Skip to content
This repository has been archived by the owner on Jan 13, 2022. It is now read-only.

AWS Support #5

Open
sagaian opened this issue Dec 6, 2015 · 1 comment
Open

AWS Support #5

sagaian opened this issue Dec 6, 2015 · 1 comment

Comments

@sagaian
Copy link

sagaian commented Dec 6, 2015

Hi - I think this may have issues running on AWS as AWS offers instances with GRID K520 with compute capability of only 3. Here's the output of the device query on g2.8xlarge instance type:

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 7.5, CUDA Runtime Version = 7.5, NumDevs = 4, Device0 = GRID K520, Device1 = GRID K520, Device2 = GRID K520, Device3 = GRID K520

I am able to train the model, but not test it as I run into an issue similar to the one described here:

https://groups.google.com/forum/#!msg/torch7/aO7Q6A2RxHs/x5CtTh_mCQAJ

I get an error "invalid device function" which according to the thread results from only 3.5 and up being supported. Just wanted to know if there was a work around. Thanks!

@supchop
Copy link

supchop commented Apr 18, 2016

Sorry for the extremely late response. Were you able to resolve the issue? Unfortunately I have never worked with AWS.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants