-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
split the tests in different files, added some tests
- Loading branch information
konkam
committed
Nov 13, 2018
1 parent
36c2b0e
commit a555fe3
Showing
4 changed files
with
215 additions
and
77 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,124 @@ | ||
| @testset "test marginal likelihood computation" begin | ||
|
|
||
| Random.seed!(0) | ||
|
|
||
| Δt = 0.1 | ||
| δ = 3. | ||
| γ = 2.5 | ||
| σ = 4. | ||
| Nobs = 2 | ||
| Nsteps = 4 | ||
| λ = 1. | ||
| Nparts = 10 | ||
| α = δ/2 | ||
| β = γ/σ^2 | ||
|
|
||
| time_grid = [k*Δt for k in 0:(Nsteps-1)] | ||
| times = [k*Δt for k in 0:(Nsteps-1)] | ||
| X = FeynmanKacParticleFilters.generate_CIR_trajectory(time_grid, 3, δ*1.2, γ/1.2, σ*0.7) | ||
| Y = map(λ -> rand(Poisson(λ), Nobs), X); | ||
| data = zip(times, Y) |> Dict | ||
| Mt = FeynmanKacParticleFilters.create_transition_kernels_CIR(data, δ, γ, σ) | ||
| Gt = FeynmanKacParticleFilters.create_potential_functions_CIR(data) | ||
| logGt = FeynmanKacParticleFilters.create_log_potential_functions_CIR(data) | ||
| RS(W) = rand(Categorical(W), length(W)) | ||
|
|
||
| Random.seed!(0) | ||
| @test Mt[0.1](3) ≈ 8.418659447049441 atol=10.0^(-7) | ||
| @test Mt[0.1](3.1) ≈ 2.1900629888259893 atol=10.0^(-7) | ||
| @test Mt[0.2](3.1) ≈ 2.6844105017153863 atol=10.0^(-7) | ||
| @test Mt[time_grid[3]](3.1) ≈ 1.3897782586244247 atol=10.0^(-7) | ||
|
|
||
| Random.seed!(0) | ||
| pf = FeynmanKacParticleFilters.generic_particle_filtering1D(Mt, Gt, Nparts, RS) | ||
|
|
||
| Random.seed!(0) | ||
| pf_dict = FeynmanKacParticleFilters.generic_particle_filtering(Mt, Gt, Nparts, RS) | ||
|
|
||
| W = pf["W"] | ||
| w = pf["w"] | ||
|
|
||
|
|
||
| marginal_lik_factors = FeynmanKacParticleFilters.marginal_likelihood_factors(pf) | ||
| # println(marginal_lik_factors) | ||
| res = [ 0.005063925135653128, 0.0013145849369714938, 0.014640244207811792, 0.0017270473953601316] | ||
| for k in 1:Nsteps | ||
| @test marginal_lik_factors[k] ≈ res[k] atol=10.0^(-7) | ||
| end | ||
| @test FeynmanKacParticleFilters.marginal_likelihood(pf, FeynmanKacParticleFilters.marginal_likelihood_factors) ≈ prod(res) atol=10.0^(-7) | ||
|
|
||
| marginal_lik_factors = FeynmanKacParticleFilters.marginal_likelihood_factors(pf_dict) | ||
| # println(marginal_lik_factors) | ||
| res = [ 0.005063925135653128, 0.0013145849369714938, 0.014640244207811792, 0.0017270473953601316] | ||
| for k in 1:Nsteps | ||
| @test marginal_lik_factors[k] ≈ res[k] atol=10.0^(-7) | ||
| end | ||
| @test FeynmanKacParticleFilters.marginal_likelihood(pf_dict, FeynmanKacParticleFilters.marginal_likelihood_factors) ≈ prod(res) atol=10.0^(-7) | ||
|
|
||
| Random.seed!(0) | ||
| pf_logweights = FeynmanKacParticleFilters.generic_particle_filtering_logweights1D(Mt, logGt, Nparts, RS) | ||
| Random.seed!(0) | ||
| pf_logweights_dict = FeynmanKacParticleFilters.generic_particle_filtering_logweights(Mt, logGt, Nparts, RS) | ||
|
|
||
|
|
||
| marginal_loglik_factors = FeynmanKacParticleFilters.marginal_loglikelihood_factors(pf_logweights) | ||
| # println(marginal_loglik_factors) | ||
| res = [ -5.285613377888339, -6.634234300460378, -4.223981089726635, -6.361342036441921] | ||
| for k in 1:Nsteps | ||
| @test marginal_loglik_factors[k] ≈ res[k] atol=5*10.0^(-5) | ||
| end | ||
| marginal_loglik_factors = FeynmanKacParticleFilters.marginal_loglikelihood_factors(pf_logweights_dict) | ||
| res = [ -5.285613377888339, -6.634234300460378, -4.223981089726635, -6.361342036441921] | ||
