How to vstack matrixs in numpy with different numbers of columns

[scidam]

You cann't do this using numpy's vstack, but desired behavior could be achieved using Pandas, e.g.

import numpy as np
a = np.array([["A1", "B1", "C1"], ["A1", "B1", "C1"]])
b = np.array([["A2", "B2"], ["A2", "B2"]])
c = np.array([["A3", "B3"], ["A3", "B3"]])
a = pd.DataFrame(a, columns=None)
b = pd.DataFrame(b, columns=None)
pd.concat([a,b]).fillna('').values