R2

Definition

The correctness metric for the Non-human Primate Motor Prediction task.

During evaluation, the coefficient of determination (\(R^2\)) for the \(X\) and \(Y\) velocities are averaged to report the correctness score for each session, where \(n\) is the number of labeled points in the test split of the session, \(y_i\) is the ground-truth velocity, \(\hat{y_i}\) is the predicted velocity, and \(\bar{y}\) is the mean of the ground-truth velocities. The \(R^2\) from sessions for each NHP are averaged, producing two final correctness scores.

\[R^2 = 1 - \frac{\sum_{i=1}^{n} (y_i - \hat{y_i})^2}{\sum_{i=1}^{n} (y_i - \bar{y})^2}\]

Implementation Notes

This is a two-dimensional implementation of R2.