Database Reference
In-Depth Information
Mean Squared Error and Root Mean Squared
Error
MSE is the average of the squared error that is used as the loss function for least squares re-
gression:
It is the sum, over all the data points, of the square of the difference between the predicted
and actual target variables, divided by the number of data points.
RMSE is the square root of MSE. MSE is measured in units that are the square of the target
variable, while RMSE is measured in the same units as the target variable. Due to its for-
mulation, MSE, just like the squared loss function that it derives from, effectively penalizes
larger errors more severely.
In order to evaluate our predictions based on the mean of an error metric, we will first
make predictions for each input feature vector in an RDD of
LabeledPoint
instances by
computing the error for each record using a function that takes the prediction and true tar-
get value as inputs. This will return a
[Double]
RDD that contains the error values. We
can then find the average using the
mean
method of RDDs that contain
Double
values.
Let's define our squared error function as follows:
def squared_error(actual, pred):
return (pred - actual)**2
