(l-onnx-doc-Upsample)=

# Upsample


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

## Upsample - 10

### Version

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

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

### Summary

Upsample the input tensor.
Each dimension value of the output tensor is:
  output_dimension = floor(input_dimension * scale).

### Attributes

* **mode - STRING** (default is `'nearest'`):

  Two interpolation modes: nearest (default), and linear (including bilinear, trilinear, etc)

### Inputs

- **X** (heterogeneous) - **T**:

  N-D tensor
- **scales** (heterogeneous) - **tensor(float)**:

  The scale array along each dimension. It takes value greater than or equal to 1. The number of elements of 'scales' should be the same as the rank of input 'X'.

### Outputs

- **Y** (heterogeneous) - **T**:

  N-D tensor after resizing

### Type Constraints

* **T** in ( `tensor(bool)`, `tensor(complex128)`, `tensor(complex64)`, `tensor(double)`, `tensor(float)`, `tensor(float16)`, `tensor(int16)`, `tensor(int32)`, `tensor(int64)`, `tensor(int8)`, `tensor(string)`, `tensor(uint16)`, `tensor(uint32)`, `tensor(uint64)`, `tensor(uint8)` ):

  Constrain input 'X' and output 'Y' to all tensor types.

```{toctree}
text_diff_Upsample_9_10
```

(l-onnx-op-upsample-9)=

## Upsample - 9

### Version

- **name**: [Upsample (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#Upsample)
- **domain**: `main`
- **since_version**: `9`
- **function**: `False`
- **support_level**: `SupportType.COMMON`
- **shape inference**: `True`

This version of the operator has been available
**since version 9**.

### Summary

Upsample the input tensor.
Each dimension value of the output tensor is:
  output_dimension = floor(input_dimension * scale).

### Attributes

* **mode - STRING** (default is `'nearest'`):

  Two interpolation modes: nearest (default), and linear (including bilinear, trilinear, etc)

### Inputs

- **X** (heterogeneous) - **T**:

  N-D tensor
- **scales** (heterogeneous) - **tensor(float)**:

  The scale array along each dimension. It takes value greater than or equal to 1. The number of elements of 'scales' should be the same as the rank of input 'X'.

### Outputs

- **Y** (heterogeneous) - **T**:

  N-D tensor after resizing

### Type Constraints

* **T** in ( `tensor(bool)`, `tensor(complex128)`, `tensor(complex64)`, `tensor(double)`, `tensor(float)`, `tensor(float16)`, `tensor(int16)`, `tensor(int32)`, `tensor(int64)`, `tensor(int8)`, `tensor(string)`, `tensor(uint16)`, `tensor(uint32)`, `tensor(uint64)`, `tensor(uint8)` ):

  Constrain input 'X' and output 'Y' to all tensor types.

```{toctree}
text_diff_Upsample_7_10
text_diff_Upsample_7_9
```

(l-onnx-op-upsample-7)=

## Upsample - 7

### Version

- **name**: [Upsample (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#Upsample)
- **domain**: `main`
- **since_version**: `7`
- **function**: `False`
- **support_level**: `SupportType.COMMON`
- **shape inference**: `True`

This version of the operator has been available
**since version 7**.

### Summary

Upsample the input tensor.
Each dimension value of the output tensor is:
  output_dimension = floor(input_dimension * scale).

### Attributes

* **mode - STRING** (default is `'nearest'`):

  Two interpolation modes: nearest (default), and linear (including bilinear, trilinear, etc)

* **scales - FLOATS** (required) :

  The scale array along each dimension. It takes value greater than or equal to 1. The number of elements of 'scales' should be the same as the rank of input 'X'.

### Inputs

- **X** (heterogeneous) - **T**:

  N-D tensor

### Outputs

- **Y** (heterogeneous) - **T**:

  N-D tensor after resizing

### Type Constraints

* **T** in ( `tensor(bool)`, `tensor(complex128)`, `tensor(complex64)`, `tensor(double)`, `tensor(float)`, `tensor(float16)`, `tensor(int16)`, `tensor(int32)`, `tensor(int64)`, `tensor(int8)`, `tensor(string)`, `tensor(uint16)`, `tensor(uint32)`, `tensor(uint64)`, `tensor(uint8)` ):

  Constrain input and output types to all tensor types.

```{toctree}
text_diff_Upsample_1_10
text_diff_Upsample_1_9
text_diff_Upsample_1_7
```

(l-onnx-op-upsample-1)=

## Upsample - 1

### Version

- **name**: [Upsample (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#Upsample)
- **domain**: `main`
- **since_version**: `1`
- **function**: `False`
- **support_level**: `SupportType.EXPERIMENTAL`
- **shape inference**: `False`

No versioning maintained for experimental ops.

### Summary

Upsample the input tensor.
The width and height of the output tensor are:
  output_width = floor(input_width * width_scale),
  output_height = floor(input_height * height_scale).
Example:
  Given `data` tensor, width_scale, height_scale, mode,
  Upsample the input 4-D tensor in nearest mode:
  data = [[[
      [1, 2],
      [3, 4]
  ]]]
  width_scale = 2
  height_scale = 2
  mode = "nearest"
  output = [[[
      [1, 1, 2, 2],
      [1, 1, 2, 2],
      [3, 3, 4, 4],
      [3, 3, 4, 4]
  ]]]

### Attributes

* **height_scale - FLOAT** (required) :

  The scale along height dimension. It takes value greater than or equal to 1.

* **mode - STRING** (default is `'nearest'`):

  Two interpolation modes: nearest(default), bilinear

* **width_scale - FLOAT** (required) :

  The scale along width dimension. It takes value greater than or equal to 1.

### Inputs

- **X** (heterogeneous) - **T**:

  4-D tensor, [N,C,H,W]

### Outputs

- **Y** (heterogeneous) - **T**:

  4-D tensor after resizing, [N,C,H,W]

### Type Constraints

* **T** in ( `tensor(bool)`, `tensor(double)`, `tensor(float)`, `tensor(float16)`, `tensor(int32)`, `tensor(int64)` ):

  Constrain output types to bool, int32, int64, float16, float, double tensors.