#🏆┊competition-general

1 messages · Page 1 of 1 (latest)

somber tartan
#

hello, world

#

Let's get this channel kicked off! I'd love to hear what everyone's first competition was like - the good, the bad, and the ugly 🙂

clever flax
somber tartan
#

Expectation | Reality 🥲

void helm
#

Ohh I agree @clever flax. I did my playground series and I couldn't understand what's the use of splitting the train set into train/test when there is already test set for submission. It took my days to figure out and I had to go through many codes already submitted to finally understand.

ruby aurora
somber tartan
somber tartan
lost talon
sharp carbon
#

Lets start with the bad, my first shake-up ! "Microsoft malware prediction" final leader-board was terrible for me and many kagglers, but i learnt a lot ... i think LOFO feature importance was one of the research results of this competition. (It was all about "Centos_OSVersion" feature , see the github link)

image.png
Microsoft Malware Prediction

Can you predict if a machine will soon be hit with malware?

GitHub
GitHub - aerdem4/lofo-importance: Leave One Feature Out Importance

Leave One Feature Out Importance. Contribute to aerdem4/lofo-importance development by creating an account on GitHub.

woeful ermine
#

I am a bit of an outlier as I haven't completed any competitions properly despite being on kaggle many years... 🤣 what usually happens to me is that I get too hung up on data munging/learning how to work with unusual data formats such as satellite imagery, and by the time it got to modelling I was either bored of the problem or the competition is over... However, I do learn a lot just by this sort of experimenting, and since most real world data science problems start with messy data, I would say even if you don't like competitions there are a lot that you can learn from them :))

somber tartan
jagged jetty
ruby aurora
sharp carbon
#

I think current active playground series - Season 3, Episode 20 Predict CO2 Emissions in Rwanda competition (started 2 days ago) is a good start to practice a full pipeline regression problem . I remember i started my journey in kaggle with "titanic" and "house price" playground competitions , like many novices i was a bit confused about submissions , cross validation , model selection and ... but didn't give up, decided to read other participants notebooks( initially EDA notebooks) , this idea helped me a lot. After that i participated in featured completions , and got my first solo silver medal in LANL Earthquake Prediction . So be persistent. Don't give up on your training and practices. There will be times when things get tough, but you need to keep going and enjoy your "Aha... Moments" !

Predict CO2 Emissions in Rwanda

Playground Series - Season 3, Episode 20

LANL Earthquake Prediction

Can you predict upcoming laboratory earthquakes?

tame shuttle
#

Thanks @sharp carbon for the insight

tacit matrix
runic rune
astral egret
scenic bone
somber tartan
somber tartan
somber tartan
astral egret
lost talon
merry carbon
runic rune
#

So, what was the thing people did to achieve 0.06 public LB and got >5 for private LB for ICR competition think

true wraith
#

Excited 😁

finite sonnet
#

What a wild shake up KEKW I almost participated in that one for a little bit

runic rune
#

I completely skipped it lol

#

They probably don't even know they won holyfuck

pallid night
#

I roughly spent 5 days on the competition. I was pretty sure this would happen in the very start of it. Anyway, I learned something new in the competition as well, so worthwile 🙂

finite sonnet
#

I might take a day to just poke around and see if there is some principled approach someone couldve taken to get a consistently good result.

My thought is that someone couldve probably done something with very smoothed and unconfident predictions that wouldve done consistently. Maybe something where you only even moved the prediction if there was wide consensus from many models. Otherwise just default to 0.5 or some other safe bet to minimize the damage of anything wrong

silver olive
#

these ~1000 row tabular comps always surprises me, and I can't hold myself from joining

plucky geyser
#

The most curious thing that has happened to me is that if we define as training CV+Public LB, as validation Private LB, submissions as epochs and we plot everything as it was a learning curve, we can see when the overfitting starts.
Quite few lessons learned after having handled such a small dataset.

merry carbon
#

How did you see their solutions

sharp carbon
pallid night
sharp carbon
runic rune
#

The main thing is doing that thresholding stuff to boost public score completely obliterates your model

#

I don’t get why does that make sense to do

crisp pier
#

I get that it sounds insane to drop data based off time valuesin a medical competition, but is it really that insane when these data points were clearly way off, they were really far from all the other data when I was looking at them through umap representation.

#

and it wasn't some random thing, I tried 100 different seeds and these data points were always largely separated from main data

#

Yeah, I too now think that it was ballsy to drop these rows, didn't think about it when I did it, I guess these data points were poorly labelled or something, I'm pretty sure they appear somewhere in the beginning of covid pandemic too

#

It also seems that most top has general case CV solutions, top1 solution is really great too, I gotta check it out later

pallid night
crisp pier
#

his solution is public, you could try it out

#

but yeah that would be one hell of a notebook

pallid night
#

going to run it 🙂

crisp pier
#

Btw, did anyone else try to understand tabpfn source code?

#

I might be tripping, but I don't think it's training anywhere, it just checks if data is ok to pass to tabpfn and then it just concats whatever you passed in fit with whatever you want to predict and just predicts it with pretrained model

#

I don't think their models are available to run not in inference

#

at least I tried to load it to try fine tune but it said model was inference only or something, I can't remember

pallid night
#

yes..

finite sonnet
#

I just wonder with certain things like the null epsilon dropping if there was any possible way to know that was a good idea with the information we originally had? Like if it didnt show anywhere on the training set or public lb set that this was beneficial and only showed up on the private test set because data was sampled in a different way and we didnt know about this then how could one have possibly made this bet?

crisp pier
#

although I get what you are saying

runic rune
#

One key question is "why does submission error count as 1 submission when we only have 1 submission a day"

#

This submission error is not easy to find the problem too [like forgetting to set index = False]

#

There are hundreds of discussion posts on this

lucid valley
#

Well data points with unknown date and all target to 0. Not so weird to drop them in general

#

I actually have a wilder hypothese regarding data quality. The point are not independant that is you have patients that appears multiple time. First as sane then as ill. This is relatively classic (help collecting data, see the evolution over time). But for the ML algo this is not good as you have points with close features but different targets. This is also why we have unstable cv. Because it wildly depends how the observations of the same patient are split.

#

(This is target over time) see the pattern repeating between 2012 - 2016. It’s like we had a bunch of patient as 0 then as 1, one year later.

barren dune
#

Anyone wants to team up with me for the housing problem?

loud forge
#

Sure

hushed echo
somber tartan
somber tartan
dusty temple
#

Is there any schedule for future competitions? I am hoping to join the next competition right at the beginning

lost talon
dusty temple
#

I see. Can we reasonably expect there to be a competition starting when another one ends, or is it just random?

lost talon
dusty temple
#

Alright, thanks for the info

slate turtle
dusty temple
#

Where could I find the playground competitions?

#

Never mind, I have figured it out

rigid shadow
runic rune
#

@real delta I don't seem to see any nulls in the sample submission

dawn stratus
#

I am getting error--- script file 'C:\Users\HP\anaconda3\Scripts\conda-script.py' is not present.

#

How to solve this

runic rune
wild ember
#

how to solve titanic dataset for accuracy 1

zinc rampart
dapper surge
dapper surge
lost talon
dapper surge
clear grove
molten radish
cyan gust
#

Hi All. I am struggling to do one thing. If I train a model on my local machine, how can i use it in Kaggle competition notebooks if they do not have access to internet?

slate turtle
cyan gust
slate turtle
cyan gust
wide scroll
#

In the past years, I have actively participated in multiple Tabular Playground Series competitions. During this process, I have built a Kaggle Pipeline for my personal use.

Here is my kaggle pipeline, my Kaggle Pipeline (https://github.com/arnabbiswas1/kaggle_pipeline_tps_aug_22)). This is an Open Source Python based pipeline (193 ⭐ at github) for Kaggle tabular data competition. Although it is customized for Kaggle TPS August 2022, with limited code changes, this project can be used as a pipeline for any tabular data competition. This project includes APIs for most of the ML competition related tasks:

  1. data processing
  2. visualization
  3. feature engineering
  4. training
  5. ensembling
  6. feature selection
  7. hyperparameter optimization
  8. experiment tracking
  9. submission of prediction to kaggle

Here is the discussion at Kaggle: https://www.kaggle.com/competitions/titanic/discussion/435856

Titanic - Machine Learning from Disaster

Start here! Predict survival on the Titanic and get familiar with ML basics

potent flare
dire elm
#

Hello everyone. Great to be here

rich relic
#

Hello guys. I'm new here in discord channel, I would like to share with you guys two community competitions with fully sponsored visit to Italy for the first place winner. Please feel free to participate in them: https://kaggle.com/competitions/oemc-hackathon-global-fapar-modeling
https://kaggle.com/competitions/oemc-hackathon-eu-land-cover-classification

OEMC Hackathon: Global FAPAR Modeling

Challenge for modeling Fraction of Absorbed Photosynthetic Active Radiation (FAPAR) using ground stations and Earth Observation data

OEMC Hackathon: EU Land Cover Classification

ML challenge for classifying 72 multi-annual land cover classes for Europe using LUCAS ground-truth samples and Earth Observation data

real lion
#

@wide scroll i want to join you man,,I want to learn n grow

pallid night
#

so I just completed intermediate machine learning course on kaggle, and believe that I can participate in all the starting begineer competitions, ngl

#

still far away from participating in actual competitions but I will get there-

slate turtle
# pallid night so I just completed intermediate machine learning course on kaggle, and believe ...

I think you can already go for any competition you want, as there is no particular competency requirement for participation. Supposedly I know how to compete after having done it 30+ times, yet I still placed #2555 in one of the recent competitions. It has no negative effect on me, and I simply moved on to the next one. As long as you don't take things too seriously, I think you can start right now with featured competitions.

pallid night
viral herald
#

Hey guys! Does anyone know any really good resources to learn Data Science and Machine Learning theory?

molten radish
#

Anybody working on kaggle competition ribonaza RNA ?

runic rune
#

What is the longest time you have waited for your submission to be scored after submitting it ?

flint wyvern
flint veldt
#

Hello Dear kagglers , i am looking for someone to collaborate with me in competitions , anyone who wants to join dm me. Thank You ❤️

sullen loom
thick hedge
#

@viral herald

#

cousera and kaggle lol

#

freecodecamp

stray wedge
neon gazelle
molten radish
#

One of the competitions I completed got good mse, as I'm experimenting with adding features it getting improved, but need someone with domain knowledge so that I can achieve lowest mse

pearl notch
molten radish
pearl notch
real lion
#

hi guys..anyone want to join me in forming a team for google mobile decimeter challenge?

gilded radish
#

Hello everyone!
I have looked over many competitions and found that several topics like NLP, Image Processing, and Deep Learning are very crucial knowledge to success in competitions. I have completed several courses on kaggle. But they are not enough for a good idea in statistics and coding. Please suggest good free resources to learn.

past tree
umbral bluff
# gilded radish Hello everyone! I have looked over many competitions and found that several topi...

^ additionally for more in depth look into deep learning https://youtube.com/playlist?list=PLAqhIrjkxbuWI23v9cThsA9GvCAUhRvKZ&si=w0b0A3IftxhaPbYR Andrew karpathys videos and additionally http://introtodeeplearning.com/ mits mini intro to deep learning course

YouTube
Neural Networks: Zero to Hero
MIT Deep Learning 6.S191
MIT Deep Learning 6.S191

MIT's introductory course on deep learning methods and applications.

feral tusk
#

Is there any chance that an existing competition will be reworked? i'm talking about neurips unlearning challenge and it's flawed metric. If not, may final leaders be reconsidered after the competition is over?

wild spade
#

Hello mate. I want to ask about the kaggle competition submission. As some of the test data content missing value, if i delete them and make prediction, the submission file will be somewhat shorter than the 'submission sample' file (lack some observation). My question is: How Kaggle treat these missing lines in my prediction file?, how is it affect my score of prediction?

#

So the only way to deal with missing value in test data is to impute it, or remove column if the column is omit by dimentional reduction in training set

#

Right?

#

Thank you for the insight

rigid jewel
#

Hello, I have participated in Bengali.AI competition. I ran on CPU and my notebook has been running for more than 5 hours now. Is this normal?

