Hyperparameters in Machine Learning

Model hyperparameters are variables which should be selected before training the given machine learning model. Generally, they are determined manually. Different combinations of hyperparameters for same algorithm and training data lead to different models. They cannot be learned from data.
Examples of hyperparameters for different algorithms:

  1. Random Forests:

    • Number of decision trees
    • Maximum depth of decision trees
  2. K-means Clustering:

    • Number of centroids
  3. Linear Regression:

    • Regularizer

You can use the Grid Search module from scikit-learn as well to find the optimum hyperparameters for the model. Here is the link of the documentation for reference : https://scikit-learn.org/stable/modules/grid_search.html