Exploring the Deep Learning Framework PyTorch - Stephanie Kim

preview_player
Показать описание
STEPHANIE KIM | SOFTWARE ENGINEER AT ALGORITHMIA

Users rapidly adopted PyTorch 1.0 for many reasons. PyTorch is intuitive to learn, and its modularity enhances debugging and visibility. Additionally, unlike other frameworks such as Tensorflow, PyTorch supports dynamic computation graphs that allow network behavior changes on the fly. This talk showcases PyTorch benefits like TorchScript, which allows models to be exported in non-Python environments. We’ll also discuss pre-release serialization and performance issues, as well as the increased focus on optimization and performance in PyTorch 1.0.
Рекомендации по теме
Комментарии
Автор

Really Helpful for beginners and people trying to move from tensorflow to Pytorch 👌👌

Skyler_twilight