#

If not, What should I do. This is my first competition

ocean python
#

Hi guys I am looking for a team for HackCBS 6.0
PLease drop a ping if anyone has a team\

vapid oriole
#

hi, is everyone getting into that ML ctf competition ?

desert scarab
#

Hi Maskman, I am trying to get started. Not sure how to get past the test. I will figure it out though!

odd turtle
#

I'm in the same place, trying to figure out Cluster 1

desert scarab
#

I might have to do lots more googling

amber stirrup
#

Hi Maskman, I'm in. Currently solved 9

rocky rune
bitter isle
abstract leaf
#

Hi, Is it legal to use the dataset competition as publication like paper or journal or student thesis?

slate turtle
nova cape
#

Hello! I've been using kaggle a lot as an online jupiter notebook for my private data but i've been thinking about starting doing some competitions.
I'm more specifically looking for time series prediction competitions.
Do you have some to recommend? 🙂
(please @ me if you answer i dont always check the channel)

nova pond
#

@nova cape titanic and space titanic are always good ones to begin with

nova cape
lost talon
#

@nova cape Here are competitions tagged with "time series analysis" on Kaggle: https://www.kaggle.com/competitions?tagIds=13209

(There are probably others that aren't listed here, but this should be a good start)

Kaggle Competitions

Kaggle is the world’s largest data science community with powerful tools and resources to help you achieve your data science goals.

nova cape
#

Right thanks!
ig i'll just start from most popular from those!

sonic forum
neon hawk
halcyon grail
lean quail
#

Hi guys, am I missing something or there is no group created for the competition data-rush-2023
It is a kudos competition, is that the reason no-one participates in that, I can see 0 discussions in that competition

patent dirge
#

What's a kudos competition?

#

and I guess what is a kudo haha

lost talon
#

@patent dirge It means competitions without cash prizes. Kudos means "thanks".

tame rock
spring sluice
#

Hi!!
I need to find competition that Find your ideal payment by changing loan amount, interest rate and term and seeing the effect on payment amount.
D you know ?

craggy olive
#

@spring sluice just a formula will do 😊

spring sluice
craggy olive
#

@spring sluice take a loan, I guess 😂

main vigil
#

Hello everyone i am new on kaggle and i want a team to start LLM Detect ai Generated text

pallid night
#

Hey, just found out about this server

random heart
#

Hey everyone. I'm new to Kaggle, and I've been getting warmed up with the Titanic competition. I've experimented with several models and methods of eda, but haven't been able to get past 76%. Is there a channel or discussion that anyone can recommend?

peak berry
#

Hey guys, we are currently of team of 2, trying to take LLM Detect ai Generated text Competition.

#

Anyone thinking of joining?

lean quail
pallid skiff
#

Hi, how do I join a competition?

obsidian lion
#

I notice a lot of competition winners use stacking /ensembles of high powered machine learning algorithms. Why does this technique work so well?

tall galleon
#

Mathematically, if two models are created independent of each other, and have the same accuracy, an ensemble of both models models will have identical bias, and improved variance, relative to the performance of either model alone

It's sometimes difficult to create truly "independent" models, but essentially combining model predictions is almost never a bad thing.

I'm sure there are some Youtube videos on the intuition behind this, most ML/SL textbooks will also have a detailed mathematical explanation for why ensemble modeling is statistically favorable as well

pallid night
#

Yes, agreed, thank you. However, the "bad" thing would be "increased CPU/memory usage".

#

Unless maybe, if you have a large number of features, and you give each inner model only access to a small subset of the total number of features.

#

It's an interesting strategy for sure.

mystic galleon
#

Anyone cares to explain "Hill Climbing" algorithm intuitively? I notice many top Kagglers include this in their solution. It seems "Hill Climbing" algorithm is becoming popular algorithm of choice recently especially during this latest playground competition

lone torrent
#

where can i find the channel for Mohs Hardness Regression

sand locust
#

guys i need some help

obsidian lion
#

Could someone help me speed up a computation?

grave fable
woeful ermine
frigid socket
hollow hull
#

Just curious does kaggle only have playground series as competitions ?

lean quail
hollow hull
#

I have

#

But all I see is this playground series

tired oasis
#

Hi everyone, I'm new to Kaggle but I have done a few projects on Forecasting, Prediction modules.

I want to make some friends to work with and pursue my career.

dawn rock
#

Ok

iron jungle
#

Are you already stuck on "Scoring..." when you submit ? I haven't done anything very resource-intensive, so I don't understand why it's taking so long.

merry fog
#

Could always try refresh and re-submit? Also worth checking your submission is meeting the expected format

iron jungle
trim flame
#

hello ! i am new here

iron jungle
#

I wanted to know, on a kaggle competition leaderboard, if you have the same score with someone, how is the one in front and the one behind calculated?

dawn rock
#

Me too, I change the parameters and can't submit 😮‍💨😮‍💨😮‍💨

#

Submit failed

pallid night
#

the displayed values are rounded off, but the ranking is done on the full precision scores

open merlin
#

is there any info when santa 2023 will start? im really excited 😊

lost talon
iron jungle
ashen holly
teal pagoda
ashen holly
teal pagoda
#

you should refer the dataset for that, the dataset has range as well the the units

#

ill link the dataset down below

#

wait they don't have the normal rate but they do have the units

ashen holly
# teal pagoda ill link the dataset down below

I’m aware and did attach the link of the original dataset and the competition dataset.
As you mentioned it’s doesn’t include the normal range but that isn’t my issue at all
I listed the 3 problems I’m facing in the discussion, please check it.

ashen holly
teal pagoda
#

great!!

ashen holly
#

How do i make my score visible on my notebook?

teal pagoda
#

after running your notebook and saving it's version, you can submit the output or submission file to the competititon by clicking on the three dots on the competetion page and pressing on the submit option in the drop down

grave fable
#

Hi guys, I'm looking for online learning resources (courses/books/articles) and would like to know which have been crucial for your career and Kaggle "mastery"?

jovial basaltBOT
#
levi_511 has been warned

Reason: Spam

#
levi_511 has been banned

Reason: Too many infractions

bright prairie
supple leaf
#

Hi I'm new to Data science i'm doing house prices prediction competition but i cant remove the nan values in the columns with float data type Can someone help me

lean quail
supple leaf
#

thanks @lean quail i have resolved it it will work when i select it by columns and do fiina

clear ginkgo
#

i was creating a model for image classification but the code keeps on running out of memory if i try running it on GPU or takes superrrr long if i run it on TPU, i tried fiddling with batch size, image size in pixels, reducing the conv2d and maxpooling layers but nothing can make it faster, i dont understand if it is normal for the program to take close to an hour's time for 2gb of data or am i doing something wrong ?

brittle cairn
inland cloud
#

Hey, I'm looking for a project to do for my regional science fair. Any suggestions/ideas?

trail leaf
#

Hi i want to do the LLM - Detect AI Generated Text project but I am confused at how my model will be getting graded. I have dont other competitions before a while ago and I converted the data to a csv called submission.csv and my model got graded. Can someone please explain how to submit a model and how it gets graded

dusk shuttle
#

Please help me on how to submit my notebook for scoring . Has been failing for the last three days. This is my first time. Please 🙏

left crag
#

You can try the titanic contest as it will walk you through each step

teal pagoda
#

if you can brief me with what exact problem you're facing maybe i can help

teal pagoda
dusk shuttle
dusk shuttle
#

Anyone with a clue on how to ccess logs in kaggle. Can I see the logs of my notebook?

obsidian lion
dusk shuttle
#

I have seen it but there is no much useful info. The logs there are not when the notebook is submitted.

native blade
#

Hello Dears,
I want to submit UBC Ovarian Cancer Subtype notebook, but there is a problem with submission, it failed after about 30 seconds and showed this message "Notebook Threw Exception",
It's really short time and definitely doesn't relate to submission I think.
please help, just 2 days left

hallow wing
#

👋 Hi, everyone!
🤝 I'm looking for friends to:
💻 Join me in coding.
🏆 Participate in Kaggle competitions.
📩 If you're interested, please DM me.
🕒 Preferably in the Vienna timezone (GMT +2).
🗣️ Fluent in English language only.

ashen holly
#

I believe teaming with someone on the other side of the globe is more wise

#

You now work all day long on code.

coral drum
uncut rune
#

Hi!
I'm looking for friends to:
Join me in coding.
Participate in Kaggle competitions.
If you're interested, please DM me.
Fluent in English language only.
||def didnt copy paste this||

brittle cairn
#

Hey guys Paulos here. I'm pretty hardcore studying about #🚀┊spaceship-titanic and approaching it in R. I'm aiming to get a >= 90 % score in the competition. Only 2 people have done it (as far as the leaderboard is right now, remember that only 2 month old max submissions are still in the leaderboard)

Anyways, if you are interested in approaching this problem the way I'm doing it just ping me and I will contact you and talk about it. I'm really excited and I'm learning a lot along the way. Love it.

verbal latch
#

@hallow wing Hi I'm interested in team up

distant orbit
#

hi gyz ||:')||
i'm a new kaggler
just know the basics
interested to participate in Kaggle competitions, if you tho, shout a dm out to me

dry delta
#

Hi, anyone want to collaborate on kaggle, I am new to it.

#

interested to participate in Kaggle competitions, if you tho, shout a dm out to me

long latch
#

Hi everyone, I'm new here i would love to participate in Kaggle competitions, hit me up!

gilded thorn
#

is there anyone looking for a teammate for the playground competitions? I have a small amount of experience; looking for some teams that are willing to work on the more introductory competitions. Please DM me if you would like to colaborate!

slender birch
brittle cairn
#

Ok forget about it, those are getting started competitions not playground ones

stark solstice
#

hii i have recently complete machine learning a-z course from udemy and andrew ng course 2 times and studying b.tech from cs(Ai-Ml) . and i want to make a group of those people who really want to do good in this field give daily 3-4 hours++ on projects

jade ether
#

Hi, I want to team up to participate in kaggle competitions

noble shale
proven wedge
#

Anyone know a good beginner competition for image classification?

open dune
#

Hi , I want to know what are the best beginners competitions for machine learning , I have done the Titanic and the spaceship Titanic one is there any other recommendations on what competitions should I go next would be so helpful

proven wedge
open dune
violet shard
#

Anyone know a good beginner competition for quant?

wind snow
#

hello guys

#

The Learning Agency Lab - PII Data Detection anyone working on this project??

obtuse whale
#

Can anyone suggest me beginner competition for data processing and data visualisation?

turbid atlas
runic rune
prisma warren
#

Hello, Am new here and looking for any team willing to Accommodate a new member. Data science & machine learning .Am ready to join right away

prisma warren
mortal vessel
#

Hello! Looking for a team for HMS Harmful Brain Activity. I'm aware it's almost over but would like to use the data while it's available. Please dm if interested.

scenic barn
#

Hello I am interested in participating in fine-grained image recognition (FGIR) related competitions such as geolifeclef-2024 and planttraits2024 or any of the others mentioned here:
https://sites.google.com/view/fgvc11/competitions?authuser=0
https://www.imageclef.org/LifeCLEF2024
I am a PhD student doing research in this area for a while and I hope I can apply it to more practical scenarios
Please DM me if interested

FGVC11 Workshop - Competitions

Running
FGVC11 hosts several competitions. Please click on the links below for more information about each competition.

rain magnet
#

can any one know about kaggle spaceship titanic

#

can any one know about kaggle spaceship titanic

austere barn
#

Hey guys, I wanted to ask what classifiers for Titanic prediction should I use to get higher accuracy? So far the highest accuracy I have got with an optimized Decision Tree is 0.7799. I have also used Logistic regression,KNN,SVM,Random forest but they got lesser accuracy.

mystic galleon
tepid comet
#

Hey ! I have a problem of multilabel image classification but 2 labels the percentage of 1 is about 95% . I think this will harm the training . how can I augment the the percentage of 0 ?

#

generally when I do data augmentation I do it randomly using image data generator but how can I augment the part of images when it has not a person or machine !

frigid vortex
#

Hey, how do I change competition team?

golden quarry
#

Btw, how can I deal with the name column of the Titanic prediction? I see various notebooks dropping it, but felt that it would provide some information if parsed properly 🧐

true belfry
sturdy smelt
#

Hey,
Looking for a team mate for the competition : ⁠hms-harmful-brain-activity-clas…
I'm familiar with eda...and sklearn machine learning models...
If interested, then plz respond...
i know the competition is about to end...but 10-15 days are sufficient to make a significant contribution

pallid ether
glass sedge
pallid ether
#

But anyways I have a problem with the "House price prediction" competition. I can not use the csv or download it in any possible way. My error is 'ParserError: Error tokenizing data. C error: Expected 1 fields in line 9, saw 2'. Does anyone know how to solve it?

bitter lintel
bitter lintel
velvet cradle
sullen frost
weary isle
#

Hello I am completely new to data science and programming, I am doing the Intro to programming course ( + started learning python through youtube very recently ) and was wondering if I should wait until I get the hang of basics to understand the code for titanic or just follow the tutorial and do it

Thank you ❤️

lucid valley
weary tide
#

Hello dear members. Am new to Data Science and to Kaggle too. Just started the competition. Hope to make some new cute friends here.❤️

midnight parrot
dense hazel
#

@midnight parrot @weary tide Hey guys i am new. i would like to join you !!

zenith niche
#

Hi....
I just started using my kaggle account this week actively...I was inactive but now I am trying to stay as active as possible...
There is a competition whose objective us to convert 2d to 3d images using ml...if anyone has any experience in this regard or have done...pls dm me,I would be really grateful I just want to learn and grow as much as possible...u can even dm me if u want to discuss anything regarding ml or ai...lets be friends...

zenith niche
midnight parrot
snow egret
#

Im trying my first competition. do you need to clean up the test data like you did the training data?

dense hazel
dense hazel
snow egret
# dense hazel no you dont need to do that.

