Challenges and Leaderboards

2022

Challenge 2: Predict wildfires

Description: In this challenge sponsored by AltaML, individuals or teams of two competed to develop the best python based machine learning algorithm in order to determine whether or not a wildfire will occur the next day in areas of Alberta based on a variety of factors. Teams were provided with a historical dataset from Alberta Wildfire in order to train their algorithm. This hands-on challenge will allow people ranging from newbie coders to seasoned Machine Learning developers to try something new with data that has an impact on Albertans.

Timeline: March 1 – 22, 2022

Winners:

1st Place: Leonardo Torres

2nd Place: Victor Silva

3rd Place: Michael Ogezi

2021

Challenge 1: sell more cars

Description: Congratulations on deciding to become a tech entrepreneur and starting your first consulting business! After doing some extensive market research in Alberta, you have found that the hottest up and coming industry right now is automotive sales! Thinking long and hard about how you can get your piece of the pie, you have come up with your big idea: You will create a machine learning model to predict people’s car purchasing preferences for marketing!

Timeline: January 28 – February 8, 2021

Evaluation: Percentage of correct preference predictions

Winners: Logan Fairgrieve-Park, Mohamed Mujammil Mohamed Yacoob (Tied)

Winning Code

2020

Challenge 0: SMLB Inaugural Challenge

Description: In this challenge, individuals or teams of up to three will compete to develop the best python based machine learning algorithm in order to determine the price of  Edmonton AirBnbs based on a number of descriptive features such as number of guests allowed, neighbourhood, number of beds and more.  Teams will be provided with a training dataset in order to train their algorithm so that when given the test set, a prediction will be made about the price of each AirBnb.

Timeline: October 23 – November 8, 2020

Evaluation: Each submission will be evaluated using root mean square error, where the algorithm’s predictions will be evaluated against the test set’s data.  The algorithm with the lowest root mean square error will be chosen as the winner of the competition.

Winners:

1st Place: Navya Gururaj Rao and Varun Ranganathan

2nd Place: Lee Borschneck

Winning Code