GaussianProcessRegressor - b-reg - expsine - cdist#

Fitted on a problem type b-reg (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=0)

index

0

skl_nop

1

onx_size

3250

onx_nnodes

11

onx_ninits

9

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_com.microsoft

1

onx_op_Reshape

1

onx_size_optim

3250

onx_nnodes_optim

11

onx_ninits_optim

9

fit_X_train_.shape

(112, 4)

fit_y_train_.shape

112

fit_log_marginal_likelihood_value_.shape

1

fit_L_.shape

(112, 112)

fit_alpha_.shape

112
%0 X X float((0, 4)) kgpd_CDist CDist (kgpd_CDist) metric=b'euclidean' X->kgpd_CDist GPmean GPmean float((0, 1)) kgpd_CDistcst kgpd_CDistcst float32((112, 4)) [[ 4.3017502e+00  3.9453187e+00  9.2195314e-01  1.... kgpd_CDistcst->kgpd_CDist kgpd_Divcst kgpd_Divcst float32((1,)) [7400.744] kgpd_Div Div (kgpd_Div) kgpd_Divcst->kgpd_Div kgpd_Mulcst kgpd_Mulcst float32((1,)) [3.1415927] kgpd_Mul Mul (kgpd_Mul) kgpd_Mulcst->kgpd_Mul kgpd_Divcst1 kgpd_Divcst1 float32((1,)) [0.00318633] kgpd_Div1 Div (kgpd_Div1) kgpd_Divcst1->kgpd_Div1 kgpd_Powcst kgpd_Powcst float32((1,)) [2.] kgpd_Pow Pow (kgpd_Pow) kgpd_Powcst->kgpd_Pow kgpd_Mulcst1 kgpd_Mulcst1 float32((1,)) [-2.] kgpd_Mul1 Mul (kgpd_Mul1) kgpd_Mulcst1->kgpd_Mul1 gpr_MatMulcst gpr_MatMulcst float32((112,)) [-0.01557005 -0.01321393  0.07640724 -0.00190291 -... gpr_MatMul MatMul (gpr_MatMul) gpr_MatMulcst->gpr_MatMul gpr_Addcst gpr_Addcst float32((1, 1)) [[0.]] gpr_Add Add (gpr_Add) gpr_Addcst->gpr_Add Re_Reshapecst Re_Reshapecst int64((2,)) [-1  1] Re_Reshape Reshape (Re_Reshape) allowzero=0 Re_Reshapecst->Re_Reshape kgpd_dist kgpd_dist kgpd_dist->kgpd_Div kgpd_CDist->kgpd_dist kgpd_C03 kgpd_C03 kgpd_C03->kgpd_Mul kgpd_Div->kgpd_C03 kgpd_C02 kgpd_C02 kgpd_Sin Sin (kgpd_Sin) kgpd_C02->kgpd_Sin kgpd_Mul->kgpd_C02 kgpd_output02 kgpd_output02 kgpd_output02->kgpd_Div1 kgpd_Sin->kgpd_output02 kgpd_C01 kgpd_C01 kgpd_C01->kgpd_Pow kgpd_Div1->kgpd_C01 kgpd_Z0 kgpd_Z0 kgpd_Z0->kgpd_Mul1 kgpd_Pow->kgpd_Z0 kgpd_C0 kgpd_C0 kgpd_Exp Exp (kgpd_Exp) kgpd_C0->kgpd_Exp kgpd_Mul1->kgpd_C0 kgpd_output01 kgpd_output01 kgpd_output01->gpr_MatMul kgpd_Exp->kgpd_output01 gpr_Y0 gpr_Y0 gpr_Y0->gpr_Add gpr_MatMul->gpr_Y0 gpr_C0 gpr_C0 gpr_C0->Re_Reshape gpr_Add->gpr_C0 Re_Reshape->GPmean

previous

GaussianProcessRegressor - ~b-reg-std-NSV-64 - rbf - cdist

next

GaussianProcessRegressor - b-reg - dotproduct -