build_AF tries to set elements of sparse matrix using loc which is not allowed

[majeau-bettez]

build_AF tries to set elements of sparse matrix using loc which is not allowed

[jakobsarthur]

So the above solution is not quite feasible as we're dealing with rows in the code instead of columns. An easy fix seems to be to take the transpose but that takes too long (13.5 s +/- ~100 ms) on my machine. So ended up with the following solution which is definitely more memory intensive (the product of self.A.sparse.to_dense() takes almost 2000MB instead of ~124MB).
dtypes = self.A.dtypes
self.A = self.A.sparse.to_dense()
self.A.loc[bo_cutoff,:] = self.A.loc[bo_cutoff,:].abs()
self.A = self.A.astype(dtypes)

[andthewatersays]

Hello,

I have the same problem with the "TypeError: SparseArray does not support item assignment via setitem".
I have tried to add as you suggest this 4 lines (dtypes = self.A.dtypes
self.A = self.A.sparse.to_dense()
self.A.loc[bo_cutoff,:] = self.A.loc[bo_cutoff,:].abs()
self.A = self.A.astype(dtypes)) but now I have a new error :

ValueError Traceback (most recent call last)
in
----> 1 parser.ecospold_to_Leontief(with_absolute_flows=True)

~/opt/anaconda3/lib/python3.7/site-packages/ecospold2matrix/ecospold2matrix.py in ecospold_to_Leontief(self, fileformats, with_absolute_flows, lci_check, rtol, atol, imax, characterisation_file, ardaidmatching_file)
432
433 # Save system to file
--> 434 self.save_system(fileformats)
435
436 # Read/load lci cummulative emissions and perform quality check

~/opt/anaconda3/lib/python3.7/site-packages/ecospold2matrix/ecospold2matrix.py in save_system(self, file_formats)
2202 STR_header = STR_header.reshape((1, -1))
2203
-> 2204 C = scipy.sparse.csr_matrix(self.C.sparse.to_coo())
2205 IMP_header = self.IMP.columns.values
2206 IMP_header = IMP_header.reshape((1, -1))

~/opt/anaconda3/lib/python3.7/site-packages/pandas/core/arrays/sparse.py in to_coo(self)
2225 from scipy.sparse import coo_matrix
2226
-> 2227 dtype = find_common_type(self._parent.dtypes)
2228 if isinstance(dtype, SparseDtype):
2229 dtype = dtype.subtype

~/opt/anaconda3/lib/python3.7/site-packages/pandas/core/dtypes/cast.py in find_common_type(types)
1136
1137 if len(types) == 0:
-> 1138 raise ValueError("no types given")
1139
1140 first = types[0]

ValueError: no types given

Did you manage to solve your error or did you have other errors as I have ?