How to compute SSE in python

Question

I need to fit a linear model to Wine Quality dataset. And then find MSE for each k-fold. Following is my code

        regressor = LinearRegression()
        regressor.fit(Features_train, Quality_train)
        scores = cross_val_score(regressor, Features, Quality, cv=10,
             scoring='mean_squared_error') 
         print scores

Problem here is that one or two values of MSE are negative. Following is the scores array:

[-0.47093648 -0.40001874 -0.46928925 -0.4317235  -0.37665658 -0.52359841
 -0.40046081 -0.42944953 -0.36179521 -0.48792052]*

According to the formula, it shoud not be negative.

score 1 · Accepted Answer · answered Nov 07 '17 at 14:19

1

Do refer to this thread below: scikit-learn cross validation, negative values with mean squared error

In summary, this is supposed to happen. The actual MSE is simply the positive version of the number you're getting. Hope this helps!

answered Nov 07 '17 at 14:19

Joel Lee

929
8
17

How to compute SSE in python

1 Answers1