I installed Theano on a Windows machine and configured the instructions .
I placed the following .theanorc.txt file in the C: \ Users \ my_username folder:
#!sh [global] device = gpu floatX = float32 [nvcc] fastmath = True # flags=-m32 # we have this hard coded for now [blas] ldflags = # ldflags = -lopenblas # placeholder for openblas support
I tried to run the test, but could not run it on the GPU. I think the values ββfrom .theanorc.txt are not readable because I added the line config.device and it prints out "cpu".
Below is a basic script test and output:
from theano import function, config, shared, sandbox import theano.tensor as T import numpy import time print config.device vlen = 10 * 30 * 768
exit:
pydev debugger: starting (pid: 9564) cpu [Elemwise{exp,no_inplace}(<TensorType(float64, vector)>)] Looping 1000 times took 10.0310001373 seconds Result is [ 1.23178032 1.61879341 1.52278065 ..., 2.20771815 2.29967753 1.62323285] Used the cpu
I successfully installed the CUDA Toolkit but could not install pyCUDA. I think Theano should work without installing pyCUDA.
I would be very grateful if anyone could help solve this problem. I followed these instructions, but I donβt know why the configuration values ββin the program do not match the values ββin the .theanorc.txt file.
source share