OneVsRestClassifier - ~m-label - logreg - {‘zipmap’: False}#

Fitted on a problem type ~m-label (see find_suitable_problem), method predict_proba matches output . Model was converted with additional parameter: <class 'sklearn.multiclass.OneVsOneClassifier'>={'zipmap': False}.

OneVsRestClassifier(estimator=LogisticRegression(random_state=0,
                                             solver='liblinear'),
                n_jobs=8)

index

0

skl_nop

4

skl_ncoef

6

skl_nlin

4

onx_size

1906

onx_nnodes

17

onx_ninits

5

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_ai.onnx.ml

1

onx_op_Cast

2

onx_op_ZipMap

1

onx_op_Reshape

1

onx_size_optim

1906

onx_nnodes_optim

17

onx_ninits_optim

5

fit_classes_.shape

3

fit_n_classes_

3

fit_estimators_.size

3

fit_estimators_.intercept_.shape

1

fit_estimators_.classes_.shape

2

fit_estimators_.coef_.shape

(1, 4)

fit_estimators_.n_iter_.shape

1
%0 X X float((0, 4)) LinearClassifier2 LinearClassifier (LinearClassifier2) classlabels_ints=[0 1] coefficients=[ 1.5634936  -0.16... intercepts=[ 0.951305 -0.951305... multi_class=1 post_transform=b'LOGISTIC' X->LinearClassifier2 LinearClassifier1 LinearClassifier (LinearClassifier1) classlabels_ints=[0 1] coefficients=[-0.54714024  1.21... intercepts=[-0.84863716  0.8486... multi_class=1 post_transform=b'LOGISTIC' X->LinearClassifier1 LinearClassifier LinearClassifier (LinearClassifier) classlabels_ints=[0 1] coefficients=[-0.2891426 -0.935... intercepts=[ 0.6001405 -0.60014... multi_class=1 post_transform=b'LOGISTIC' X->LinearClassifier output_label output_label int64((0, 3)) output_probability output_probability [{int64, {'kind': 'tensor', 'elem': 'float', 'shape': }}] starts starts int64((1,)) [1] Slice Slice (Slice) starts->Slice starts->Slice Slice1 Slice (Slice1) starts->Slice1 starts->Slice1 Slice2 Slice (Slice2) starts->Slice2 starts->Slice2 ends ends int64((1,)) [2] ends->Slice ends->Slice1 ends->Slice2 thresh thresh float32((1, 3)) [[0.5 0.5 0.5]] Sub Sub (Sub) thresh->Sub zero zero int64(()) 0 Clip Clip (Clip) zero->Clip shape_tensor shape_tensor int64((2,)) [-1  3] Reshape Reshape (Reshape) shape_tensor->Reshape label_2 label_2 probability_tensor2 probability_tensor2 Normalizer2 Normalizer (Normalizer2) norm=b'L1' probability_tensor2->Normalizer2 LinearClassifier2->label_2 LinearClassifier2->probability_tensor2 label_1 label_1 probability_tensor1 probability_tensor1 Normalizer1 Normalizer (Normalizer1) norm=b'L1' probability_tensor1->Normalizer1 LinearClassifier1->label_1 LinearClassifier1->probability_tensor1 label_0 label_0 probability_tensor probability_tensor Normalizer Normalizer (Normalizer) norm=b'L1' probability_tensor->Normalizer LinearClassifier->label_0 LinearClassifier->probability_tensor proba_2 proba_2 proba_2->Slice2 Normalizer2->proba_2 proba_1 proba_1 proba_1->Slice1 Normalizer1->proba_1 proba_0 proba_0 proba_0->Slice Normalizer->proba_0 probY_0 probY_0 Concat Concat (Concat) axis=1 probY_0->Concat Slice->probY_0 probY_1 probY_1 probY_1->Concat Slice1->probY_1 probY_2 probY_2 probY_2->Concat Slice2->probY_2 probabilities probabilities ZipMap ZipMap (ZipMap) classlabels_int64s=[0 1 2] probabilities->ZipMap probabilities->Sub Concat->probabilities ZipMap->output_probability threshed threshed Sign Sign (Sign) threshed->Sign Sub->threshed signed signed Cast Cast (Cast) to=7 signed->Cast Sign->signed signed_int64 signed_int64 signed_int64->Clip Cast->signed_int64 label1 label1 label1->Reshape Clip->label1 label label Cast1 Cast (Cast1) to=7 label->Cast1 Reshape->label Cast1->output_label

previous

OneVsRestClassifier - m-cl - logreg - {‘zipmap’: False}

next

multioutput