# helia_edge.layers.squeeze_excite

## helia_edge.layers.squeeze_excite.se_layer

`function` · `python`

```python
se_layer(
    ratio: int = 8,
    name: str | None = None,
    squeeze_activation: str | Callable = 'relu6',
    excite_activation: str | Callable = 'hard_sigmoid',
) -> keras.Layer
```

Squeeze & excite functional layer

Implements Squeeze and Excite block as in
[Squeeze-and-Excitation Networks](https://arxiv.org/pdf/1709.01507.pdf).

**Parameters**

| Name | Type | Default | Description |
| --- | --- | --- | --- |
| ratio | Expansion ratio | 8 | Expansion ratio. Defaults to 8. |
| name | str \| None | None | Block name. Defaults to None. |
| squeeze_activation | str \| Callable | 'relu6' | Squeeze activation. Defaults to "relu6". |
| excite_activation | str \| Callable | 'hard_sigmoid' | Excite activation. Defaults to "hard_sigmoid". |

**Returns**

| Name | Type | Description |
| --- | --- | --- |
|  | keras.Layer | keras.Layer: Functional SE layer |

Source: `helia_edge/layers/squeeze_excite.py:6`