(l-onnx-doc-IsInf)=

# IsInf


(l-onnx-op-isinf-20)=

## IsInf - 20

### Version

- **name**: [IsInf (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#IsInf)
- **domain**: `main`
- **since_version**: `20`
- **function**: `False`
- **support_level**: `SupportType.COMMON`
- **shape inference**: `True`

This version of the operator has been available
**since version 20**.

### Summary

Map infinity to true and other values to false.

### Attributes

* **detect_negative - INT** (default is `1`):

  (Optional) Whether map negative infinity to true. Default to 1 so that negative infinity induces true. Set this attribute to 0 if negative infinity should be mapped to false.

* **detect_positive - INT** (default is `1`):

  (Optional) Whether map positive infinity to true. Default to 1 so that positive infinity induces true. Set this attribute to 0 if positive infinity should be mapped to false.

### Inputs

- **X** (heterogeneous) - **T1**:

  input

### Outputs

- **Y** (heterogeneous) - **T2**:

  output

### Type Constraints

* **T1** in ( `tensor(bfloat16)`, `tensor(double)`, `tensor(float)`, `tensor(float16)`, `tensor(float8e4m3fn)`, `tensor(float8e4m3fnuz)`, `tensor(float8e5m2)`, `tensor(float8e5m2fnuz)` ):

  Constrain input types to float tensors.
* **T2** in ( `tensor(bool)` ):

  Constrain output types to boolean tensors.

### Examples

#### _infinity

```python
import numpy as np
import onnx

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

x = np.array([-1.2, np.nan, np.inf, 2.8, -np.inf, np.inf], dtype=np.float32)
y = np.isinf(x)
expect(node, inputs=[x], outputs=[y], name="test_isinf")
```

#### _positive_infinity_only

```python
import numpy as np
import onnx

node = onnx.helper.make_node(
    "IsInf", inputs=["x"], outputs=["y"], detect_negative=0
)

x = np.array([-1.7, np.nan, np.inf, 3.6, -np.inf, np.inf], dtype=np.float32)
y = np.isposinf(x)
expect(node, inputs=[x], outputs=[y], name="test_isinf_positive")
```

#### _negative_infinity_only

```python
import numpy as np
import onnx

node = onnx.helper.make_node(
    "IsInf", inputs=["x"], outputs=["y"], detect_positive=0
)

x = np.array([-1.7, np.nan, np.inf, -3.6, -np.inf, np.inf], dtype=np.float32)
y = np.isneginf(x)
expect(node, inputs=[x], outputs=[y], name="test_isinf_negative")
```

#### _infinity_float16

```python
import numpy as np
import onnx

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

x = np.array([-1.2, np.nan, np.inf, 2.8, -np.inf, np.inf], dtype=np.float16)
y = np.isinf(x)
expect(node, inputs=[x], outputs=[y], name="test_isinf_float16")
```

```{toctree}
text_diff_IsInf_10_20
```

(l-onnx-op-isinf-10)=

## IsInf - 10

### Version

- **name**: [IsInf (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#IsInf)
- **domain**: `main`
- **since_version**: `10`
- **function**: `False`
- **support_level**: `SupportType.COMMON`
- **shape inference**: `True`

This version of the operator has been available
**since version 10**.

### Summary

Map infinity to true and other values to false.

### Attributes

* **detect_negative - INT** (default is `1`):

  (Optional) Whether map negative infinity to true. Default to 1 so that negative infinity induces true. Set this attribute to 0 if negative infinity should be mapped to false.

* **detect_positive - INT** (default is `1`):

  (Optional) Whether map positive infinity to true. Default to 1 so that positive infinity induces true. Set this attribute to 0 if positive infinity should be mapped to false.

### Inputs

- **X** (heterogeneous) - **T1**:

  input

### Outputs

- **Y** (heterogeneous) - **T2**:

  output

### Type Constraints

* **T1** in ( `tensor(double)`, `tensor(float)` ):

  Constrain input types to float tensors.
* **T2** in ( `tensor(bool)` ):

  Constrain output types to boolean tensors.