基于标签与基于位置的访问

A Series has both labels and positions, so there are two ways to grab a value: by its label or by its position number. Knowing which to use keeps your code clear. 🎯

Here is a small Series with string labels. We will use it to compare both access styles throughout this lesson.

The .loc accessor looks values up by their index label. You ask for a name and pandas returns that named value.

The .iloc accessor works by integer position instead, counting from zero like a normal list. Position 1 is the second value.

With our Series, s.loc['Bo'] and s.iloc[1] return the same number. They reach it differently: one by name, one by spot.

Position slices behave like list slices: the end is excluded. So iloc[0:2] returns the first two values only.

Label slices are different: with .loc, the end label is included. Asking from 'Ada' to 'Bo' returns both endpoints.

That inclusive end is the most common mix-up. Remember: iloc drops the last position, while loc keeps the last label. 🧠

Pass a list of labels to .loc to pull many values together. The result is a smaller Series in the order you asked.

If your index is integers, plain s[2] gets ambiguous. Using explicit .loc or .iloc removes all doubt about what you meant.

Reach for .loc when you know the name, and .iloc when you know the spot. Being explicit makes your selections impossible to misread.

Test your grip on the two accessors.

You now choose .loc for labels and .iloc for positions, and you remember that loc slices include the end while iloc slices do not. Solid! 🎉