update text to match the new example; closes #1

content/version-control/branching-and-committing.md

@@ -49,7 +49,7 @@ Illustration of what we want to achieve in this exercise.
 1. First create a new branch and then either add a new file or modify an
    existing file and commit the change.  Make sure that you now work **on your
    copy** of the example repository. In your new commit you can share a Git or
-   programming trick you like.
+   programming trick you like or improve the documentation.
 1. In a new commit, modify the file again.
 1. Switch to the `main` branch and create a commit there.
 1. Browse the network and locate the commits that you just created ("Insights" -> "Network").
@@ -58,6 +58,16 @@ Illustration of what we want to achieve in this exercise.
 1. Try to rename the branch that you created and then browse the network again.
 1. Try to create a tag for one of the commits that you created (on GitHub,
    create a "release").
+1. Optional: If this was too easy, you can try to create a new branch and on
+   this branch work on one of these new features:
+   - The random seed is now set to a specific number (42). Make it possible to
+     set the seed to another number or to turn off the seed setting via the command line interface.
+   - Move the code that does the majority vote to an own function.
+   - Write a test for the new majority vote function.
+   - The `num_neighbors` in the code needs to be odd. Change the code so that it stops with an error message if an even number is given.
+   - Add type annotations to some functions.
+   - When calling the `scatter_plot` function, call the function with named arguments.
+   - Add example usage to README.md.
 :::
 
 The solution below goes over most of the answers, and you are
@@ -208,7 +218,7 @@ it makes sense later on.
 ::::{group-tab} GitHub
 A nice way to compare branches is to add `/compare` to the URL of the repository,
 for example (replace USER):
-https://github.com/**USER**/planets/compare
+https://github.com/**USER**/classification-task/compare
 ::::
 
 ::::{group-tab} VS Code
@@ -241,7 +251,7 @@ This is similar to above, but not only between branches.
 ::::{group-tab} GitHub
 Following the `/compare`-trick above, one can compare commits on GitHub by
 adjusting the following URL:
-https://github.com/**USER**/planets/compare/**VERSION1**..**VERSION2**
+https://github.com/**USER**/classification-task/compare/**VERSION1**..**VERSION2**
 
 Replace `USER` with your username and `VERSION1` and `VERSION2` with a commit hash or branch name.
 Please try it out.

content/version-control/browsing.md

@@ -55,15 +55,15 @@ It is also possible to do this: to clone a forked repository to your computer.
 
 At all times you should be aware of if you are looking at **your repository**
 or the **upstream repository** (original repository):
-- Your repository: https://github.com/**USER**/planets
-- Upstream repository: https://github.com/**workshop-material**/planets
+- Your repository: https://github.com/**USER**/classification-task
+- Upstream repository: https://github.com/**workshop-material**/classification-task
 
 :::{admonition} How to create a fork
-1. Go to the repository view on GitHub: <https://github.com/workshop-material/planets>
+1. Go to the repository view on GitHub: <https://github.com/workshop-material/classification-task>
 1. First, on GitHub, click the button that says "Fork". It is towards
    the top-right of the screen.
 1. You should shortly be redirected to your copy of the repository
-   **USER/planets**.
+   **USER/classification-task**.
 :::
 
 :::{instructor-note}
@@ -95,7 +95,7 @@ Work on this by yourself or in pairs.
 In this case you will work on a fork.
 
 You only need to open your own view, as described above.  The browser
-URL should look like https://github.com/**USER**/planets, where
+URL should look like https://github.com/**USER**/classification-task, where
 USER is your GitHub username.
 ::::
 
@@ -110,7 +110,7 @@ locally.
 open), go to File → New Window.
 1. Under "Start" on the screen, select "Clone Git Repository...". Alternatively
    navigate to the "Source Control" tab on the left sidebar and click on the "Clone Repository" button.
-1. Paste in this URL: `https://github.com/USER/planets`, where
+1. Paste in this URL: `https://github.com/USER/classification-task`, where
    `USER` is your username.  You can copy this from the browser.
 1. Browse and select the folder in which you want to clone the
    repository.
@@ -134,30 +134,30 @@ locally.
    (`cd ~/code` for example, if the `~/code` directory is where you
    store your files).
 1. Run the following command: `git clone
-   https://github.com/USER/planets`, where `USER` is your
+   https://github.com/USER/classification-task`, where `USER` is your
    username.  You might need to use a SSH clone command instead of
    HTTPS, depending on your setup.
-1. Change to that directory: `cd planets`
+1. Change to that directory: `cd classification-task`
 ::::
 :::::
 ::::::
 
 
 :::{exercise} Exercise: Browsing an existing project (20 min)
 
-Browse the [example project](https://github.com/workshop-material/planets) and
+Browse the [example project](https://github.com/workshop-material/classification-task) and
 explore commits and branches, either on a fork or on a clone.  Take notes and
 prepare questions.  The hints are for the GitHub path in the browser.
 
 1. Browse the **commit history**: Are commit messages understandable?
    (Hint: "Commit history", the timeline symbol, above the file list)
 1. Compare the commit history with the **network graph** ("Insights" -> "Network"). Can you find the branches?
-1. Try to find the **history of commits for a single file**, e.g. `simulate.py`.
+1. Try to find the **history of commits for a single file**, e.g. `generate_predictions.py`.
    (Hint: "History" button in the file view)
-1. **Which files include the word "position"**?
+1. **Which files include the word "training"**?
    (Hint: the GitHub search on top of the repository view)
-1. In the `simulate.py` file,
-   find out who modified the "time step"
+1. In the `generate_predictions.py` file,
+   find out who modified the evaluation of "majority_index"
    last and **in which commit**.
    (Hint: "Blame" view in the file view)
 1. Can you use this code yourself? **Are you allowed to share
@@ -178,7 +178,7 @@ The most basic thing to look at is the history of commits.
 
