Special topics in deep learning for big data

ssh access

If you have accout for server @HYU_PHY, you can access the server using the following command:

ssh -X -Y {YourID}@{ServerIP}

Environment set up

If you don't have accout, follow these steps to set up the necessary environment:

Prerequisite

• Python == 3.6.5
• cudnn == 7
• cudatoolkit == 9.0

Anaconda set up

You need to install anaconda.

# download conda installation file by wget
wget https://repo.anaconda.com/archive/Anaconda3-2022.10-Linux-x86_64.sh

sh Anaconda3-2022.10-Linux-x86_64.sh

# add conda path
export PATH=/opt/anaconda3/bin/:$PATH

In your conda environment, need to install tensorflow, Keras, Jupyter notebook, ipykernel. You can set up the environment using the provided 'env.yaml' file.

# create conda environment with env.yaml file
conda env create --file env.yaml
conda activate py36-cuda9.0 # this env name is set in env.yaml file.

Usage

Jupyter Notebook

Jupyter Notebook is a web-based interactive computing platform that allows you to create and share documents that contain live code, equations, visualizations, and narrative text.

To use Jupyter Notebook:

1. At the server

conda activate py36-cuda9.0
jupyter notebook --no-browser

You should check which port number is opened for jupyter notebook.

2. At your local computer

ssh -NL {Port}:localhost:{Port} {YourID}@{ServerIP}

3. Open a new browser and copy and paste the url from the step 1.