LinearSolve.jl/test/resolve.jl at 3d4cfa4d6aa322b2124cdd2f35e35cc0ae6dd689 · SciML/LinearSolve.jl · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
using LinearSolve, LinearAlgebra, SparseArrays, InteractiveUtils, Test
using LinearSolve: AbstractDenseFactorization, AbstractSparseFactorization,
    BLISLUFactorization, CliqueTreesFactorization,
    AMDGPUOffloadLUFactorization, AMDGPUOffloadQRFactorization,
    SparspakFactorization

# Function to check if an algorithm is mixed precision
function is_mixed_precision_alg(alg)
    alg_name = string(alg)
    return contains(alg_name, "32Mixed") || contains(alg_name, "Mixed32")
end

for alg in vcat(
        InteractiveUtils.subtypes(AbstractDenseFactorization),
        InteractiveUtils.subtypes(AbstractSparseFactorization)
    )
    if alg in [PardisoJL]
        ## Pardiso has extra tests in test/pardiso/pardiso.jl
        continue
    end
    @show alg
    if !(
            alg in [
                DiagonalFactorization,
                CudaOffloadFactorization,
                CudaOffloadLUFactorization,
                CudaOffloadQRFactorization,
                CUSOLVERRFFactorization,
                AppleAccelerateLUFactorization,
                MetalLUFactorization,
                FastLUFactorization,
                FastQRFactorization,
                CliqueTreesFactorization,
                BLISLUFactorization,
                AMDGPUOffloadLUFactorization,
                AMDGPUOffloadQRFactorization,
            ]
        ) &&
            (
            !(alg == AppleAccelerateLUFactorization) ||
                LinearSolve.appleaccelerate_isavailable()
        ) &&
            (!(alg == MKLLUFactorization) || LinearSolve.usemkl) &&
            (!(alg == OpenBLASLUFactorization) || LinearSolve.useopenblas) &&
            (!(alg == RFLUFactorization) || LinearSolve.userecursivefactorization(nothing)) &&
            (!(alg == RF32MixedLUFactorization) || LinearSolve.userecursivefactorization(nothing)) &&
            (!(alg == MKL32MixedLUFactorization) || LinearSolve.usemkl) &&
            (!(alg == AppleAccelerate32MixedLUFactorization) || Sys.isapple()) &&
            (!(alg == OpenBLAS32MixedLUFactorization) || LinearSolve.useopenblas) &&
            (!(alg == SparspakFactorization) || false) &&
            (
            !(alg == ParUFactorization) ||
                Base.get_extension(LinearSolve, :LinearSolveParUExt) !== nothing
        ) &&
            (
            !(alg == ElementalJL) ||
                Base.get_extension(LinearSolve, :LinearSolveElementalExt) !== nothing
        )
        A = [1.0 2.0; 3.0 4.0]
        alg in [
            KLUFactorization, UMFPACKFactorization, SparspakFactorization,
            ParUFactorization,
        ] &&
            (A = sparse(A))
        A = A' * A
        @show A
        alg in [CHOLMODFactorization] && (A = sparse(Symmetric(A, :L)))
        alg in [BunchKaufmanFactorization] && (A = Symmetric(A, :L))
        alg in [LDLtFactorization] && (A = SymTridiagonal(A))
        b = [1.0, 2.0]
        prob = LinearProblem(A, b)
        linsolve = init(
            prob, alg(), alias = LinearAliasSpecifier(alias_A = false, alias_b = false)
        )

        # Use higher tolerance for mixed precision algorithms
        expected = [-2.0, 1.5]
        if is_mixed_precision_alg(alg)
            @test solve!(linsolve).u ≈ expected atol = 1.0e-4 rtol = 1.0e-4
            @test !linsolve.isfresh
            @test solve!(linsolve).u ≈ expected atol = 1.0e-4 rtol = 1.0e-4
        else
            @test solve!(linsolve).u ≈ expected
            @test !linsolve.isfresh
            @test solve!(linsolve).u ≈ expected
        end

        A = [1.0 2.0; 3.0 4.0]
        alg in [
            KLUFactorization, UMFPACKFactorization, SparspakFactorization,
            ParUFactorization,
        ] &&
            (A = sparse(A))
        A = A' * A
        alg in [CHOLMODFactorization] && (A = sparse(Symmetric(A, :L)))
        alg in [BunchKaufmanFactorization] && (A = Symmetric(A, :L))
        alg in [LDLtFactorization] && (A = SymTridiagonal(A))
        linsolve.A = A
        @test linsolve.isfresh

        # Use higher tolerance for mixed precision algorithms
        if is_mixed_precision_alg(alg)
            @test solve!(linsolve).u ≈ expected atol = 1.0e-4 rtol = 1.0e-4
        else
            @test solve!(linsolve).u ≈ expected
        end
    end
end

A = Diagonal([1.0, 4.0])
b = [1.0, 2.0]
prob = LinearProblem(A, b)
linsolve = init(
    prob, DiagonalFactorization(),
    alias = LinearAliasSpecifier(alias_A = false, alias_b = false)
)
@test solve!(linsolve).u ≈ [1.0, 0.5]
@test solve!(linsolve).u ≈ [1.0, 0.5]
A = Diagonal([1.0, 4.0])
linsolve.A = A
@test solve!(linsolve).u ≈ [1.0, 0.5]

A = Symmetric(
    [
        1.0 2.0
        2.0 1.0
    ]
)
b = [1.0, 2.0]
prob = LinearProblem(A, b)
linsolve = init(
    prob, BunchKaufmanFactorization(),
    alias = LinearAliasSpecifier(alias_A = false, alias_b = false)
)
@test solve!(linsolve).u ≈ [1.0, 0.0]
@test solve!(linsolve).u ≈ [1.0, 0.0]
A = Symmetric(
    [
        1.0 2.0
        2.0 1.0
    ]
)
linsolve.A = A
@test solve!(linsolve).u ≈ [1.0, 0.0]

A = [
    1.0 2.0
    2.0 1.0
]
A = Symmetric(A * A')
b = [1.0, 2.0]
prob = LinearProblem(A, b)
linsolve = init(prob, CholeskyFactorization(), alias = LinearAliasSpecifier(alias_A = false, alias_b = false))
@test solve!(linsolve).u ≈ [-1 / 3, 2 / 3]
@test solve!(linsolve).u ≈ [-1 / 3, 2 / 3]
A = [
    1.0 2.0
    2.0 1.0
]
A = Symmetric(A * A')
b = [1.0, 2.0]
@test solve!(linsolve).u ≈ [-1 / 3, 2 / 3]