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

Jump to bottom

Update Turing inference #48

Merged
merged 2 commits into from
Jul 4, 2018
Merged

Update Turing inference #48

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0acb313
Update Turing inference
xukai92 Jun 25, 2018
df81743
set stable Turing version in REQUIRE
xukai92 Jul 3, 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
2 changes: 1 addition & 1 deletion REQUIRE
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@ DiffEqBase 1.7.0
Mamba
Stan
Distributions
Turing
Turing 0.4.3
MacroTools
Optim
RecursiveArrayTools
Expand Down
16 changes: 13 additions & 3 deletions src/turing_inference.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,31 +2,41 @@ function turing_inference(prob::DEProblem,alg,t,data,priors = nothing;
num_samples=1000, epsilon = 0.02, tau = 4, kwargs...)

bif(vi, sampler, x=data) = begin
_lp = 0.0
N = length(priors)
_theta = Vector(N)

for i in 1:length(priors)
_theta[i] = Turing.assume(sampler,
_theta[i], __lp = Turing.assume(sampler,
priors[i],
Turing.VarName(vi, [:bif, Symbol("theta$i")], ""),
vi)
_lp += __lp
end

theta = convert(Array{typeof(first(_theta))},_theta)
σ = Turing.assume(sampler,

σ, __lp = Turing.assume(sampler,
InverseGamma(2, 3),
Turing.VarName(vi, [:bif, :σ], ""),
vi)
_lp += __lp

p_tmp = problem_new_parameters(prob, theta); sol_tmp = solve(p_tmp,alg;saveat=t,kwargs...)

for i = 1:length(t)
res = sol_tmp.u[i]
# x[:,i] ~ MvNormal(res, σ*ones(2))
Turing.observe(
__lp = Turing.observe(
sampler,
MvNormal(res, σ*ones(length(prob.u0))), # Distribution
x[:,i], # Data point
vi
)
_lp += __lp
end

vi.logp = _lp
vi
end

Expand Down