User Tools

Site Tools


cs501r_f2016:openlabtf

Tensorflow on open lab machines

Here are some notes for getting Tensorflow installed and running on the open lab machines.

First of all, you can always just use my anaconda installation, which already has Tensorflow 0.10 installed. It can be found at

/users/faculty/wingated/anaconda2

In order to use it, you should be able to add the following line to your ~/.bashrc:

export PATH=“/users/faculty/wingated/anaconda2/bin:$PATH”

If for whatever reason you would like your own anaconda installation, here are the exact steps I took to install it and Tensorflow:

wget https://repo.continuum.io/archive/Anaconda2-4.1.1-Linux-x86_64.sh
 
bash ./Anaconda2-4.1.1-Linux-x86_64.sh 
. ~/.bashrc
 
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
pip install --upgrade $TF_BINARY_URL
pip install -I --upgrade setuptools
pip install --upgrade $TF_BINARY_URL

(When anaconda asks you if you want it to modify your .bashrc, I answered yes. All it does is append that PATH modification line.)

Note that this installs the CPU-only version.

Note that no sudo commands are used – that's because the pip commands I used were the anaconda ones.


Fellow lab mate Darin Costello reports that the following worked for him: given a working anaconda install, he used

conda install -c conda-forge tensorflow

cs501r_f2016/openlabtf.txt · Last modified: 2016/09/20 15:15 by wingated