GaussianProcessRegressor - ~b-reg-NF-std-64 - expsine - cdist#

Fitted on a problem type ~b-reg-NF-std-64 (see find_suitable_problem), method predict matches output . Model was converted with additional parameter: <class 'sklearn.gaussian_process._gpr.GaussianProcessRegressor'>={'optim': 'cdist'}.

GaussianProcessRegressor(alpha=20.0,
                     kernel=ExpSineSquared(length_scale=1, periodicity=1),
                     random_state=42)

index

0

skl_nop

1

onx_size

714

onx_nnodes

8

onx_ninits

2

onx_doc_string

onx_ir_version

8

onx_domain

ai.onnx

onx_model_version

0

onx_producer_name

skl2onnx

onx_producer_version

1.11.1

onx_

15

onx_size_optim

581

onx_nnodes_optim

6

onx_ninits_optim

2
%0 X X double((0, 4)) Sh_Shape Shape (Sh_Shape) X->Sh_Shape Sh_Shape1 Shape (Sh_Shape1) X->Sh_Shape1 GPmean GPmean double((0, 1)) GPcovstd GPcovstd double(('?',)) Re_ReduceSumcst Re_ReduceSumcst int64((1,)) [1] Re_ReduceSum1 ReduceSum (Re_ReduceSum1) keepdims=0 Re_ReduceSumcst->Re_ReduceSum1 Re_ReduceSum ReduceSum (Re_ReduceSum) keepdims=1 Re_ReduceSumcst->Re_ReduceSum Ad_Addcst Ad_Addcst float64((1,)) [1.] Ad_Add Add (Ad_Add) Ad_Addcst->Ad_Add Sh_shape0 Sh_shape0 Co_ConstantOfShape ConstantOfShape (Co_ConstantOfShape) value=[0.] Sh_shape0->Co_ConstantOfShape Sh_Shape->Sh_shape0 Sh_shape02 Sh_shape02 Co_ConstantOfShape1 ConstantOfShape (Co_ConstantOfShape1) value=[0.] Sh_shape02->Co_ConstantOfShape1 Sh_Shape1->Sh_shape02 Co_output0 Co_output0 Co_output0->Re_ReduceSum Co_ConstantOfShape->Co_output0 Co_output02 Co_output02 Co_output02->Re_ReduceSum1 Co_ConstantOfShape1->Co_output02 Re_reduced0 Re_reduced0 Re_reduced0->Ad_Add Re_ReduceSum1->Re_reduced0 Re_ReduceSum->GPmean Ad_C0 Ad_C0 Sq_Sqrt Sqrt (Sq_Sqrt) Ad_C0->Sq_Sqrt Ad_Add->Ad_C0 Sq_Sqrt->GPcovstd

previous

GaussianProcessRegressor - ~b-reg-NF-cov-64 - rbf - cdist

next

GaussianProcessRegressor - ~b-reg-NF-std-64 - dotproduct -