I cleaned up all the test data, then i did feature scaling excluding the id and the field i am supposed to be "guessing" . then i split the data not 100% sure i understand what that does. then i did a tuning using LinearRegression followed by fit and predict.

#

I guess the question is what next and how do i know if it worked.

#

im really suprised i got this far 🙂

dense hazel
#

@snow egret you are doing great as A Newbie. Kudos 👏🏻

reef mural
#

Which competition would I learn the most as practice if I am going to build a machine learning model for predicting how likely horses are to win a race in horse racing?

#

And would i learn more from an active or completed competition?

chrome girder
fading berry
#

Hi Everyone! I am starting my journey in ML- AI, starting off with statistics for ML. Would anyone like to join me? It will be healthy competitive way to learn, track and explore.

hearty arch
languid dew
deep trellis
torpid sentinel
#

🌟 Hey everyone! 🌟

I'm thrilled to present my latest notebook: "Gemma, LLM, Kaggle Solution Writeup Agent"! 🚀

Ever wondered how top performers crack the code in Kaggle competitions? Join me as we'll unlock the secrets behind winning solutions, powered by Gemma, Google's Open Model.
We'll unravel the strategies, techniques, and secrets behind those winning submissions. With Gemma we'll dive deep into the heart of data science excellence. 🌐

Throughout the notebook, we'll:

🔍 Deep dive into Kaggle solution write-ups, dissecting the approaches of top performers.
🎓 Explore innovative methods for data generation, few-shot prompting, and fine-tuning LLMs to boost model performance.
💡 Break down complex concepts with real-life examples, making them accessible to everyone interested in data science.
Discover how to upload your fine-tuned models to the Kaggle Model Hub, opening up opportunities for collaboration and knowledge-sharing within the community.

Check out the notebook right here: https://www.kaggle.com/code/ianakoto/gemma-llm-kaggle-solution-writeup-agent

Gemma, LLM, Kaggle Solution Writeup Agent

Explore and run machine learning code with Kaggle Notebooks | Using data from multiple data sources

wary dragon
#

Is there a way to submit a notebook to competition without it consuming GPU quota?

boreal iris
wary dragon
main vigil
#

Somenone can tell me a project with source code that i can put on my resume

indigo river
#

is there a sequence of competition on kaggle that I can follow from beginner to advanced?

#

I think I can start from here

image.png
#

but on leetcode we can sort easy to hard

steep epoch
#

I feel like there should be a pinned message which gives us or guides us as to how to generally approach a competition including doing Research, Data collection etc, it will be really helpful for us beginners!

dense meadow
#

hey, for a notebook competition, once I submit a notebook, I am getting an error that it's not able to find my custom model which I have supplied with the notebook.
I have ran the notebook myself in kaggle and it seems to work there. Any tips, suggestions?

runic rune
maiden sigil
#

Good day! I am an Applied Computer Science student, currently in my second year with a focus on Artificial Intelligence. For our Deep Learning Project, we have to choose a competition on Kaggle where we have to use at least one of the following:

  • MLP
  • CNN
  • RNN
  • Auto-encoder
  • NLP
    As we are very limited in time (only 10 days), they have advised us against using CNN, as training would take too long. So my question is which of the current competitions would you suggest me to choose? Thank you all!
pale gyro
#

Hey everyone !!
I hope everyone is doing great
This is my first ever Kaggle competition, hence I don't have prior experience with respect to submissions. I just wanted to ask how does submission work ? Do we have to upload the model or a pipeline (script) somewhere ? (I could just see an option to upload the notebook) . Apologies if the question is repeated

#

It would be of great help if anyone of you could provide some necessary information about the same

fathom fossil
#

that code on kaggle

#

you can code locally but

#

have to upload the notebook to kaggle

#

in some competitions you can submit the submission.csv ( i think in the playground compeititons

fathom fossil
#

you have a lot more options in non prize comps

maiden sigil
fathom fossil
# maiden sigil Yes, it may be both prize and non prize series!

MLP: Titanic, Prosumers, Home Credit Default Risk, Digit Recognizer, Regression with an Abalone Dataset, Store Sales Time Series Forecasting, Leash Bio, and possibly others.
RNN: Prosumers, Store Sales Time Series Forecasting, BirdCLEF 2024, and possibly others.
Auto-encoder: Predict Energy Behavior of Prosumers, Leash Bio, and potentially others.
NLP: Titanic, PII Detection & Removal from Educational Data, NLP Getting Started, Automated Essay Scoring 2, and Contradictory, My Dear Watson.

you can try these on the above compeitions

maiden sigil
#

Thank you so much!

nova cape
#

anyone know of a time series competition where exogenous data was useless and where the given time series to predict was disgustingly bad?

#

(Thats what the data im working on looks like so im trying to find winning notebooks to see how they went about it )

keen needle
#

Is there a channel for the WiDS Datathon 2024 Challenge #1 for cancer diagnosis

brave wren
fathom fossil
obsidian badge
#

Good day everyone, I have a few questions about my model for the Titanic competition. I extracted the titles of each person and put them as extra features. Then I normalized my data and did PCA. Then I took the 10 best principle components and fit a few models on these. All of my models are performing very badly even after doing extensive grid search with each model. Does anyone have any tips?

hot field
#

what do you mean by badly exactly?

gritty frost
#

I am planning to participate in "Regression with an Abalone Dataset" kaggle competition. Any suggestions and guidance please.

weary moat
#

Hi! I'm participating in my first Kaggle Code Competition, so I'd be greatful if you could help me, a beginner, understand the requirements for an elligible submission.

From what I've read in the Kaggle documentation, it is necessary for the submission notebook to be ran "top to buttom" in less than 9 hours of CPU / GPU runtime. That means that in my submission I should train the model on the training dataset and also predict on the test dataset in less than 9 hours, right? Or I'm getting it all wrong?!

I'm asking this question because I've noticed in this and other competitions' code tabs that there are public inferrence-only notebooks that import model(s) trained elsewhere (uploaded as Kaggle datasets) and use them directly to predict on the test dataset. This shortens the total runtime of those notebooks. Is this kind of notebook allowed to be a final submission? Or is this just a way to avoid exhausting the GPU weekly quota while also allowing one to see how well their predictions perform on the public leaderboard and also making certain notebooks public for the community without revealing too much of the training process used.

If this isn't allowed, then how are my submissions supposed to compete with these sped-up notebooks, with high public scores, especially in the efficiency section of the contest?

Thank you in advance!

stark elm
#

Hi everyone! I'm excited to announce that I'm participating in my first Kaggle Competition, WiDs Datathon Challenge #2! I've just begun my journey into machine learning and modeling, and I'm eager to learn more.

I'm wondering if anyone would be available to chat about the fundamentals of training a model and share some knowledge with me. I believe that by learning and teaching each other, we can all improve our skills together.

If you're interested, please don't hesitate to reach out to me.
Thank you in advance!

zenith niche
#

Hi can anyone like give some idea regarding structure to motion model...like from basics how can I implement it...I have a month time

#

It would be of great help...pls dm me

boreal iris
jovial basaltBOT
#
.paitom has been warned

Reason: Bad word usage

#
.paitom has been banned

Reason: Too many infractions

wind snow
cunning dragon
#

Yes.

#

I am.

coarse vortex
#

Hi, If there is anyone from Kaggle team.
I wonder why can't we make private competitions public again?

pastel juniper
lucid glen
#

Does anyone know I cannot submit my notebook?

Screenshot_2024-05-31_123854.png
pseudo pond
lucid glen
brazen stratus
#

hi

vivid rose
#

Do we know what new competitions are coming up in following months?

vivid rose
#

Hello Kagglers! kaggle

If you are about to start a competition, I offer you to join my team to solve "LMSYS Chatbot Arena Human Preference Prediction". Because more minds are better than one! grandmaster

Apart from solving the LMSYS competition I want to create some kind of short-term "mastermind group". A mastermind group, in the context of building a career in IT, is a peer-to-peer mentoring concept used to help members solve their problems with input and advice from the other group members. The idea is to bring together like-minded individuals who are focused on mutual growth and success within the IT field. halite_green_ship

The activities of the mastermind group will include working together on the LMSYS competition, checking in on each other’s individual projects, providing feedback and help each other grow! woohoo

If you are interested, feel free to DM me. kerneler

shut cosmos
#

@vivid rose interested

jovial basaltBOT
#
archangel7431 has been warned

Reason: Bad word usage

#
archangel7431 has been banned

Reason: Too many infractions

hardy carbon
ocean rune
jade frigate
#

Hello people! My name is paul and I'm currently on going for the RSNA 2024 lumbar spine.... competition, but I need some help. I understand the data given to me by the competition, but I don't get how to format the x_train/test and/or the y_train/test. Basically, I don't know what to do with the data I'm given, I'm stuck with what I should load the model and what the output should be. Please help if you can.

night fable
#

Well here's what I did:
I iterated over the train.csv, fetched the series_id from train_series.csv and loaded the image in the corresponding folder path in a tensor

night fable
past moth
#

You don't necessarily have to reconstruct it as a 3d image though. You might find it easier to treat it similar to video classification

jovial basaltBOT
#
diablo931 has been warned

Reason: Bad word usage

#
diablo931 has been banned

Reason: Too many infractions

night fable
#

im just doing it to learn but probably the best way is to use a pretrained neural network for MRI images

past moth
#

Starting off with a pretrained backbone can save you time for sure, but you will mostly likely still need to do quite a bit of training

old topaz
#

Hi guys I am pretty new to DL. I build MLP and CNN from scratch using numpy, but not sure how to apply them with the various datasets. I want to have a good foundation before building models in the Rsna 2024 lumbar competition, so can someone have any recommended resources for this image competition?

pallid night
#

I am interested in collaborating with a team for some upcoming data science competitions and hackathons. Anyone up?

rigid sonnet
rocky zinc
#

hey , I wanted to ask, is it possible to solve "Titanic - Machine Learning from Disaster" problem with neural network ?

pallid night
#

@rocky zinc You may search on YT there a lot of tutorials over there for the same

rocky zinc
stiff condor
#

I have my LinkedIn set up, but I don't have any ideas on how to apply for internships. I need to join urgently, and I'm unable to find any good opportunities yet. Any suggestions?

pallid night
pallid night
rocky zinc
#

Hope this helps

pallid pebble
#

Where is the channel for Cancer detection competition

trail cave
#

Hi everyone, I’m new here and need some help. Can someone please explain what I need to do for this Skin Cancer competition in a simple way?

  1. What do I need to do? - Develop a program to identify dangerous skin spots from images.
  2. What language and tools should I use? - Use Python and tools like TensorFlow or PyTorch.
  3. Which data should I use? - Use the Training and Test data provided in the competition files or data from outside.

Thanks a lot!

strong karma
#

I'm calculating pAUC_80 using the implementation found on the icic skin cancer competition, but it is not matching up with the outputted score, any ideas on how to fix this? I want to test my models true value without wasting a submission

lusty osprey
ocean rune
strong karma
#

Do you reccomend deep learning?

#

I have tried both but lgbm seems to be the best method

#

Resnet got me terrible scores

lucid valley
#

On tabular data tree boosting sees the way to go

strong karma
#

Has anyone used image data with high score yet?

#

From what I can see most people are using lgbm

true trench
#

its all extremely close to eachother with slightly different params if you look at the open code

past moth
#

It's basically distributed Graduate Student Descent. People grab the best performing publicly available approach, tweak some hyperparams and hope for the best

#

Sometimes one of those hyperparams puts the performance at a new order of magnitude and then becomes part of the baseline for future GSD iterations

strong karma
#

😭 I tried with other solutions and nothing I made compares to the open code

spring idol
gilded drum
#

On July 1, The Global Multimedia Deepfake Detection Challenge 2024 (The Challenge) officially launched. This competition is co-hosted by the INCLUSION·Conference on the Bund and Ant Group, and organized by Ant Digital Technologies.

