lerp

→Tensor

lerp(input: Tensor)

Linearly interpolate between input and an endpoint with a weight.

Computes the affine combination

\text{out} \;=\; x + w \cdot (\text{end} - x) \;=\; (1 - w) \cdot x + w \cdot \text{end} ,

where $w$ is the interpolation weight. weight may be a scalar or a tensor broadcastable to input.

inputTensor

Start point.

Tensor

Interpolated tensor of the broadcast shape.

With w = 0 the result equals input; with w = 1 it equals end; out-of-range weights extrapolate beyond the endpoints.

>>> import lucid
>>> a = lucid.tensor([0.0, 0.0, 0.0])
>>> b = lucid.tensor([1.0, 2.0, 4.0])
>>> lucid.lerp(a, b, 0.5)
Tensor([0.5, 1. , 2. ])