Quick Introduction to Underfitting & Overfitting
Nov 12, 2020
In real world data sets Overfitting or Underfitting are the biggest problems that are faced by Machine Learning Engineers. Before going any further, first let’s clear out the concept of underfitting and overfitting.
What is underfitting?
When your model is neither performing well on training data nor testing data than your model is likely to underfit. Why does it happen? Because your model is not complex enough to learn the relationship between the features (X) of the data set and the target variable (Y).
What is overfitting?
When you are having high accuracy on training data but the accuracy is very low on testing data than the model is said to be overfitting. Main reason of overfitting is the class imbalance.
