Applied Machine Learning With Python

Compare Python and R as data science languages, with SQL and Scala for data engineering. Learn the ML workflow from business understanding to deployment, including data understanding and preparation.

Outline the common approach to machine learning model building for supervised and unsupervised tasks, separating predictors (x) and target (y) and using train-validation-test splits.

Apply linear regression in Excel to build a beginner-friendly machine learning model, understand regression concepts, evaluation metrics like R-squared and adjusted R-squared, and check assumptions such as multicollinearity and homoscedasticity.

Learn to build and evaluate linear regression models in MS Excel, from data preparation, correlation analysis, and multicollinearity checks to fitting the model, interpreting r-squared and p-values, and validating residuals.

Explore the core principles of logistic regression for binary classification, including probability, odds, logit, and the sigmoid function, with hands-on practice and evaluation via confusion matrices.

Explore the limitations of machine learning in Excel, where only linear regression is native; logistic regression is possible with formulas and solver, but large datasets benefit from Python or R.

Explore practical machine learning with Python through hands-on tutorials in Google Colab, covering linear and logistic regression, KNN, SVM, and tree-based methods like random forest and gradient boosting.

Learn linear regression with Python, including fitting a regression line, understanding model coefficients and intercept, and evaluating performance with root mean square error on train and test data.

Learn logistic regression for classification in Python using the Titanic survival dataset in Google Colab. Explore model fitting, coefficients and intercept, and assess accuracy while noting overfitting and confusion matrix.

Explore k-nearest neighbors in python by implementing a custom euclidean distance classifier and validating it against scikit-learn on the iris dataset in Google Colab, optimizing the k value.

Explore support vector machines in Python with hands-on Google Colab tutorials, covering SVM fundamentals, kernels (linear and RBF), key parameters (C, gamma), model training, and tuning for improved accuracy.

Explore tree-based algorithms, including decision trees, random forests, and boosting, to build accurate, interpretable models for classification and regression while managing overfitting through pruning.

Explore tree-based algorithms and ensemble methods, including how splits use chi-square, gini index, and information gain, then build a titanic survival classifier in colab and discuss overfitting, bagging, and boosting.

Master the random forest’s bagging of trees, tune random state and OOB score to curb overfitting, and practice in Google Colab with a survival dataset using accuracy and roc auc.

Understand gbm and gradient boosting in python with google colab, building a weak learner into a classifier. Learn data preparation, model fitting, and evaluating accuracy on train and test sets.

Explore key evaluation metrics for machine learning models, including confusion matrix metrics, precision, recall, F1, ROC AUC, log loss, RMSE, R square, adjusted R square, and lift/gain charts.