define functions with def, pass parameters, and return results to create modular, reusable python code; explore scope and lifetime of variables and importing modules.

Explore Python string manipulation, including concatenation, slicing, and formatting with f-strings; learn split, join, replace, and strip; master regex with re, find all, and sub for NLP data cleaning.

Learn pythonic code with descriptive variable names, modular design, and built-in features, list comprehensions, lambda, map, filter, reduce, and build a command-line task manager with file handling.

Learn week two data science essentials, from NumPy for numerical computing to pandas for data manipulation and cleaning, with visualization using Matplotlib and Seaborn and an EDA project.

Explore NumPy for numerical computing in AI and data science, covering array creation, manipulation, linear algebra basics, and hands-on exercises with zeros, ones, linspace, reshape, indexing, slicing, and element-wise operations.

Learn pandas for data manipulation by working with series and data frames, loading and saving csv and excel files, and performing core operations like head, tail, and describe.

Learn how to group data by category in pandas and apply mean, max, min, and custom aggregations. Use pivot tables to reshape data and compute summary statistics.

Execute an end-to-end exploratory data analysis project by cleaning data, handling missing values, removing duplicates, transforming and aggregating data, and visualizing patterns with matplotlib and seaborn, including Titanic dataset analyses.

Explore vectors and matrices, perform element-wise operations, scalar multiplication, and matrix multiplication using NumPy. Learn identity, zero, diagonal matrices, and matrix-vector multiplication.

Delve into advanced linear algebra concepts, including determinants, inverses, eigenvalues and eigenvectors, and matrix decomposition with NumPy, covering SVD, PCA, and dimensionality reduction for AI.

Explore derivatives and optimization in machine learning, including first and partial derivatives, gradients, and gradient descent, with Python examples using SymPy and a linear regression case.

Master integrals and optimization in machine learning through practical Python examples, covering definite and indefinite integrals, probability distributions, regularization terms, SGD variants, and Adam.

Explore probability basics, conditional probability, and Bayes' theorem, then model Gaussian, Bernoulli, binomial, and Poisson distributions with Python, applying them to machine learning.

Explore central tendency and dispersion, including mean, median, mode, variance, and standard deviation, with Python examples; apply hypothesis testing, confidence intervals, and t tests.

Explore linear regression from scratch with gradient descent, mean squared error, and r-squared evaluation, implementing a full mini project from data generation to model tuning.

Explore week four probability and statistics for machine learning, covering probability theory, random variables, distributions, hypothesis testing with p-values, and regression for real-world data.

Learn to apply one-sample, two-sample, and paired t tests to compare means; perform chi square tests for independence and goodness of fit; and run anova for three or more groups.

Explore week five of machine learning fundamentals, covering supervised learning, regression models (polynomial and regularization), classification with logistic regression, model evaluation, cross validation, and k nearest neighbors.

Explore machine learning foundations and terminology, including supervised, unsupervised, and reinforcement learning, and learn about features, targets, train-test splits, and basic visualizations.

Explore supervised learning and regression models, including linear regression, cost functions, and gradient descent, with hands-on training and evaluation using mean squared error and R square for predictions.

Apply polynomial regression to model nonlinear relationships by transforming features to polynomial terms, and use ridge and lasso regularization to prevent overfitting while evaluating squared error on California housing data.

Explore classification in supervised learning, including binary, multi-class, and multi-label tasks, and learn logistic regression with the sigmoid function, decision boundary, and evaluation metrics.

Evaluate regression and classification models using MSE, MAE, RMSE, accuracy, precision, recall, and F1; explore cross validation methods such as k-fold, stratified k-fold, and LOOCV, plus confusion matrices.

Explore regression and classification evaluation metrics, including MAE, MSE, RMSE, R2, accuracy, precision, recall, F1, and ROC AUC, with hands-on exercises using iris data and California housing datasets.

Explore an end-to-end supervised learning mini project that builds regression and classification models through a data pipeline with eda, preprocessing, model selection, and evaluation on California housing and telco churn.

Explore feature engineering essentials—transform raw data into meaningful inputs, apply encoding, scaling, transformation, and feature selection; plan techniques using a Titanic dataset exercise to enhance model performance and interpretability.

Explore scaling and normalization to transform numerical features, compare min-max scaling and z-score standardization, and observe their effects on iris dataset classifiers using scikit-learn.

Encode categorical variables with one hot encoding and label encoding. Assess nominal and ordinal features, high cardinality with frequency and target encoding, and observe effects on model performance.

Explore feature creation and transformation to boost model performance by deriving date-time components, interactions, and aggregations, then compare linear and polynomial models on a bike sharing dataset.

Explore ensemble learning techniques—from bagging, random forests, gradient boosting, XGBoost, Lightgbm, to Catboost—plus handling imbalanced data, culminating in a real-data model comparison project.

Explore bagging and random forests as robust ensemble methods, using bootstrap sampling and feature randomness to reduce variance and overfitting, with a hands-on breast cancer classification exercise using grid search.

Explore XGBoost, the extreme gradient boosting framework, focusing on speed, missing data handling, regularization, and hyperparameter tuning to outperform traditional gradient boosting in binary classification tasks.

Explore Lightgbm and Catboost in practice, comparing gradient boosting on large datasets with categorical features, and evaluating models on the Titanic dataset to highlight speed, memory use, and overfitting controls.

Explain imbalanced data challenges in classification and use Smote oversampling and class weights to boost ROC AUC and F1, with fraud detection examples.

Compare bagging and boosting ensemble models on the telco churn dataset, preprocess the data, address imbalance with smote, and evaluate performance using roc auc and classification metrics.

Explore the difference between parameters and hyperparameters, and discover how tuning hyperparameters like learning rate, max depth, and estimators improves model performance and efficiency using the iris dataset.

Explore the fundamentals of deep learning and neural networks, including automatic feature extraction, forward and backpropagation, and common activation functions. Practice with TensorFlow and PyTorch on MNIST, CIFAR-10.

Learn to build and train neural networks with TensorFlow and Keras, using sequential and functional models, define layers, compile, train and evaluate on MNIST, and save the trained model.

Learn to build and train neural networks with PyTorch using tensors, autograd, and the nn module, plus MNIST data handling, loss, optimizer, training, evaluation, and saving models.

Develop a convolutional neural network for cifar-10 image classification, including data normalization, one-hot encoding, and dropout. Experiment with activation functions, optimizers, and learning-rate schedules to optimize accuracy and generalization.