Global Data Science Forum

Call for Code Spot Challenge for Wildfires

By Susan Malaika posted Tue November 10, 2020 07:12 AM

  

News

Hear a comparison of Model Approaches
Apr 5, 2021
https://www.crowdcast.io/e/predicting-australian

Crowdcast-Wildfires-2021-04-05-f-Susan-Assembled-Slides.pdf

Read about a comparison of Model Approaches
Mar 25, 2021
https://developer.ibm.com/technologies/artificial-intelligence/blogs/call-for-code-spot-challenge-for-wildfires-predictions-comparing-approaches/


The Final Results Posted on the Leaderboard
Mar 2 , 2021

The Finale - and Next Steps

Hear from IBMers building models for the wildfires challenge
Feb 22, 2021 -

https://www.crowdcast.io/e/call-for-code-spot-7

A Summary of Self Study Materials is Available

Jan 26, 2021 - You can find a list of self-study materials here

The Evaluation Script is Now Public

Jan 16, 2021 - You can find the challenge evaluation script here https://github.com/Call-for-Code/Spot-Challenge-Wildfires/tree/main/evaluation_scripts and if you consult the readme you can run it yourself.

Register for Office Hours - Tuesdays 8am US Eastern

From Dec 15, 2020 - Jan 26, 2021 -

To get the latest news, find resources etc - You have to register for each session individually
https://us02web.zoom.us/meeting/register/tZYuduGhrjsuGtyq1AGldJzq_3TJfmJiFF3G




Summary

Nearly 3 billion animals were affected by Australia's worst wildfire season that burned from July 2019 through March 2020 estimates Chris Dickman, a professor of ecology at the University of Sydney. The human cost to Aboriginal and Torres Island Australians, who lost their homes and their sacred sites, is devastating.

Join data scientists to develop models focused on forecasting wildfires in Australia for the upcoming wildfire season and enter the chance to win 5K US Dollars. To get you started we're releasing historical data sets extracted from Weather Operations Center Geospatial Analytics component (PAIRS Geoscope)  Our goal is to better understand the application of machine learning techniques in this domain.

Useful Links


      You can watch prior Crowdcasts


        Overview of Challenge


        Wildfires are among the most common forms of natural disaster in some regions, including Siberia, California, and Australia.  It is important to improve forecasting for wildfires for a number of reasons:

        1. To prepare and respond
        2. To understand the root causes
        3. To help to mitigate wildfires in the future

         

        At the Digital Developer Conference Data & AI, you'll hear about the Call for Code Spot Challenge on Wildfires. The objective will be to forecast wildfires in Australia during the month of February 2021 in order to better understand the application of machine learning techniques in this domain. We are excited to share an extract from Weather Operations Center Geospatial Analytics component (PAIRS Geoscope) with some of the data going back to 2005, and sessions to help you get started. Do join us. You can see a list of the sessions that were presented at the end of this blog.

        The Datasets

        Predict 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
        • NT=Northern Territory
        • QL=Queensland
        • SA=Australia
        • TA=Tasmania
        • VI=Victoria
        • WA=Western Australia

        To forecast the wildfires, you will be given 5 datasets, extracted from  Weather Operations Center Geospatial Analytics component (PAIRS Geoscope), which you can augment with other open datasets. You will also be given opportunities to try out your predictions before February in earlier stages of the contest.

        Note that there is no hidden data in this contest. You will be predicting  wildfires in February 2021 during January 2021. The leaderboard will check how closely your prediction matches with reality.

        The datasets and accompanying readme and slides are available via GitHub https://github.com/Call-for-Code/Spot-Challenge-Wildfires together with a starter notebook.

        • Landclasses Australia by region (static throughout the contest)
        • Normalized Vegetation index Australia by region
        • Weather
        • Weather forecasts
        • Wildfires

        The datasets will be refreshed through the contest at particular times (see the timeline table below). Contestants can incorporate other open datasets  into their model preparation.

        The Prize

        One winner at the top of the final leaderboard on March 1, 2021 (or when IBM declares the contest closed) gets 5K US Dollars.

        Registering for the Contest

        Step 1  - Register for the Call for Code Spot Challenge for Wildfires at https://developer.ibm.com/dwwi/jsp/register.jsp?eventid=cfc-2020-SP-wildfire - If you are working as a team on the Spot Challenge, please make sure each team member is individually registered.

        Step 2 - You will then register yourself for the contest leaderboard If you are working as a team on the Spot Challenge, please make sure each team member is individually registered. You can then form your team on the leaderboard.

        You can:

        How to Win

        Predict the size of the fire area in km square by region in Australia for each day in February 2021.

        Submit your prediction in January 2021 to the public leaderboard during the final stage "predict Feb 2021 (Feb 1-28)".  The date range when you can submit your final predictions is 2020-01-23 to 2021-01-31. If you examine the timeline closely, you will see that there is a plan to refresh the datasets  on 2020-01-29 which enables you to review your model again during the last days of January.

