Whether NVIDIA Quadro K620 is enough to run the deep learning codes smoothly with less time in Matlab?

22 次查看（过去 30 天）
rcjr15
rcjr15 2017-7-25
评论： Ananya Roy 2019-4-5
Whether NVIDIA Quadro K620 is enough to run the deep learning codes smoothly with less time in Matlab?

请先登录，再回答此问题。

采纳的回答

Joss Knight
Joss Knight 2017-7-28
The K620 is capable of doing Deep Learning in MATLAB but it only has 2GB of memory. This limits what you can do, in particular, the size of network that you can train.

更多回答（1 个）

Kristen Amaddio
Kristen Amaddio 2017-7-27
According to the documentation below, using a GPU with deep learning in MATLAB requires a CUDA-enabled NVIDIA GPU with compute capability 3.0 or higher:
https://www.mathworks.com/help/nnet/ug/deep-learning-in-matlab.html#bvmt9z2
According to the NVIDIA website, the Quadro K620 is CUDA-enabled and has compute compatibility 5.0, so this appears to satisfy the specified requirements:
https://developer.nvidia.com/cuda-gpus

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by