FML

Fundamentals of Machine Learning - course notes, code and auxiliary material

Purpose

Use to store and track Jupyter notebooks and other core material.

Licencing

GPL 3.0 applies by default to all files on this repository that do not explicitly claim their own licence (e.g. Creative Commons).

Binder

Use these notebooks on https://mybinder.org/v2/gh/variationalform/FML.git/HEAD...

Git - usage notes

This repo was originally set up on git hub and then, locally,

cd /to/parent/directory
git clone git@github.com:variationalform/FML.git

Then, for example,

cd FML
git add notebooks/*
git add *
git add .gitignore 
git reset    # to undo and start again
git status
git commit -m 'Initial commit for notebooks'
git push
git log -p
git log --oneline

Note that .gitignore is a hidden file. It's worth collaborating with it though.

From https://stackoverflow.com/questions/54825213/git-log-source-in-pretty-format, this is a stylised log output.

git log --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %s %Cgreen(%cr) %C(bold blue)<%an>%Creset' --abbrev-commit

Some More Useful Stuff

jupyter notebook list # obtain the notebook token if it gets requested.
git reflog
git fetch && git diff main origin/main
git fetch && git diff main origin/main --compact-summary

If changes are made remotely, sync them locally and see what changed with,

git pull            # changes the lovcal branch
git fetch           # doesn't change local branch, just remote
git log --patch     # see what changed
git whatchanged     # deprecated
git log --pretty=oneline
git log --pretty=short
git log --pretty=medium
git log --pretty=full
git log --pretty=fuller

If changes are made in a pulled branch (e.g. by https://github.com/xiaochuany, try this:

# Step 1: From your project repository, check out a new branch and test the changes.

git checkout -b xiaochuany-main main
git pull https://github.com/xiaochuany/FML.git main

# Step 2: Merge the changes and update on GitHub.

git checkout main
git merge --no-ff xiaochuany-main
git push origin main

Creating a presentation from a jupyter notebook:

in the notebook select
view -> cell Toolbar -> slideshow
For each cell select slide type

Then:
# command to convert notebook to presentation
jupyter nbconvert --to slides presentation.ipynb

In the browser do this...
file:///home/.../presentation.slides.html?print-pdf#/

which will then be available to save as PDF.

Main reference for the above was https://mljar.com/blog/jupyter-notebook-presentation/