This accounts for about 60% of the Machine Learning Methods we have.

By definition a parametric model is one that has fixed parameters to learn, i.e. weights in Linear Regression: $w_0, w_1, ..., w_n$. Conversely, a non-parametric model does not have a fixed number of parameters to learn: K-means clustering for example just clusters the data as best as it can.

We can list some more models:

  1. Linear Regression
  2. Ridge Regression
  3. Lasso Regression
  4. Logistic Regression
  5. Neural Networks
  6. Perceptron
  7. Naive Bayes