Machine Learning Bootcamp: Python, Projects & Deployment

Explore the hierarchy of ai, ml, dl, and genai, and distinguish supervised, unsupervised, and reinforcement learning while outlining an end-to-end machine learning project workflow.

Master supervised learning with labeled data, covering classification and regression, and key models like logistic regression, decision trees, and random forests to predict categories or numerical values.

Explore unsupervised learning with unlabeled data, uncover hidden patterns through clustering and association, using k-means, dbscan, gmm, and market basket analysis for customer insights.

Explore the three main machine learning types: supervised, unsupervised, and reinforcement, along with their definitions and real-world examples like spam filters, diabetes prediction, and robotics.

Explore eight major steps of machine learning project workflow from problem statement to deployment, including data collection, analysis and visualization, preprocessing, model training, evaluation, and showcasing with GitHub and Streamlet.

Learn how to set up a Python environment for writing code, notebooks, and data analysis to build AI, machine learning, and generative AI projects, using Colab, Jupyter, and VS Code.

Explore Google Colab, a free cloud workspace to write and run Python for data science, machine learning, and AI directly in your browser, with GPU access and notebook sharing.

Install Python on Windows, add python.exe to path, verify with python --version and pip list, and set up JupyterLab or Notebook to work locally with NumPy, Pandas, Matplotlib, and scikit-learn.

Install Visual Studio Code, configure Python and Jupyter extensions, and learn to create and use virtual environments (.venv) with activation and the correct interpreter for code and notebooks.

Explore Python basics from variables and operations to data structures, functions, and classes, culminating in a calculator project that prepares you for data analysis and machine learning.

Learn Python basics by practicing print and input statements, creating variables, and exploring data types like int, float, string, and Boolean in a Jupyter notebook or VS Code.

Learn data type conversion in Python by turning integers to floats, floats to integers, and values into strings or booleans, using int(), float(), and str(), with input handling notes.

Explore Python operators from arithmetic to logical and assignment, learn how to compare values, test membership, and verify identity in lists with practical code examples.

Learn how to use Python if-else statements to make decisions based on conditions, including if, else, elif, and nested tests, with examples like age checks, voting eligibility, and grading.

Learn Python loops to automate tasks with for and while, start and stop semantics for range, and break or continue controls.

Explore Python data structures (lists, tuples, sets, and dictionaries), understanding mutability, ordering, and indexing, with practical examples of creation, access, updates, and iteration.

Explore reading, writing, and appending text files in Python with open and the with statement, and prepare for csvs and pandas later.

Learn how to handle Python errors with try and except blocks, provide friendly messages for value errors and division by zero, and use finally to run code no matter what.

Learn how to define and call Python functions with def, pass parameters, return values, use default parameters, and manage local and global variables for reusable, clean code.

Explore Python classes as blueprints for objects, define attributes and methods with init, create instances, and use inheritance to extend behavior with examples like dog and puppy.

Explore core data libraries for machine learning with NumPy, Pandas, Matplotlib, and Seaborn. This warm-up helps you import, explore, and visualize data before modeling with scikit-learn.

Master NumPy basics for data science and machine learning by creating and manipulating arrays (1d, 2d, 3d), indexing, slicing, and performing elementwise and matrix operations.

Learn pandas basics, including series and data frames, loading CSV files, inspecting data, selecting and filtering, creating new columns, and performing descriptive statistics to prepare data for machine learning.

Learn the basics of Matplotlib to create line, bar, pie, scatter, histogram plots, customize styles, and build multi-plot figures for exploratory data analysis.

Learn Seaborn basics to create clean, modern visualizations with simple code, including histograms, kde, box plots, count plots, scatter and pair plots, and a correlation heat map.

Explore the iris dataset with numpy, pandas, matplotlib, and seaborn to perform a mini data analysis workflow, visualize features, and map species labels.

Discover why math matters for machine learning, and how linear algebra, statistics, calculus, and probability give intuition, enable data modeling, and guide training with gradients.

Master algebra basics for machine learning by defining variables and equations that relate inputs X to outputs Y, with weight (slope) and bias (intercept) guiding linear and logistic regression.

Learn linear algebra basics for machine learning: scalars, vectors, matrices, and tensors; data shapes, weights, and bias, with diabetes data and dot product and matrix multiplication.

Learn hands-on linear algebra in Python using NumPy to represent data as scalars, vectors, matrices, and tensors, and format machine learning data with X and Y for model fitting.

Explore dot product, euclidean distance, and matrix multiplication, and see how WX plus B drives linear regression, with a practical workflow that splits data into features and targets.

