Welcome to Oracle Database’s GHW: AI/ML Week Challenges!
Hello hackers! This week, we’re going to learn how to leverage several of Oracle’s database technologies through a series of fun and interactive challenges!
Getting Help
- If you have any questions about Oracle or their Global Hack Week challenges, head to the MLH Discord and find the #ask-oracle channel!
- Each coding challenge is accompanied by a LiveLab tutorial that will walk you through each challenge step by step
- If you need additional resources, you can find them at the bottom of this page!
Registration Challenges
Registration Challenge 1
Sign up and download 23ai VirtualBox
Head over to the VirtualBox signup page to get set up with a new account.
Registration Challenge 2
Create an account and Sign in for LiveLabs
Create a LiveLabs account so you can leverage Oracle’s LiveLab tutorials
Coding Challenges
Coding Challenge 1
Build a RAG application in 7 easy steps with LangChain and Oracle AI Vector Search
Objectives:
- Learn how to use the popular open source Python LangChain framework to search your PDF documents with natural language.
- The application will load a chosen PDF document, chop it up into chunks, vectorize and index those chunks.
- Build a simple ChatBot interface to allow natural language questions to be asked about data in your PDF documents.
- This RAG [retrieval augmented generation] architecture is powered by Oracle AI Vector Search, a feature of Oracle Database 23ai.
Documentation:
Coding Challenge 2
An introduction to Oracle AI Vector Search using SQL
Objectives:
- Learn the fundamentals of vector search and how it can be applied to similarity search, RAG [retrieval augmented generation] and finding outliers.
- Learn how to create, query and modify vectors using SQL.
- See how vector search uses a ‘closest match given the available data’ approach.
- See how that you can combine vector search with relational queries for advanced attribute filtering.
Documentation:
Coding Challenge 3
Get started with Oracle Machine Learning Fundamentals on Oracle Autonomous Database
Objectives:
- Get a quick tour of Oracle Machine Learning technologies on Autonomous Database.
- Use OML Notebooks to create and evaluate models and score data using SQL, Python and R.
- Use OML Services REST API to deploy models and score data. Use AutoML UI for a no-code machine learning experience.
Documentation:
Coding Challenge 4
Introduction to Oracle Machine Learning for Python on Autonomous Database
Objectives:
- In this hands-on lab, experience Oracle Machine Learning for Python on Oracle Autonomous Database.
- OML4Py supports scalable in-database data exploration and preparation using native Python syntax, invocation of in-database algorithms for model building and scoring, and embedded execution of user-defined Python functions from Python or REST APIs.
- OML4Py also includes the AutoML interface for automated algorithms and feature selection, and hyperparameter tuning. Join us for this tour of OML4Py.
Documentation:
Coding Challenge 5
MySQL : Machine Learning for Beginners using HeatWave AutoML
Objectives:
- Discover how HeatWave’s built-in capabilities enable the development of machine learning models directly within the MySQL database.
- HeatWave ML simplifies machine learning for both novice users and experienced practitioners.
- By providing the data, HeatWave ML analyzes its characteristics and creates an optimized machine learning model for generating predictions and explanations.
- In this challenge, participants will create and use a predictive machine learning model.
- The process includes preparing data, training a model using the ML_TRAIN routine, and generating predictions and explanations with the ML_PREDICT_ and ML_EXPLAIN_ routines.
- Finally, participants will assess the model’s quality using the ML_SCORE routine and view model explanations to understand the workings of their model.
- All these routines are executed within the HeatWave MySQL Database.
Documentation:
Resources
Hands-On Labs
Dev Gym
Documentation