FMeasure Example: Output All Classes | Teradata Vantage - FMeasure Example: Output All Classes - Teradata Vantage

Input

The input table, computers_category, has five attributes of personal computers—price, speed, hard disk size, RAM, and screen size. The table has 500 rows, categorized into five price groups—SPECIAL, SUPER, HYPER, MEGA and UBER. The predicted_compcategory values can be output by a classification function, such as KNN (ML Engine).

SQL Call

SELECT * FROM FMeasure(
ON computers_category PARTITION BY 1
USING
ObservationColumn ('expected_compcategory')
PredictColumn ('predicted_compcategory')
Beta (1.0)
) AS dt;

Output

 class   precision          recall             beta fmeasure           
 ------- ------------------ ------------------ ---- ------------------ 
 hyper   0.9368421052631579               0.89  1.0 0.9128205128205129
 mega    0.9230769230769231  0.935064935064935  1.0 0.9290322580645162
 special            0.84375 0.8852459016393442  1.0              0.864
 super   0.9358974358974359  0.954248366013072  1.0 0.9449838187702266
 uber     0.896551724137931             0.8125  1.0 0.8524590163934426
 -AVG-                0.918              0.918  1.0              0.918

Download a zip file of all examples and a SQL script file that creates their input tables.