Reviewing which metrics are mentioned in the README

[tennlee]

The purpose of the table of included metrics in the README is to be a short but interesting summary of some of the metrics contained in the package to give a quick sense of what's included. The index page of included metrics provides a complete listing. The table in the README has become overly long, as each new PR tends to add to it, without removing anything.

The README should be shortened, including some of the most interesting and a few more common metrics, and readers should be directed to the index listing for a complete listing.

[nicholasloveday]

One option could be to consolidate

Probability of Detection (POD), False Alarm Ratio (FAR), Probability of False Detection (POFD), Success Ratio, Accuracy, Peirce's Skill Score, Critical Success Index (CSI), Gilbert Skill Score, Heidke Skill Score, Odds Ratio, Odds Ratio Skill Score, F1 score, Symmetric Extremal Dependence Index, FIxed Risk Multicategorical (FIRM) Score.

Into

Binary contingency table (confusion matrix) metrics and the FIxed Risk Multicategorical (FIRM) Score.

[nicholasloveday]

And shorten

Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Additive Bias, Multiplicative Bias, Pearson's Correlation Coefficient, Flip-Flop Index, Quantile Loss, Murphy Score, families of consistent scoring functions for quantiles and expectiles, Threshold Weighted Squared Error, Threshold Weighted Quantile Score, Threshold Weighted Absolute Error, Threshold Weighted Expectile Score, Threshold Weighted Huber Loss.

to

MAE, MSE, RMSE, Additive Bias, Multiplicative Bias, Pearson's Correlation Coefficient, Flip-Flop Index, Quantile Loss, Murphy Score, Threshold Weighted Squared Error, Threshold Weighted Quantile Score, Threshold Weighted Absolute Error, Threshold Weighted Expectile Score, Threshold Weighted Huber Loss.

or

MAE, MSE, RMSE, Additive Bias, Multiplicative Bias, Pearson's Correlation Coefficient, Flip-Flop Index, Quantile Loss, Murphy Score, and threshold weighted scores for expectiles, quantiles and Huber Loss

[Steph-Chong]

And shorten

Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Additive Bias, Multiplicative Bias, Pearson's Correlation Coefficient, Flip-Flop Index, Quantile Loss, Murphy Score, families of consistent scoring functions for quantiles and expectiles, Threshold Weighted Squared Error, Threshold Weighted Quantile Score, Threshold Weighted Absolute Error, Threshold Weighted Expectile Score, Threshold Weighted Huber Loss.

to

MAE, MSE, RMSE, Additive Bias, Multiplicative Bias, Pearson's Correlation Coefficient, Flip-Flop Index, Quantile Loss, Murphy Score, Threshold Weighted Squared Error, Threshold Weighted Quantile Score, Threshold Weighted Absolute Error, Threshold Weighted Expectile Score, Threshold Weighted Huber Loss.

or

MAE, MSE, RMSE, Additive Bias, Multiplicative Bias, Pearson's Correlation Coefficient, Flip-Flop Index, Quantile Loss, Murphy Score, and threshold weighted scores for expectiles, quantiles and Huber Loss

I like @nicholasloveday's suggestions for both the categorical and continuous categories. In terms of the continuous ones, I probably have a preference for the second suggestion (because it is the shortest). I know I pushed for including the full terms for MAE, MSE and RMSE in the first place, but given the length of the table now, I agree using the acronyms for these three makes sense.

One thought - at the moment percent bias is the only continuous metric not covered. As such, should percent bias be added to the continuous section?

[tennlee]

I like the final option. I don't think we need to include percent bias at this stage. I'll put up a PR for the final option unless someone else gets to it first and we can look at it rendered on 228. I'll try to get to it this afternoon.

[tennlee]

With percent bias, I'm thinking that it may seem different when NSE, KGE and PIT are all available, and we can revisit the readme list then.