I solved the test task from KazanExpress on multiclass products classification. Each product has its title, description, characteristics, images and other data that is usually included into product cards on the marketplace. I developed a multimodal model
which uses embeddings from ruBERT
, FastText
, Wiki2Vec
and feature maps obtained from ResNet
pretrained on the ImageNet. My solution performs 0.884 on f1 weighted score on more than 750 classes.
-
Notifications
You must be signed in to change notification settings - Fork 0
KKroliKK/Products-Classification
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Test task from KazanExpress 2023.
Resources
Stars
Watchers
Forks
Releases
No releases published