Stack Overflow
Stack Overflow

Questions tagged [theano-cuda]

theano-cuda relates to the installation, configuration, and use of the CUDA backend for Theano

144 questions
12
votes
1 answer

GPU is lost during execution of either Tensorflow or Theano code

When training either one of two different neural networks, one with Tensorflow and the other with Theano, sometimes after a random amount of time (could be a few hours or minutes, mostly a few hours), the execution freezes and I get this message by…
Mega
  • 495
  • 5
  • 16
10
votes
1 answer

How to enable Keras with Theano to utilize multiple GPUs

Setup: Using a Amazon Linux system with a Nvidia GPU I'm using Keras 1.0.1 Running Theano v0.8.2 backend Using CUDA and CuDNN THEANO_FLAGS="device=gpu,floatX=float32,lib.cnmem=1" Everything works fine, but I run out of video memory on large…
Ray
  • 40,256
  • 21
  • 101
  • 138
8
votes
3 answers

WARNING (theano.sandbox.cuda): CUDA is installed, but device gpu is not available (error: cuda unavailable)

in Ubuntu MATE 16.04 I'm trying to run the deep-learning python examples here using the GPU: testing Theano with GPU I did run the example code, THEANO_FLAGS=mode=FAST_RUN,device=gpu,floatX=float32 python check1.py but it seems that it is used the…
valerio_sperati
  • 793
  • 2
  • 9
  • 23
7
votes
2 answers

Conda keeps trying to install all optional dependencies?

When installing Theano anaconda automatically tries to install pygpu despite this being an optional dependency. I have deleted the .theanorc file from my windows user directory. Also when running my application Theano tries to load from the GPU.…
K-Dawg
  • 3,013
  • 2
  • 34
  • 52
7
votes
3 answers

keras predict is very slow

I am working on a reinforcement learning task and decided to use keras NN model for Q value approximation. The approach is common: after each action the reward is stored in a memory replay array, then I take random sample from it and fit the model…
Serhiy
  • 4,357
  • 5
  • 37
  • 53
6
votes
0 answers

Theano uses gpu from the command line, but not from PyCharm

I am trying to make Theano use the gpu on my Linux machine. It works from the command line, but not from Pycharm. Both are using Python 3.5 from the same folder of my machine. I am testing this script: from theano import function, config, shared,…
Ginger
  • 8,320
  • 12
  • 56
  • 99
5
votes
1 answer

Can not use cuDNN on context None, fatal error: cudnn.h: No such file or directory

I am trying to configure theano to use gpu on my windows machine. I have set up .theanorc to use device= gpu but when I run some code that should utilize the gpu, I get the following error: Can not use cuDNN on context None: cannot compile with…
A_Matar
  • 2,210
  • 3
  • 31
  • 53
5
votes
2 answers

Installing CUDNN for general usage without root access

I use a server which has CUDA 7.5. But the server does not involve CUDNN. Is it possible to install CUDNN, and set all the linkings with CUDA, without root access, for the usage of all applications on ubuntu 14.04? I have implemented the solution on…
yusuf
  • 3,591
  • 8
  • 45
  • 86
4
votes
1 answer

Overriding existing lock by dead process '12224' (I am process '2516')

Using Theano backend. (57720L, 51L) (57720L, 5L) (850L, 51L) (850L,) WARNING (theano.gof.compilelock): Overriding existing lock by dead process '12224' (I am process '2516') what is the meaning of that, and how to deal with it.
L.Zhang
  • 41
  • 3
4
votes
1 answer

Share GPU memory for different users in keras and tensorflow

We had only one GPU installed with CUDA drivers and whenever one user runs the code, the whole memory is assigned to that user. And the other users are unable to use the GPU. Is there a way to get rid of this behavior?
Rajesh
  • 221
  • 4
  • 16
3
votes
0 answers

theano with GPU support

I installed cuda and cudnn following the instructions here: https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html following that I set up a conda environment with python3.8 and installed theano 1.0.4. Then I createa a .theanorc in my…
hadron
  • 457
  • 4
  • 19
3
votes
0 answers

How can I solve theano GPU pygpu error on Google Colab?

I'm using theano gpu on Google Colab and I get this error: ERROR (theano.gpuarray): pygpu was configured but could not be imported or is too old (version 0.7 or higher required) I've set: import os os.environ["THEANO_FLAGS"]="device=cuda,…
Christopher Pironti
  • 141
  • 1
  • 4
3
votes
0 answers

Theano cannot find libcuda.so

I have installed theano using miniconda: conda install theano This installed libgpuarray: 0.7.6-h14c3975_0 mkl-service: 1.1.2-py27hb2d42c5_4 pygpu: 0.7.6-py27h3010b51_0 theano: 1.0.2-py27h6bb024c_0 I then run my program and get the…
hyperio
  • 401
  • 3
  • 13
3
votes
1 answer

theano GPU support: PTX JIT compiler library not found

I am installing libgpuarray v0.7.4. When I import theano, I got the following errors: I am using the version of: (1) Theano from github (rel-1.0.0rc1), (2) CUDA 9.0 (I am sure CUDA / cusolver works, I installed pycuda, scikit-cuda and could run them…
wh0
  • 510
  • 1
  • 6
  • 19
3
votes
0 answers

Setting the path to CUDA in THEANO_FLAGS

I try to set the THEANO_FLAGS and then import Theano. This always results in an error. In [1]: import os In [2]: os.environ["THEANO_FLAGS"] = "cuda.root=/usr/local/cuda,device=cuda,floatX=float32" In [3]: import theano ERROR (theano.gpuarray):…
Andy
  • 1,072
  • 2
  • 19
  • 33
1
2 3
…
9 10