10,000pts
Timeline
Begin
2020/04/08
2020/04/08
Finish
2020/05/31
2020/05/31
Complete
2020/06/14
2020/06/14
Competition start: 2020/04/08 00:00:00
Competition closes on: 2020/05/31 00:00:00
Final Submission Limit: 2020/06/14 00:00:00
This competition has a total duration of 3 months, within which you will be able to make your submissions and obtain results automatically. Once the first part of the competition is over, you will have one week to choose your best model and submit it to be graded and considered for cash or points prizes.
This competition does not have a "Late Submission" option.
This competition does not have a "Late Submission" option.
Description
The essential basis of this competence is to be able to predict the number of people who have been affected (confirmed cases) or unfortunately declared dead by the COVID virus19 , in Latin America. If you can identify factors in your analysis that may affect the rate of transmission of COVID-19, you can share them.
Presentation file
We understand that this is a serious situation, and in no way do we want to trivialize the human impact this crisis is causing by predicting deaths. Our goal is to provide better methods of estimation that can help medical and government institutions prepare and adjust as pandemics develop.
Time Line
Presentation file
We understand that this is a serious situation, and in no way do we want to trivialize the human impact this crisis is causing by predicting deaths. Our goal is to provide better methods of estimation that can help medical and government institutions prepare and adjust as pandemics develop.
Time Line
- 04/08/2020 - Launch of the competition
- 08/04/2020 - 31//05/2020 - You work the model and test it against test.csv, we update it weekly with the new data from train.csv that we will launch. We are publishing a public table of positions in that period of time.
- 31//05/2020 - Final presentation file deadline.
- 06/01/2020 - 06/30/2020 - Evaluation period.
- 30/06/2020 we show the private position table
The initial data set for the model training will be from February 26 to April 08, 2020.
This dataset will be updated weekly until May 31st.
Evaluation
The metric for the evaluation of results will be the Root Mean Squared Logarithmic Error (RMSLE):
where:
n is the total number of observations pi is the prediction ai is the current value
The final score will be the average of the result of the RMSLE metric applied on the 2 columns (ConfirmedCases, Fatalities).
The results presentation file (Submission.csv) must have the following format
ForecastId,ConfirmedCases,Fatalities 1,10,0 2,10,0 3,10,0 etc.
For each ForecastID of the test set, the sum of the ConfirmedCases and Fatalities variables must be predicted, in the date range of the test.csv
Rules
Competitors can register and submit solutions as individuals not as teams (at least for now).
Since this is a learning challenge (the competitor will earn points, not cash), other than the rules in the DataSource Terms of Use, no other particular rules apply.
Maximum 10 solutions submitted per day.
Note: We reserve the right to modify these rules at any time as needed.
Since this is a learning challenge (the competitor will earn points, not cash), other than the rules in the DataSource Terms of Use, no other particular rules apply.
Maximum 10 solutions submitted per day.
Note: We reserve the right to modify these rules at any time as needed.
