# `Monty.Attention.Decay`
[🔗](https://github.com/d-led/ex.tbp.monty/blob/v0.1.0/packages/ex_tbp_monty_core/lib/monty/attention/decay.ex#L1)

How a weight that was not refreshed this step loses its say: Monty's
`attention/decay.py`.

Attention has to forget, or everything ever looked at stays as interesting as
the thing in front of the sensor. A decay is what forgets, and this module
holds the one thing the file holds besides its classes: the rate a config means
when it says `linear` and names none.

The decays themselves are `Monty.Attention.Decay.NoopDecay` and
`Monty.Attention.Decay.LinearWeightDecay`, and what they have in common is
`Monty.Attention.Decay.VoxelGridWeightDecay`.

# `default_linear_weight_decay_rate`

```elixir
@spec default_linear_weight_decay_rate() :: float()
```

Default per-step decay toward zero for linear decay.

---

*Consult [api-reference.md](api-reference.md) for complete listing*