top of page

News to help your R&D in artificial intelligence, machine learning, robotics, computer vision, smart hardware

As an Amazon Associate I earn

from qualifying purchases

Writer's picturemorrislee

Training with modified data like using 10 times more data

Updated: Jun 22, 2021

Training with modified data like using 10 times more data


How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers

arXiv paper abstract https://arxiv.org/abs/2106.10270


To help with augmention, AugLy https://github.com/facebookresearch/AugLy has over 100 augmentations for images, video, text, and audio. They also include emoji, fonts, and screenshot templates to assist in augmentation.



Vision Transformers (ViT) have been shown to attain highly competitive performance for a wide range of vision applications, such as image classification, object detection and semantic image segmentation.


... empirical study in order to better understand the interplay between the amount of training data, AugReg, model size and compute budget.


... find that the combination of increased compute and AugReg can yield models with the same performance as models trained on an order of magnitude more training data: we train ViT models of various sizes on the public ImageNet-21k dataset which either match or outperform their counterparts trained on the larger, but not publicly available JFT-300M dataset.


Please like and share this post if you enjoyed it using the buttons at the bottom!


Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact

Web site with my other posts by category https://morrislee1234.wixsite.com/website

19 views0 comments

Comments


ClickBank paid link

bottom of page