Selecting with loc and iloc

Two ways to point at a row

pandas gives you two accessors and it matters which you use:

• .loc selects by label — the index value and the column name.
• .iloc selects by integer position, like a list.

df.loc[2]                    # the row labelled 2
df.loc[2, "city"]            # one cell
df.loc[1:3]                  # rows 1 to 3 INCLUSIVE — labels, not positions
df.loc[:, ["city", "temp"]]  # all rows, two columns

df.iloc[0]                   # the first row
df.iloc[0, 1]                # first row, second column
df.iloc[0:2]                 # the first two rows — exclusive, like a list

The trap worth memorising: .loc slices are inclusive of the end, .iloc slices are not. df.loc[1:3] gives three rows; df.iloc[1:3] gives two.

Your turn: pull out the first city by position, the temperature of the row labelled 2 by label, and the first two rows.