theano-cuda relates to the installation, configuration, and use of the CUDA backend for Theano
Questions tagged [theano-cuda]
144 questions
2
votes
1 answer
What's the effect of CNMeM enabled but 'cuDNN not available' in Theano?
I have the following code based on Theano example:
from theano import function, config, shared, sandbox
import theano.tensor as T
import numpy
import time
vlen = 10 * 30 * 768 # 10 x #cores x # threads per core
iters = 1000
rng =…

neversaint
- 60,904
- 137
- 310
- 477
2
votes
1 answer
How to use theano with GPU on OSX?
OS: OSX 10.11.4
GPU: GeForce GT 750M, CUDA7.5
Language: Python Anaconda 2.7
I have followed theano's instructions to setup CUDA on my Mac:
set PATH: export PATH=/Developer/NVIDIA/CUDA-7.5/bin:$PATH
set LD_LIBRARY_PATH: export…

Junhong Xu
- 145
- 1
- 2
- 13
2
votes
1 answer
Import theano with gpu fails, because "out of memory" error
As soon as I try to import theano library I get this error:
>>> import theano
ERROR (theano.sandbox.cuda): ERROR: Not using GPU. Initialisation of device gpu failed:
initCnmem: cnmemInit call failed! Reason=CNMEM_STATUS_OUT_OF_MEMORY.…

Higgcz
- 55
- 2
- 7
2
votes
1 answer
Pycuda Compilation Error
I am on windows machine with python 2.7 ( 32 bit ) and pycuda with cuda 7.5 whl installed . I get error while running a sample program to test pycuda.
Traceback (most recent call last):
File "C:\Users\newbie\Desktop\roo.py", line 82, in
…

Rahul rao
- 45
- 5
2
votes
1 answer
How to get AdvancedSubtensor on GPU
I have some subtensor and for some reason, Theano cannot transfer it to the GPU.
Some sample code:
import numpy
import theano
import theano.printing
import theano.compile.io
import theano.compile.function_module
import theano.tensor as T
from…

Albert
- 65,406
- 61
- 242
- 386
1
vote
1 answer
How to provide the dynamic library used for blas for Theano on Google Colab?
As the title clearly describes the situation I'm facing, I'm getting the following warning on Google Colab while using the Theano as the backend of Keras:
WARNING (theano.tensor.blas): We did not find a dynamic library in the library_dir of the…

talha06
- 6,206
- 21
- 92
- 147
1
vote
0 answers
Can't put GPU as Theano device in Windows
My pc specs are a GPU NVIDIA 1050ti in a windows 10.
I installed the CUDA drivers (had to install visual studio with c++ tools) and run the tests, created a new conda environment with with theano and pygpu (python 3.6.3). Run the script:
from theano…

Joaquim Ferrer
- 603
- 6
- 23
1
vote
0 answers
How to convert parameters saved from theano model (.pkl / .npz) to tensorflow format .ckpt?
I have found this link Converting Theano-based Keras model definition to TensorFlow, but I do not use keras at all.
Are there other ways to do that?

j35t3r
- 1,254
- 2
- 19
- 53
1
vote
1 answer
mod.cu(66): error: identifier "cudnnSetFilterNdDescriptor_v4" is undefined
When I compile my python produce written with theano in Linux.I got a compile error in the picture.All path set well.This error suddenly appeared with no operation.

Cui
- 21
- 1
1
vote
0 answers
theano fails to compile cuda but the python code runs using GPU
I am trying to run a theano simple code on Ubuntu 16.04 with Cuda 8.0 on NVIDIA 1060 GPU within a python virtual environment created by anaconda. The following is my theanorc file:
[global]
floatX = float32
device = cuda
The code I am trying to run…

Mohammadreza
- 450
- 2
- 4
- 14
1
vote
0 answers
Installing theano 0.9 with GPU support on Ubuntu 14.04, TitanX pascal, Cuda8.0
It's driving me crazy. I'm following this guide to install the required libgpuarray. But I keep on getting errors.
In this guide:
cd
rm -rf build Build
mkdir Build
cd Build
cmake .. -DCMAKE_INSTALL_PREFIX=~/.local…

mcExchange
- 6,154
- 12
- 57
- 103
1
vote
1 answer
Theano cannot import pygpu/cudnn (macOS, CUDA 8)
The problem is that I can't import theano. I know there's several other questions about same issue but none of them have worked for me and usually they are either with older versions or different OS. I've been trying to fix this for a few days now…

jimmy
- 1,413
- 2
- 13
- 23
1
vote
1 answer
MemoryError with python using keras and theano
I tried this keras tutorial. I'm using theano on my other project, so I changed keras to use theano and not tenorflow. But when I run this tutorial I get this error first:
nvcc fatal : Cannot find compiler 'cl.exe' in PATH
And after some time,…

Domjan Barić
- 11
- 2
1
vote
1 answer
GPU slower than CPU when it comes to theano prediction?
I was using keras + theano to predict labels from a VGG pre-trained model on NVidia TK1.
I am getting a faster prediction time from CPU than from GPU in predictions. If my memory is correct, prediction also involves a lot of number crunching in a…

Winston Chen
- 6,799
- 12
- 52
- 81
1
vote
0 answers
Running Multiple Gpus in theano jupyter notebooks, implementing theano.gpyarray.use
I have a linux system with three gpus. I am using keras with theano to run cnn's, In the past when I was using Theano 8.+ , I was able to assign a particular gpu to jupyter notebook window using the following:
import…

JerryC
- 11
- 2