Data Science improvements (first block)

docs/data-science/basics/setup.md

@@ -143,6 +143,18 @@ uv init --vcs none  # (1)!
     By default `--vcs git` is set, which initializes a git repository. Since
     git is not within the scope of this project, we set `--vcs` to none.
 
+???+ warning "Restart VS Code if command fails"
+
+    If the command returns an error saying `uv` was not found, close and reopen
+    VS Code. This allows your system to recognize the newly installed `uv` 
+    executable. Then run the command again.
+
+???+ tip "Navigate command history"
+
+    There's no need to re-type old commands. Press ++arrow-up++ / 
+    ++arrow-down++ to cycle through previously executed commands in the 
+    terminal.
+
 This initializes the project. `uv` creates a few files in your folder. Your
 workspace should look like this:
 

docs/data-science/data/basics.md

@@ -71,6 +71,26 @@ Examples include number of students (5) or age (22).
     A simple rule of thumb: If you can meaningfully have fractional values, it's
     continuous. If counting whole units makes more sense, it's discrete.
 
+???+ warning "Numbers aren't always numerical data"
+
+    Just because data is stored as numbers doesn't make it numerical. 
+    Consider ZIP codes, their mean is mathematically possible but conceptually 
+    meaningless.
+
+    ```python
+    zip_codes = pd.Series([6020, 1050, 6011, 1010])
+    print(f"Average ZIP code: {zip_codes.mean()}")  # Makes no sense!
+    ```
+
+    ```title=">>> Output"
+    Average ZIP code: 3522.75 
+    ```
+
+    If you can't meaningfully add, subtract or average the values, it's 
+    categorical data in disguise.
+
+    Other examples are customer IDs or coordinates.
+
 ### Categorical (Qualitative)
 
 Categorical data represents qualities or characteristics that place
@@ -160,8 +180,8 @@ How many rows and columns has the `penguin` dataset?
 - [ ] 5 rows and 8 columns
 - [x] 344 rows and 7 columns
 
-The data set has 344 rows (penguins) and 7 columns (features). Use `data.shape`
-to quickly get the datasets dimensions.
+The data set has 344 rows (penguins) and 7 columns (features). Use 
+`penguins.shape` to quickly get the datasets dimensions.
 </quiz>
 
 ???+ question "Identify attribute types"