Visualization transforms for convolutional neural networks
Автор: Suhas Sreehari
Загружено: 2020-08-18
Просмотров: 91
Convolutional neural networks (CNNs) are powerful algorithms for dealing with visual data, such as images and videos. However, many datasets around us are not images. For example, in healthcare, banking, pharma... a loan default dataset, or an Experian attribute file, or a heart arrythmia dataset -- none of these are explicitly images. We show that these important tabular datasets can be transformed into meaningful images first and then we are able to effectively apply CNNs for a variety of ML tasks.
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке: