	Returns the names of all outputs.
`output_names_shapes`	Returns the names and shapes of all outputs. This method assumes all inputs are tensors.
`output_names_shapes_types`	Returns the names, shapes, types of all outputs. This method assumes all inputs are tensors. It does not …
`shape`	returns shape
`shape`	returns shape
`shape`	returns shape
`type`	returns type
`type`	returns type
`type`	returns type

Methods#

method	truncated documentation
`__init__`
`__init__`
`__init__`
`get_inputs`	onnxruntime API
`get_outputs`	onnxruntime API
`run`	onnxruntime API
`run_in_scan`	Instance to run in operator scan.

Documentation#

Inspired from sklearn-onnx, handles two backends.

source on GitHub

class mlprodict.testing.test_utils.utils_backend_python.MockVariableName(name)#

Bases: object

A string.

__init__(name)#

property shape#: returns shape

property type#: returns type

class mlprodict.testing.test_utils.utils_backend_python.MockVariableNameShape(name, sh)#

Bases: MockVariableName

A string and a shape.

__init__(name, sh)#

property shape#: returns shape

class mlprodict.testing.test_utils.utils_backend_python.MockVariableNameShapeType(name, sh, stype)#

Bases: MockVariableNameShape

A string and a shape and a type.

__init__(name, sh, stype)#

property type#: returns type

class mlprodict.testing.test_utils.utils_backend_python.OnnxInference2(onnx_or_bytes_or_stream, runtime=None, skip_run=False, inplace=True, input_inplace=False, ir_version=None, target_opset=None, runtime_options=None, session_options=None, inside_loop=False, static_inputs=None, new_outputs=None, new_opset=None, existing_functions=None)#

Bases: OnnxInference

onnxruntime API

get_inputs()#: onnxruntime API

get_outputs()#: onnxruntime API

run(name, inputs, *args, **kwargs)#: onnxruntime API

run_in_scan(inputs, attributes=None, verbose=0, fLOG=None)#: Instance to run in operator scan.

mlprodict.testing.test_utils.utils_backend_python.compare_runtime(test, decimal=5, options=None, verbose=False, context=None, comparable_outputs=None, intermediate_steps=False, classes=None, disable_optimisation=False)#

The function compares the expected output (computed with the model before being converted to ONNX) and the ONNX output produced with module onnxruntime or mlprodict.

Parameters:

test – dictionary with the following keys: - onnx: onnx model (filename or object) - expected: expected output (filename pkl or object) - data: input data (filename pkl or object)
decimal – precision of the comparison
options – comparison options
context – specifies custom operators
verbose – in case of error, the function may print more information on the standard output
comparable_outputs – compare only these outputs
intermediate_steps – displays intermediate steps in case of an error
classes – classes names (if option ‘nocl’ is used)
disable_optimisation – disable optimisation the runtime may do

Returns:

tuple (outut, lambda function to run the predictions)

The function does not return anything but raises an error if the comparison failed.