Bias vs Variance

Summary

When we train a model, we split data into a testing set and a training set. We train our model against the training set and then evaluate the predictions of the model against the actual values in the testing set.

Bias

Bias relates to a model's difference between the true relationship between the data and the predicted value in the training set.

Variance

Variance relates to the difference between a model's prediction to the actual value.

Overfitting

Overfitting is a problem where a model adapts too much to the training set of the data and lacks the ability to predict the testing data.