Can't run rbm/runall.sh

[hlzz]

It seems that this project doesn't have support for CUDA 5.5. When I compiled the cudamat with CUDA 5.5. It reported that 'uint is undefined'. So I added '#define uint unsigned int' in the header. Then after I compiled libcudamat_conv.so and libcudamat.so, I failed on 'python test_cudamat.py'. I didn't got error message, but I got the following message:

Run 0 tests in 0.000s

I ignored this and tried to run the sample code. I mean the runall.sh in /deepnet/examples/rbm, then it told me 'No GPU board available'. I don't know how to solve this. By the way, I use mac OS 10.8.4. Should I go back to older version of CUDA (such as 4.2, 5.0) or find a linux OS? Thanks!

[hlzz]

I degraded my cuda version to 5.0 but it still doesn't work. I know, it seems that at present this library doesn't support mac os, right?

[odieXin]

I met exactly the same problem. Did you get a solution?

[zuiwufenghua]

I used the Ubuntu 12.04 CUDA5.5,however it showed that 'No GPU board available'

[ansyral]

got the same problem with UBUNTU 10.04 CUDA4.2. Also show that"No GPU board available"

[OpenHero]

def LockGPU(max_retries=10): #for retry_count in range(max_retries):

board = gpu_lock.obtain_lock_id()

if board != -1:

break

sleep(1)

board = 0 if board == -1: print 'No GPU board available.' sys.exit(1) else: cm.cuda_set_device(board) cm.cublas_init() return board

def FreeGPU(board): cm.cublas_shutdown() #gpu_lock.free_lock(board)

[pannous]

Shouldn't it fall back to to eigenmat instead of quitting with No GPU board available. ?

me:~/dev/ai/deepnet/deepnet/examples/dbm$ ./runall.sh No GPU board available.

[OpenHero]

fixed it with upload code.

[KerrChen]

no gpu board error doesn't show again,but there is another error,total size of new array must be unchanged,did you have the same puzzle?

[Rietaros]

I run for DBN.

I'm using CUDA 7.5 nvidia 352..39.

And have the same problem. While I run gpu_lock.py --id it comes with output -1

Any suggestion?