JupyterHub examples

Notebooks, databases and ppt to run examples. The recorded lunchbox session can be found here

Python notebook

To be able to run the Python notebook, you need to install a correct conda environment yourself. In this case an environment called p39env:

conda create -n p39env python=3.9 tensorflow-gpu ipykernel

Add this kernel to your VSC_HOME/.local folder:

source activate p39env
python -m ipykernel install  --prefix=${VSC_HOME}/.local/ --name 'p39env'

Now, the kernel should appear in your kernel list and work fine. The matplotlib package is not yet installed, but will be installed in the notebook.

In the notebook, we use a locally stored pickle dataset to show that we can read from VSC_DATA. As it is too big to save ont GitHub, you can download and save it as follows (in a Python interpreter):

import os
import pickle
import tensorflow as tf

os.chdir('jupyterHUB_examples')
os.mkdir('py_notebook_ex')

dt = tf.keras.datasets.mnist.load_data()
with open('mnist.pickle', 'wb') as f:
    pickle.dump(dt, f)

Now you should be able to run the notebook without any problems.

R notebook

Similar as for the Python notebook, first install a R conda environment:

conda create -n r41env -c conda-forge r-base r-ggplot2 r-factoextra jupyter_client r-irkernel

Add the kernel to your VSC_HOME/.local folder:

source activate r41env
Rscript -e 'IRkernel::installspec(prefix="${VSC_HOME}/.local/", name="r41env", 	displayname="r41env")’

This kernel should be available in your kernel list now as well.