| for k in 1:Nsteps | ||
| @test marginal_loglik_factors[k] ≈ res[k] atol=5*10.0^(-5) | ||
| end | ||
|
|
||
| @test FeynmanKacParticleFilters.marginal_loglikelihood(pf_logweights, FeynmanKacParticleFilters.marginal_loglikelihood_factors) ≈ sum(res) atol=5*10.0^(-5) | ||
|
|
||
| @test FeynmanKacParticleFilters.marginal_loglikelihood(pf_logweights_dict, FeynmanKacParticleFilters.marginal_loglikelihood_factors) ≈ sum(res) atol=5*10.0^(-5) | ||
| # | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling1D(Mt, Gt, Nparts, RS) | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive_dict = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling(Mt, Gt, Nparts, RS) | ||
|
|
||
| marginal_lik_factors = FeynmanKacParticleFilters.marginal_likelihood_factors_adaptive_resampling(pf_adaptive) | ||
| # println(marginal_lik_factors) | ||
| res = [0.005063925135653128, 0.0013145849369714936, 0.014640244207811792, 0.0020015945094952942] | ||
| for k in 1:Nsteps | ||
| @test marginal_lik_factors[k] ≈ res[k] atol=10.0^(-7) | ||
| end | ||
| @test FeynmanKacParticleFilters.marginal_likelihood(pf_adaptive, FeynmanKacParticleFilters.marginal_likelihood_factors) ≈ prod(res) atol=10.0^(-7) | ||
|
|
||
| marginal_lik_factors = FeynmanKacParticleFilters.marginal_likelihood_factors_adaptive_resampling(pf_adaptive_dict) | ||
| # println(marginal_lik_factors) | ||
| res = [0.005063925135653128, 0.0013145849369714936, 0.014640244207811792, 0.0020015945094952942] | ||
| for k in 1:Nsteps | ||
| @test marginal_lik_factors[k] ≈ res[k] atol=10.0^(-7) | ||
| end | ||
| @test FeynmanKacParticleFilters.marginal_likelihood(pf_adaptive_dict, FeynmanKacParticleFilters.marginal_likelihood_factors) ≈ prod(res) atol=10.0^(-7) | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive_logweights = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling_logweights1D(Mt, logGt, Nparts, RS) | ||
| marginal_loglik_factors = FeynmanKacParticleFilters.marginal_loglikelihood_factors_adaptive_resampling(pf_adaptive_logweights) | ||
| # println(marginal_loglik_factors) | ||
| res = [ -5.285613377888339, -6.634234300460378, -4.223981089726635, -6.213811161313297] | ||
| for k in 1:Nsteps | ||
| @test marginal_loglik_factors[k] ≈ res[k] atol=5*10.0^(-5) | ||
| end | ||
| @test FeynmanKacParticleFilters.marginal_loglikelihood(pf_adaptive_logweights, FeynmanKacParticleFilters.marginal_loglikelihood_factors_adaptive_resampling) ≈ sum(res) atol=5*10.0^(-5) | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive_logweights_dict = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling_logweights(Mt, logGt, Nparts, RS) | ||
| marginal_loglik_factors = FeynmanKacParticleFilters.marginal_loglikelihood_factors_adaptive_resampling(pf_adaptive_logweights_dict) | ||
| # println(marginal_loglik_factors) | ||
| res = [ -5.285613377888339, -6.634234300460378, -4.223981089726635, -6.213811161313297] | ||
| for k in 1:Nsteps | ||
| @test marginal_loglik_factors[k] ≈ res[k] atol=5*10.0^(-5) | ||
| end | ||
| @test FeynmanKacParticleFilters.marginal_loglikelihood(pf_adaptive_logweights_dict, FeynmanKacParticleFilters.marginal_loglikelihood_factors_adaptive_resampling) ≈ sum(res) atol=5*10.0^(-5) | ||
|
|
||
|
|
||
| end |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,85 @@ | ||
| @testset "test sampling from the particle filter algorithm for CIR process" begin | ||
|
|
||
| Random.seed!(0) | ||
|
|
||
| Δt = 0.1 | ||
| δ = 3. | ||
| γ = 2.5 | ||
