Learn various concepts that will be useful in creating functional machine learning models like model selection and tuning, model performance measures, ways of regularisation, etc.

Feature engineering

Feature engineering is transforming data from the raw state to a state where it becomes suitable for modelling. It converts the data columns into features that are better at representing a given situation in terms of clarity. Quality of the component in distinctly representing an entity impacts the model’s quality in predicting its behaviour. In this module, you will learn several steps involved in Feature Engineering.

Model selection and tuning

This module will teach you which model best suits architecture by evaluating every individual model based on the requirements.

Model performance measures

In this module, you will learn how to optimise your machine learning model’s performance using model evaluation metrics.

Regularising Linear models

In this module, you will learn the technique to avoid overfitting and increase model interpretability.

ML pipeline

This module will teach you how to automate machine learning workflows using the ML Pipeline. You can operate the ML Pipeline by enabling a series of data to be altered and linked together in a model, which can be tested and evaluated to achieve either a positive or negative result.

Bootstrap sampling

Bootstrap Sampling is a machine learning technique to estimate statistics on population by examining a dataset with replacement.

Grid search CV

Grid search CV is the process of performing hyperparameter tuning to determine the optimal values for any machine learning model. The performance of a model significantly depends on the importance of hyperparameters. Doing this process manually is a tedious task. Hence, we use GridSearchCV to automate the tuning of hyperparameters.

Randomized search CV

Randomized search CV is used to automate the tuning of hyperparameters similar to Grid search CV. Randomized search CV is provided for a random search, and Grid search CV is provided for a grid search.

K fold cross-validation

K-fold cross-validation is a way in ML to improve the holdout method. This method guarantees that our model’s score does not depend on how we picked the train and test set. The data set is divided into k number of subsets, and the holdout method is repeated k number of times.