 * This is visible from a button in the repository view.  We see every
   change, when, and who has committed.
-* Every change has a unique identifier, such as `244c993`.  This can
+* Every change has a unique identifier, such as `79ce3be`.  This can
   be used to identify both this change, and the whole project's
   version as of that change.
 * Clicking on a change in the view shows more.
@@ -263,12 +263,12 @@ for a single file.
 :::::{tabs}
 
 ::::{group-tab} GitHub
-Navigate to the file view: Main page → simulate.py.
+Navigate to the file view: Main page → generate_predictions.py.
 Click the "History" button near the top right.
 ::::
 
 ::::{group-tab} VS Code
-Open simulate.py file in the editor.  Under the file browser,
+Open generate_predictions.py file in the editor.  Under the file browser,
 we see a "Timeline" view there.
 ::::
 
@@ -277,14 +277,14 @@ The `git log` command can take a filename and provide the log of only
 a single file:
 
 ```
-$ git log simulate.py
+$ git log generate_predictions.py
 ```
 ::::
 
 :::::
 
 
-### (4) Which files include the word "position"?
+### (4) Which files include the word "training"?
 
 Version control makes it very easy to find all occurrences of a
 word or pattern. This is useful for things like finding where functions or
@@ -293,7 +293,7 @@ variables are defined or used.
 :::::{tabs}
 ::::{group-tab} GitHub
 We go to the main file view.  We click the Search magnifying
-class at the very top, type "position", and click enter. We see every
+class at the very top, type "training", and click enter. We see every
 instance, including the context.
 
 :::{admonition} Searching in a forked repository will not work instantaneously!
@@ -306,16 +306,16 @@ Start it, continue with other steps, then come back to this.
 
 ::::{group-tab} VS Code
 If you use the "Search" magnifying class on the left sidebar, and
-search for "position" it shows the occurrences in every file. You can
+search for "training" it shows the occurrences in every file. You can
 click to see the usage in context.
 ::::
 
 ::::{group-tab} Command line
 `grep` is the command line tool that searches for lines matching a term
 ```console
-$ git grep position          # only the lines
-$ git grep -C 3 position     # three lines of context
-$ git grep -i position       # case insensitive
+$ git grep training          # only the lines
+$ git grep -C 3 training     # three lines of context
+$ git grep -i training       # case insensitive
 ```
 ::::
 
@@ -344,8 +344,8 @@ line version, after opening a terminal.
 ::::{group-tab} Command line
 These two commands are similar but have slightly different output.
 ```console
-$ git annotate simulate.py
-$ git blame simulate.py
+$ git annotate generate_predictions.py
+$ git blame generate_predictions.py
 ```
 ::::
 

content/version-control/motivation.md

@@ -33,17 +33,17 @@ of them?):
 
 ## Demonstration
 
-- Example repository: <https://github.com/workshop-material/planets>
+- Example repository: <https://github.com/workshop-material/classification-task>
 - Commits are like **snapshots** and if we break something we can go back to a
   previous snapshot.
 - Commits carry **metadata** about changes: author, date, commit message, and
   a checksum.
 - **Branches** are like parallel universes where you can experiment with
   changes without affecting the default branch:
-  <https://github.com/workshop-material/planets/network>
+  <https://github.com/workshop-material/classification-task/network>
   ("Insights" -> "Network")
 - With version control we can **annotate code**
-  ([example](https://github.com/workshop-material/planets/blame/main/simulate.py)).
+  ([example](https://github.com/workshop-material/classification-tas/blame/main/generate_predictions.py)).
 - **Collaboration**: We can fork (make a copy on GitHub), clone (make a copy
   to our computer), review, compare, share, and discuss.
 - **Code review**: Others can suggest changes using pull requests or merge
@@ -70,15 +70,15 @@ Image created using <https://gopherize.me/>
 
 - **Collaboration**: review, compare, share, discuss
 
-- [Example network graph](https://github.com/workshop-material/planets/network)
+- [Example network graph](https://github.com/workshop-material/classification-task/network)
 
 
 ## Talking about code
 
 Which of these two is more practical?
-1. "Clone the code, go to the file 'simulate.py', and search for 'force_between_planets'.
-   Oh! But make sure you use the version from September 2024."
-1. Or I can send you a permalink: <https://github.com/workshop-material/planets/blob/1343ac0/simulate.py#L31C5-L39>
+1. "Clone the code, go to the file 'generate_predictions.py', and search for 'majority_index'.
+   Oh! But make sure you use the version from January 2025."
+1. Or I can send you a permalink: <https://github.com/workshop-material/classification-task/blob/79ce3be8/generate_predictions.py#L25-L28>
 
 
 ## What we typically like to snapshot

content/version-control/practical-advice.md

@@ -27,7 +27,7 @@ this is based on a discussion in #123
 - **Why something was changed is more important than what has changed.**
 - Cross-reference to issues and discussions if possible/relevant.
 - Bad commit messages: "fix", "oops", "save work"
-- Bad examples: [http://whatthecommit.com](http://whatthecommit.com)
+- Just for fun, a page collecting bad examples: [http://whatthecommit.com](http://whatthecommit.com)
 - Write commit messages that will be understood
   15 years from now by someone else than you. Or by your future you.
 - Many projects start out as projects "just for me" and end up to be successful projects

content/version-control/sharing.md

@@ -297,3 +297,5 @@ accessible**, consider making your code **citable and persistent**: Get a
 persistent identifier (PID) such as DOI in addition to sharing the code
 publicly, by using services like [Zenodo](https://zenodo.org) or similar
 services.
+
+More about this in {doc}`../publishing`.