Data Science Zero to Hero: Data Science Course from Scratch

Install and set up Python using Anaconda, launch Jupyter Notebook, and write your first Python code, exploring Anaconda Navigator and Spyder, plus essential editing features.

Explore Python operators, including arithmetic, assignment, comparison, and logical operators, and learn variable naming rules and how to run code in Jupyter notebooks.

Explore Python data types with a focus on numeric types, including integers, floats, long integers, and complex numbers, and learn how boolean and string types fit into programming.

Explore Python's string data type, understand indexing and slicing from both ends, and apply operators and methods like strip, len, lower, upper, capitalize, and split to manipulate text data.

Learn how to declare and index Python lists, and distinguish mutable lists from immutable strings. Master list methods such as append, pop, extend, and explore nested lists and list comprehension.

Discover how tuples and sets work in Python. Learn tuple immutability, how to declare and access tuple elements, and how sets deduplicate and support operations like union and intersection.

Learn Python dictionaries as key-value pairs, including creation, updating, and access by keys, and viewing keys and values with keys() and values(), illustrated by Aman's name, id, and subject.

Learn how control statements in Python drive program flow, starting with the if statement. Explore if else and nested if else, and master indentation and colon rules for correct execution.

Explore Python functions, including defining with def, function arguments, and return values; differentiate user defined vs system defined functions, and use lambda for inline operations.

Run Python commands and scripts from the command line using the Anaconda Prompt and Python interpreter, including executing .py files and handling user input for deployment and data science.

Learn statistics and mathematics vital for data science, covering descriptive and inferential statistics, measures of central tendency and dispersion, variance, standard deviation, and Python demonstrations using pre-existing functions.

Explore percentiles and box plots to understand data distribution, including the five-number summary (min, Q1, Q2, Q3, max), interquartile range, and outlier boundaries, and learn to plot in Python.

Perform basic data analysis in python with pandas, numpy, matplotlib, and seaborn; load mpg data, inspect shape and columns, and plot box, scatter, and distribution charts with descriptive stats.

Explore data distribution and the normal distribution, including mean, median, mode, and standard deviation, illustrated with a histogram and the 68%, 95%, and 99.7% rules.

Learn hypothesis testing as a core inferential statistics tool, with null and alternate hypotheses, p-values, and a 5% threshold using z tests and t tests on sample data.

Explore probability as the numerical representation of the likelihood of an event, and define event space and sample space; learn odds in favor via p/(1-p) with coin toss and dice.

Explore linear equations and equation of a line, including slope and intercept. Understand correlation, its range from minus one to plus one, and how to assess relationships between variables.

Master the basics of the structured query language (SQL), learn how to write simple queries, connect SQL with Python, and perform joins to retrieve and analyze data from databases.

Discover how machine learning learns patterns from data using algorithms to create models, and compare supervised, unsupervised, and reinforcement learning with real-world examples.

Begin an end-to-end data science project by importing packages, cleaning and exploring data, engineering features, encoding and creating dummy variables, training a linear regression model, and evaluating with r-squared.

Data science zero to hero teaches logistic regression for classification, contrasting it with linear regression, and covers model evaluation using a confusion matrix, accuracy, precision, and recall.

Learn end-to-end logistic regression in Python to predict ad clicks using feature engineering and data preprocessing, including train-test split, dummy variables, and evaluation with confusion matrix and classification report.

Examine the decision tree algorithm for classification and regression, building a tree from weather data with splits on wind speed and temperature. Explore bias-variance trade-offs and how splits drive predictions.

Master ensemble learning by combining multiple models to reduce bias and variance. Explore bagging, boosting, and stacking, with examples like random forest, AdaBoost, and gradient boosting.

Master ensemble learning with bagging and boosting on diabetes data, covering data import, cleaning, feature engineering, model tuning, and evaluation using random forest and AdaBoost.

Learn how clustering segments customers into buckets to guide targeted campaigns, using unsupervised learning and techniques like k-means and euclidean distance.

Explore k-means clustering and the elbow method to identify optimal customer segments. Implement in Python using annual income and spending score, visualize clusters, and interpret within-cluster sum of squares.

Understand model deployment in Python using Flask with storing and loading models, building a front end, and bridging storage and access using pickle or joblib.

Explore how generative ai reshapes data science. Learn the neural networks that power generative ai and complete projects in generative ai to solidify your understanding.

Explore artificial neural network concepts. Map data through neurons across input, hidden, and output layers in a fully connected network, training with trainable parameters to minimize error and activation functions.

Explore key neural network types in data science zero to hero, including feed-forward networks, CNNs, recurrent networks, and transformers, and learn how attention powers modern large language models like ChatGPT.

Build a retrieval augmented generation pipeline by creating a knowledge source in a vector database, ingesting documents, and using context to guide the language model's answers.