IsNaN#

IsNaN - 13 #

Version

name: IsNaN (GitHub)
domain: main
since_version: 13
function: False
support_level: SupportType.COMMON
shape inference: True

This version of the operator has been available since version 13.

Summary

Returns which elements of the input are NaN.

Inputs

X (heterogeneous) - T1: input

Outputs

Y (heterogeneous) - T2: output

Type Constraints

T1 in ( tensor(bfloat16), tensor(double), tensor(float), tensor(float16) ): Constrain input types to float tensors.
T2 in ( tensor(bool) ): Constrain output types to boolean tensors.

Examples

default

node = onnx.helper.make_node(
    "IsNaN",
    inputs=["x"],
    outputs=["y"],
)

x = np.array([3.0, np.nan, 4.0, np.nan], dtype=np.float32)
y = np.isnan(x)
expect(node, inputs=[x], outputs=[y], name="test_isnan")

Differences

`0`	`0`	`Returns which elements of the input are NaN.`	`Returns which elements of the input are NaN.`
`1`	`1`
`2`	`2`	`Inputs`	`Inputs`
`3`	`3`
`4`	`4`	`* X (heterogeneous) - T1:`	`* X (heterogeneous) - T1:`
`5`	`5`	`input`	`input`
`6`	`6`
`7`	`7`	`Outputs`	`Outputs`
`8`	`8`
`9`	`9`	`* Y (heterogeneous) - T2:`	`* Y (heterogeneous) - T2:`
`10`	`10`	`output`	`output`
`11`	`11`
`12`	`12`	`Type Constraints`	`Type Constraints`
`13`	`13`
`14`	`14`	`* T1 in (`	`* T1 in (`
	`15`		`tensor(bfloat16),`
`15`	`16`	`tensor(double),`	`tensor(double),`
`16`	`17`	`tensor(float),`	`tensor(float),`
`17`	`18`	`tensor(float16)`	`tensor(float16)`
`18`	`19`	`):`	`):`
`19`	`20`	`Constrain input types to float tensors.`	`Constrain input types to float tensors.`
`20`	`21`	`* T2 in (`	`* T2 in (`
`21`	`22`	`tensor(bool)`	`tensor(bool)`
`22`	`23`	`):`	`):`
`23`	`24`	`Constrain output types to boolean tensors.`	`Constrain output types to boolean tensors.`

IsNaN - 9 #