The regrid operator selects non-overlapping subarrays and performs aggregation on them.

Synopsis

regrid(array,grid_1 [, grid_2,...,grid_N],
aggregate_call_1 [, aggregate_call_2,...,aggregate_call_N])

Summary

The regrid operator partitions the cells in the input array into blocks. For each block, regrid applies a specific aggregate operation over the value(s) of some attribute in each block.

Note that the chunk size does not have to be a multiple of the grid size, and the grids may span array chunks. However, processing is most efficient if chunk size is a multiple of the grid size, and grids do not span array chunks.

Inputs

The regrid operator takes the following arguments:

array: A source array with one or more attributes and one or more dimensions.
grid_n: A list of block sizes, one per dimension.
aggregate_call_n: One or more aggregate calls.

Example

To divide a 4×4 array into 4 equal partitions and calculates the average of each one, do the following. This process is known as spatial averaging.

Create an array m4x4:
AFL% CREATE ARRAY m4x4 <val:double>[i=0:3; j=0:3];

The output is:
Query was executed successfully
Run this command, which increments val from 0 to 15:
AFL% store(build (m4x4, i*4+j), m4x4);

The output is:
{i,j} val {0,0} 0 {0,1} 1 {0,2} 2 {0,3} 3 {1,0} 4 {1,1} 5 {1,2} 6 {1,3} 7 {2,0} 8 {2,1} 9 {2,2} 10 {2,3} 11 {3,0} 12 {3,1} 13 {3,2} 14 {3,3} 15
Regrid m4x4 into four (2*2) partitions and find the average of each partition.
AFL% regrid(m4x4, 2,2, avg(val) as Average);

The output is:
{i,j} Average {0,0} 2.5 {0,1} 4.5 {1,0} 10.5 {1,1} 12.5
The regrid of 2x2 causes the following shape:
and the averages are computed as:
(0 + 1 + 4 + 5) / 4 = 2.5
(2 + 3 + 6 + 7) / 4 = 4.5
(8 + 9 + 12 + 13) / 4 = 10.5
(10 + 11 + 14 + 15) / 4 = 12.5
which results in:
Remove the m4x4 array
AFL% remove(m4x4);

The output is:
Query was executed successfully