## models.lift()


Lift table: rows sorted by predicted rate, cut into bands of about


Usage


``` python
models.lift(
    y,
    pred,
    exposure=None,
    bands=10,
)
```


equal exposure.


## Parameters


`y: list of float`  

`pred: list of float`  

`exposure: list of float = None`  

`bands: int = ``10`  


## Returns


`list of dict`  
[exposure](reserving.ClarkFit.md#prospicio.reserving.ClarkFit.exposure), `expected` and `actual` per band.