Error when importing theano using GPU

Question

I am trying to use theano for python 2.7 installed through anaconda on a CentOS 5.6. I am able to import theano if only using cpu, but when I try to run it using gpu I get hte error:

>>> import theano
Using gpu device 0: Tesla K20m
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/data/ASR1/yunwang/software/anaconda/lib/python2.7/site-packages/theano/__init__.py", line 98, in <module>
    theano.sandbox.cuda.tests.test_driver.test_nvidia_driver1()
  File "/data/ASR1/yunwang/software/anaconda/lib/python2.7/site-packages/theano/sandbox/cuda/tests/test_driver.py", line 39, in test_nvidia_driver1
    if not numpy.allclose(f(), a.sum()):
  File "/data/ASR1/yunwang/software/anaconda/lib/python2.7/site-packages/theano/compile/function_module.py", line 606, in __call__
    storage_map=self.fn.storage_map)
  File "/data/ASR1/yunwang/software/anaconda/lib/python2.7/site-packages/theano/compile/function_module.py", line 595, in __call__
outputs = self.fn()
TypeError: Argument not a CudaNdarray
Apply node that caused the error: GpuCAReduce{add}{1}(<CudaNdarrayType(float32, vector)>)
Inputs types: [CudaNdarrayType(float32, vector)]
Inputs shapes: [(10000,)]
Inputs strides: [(1,)]
Inputs values: ['not shown']

HINT: Re-running with most Theano optimization disabled could give you a back-trace of when this node was created. This can be done with by setting the Theano flag 'optimizer=fast_compile'. If that does not work, Theano optimizations can be disabled with 'optimizer=None'.
HINT: Use the Theano flag 'exception_verbosity=high' for a debugprint and storage map footprint of this apply node.

Any ideas on how to fix that? I have been trying everything, from reinstalling cuda, anaconda,python...

Answer 1

I have found the problem after reading this link . Apparently theano creates a cache folder .theano that was causing the issue. It was fixed after removing this folder.