NautilusServer

From Deep Depth 116E167 Project Documentation
Revision as of 18:24, 14 August 2017 by Damienjadeduff (talk | contribs)

Jump to: navigation, search

Accessing

IP: 160.75.27.83

SSH port: 1542

Access from: ITU or ITU VPN.

VPN Help

SSH help

The SSH command to connect from a Unix environment:

   ssh -X -p 1542 hossein@160.75.27.83

Switch meanings:

  • -p 1542: connect on port 1542
  • -X: This allows you to run X applications. Omit it if you will be pure command line.

Note: to check that X forwarding is working, once you have connected, try running on the server the command:

   xeyes

Or:

   dolphin

You could for example run spyder like this. But there can be some latency across the network.

Setting up a deep learning environment

Install anaconda

   export ANACONDA_PATH_PARENT=$HOME/software
   export ANACONDA_PATH=$ANACONDA_PATH_PARENT/anaconda3
   export ANACONDA_INSTALLER=Anaconda3-4.3.1-Linux-x86_64.sh
   mkdir -p ~/tmp
   cd ~/tmp
   mkdir -p $ANACONDA_PATH_PARENT
   wget https://repo.continuum.io/archive/$ANACONDA_INSTALLER
   sudo bash $ANACONDA_INSTALLER -b -p $ANACONDA_PATH
   export PATH=$ANACONDA_PATH/bin:$PATH
   echo PATH: $PATH
   echo >> ~/.bashrc
   echo export PATH=$ANACONDA_PATH/bin:\$PATH >> ~/.bashrc

Install tensorflow and keras

These will be installed in a conda environment called deep:


   export ENVNAME=deep
   conda create --name $ENVNAME
   source activate $ENVNAME
   conda install theano keras tensorflow tensorflow-gpu opencv pillow spyder matplotlib

To check Keras is working:

   python -c "from keras.models import Sequential;Sequential()"

To check the GPU is working with tensorflow, xheck this first (it should list two GPUs):

   nvidia-smi


Then make sure the following script runs and finds one CPU and two GPUS: https://bitbucket.org/damienjadeduff/uhem_keras_tf/src/master/sariyer_python3/test_tf_gpu.py

Run it like this:

   python test_tf_gpu.py

Server Construction

Software

OS

Kubuntu 16.04.3 LTS

Graphics Drivers

Nvidia 384.59 drivers installed using runfile NVIDIA-Linux-x86_64-384.59.run

Installed using (to keep using the integrated graphics as main display graphics):

   sudo ./NVIDIA-Linux-x86_64-370.28.run --no-opengl-files --no-x-check --disable-nouveau


CUDA Drivers

Installed using

   cuda_8.0.61.2_linux.run