There are plenty of links for simple, practical, and efficient coding solutions, hand-picked books, and videos organized within modules below to cover; Data Structures, Computer Science Algorithms, Data Analysis Examples, and Machine Learning Applications.

Note: Starting from Module2, make sure to have a working Ipython (jupyter) notebook to be able to view, use, and edit the .ipynb files. It is recommended to install Miniconda and then work on the environments you need. With the conda package management system, install: 'numpy', 'scipy', 'matplotlib', 'sklearn', 'torch', and 'pandas' libraries to be able to run the code in the jupyter notebooks.

Wait! I want to learn data science & machine learning step by step

Module1: Start Here! Learn Coding with Python & Javascript

Fix your typing first. Start learning Python right away. Study and practice coding a few hours a day, every day. Cover every single topic in this document thoroughly and create your .py notebooks.If you want to go deep in coding, also study Javascript in parallel.

Done? Now you know how to code! Click here to quit the data science track and become a web developer instead!

Module3: Data Structures, CS Algorithms & Intro to ML

Study these DS & CS Algorithms:

BST, Decision Trees, Random Forest, Linked Lists, Graph SPT & Heap, Graph MST, DFT & FFT, Sorting Algorithms, Bipartite Matching & Max Flow, Dynamic Programming, Stacks, Queues & Heaps, BFS & DFS, Greedy Algorithms, ML: Intro to NLP, ML: KNN, ML: Kmeans++, ML: 2LayerNN

together with: CS231 Then study: R, Big Data with Scala & Spark

+Learn: NoSQL/SQL, data warehousing, hadoop, Java, and cloud computing to become a data or software engineer.

Module5: Create Your Own Portfolio with EDA & ML Solutions:

Binary Logistic Regression

"The Framingham Study is a longitudinal investigation of constitutional and environmental factors influencing the development of CVD in men and..." More about data...

Multi-Class Logistic Regression

"The MNIST database of handwritten digits has a training set of 60,000 and a test set of 10,000 examples. Each instance is 28x28 pixels..." More about data...

Classification with Decision Tree & XGBoost

"Features are computed from a digitized image of a fine needle aspirate (FNA) of a breast mass. Data has information about 569 patients..." More about data...

Anomaly Detection with Time Series

"Real data: This is data for which we know the anomaly causes; no hand labeling reveals the ambient hourly temperature in an office setting..." More about data...

Bitcoin Price Prediction

"Bitcoin prices are downloaded and visualized. Arima (statistical) and LSTM (machine learning) models are written from scratch and tr..." More about data...

Audio Processing & Laughter Analysis

"There are 22 .wav files in the 'laugh' and 'laugh_test' sub-folders in the Audio Dataset Folder. Some very sincere and some brutally fake..." More about data...

Extracting Image Features with VAE

"Created a Variational Auto Encoder (VAE) and fed the Yale Face Database to this model to extract the average facial features of this dataset..." More about data...

Self Supervised Learning for Auto-Parking

"A truck is created in a frame of 400x300 pixel area for about 100,000 scenarios (at different coordinates) and a random steering angle (wheel input)..." More about design...