AI Club’s

Project Workshop

• Build your own ML project from scratch�
• Add to your portfolio�
• Compete for prizes

Noah Pragin

Project Manager

Lauren Gliane

Project Officer

Kellen Sullivan

Project Officer

Ally Chen

Project Officer

A senior majoring in CS with a custom option focusing on robotics and AI. Experience includes TAing for the machine learning course and working on autonomous vehicles.

A 3rd year in Computer Science with a Data Science focus. Joined AI Club to learn AI and help make AI education and projects accessible and welcoming at OSU.

A senior majoring in CS with a custom AI focus and minoring in math. Experience includes completing multiple AI projects during a data engineering internship.

A sophomore in CS with a custom focus in AI and Finance, with a Finance minor. Joined AI Club to gain early hands-on experience with AI, an opportunity often limited to upperclassmen.

​

Project Officers

What We’re Building

Goal: Complete ML Project in 10 Weeks

• Work on a guided machine learning project of your choice. No prior experience necessary and we'll help you through every step!

What You’ll Gain:

• Portfolio Project: A complete ML project to showcase
• Hands-on Experience: Real-world data science workflow
• Resume Material: Demonstrable AI/ML skills
• Competition: Chance to win prizes, AI Club merch and either compute credits or an Amazon gift card!
• Community: Work alongside other learners

10-Week Project Timeline Pt. 1

Setup and Tutorials

Problem & Dataset Selection

Data Analysis & Preprocessing

Model Comparison & Tuning

Set up your computer to work on your project and complete tutorials for new tools

Train 3+ models, basic hyperparameter tuning

Define problem, select dataset, identify potential models

Explore data, create training, validation, and test splits, feature engineering

1

2

3

4

10-Week Project Timeline Pt. 2

Catch-up / Get ahead

Final Model Selection

Model Analysis

Deployment Implementation

Presentations & Winner Selection

Catch up on anything you missed, polish your model, plan your deployment approach

Build your interface, integrate your model

Comprehensive tuning and test set evaluation on selected model

Bias detection, interpretability, error analysis

Present your projects and vote on a winner!

5

6

7

8/9

10

Team Structure & Logistics

• Team Size
• Work solo or in groups, whatever works best for you!
• Solo projects are absolutely doable and encouraged.
• Groups larger than 3 may have a hard time evenly splitting work
• Matchmaking
• Have friends? Form your own group!
• Flying solo? We’ll help connect you with others!
• Want to work alone? Perfect!
• Scan the QR code and fill out the form to�help us build teams

Support I

• During Meetings
• We will walk around during meetings to check in and assist you
• Recitation-style meetings, where we are happy to give you a hand if you ask

​

• Outside of Meetings
• Noah’s office hours: 10-11am T+Th in Johnson 121
• Send questions to Discord!

Support II

• The Project Workshop page of the AI Club website has our slides, tutorials, and example project!
• osu-ai.club/project-workshop

​

• The first and last slide will always have a QR code to our website

Packages / Tools

• NumPy
• Numerical processing, used under the hood by everything else
• Pandas
• Data organization and manipulation, used to aggregate your data before feeding it into a model
• MatPlotLib
• Plots and visualizations, used to visualize your dataset and model performance
• Scikit-learn
• Machine learning models and data preprocessing, used for the training and evaluating models

Tips for Success

• Ask questions!
• ML is a broad topic, and there is no way for us to cover everything
• Find something that interests you!
• Tell us if you see anything wrong or inconsistent!
• Consistency > Intensity

Feedback Survey

• Let us know how today’s meeting went!

Week 1: Setting Up & Learning New Skills

• Definition of Done
• Starting to think about what problem you want to work on
• Access to ENGR servers
• Code editor set up
• Python ready to go
• Tutorials complete for tools you’re unfamiliar with
• Extra Credit
• Problem identified and defined
• Dataset selected and accessible

What Makes a Good Problem?

• Clean, available datasets
• There is no model without data to learn from
• Interesting to you
• Some level of interactivity
• Classification problem?
• Best suited to beginners, strongest support from us

Limitations of Machine Learning

• Supervised ML Problem Types
• Classification: Is this email spam or not spam?
• Regression: What will the house price be?
• How to Frame a Problem for ML
• Predict [specific outcome] based on [available data]
• Bad: Detect COVID-19
• Good: Predict COVID-19 severity based on lung CT images
• Guiding Questions
• Is your prediction target specific?
• Do you have data to learn from?
• Can you quantitatively measure success?

Project Ideas & Examples

• Pet Breed Classification
• Problem: Identify cat and dog breeds from photos
• Dataset: Oxford-IIT Pet Dataset
• Models: CNN, ResNet
• Heart Disease Detection
• Problem: Predict heart disease risk from medical data
• Dataset: Cleveland Heart Disease Dataset
• Models: Random Forest, Logistic Regression, SVM
• Network Security
• Problem: Detect malicious network connections
• Dataset: KDD Cup 99, NSL-KDD
• Models: Decision Trees, Neural Networks, Ensemble

Tutorials (Ordered)

• Environment Setup
• Skim this one even if you’re set up, it calls out some good extensions for your code editor
• NumPy
• Pandas
• Scikit-learn
• MatPlotLib?

Let’s Build Something Amazing!

• Action Items
• Start thinking about a problem to tackle!
• Computer set up and ready to go
• Complete tutorials on unfamiliar tools
• Resources
• Tutorials on our website; scan the QR code!
• Project idea slides
• Questions? Stuck?
• Raise your hand! We’re here to help