parameter.SigmoidBoundedParameter

parameter.SigmoidBoundedParameter(value, low, high)

Parameter with sigmoid-based bounds enforcement.

Uses sigmoid transformation to map from unconstrained real space to bounded interval [low, high]. This provides smooth gradients and avoids the gradient issues of hard clipping.

Parameters

Name	Type	Description	Default
value	Union[float, int, jnp.ndarray]	The initial parameter value (will be clipped to bounds).	required
low	float	Lower bound for the parameter value.	required
high	float	Upper bound for the parameter value.	required

Examples

>>> param = SigmoidBoundedParameter(0.7, low=0.0, high=1.0)
>>> param.__jax_array__()  # Returns value in [0, 1]
>>> # Gradients flow smoothly through sigmoid transformation

Methods

Name	Description
tree_flatten	Flatten for JAX pytree registration.
tree_unflatten	Unflatten for JAX pytree registration.

tree_flatten

parameter.SigmoidBoundedParameter.tree_flatten()

Flatten for JAX pytree registration.

tree_unflatten

parameter.SigmoidBoundedParameter.tree_unflatten(aux_data, children)

Unflatten for JAX pytree registration.

# parameter.SigmoidBoundedParameter { #tvboptim.parameter.SigmoidBoundedParameter }

```python
parameter.SigmoidBoundedParameter(value, low, high)
```

Parameter with sigmoid-based bounds enforcement.

Uses sigmoid transformation to map from unconstrained real space to
bounded interval [low, high]. This provides smooth gradients and
avoids the gradient issues of hard clipping.

## Parameters {.doc-section .doc-section-parameters}

| Name   | Type                                                                                                         | Description                                              | Default    |
|--------|--------------------------------------------------------------------------------------------------------------|----------------------------------------------------------|------------|
| value  | [Union](`typing.Union`)\[[float](`float`), [int](`int`), [jnp](`jax.numpy`).[ndarray](`jax.numpy.ndarray`)\] | The initial parameter value (will be clipped to bounds). | _required_ |
| low    | [float](`float`)                                                                                             | Lower bound for the parameter value.                     | _required_ |
| high   | [float](`float`)                                                                                             | Upper bound for the parameter value.                     | _required_ |

## Examples {.doc-section .doc-section-examples}

```python
>>> param = SigmoidBoundedParameter(0.7, low=0.0, high=1.0)
>>> param.__jax_array__()  # Returns value in [0, 1]
>>> # Gradients flow smoothly through sigmoid transformation
```

## Methods

| Name | Description |
| --- | --- |
| [tree_flatten](#tvboptim.parameter.SigmoidBoundedParameter.tree_flatten) | Flatten for JAX pytree registration. |
| [tree_unflatten](#tvboptim.parameter.SigmoidBoundedParameter.tree_unflatten) | Unflatten for JAX pytree registration. |

### tree_flatten { #tvboptim.parameter.SigmoidBoundedParameter.tree_flatten }

```python
parameter.SigmoidBoundedParameter.tree_flatten()
```

Flatten for JAX pytree registration.

### tree_unflatten { #tvboptim.parameter.SigmoidBoundedParameter.tree_unflatten }

```python
parameter.SigmoidBoundedParameter.tree_unflatten(aux_data, children)
```

Unflatten for JAX pytree registration.