Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

affinity propagation: add comparison test with sklearn implementation #132

Merged
merged 3 commits into from
Nov 20, 2018
Merged

affinity propagation: add comparison test with sklearn implementation #132

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
248b886
add comparison with python results to illustrate the bug: https://git…
xiuliren Nov 19, 2018
1fe1f3f
compare label using rand index
xiuliren Nov 20, 2018
d6dbafc
make test code more formal
xiuliren Nov 20, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
84 changes: 65 additions & 19 deletions test/affprop.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,32 +5,78 @@ using Distances
using Clustering
using LinearAlgebra
using Random
using Statistics

@testset "affinityprop() (affinity propagation)" begin

Random.seed!(34568)
Random.seed!(34568)

d = 10
n = 500
x = rand(d, n)
S = -pairwise(Euclidean(), x, x)
d = 10
n = 500
x = rand(d, n)
S = -pairwise(Euclidean(), x, x)

# set diagonal value to median value
S = S - diagm(0 => diag(S)) + median(S)*I
# set diagonal value to median value
S = S - diagm(0 => diag(S)) + median(S)*I

R = affinityprop(S)
R = affinityprop(S)

@test isa(R, AffinityPropResult)
k = length(R.exemplars)
@test 0 < k < n
@test length(R.assignments) == n
@test all(R.assignments .>= 1) && all(R.assignments .<= k)
@test all(R.assignments[R.exemplars] .== collect(1:k))
@test isa(R, AffinityPropResult)
k = length(R.exemplars)
@test 0 < k < n
@test length(R.assignments) == n
@test all(R.assignments .>= 1) && all(R.assignments .<= k)
@test all(R.assignments[R.exemplars] .== collect(1:k))

@test length(R.counts) == k
@test sum(R.counts) == n
for i = 1:k
@test R.counts[i] == count(==(i), R.assignments)
end
@test length(R.counts) == k
@test sum(R.counts) == n
for i = 1:k
@test R.counts[i] == count(==(i), R.assignments)
end

#= compare with python result
the reference assignments were computed using python sklearn:
```julia
using PyCall

@pyimport sklearn.cluster as cl
af = cl.AffinityPropagation(affinity="precomputed")[:fit]( S )

ref_assignments = af[:labels_] .+ 1
```
=#
ref_assignments = [11, 47, 23, 1, 3, 8, 54, 2, 6, 29, 11, 43, 2, 20, 13, 40,
3, 48, 3, 32, 19, 59, 4, 28, 59, 24, 5, 36, 6, 23, 49, 19,
42, 14, 40, 29, 20, 49, 53, 48, 51, 55, 57, 32, 32, 10, 48,
12, 13, 7, 1, 57, 52, 7, 3, 47, 43, 10, 53, 40, 52, 8, 42,
15, 23, 47, 32, 9, 10, 42, 52, 43, 49, 32, 49, 12, 5, 53,
54, 58, 14, 21, 7, 16, 14, 17, 12, 15, 38, 15, 33, 11, 27,
42, 56, 14, 42, 41, 34, 26, 55, 46, 47, 9, 25, 23, 19, 47,
34, 38, 53, 10, 9, 9, 43, 59, 43, 51, 12, 39, 29, 12, 13,
9, 52, 1, 13, 36, 11, 21, 58, 38, 2, 31, 18, 19, 16, 14, 6,
50, 15, 16, 21, 26, 17, 21, 43, 25, 18, 39, 57, 19, 41, 4,
31, 20, 15, 16, 34, 10, 28, 45, 21, 13, 13, 21, 43, 2, 46,
52, 12, 26, 21, 14, 52, 21, 36, 6, 22, 25, 45, 25, 12, 1, 3,
25, 31, 55, 28, 32, 9, 32, 44, 33, 6, 17, 2, 58, 32, 38, 20,
3, 12, 51, 23, 51, 1, 32, 12, 35, 13, 39, 48, 40, 7, 57, 4,
38, 51, 57, 24, 20, 30, 25, 28, 55, 32, 2, 28, 26, 27, 23,
46, 46, 9, 12, 49, 52, 17, 24, 44, 23, 54, 46, 9, 28, 32,
20, 27, 22, 57, 60, 61, 59, 56, 41, 8, 26, 20, 12, 36, 26,
23, 48, 17, 29, 19, 41, 20, 4, 29, 55, 43, 11, 24, 52, 42,
30, 37, 32, 11, 59, 6, 53, 59, 33, 52, 11, 31, 3, 52, 36,
34, 40, 23, 48, 50, 16, 32, 52, 55, 47, 56, 25, 13, 12, 33,
35, 20, 57, 61, 47, 40, 29, 31, 34, 37, 19, 2, 59, 57, 35,
44, 50, 30, 36, 27, 25, 50, 5, 38, 61, 8, 60, 23, 60, 37,
9, 6, 9, 25, 33, 23, 27, 55, 25, 7, 38, 55, 12, 44, 29, 7,
14, 34, 56, 39, 43, 14, 27, 17, 29, 51, 38, 40, 41, 42,
16, 42, 14, 19, 44, 43, 34, 60, 44, 20, 3, 45, 61, 57, 33,
46, 44, 40, 22, 8, 12, 43, 46, 20, 26, 43, 7, 61, 47, 59,
38, 48, 1, 49, 48, 35, 22, 50, 44, 43, 1, 53, 50, 52, 53,
41, 49, 6, 51, 51, 40, 5, 12, 20, 58, 57, 16, 1, 57, 58,
34, 17, 52, 25, 44, 20, 41, 50, 58, 53, 47, 58, 14, 40, 39,
10, 53, 1, 34, 54, 53, 49, 60, 31, 54, 35, 39, 23, 10, 55,
12, 56, 57, 48, 57, 15, 35, 23, 13, 10, 22, 4, 25, 17, 47,
5, 36, 51, 6, 44, 26, 27, 40, 58, 59, 15, 55, 19, 21, 7,
32, 45, 13, 41, 47, 30, 36, 1, 60, 43, 48, 61, 51]
@test randindex(R.assignments, ref_assignments)[2] == 1.0
end