| σ = 4. | ||
| Nobs = 2 | ||
| Nsteps = 4 | ||
| λ = 1. | ||
| Nparts = 10 | ||
| α = δ/2 | ||
| β = γ/σ^2 | ||
|
|
||
| time_grid = [k*Δt for k in 0:(Nsteps-1)] | ||
| times = [k*Δt for k in 0:(Nsteps-1)] | ||
| X = FeynmanKacParticleFilters.generate_CIR_trajectory(time_grid, 3, δ*1.2, γ/1.2, σ*0.7) | ||
| Y = map(λ -> rand(Poisson(λ), Nobs), X); | ||
| data = zip(times, Y) |> Dict | ||
| Mt = FeynmanKacParticleFilters.create_transition_kernels_CIR(data, δ, γ, σ) | ||
| Gt = FeynmanKacParticleFilters.create_potential_functions_CIR(data) | ||
| logGt = FeynmanKacParticleFilters.create_log_potential_functions_CIR(data) | ||
| RS(W) = rand(Categorical(W), length(W)) | ||
|
|
||
| Random.seed!(0) | ||
| @test Mt[0.1](3) ≈ 8.418659447049441 atol=10.0^(-7) | ||
| @test Mt[0.1](3.1) ≈ 2.1900629888259893 atol=10.0^(-7) | ||
| @test Mt[0.2](3.1) ≈ 2.6844105017153863 atol=10.0^(-7) | ||
| @test Mt[time_grid[3]](3.1) ≈ 1.3897782586244247 atol=10.0^(-7) | ||
|
|
||
| Random.seed!(0) | ||
| pf = FeynmanKacParticleFilters.generic_particle_filtering1D(Mt, Gt, Nparts, RS) | ||
|
|
||
| Random.seed!(0) | ||
| pf_dict = FeynmanKacParticleFilters.generic_particle_filtering(Mt, Gt, Nparts, RS) | ||
|
|
||
| W = pf["W"] | ||
| w = pf["w"] | ||
|
|
||
| @test typeof(pf) == Dict{String,Array{Float64,2}} | ||
| for i in 1:size(W,2) | ||
| @test pf["W"][1,i] ≈ [1.58397e-6, 0.000109003, 0.247537, 0.332939][i] atol = 10^(-6) | ||
| end | ||
| for i in 1:size(w,2) | ||
| @test pf["w"][1,i] ≈ [8.021083116860762e-8, 1.4329312817343978e-6, 0.03624009164218452, 0.005750007892716746][i] atol = 10^(-10) | ||
| end | ||
|
|
||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions1D(pf, 10, 2)) == 10 | ||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions(pf_dict, 10, 2)) == 10 | ||
|
|
||
| Random.seed!(0) | ||
| pf_logweights = FeynmanKacParticleFilters.generic_particle_filtering_logweights1D(Mt, logGt, Nparts, RS) | ||
| Random.seed!(0) | ||
| pf_logweights_dict = FeynmanKacParticleFilters.generic_particle_filtering_logweights(Mt, logGt, Nparts, RS) | ||
|
|
||
| @test typeof(pf_logweights) == Dict{String,Array{Float64,2}} | ||
| @test typeof(pf_logweights_dict) == Dict{String,Any} | ||
|
|
||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions_logweights1D(pf_logweights, 10, 2)) == 10 | ||
|
|
||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions_logweights(pf_logweights_dict, 10, 2)) == 10 | ||
| # | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling1D(Mt, Gt, Nparts, RS) | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive_dict = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling(Mt, Gt, Nparts, RS) | ||
|
|
||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions1D(pf_adaptive, 10, 2)) == 10 | ||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions(pf_adaptive_dict, 10, 2)) == 10 | ||
|
|
||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive_logweights = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling_logweights1D(Mt, logGt, Nparts, RS) | ||
|
|
||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions_logweights1D(pf_adaptive_logweights, 10, 2)) == 10 | ||
|
|
||
| Random.seed!(0) | ||
| pf_adaptive_logweights_dict = FeynmanKacParticleFilters.generic_particle_filtering_adaptive_resampling_logweights(Mt, logGt, Nparts, RS) | ||
| @test length(FeynmanKacParticleFilters.sample_from_filtering_distributions_logweights(pf_adaptive_logweights_dict, 10, 2)) == 10 | ||
|
|
||
|
|
||
| end |