pandas_fu.md

Reading a random subset of a huge file

df = pd.read_csv('file.csv', skiprows = n > 0 and np.random.rand() > 0.01)