Numpy functions#

Similar to pandas methods, all numpy functions in the main namespace have been added as methods to MulDataFrame and MulSeries using the call method. For exmaple, np.log1p is added but numpy.linalg.matmul is not. There is some overlap between numpy functions and pandas methods. In this case, pandas methods take precedence over numpy functions.

>>> mf
(3, 2)    g  7  6
          f  5  3
             c  d
--------  ---------
   x  y      c  d
a  1  2   a  1  2
b  3  6   b  8  9
b  5  6   b  8  7
>>> mf.min()
(2,)     Series([], Name: min, dtype: object)
-------  --------------------------------------
   f  g     min
c  5  7  c    1
d  3  6  d    2
>>> mf.call(np.min)
1
>>> mf.log1p()
(3, 2)    g         7         6
          f         5         3
                    c         d
--------  ---------------------
   x  y             c         d
a  1  2   a  0.693147  1.098612
b  3  6   b  2.197225  2.302585
b  5  6   b  2.197225  2.079442
>>> mf.df.log1p()
AttributeError

pandas.DataFrame.min returns the minimum values of each column by default. numpy.min returns the minimum value of all values by default. The above example shows that mf.min uses pandas.DataFrame.min under the hood.

As explained in Proper function, functions must satisfied certain rules to be used with the call method. Users should check numpy API to decide whether a numpy function in the main namespace can be called as a method of MulSeries or MulDataFrame. As a matter of fact, most of the functions do work with the call method.

Show Source