The Challenge aims to invite participants to develop, test, and further evolve more accurate, effective, and innovative detection models against various types of Deepfake attacks in real-world scenarios, as well as to motivate innovative defense strategies and improve Deepfake recognition accuracy.

https://www.kaggle.com/organizations/inclusionconference-on-the-bund

INCLUSION·Conference on the Bund Profile | Kaggle

International exchange and industrial growth

clear ginkgo
#

can i dm a competitions master/expert, i want guidance on how to get better in competitions...

simple sun
#

Anyone has any idea how i could improve on public score

i working on Leash belka SMILES prediction

sullen frost
#

I get this error while downloading the data via API, any suggestions?

403 - Forbidden - You must accept this competition's rules before you'll be able to download files.

PS: I have already accepted the rules for the competition

sullen frost
#

@lofty bear, @random umbra or any Kaggle Staff can you please help

rocky zinc
sullen frost
simple sun
#

I am interested in collaborating with a team for some upcoming data science competitions and hackathons. Anyone up?

brittle basin
simple sun
runic rune
simple pendant
#

hey

weak jungle
#

Hi

lost talon
runic rune
unborn mason
#

hi i had a question about my work on the current playground series if anyone is around to help me out it would be much appreciated

tiny ginkgo
#

I am wondering why no competitions started in the last month and why there are only five active competitions recently

jovial basaltBOT
#
gabrielpradyumna has been warned

Reason: Spam

stable vector
snow lily
#

Hi ! I'm a first-timer in a Kaggle competition. Getting my head around the skin cancer dataset. Can I ask a stupid question?

#

The AI model we are developing is for an end user to take a mobile photo at home and submit this. So there won't be any other information provided to our model for inference. So is there anything useful to us about all the metadata in the dataset? Surely we just focus on the photos and labels?

bitter lintel
leaden vigil
opal whale
#

Hi, Kaggle community!

My friend Alex and I are developing a tool to make deep neural network training interactive, user-friendly, and deterministic.

Are you tired of the guesswork and never-ending experimentation?
Are you seeking an alternative methodology that makes the process less painful and more understandable?

We'd love to hear from you! Please share your thoughts and help us to upgrade the current methodology.

LINK: https://calendly.com/graybx/30min <<

Calendly
Demo | Graybox - Luigi d'Ovidio
sullen frost
opal whale
# sullen frost Making things deterministic is not at all possible unless you fix seeds. Talking...

The difference between our solution and Weights&Biases is that during training, you can tweak and change hyperparameters, change the architecture, set the learning rate individually, re-initialize dead/low impact neurons, and deny-list/allowlist data samples. This is supported by suggestive neuron-level statistics, so the changes you make are rather informed than guesses. For instance, you can grow the learning capacity of the network (a.k.a adding neurons to a layer) every time it plateaus, thus reaching convergence with overall fewer FLops.

sullen frost
opal whale
ancient shore
#

hello everyone! about 10 months ago I remember seeing some competitions about quantifying uncertainty of predictions. If I'm not mistaken, it was the first episode of a new series of playground competitions. The problem is I can't find them anymore. Does anyone know what happened to them?

runic rune
shut pier
#

hi I just started my first nlp competition about tweet disaster. I saw a bunch of people reached the highest score. I wonder which kind of model did the job for you? Was it deep learning? machine learning? or already trained model?

main storm
#

So I am training a model for skin_cancer detection dataset which has images w.r.t target variable.When I am using any pretrained model such as resnet or efficient net it's giving me below par accuracy of 56% but when I am using a random cnn it's giving 90%? Why

tiny ginkgo
#

Now there are only 4 active competitions which really make me wonder if there is something going now

#

I believe we usually used to have like 7 or 8 active competitions at least

tiny ginkgo
main storm
tiny ginkgo
#

Also, I didn't work on this competition so I don't know good the score 90% is

#

But it is possible that there was data leakage in the second case where you used random cnn which lead to this optimistic score

tiny ginkgo
runic rune
#

at least, the obvious way is to use tabular data

main storm
#

How much time will submision take for this skin cancer detection competition?

outer fog
#

Is there any new competition on kaggle, that we can join

tiny abyss
#

Regression of Used Car Prices seems to be good for newbies.

inland garden
tiny ginkgo
ancient robin
#

Does anyone have information about if DEFCON will launch a CTF competition at Kaggle this year?

vestal sun
#

Is there any ongoing competition?

tiny ginkgo
gritty tapir
#

Is there any course on coursera that can help me get started with the kaggle competitions?

viscid helm
#

How do teams typically share code while developing an analysis and solution for a competition?
Do they use a GItHub repository, or is some Kaggle notebook more typically used?

viscid helm
main storm
#

Anyone working in Amazon ml challenge

grizzled storm
#

Is anyone working on the Store Sales - Time Series Forecasting competition, I am just beginning, and I would like to join

tulip snow
#

@main storm yep I am

sturdy drift
vivid kite
#

Hello folks anyone interested in taking part in any upcoming kaggle competition pls dm me asap

fickle hamlet
vestal sun
#

can anyone help me ? whenever i went for submit it show inference error and got reject

keen violet
#

Hi @vivid kite. Are you still looking for mates for competition. I am in. Which competition are you planning for?

bleak glade
#

I got a position of 2000 in a competetion with 2500 members 😭😭
finna go jump from a tower

honest wigeon
#

Hi everyone,
I'm planning to participate in the "Child Mind Institute — Problematic Internet Use" Competition. This is my first ever Kaggle competition. I'm not sure how to get started tho?

nocturne quarry
#

Hey friends! 🌱👩🏽‍💻👨🏽‍💻

My team and I recently participated in a Climate Risk Research Challenge, and we developed DEWSClim, an early warning system app to help farmers tackle climate risks and improve food security. 💡🌾

We’ve entered the Gemini API Developer Competition, and I need your help! Please vote for our project to help us take DEWSClim from idea to reality. Voting ends September 30th, so your support means everything!

Link to vote: https://ai.google.dev/competition/projects/dewsclim

Thanks so much!

Google AI for Developers
DEWSclim  |  Gemini API Developer Competition  |  Google AI for Dev...

DEWSClim: Smart farming, rich harvests.

zinc geyser
#

when winning competition, do we need to presented our solutions to the hosting company?

zealous atlas
#

Just started my second competition, I noticed on the leader board there is rf_benchmark.csv with a different icon. Does anyone know what this is? I am guessing it means random forest benchmark. Is this an autogennerated prediction?

twilit marlin
#

Hi guys, is anyone interested in teaming up with me for the BrisT1D Blood Glucose Prediction Competition?

vestal sun
#

Feel free to contact me for teaming up remember I'm still a learner

late pewter
#

im a learner as well been doing datascience for about 7 months now im willing to team up if anyone is available

autumn sparrow
#

looking for a team mate for the Child Mind Institute competition to learn more anyone wanna teamup ?!

austere mural
#

I'm Zeineb currently pursuing my master's in data science. I'm still quite new to generative AI, but I've been diving in! I recently worked with LLMs and built a little AI Q&A chatbot for workout and diet recommendations using Mistral 7B, which I finetuned with some scraped data from Quora and Reddit. i also build a facial recognition using ython and deep face

if there an intersting project i would love to be a collaborator 🤗🚀

sullen frost
#

Hello, I’m Sarvesh, a PhD student in AI at IIT Bombay and currently working as a Student Researcher at Google. I am seeking collaborators for various Kaggle challenges and conference competitions, as well as research collaborations.

To collaborate, you should be well-versed in coding, particularly with experience in large language models (LLMs) and frameworks like PyTorch or JAX. For research collaborations, a strong mathematical background is also necessary.

Please note that due to time constraints, I will primarily provide guidance rather than direct involvement. If you meet these criteria and are interested, feel free to get in touch.

ionic echo
#

I was participating in the EEDI competition, any idea why the cell randomly stops running while loading the model?

image.png
safe hound
#

is this normal I cannot use the file upload feature? do i have to use a kaggle notebook vs local dev ?

Screenshot_2024-10-11_at_07.34.12.png
safe hound
#

the kaggle notebook runs out of memory, do i have to pay to use google notebooks to compete?

plush hawk
#

ayo, I solved ARC but I need to wait 9 hours, what do

#

if you ever wanted to know what it feels like to stare at 600k, ask me anything

Screenshot_2024-10-14_at_2.02.08_AM.png
plush hawk
#

8 hours now...

true trench
#

600k?

plush hawk
#

$1,000,000 AUD

#

I'd finally be able to take a break

warm heart
#

??

plush hawk
#

This is my model doing the thing with no gpu accelleration

Correlation_Matrix_for_Train_Dataset.png Distribution_of_Median_House_Value_by_Income_Group_-_Test_Dataset.png Median_Income_vs_Median_House_Value_-_Test_Dataset.png Correlation_Matrix_for_Test_Dataset.png Distribution_of_Median_House_Value_by_Income_Group_-_Train_Dataset.png Median_Income_vs_Median_House_Value_-_Train_Dataset.png
warm heart
#

wow

#

dm

true trench
vast stag
#

I don't understand how to generate my submission.csv file for the competions.

agile dove
ancient cedar
#

Anyone can tell me what is this project based learning what are the requirements to start building anything and how people follow this path?

lilac yew
obtuse bison
#

hello, i am doing my first competition and i have a question. when i joined the competition i created a notebook for it where it's linked to the competition, and i see that there's a code section where people show their EDA and stuff. i know that if you're high on the leaderboard you're probably not showing how you're doing it, so how do you submit without showing others your code? do people have a public notebook for EDA and then submit a different way?

glacial pike
obtuse bison
#

@glacial pike thank you!!

warped heron
warped heron
chilly berry
#

Looking to join a team for this math competition!

pallid night
ancient cedar
south osprey
#

has anyone done the House Prices - Advanced Regression Techniques competition?

warped heron
torpid cargo
#

Hey everyone... im have decent experience in ML and AI but havent ever done any sort of challenges and would like to start with the Kaggle ones, Im not an expert for sure but how to find challenges that are not too easy and not too difficult?

warped heron
torpid cargo
gleaming jetty
craggy star
#

Can we use cohere llms ?

surreal ermine
#

hello,
I am working on a loan approval prediction project and am considering modifying features by removing less important ones or engineering new ones based on their importance. If I apply these feature modifications consistently to the training and test datasets, will my notebook be eligible for acceptance within the competition?

Additionally, as I am new to Kaggle competitions, I would appreciate an overview of the typical roles and responsibilities that participants might take on.

drifting flint
#

Hello everyone...I am learning ML ,but do not have any experience yet.i am new in discord and kaggle..how can I use it to improve my learning and kickstart my career.please guide me about this platform

warped heron
gleaming jetty
craggy star
ripe monolith
#

Hello everyone! Although the WiDS competition has ended for a while now, I’m excited to finally share my work. You can find the details in my GitHub and Kaggle links below. I’d love to hear your feedback!

https://github.com/drkbluescience/WiDS2024_Challenge2_MetastaticDiagnosisRegression
https://www.kaggle.com/code/enisezengin/wids-24c-2-metastaticdiagnosisregression-80-154

Thank you!

GitHub
GitHub - drkbluescience/WiDS2024_Challenge2_MetastaticDiagnosisRegr...

This notebook presents an exploratory data analysis (EDA) and regression modeling approach for the WiDS Datathon 2024 Challenge #2. - drkbluescience/WiDS2024_Challenge2_MetastaticDiagnosisRegression

WiDS'24C_2_MetastaticDiagnosisRegression_80.154

Explore and run machine learning code with Kaggle Notebooks | Using data from multiple data sources

silent thorn
silent thorn
craggy star
silent thorn
craggy star
# silent thorn i have already started pandas

I would still recommend do join that ai club after u register (or even if u not consider registering do install pycharm and there re nice builtin numpy/pandas and other data science courses in the IDE

wraith basin
#

Does anyone know why plagiarism is allowed in kaggle competitions? In every competition you look at, the top 10% in public leaderboard always have the same scores because they've forked the highest scoring solution

#

and this behaviour is even more encouraged because you gain notebook medals when sharing these types of notebooks

true trench
#

yeah, in ARC, the entries on ranks 42-813 (772 entries in total) all have a score of 26 😂
With medals being given out until rank 134, the majority of the medals will go to copycats.

wraith basin
#

rank 42 with 6 days to go aswell, how depressing

fierce marsh
#

Is anyone taking part in Meesho Competition?

lost talon
# wraith basin Does anyone know why plagiarism is allowed in kaggle competitions? In every comp...

The goal of Kaggle competitions is to get the best possible solutions, public sharing during the start and middle of a competition can help accelerate all teams and leads to better learning experiences for participants and higher quality final top solutions.

