Deep Learning Oracle

Installing Anaconda in OCI GPU instance

Now that you know how to create an Oracle Cloud Infrastructure GPU instance, the next steps are install Anaconda and use Jupyter Notebook to develop or test your AI projects.

First of all, go to your Oracle Cloud Account and add the following Ingress Rule in your Security List (Networking > Virtual Cloud Networks > Virtual Cloud Network Details > Security Lists > Security List Details).

Using a Terminal, access your Ubuntu instance and download the latest version of Ananconda. Today the latest version is Anaconda 2018.12.

curl -O

When the download is complete, start the installation.

bash -b

Run the following commands to use the conda command.

echo -e '\nexport PATH="$HOME/anaconda3/bin:$PATH"' >> $HOME/.bashrc

source ~/.bashrc

Anaconda has been installed! We can continue to configure the environment.

Create a new environment with Python 3.6, Tensorflow (GPU version) and Keras.

conda create -n myenv python=3.6 tensorflow-gpu keras -y

conda install nb_conda -y

Update the firewall using iptables to open the port 8888.

sudo iptables -I INPUT -p tcp -s --dport 8888 -j ACCEPT

sudo service netfilter-persistent save

Generate the Jupyter configuration and open it.

jupyter notebook --generate-config

vi /$HOME/.jupyter/

Add the following code at the beginning of the file.

c = get_config()
c.NotebookApp.ip = ''
c.NotebookApp.open_browser = False
c.NotebookApp.port = 8888

Run the following command to add a password to access Jupyter environment.

jupyter notebook password

Start the Jupyter Notebook.

jupyter notebook &

Done! To access your Jupyter environment, copy and paste the URL into a browser (do not forget to use public IP).

Now you have a environment to test your projects!

Have a good time!