Non-Parametric vs Parametric Models –
A model being non-parametric does not mean it do not have parameters. In fact they tend to have a lot of them. They are called non-parametric because the number of parameters is not determined prior to training, so the model structure is free to stick closely to the data. One example of this is Decision Tree which is why they tends to overfit if not regularized.
In contrast, the parametric models such as linear regression has a predetermined numbers of parameters so it’s degree of freedom is limited, reducing the risk of overfitting but increasing the risk of underfitting.