Your goal should be to learn from what is public and build on it to secure higher results. Usually this process will push a lot of simple "fork and submit" users out of the medal range, but unfortunately that doesn't always happen. It's unfortunate, but the alternatives (eg. blocking public sharing) would be worse for most participants and most hosts.

wraith basin
# lost talon The goal of Kaggle competitions is to get the best possible solutions, public sh...

In that case you should completely rework the medal and rank system. It's a joke that the best way to earn titles is by forking and submitting.

"Usually this process will push a lot of simple "fork and submit" users out of the medal range, but unfortunately that doesn't always happen." - It literally always happens in every competition because people just try to farm medals. In the scenario from the user above, you get pushed to the bottom 50% if you don't beat the highest scoring notebook.

I understand the point about knowledge sharing, but it completely destroys the point of competing and trying to get medals imo

lost talon
# wraith basin In that case you should completely rework the medal and rank system. It's a joke...

If you look at the data (check out the metakaggle dataset) I think you'll find that isn't the case. For one thing, the results from the public leaderboard often don't result in good outcomes on the private leaderboard (due to overfitting etc).

Bronze medals are certainly possible to get with a little luck and forking, silver medals are very rarely awarded this way, and gold never are. To get to master or grandmaster tiers (the ones that actually matter) you need gold medals, which are still very hard to get.

Bronze medals and the expert tier represent effort at Kaggle, but you are right that sometimes people will find easier ways to win things. It doesn't tend to help them though, ultimately a bronze medal on Kaggle is only really valuable if you learned something in the process or can explain your project to a recruiter to get a job. Simply forking won't get you very far in the end. We would love to prevent this sort of behavior but ultimately we are making a tradeoff to support the benefits of sharing.

If you have specific suggestions for changes you'd like to see to the system we are always open to listening to feedback.

runic rune
#

Is anyone’s notebooks being stuck on “Queued” while committing for quite some time ?

lost talon
visual sigil
#

hello guys i have a question , is it possible to import ollama to my external ssd ?? and train my model using my external ssd

grizzled holly
#

anyone has recommendations of EDA handbooks? new guy needs helpyou_got_it_dude

austere mural
#

Heyy guys i'm looking for hackathon to patrticipate in especially in LLMs you know any?

unborn void
true trench
#

@slim plaza

Is scam message he sent in all channel

slim plaza
#

Which other channels? @true trench

crisp marsh
#

Hello guys !! I had a question, who determines what medals we recieve? the organizer or kaggle?

#

I recently got a bronze medal on the long context gemini competition, I wanted to ask how I can improve this, since this competition is not ranking based

runic rune
#

wonder if there's a long queue in the submissions...I have a sub that is still scoring after X hours although I expect it to complete in X-2 hours...and a commit that has been stuck on queue for 15 minutes

heady bane
#

What steps would a parent need to take in order to allow a minor to participate in a contest?

#

my parent filled out a google form but is there anything weshould do after that?

gentle grotto
heady bane
#

or can you just participate

dusk shuttle
#

Hello Everyone. My name is Bukenya Lukman. Reaching out to everyone who can help. I' am a Ugandan student.
Looking for ways to raise tuition such that I can graduate (Ms in Software Engineering and Data Communication) this coming January , 2025.
Any support is highly appreciated. Help me Share with Friends and colleagues
I have explained my story in my GoFundMe campaign here
https://gofund.me/a598102e

#

Any help, even sharing will be highly appreciated

safe dove
#

Hello everyone, i am new to kaggle competition. I am getting started with spaceship competition. Can anyone help, how to handle "nan" values in different columns ? shall that be removed or that also can be trained?

lost torrent
#

Hello, I am setting up a kaggle community competition, but I would like it to have two stages, one stage where participants train their models and submit their methodology and the second stage where participants submit their solutions each week to predict the outcome of games. So the submission and evaluation will be ongoing and live. Can I host such a competition? It seems the "launch checklist" has certain requirements that need to be met, which are not relevant here. Thank you in advance

twin dust
uneven rose
#

Anyone with a Kaggle competition medal, DM me! Let’s team up and try to snag a medal together ASAP.

bitter lintel
lethal canyon
#

Guys I am stuck with "Regression with an Insurance Dataset" competition ,it has 1,200,000 row so when I visualize the data I cannot understand anything so I cannot remove or handle the outliers and this is my first time to deal with such a big data so any suggestions ?

bitter lintel
slate sand
#

hi , how to proceed with the problem ...like to we consider the state data of the policy and treat it like time series data ...and by creating the window do we predict the insurance amount ...can someone help me with this

wheat compass
#

Well, I mind to ask this question. But, if there is anybody who knows kaggle well, could you tell me how I can get prize in the competition after winning?
Payoneer, or paypal, or crypto?
Sorry,

true hollow
#

hi, so i was wondering, what does the score represent in competition leaderboards? is it the accuracy on test data? or something else? or is it up for each competition to define? (if so then what is it for the spaceship titanic competition?)

image.png
midnight garden
#

up to each comp to define

#

on the titanic i think its accuracy

lost talon
sinful latch
#

what can be the possible reason for failed submission in a competition, if the version saving have no error.

#

error: Notebook Threw Exception

#

really cant understand much via explanation given in Faq.

bitter lintel
stone palm
#

Hi @lofty bear
I am very eager to know if there will be season5 for playground series starting Jan 2025.
Can you enlighten us about it ?
Happy holidays and a very happy new year halite

bitter lintel
pallid night
#

is that normal?

Screenshot_2024-12-25_at_1.35.30_AM.png
lucid relic
solemn compass
lucid relic
lucid relic
pallid night
wicked pond
#

Hello,everyone, i am just a rookie in this field. i wanna know how to prepare this item if i just have a little basis. i dont have some high goals ,i just wanna have a basic recognization of this item. Thank u for ur suggestions and advice!

wheat compass
#

Hello, kagglers.
I have one question.
I need to make a news sentiment ananlysis tools.
Now, I have 2 methods, first method is to use open source models such as deberta or roberta and second method is to use HMM(I heard IBM used this method).
Which method should I select?

wicked pond
solemn compass
#

Hii, I have a imbalanced dataset. Target column has a two category. One is 1.5% and second is 98.5%. And dataset shape is (100000,32). There is 1lkh rows. If i use smote then data is increase in high numbers of rows. So, according all of you what should i do ?

surreal onyx
#

Hi @solemn compass . Without smote, in a simple way, try multiply these records by 65.3 times to balance the dataset for testing.

pallid night
#

hello. where can i find datasets that are in demand? looking to contribute

simple dome
#

hello can anyone help
im new in discord
how solve this issue

#

hello

manic otter
#

I'm a 17 year old who has been doing ML projects for about a year now, but don't know how to improve from here and get better at competitions. I am willing to pay masters, grandmasters, or anyone who can prove that they can teach me how to improve my ML skills for an hour of their time.

#

Please dm me if interested

craggy star
manic otter
#

yup

#

you?

craggy star
wheat compass
#

Hi, everybody.
I'm looking for ar related to project where that can detect palne and argument the objects on that.
If there is anybody who knows, please tell me.
Thanks.

torn hazel
#

Hi everyone!
I’m looking to form a team for the WSDM Cup - Multilingual Chatbot Arena Challenge and would like to collaborate with 2-3 motivated individuals. If you're interested in joining forces and taking on this exciting challenge together, feel free to reach out to me via message!😋

ancient cedar
bitter lintel
bitter lintel
# manic otter I'm a 17 year old who has been doing ML projects for about a year now, but don't...

He'll charge you 70 bucks, https://www.amazon.com/Hands-Machine-Learning-Scikit-Learn-TensorFlow/dp/1492032646/ref=sr_1_1?crid=289MDXUG37E9J&dib=eyJ2IjoiMSJ9.xngn_0-9hkeaaBfChRVK4nNsdt7xyp6x3azvSdRWWXQ5xaf9t82Orefwoj8AgZUu4t2WDJfquFcr4sXyM7W8pw.8BCiJN0dTvDwRk52OdIiaaY_ywCCc7ZqYLoqN5LsObg&dib_tag=se&keywords=hands+on+sci-kit+keras&qid=1736525094&sprefix=hands+on+sci-kit+keras%2Caps%2C96&sr=8-1

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow:...

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems

ashen pike
#

hello @everyone I have joined AI mathematical Olympiad and now I am little bit confused about dataset and other files that are showing in side bar, i want to know how start with this problem, where can i find dataset, here are two one is reference showing only test showing 3-5 problem, test set showing i think approximately 10 problem and one pdf contain 10 problem so can anyone guide me about how to start working on this from where can i get dataset and more. kindly help me thanks.

deep charm
thick heron
#

Hi,
A question for "WSDM Cup - Multilingual Chatbot Arena"
Can i pretraining model before create notebook to summit?

#

pretrain dataset not related with competitions

bitter lintel
thick heron
candid wagon
#

Hello! Does anyone have recommended peft install recommendation? I used pip but when importing peft, got this error: ImportError: cannot import name 'GatedRepoError' from 'huggingface_hub.errors' (/usr/local/lib/python3.10/dist-packages/huggingface_hub/errors.py)

candid wagon
#

after trials and erros: huggingface_hub=0.25.0 is compatible with peft=0.14

mystic raptor
#

I am not able to verify my account

#

what should i do ! help

mystic raptor
#

Any Kaggle mod can help?

serene violet
#

I'm participating in a competition that ends on January 18th. Do you know what time does the competition finished? I've looked into it in the Kaggle site but I can't find it. I live in a UTC + 1 time area.

sub.png
bitter lintel
# mystic raptor I am not able to verify my account

I ran into this issue, and it turns out that i had been loggging in with two different email accounts, inadvertently . You know, when asked to sign in, you are given a plethora of ways too: gmail, facebook, email, etc. Signed off and logged in with only one, and things worked smooth.

#

I got this error message too when it was a problem. Click on your PFP when on Kaggle site to see if it's there.

Screenshot_2025-01-13_at_12.46.51_PM.png
royal timber
#

can I join Kaggle competition via this channel?

lost talon
royal timber
#

so this competition gives reward?

bitter lintel
royal timber
fleet shadow
#

Hi everyone I hope all are doing fine, I need suggestion for next simple project for beginners after the famous Titanic. can i have a list for for beginners. Thank you

#

for ML

bitter lintel
fleet shadow
hollow tundra
#

Hi, everybody.
I'm trying to fine tune BERT model with datasets.
If Someone who has similar project please help me.
Thanks.

lyric rune
#

Hello Everyone , I am new to kaggle competitions, i was looking at sample notebooks for https://www.kaggle.com/competitions/ai-mathematical-olympiad-progress-prize-2/overview this competition , i see people loading models from input directory , but i dont see any language models on my directory , is there anything i am missing or do i have to train my own model and upload for inference or can load from huggingface?

AI Mathematical Olympiad - Progress Prize 2

Solve national-level math challenges using artificial intelligence models

hollow tundra
tepid sun
#

Hi, @hollow tundra
I am new here as well and would be inrested in working with other people.

robust inlet
#

Hi, I am new here and am interested in working with others

pure bear
#

Hi! I’m new to this field and have recently completed a course in data science, machine learning, and deep learning. I’m excited to get started and eager to join a team where I can gain real-world experience. Let’s roll!
Looking forward to joining a team if anyone is open to having me on board!

solar hinge
#

@robust inlet @pure bear
Hello
I am also new to kaggle and a beginner in machine learning, let's decide a competition and participate in one, starting with playground competition would be good

gloomy canopy
#

DeepSeek just shocked the AI world by beating ChatGPT & Claude in benchmarks🤯

Everyone's rushing to try this game-changing model that's 30x cheaper than existing solutions.

We’re doing a deep dive into how it’s actually possible!

Date: Tomorrow, Jan 30th, 9 PM IST

Register Now: https://lu.ma/ael5tq70

WhatsApp_Image_2025-01-29_at_4.19.58_PM.jpeg
olive quarry
#

how to download whole checkpoint-9422 directory quickly?

image.png
olive quarry
#

oh, it is kaggle kernels output <user>/<version-name> -p /path/to/dest

verbal latch
sinful latch
#

after saving a version of my notebook, and submit it to competition, can i shutdown my system while notebook is still running for submission??

smoky raft
#

hi

#

can someone help me with starting a competition?

#

i am new kaggle and i wanna do some projects but doesn't know where to start!

pure bear
west badge
craggy star
#

When joining a competition as individual, and when some others join my team, and when there's limit to no. of submissions per day can each teammate can submit individually that no. of times or in total we can submit that no. of times in a day ?

west badge
#

I don't know actually, but I assume limit submissions is per team

#

GPU/TPU quota per person

craggy star
west badge
#

I guess, but there is a limit on added submissions too, so you have to think it carefully

#

or go to learning experience and just go to a team when you can

lucid relic
radiant gull
gloomy canopy
#

🎙️ The Secret to Starting your Podcast in 2025

Hint: You don't need expensive equipment anymore!

