CGIAR Root Volume Estimation Challenge
Helping Africa
$15 000 USD
Completed (over 1 year ago)
Computer Vision
Prediction
1070 joined
257 active
Start
24 Jan 25
Close
09 Mar 25
Reveal
10 Mar 25
'%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)
Leaderboard Sealed
Connect · 1 Apr 2025, 13:30 · 23
Dear Zindians,
Well done on competiting in a particularly complex challenge!
Zindi remains committed to delivering valuable solutions to our clients and partners.
While we uphold high standards of usability and value, we also recognize the complexity
of this challenge and appreciate the dedication and effort each of you put into your
submissions.
Following our review of the CGIAR Root Volume Estimation Challenge, winners were
selected based on the following criteria:
- Usability & Usefulness: The proposed solution must be practical and valuable for the client (feature selection / engineering).
- Correctness of Preprocessing: Image merging must be performed accurately, ensuring proper alignment of the left and right image segments and stacking the images vertically.
We sincerely appreciate your participation and hard work in this challenge.
Congratulations to everyone 🎉
Seems like the rules were changed after the challenge - adding subjective criteria. These should have been declared at the beginning of the challenge.
But what is done is done. Terrible competitions happen from time to time.
This really hurts me 🫤. The rules were changed after the competition. The client could have used the top 30 solutions that seemed more appropriate since He only wants image models. But changing the scores seems intolerable to me. The leaderboard should be the leaderboard. The private top 3 score seems artificial to me, just to get them ahead. I'm going to leave the platform for a while.
Oh you don't have to. It happens from time to time. Please don't leave the platform. There are more competitions to come. I am sure they will go over the rules for the next competitions.
Thank you @CodeJoe. I will get over it soon
Inserting private scores of 1, 1.1 and 1.2 is just bad. Terrible judgment call there. You could have just announced prize winners without creating a fake PB.
Exactly, they could have given their money to whoever they thought deserved it, why taint the leaderboard with fake scores? We compete to build the best models and this isn't something the organisers can decide based on preference.
As I said. This is roberry. It is no more science but taste. Had we know it in the beginning, we wouldn't had had competed.
It's sad really. Didn't think this sort of thing could happen, all the same we'll continue to build the best models we can and hope for fair judgement 😔
Yes Lad, don't stop. You did amazingly well. More competitions are there. Keep on building the best models you can.
Thanks man, see you at the top of the leaderboard in competitions to come 👊
Game on Brudda 🙌 ! Let's do this.
The competition really felt very strange, the winner was not judged based on the private score, but based on the source code assessment when the competition was over.
It is Roberry again --- (i.e. Smart Energy Supply Scheduling for Green Telecom Challenge )
I don't really understand what the end of this competition rule is. My model uses image encode with timm and combines it with start, end, genotype and stage values and uses lgbm for the final prediction. Does Zindi only want a solution using the image without using other values (start, end, genotype, stage)?, or must use CNN?, or something else. Please explain so that we can anticipate problems related to the rules that may recur in future competitions. thank you for your response.
That's being said. You were robed @sys_ts__. You was in the inital top 3 ? To me, it is an image based solution. I think they should create a downvote button.
No, i was initial top 4
I second this point:
Please explain so that we can anticipate problems related to the rules that may recur in future competitions. thank you for your response.
Hey @AJoel could the winning solutions be open sourced so that we can learn from them? Atleast that will bring tranparency to the below issue which I think was the main judging point?
Correctness of Preprocessing: Image merging must be performed accurately, ensuring proper alignment of the left and right image segments and stacking the images vertically.
@zinidi painful, funny and laughable at the same time....🤣🤣🤣🤣🤣 Zindiiiiiiiiiiii not again..... Can @zindi respond to this simple question below coming from @sys_ts__ :
Please explain so that we can anticipate problems related to the rules that may recur in future competitions. thank you for your response.
if zindi selected less sentitive to outliers metric for evaluation small test sample, this couldn't happened.
couldn't agree more. I sometimes don't understand the choice of Zindi metrics at times.
In my opinion, Zindi should have removed the Start, End, Genotype and Stage columns from the Test data from the start of the competition (or they forgot to remove them), because it seems they reject the use of these columns for prediction, Zindi only allows images to be used as the only feature. So the existence of these columns in the Test data causes confusion.
Thank you for your complete explanation Amy.