Farm Pin Crop Detection Challenge
'%3e%3cpath fill='%23001489' d='M0 0v6h9V0z'/%3e%3cpath fill='%23e03c31' d='M0 0v3h9V0z'/%3e%3cg stroke='%23fff' stroke-width='2'%3e%3cpath id='d' d='m0 0 4.5 3L0 6m4.5-3H9'/%3e%3cuse xlink:href='%23b' stroke='%23ffb81c' clip-path='url(%23c)'/%3e%3c/g%3e%3cuse xlink:href='%23d' fill='none' stroke='%23007749' stroke-width='1.2'/%3e%3c/g%3e%3c/svg%3e)
Helping South Africa
$11 000 USD
Completed (over 6 years ago)
Classification
Earth Observation
1469 joined
42 active
Start
Mar 04, 19
Close
Sep 15, 19
Reveal
Sep 16, 19
Last minute change of rules?
Help · 16 Sep 2019, 05:56 · 1
Dear Zindi,
I would like to have an official statement that the leadboard is constructed based on 100% of the data set.
"Note that there is Public and Private Leaderboards. The Public Leaderboard excludes approximately 50% of the test dataset. While the competition is open, the Public Leaderboard will rank the submitted solutions by the accuracy score they achieve. Upon close of the competition, the Private Leaderboard, which covers 100% of the test dataset, will be made public and will constitute the final ranking for the competition."
The unveiling of the private leaderboard is popularly known as "the shakeup". Models built to generalise well result in similar or even better performance on the private leaderboard. I have been a victim of overfitting to the public leaderboard where I was 1st position then significantly dropped to 21st position because my model didnt generalise well enough. I hope I was able to provide some clarity.