Join our FREE workshop and learn how top creators are launching podcasts using just AI.

📅** Thursday, Feb 13 at 9PM IST**

Register Now: https://lu.ma/wlnvyebn?tk=eMLWC6

For AI Resources Join: https://t.me/BuildFastWithAI

Create Your Own Podcast Studio with AI 🎙️ · Zoom · Luma

Want to start a podcast but don't have professional recording equipment? Curious about how AI can be your personal voice artist?
Join us for an exciting…

Telegram
Build Fast With AI - Free AI Resources

Build Fast With AI is a Generative AI focused start-up run by IIT Delhi alumni to deliver cutting-edge Gen AI solutions. --> Latest Gen AI Tools --> Gen AI roadmap & materials --> Workshop & Event Links Website : buildfastwithai.com Contact : @nag2mani

verbal latch
wide pulsar
#

Any new competitions coming soon?

pure bear
azure loom
glad forge
#

Hi, everyone! I am new here. I want to participate in March Machine Learning Mania 2025 competition. Can I know what to do in this. I am first time participating in the competition

glad forge
#

Hey, anyone interested in march mania competition

cunning birch
#

i am

cunning birch
cloud bison
obtuse night
verbal latch
left plinth
#

Hey everyone. I'm looking for people to learn, discuss and create together!

pale mango
#

surely the top three people in a competition having the exact same score means something is off

solar hinge
earnest kite
snow anvil
#

hello @everyone, My query is regarding WSDM Cup I need your serious help, I can't tolerate this. Whenever I am trying to train any LLM in Kaggle. the kernel just thrashes me an OOM error. This is really frustrating, I am seeing people using the same configuration but they are able to train their model.

I know someone will say, you should lower the batch size, etc etc. But even with reduced batch size the error still persists. Please help out or join my sorrow if you have experienced the same


class Config:
    # Model
    model = "unsloth/gemma-2-9b-it-bnb-4bit"
    
    # Data params
    max_len = 2048
    per_device_train_batch_size = 2
    per_device_eval_batch_size = 4

    # Training params
    gradient_accumulation_steps = 2
    n_epochs = 1
    freeze_layers_cnt = 15
    lr = 3e-5
    warmup_steps = 20
    optimizer = "adamw_8bit"
    
    # LORA Params
    lora_r = 16
    lora_alpha = 32
    lora_dropout = 5e-2
    lora_bias = "none"
    seed = 42
image.png
steady estuary
#

Hello guys!

dusky crescent
#

hi guys my first ever submission on kaggle, is this normal running time ?

image.png
simple garnet
#

Try restarting the kernel

sinful latch
#

In lux-ai S3, We can only have two submission notebooks running at any time for competing with other agents, how do i decide which two those notebooks should be, right now it disabled the old ones, and run only last 2.

#

these bottom 2 submissions are disabled

image.png
molten radish
#

Hey, i need help

We have long conveyor like 5 kms long
In which there are idlers around 4 to 5k
We divided conveyor section with imaginary line let's say every 20m , in that there are around 10-12 idlers of same dimensions.

We have normal data that is after replacement of idlers, and have abnormal data that is before replacement , we have dataset in the form of real positive fft that is each row contains list with 5k integers

If i train 1dcnn based auto encoder or vae it works section wise, like I can see higher reconstruction error in abnormal data. But it is impossible to create model for every section it will be computationally very expensive. I want single model that will work entire conveyor, but when I combine all data and train then it won't generalise well.

Also I tried extracting statistical features like kurtosis , skewness etc and trained dense vae but no luck what can I do ?

Note: i can see abnormality in normal data too. Even after cleaning it becomes more sensitive to normal data as well tell me better approach if you have any experience related to similer problem

odd shadow
bitter lintel
molten radish
bitter lintel
molten radish
# bitter lintel Are you using autoencoding to reduce the amounts of dimensions you are processin...

i tried AE, VAE using 1dcnn data, also tried supervised way like labeling 1 for normal and 0 for abnormal, tried pretained computer vision models by converting list into imagess mapping each frequency amplitude with different color, tried regression approach by assigning one target variable and predicting error, not much luck it get worst as we combine multiple idlers, but thing is we want single model to work all over

bitter lintel
molten radish
bitter lintel
molten radish
bitter lintel
#

Are you using LSTM layers in your processing NN?

molten radish
bitter lintel
#

But LSTM are designed to excel at spatiotemporal processing, such as say FFT bin sequences

#

Its kind of their Raison D'etre

#

It would help if you clearly stated your problem state, desired solution, and approach to it

molten radish
molten radish
bitter lintel
#

Again, rather than dealing with absolute values, use my method and values become relative to a norm. You can even change the size of windows (bins) to establish that norm.

#

It doesn't introduce artifacts (autoencoders do), nor additional computations (except for processing the data)

#

Outliers will tend towards 1, or if directionality is taken into account, -1 too. Then you can do whatever you want with data

molten radish
bitter lintel
#

Here, a sample application w/ a trading bot I've been developing: metrics['open_high_dist'] = (ohlcv_data['High'] - ohlcv_data['Open']) / (ohlcv_data['High'])

molten radish
#

in some im getting 66% rate and high profit too

#

this is the sample

image.png
bitter lintel
#

I'm shooting for a system that's mid point, of predicted highs and lows, always falls within band of target high and low candlestick band. And then figuring how low I can go (buying) from predicted midpoint, and how high (selling). A system that, while not making maximum profit, always makes one 😁

#

A bullish bear 😂

molten radish
#

also u can pretrain selfsupervised model on lots of stock if u have gpu memory like 80gb

#

i created smaller one sometimes it consistent and sometimes it dont, so i made losses too

bitter lintel
molten radish
bitter lintel
#

getting close to training the system this week. my categories

Screenshot_2025-03-02_at_6.17.11_AM.png
molten radish
#

See this one rsi and rsi anomaly when anomaly score high and rsi low buy signal

#

When rsi high and anomaly score high buy put signal like that

molten radish
bitter lintel
# molten radish damn crazy

Plans to train multiple NNets and than Ensemble Method style, combine there outputs when predicting for an asset. At least, that's the gist.

molten radish
bitter lintel
molten radish
bitter lintel
#

Are you using monolithic Neural Nets. A single net. With one big massive input layer? Hoping something helpful comes out?

molten radish
#

also for cross verification i created 50 small models which will follow iid condition of clt, using variable features in each models, to get forecasting closeer to expected true value

bitter lintel
#

Okay, let's imagine you have 13 weeks of data. You could have multiple input layers. One processes 2 weeks worth of dat, one 3, then 5, then 8, then 13. Seprately at first. Then those "channels" , you concatenate them and process that. Each would learn about different influences. Short, medium and long term

molten radish
#

but not consistent

#

different windows

bitter lintel
#

But windows are all part of same Net. Just in beginning of pipeline, they are first processed in parallel, not toghether. Are you doing this. In Keras, you'd make use of a Functional Model

molten radish
#

my forcasting model will surpass if i get order book data but it is impossible to get also it is costly

bitter lintel
molten radish
bitter lintel
molten radish
#

thing is i need order book data which is costly like millions

bitter lintel
#

I'm trying to build a predictive tool, that each week sets up buy-sell pairs, before the week has started

molten radish
sinful latch
#

anyone here participating in CIBMTR competition, need to ask a query about, prediction submission.
I have trained model to predict 'efs' and 'efs_time', how do i calculate risk scores for prediction col.

molten radish
cunning scarab
#

It provides historical data

chilly portal
chilly portal
#

but version of notebook run and saved successfully I need some help in this problem

image.png
sinful latch
chilly portal
sinful latch
# chilly portal What do you advise me to do?

preliminary checks, empty old stuff from gpu memory(not needed anymore), when not needed, use T4x2 for more availability of vram.
at last check at what step your vram is booming, is it preprocessing, inference. just divide this process in multiple steps. it will increase notebook time, but will get the job done.
PS: make sure you are using both T4-gpu

sinful latch
#

what is the submission file size limit in competitions?

hard kestrel
#

is this the right channel to ask questions/help for my competitions?

pallid night
#

Any latest updates?

clever anchor
craggy star
glass shoal
granite valley
#

Quick question on notebook internet requirements: if I have to install something not present in the stock environment (eg some more torch modules), how will that qualify for no internet competitions?

bitter lintel
#

I think, no internet connection refers to a running program. I think when running it, you can install modules from multiple sources. For instance, running a routine that makes a pip install call. I think no calls to Internet mean none after setup and program running. Checking with chatGPT now:

#

Yeah, you’re mostly on point.

Here’s how it works on Kaggle competitions:

✅ “No internet connections” generally means:
• Once your code is running in the Kaggle environment (during training or inference), it cannot access the internet—no external API calls, no downloading files from the web, etc.
• Security sandbox – They want to ensure your solution is self-contained and reproducible.

✅ BUT Kaggle does allow certain things beforehand:
• You can install Python libraries via !pip install or !apt-get install in the Kaggle Notebook cells before running your actual solution, as long as the packages are hosted on PyPI or Ubuntu repositories.
• You can upload custom .whl, .tar.gz, or .zip packages as Kaggle “Dataset” files and load them locally.

So, if Kaggle doesn’t have a package pre-installed:
1. Try doing !pip install package_name in a notebook cell (works offline since Kaggle mirrors some PyPI packages internally).
2. If it’s a custom/private or less common package, download the library manually on your machine, zip it up, and upload it to Kaggle as a dataset. Then, you can do sys.path.append("/kaggle/input/your-uploaded-library-folder") and import it directly.

#

💡 TL;DR:
• Kaggle blocks network access during runtime, but you can still prep your environment with installs and uploads before execution.

granite valley
#

Thanks for the help! I ended up uploading the wheel files as a dataset and all is well now, submission worked fine

granite valley
#

One more question, when making a submission is run time deducted from my total allocated GPU time?

lucid relic
radiant flame
#

does anyone have space to add me to their team, i am kinda new to this, and i dont understand what to do

cosmic crest
radiant flame
granite valley
#

Competition specific question already asked in #stanford-rna-3d-folding but I'm asking here as well since I think this can apply to all competitions:

Can I use a separate notebook to build my dataset to avoid reducing my available training time? Given I make the result public

craggy star
pallid night
#

Can anyone please help me make a model that is over 90% accurate with this data? I am having no more than 80% accuracy. Is there anyone who can help me?

📎 mnist_train.csv
red marsh
peak seal
#

overfitted, run with less epoch's

pallid night
pallid night
#

With this classification data, my model is performing terribly like I only scored 0.0932 in the Kaggle competition. RandomForestClassifier gives me the highest score but I am stuggling to have a better score like those who are at the top of the leaderboard

📎 train.csv image.png
#

Can anyone help me to do well?

lucid relic
craggy star
#

No other national aio doing even practice contests/comps at kaggle BD is great

half lantern
#

What imbalanced class methods do you guys prefer for maximizing AUC? I have tried SMOTE and NearMiss, and also used scale_pos_weight in Xgboost

true trench
#

typically we can balance the test set such that it more closely resembles the training set in terms of distribution. This then gives us a better AUC

bitter lintel
#

Hopefully, ARC-AGI competition comes back soon. They said it would launch Q1 2025, but that's about to end in 5 days. 😅

craggy star
#

From a long time 🤧

lucid relic
craggy star
bitter lintel
rigid light
#

Please, I was recently working on a 'fraud detection' dataset and I needed to fill in the missing values for a qualitative variable: my 'Devise_used' column had 50100 rows per

#

s'il vous plait je travaillais récemment sut un dataset 'fraud détection' et il fallait compléter les valeurs manquantes pour une variable qualitative : ma colonne 'Devise_used' avait 50100 lignes par contre le nombre de valeur abérante est de 2498 je ne sais pas si il faut compléter par le mode ou juste laisser les valeurs aberantes comme valeurs significatives puisque je me dit que forcement c'est une colonne très importante

ocean elbow
craggy star
#

Did the official imagenet comp get discontinued after 2017 ?

red ore
#

hello

inner yew
#

anyone want to do Image Matching 2025?

pallid night
inner yew
craggy star
pure ruin
#

Yeah

lusty marsh
#

Hey, anyone interested in ML competition

left wraith
#

Today I can't get any kind of mail for day 5 tasks and assignments

sacred stone
pallid night
#

i am interested

strange bridge
#

I'm interested

pure glacier
#

I'm interested

earnest kite
grizzled tartan
#

@ Divine Tree Even me have not gotten any email too

open cape
#

🚨 Final Call – Capstone Team Form 🚨

Hey everyone!
This is the last reminder to fill out the form if you're interested in joining my Capstone project team focused on AI/ML-based real-world solutions.

