Machine Learning with Python - Complete Course & Projects

Explore numpy, a python library for arrays and matrices, with functions for linear algebra and Fourier transforms, and learn array creation, dimensions, and indexing.

Explore numpy arrays with one- and two-dimensional slicing, indexing, and iteration, and apply for loops to perform conditions and math operations, paving the way to matrix operations.

Learn how feature scaling normalizes diverse data ranges through normalization and standardization, preparing features like years of experience and salary in a consistent range for reliable models.

Learn to scale data in Python using numpy arrays, min-max scaling with MinMaxScaler, and standard scaling, applying fit and transform, and box plots with ranges 0–1 and −1 to 1.5.

Teach label encoding by converting categorical values to numbers from zero to n-1 using a pandas dataframe. Demonstrate fit_transform encoding gender male, female, and unknown as 1, 0, and 2.

Identify outliers and remove them in Python using pandas and numpy by computing Q1, Q3, and IQR to set upper and lower limits, then apply drop.

Learn to perform a train-test split in Python with sklearn, dividing data into X and Y for train and test sets (0.3 size) to train models and evaluate predictions.

Explore regression metrics MSE and RMSE, quantify how far data points deviate from the regression line, and implement these metrics in Python using numpy.

Understand how a confusion matrix visualizes a classifier's performance, with true positives, false positives, true negatives, and false negatives in illness detection. Learn to compute the accuracy score in Python.

Perform feature engineering by creating gender dummy variables and dropping one column, defining X and y. Split data with 30% test, apply standard scaling, and output model accuracy.

Explore simple and multiple linear regression to model the relationship between a scalar response and explanatory variables using a straight-line equation with an intercept and slope, plus an error term.

Explore simple linear regression using a dataset of experience and income to estimate an intercept and beta, plot observations, predict salaries, and perform hypothesis tests or confidence interval estimations.

Learn simple linear regression by deriving the y hat equation beta0 plus beta1 x, identifying the intercept and slope, and defining the error term as y minus y hat.

Learn to build a linear regression model in Python, including data prep, outlier removal, train-test split, scaling, fitting, predicting, and evaluating with mean absolute error and root mean squared error.

Apply logistic regression to a classification task using Pandas and NumPy, one-hot encode gender, use salary and age to predict purchased, and evaluate with scaling and a train/test split.

Explore the k nearest neighbors algorithm, a nonparametric supervised learner that uses proximity to classify or predict data points. See how varying k affects classification decisions with examples.

Learn to implement k-nearest neighbors for classification in python, scale features, perform train/test split, and compare k values with elbow method plots, then select three and evaluate accuracy.

Explore support vector machines, supervised learning algorithms for classification and regression. Visualize linear and non-linear SVMs with support vectors and the maximum margin hyperplane, and preview Python usage.

This lecture demonstrates support vector classification with grid search on a prepared dataset, covering dummy variables, scaling, train-test split, and kernel tuning (linear, rbf, polynomial) to maximize accuracy.

Apply support vector regression in Python with data prep, scaling, and outlier removal, then tune parameters via grid search and compare mean absolute error and root mean squared error.

Learn how decision trees use an algorithmic approach to split data based on criteria, enabling classification and regression in a nonparametric supervised learning method.

Demonstrates decision tree coding for classification, including preprocessing with dummy variables, one-hot encoding, label encoding, scaling, train-test split, and evaluating accuracy of 0.86 using a decision tree classifier.

Discover how random forest uses ensemble learning of multiple decision trees to improve supervised learning for classification and regression tasks.

Apply random forest regression to data, scale features, split into train and test sets, and tune hyperparameters with grid search while evaluating mean absolute error and root mean squared error.

Explore K means clustering, an unsupervised learning method that groups data points by minimizing cluster variance, computing centroids, and reassigning points to the nearest centroid until stable.

Analyze a booking dataset to predict cancellations using data visualization, feature engineering, and multiple machine learning models, then apply predictions to a new customer scenario.

Import seaborn and matplotlib to visualize data distributions, plot value counts and pie charts for meal plan, room type, and booking status, and select key features for modeling.

Engineer features and build models using label encoding, train-test split, and standardization; compare logistic regression, k-nearest neighbors, and random forest with accuracy scores.