Accuracy of the model

Course Q&A Deep Learning Specialization Sequence Models
1

When model oerfits the data, then we get less accuracy of the model, then , why we simply remove those observation considering it as a outlier, and after that fitting model we will good accuracy of the model and it doesn’t overfits also.

Is that solution okay ?

2

No. Removing perfectly valid examples just so the accuracy looks better is not a good solution.

1 Like