RegressionConformalInterval<TOutput>
Models & Types · AiDotNet.Models.Results
Represents a conformal prediction interval for regression-style outputs.
For Beginners
Instead of a single number, you get a range where the true value is likely to fall.
How It Works
The interval is typically computed as [prediction - q, prediction + q] where q is a quantile of calibration residuals.
Constructors
| Constructor | Summary |
|---|---|
RegressionConformalInterval(TOutput lower, TOutput upper) |
Initializes a new instance of the RegressionConformalInterval class. |
Properties
| Property | Summary |
|---|---|
TOutput Lower { get; } |
Gets the lower bound of the interval. |
TOutput Upper { get; } |
Gets the upper bound of the interval. |