Most CloudBank Classroom hubs start you in the same standard image, so a notebook that runs on one
hub generally runs on another. It is built from
cal
What is installed¶
Python, with the usual teaching and analysis stack: datascience, pandas, numpy, scipy,
matplotlib, seaborn, plotly, scikit-learn, statsmodels, sympy, geopandas and
openpyxl. pytorch, transformers and tokenizers are there too, which is enough for small
models on CPU.
JupyterLab, with jupyterlab-git, jupytext, jupyterlab_rise for presenting notebooks,
jupyter-archive for downloading folders, and the resource indicator described in
Memory and Storage Limits.
Course tooling: nbgitpuller for distributing materials, and otter-grader.
Other languages and editors. R and RStudio are available through the launcher, as is VS Code
via code-server. xeus-cling provides a C++ kernel.
Command-line tools for terminal work: git, curl, wget, vim, nano, htop, tree,
rsync, build tools (build-essential, cmake), debuggers (gdb, valgrind), and pandoc
plus a TeX installation so notebooks export to PDF.
Checking what you have¶
The surest answer comes from the environment itself rather than from this page. In a notebook:
import pandas as pd
pd.__version__Or in a terminal, to see everything at once:
pip listIf something you need is missing, Requesting Packages covers both the quick fix and the permanent one.