There used to be a tutorial here for using Torch7 on EC2, but it's now outdated. It is best to use an EC2 image that already has Torch7 and CUDA stuff preinstalled.
Last active
November 23, 2015 03:11
-
-
Save y0ast/1e80daf19a6d4e210706 to your computer and use it in GitHub Desktop.
Tutorial for using Torch7 on Amazon EC2 GPUs
Good that it's solved!
Thanks for the tutorial!
Unfortunately it doesn't seem to be working for me:
$ th -l cutorch -e "cutorch.test()"
could not load cutorch, skipping
[string "cutorch.test()"]:1: attempt to index global 'cutorch' (a nil value)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi, I followed the instructions on this page and got the following error:
/usr/local/bin/luajit: unable to initialize cublas
stack traceback:
[C]: at 0x7f68cfe679e0
[C]: in function 'require'
/usr/local/share/lua/5.1/cutorch/init.lua:2: in main chunk
[C]: in function 'require'
cuda_test.lua:1: in main chunk
[C]: in function 'dofile'
/usr/local/lib/luarocks/rocks/trepl/scm-1/bin/th:129: in main chunk
[C]: at 0x00405f50
Update:
The problem was in the gpu drivers provided by Amazon.