onnx_test_reduce_max_do_not_keepdims_example.go

package onnxtest

// this file is auto-generated... DO NOT EDIT

import (
	"github.com/godshen/onnx-go/backend/testbackend"
	"gorgonia.org/tensor"
)

func init() {
	testbackend.Register("ReduceMax", "TestReduceMaxDoNotKeepdimsExample", NewTestReduceMaxDoNotKeepdimsExample)
}

// NewTestReduceMaxDoNotKeepdimsExample version: 3.
func NewTestReduceMaxDoNotKeepdimsExample() *testbackend.TestCase {
	return &testbackend.TestCase{
		OpType: "ReduceMax",
		Title:  "TestReduceMaxDoNotKeepdimsExample",
		ModelB: []byte{0x8, 0x3, 0x12, 0xc, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2d, 0x74, 0x65, 0x73, 0x74, 0x3a, 0x9a, 0x1, 0xa, 0x38, 0xa, 0x4, 0x64, 0x61, 0x74, 0x61, 0x12, 0x7, 0x72, 0x65, 0x64, 0x75, 0x63, 0x65, 0x64, 0x22, 0x9, 0x52, 0x65, 0x64, 0x75, 0x63, 0x65, 0x4d, 0x61, 0x78, 0x2a, 0xb, 0xa, 0x4, 0x61, 0x78, 0x65, 0x73, 0x40, 0x1, 0xa0, 0x1, 0x7, 0x2a, 0xf, 0xa, 0x8, 0x6b, 0x65, 0x65, 0x70, 0x64, 0x69, 0x6d, 0x73, 0x18, 0x0, 0xa0, 0x1, 0x2, 0x12, 0x27, 0x74, 0x65, 0x73, 0x74, 0x5f, 0x72, 0x65, 0x64, 0x75, 0x63, 0x65, 0x5f, 0x6d, 0x61, 0x78, 0x5f, 0x64, 0x6f, 0x5f, 0x6e, 0x6f, 0x74, 0x5f, 0x6b, 0x65, 0x65, 0x70, 0x64, 0x69, 0x6d, 0x73, 0x5f, 0x65, 0x78, 0x61, 0x6d, 0x70, 0x6c, 0x65, 0x5a, 0x1a, 0xa, 0x4, 0x64, 0x61, 0x74, 0x61, 0x12, 0x12, 0xa, 0x10, 0x8, 0x1, 0x12, 0xc, 0xa, 0x2, 0x8, 0x3, 0xa, 0x2, 0x8, 0x2, 0xa, 0x2, 0x8, 0x2, 0x62, 0x19, 0xa, 0x7, 0x72, 0x65, 0x64, 0x75, 0x63, 0x65, 0x64, 0x12, 0xe, 0xa, 0xc, 0x8, 0x1, 0x12, 0x8, 0xa, 0x2, 0x8, 0x3, 0xa, 0x2, 0x8, 0x2, 0x42, 0x2, 0x10, 0x9},

		/*

		   &ir.NodeProto{
		     Input:     []string{"data"},
		     Output:    []string{"reduced"},
		     Name:      "",
		     OpType:    "ReduceMax",
		     Attributes: ([]*ir.AttributeProto) (len=2 cap=2) {
		    (*ir.AttributeProto)(0xc000119300)(name:"axes" type:INTS ints:1 ),
		    (*ir.AttributeProto)(0xc000119400)(name:"keepdims" type:INT )
		   }
		   ,
		   },


		*/

		Input: []tensor.Tensor{

			tensor.New(
				tensor.WithShape(3, 2, 2),
				tensor.WithBacking([]float32{5, 1, 20, 2, 30, 1, 40, 2, 55, 1, 60, 2}),
			),
		},
		ExpectedOutput: []tensor.Tensor{

			tensor.New(
				tensor.WithShape(3, 2),
				tensor.WithBacking([]float32{20, 2, 40, 2, 60, 2}),
			),
		},
	}
}