(l-onnx-doc-ReverseSequence)=

# ReverseSequence


(l-onnx-op-reversesequence-28)=

## ReverseSequence - 28

### Version

- **name**: [ReverseSequence (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#ReverseSequence)
- **domain**: `main`
- **since_version**: `28`
- **function**: `False`
- **support_level**: `SupportType.COMMON`
- **shape inference**: `True`

This version of the operator has been available
**since version 28**.

### Summary

Reverse batch of sequences having different lengths specified by `sequence_lens`.

For each slice i iterating on batch axis, the operator reverses the first sequence_lens[i] elements on time axis,
and copies elements whose index's beyond sequence_lens[i] to the output. So the output slice i contains reversed
sequences on the first sequence_lens[i] elements, then have original values copied for the other elements.

Example 1:
  input = [[0.0, 4.0, 8.0,  12.0],
           [1.0, 5.0, 9.0,  13.0],
           [2.0, 6.0, 10.0, 14.0],
           [3.0, 7.0, 11.0, 15.0]]
  sequence_lens = [4, 3, 2, 1]
  time_axis = 0
  batch_axis = 1

  output = [[3.0, 6.0, 9.0,  12.0],
            [2.0, 5.0, 8.0,  13.0],
            [1.0, 4.0, 10.0, 14.0],
            [0.0, 7.0, 11.0, 15.0]]

Example 2:
  input = [[0.0,  1.0,  2.0,  3.0 ],
           [4.0,  5.0,  6.0,  7.0 ],
           [8.0,  9.0,  10.0, 11.0],
           [12.0, 13.0, 14.0, 15.0]]
  sequence_lens = [1, 2, 3, 4]
  time_axis = 1
  batch_axis = 0

  output = [[0.0,  1.0,  2.0,  3.0 ],
            [5.0,  4.0,  6.0,  7.0 ],
            [10.0, 9.0,  8.0,  11.0],
            [15.0, 14.0, 13.0, 12.0]]

### Attributes

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

  (Optional) Specify which axis is batch axis. Must be one of 1 (default), or 0.

* **time_axis - INT** (default is `0`):

  (Optional) Specify which axis is time axis. Must be one of 0 (default), or 1.

### Inputs

- **input** (heterogeneous) - **T**:

  Tensor of rank r >= 2.
- **sequence_lens** (heterogeneous) - **tensor(int64)**:

  Tensor specifying lengths of the sequences in a batch. It has shape `[batch_size]`.

### Outputs

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

  Tensor with same shape of input.

### Type Constraints

* **T** in ( `tensor(bfloat16)`, `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)` ):

  Input and output types can be of any tensor type.

### Examples

#### _reversesequence_time

```python
import numpy as np
import onnx

node = onnx.helper.make_node(
    "ReverseSequence",
    inputs=["x", "sequence_lens"],
    outputs=["y"],
    time_axis=0,
    batch_axis=1,
)
x = np.array(
    [
        [0.0, 4.0, 8.0, 12.0],
        [1.0, 5.0, 9.0, 13.0],
        [2.0, 6.0, 10.0, 14.0],
        [3.0, 7.0, 11.0, 15.0],
    ],
    dtype=np.float32,
)
sequence_lens = np.array([4, 3, 2, 1], dtype=np.int64)

y = np.array(
    [
        [3.0, 6.0, 9.0, 12.0],
        [2.0, 5.0, 8.0, 13.0],
        [1.0, 4.0, 10.0, 14.0],
        [0.0, 7.0, 11.0, 15.0],
    ],
    dtype=np.float32,
)

expect(
    node,
    inputs=[x, sequence_lens],
    outputs=[y],
    name="test_reversesequence_time",
)
```

