Relataly Public Python Tutorials

A collection of practical Jupyter notebooks for applied machine learning, deep learning, generative AI, and analytics.

Developed alongside technical articles on Relataly.com, this repository has grown to nearly 160 stars and 90+ forks.

What you will find

Area	Examples
Applied machine learning	Anomaly detection, feature selection, hyperparameter tuning, and churn prediction
Deep learning	Neural networks, computer vision, and recurrent models
Time series	Forecasting and stock-market prediction
Generative AI	OpenAI and ChatGPT examples, prompt engineering, and API-based workflows
Recommender systems	Collaborative filtering and content-based methods
Analytics and data engineering	Visualization, geographic analysis, and PySpark

Tutorials

Data Visualization

• 042 - Visualizing Stock Market Structures using Cluster Analysis
• 070 - Geographic Heatmaps using Python
• 071 - Color-Coded Cryptocurrency Price Charts
• 072 - Bitcoin Logarithmic Regression Curves and Halving Dates
• 073 - Seaborn Plots Overview
• 112 - Correlation Matrix for COVID-19 and Financial Assets

Time Series Forecasting and Regression

• 001 - Forecasting US Beer Sales with Auto ARIMA
• 003 - Univariate Forecasting with Recurrent Neural Networks
• 004 - Adjusting Prediction Intervals
• 005 - Multi-Step Rolling Forecasting
• 006 - Multi-Output Regression
• 007 - Multivariate Time Series Forecasting with RNNs
• 008 - Feature Engineering for Multivariate Models
• 009 - Measuring Regression Model Performance
• 011 - Time Series Forecasting using Prophet
• 050 - Exploratory Feature Preparation for Car Sales Price Forecasting

Classification

• 017 - Permutation Feature Importance for Customer Churn
• 018 - Classifying Criminal Activity in San Francisco
• 019 - Classifying Shopper Buying Intention with Logistic Regression
• 020 - Measuring Classifier Performance
• 021 - Unsupervised Feature Selection
• 022 - Predicting Milling Machine Malfunctions

Clustering

• 041 - K-Means Clustering
• 043 - Customer Segmentation with Hierarchical Clustering

Anomaly Detection

• 026 - Credit Card Fraud Detection with Isolation Forests

Recommender Systems

• 031 - Movie Recommender using Collaborative Filtering
• 032 - Movie Recommender using Content-Based Filtering

Computer Vision

• 200 - Binary Classification of Cats and Dogs

OpenAI

• 601 - Prompt Augmentation for DALL-E using ChatGPT
• 602 - Comparing OpenAI Performance on Classification Tasks
• 603 - Simulating a Conversation between ChatGPT Agents
• 604 - Custom ChatGPT with Azure Cosmos DB and Embeddings
• 605 - Vector and Hybrid Movie Search with Azure AI Search

Natural Language Processing and Bots

• 024 - Building an OpenAI News Bot
• 025 - Building a Crypto Signal Bot
• 700 - Sentiment Analysis with Naive Bayes and Logistic Regression

Distributed Analytics

• 300 - Analyzing Zurich Weather Data with PySpark

Hyperparameter Tuning

• 015 - Regression Hyperparameter Tuning with Randomized Search
• 016 - Random Forest Hyperparameter Tuning with Grid Search

Responsible AI

• 800 - Fair Machine Learning with Fairlearn

Using the notebooks

• Browse the repository for individual .ipynb projects.
• Open a notebook locally or in a compatible hosted notebook environment.
• Review its imports and install dependencies in an isolated Python environment.
• Consult the related article on Relataly.com when available.

The notebooks were published over time and may reflect APIs or library versions current at their original publication date. Validate dependencies and outputs before using them in production.

Related repository

Relataly Public Python API Tutorials

Star history

License

This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License.