Spark ML: Evaluating the model

- [Instructor] So, continuing on,…we got a result of only a little over 29% accuracy…and that's just really not good enough for…us to use this as a model.…So, do we have some capability that we can use to evaluate…other types of solutions to this problem?…And we do!…So, we're going to use this method…called Regression Metrics.…We can get more insight into our model performance.…So, just of note,…regression metrics requires input formatted…as tuples of doubles,…where the first item is the prediction…and the second item is the observation.…

In this case, the count, how many markets?…Once you have mapped these values from Hold Out,…you can directly pass them…to the Regression Metrics Constructor.…Now, the idea with this…is that we're going to have this holding area,…so that we can see what it is…we want to get out of this model.…And then, we're going to apply some new methods against it.…So, the key line of this is mapped equals holdout select…from our prediction, the count…from the rdd-map into a lambda…and then pass the values.…