API REFERENCE/omnibias-keras

omnibias.keras.blocks.linear

0.0.2a1AlphaPython >=3.10Apache-2.0

Keras 3 operator layers and activation paths implemented with keras.ops for TensorFlow, JAX, or Torch.

import omnibias.keras.blocks.linear

Module contract & mathematical conventions

Operator-typed dense layer: cmbDense.

A drop-in for keras.layers.Dense followed by an OperatorBlock on the output features. Mirrors omnibias.torch.blocks.linear.cmbLinear.

Shared formulas do not imply universal floating-point identity. Verify dtype, shape, derivative-order support, and the stated numerical assumptions for the path you use.

Module source ↗Source revision 9a2d73df0573Running reference examples →

CLASS

cmbDense

omnibias.keras.cmbDenseSource ↗

class cmbDense(layers.Layer)

keras.layers.Dense + per-feature OperatorBlock.

Parameters

units : int Number of output features. op : OpName, default "identity" base : str or ActivationSpec, default "sigmoid" use_bias : bool, default True Dense-layer bias (independent of the OMBU bias terms). block_kwargs : dict, optional Extra keyword arguments passed to OperatorBlock.

Parameters

Name	Type & default	Contract
unitspositional-or-keyword	intRequired	int Number of output features.
oppositional-or-keyword	OpNameDefault: 'identity'	OpName, default "identity"
basepositional-or-keyword	str | ActivationSpec[Any]Default: 'sigmoid'	str or ActivationSpec, default "sigmoid"
use_biaspositional-or-keyword	boolDefault: True	bool, default True Dense-layer bias (independent of the OMBU bias terms).
block_kwargspositional-or-keyword	dict[str, Any] | NoneDefault: None	dict, optional Extra keyword arguments passed to OperatorBlock.
kwargsvariadic keyword	AnyRequired	See the operation contract and implementation below.

Executed regression · cmbdense shape and finite

Recorded coverage confirms this test executed the API body. This example belongs to the pinned upstream test suite. It may use fixtures and imports from its full file. Run it from that checkout with its test dependencies installed.

Full file, imports & fixtures ↗

def test_cmbdense_shape_and_finite() -> None:
    x = np.random.default_rng(5).normal(size=(8, 6))
    layer = cmbDense(units=16, op="identity", base="tanh")
    out = _np(layer(_t(x)))
    assert out.shape == (8, 16)
    assert np.all(np.isfinite(out))

python -m pytest "packages/omnibias-keras/tests/test_blocks.py::test_cmbdense_shape_and_finite" -q

Fields & methods

__init__method

__init__(self, units: int, op: OpName='identity', base: str | ActivationSpec[Any]='sigmoid', use_bias: bool=True, block_kwargs: dict[str, Any] | None=None, **kwargs: Any) -> None

See the class contract above and the exact method implementation below.

Parameters

Name	Type & default	Contract
unitspositional-or-keyword	intRequired	See the operation contract and implementation below.
oppositional-or-keyword	OpNameDefault: 'identity'	See the operation contract and implementation below.
basepositional-or-keyword	str | ActivationSpec[Any]Default: 'sigmoid'	See the operation contract and implementation below.
use_biaspositional-or-keyword	boolDefault: True	See the operation contract and implementation below.
block_kwargspositional-or-keyword	dict[str, Any] | NoneDefault: None	See the operation contract and implementation below.
kwargsvariadic keyword	AnyRequired	See the operation contract and implementation below.

Returns: None

Executed regression · cmbdense shape and finite

Recorded coverage confirms this test executed the API body. This example belongs to the pinned upstream test suite. It may use fixtures and imports from its full file. Run it from that checkout with its test dependencies installed.

Full file, imports & fixtures ↗

def test_cmbdense_shape_and_finite() -> None:
    x = np.random.default_rng(5).normal(size=(8, 6))
    layer = cmbDense(units=16, op="identity", base="tanh")
    out = _np(layer(_t(x)))
    assert out.shape == (8, 16)
    assert np.all(np.isfinite(out))

python -m pytest "packages/omnibias-keras/tests/test_blocks.py::test_cmbdense_shape_and_finite" -q

Method implementation

    def __init__(
        self,
        units: int,
        op: OpName = "identity",
        base: str | ActivationSpec[Any] = "sigmoid",
        use_bias: bool = True,
        block_kwargs: dict[str, Any] | None = None,
        **kwargs: Any,
    ) -> None:
        super().__init__(**kwargs)
        self.units = units
        self._op = op
        self._base_name = base if isinstance(base, str) else base.name
        self._use_bias = use_bias
        self._block_kwargs = block_kwargs or {}
        self.dense = layers.Dense(units, use_bias=use_bias)
        self.block = OperatorBlock(
            op=op, base=base, channels=units, **self._block_kwargs
        )

View source ↗

callmethod

call(self, x: Any) -> Any

See the class contract above and the exact method implementation below.

