Metrics 0.2.0 r2 - CyrilB1531/lodestar GitHub Wiki
Lodestar.Metrics 0.2.0. This page is frozen at that release. Read the current documentation for what
mainsays now. A link to a decision or a migration page followsmain, and leaves the archive.
R2
The coefficient of determination: how much better the model is than always predicting the mean, as a unitless score.
Members
| Member | What it does |
|---|---|
R2.Score |
One score for the whole prediction: 1 minus the residual variance over the truth's variance. |
R2.PerOutput |
One score per output, unreduced. |
R2.VarianceWeighted |
One score, each output counted in proportion to how much its own truth varies. |