module testing.test_utils.utils_backend_common

Short summary

module mlprodict.testing.test_utils.utils_backend_common

Inspired from sklearn-onnx, handles two backends.

source on GitHub

Classes

class	truncated documentation
ExpectedAssertionError	Expected failure.
OnnxBackendAssertionError	Expected failure.
OnnxBackendMissingNewOnnxOperatorException	Raised when onnxruntime or mlprodict does not implement a new operator defined in the latest onnx. …
OnnxRuntimeMissingNewOnnxOperatorException	Raised when a new operator was added but cannot be found.

Functions

function	truncated documentation
_compare_expected	Compares the expected output against the runtime outputs. This is specific to onnxruntime or mlprodict. …
_create_column	Creates a column from values with dtype
_post_process_output	Applies post processings before running the comparison such as changing type from list to arrays.
compare_outputs	Compares expected values and output. Returns None if no error, an exception message otherwise.
evaluate_condition	Evaluates a condition such as StrictVersion(onnxruntime.__version__) <= StrictVersion('0.1.3')
extract_options	Extracts comparison option from filename. As example, Binarizer-SkipDim1 means options SkipDim1 is enabled. …
is_backend_enabled	Tells if a backend is enabled. Raises an exception if backend != ‘onnxruntime’. Unit tests only test models against …
load_data_and_model	Loads every file in a dictionary {key: filename}. The extension is either pkl and onnx and determines how it …

Documentation

Inspired from sklearn-onnx, handles two backends.

source on GitHub

exception mlprodict.testing.test_utils.utils_backend_common.ExpectedAssertionError

Bases: Exception

Expected failure.

source on GitHub

exception mlprodict.testing.test_utils.utils_backend_common.OnnxBackendAssertionError

Bases: AssertionError

Expected failure.

source on GitHub

exception mlprodict.testing.test_utils.utils_backend_common.OnnxBackendMissingNewOnnxOperatorException

Bases: OnnxBackendAssertionError

Raised when onnxruntime or mlprodict does not implement a new operator defined in the latest onnx.

source on GitHub

exception mlprodict.testing.test_utils.utils_backend_common.OnnxRuntimeMissingNewOnnxOperatorException

Bases: OnnxBackendAssertionError

Raised when a new operator was added but cannot be found.

source on GitHub

mlprodict.testing.test_utils.utils_backend_common._compare_expected(expected, output, sess, onnx_model, decimal=5, verbose=False, classes=None, **kwargs)

Compares the expected output against the runtime outputs. This is specific to onnxruntime or mlprodict.

source on GitHub

mlprodict.testing.test_utils.utils_backend_common._create_column(values, dtype)

Creates a column from values with dtype

mlprodict.testing.test_utils.utils_backend_common._post_process_output(res)

Applies post processings before running the comparison such as changing type from list to arrays.

source on GitHub

mlprodict.testing.test_utils.utils_backend_common.compare_outputs(expected, output, verbose=False, **kwargs)

Compares expected values and output. Returns None if no error, an exception message otherwise.

source on GitHub

mlprodict.testing.test_utils.utils_backend_common.evaluate_condition(backend, condition)

Evaluates a condition such as StrictVersion(onnxruntime.__version__) <= StrictVersion('0.1.3')

source on GitHub

mlprodict.testing.test_utils.utils_backend_common.extract_options(name)

Extracts comparison option from filename. As example, Binarizer-SkipDim1 means options SkipDim1 is enabled. (1, 2) and (2,) are considered equal. Available options: see dump_data_and_model.

source on GitHub

mlprodict.testing.test_utils.utils_backend_common.is_backend_enabled(backend)

Tells if a backend is enabled. Raises an exception if backend != ‘onnxruntime’. Unit tests only test models against this backend.

source on GitHub

mlprodict.testing.test_utils.utils_backend_common.load_data_and_model(items_as_dict, **context)

Loads every file in a dictionary {key: filename}. The extension is either pkl and onnx and determines how it it loaded. If the value is not a string, the function assumes it was already loaded.

source on GitHub