Parameters

Name	Type & default	Contract
xpositional-or-keyword	AnyRequired	See the operation contract and implementation below.

Returns: Any

Executed regression · cmbdense shape and finite

Recorded coverage confirms this test executed the API body. This example belongs to the pinned upstream test suite. It may use fixtures and imports from its full file. Run it from that checkout with its test dependencies installed.

Full file, imports & fixtures ↗

def test_cmbdense_shape_and_finite() -> None:
    x = np.random.default_rng(5).normal(size=(8, 6))
    layer = cmbDense(units=16, op="identity", base="tanh")
    out = _np(layer(_t(x)))
    assert out.shape == (8, 16)
    assert np.all(np.isfinite(out))

python -m pytest "packages/omnibias-keras/tests/test_blocks.py::test_cmbdense_shape_and_finite" -q

Method implementation

    def call(self, x: Any) -> Any:
        return self.block(self.dense(x))

View source ↗

get_configmethod

get_config(self) -> dict[str, Any]

See the class contract above and the exact method implementation below.

Returns: dict[str, Any]

Method implementation

    def get_config(self) -> dict[str, Any]:
        config = super().get_config()
        config.update(
            {
                "units": self.units,
                "op": self._op,
                "base": self._base_name,
                "use_bias": self._use_bias,
                "block_kwargs": self._block_kwargs,
            }
        )
        return config

View source ↗

unitsattribute

units: int = units

Number of output features.

View source ↗

denseattribute

dense = layers.Dense(units, use_bias=use_bias)

Public instance attribute assigned by __init__. The source assignment is dense = layers.Dense(units, use_bias=use_bias); it describes instance state, not a literal constructor default.

View source ↗

blockattribute

block = OperatorBlock(op=op, base=base, channels=units, **self._block_kwargs)

Public instance attribute assigned by __init__. The source assignment is block = OperatorBlock(op=op, base=base, channels=units, **self._block_kwargs); it describes instance state, not a literal constructor default.

View source ↗

Read the implementation

class cmbDense(layers.Layer):
    """``keras.layers.Dense`` + per-feature :class:`OperatorBlock`.

    Parameters
    ----------
    units : int
        Number of output features.
    op : :class:`OpName`, default ``"identity"``
    base : str or :class:`ActivationSpec`, default ``"sigmoid"``
    use_bias : bool, default True
        Dense-layer bias (independent of the OMBU bias terms).
    block_kwargs : dict, optional
        Extra keyword arguments passed to :class:`OperatorBlock`.
    """

    def __init__(
        self,
        units: int,
        op: OpName = "identity",
        base: str | ActivationSpec[Any] = "sigmoid",
        use_bias: bool = True,
        block_kwargs: dict[str, Any] | None = None,
        **kwargs: Any,
    ) -> None:
        super().__init__(**kwargs)
        self.units = units
        self._op = op
        self._base_name = base if isinstance(base, str) else base.name
        self._use_bias = use_bias
        self._block_kwargs = block_kwargs or {}
        self.dense = layers.Dense(units, use_bias=use_bias)
        self.block = OperatorBlock(
            op=op, base=base, channels=units, **self._block_kwargs
        )

    def call(self, x: Any) -> Any:
        return self.block(self.dense(x))

    def get_config(self) -> dict[str, Any]:
        config = super().get_config()
        config.update(
            {
                "units": self.units,
                "op": self._op,
                "base": self._base_name,
                "use_bias": self._use_bias,
                "block_kwargs": self._block_kwargs,
            }
        )
        return config

Related APIsomnibias.keras.OperatorBlock↗

Other public import paths

• omnibias.keras.cmbDense
• omnibias.keras.blocks.cmbDense
• omnibias.keras.blocks.linear.cmbDense

← Package overviewRunnable learning examples →