0 Follower
0 Following
tearth
Location
FI
Badges
2
0
0
Activity
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
Mon
Wed
Fri
Ratings Progression
Loading...
Challenge Categories
Loading...
Challenges Entered
Completed
AIcrowd
Learn Prompting
Towards AI
Preamble
Amazon Search
Amazon Prime Air
Understand semantic segmentation and monocular depth estimation from downward-facing drone images
Latest submissions
See All| failed | 218779 | ||
| failed | 218739 | ||
| failed | 218727 |
Seerave Foundation
A benchmark for image-based food recognition
Latest submissions
DEEP
Help improve humanitarian crisis response through better NLP modeling
Latest submissions
See All| graded | 58130 | ||
| graded | 58117 | ||
| graded | 58116 |
Estimate depth in aerial images from monocular downward-facing drone
Latest submissions
See All| failed | 218779 | ||
| failed | 218739 | ||
| failed | 218727 |
| Participant | Rating |
|---|
| Participant | Rating |
|---|
-
DenisVorotyntsev AMLD 2020 - Transfer Learning for International Crisis ResponseView
Amazon KDD Cup '23: Multilingual Recommendation Ch
AMLD 2020 - Transfer Learning for International...
Rssfete and tearth: Thank you so much
Over 4 years agoI’d like to say thanks to the organizers of the competition and everyone who was actively participating in it. My solution is simple, yet I think every team from top 5 used more or less the same approach: fine-tuning of pretrined transformer. I’m not sure that I could share more details (architecture, hyperparameters, and tricks) before the conference (it is stated in rules).
tearth has not provided any information yet.
Question about if we could use other information from the network, such as extra information from Amazon.com to train our model?
12 months agoI am also interested. @dipam could you clarify, are participants allowed to use any data outside of train/val/test dataset. For example, using product id (e.g. B07WSY3MG8) we could parse image of the product from amazon.com, reviews, etc. Are we allowed to use it?