Modern Scala Projects

A recap of how we looped through a ranger of Epsilons, the best error term, and the best F1 measure

We implemented a loop just prior to arriving here. Here is a breakdown of those steps in pseudocode:

for (errorTerm <- maxMinPair._2 to maxMinPair._1 by stepsize) {//Step 1: We broadcast the error term (epsilon) into Spark//Step 2: We generate predictions //Step 3: We will crossjoin the final predictions dataframe with our initial Testing Dataframe//Step 4: We calculate False Negatives, True Negatives, False Negatives and True Positives//Step 5: Calculate Precision and Recall//Step 6: Calculate F1Step 7: Return Best Error Term and Best F1 Measure}

In the preceding Step 3, we derived thelabelsAndPredictions dataframe. In Step 4, we set out to ...