|IBM Announces Code Spot Challenge For Wildfires|
|Written by Kay Ewbank|
|Friday, 13 November 2020|
IBM has announced the Code Spot Challenge on Wildfires. The objective will be to forecast wildfires in Australia during the month of February 2021, with the aim of improving and gaining better understanding of how machine learning techniques can be used for this type of problem. One winner at the top of the final leaderboard will be awarded $5K USD.
IBM has created the challenge in response to recent wildfires in regions including California and Australia. The team organizing the challenge says it's important to improve forecasting for wildfires to improve the preparation and response when fires occur, to understand the root causes, and to help to mitigate wildfires in the future.
Nearly 3 billion animals were affected by Australia's worst wildfire season that burned from July 2019 through March 2020, according to estimates by Chris Dickman, a professor of ecology at the University of Sydney.
The details of the challenge were announced at IBM's Digital Developer Conference Data & AI on November 10.
Entrants to the challenge will be asked to forecast wildfires in Australia during the month of February 2021 using machine learning techniques. To help competitors prepare, IBM has shared an extract from Weather Operations Center Geospatial Analytics component (PAIRS Geoscope) with some of the data going back to 2005, and several sessions from the virtual conference are available on demand explaining what to do. At some point during the contest, an IBM tool such as Watson Studio or AutoAI should be used during the model development, training, etc.
The data set is made up of daily timeseries for the nine regions showing historical wildfires, historical weather, historical weather forecasts, historical vegetation index, and land classes. There's also a Jupyter notebook that shows how to load and explore the data.
The data should be used by systems to [redict the size of the fire area in km squared by region in Australia for each day in February 2021.
The regions are:
NSW=New South Wales
Alongside the five datasets that are provided, entrants can augment the data with other open datasets. They will also be given opportunities to try out their predictions before February in earlier stages of the contest. There's a competition leaderboard that will check how closely entrant's predictions match with reality.
Summer SALE Kindle 9.99 Paperback $10 off!!
or email your comment to: email@example.com
|Last Updated ( Friday, 13 November 2020 )|