Explore statistics basics for machine learning to understand data with central tendency, spread, distribution, and relationships. Learn how to detect outliers, scale features, and assess model performance using these concepts.

Explore real data statistics with tips dataset: compute mean, median, mode, range, variance, standard deviation, and interquartile range (iqr); visualize distributions and identify skew and outliers using pandas and seaborn.

Apply descriptive statistics to understand data distribution and skewness, identify and quantify outliers, and analyze feature relationships with a correlation heat map for ML pre-processing.

Explore calculus basics for machine learning, including how changes in weights affect loss, the roles of slope, tangent, and gradient, and how gradient descent optimizes models.

Practice calculus concepts essential for machine learning with hands-on line and curve slope tasks. Explore slope estimation and gradient descent intuition with plots, loss, and code.

Explore loss functions and gradient descent to update model parameters and reduce mean squared error, illustrated with numpy calculations and matplotlib visualizations of weight updates.

Explore probability basics as the language of uncertainty in machine learning, covering random events, sample spaces, conditional probability, distributions, and models from Naive Bayes to deep learning.

Explore probability hands-on with Python by simulating coin tosses and dice, studying conditional probability and independence, visualizing distributions with NumPy, pandas, and matplotlib, and connecting to machine learning predictions.

Load data from CSVs, Excel, Kaggle, databases, and APIs, then recap the full ML workflow—from problem statement to deployment, covering exploratory data analysis, preprocessing, training, evaluation, and optimization.

Explore how a data set powers machine learning, detailing samples, features, and a target; distinguish data types, and learn to split data into training, validation, and test sets.

Master loading data from csv, Excel, and parquet formats into pandas, selecting columns and sheets, handling tab separators, previewing rows, and setting up environments for seamless analysis.

Load datasets from Kaggle and the UCI machine learning repository using Python libraries and APIs in Google Colab, authenticating with a Kaggle API token stored as an environment variable.

Connect to a Postgres database from Python and load data into a pandas data frame. Use SQLAlchemy and psycopg2 binary with dbware to pull the diabetes data set into Python.

Master exploratory data analysis to understand data, uncover patterns, identify issues, and guide preprocessing for modeling, by examining feature types, data quality, and univariate to multivariate analyses.

Explore data quality checks to ensure reliable machine learning results, focusing on data integrity and preparation for model training and deployment. Apply practical steps using Python.

Apply univariate analysis to the Titanic dataset, examining numerical features like age and fare and categorical features like sex and embark, and visualize distributions to reveal outliers and missing values.

Explore multivariate analysis by using pair plots, heat maps, and cluster maps to reveal interactions among age, fare, family size, and survival in the Titanic dataset, guiding feature engineering.

Detect outliers using box plots, the IQR method, and Z-scores. Compare statistical measures for data with and without outliers in the Titanic dataset to understand impact on distribution and models.

Learn data preprocessing basics, including train/test split, imputation of missing values, encoding categories, handling outliers, and feature scaling for reliable model performance.

Master encoding categorical features for machine learning in Python, transforming categorical data into model-ready input for training, evaluation, and deployment.

Identify and cap outliers with IQR bounds, apply train bounds to train and test data, and prepare numeric features before scaling in a Titanic dataset workflow.

Identify and address imbalanced data in diabetes classification using undersampling, oversampling, and SMOTE on the training set, with a proper train-test split to avoid data leakage.

Master standardization and normalization for numerical features using standard scalar and min-max scaler. Fit on training data and transform test data to prevent data leakage, illustrated with diabetes.csv.

Build a full end-to-end data preprocessing pipeline for a loan approval dataset, including missing-value imputation, outlier handling, encoding, imbalance correction, scaling, and initiating with a train-test split.

Explore classification in supervised learning, including binary and multiclass tasks, probability outputs, and decision thresholds. Grasp why evaluation matters for real-world AI and how data, training, and optimization impact performance.

Explore intuition behind two key classification models—logistic regression and a support vector machine—covering linear boundaries, sigmoid-based probabilities, and use cases for binary and text classification.

Explore the intuition of k-nearest neighbors and Naive Bayes, two classification approaches—instance-based and probability-based—covering feature space, distance measures, independence assumption, and text classification use cases.

Discover how tree-based models like decision trees and random forests classify data with rule-based questions, ensemble voting, interpretability, and resilience to overfitting on nonlinear tabular data.

Compare major classification models to decide which to use for your data and constraints, covering logistic regression, SVM, KNN, Naive Bayes, decision trees, random forest, and XGBoost.

Explore classification evaluation metrics, from accuracy and imbalanced-data pitfalls to precision, recall, F1-score, confusion matrix, and ROC-AUC, with threshold concepts.