🗓️ Deadline: April 9, 2025 - 11:59PM (EST)
📄 Form Link: https://docs.google.com/forms/d/1UKhtfdgH8Z1CkGXXRIsQ2uaDYC5WBJiXhshn6fj3U4g

Looking forward to connecting with like-minded folks!
📧 Reach out if you have questions: snkp.globalcollege@gmail.com

Google Docs
Capstone Project-(generative AI-Intensive Course)

Hi! I'm Shashank Pandey, forming a team of 3–4 members for our Capstone project focused on AI/ML-based end-to-end solutions in domains like finance, healthcare, and supply chain.
If you’re interested in collaborating, please fill out this quick form to introduce yourself.
⏳ Deadline: April 9th 2025
📬 I’ll reach out to selected individ...

wispy path
#

hello, i have uploaded this notebook to the competition 'Stock_Research_With_LangChain_Gemini' could you confirm you got it> it's my first competition and i just click on 'Submit to competition'

chilly narwhal
#

🚨 URGENT HELP NEEDED: Rossmann Store Sales Kaggle Final Project

Hey everyone 👋 I'm a student working on my final project using the Rossmann Store Sales dataset. I need a Google Colab notebook that gets me into the top 5% on the private leaderboard.

🎯 Target: Private Score < 0.11700
💻 Must be:

  • Fully working Colab notebook (clean & optimized)
  • Feature engineered with lags, rolling means, holidays, etc.
  • Optuna hyperparameter tuning
  • Final model: LightGBM or ensemble (LGBM + XGB + CatBoost)
  • No bugs / ready to submit
    🕒 Deadline: Today
    💸 Willing to pay for the right notebook

Please DM me if you have a strong solution or are available to help 🙏

prime latch
#

Hey guys,

my teammate is'nt able to join kaggle competition. it says as shown in the image. can somone please help? he did register and attend the webinars and can also access assignment notebooks but yet he isnt able to join the competition. can someone please help on what he can do. Thanks.

WhatsApp_Image_2025-04-16_at_15.12.46.jpeg
dense pebble
#

Completed and submitted my project - AI Meta Data Generator for images

Here is the step by step guide and blog of my project

Video - https://www.youtube.com/watch?v=iReYY5vYzqQ
Blog - https://yaitoolbox.com/ai-driven-art-metadata-generation-build-your-free-model-on-kaggle/

YouTube
YAi Tool Box
Kaggle Generative AI Competition | Build AI Model For Free

AI-Driven Art Metadata Generation: Build Your Free Model On Kaggle

Meta Data Generator Notebook : https://www.kaggle.com/code/veeramallavijaigopal/ai-metadata-generator-for-images
To Explore Kaggle - https://kaggle.com
Google AI Studio - https://aistudio.google.com/
Step by Step Guide - https://yaitoolbox.com/ai-driven-art-metadata-generation-b...

▶ Play video
YAi Tool Box
yaitoolbox.com
AI-Driven Art Metadata Generation: Build Your Free Model On Kaggle ...

AI-Driven Art Metadata Generation: Build Your Free Model on Kaggle. Step-by-step guide to creating your model with Gemini AI for art sales

void gorge
#

can i submit the collab notebook link rather than kaggle notebook

void gorge
dense pebble
craggy star
#

that code block decorator style how did u do that i couldnt find any such option

dense pebble
#

Insert the image in the napkin and select the image to get decorative option

craggy star
#

I see too tedious

dense pebble
#

Today generative ai capstone projects will be announced by kaggle officials @ 9:30 pm IST

#

Good luck every one

olive quarry
#

if I want to use C/C++ in a competition I should build C/C++ code in the notebook or usage pre-built binaries is possible?

half yarrow
#

hello there i am new to competetions is it alright if i discuss any ways to solve a particular competetion or is it against the rules?

#

currently i am working on llm classification finetuning and i have some doubts as i am a beginner.
can i ask those here?

craggy star
half yarrow
bitter lintel
verbal latch
#

Hello people !

I have a general question, do you usually use the Kaggle notebook ? or do you rather use google collab or any other platform ?

Any tips for organizing notebooks when starting a competition ?

Am a beginner and I find it hard to resume my work whenever i pause it (have to re-run all the cells etc)

Thanks Tasty

foggy portal
#

Should I encode categorical features of my dataset into numerical features and then standard scale those too or should I leave the categorical features out?

foggy portal
#

Its my first competition as well

kind moat
foggy portal
kind moat
bitter lintel
# foggy portal would u mind telling me what a dependent variable is?

In an experiment, it is a variable that is kept constant. Independent variables are changed for each instance of an experiment. It's a term from scientific research methodology. The goal is to see the effect of Independent variable variations on a constant (dependent variable).

bitter lintel
turbid mortar
calm oak
#

Hi everyone! I'm new to Kaggle and excited to start participating in competitions. Could anyone share what skills are most important for getting started and doing well? I'm learning Python, data analysis, and basic ML. Any advice or beginner-friendly tips would really help thanks

bitter lintel
calm oak
bitter lintel
#

They're good for "learning to swim"

calm oak
bitter lintel
#

Both books will allow you to get deep, w/o needing a deep understanding b4 hand, if that is your question

#

So to answer your question, yes

calm oak
#

thanks
i also wanted to ask about the difference between tensorflow and pytorch,

bitter lintel
calm oak
bitter lintel
#

no one is b etter when it coes to ANNs

#

maybe this will help

Screenshot_2025-05-04_at_6.55.54_PM.png
#

so @calm oak , how did you get into AI/ML

calm oak
bitter lintel
calm oak
#

right now i am learning pytorh

bitter lintel
#

the rest is easy

bitter lintel
#

As in, you could code in C++, or Python, but princiles remain the same

calm oak
bitter lintel
calm oak
olive quarry
undone granite
#

is there any famous competition about video ai in recent years?

vague snow
foggy portal
#

General ML question: If i have 2 columns of categorical data, would it make sense to one hot encode both of those columns? I'm asking cus now there will be a bunch of columns with 0s and 1s for the features of both columns and im wondering if that would be enough to differentiate them

merry tulip
bitter lintel
clever shadow
#

Is anyone doing stand form rna folding comp this year?

tidal aspen
clever shadow
#

Ive made some progress

#

I’ll dm you

lavish quest
#

Hello. I have to finish a project which is a competition on Kaggle from 2019. The goal of the project is to classify 5 stages of an eye disease. I've already got a working code that scores 0.88 qwk. Now I have to use different color spaces and convert the images from the dataset to 6 channel colorspace from the traditional 3 channel RGB. I have tried this and got a result 0.59qwk. I am thinking I do something wrong. The goal is to reach 0.90+ qwk. Is anyone interested in helping me with this task?

idle timber
#

hello. I am trying to take part in OpenAI to Z Challenge. I am unable to verify my kaggle account using persona. It failed and now says to contact support but i haven't received a response from them. Any help?? Also I don't see an option to retry either.

deft hemlock
#

Hi Everyone

I’m putting together a high-impact AI project — AgriIntel — a Smart Farming Assistant built to solve real problems faced by small-scale farmers like my own family.

It’s not just a demo or a college project. AgriIntel is being developed as a serious portfolio product to help all team members land strong remote roles in ML/Data Science by showcasing real-world impact, engineering ability, and product thinking.

We're building:
🌾 Crop Recommendation Engine (Soil, pH, Rainfall, etc.)
🧠 Crop Price Forecasting using Time Series (Prophet/LSTM)
🌿 Leaf Disease Detection using Computer Vision (YOLOv5 or MobileNet)
🗣️ Hindi Voice Assistant (Whisper + gTTS)
📊 Insightful Dashboard (Streamlit/React)

This is your chance to:
✅ Build a project recruiters will ask about
✅ Add something unique to your GitHub & resume
✅ Work in a sprint-style, outcome-driven team
✅ Contribute to something that creates real-world impact

🔗 Full project pitch & roadmap:
https://docs.google.com/document/d/1bY6L_xb5gjAUrqiewkz6BcTMldzJd-kZvS6dn1lNazY/edit?usp=sharing

If you're serious about standing out and want to be part of something meaningful, drop me a message. Let’s build AgriIntel together 🌱

Best,
Dinesh Kumar
🔗 LinkedIn: https://www.linkedin.com/in/dinesh-kumar-775575222/

Join_AgriIntels_AI_Farming_Revolution.png
craggy star
bitter lintel
craggy star
foggy portal
#

@bitter lintel hey super sorry for the tag, its actually my first competition. I'm really struggling to make a submission cus the submission keeps showing me an error. My submission.csv file looks the exact same as the sample submission i believe. The values in there are not good at all, but i don't think there's something error or anything in there. Would you mind helping me out a bit to get a submission score of some kind?

bitter lintel
# foggy portal <@817415799261888522> hey super sorry for the tag, its actually my first competi...

Hard to tell. There could be a slight error in your submitted file itself, related to formatting. A missing comma, colon, bracket, or key. Make sure for one particular answer that makes part of your submission, it is well formatted. Also between the answers. Check the csv file you are generating to verify. BTW, you may want to code a python class that generates the csv itself from your predicted data, consistently, and ensures that all answers are always correctly formatted, individually and across each other. Basically, it can create the sample submission to a tee, if required to.

#

Or maybe, you just aren't submitting correctly (not the submission itself, but code that submits the submission). Check that out too. Search "Code" for competition to find examples of this (usually you can find a few examples; use search tool i necessary). Anyway, best advice I can give. And you needn't apologize for asking for help. That's the whole point, amongst others, of this server. Good luck.

foggy portal
#

and not in the code itself

bitter lintel
# foggy portal and not in the code itself

I was giving you two things to check for, but yes, it's probably not the code itself. If you want to verify this, if you can run the code itself in a notebook, WITHOUT the part that handles submitting it, you can assure yourself that this is NOT the problem. However it may be formatting, and a few others. Here's a list of things to check up on:

#
  1. Wrong file format
    You submitted a file with the wrong format (e.g., wrong extension or structure).
    Fix: Match the sample submission exactly, including file extension.
    2. Invalid or missing columns
    Your submission is missing required columns or has incorrect column names.
    Fix: Use the exact column names provided in the sample submission file.
    3. Wrong number of rows
    Your submission has more or fewer rows than expected.
    Fix: Make sure there’s exactly one row per required ID.
    4. Non-numeric values where numbers are expected
    You have invalid values like “NaN”, “undefined”, or text in numeric columns.
    Fix: Ensure all prediction values are valid numbers.
    5. Duplicate IDs or malformed rows
    Your file contains duplicate or missing IDs, or rows that aren’t formatted properly.
    Fix: Ensure all IDs are present, unique, and ordered as required.
    6. Wrong delimiter or encoding
    Your file uses the wrong delimiter (e.g., semicolon instead of comma) or incorrect encoding.
    Fix: Save your file as UTF-8 encoded CSV using commas as separators.
    7. For code competitions: file too large or timeouts
    Your notebook exceeds allowed time or memory limits.
    Fix: Optimize your code and reduce resource usage.
foggy portal
foggy portal
foggy portal
#

I also put my submission file and and sample submission file on llms to check if there are any minor differences in anything, couldn't find any

foggy portal
#

Okay I think I found it, the data on my produced csv are all float 32 type of data whereas on sample submission its float 64 type of data. That could be the problem right? @bitter lintel

foggy portal
#

Nvm that didn't work

bitter lintel
bitter lintel
# foggy portal Nvm that didn't work

I don't know what to tell you. Try making a submission with dummy values. Create and use a python class that uses random values for answers, and creates your csv. You can then use it to generate a submission that includes your predicted values. As i said b4.

foggy portal
foggy portal
#

It's just frustrating cus i've made 7 attempts so far, it keeps showing this error

foggy portal
#

Sorry for pings again @bitter lintel are negative values in my predictions a problem? Since I think there are logs involved in the scoring criteria. If it somewhere takes in a negative log that could result in an error. Do you think that's possible? What to do in that case

forest thistle
#

@gloomy zodiac please check DM

cedar patrol
#

https://docs.google.com/forms/d/e/1FAIpQLSf-oaubYrc8SkKUxRuGg9oKWsvW42bI8_j9IcRvh4k1fJpjLg/viewform?usp=dialog

i am working on a major project..
i need your suggestion for proressing further..

it will hardly take 10 seconds..

Google Docs
Survey for NutriAi

NutriAI — Your FREE AI-Powered Nutrition & Calorie Tracker 🍎
Are you someone who wants to eat healthier, lose/gain weight, or just understand what’s on your plate?
We’re building NutriAI, a completely FREE and intelligent calorie tracking app designed for everyone—from fitness beginners to health enthusiasts.

