AI Python for Beginners

Explore object oriented programming in Python, including classes, objects, methods, constructors, inheritance, polymorphism, encapsulation, and data abstraction, with hands-on examples like car and person classes.

Explore essential Python libraries for data science, including NumPy for arrays and math, Pandas for data frames, Matplotlib and Seaborn for visualization, and scikit-learn for machine learning.

Analyze distributions in the Titanic dataset with histograms and bar plots using pandas, matplotlib, and seaborn, focusing on age and passenger class, and prepare clean, standardized data for machine learning.

Explore feature distribution analysis on the Titanic dataset by visualizing numerical and categorical features with histograms and bar plots using pandas, matplotlib, and seaborn.

Analyze the relationship between numerical features with a correlation heat map to reveal correlations, interdependencies, and guide feature selection using seaborn, annotations, and a cool-warm color scheme.

Analyze the Titanic data set to explore features and survival outcomes, generate synthetic data with NumPy, and visualize results using pandas, Matplotlib, and seaborn.

Create and visualize simulated stock market data with pandas and matplotlib using ranges and price trends. Load CSV data into dataframes and plot scatter, correlation heat maps, and time series.

Apply one-hot encoding to categorical features and standard scaling to numerical features to prepare data for machine learning models, using feature engineering and transformation techniques.

Explore machine learning fundamentals, including supervised, unsupervised, and reinforcement learning, and the workflow from data collection and preprocessing to model deployment with regression, classification, and neural networks.

Learn polynomial regression theory, implementation in Python, and when to use it for non-linear relationships that linear regression underfits, with visualization and evaluation using MSE and R2 square.

Explore multiple linear regression—its concept, mathematics, and implementation for predicting house prices from area, bedrooms, and age using synthetic data; evaluate with mean square error and R2.

Explore the filter method with select k best using f regression to identify the top three features, then apply wrapper rfe and embedded lasso cv for feature selection.

Use Lasso CV to automatically select key features for price prediction with L1 regularization. Build a pipeline with scaling, encoding, and polynomial features to cross-validate and boost model performance.

Learn logistic regression basics, the sigmoid-based probability for binary classification, with the cost function and binary cross entropy, implemented in Python and evaluated by accuracy, precision, recall, F1, and ROC.

Explore hands-on logistic regression: import libraries, generate synthetic binary data, split into training and testing sets, train a logistic model, and evaluate with accuracy, confusion matrix, ROC curve, and AUC.

Explore the mathematical intuition behind decision trees by examining entropy, gini impurity, and information gain, and learn how impurity and purity guide node splits.

Explore the mathematical intuition of support vector machines, including hyperplanes, margins, hard and soft margin concepts, and the kernel trick for non-linear separation.

Master the theory and practical implementation of support vector machines in Python with scikit-learn, exploring linear and kernel methods for binary and multi-class classification, the maximal margin, and decision-boundary visualization.

Explore the Naive Bayes classifier and its probabilistic intuition, including Bayes' theorem, conditional independence, and the Gaussian, multinomial, and Bernoulli variants used for spam filtering, sentiment analysis, and document classification.