Reputation: 219
I want to implement the CSCmatrix (Compressed Sparse Coloumns) for doing this I need to read matrix given in this form :
{{1,2,3,4},{5,6,7,8}};
reading columns by columns! and not row by row (in this case i can use a simple costructor like in the follow )
template<typename T>
inline constexpr CSRmatrix<T>::CSRmatrix(std::initializer_list<std::initializer_list<T>>&& row ) noexcept
{
this->rows = row.size();
auto itr = *(row.begin());
this->cols = itr.size();
std::size_t i=0, j=0, w=0;
ia_.resize(rows+1);
ia_[0] = 0;
for(auto & r : row)
{
j=0 ; w =0 ;
for(auto & c : r)
{
if( c != 0.0 )
{
a_.push_back(c);
ja_.push_back(j);
w++;
}
j++;
}
i++;
ia_[i] = ia_[i-1] + w ;
}
}
this read row by row and stored the matrix in CSR format (compressed sparse row) could somebody help me about ? I have not idea ! thanks in advance
my data will be is stored in a single vector in this order {1,5,2,6,3,7,4,8}
Upvotes: 0
Views: 185
Reputation: 31599
If your data is declared as say vector<vector<T>> data
, you can simply put data(v)
in the initialization. But if you want to switch the rows and columns then use a temporary variable and switch the rows and columns.
Edit, or use arr
to put data in a single array
This code assumes all rows in the initialization data are the same size
template<typename T> class matrix
{
public:
//std::vector<std::vector<T>> data;
std::vector<T> arr;
matrix(std::initializer_list<std::vector<T>> v) //: data(v)
{
if(!v.size()) return; //invalid input
std::vector<std::vector<T>> temp(v);
//test to make sure all row are the same size
//use the size of row[0] for all other rows:
for(size_t c = 0; c < temp[0].size(); c++)
{
for(size_t r = 0; r < temp.size(); r++)
{
if(c < temp[r].size())
arr.push_back(temp[r][c]);
//else {invalid input}
}
}
for(size_t c = 0; c < temp[0].size(); c++)
for(size_t r = 0; r < temp.size(); r++)
arr.push_back(temp[r][c]);
}
};
int main()
{
matrix<int> m = { { 1,2,3,4 },{ 5,6,7,8 } };
for(auto e : m.arr) std::cout << e << ",";
std::cout << "\n";
return 0;
}
Output
1,5,2,6,3,7,4,8,
Upvotes: 1