        You will be predicting wildfires in February 2021 during January 2021. In February 2021, the leaderboard will check weekly how closely your prediction matches with reality: the actual fires.  Please note that the actual fire information will become available in February 2021 by NASA. The raw data (the actual fires) provided by NASA will be processed in the same way as the training data as detailed in Readme_Docs. The overall scoring is based on two metrics, the mean absolute error (MAE) and the root mean square errror (RMSE) between the forecasted and the actual estimated fire area. The total score will be weighted 80% towards MAE and 20% towards RMSE

         

        Contest Stages and Submission Time Line

        There will be four main contest stages  - the first three stages are for practice. The final stage is what the contestants will be measured on.

        • Development - Try the platform - Predict Feb 2020  - we will keep this stage open until the end of the contest
        • Predict Jan 2021 week 3 (Jan 16-22)
        • Predict Jan 2021 week 4 (Jan 23-29)
        • Predict Feb 2021 (Feb 1-28)

        The following table describes the four stages including:

        • When the data is refreshed
        • When you can make submissions
        • The maximum number of submissions you can make

          Data Refresh Submissions take place Contest Stage Max Allowed number of submissions
          Available on 2020-11-10 Base Data - starts between 2005 - 2015 until 2020-10-31 2020-11-10 2021-01-09
          2021-02-28
          Development - Try the platform - Predict Feb 2020 (the first 28 days) daily 10, weekly 50, total 100
          Available on 2021-01-09 Refresh data to include up until 2021-01-08 (with some exceptions - see notes in zip file) 2021-01-10 2021-01-15
          2021-01-19
          Predict Jan 2021 week 3 (Jan 16-22) daily 5, weekly 35, total 35
          Available on 2021-01-18 Refresh data to include up until 2021-01-14 (see zip files)

          2021-01-16 2021-01-22

          2021-01-26

          Predict Jan 2021 week 4 (Jan 23-29) daily 5, weekly 35, total 35
          Available on 2021-01-23 Refresh data to include up until 2021-01-22 (see zip files)

          2020-01-23 2021-01-31

          2021-02-02

          Predict Feb 2021 (Feb 1-28) daily 3, weekly 3, total 3
          Available on 2021-01-30 Refresh data to include up until 2021-01-29 As above As above As above


          The Terms and Conditions

           Please make sure you have agreed to the Participation Agreement for the Call for Code Spot Challenge for Wildfires before you start submitting to this leaderboard. (See registering for the Contest Section)

          • No IBMers or Red Hatters can participate
          • A contestant can have one account only on the leaderboard and can be in one team only after the first Develop stage
          • The maximum team size is 5
          • Teams must be registered on this leaderboard by January 8, 2021 January 30, 2021
          • No team mergers are allowed
          • IBM can restrict the number of teams competing
          • No sharing of notebooks and models privately between teams unless you make the content available to all
          • The leaderboard determines the winner on March 1, 2021 or when IBM declares the contest closed
          • 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 top 5 contestants on the final leaderboard will be asked to share their notebook on Watson Studio and provide information on the tools they used as well as any other open datasets they incorporated

          We hope that you will join us.



          Conference Sessions in November 2020 that related to the challenge

          Sessions at the conference in track 4 that will help you get started with the contest:

          The Conference slack is here http://digitaldevcon.slack.com if you need an invitation to the slack go to http://ibm.biz/devcon-ai-slack. There is a lab channel #ddc-ai-competitions and a help channel #ddc-ai-help in the workspace

          Summary of Track 4 - Community : Contests and Open Source

          Talk Title

          Speakers

          Introducing the Call for Code Spot Challenge for Wildfires Dataset and Contest

          Hendrik Hamann Chief Scientist for Geoinformatics and PAIRS Geoscope, Omid Meh, Developer Advocate, and Sundar Saranathan Software Architect, IBM

          Getting Started with the Wildfires Dataset

          Margriet Groenendijk, Developer Advocate, IBM

          Getting Started with AutoAI and the Wildfires Dataset

          Gregory Bramble, Research Software Engineer, IBM

          Building and using stacked machine learning models -  a proven path to more accurate models

          David Carew, Developer, IBM

          Creating Inclusive IT Language – A Fireside Chat

          Priyanka Sharma, GM CNCF , Dale Davis Jones VP & DE, IBM

          Ways you can get involved in Women in Data Science (WiDS) 

          Karen Matthys, Executive Director, ICME, Stanford University

          What’s Next in Open Source Data Science & AI

          Todd Moore, VP IBM, Ibrahim Haddad, Executive Director, LF AI & Data Foundation Lisa Seacat, DE IBM


        #Featured-area-3
        #Featured-area-3-home
        #call-for-code
        #wildfires

        0 comments
        1878 views

        Permalink