Data Fields | |
| FloatComplex * | data |
| all data values are complex | |
| int | nX |
| number of rows in xy-plane | |
| int | nY |
| number of columns in xy-plane | |
| int | nZ |
| depth | |
| int | nT |
| number of measurements | |
| int | nV |
| if each data point is now a vector, then nV is the dimension of the vector usually (x,y,z) | |
| MatrixType | mtype |
| type of matrix: always complex | |
| bool | shift |
| boolean: whether to shift data after inverse Fourier Transform | |
| int | nYOriginal |
| original dimension of y, because sometimes the xy-plane has been Fourier transformed | |
Four dimensional matrix that keeps complete input and output data. This is a main datatype for the 3d problem. The input kernel data, input map data, solution data, averaging kernel data and covariance data are all kept in a matrix4d object. During the inversion, pointers to 2d sub-matrices are passed around and inserted to accumulate full solutions. Map-depth data is kept contiguous to make the inversion run faster. This is done at the expense of data in the xy-plane. Since input and output to FITS files uses the xy-plane, the expense is incurred during I/O which is inherently slow anyway.
| FloatComplex* MAP_DEPTH_MATRIX::data |
all data values are complex
type of matrix: always complex
number of measurements
if each data point is now a vector, then nV is the dimension of the vector usually (x,y,z)
number of rows in xy-plane
number of columns in xy-plane
original dimension of y, because sometimes the xy-plane has been Fourier transformed
depth
boolean: whether to shift data after inverse Fourier Transform
1.7.0
