Login to the Instance

After waiting for a couple minutes for your instance to launch, you can access it! In order to ssh into your instance run the following command in gcloud:

gcloud compute ssh --project $PROJECT_NAME --zone $ZONE jupyter@$INSTANCE_NAME -- -L 8080:localhost:8080

On the instance you now need to install some packages that are required for the course; the following installs a Python wrapper for use of the OpenCV library:

sudo pip install opencv-python 
sudo pip3 install opencv-python 

Finally, you'll need to clone a Github repository. Run the following command to clone the first project repository that has all the project notebooks and resources:

git clone https://github.com/udacity/P1_Facial_Keypoints

Note: These GPU instances only support JupyterLab as opposed to plain Jupyter notebooks. This just means the interface you're used to in the classroom notebooks will be slightly different (and with a nav bar fornavigating between files)!

Once you're finished working on a project or just taking a break during training, don't forget to shutdown your instance.