https://github.com/iacolippo/gpu-dnn-install
Scripts and instructions to install CUDA, cuDNN and the most common deep learning frameworks.
https://github.com/iacolippo/gpu-dnn-install
cuda-toolkit cudnn install-script theano torch
Last synced: 7 months ago
JSON representation
Scripts and instructions to install CUDA, cuDNN and the most common deep learning frameworks.
- Host: GitHub
- URL: https://github.com/iacolippo/gpu-dnn-install
- Owner: iacolippo
- Created: 2017-02-28T21:09:09.000Z (about 9 years ago)
- Default Branch: master
- Last Pushed: 2017-05-14T11:48:20.000Z (almost 9 years ago)
- Last Synced: 2025-05-04T14:46:55.370Z (11 months ago)
- Topics: cuda-toolkit, cudnn, install-script, theano, torch
- Language: Shell
- Size: 89.8 KB
- Stars: 4
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# gpu-dnn-install
Author: Iacopo Poli
Description: Scripts and instructions to install CUDA, cuDNN and two of the most common deep learning frameworks ([Theano](http://deeplearning.net/software/theano/) and [Torch](http://torch.ch)).
## PREREQUISITES:
1 - Download CUDA 8.0 deb(network) file for your system [here](https://developer.nvidia.com/cuda-downloads). If you're using Ubuntu 16.04 on NV6, the file should be called
```bash
cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
```
2 - Download CuDNN 5.1 for CUDA 8.0 Linux [here](https://developer.nvidia.com/rdp/cudnn-download). You have to register first and accept the License. The file should be called
```bash
cudnn-8.0-linux-x64-v5.1.tar
```
It should work the same with new versions of CuDNN.
All the other files needed are in this repository.
## INSTALLATION
NOTE: You have to set the permission to execute the installation script files. You can do that with
```bash
chmod a+x
```
0 - Run this and check that it prints something, otherwise there is no NVIDIA hardware available.
```bash
lspci | grep -i nvidia
```
Sample output:
1 - Run
```bash
./cuda-install.sh
```
2 - add /usr/local/cuda-8.0/bin to PATH environment variable in .profile in home directory using nano or vim
```bash
export PATH="$PATH:/usr/local/cuda-8.0/bin"
```
3 - add /usr/local/cuda-8.0/lib64\ to LD_LIBRARY_PATH environment variable in .profile using nano or vim
```bash
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64"
```
For points 2 & 3 you can look at the [example file](https://github.com/iacolippo/gpu-dnn-install/blob/master/.profile) in this repository.
4 - Activate the changes using
```bash
source .profile
```
5 - Reboot the system
```bash
sudo reboot
```
6 - Reconnect to the machine via ssh and write a ```.theanorc``` file in the home directory equal to [this](https://github.com/iacolippo/gpu-dnn-install/blob/master/.theanorc). Then run the following command and check that is using gpu. It should also print a message that cuDNN is not available.
```bash
./theano-install.sh
```
Output:
7 - Run the following command.
```bash
./cudnn-install.sh
```
If you installed Theano, you can run ```python gpu-test.py``` and you should see cuDNN is now available.
Output:
8 - Install Torch
```bash
./torch-install.sh
```
Answer *yes* to anything on the terminal. At the end, enter
```bash
source ~/.bashrc
```
9 - Install Tensorflow (GPU version)
```bash
./tensorflow-install.sh
```
Check that the GPU is being used by running
```bash
python tensorflow-gpu.py
```
10 - Install Keras by running
```bash
sudo pip install keras
```
When using Tensorflow backend (default setting), the code runs on GPU automatically if one is detected.
For any question you can contact me on Twitter [@iacopo_poli](https://twitter.com/iacopo_poli).