What makes NutriAI speci...

craggy star
#

Opensource project ?

cedar patrol
#

thnx

craggy star
cedar patrol
craggy star
#

Bro 💀

cedar patrol
#

specially all those apps are.. much for western food..

craggy star
craggy star
cedar patrol
#

and when your plates are fked with 7 differnet items...
calculation its calorie based on volume and ingredient..
bro..
do u really thing there is no scope..

craggy star
#

Btw they might be for Western but they r valued n their MRR is in millions just in few months

#

-# so there's definitely scope they proving (maybe time for Asian food now)...

cedar patrol
craggy star
# cedar patrol yhh

I was trying to find the guy's traces in GitHub couldn't find anything related to him on GitHub 💀

cedar patrol
#

someone vanished him..

craggy star
clever shadow
opal smelt
#

Hello, I am new to Data Science and I want to learn it. Could you please help me with that?

bitter lintel
opal smelt
tropic hull
#

Hello

sly junco
#

heyo oyo

jolly otter
#

Hi I'm trying to use Qwen3 with vLLM, can anyone share a starter notebook for this?

hushed edge
#

why is my name not showing up on the leaderboard? I thought kaggle would have auto-selected my submissions?

Screenshot_2025-06-10_at_5.35.12_AM.png
silk sable
#

I tried everything I could to debug it. Is there anything else I can do? Any help would be greatly appreciated.

Screenshot_20250610_225501.jpg Screenshot_20250610_225511.jpg
true trench
#

@zealous terrace not sure if you are right person to ping, but maybe remove this user, because he post scam

#

He post this in all channel

high nimbus
bitter lintel
high nimbus
dry niche
#

hlo bro can anyone suggest me some video resources to learn mlops for free

pallid night
primal perch
#

hello , is there like a chill discord group of beginers who have started doing KAGGLE?

dull lion
#

This one

gentle grotto
#

me when bot

raven shell
#

Hola

high nimbus
#

🚀 Team Formation for DRW - Crypto Market Prediction 🧠📈

Hey everyone!
I’m putting together a strong, hard-working team for the DRW Crypto Market Prediction competition. The goal is clear: at least a medal, ideally more.

Looking for folks who are:
• serious about pushing for a top spot
• ready to share ideas and commit to regular iterations

If you’re motivated, eager to participate, and would like to be a part of the team, I’d love to see a starter repeater notebook with valuable insights. It’s totally fine if it’s public or builds on others’ work — but please don’t just copy-paste someone else’s notebook. What matters most is that I can clearly see your own thought process, effort, and dedication to the competition.

If that sounds like you, feel free to DM me or reply here. Let’s build something sharp together. 💪

#

Still 1-2 slots available. Going to finalize the team before 25.06.
After that hard joint work towards the medal 💪 💪 💪

keen raven
#

was up guys

kind berry
#

https://vision.hack2skill.com/event/bah2025?utm_source=hack2skill&utm_medium=homepage
hello everyone!
this is hackathon by isro! ( Indian Space Research Organisation)

Bharatiya Antariksh Hackathon 2025
Bharatiya Antariksh Hackathon 2025 | Hack2skill

Welcome to the Bharatiya Antariksh Hackathon 2025 a nationwide innovation challenge organised by the Indian Space Research Organisation ISRO with Hack2skill as the innovation partner We are calling upon undergraduates graduates postgraduates and PhD scholars from across India to bring their creativity passion and problem solving skills...

unkempt fulcrum
silk sable
unkempt fulcrum
silk sable
boreal cargo
#

what do they mean by offline-first

left stratus
#

🚨 Collab Call: Gemma 3n + Education Hackers Wanted!

Hey folks — I’m Apreddy, solo builder of Lab2Life — a global hands-on science learning platform that just shook up Bolt Hackathon 2025. Now, I'm extending it to Gemma 3n (Kaggle Hackathon) to bring offline, privacy-first, multimodal science learning to underserved kids around the world.

Looking for 1 solid collaborator who’s got:
✅ Experience running Gemma 3n locally (preferably with Kaggle or on-device setup)
✅ Knowledge of multimodal AI (images, voice, text fusion)
✅ Comfort with Edge deployment, privacy-first AI, or education tech
✅ Bonus if you know lightweight backend/devops for packaging

💼 What You’ll Do:
Help build and demo a voice+vision AI Lab Assistant powered by Gemma 3n
Assist with integration (voice prompts, object recognition from webcam, etc.)

🎯 Goal: Build something powerful. Ethical. Fundable.
DM me or drop your GitHub/Kaggle/Portfolio if you're game.
Let’s bring science to every child — even offline.

hushed edge
#

i made these submissions but I forgot to select them, but I thought it would be fine because of auto-selection. However, my name does not show up on the leaderboard
why is my name not showing up on the leaderboard? I thought kaggle would have auto-selected my submissions?

Screenshot_2025-07-01_at_2.26.54_AM.png
craggy star
#

If u don't mind just curious the production implications of Gemma 3n based apps are the apps would be 2-3gb download size and is it gonna be usual ?

thorny bolt
patent sequoia
#

Hi, is anyone here also doing the DRW crypto competition? Please feel free to message me if you are. We can discuss some strategies and maybe ways to improve our model.

cold holly
patent sequoia
atomic lion
#

Hello all! I am free too. I have tools but not team.

patent sequoia
atomic lion
#

This Gemma Seems very good. The competition is apply the models in real world, right? Don´t know exactly the area that i probably will go.

Based on what i have seen on Google site, the best shot is science. Easy to integrate Gemma to chemistry or Proteins. Most of real world stuffs are made for scientists. And they always use Python.

atomic lion
atomic lion
atomic lion
atomic lion
soft wigeon
#

I'm curious, how fast is Gemma 3n running in your devices? In my Pixel 8, it takes almost a minute to generate an answer from the input image

craggy star
#

btw cant access gemma3n from ai studio

soft wigeon
#

I downloaded it from Hugging Face

rich sparrow
patent sequoia
eager obsidian
#

BTW, what's the point of cheating in the DRW competition? Do these guys actually think that {insert the hedge fund they're dreaming about} will seriously consider their application?

keen raven
#

Some high schoolers use it as an extracurricular

#

And get into really good schools like Berkeley through the method

craggy star
lucid valley
craggy star
paper wedge
#

Can anyone help me with the submission of CMI detect behaviour competition actually I'm unable to generate the necessary parquet file from the predict function and hence unable to make the submission

boreal cargo
#

has anyone been able to run gemma3n on a Pi?

signal dew
#

I have submitted a notebook for the MAP - Charting Student Math Misunderstandings competition but its been running since 35 mins lol
Is this normal ?

craggy star
#

It's normal on avg it takes like few hrs

signal dew
# craggy star Yes

thanks!
I feel we understand a lot when we go through others solutions as well

turbid mortar
#

I wanna reach to Expert at kaggle can anyone help me

analog cove
signal dew
analog cove
compact moth
atomic lion
atomic lion
# atomic lion https://ai.google.dev/gemini-api/docs/embeddings?utm_source=gais&utm_medium=emai...

The new Gemini Embedding model is now live in the Gemini API. Start building better Retrieval-Augmented Generation (RAG), classification, and search today.

It’s our most powerful and versatile text embedding model ever, achieving top-ranking scores on the Massive Text Embedding Benchmark (MTEB) leaderboard, and is priced at $0.15 USD per million tokens. Gemini Embedding supports 100+ languages and has a controllable embedding size.

You can generate text embeddings by using the embed_content method:

from google import genai

client = genai.Client()

result = client.models.embed_content(

model="gemini-embedding-001",

contents="What is the meaning of life?"

)

print(result.embeddings)

atomic lion
# boreal cargo has anyone been able to run gemma3n on a Pi?

gemma even on PC is tooo slow...We need raspberry PI with super power. haha. May on huggingfaces have someone that changed it. I have no idea why they are asking fro mobile. A Embeed system that can run these kind of stuffs the price si skyrocket...

inland hound
boreal cargo
atomic lion
atomic lion
rapid thicket
#

Hi all!
I’m forming a team for the RSNA Intracranial Aneurysm Detection competition.

🔹 Skills we’ll use: Python, TensorFlow, Google Colab (AWS optional for architecture).
🔹 Goal: collaborate on data preprocessing, modeling, and optimization.

If interested, drop a 👍 or DM me! 🚀

earnest dust
#

Hi, I am.new to kagglr had a question regarding the submission criteria. What does it mean that my submission has to have a runtime less than 11 hours? Does it mean that the entire 2500 inferences should be completed in less than that?

And also the no internet rule. If my code relies on some packages like nibabel or idk torch then how does the no internet rule work.

Thank you for your help

feral creek
sage marsh
# earnest dust Hi, I am.new to kagglr had a question regarding the submission criteria. What do...

Yes, the entire pipeline — including all 2,500 inferences — must complete within the 11-hour runtime limit.

Regarding the no-internet rule: any form of external access (e.g., API calls, cloud-based data fetching, or live searches) is strictly prohibited. To enforce this, internet connectivity is disabled during runtime, so all required packages and data must be pre-installed and bundled with your submission

earnest dust
#

Thank you for the answersss

maiden sedge
#

Looking for any experieince experts for medical AIML project to work on a paper for ICML

runic rune
#

Is anyone getting "notebook threw exception" when using kagglehub ?

#

I suffered multiple "notebook threw exception" errors today due to this and it was fixed by hardcoding the competition directory rather than using kagglehub.competition_download 🤔

#

or maybe its because I submitted shortly after refreshing my kaggle key

echo kernel
#

anyone in the AeroClub RecSys 2025 competition ?

#

how do i make sumission

#

do i need to show the notebook or just the final file ?

#

cause i cant to handle all the dataset in the notebook

turbid pollen
#

@everyone I made my app before submission to the competition should we keep the api key or should we redact it and the competition judges will use their api keys

slender pecan
#

Hey everyone!

Long before the launch of Kaggle’s Game Arena, I developed a real-time AI chess battle platform where you can watch different AI models play against each other.

🔗 Try it here: https://adaptive-ai-chess-game.streamlit.app/

🎮 Modes available:

Human vs Adaptive AI

Adaptive AI vs Hyperbolic AI

It’s a fun and insightful way to observe how different AI strategies behave on the board — ideal for anyone into AI, reinforcement learning, or game theory.

📩 If you're interested in collaborating, expanding, or integrating it into more formal benchmarks, feel free to connect:

👤 Karthikeya Guduru
📧 karthikeyaguduru19@gmail.com
🔗 https://www.linkedin.com/in/karthikeya-guduru-70227b262/

Streamlit
Chess AI Battle

Play chess against an Adaptive AI that learns your style and skill as you play. The AI adjusts it...

pulsar mantle
#

@indigo jasper Hi Melissa, I’m with the Hailuo AI (MiniMax) team. We’re running an Ai hackathon with a $100k prize pool and would love to partner with you. Who’s the best person to speak with? Thanks, Eric

warped fox
#

Hi, me and my team are participating on the Gemma 3n Hackathon competition. When we tried to submit our writeup into the Kaggle competition website but an error was showed "Internal Error" and our writeup could not be submitted, even though we tried submitting it minutes before the deadline...

Here is the topic we created on the discussion presenting our solution and the error we faced: https://www.kaggle.com/competitions/google-gemma-3n-hackathon/discussion/597689

We invested a lot of effort into this competition throught the whole month and would like to kindly ask the Hackathon organizers and Kaggle support team to take look at this and take in consideration our submission. If anyone can help me out I would be very grateful! Thank you.

Google - The Gemma 3n Impact Challenge

Explore the newest Gemma model and build your best products for a better world

pulsar mantle
pulsar mantle
indigo jasper
slim mulch
#

Not Comp, but I encourage able users to look into Rett Syndrome research if you'd like to help some kids. This directly benefits a user within the red team community, and I'm sure they'll appreciate any effort put toward helping those with the disorder. @covert heath delete if inappropriate. 🐉

craggy star
#

its not there

pulsar mantle
craggy star
pulsar mantle
# craggy star Btw why not on kaggle

We need a hybrid hosting strategy for both the Developer and Creator tracks. Kaggle is great for the Developer track.
However, one of our team leads found that we may not be eligible to post a competition since we’re not asking users to solve a specific problem.

craggy star
pulsar mantle
craggy star
pulsar mantle
vernal drum
#

Hi i am facing an error "Notebook threw error" while competition evaluation. Notebook runs fine when i run it manually. . Does anybody know how to solve it? (Jigsaw competition)

torn cradle
#

There's likely something in the hidden dataset that crashes it, e.g., NaN values.

I found it useful to comment out pieces of code and submit such incomplete result. It should help you narrow down the parts of the code that can cause the issue.