mirror of
https://github.com/josdejong/mathjs.git
synced 2025-12-08 19:46:04 +00:00
89 lines
3.1 KiB
JavaScript
89 lines
3.1 KiB
JavaScript
// Only use native node.js API's and references to ./lib here, this file is not transpiled!
|
|
const math = require('../../../../../lib/bundleAny')
|
|
const { createCsAmd } = require('../../../../../lib/function/algebra/sparse/csAmd')
|
|
|
|
const assert = require('assert')
|
|
const approx = require('../../../../../tools/approx')
|
|
const market = require('../../../../../tools/matrixmarket')
|
|
|
|
const { add, multiply, transpose } = math
|
|
const csAmd = createCsAmd({ add, multiply, transpose })
|
|
|
|
describe('csAmd', function () {
|
|
it('should approximate minimum degree ordering, 48 x 48, natural ordering (order=0), matrix market', function (done) {
|
|
// import matrix
|
|
market.import('tools/matrices/bcsstk01.mtx')
|
|
.then(function (m) {
|
|
// symbolic ordering and analysis, order = 0
|
|
const q = csAmd(0, m)
|
|
|
|
// verify
|
|
assert(q === null)
|
|
|
|
// indicate test has completed
|
|
done()
|
|
})
|
|
.catch(function (error) {
|
|
// indicate test has completed
|
|
done(error)
|
|
})
|
|
})
|
|
|
|
it('should approximate minimum degree ordering, 48 x 48, amd(A+A\') (order=1), matrix market', function (done) {
|
|
// import matrix
|
|
market.import('tools/matrices/bcsstk01.mtx')
|
|
.then(function (m) {
|
|
// symbolic ordering and analysis, order = 1
|
|
const q = csAmd(1, m)
|
|
|
|
// verify
|
|
approx.deepEqual(q, [10, 28, 29, 24, 0, 11, 30, 6, 23, 22, 40, 46, 42, 18, 4, 16, 34, 5, 9, 39, 21, 44, 45, 43, 15, 25, 26, 27, 3, 33, 41, 19, 20, 2, 38, 32, 1, 14, 8, 13, 37, 31, 12, 36, 17, 47, 35, 7])
|
|
|
|
// indicate test has completed
|
|
done()
|
|
})
|
|
.catch(function (error) {
|
|
// indicate test has completed
|
|
done(error)
|
|
})
|
|
})
|
|
|
|
it('should approximate minimum degree ordering, 48 x 48, amd(A\'*A) (order=2), matrix market', function (done) {
|
|
// import matrix
|
|
market.import('tools/matrices/bcsstk01.mtx')
|
|
.then(function (m) {
|
|
// symbolic ordering and analysis, order = 2
|
|
const q = csAmd(2, m, false)
|
|
|
|
// verify
|
|
approx.deepEqual(q, [26, 27, 25, 44, 9, 15, 21, 33, 39, 43, 45, 3, 29, 24, 28, 47, 6, 18, 36, 0, 1, 4, 20, 2, 10, 11, 12, 8, 14, 16, 7, 13, 17, 23, 30, 34, 38, 32, 31, 41, 35, 22, 19, 37, 40, 42, 46, 5])
|
|
|
|
// indicate test has completed
|
|
done()
|
|
})
|
|
.catch(function (error) {
|
|
// indicate test has completed
|
|
done(error)
|
|
})
|
|
})
|
|
|
|
it('should approximate minimum degree ordering, 48 x 48, amd(A\'*A) (order=3), matrix market', function (done) {
|
|
// import matrix
|
|
market.import('tools/matrices/bcsstk01.mtx')
|
|
.then(function (m) {
|
|
// symbolic ordering and analysis, order = 3
|
|
const q = csAmd(3, m, false)
|
|
|
|
// verify
|
|
approx.deepEqual(q, [26, 27, 25, 44, 9, 15, 21, 33, 39, 43, 45, 3, 29, 24, 28, 47, 6, 18, 36, 0, 1, 4, 20, 2, 10, 11, 12, 8, 14, 16, 7, 13, 17, 23, 30, 34, 38, 32, 31, 41, 35, 22, 19, 37, 40, 42, 46, 5])
|
|
|
|
// indicate test has completed
|
|
done()
|
|
})
|
|
.catch(function (error) {
|
|
// indicate test has completed
|
|
done(error)
|
|
})
|
|
})
|
|
})
|