Quelltext der Seite GPU-Erweiterung für Ubuntu



auf einem System auf dem noch nicht Anaconda etc. vorinstalliert ist.

 sudo apt-get update

 sudo apt-get upgrade

 sudo apt-get install python3-pip python3-dev


 sudo apt-get install build-essential cmake git

 sudo apt-get install libopenblas-dev 

 sudo apt-get install liblapack-dev

 sudo apt-get install python-numpy python-scipy python-matplotlib

 sudo apt-get install libhdf5-serial-dev python-h5py

 sudo apt-get install graphviz

 sudo pip3 install pydot-ng

 sudo apt-get install python-opencv

----

CUDA downloaden:

https://developer.nvidia.com/cuda-90-download-archive?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu

----

CUDA-Version für 17.04 auswählen (funktioniert auch für Ubuntu 18.04)

 sudo dpkg -i cuda-repo-ubuntu1704-9-0-local_9.0.176-1_amd64.deb

 sudo apt-key add /var/cuda-repo-9-0-local/7fa2af80.pub

 sudo apt-get update

 sudo apt-get install cuda

 cuda dir: /usr/local/cuda-9.0
 cuDNN >= 7.2
 https://developer.nvidia.com/cudnn

 sudo dpkg -i libcudnn7_7.4.1.5-1+cuda9.0_amd64.deb

 sudo pip3 install tensorflow-gpu

----

verify:
 python3 -c "import tensorflow as tf; tf.enable_eager_execution(); print(tf.reduce_sum(tf.random_normal([1000, 1000])))"

----

output sollte so ähnlich aussehen:

 2018-12-11 17:19:43.988326: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this 
 TensorFlow binary was not compiled to use: AVX2 FMA
 2018-12-11 17:19:44.075230: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:964] successful NUMA node read from SysFS had 
 negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
 2018-12-11 17:19:44.075916: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1432] Found device 0 with properties: 
 name: GeForce GTX 960M major: 5 minor: 0 memoryClockRate(GHz): 1.176
 pciBusID: 0000:01:00.0
 totalMemory: 1.96GiB freeMemory: 1.72GiB
 2018-12-11 17:19:44.075953: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] Adding visible gpu devices: 0
 2018-12-11 17:22:28.528023: I tensorflow/core/common_runtime/gpu/gpu_device.cc:982] Device interconnect StreamExecutor with 
 strength 1 edge matrix:
 2018-12-11 17:22:28.528065: I tensorflow/core/common_runtime/gpu/gpu_device.cc:988]      0 
 2018-12-11 17:22:28.528072: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 0:   N 
 2018-12-11 17:22:28.528250: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device 
 (/job:localhost/replica:0/task:0/device:GPU:0 with 1477 MB memory) -> physical GPU (device: 0, name: GeForce GTX 960M, pci bus 
 id: 0000:01:00.0, compute capability: 5.0)
 tf.Tensor(-253.9591, shape=(), dtype=float32)

----
danach keras und jupyter installieren:

 sudo pip3 install keras

 sudo python3 -m pip install jupyter

[[Category:HowTo]]
[[Category:KI]]
[[Category: deep learning]]
[[Category:Jupyter Notebook]]
[[Category:Tensorflow]]
[[Category:Keras]]
[[Category:GPU]]
[[Category:Command Line]]