ERR: give better error message when providing wrong sparse matrix type in from_coo

[jorisvandenbossche]

pd.Series.sparse.from_coo only accepts a coo matrix format, but when giving it different matrix type, you get a not so useful error message:

In [11]: import scipy.sparse

In [12]: m = scipy.sparse.csr_matrix(np.array([[0,1],[0,0]]))

In [13]: pd.Series.sparse.from_coo(m)
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-13-b31c13c3a79c> in <module>
----> 1 pd.Series.sparse.from_coo(m)

~/scipy/pandas/pandas/core/arrays/sparse.py in from_coo(cls, A, dense_index)
   2015         from pandas import Series
   2016 
-> 2017         result = _coo_to_sparse_series(A, dense_index=dense_index,
   2018                                        sparse_series=False)
   2019         result = Series(result.array, index=result.index, copy=False)

~/scipy/pandas/pandas/core/sparse/scipy_sparse.py in _coo_to_sparse_series(A, dense_index)
    122     Convert a scipy.sparse.coo_matrix to a SparseSeries.
    123 
--> 124     Parameters
    125     ----------
    126     A : scipy.sparse.coo.coo_matrix

~/miniconda3/envs/dev37/lib/python3.7/site-packages/scipy/sparse/base.py in __getattr__(self, attr)
    687             return self.getnnz()
    688         else:
--> 689             raise AttributeError(attr + " not found")
    690 
    691     def transpose(self, axes=None, copy=False):

AttributeError: row not found

[fhoang7]

Happy to take a look at this!