Activity
Ratings Progression
Challenge Categories
Challenges Entered
Participant | Rating |
---|---|
benjamin_gabay | 0 |
Participant | Rating |
---|
-
NVIDIA-Merlin Amazon KDD Cup '23: Multilingual Recommendation ChallengeView
-
PWN_Patrol HackAPrompt 2023View
Amazon KDD Cup '23: Multilingual Recommendation Ch
🗂️ Using External Data: Guidelines for Data Declaration
11 months ago@hansu How did you gather that data? We looked at amazon sites, and terms of use do not allow data scrapping,
Wow, this was close
11 months agoCongrats to all, and kudos to MGTV-REC who fought till the end and came very close to us.
I can’t wait reading all the great solutions form other teams.
💬 Feedback & Suggestions
11 months agoyes, I have 5 submissions when my team mates have already submitted. We keep track of our submissi0ns to not exceed 5 per tak per day. W euse UTC time.
💬 Feedback & Suggestions
11 months agoThanks.
I also posted a while ago that the 5 submissions per 24 hours is not enforced at team level by your UI. It is only enforced at individual level. We track our submissions as a team to no exceed 5 submissions per day, but this is not convenient and it is error prone.
💬 Feedback & Suggestions
11 months agoCertainly not. On Kaggle a merge is possible only if the sum of submission counts does not exceed 5 times the number of days spent on the competition so far.
You system is an incentive for officially teaming as late as possible. I wish we had known.
🗂️ Using External Data: Guidelines for Data Declaration
11 months agoAll right, we have to make dataset available to other participants after the competition end.
Wait.
Really?
🚨 Phase 2 Is Now Live
11 months agoHi, is “9th June, 00:00:00 UTC” the start of June 9 or the end of June 9?
Also the overview pages says competition ends on June 14. Which date is the right one?
More than 5 subs a day are posible !?!
12 months agoHi, it seems that the 5 sub limit a day is enforced at the individual level and not the team level. We try ouor best to not use more than 5 subs a day as a team, but it would greatly help if the limit was explicitly enforced.
HackAPrompt 2023
I think it would make more sense if the evaluation checked for the answer within the response rather than being the entire response
11 months agobeen there, found a way eventually. Keep pushing!
🗂️ Using External Data: Guidelines for Data Declaration
11 months ago@snehananavati Do pretrained models weights (for instance form Huggingface site) count as datasets?