Huge thanks to my teammates @nymfree and @DJOE . We could have not done this without the team work.Our solution consists of four key stages:1. Exploratory Data... See more

https://github.com/littlefiveflowerstarfish/floodI'm also making videos recapping my solution. Please check the playlisthttps://www.youtube.com/playlist?list=PL... See more

Can anyone please share the winning strategy for these scores? Thanks

This competition was very interesting and challenging at the same time. I tried heavy feature engineering, undersampling, oversampling, applied images, used sev... See more

is there final code reproduceing test without post-process step? my lb score 0.002447672 -> 0.002315709 boost by https://zindi.africa/competitions/inundata-m... See more

I selected two submissions but the leaderboard is showing neither of them. Instead it is based on my best public submission which has the leak. Please help me.B... See more

[Edit]: I made an video to explain it: https://youtu.be/GDZ8-ta-P_A and yeah I used AI voice LOL[Edit 2]: I want to apologize for my previous baseless claim tha... See more

Looks like Zindi's server is out of disk space @Zindiunknown error: No space left on device @ io_write - /tmp/20250212-31221-1u3hjv0

Hey guys,Trying to submit but keep encountering this error""""""Missing entries for IDs id_j7b6sokflo4k_X_0, id_j7b6sokflo4k_X_1, id_j7b6sokflo4k_X_2, id_j7b6so... See more

Hello guys, I hope you're enjoying this challenge. I have been scratching my heed with both neural nets and boosting models. Still I couldn't break under 0.003.... See more

Competition rule

In the data description, it is stated: "Please note that some locations do not have a composite image as these flood events were before the start of Sentinel-2.... See more

Hi everyone. this competition almost come to end soon. so as usually let share cv and lb scores.mine is 0.00309 cv to 0.00305 lb

I originally thought that the same event_ids would be present in the train and test sets, however upon further inspection this does not seem to be the case. Is... See more

It says there's some train images with no floods (true). But it mentions this because some images / floods are before the sentinal images, right? So could tes... See more

It's stated in the data description that "For each location the CHIRPS precipitation data has been aggregated over a 5km radius." Does that mean each location i... See more

I've been experimenting with models that take as input the image of a location and the time series of precipitation data for that event and try to predict, for... See more

I find deeplearning tough. i've used traditional models, lb=0.0037 da_ are you here private_1x Koleshjr. are guys understanding the starter notebook codes 😅

How do you reconcile these two statements, because it seems that there will not be a flood within 40-60 weeks on either end of the 730 days??...events contain r... See more

Hello guys , hope you are all doing great . I am writing to ask if you find the composite images helpful for improving your model performance ??? For the top sc... See more

Mismatch in Event IDs:event_id formats differ across datasets (train.csv includes time-step suffix; processed_chirps_data.csv does not), causing failed merges.... See more

If we look at the data in the "id" column, are the X values ​​from 0 to 729 time data?, is the task of the model created from this competition later used to pre... See more

Reduce the dimensionality caused by high-cardinality categorical features. Also, Rare categories are grouped into a single "Other" category, reducing dimensiona... See more

Load the raw data files, inspect their structure, and identify key attributes (numerical and categorical features, missing values).Outcome: Data structure and b... See more

could anybody please tell me in metric in the competetion https://zindi.africa/competitions/inundata-mapping-floods-in-south-africa score lower is better or hig... See more