create new instance using ubuntu trusty (14.04)

make sure to give it lots of memory (try 8 vCPU’s and 52 gigabytes memory)

once the instane is created and running, launch an ssh terminal

install a compiler:

sudo apt-get install gcc

install git:

sudo apt-get install git

you can now clone in the magenta repository at any time using:

git clone https://github.com/tensorflow/magenta.git

now run the installer for the magenta environment (if this doesn’t work, try the manual instructions here):

curl https://raw.githubusercontent.com/tensorflow/magenta/master/magenta/tools/magenta-install.sh > magenta-install.sh

bash magenta-install.sh

source ~/.bashrc

this creates a virtual environment for magenta that you can now launch by typing

source activate magenta

you should see (magenta) in front of the terminal prompt now.

you can type conda list to see a list of installed packages.

for any dependencies required by magenta, make sure to activate the magenta environment before installing. for sketch_rnn, you need to install a specific version of svgwrite:

conda install -c omnia svgwrite=1.1.6

now, if you want to open a jupyter notebook, activate the magenta environment and then type:

jupyter notebook --no-browser --allow-root

which will give you a url with a token that you can paste in a browser…

…but first you need to open an ssh tunnel from your local terminal by typing:

ssh -i .ssh/google_compute_engine -L 8888:localhost:8888 yourusername@your-instance-external-ip

(you’ll need to have set up ssh keys for this to work properly - see how to create a cloud compute instance for more info)