Построяване на невронна мрежа с Pytorch
Building the neural net with Pytorch
Author(s): Nayden Nenkov, Ertan GeldievSubject(s): Social Sciences, Economy, Education, ICT Information and Communications Technologies
Published by: Шуменски университет »Епископ Константин Преславски«
Keywords: PyTtorch-an open source machine learning library written in the Lua language; Deep Learning; Neural Networks; Deep Python integration; CPU-Central Processor Unit; GPU – Graphic Processor Unit; TPU – T
Summary/Abstract: Neural Network's capabilities to support deep learning challenge software developers to produce work environments to create and implement similar tasks and methods. One such environment is Pytorch, a Python library that can be integrated with acceleration libraries such as Intel MKL and NVIDIA (CuDNN, NCCL) to increase speed. One definition of Pytorch [1] is "an open-source deep learning platform that provides a seamless path from prototyping research to deployment." The aim of the study is to demonstrate with examples the construction of a neural network with Pytorch
Journal: Научни трудове. Колеж - Добрич
- Issue Year: XI/2019
- Issue No: 1
- Page Range: 144-153
- Page Count: 10
- Language: Bulgarian
