AgriFieldNet India Challenge
Can you detect crop types in a class-imbalanced satellite image dataset?
$10 000 USD
Ended 25 days ago
179 active · 626 enrolled
It's over
Platform · 1 Nov 2022, 00:05 · 4

Finally - I am glad, this was tough.

Not too big shake-up, also, nobody I see has better private score than public score.

Congrats to Starlink on winning this one.

Discussion 4 answers

I agree, glad it is over. Very tough competition with everyone pushing till the last day.

Congrats to you guys and everyone who joined.

Woww.... Yeah.. @skaak Not too big shake-up. Very tough fight and very tough competition to crack indeed especially with the nature of the datasets provided. Except top 5 that experienced shake-up in positions, other half of the top 10 (i.e. 6th to 10th on LB) maintained their public positions till the end of the competition. Congratulations to Team Starlink. Well deserved guys!!!

Note: I strongly believe that no one uses "field_id" as feature.

Bravo to all the participants. Cheers!!!!

Tough one!! I hope the field id wasn't used as a feature.

Congratulations to the winners and to all participants. Onto the next one

Hi Skaak,

Request if you can share your approach. It is always a treat to learn from you