module onnx_conv.sklconv.svm_converters
#
Short summary#
module mlprodict.onnx_conv.sklconv.svm_converters
Rewrites some of the converters implemented in sklearn-onnx.
Functions#
function |
truncated documentation |
---|---|
Defines op_type and op_domain based on dtype. |
|
Defines op_type and op_domain based on dtype. |
|
Rewrites the converters implemented in sklearn-onnx to support an operator supporting doubles. |
|
Rewrites the converters implemented in sklearn-onnx to support an operator supporting doubles. |
Documentation#
Rewrites some of the converters implemented in sklearn-onnx.
- mlprodict.onnx_conv.sklconv.svm_converters._convert_sklearn_svm_classifier(scope, operator, container, op_type='SVMClassifier', op_domain='ai.onnx.ml', op_version=1)#
Converter for model SVC, NuSVC. The converted model in ONNX produces the same results as the original model except when probability=False: onnxruntime and scikit-learn do not return the same raw scores. scikit-learn returns aggregated scores as a matrix[N, C] coming from _ovr_decision_function. onnxruntime returns the raw score from svm algorithm as a matrix[N, (C(C-1)/2].
- mlprodict.onnx_conv.sklconv.svm_converters._op_type_domain_classifier(dtype)#
Defines op_type and op_domain based on dtype.
- mlprodict.onnx_conv.sklconv.svm_converters._op_type_domain_regressor(dtype)#
Defines op_type and op_domain based on dtype.
- mlprodict.onnx_conv.sklconv.svm_converters.new_convert_sklearn_svm_classifier(scope, operator, container)#
Rewrites the converters implemented in sklearn-onnx to support an operator supporting doubles.
- mlprodict.onnx_conv.sklconv.svm_converters.new_convert_sklearn_svm_regressor(scope, operator, container)#
Rewrites the converters implemented in sklearn-onnx to support an operator supporting doubles.