ML-Challenge

TKA Development is a team of builders who value creativity, teamwork, and a positive attitude. We believe that technical skill is important, but we don't judge people based on their background or appearance. We welcome all kinds of builders, creators, and designers, and we're particularly excited to invite developers to take on our open-ended challenge.

At TKA, we're less interested in your formal qualifications than in your ability to create amazing products and learn quickly. If you're up for the challenge, we encourage you to clone our repository and start building! Once you're ready, submit a pull request and our whole team will review your work. We can't wait to see what you create!

Instructions

1. Learn

• Feel free to learn from any resources. here are a few resources that can help you to get started:
  • UBC CPSC 330 Applied Machine Learning Playlist
  • Cornell Tech CS 5787 Applied Machine Learning Playlist
  • How to Implement Linear Regression from Scratch
  • PyTorch Beginner Tutorial - Training an Image Classification Model and Putting it Online!
• We have a Udemy account that you can borrow (reach out to [email protected]). We have access to an online course that teaches you the fundamentals of web development, feel free to reach out if you are interested!
• Get familiar with GitHub by going through the GitHub tutorial.

2. Build

• You have 1 week to complete the challenge. We can see the commits timeline
• Implementation (code) will be evaluated according to the requirements

3. Show

• Impress us with your skills
• Go beyond the requirements
• Beat the competition
• Join us
• Win

The Goal

Easy Task: Regression

Reference: delete before sharing the coding challenge https://github.com/WillKoehrsen/ai-projects/blob/master/bayesian_lm/Bayesian%20Linear%20Regression.ipynb

Features (you can implement some or all of the features):

• Get to a minimum of --% accuracy
• Show a visualisation of the regression
• ??

Medium Task: Classifier

Reference: delete before sharing the coding challenge http://slazebni.cs.illinois.edu/spring21/assignment3_part2.html

Features (you can implement some or all of the features):

• ??

Hard: Image Generation

Reference: delete before sharing the coding challenge https://colab.research.google.com/github/tg-bomze/collection-of-notebooks/blob/master/AI_Interior_Designer.ipynb#scrollTo=YgrM5beakjsm

Features (you can implement some or all of the features):

• Have an interface for users to draw their interior design

Getting Started

• Clone the challenge repository
• Create a dedicated branch
• Write your code
• Commit your changes
• Fork the challenge repository
• Issue a Pull Request
• Notify us. Please send an email to [email protected]

Suggestions

• get familiar with Jupyter Notebook and Google Colab. you can use the note book right away by trying it on your browser.
• Make it easy for us to try your app. Add instructions on how to run your demo. There's a section below where you can fill in.
• Don't be afraid if you're still a newbie. We will judge what you built adjusted with your experience. If you're just starting out, but can learn fast. We want you :)
• Be prepared to explain your decisions and your thought process in the next interview. We're curious about how you think! :)

How to run the demo

(REPLACE THIS WITH YOUR INSTRUCTIONS)