i am happy to be in the TOP 3
Congratulations to the winners but what a shake up guys !!!
Congratulations to you. Can you please share with us, your approach ?
Congratulations to you and your team. what was your approach. I can’t wait to find out 😊
Congratulations to you and your team.
Congrats :)
Congratulations!.
Congratulations!!
Congratulations!
Many congratulations to all of the top finishers.
No doubt, the shake-up was expected because we were having very less dataset and there are several data points that can go into more than one category.
For example: "I am addicted" so this can be Alcohol addiction as well as Drug addiction.
Looking forward to learn from the winners :)
Shobhit aka asinglenueron. What was ur approach. you did very well on private LB.
First of all I want to know which of my submission did well.
I tried so many weird approaches :D
Do you know, how to check the submission score on a private dataset?
You can still submit now and get the score. LB will not get updated.
I got it. Now the submission scores got changed and they are reflecting the private leaderboard scores.
So the model did best on a private dataset is:-
single model roberta-base
Converting text to lowercase : False
OOF with 20 splits
CV Score: 0.3294175537361298
Public leaderboard SCORE: 0.33584687745369
Private leaderboard SCORE: 0.311455622610637
What preprocessing did you do on your dataset.. How did you clean your dataset ?
There was no text cleaning in this version.
I did the same thing but only for 400 rows, as i have a weird habit of always taking a holdout,and 10 splits
I tried both 10 and 20, but 20 CV score was in sync with the public leaderboard score. That's why I picked 20 splits for this competition.
@asingleneuron can you share your notebook on your solution I used
Congratulations guys
Thank you
'/%3e%3c/defs%3e%3cpath fill='%23fff' d='M-120-80h240V80h-240z'/%3e%3cpath d='M-120-80h240v48h-240z'/%3e%3cpath fill='%23060' d='M-120 32h240v48h-240z'/%3e%3cg id='b'%3e%3cuse xlink:href='%23a' stroke='%23000'/%3e%3cuse xlink:href='%23a' fill='%23fff'/%3e%3c/g%3e%3cuse xlink:href='%23b' transform='scale(-1 1)'/%3e%3cpath fill='%23b00' d='M-120-24v48h101c3 8 13 24 19 24s16-16 19-24h101v-48H19C16-32 6-48 0-48s-16 16-19 24z'/%3e%3cpath id='c' d='M19 24c3-8 5-16 5-24s-2-16-5-24c-3 8-5 16-5 24s2 16 5 24'/%3e%3cuse xlink:href='%23c' transform='scale(-1 1)'/%3e%3cg fill='%23fff'%3e%3cellipse rx='4' ry='6'/%3e%3cpath id='d' d='M1 5.85s4 8 4 21-4 21-4 21z'/%3e%3cuse xlink:href='%23d' transform='scale(-1)'/%3e%3cuse xlink:href='%23d' transform='scale(-1 1)'/%3e%3cuse xlink:href='%23d' transform='scale(1 -1)'/%3e%3c/g%3e%3c/svg%3e)
Congratulations to you. Can you please share with us, your approach ?
Congratulations to you and your team. what was your approach. I can’t wait to find out 😊
Congratulations to you and your team.
Congrats :)
Congratulations!.
Congratulations!!
Congratulations!
Many congratulations to all of the top finishers.
No doubt, the shake-up was expected because we were having very less dataset and there are several data points that can go into more than one category.
For example: "I am addicted" so this can be Alcohol addiction as well as Drug addiction.
Looking forward to learn from the winners :)
Shobhit aka asinglenueron. What was ur approach. you did very well on private LB.
First of all I want to know which of my submission did well.
I tried so many weird approaches :D
Do you know, how to check the submission score on a private dataset?
You can still submit now and get the score. LB will not get updated.
I got it. Now the submission scores got changed and they are reflecting the private leaderboard scores.
So the model did best on a private dataset is:-
single model roberta-base
Converting text to lowercase : False
OOF with 20 splits
CV Score: 0.3294175537361298
Public leaderboard SCORE: 0.33584687745369
Private leaderboard SCORE: 0.311455622610637
What preprocessing did you do on your dataset.. How did you clean your dataset ?
There was no text cleaning in this version.
I did the same thing but only for 400 rows, as i have a weird habit of always taking a holdout,and 10 splits
I tried both 10 and 20, but 20 CV score was in sync with the public leaderboard score. That's why I picked 20 splits for this competition.
@asingleneuron can you share your notebook on your solution I used
Congratulations guys
Thank you