from IPython.display import display df.groupby("colName").apply(display)
grp = df.groupby['colName'] grp.describe()