For a dataframe df, one can use any of the following: 1)len(df.index) 2)df.shape[0] 3)df[df.columns[0]].count() (slowest, but avoids counting NaN values in the first column)