Request for correct options to run in cpu mode

[ghost]

Hi:

I'm interested in your word version of karpathy's rnn.

I'm new to torch/lua and have the original running error free on 64 bit mint with i7 in cpu mode (GPU will be next machine)

When I run your version of the sanity check I receive the following error:

$ th train.lua -gpuid -1

/home/pixelhead/torch/install/bin/luajit: /home/pixelhead/torch/install/share/lua/5.1/trepl/init.lua:383: ./util/SharedDropout.lua:3: attempt to call field 'CudaTensor' (a nil value)
stack traceback:
[C]: in function 'error'
/home/pixelhead/torch/install/share/lua/5.1/trepl/init.lua:383: in function 'require'
train.lua:119: in main chunk
[C]: in function 'dofile'
...head/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:145: in main chunk
[C]: at 0x00406670

I'm assuming 'CudaTensor' applies to GPU use only. Is there a modification I can make in 'SharedDropout.lua' or another script for CPU use?

Looking forward to comparing the output from the original to your version.

Cheers,

[ghost]

Found answer to my problem in issue #7 :)

@larspars Oh, figured it out,

had to change line 3 in .../word-rnn/util/SharedDropout.lua

where it says torch.CudaTensor I simply changed it to torch.Tensor and now it's good.

[andreasmarkmann]

Should there be a branch statement testing the cpuid command line parameter in order to allow use on both platforms without edit?