#### _reversesequence_batch

```python
import numpy as np
import onnx

node = onnx.helper.make_node(
    "ReverseSequence",
    inputs=["x", "sequence_lens"],
    outputs=["y"],
    time_axis=1,
    batch_axis=0,
)
x = np.array(
    [
        [0.0, 1.0, 2.0, 3.0],
        [4.0, 5.0, 6.0, 7.0],
        [8.0, 9.0, 10.0, 11.0],
        [12.0, 13.0, 14.0, 15.0],
    ],
    dtype=np.float32,
)
sequence_lens = np.array([0, 2, 3, 4], dtype=np.int64)

y = np.array(
    [
        [0.0, 1.0, 2.0, 3.0],
        [5.0, 4.0, 6.0, 7.0],
        [10.0, 9.0, 8.0, 11.0],
        [15.0, 14.0, 13.0, 12.0],
    ],
    dtype=np.float32,
)

expect(
    node,
    inputs=[x, sequence_lens],
    outputs=[y],
    name="test_reversesequence_batch",
)
```

#### _reversesequence_bfloat16

```python
import numpy as np
import onnx

node = onnx.helper.make_node(
    "ReverseSequence",
    inputs=["x", "sequence_lens"],
    outputs=["y"],
    time_axis=1,
    batch_axis=0,
)

x = np.array(
    [[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]], dtype=ml_dtypes.bfloat16
)
sequence_lens = np.array([4, 3], dtype=np.int64)
y = np.array(
    [[4.0, 3.0, 2.0, 1.0], [7.0, 6.0, 5.0, 8.0]], dtype=ml_dtypes.bfloat16
)

expect(
    node,
    inputs=[x, sequence_lens],
    outputs=[y],
    name="test_reversesequence_bfloat16",
)
```

```{toctree}
text_diff_ReverseSequence_10_28
```

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

## ReverseSequence - 10

### Version

- **name**: [ReverseSequence (GitHub)](https://github.com/onnx/onnx/blob/main/docs/Operators.md#ReverseSequence)
- **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

Reverse batch of sequences having different lengths specified by `sequence_lens`.

For each slice i iterating on batch axis, the operator reverses the first sequence_lens[i] elements on time axis,
and copies elements whose index's beyond sequence_lens[i] to the output. So the output slice i contains reversed
sequences on the first sequence_lens[i] elements, then have original values copied for the other elements.

Example 1:
  input = [[0.0, 4.0, 8.0,  12.0],
           [1.0, 5.0, 9.0,  13.0],
           [2.0, 6.0, 10.0, 14.0],
           [3.0, 7.0, 11.0, 15.0]]
  sequence_lens = [4, 3, 2, 1]
  time_axis = 0
  batch_axis = 1

  output = [[3.0, 6.0, 9.0,  12.0],
            [2.0, 5.0, 8.0,  13.0],
            [1.0, 4.0, 10.0, 14.0],
            [0.0, 7.0, 11.0, 15.0]]

Example 2:
  input = [[0.0,  1.0,  2.0,  3.0 ],
           [4.0,  5.0,  6.0,  7.0 ],
           [8.0,  9.0,  10.0, 11.0],
           [12.0, 13.0, 14.0, 15.0]]
  sequence_lens = [1, 2, 3, 4]
  time_axis = 1
  batch_axis = 0

  output = [[0.0,  1.0,  2.0,  3.0 ],
            [5.0,  4.0,  6.0,  7.0 ],
            [10.0, 9.0,  8.0,  11.0],
            [15.0, 14.0, 13.0, 12.0]]

### Attributes

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

  (Optional) Specify which axis is batch axis. Must be one of 1 (default), or 0.

* **time_axis - INT** (default is `0`):

  (Optional) Specify which axis is time axis. Must be one of 0 (default), or 1.

### Inputs

- **input** (heterogeneous) - **T**:

  Tensor of rank r >= 2.
- **sequence_lens** (heterogeneous) - **tensor(int64)**:

  Tensor specifying lengths of the sequences in a batch. It has shape `[batch_size]`.

### Outputs

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

  Tensor with same shape of input.

### 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)` ):

  Input and output types can be of any tensor type.