resolve_conj

→Tensor

resolve_conj(input: Tensor)

source edit

Materialise any pending lazy conjugation — no-op in Lucid.

Reference frameworks that maintain a lazy "is_conj" flag on complex tensors use resolve_conj to force the flag to be applied to the underlying buffer. Lucid has no such flag (see conj_physical), so this routine simply returns the input unchanged. It is provided for API parity so that ported code keeps working without behavioural drift.

Parameters

inputTensor

Input tensor.

Returns

Tensor

The input input returned unchanged (no copy, no autograd op inserted).

Notes

Equivalent to the identity map:

\text{resolve\_conj}(x) = x.

See conj_physical for an explicit, always-materialised conjugation.

Examples

>>> import lucid
>>> z = lucid.tensor([1+2j, 3-4j])
>>> lucid.resolve_conj(z) is z
True