pymrio.build_agg_matrix

pymrio.build_agg_matrix(agg_vector, pos_dict=None)

Aggregation based on mapping given in input as numerical or str vector.

Parameters

• agg_vector (list or vector like numpy ndarray) – This can be row or column vector. Length m with position given for n and -1 if values should not be included or length m with id_string for the aggregation

• pos_dict (dictionary) – (only possible if agg_vector is given as string) output order for the new matrix must be given as dict with ‘string in agg_vector’ = pos (as int, -1 if value should not be included in the aggregation)

Returns

agg_matrix – Aggregation matrix with shape (n, m) with n (rows) indicating the new classification and m (columns) the old classification

Return type

numpy ndarray

Examples

Assume an input vector with either

>>> inp1 = np.array([0, 1, 1, 2])

or

>>> inp2 = ['a', 'b', 'b', 'c']

inp1 and inp2 are equivalent, the entries are just indicators.

>>> build_agg_matrix(inp1)
>>> build_agg_matrix(inp2)

Returns

>>> array([[1., 0., 0., 0.],
>>>        [0., 1., 1., 0.],
>>>        [0., 0., 0., 1.]])

The order can be defined by a dictionary, thus

>>> pymrio.build_agg_matrix(np.array([1, 0, 0, 2]))

is equivalent to

>>> pymrio.build_agg_matrix(['b', 'a', 'a', 'c'], dict(a=0,b=1,c=2))

>>> array([[0., 1., 1., 0.],
>>>        [1., 0., 0., 0.],
>>>        [0., 0., 0., 1.]])