Hi All
I'm testing on spark2.0 Random forest example code. Here is the part
about accuracy:
MulticlassClassificationEvaluator evaluator = new
MulticlassClassificationEvaluator()
.setLabelCol("indexedLabel")
.setPredictionCol("prediction")
.setMetricName("accuracy");
double accuracy = evaluator.evaluate(predictions);
I'm reading the spark doc, find out that accuracy =
ACC=TP / TP + FP
https://spark.apache.org/docs/latest/mllib-evaluation-metrics.html
My question is how can I can the TN , FN results ? I look up the api ,
didn't find anything useful.
Thanks
Best Regards
