Working with rows and columns

Adding and changing. df['NewCol'] = values adds a new column (or updates it, if the label already exists). df.loc['label'] = values adds (or updates) a row. If you give too few values, Python raises a ValueError.

Deleting. drop() removes rows or columns: drop(label, axis=0) deletes rows; drop([labels], axis=1) deletes columns. For example, df = df.drop(['Hindi', 'Art'], axis=1) removes two columns.

Renaming. rename({old: new}, axis='index') renames row labels, and axis='columns' renames column labels. Labels you do not mention stay as they are.

Selecting.

• loc[row, column] selects by labels: df.loc['Maths', 'Asha'];
• label slices include the end: df.loc['Maths':'Science'];
• Boolean conditions: df.loc['Maths'] > 80 gives True or False for each column; a list of True/False values picks rows: df.loc[[True, False, True]] keeps rows 1 and 3.

Remember: axis 0 means rows, axis 1 means columns.