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

Use pomdps action #1

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Use pomdps action #1

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
deeca02
Use POMDPs.action()
bkraske Feb 7, 2024
9ae4a15
Add depth list
bkraske Feb 7, 2024
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
14 changes: 9 additions & 5 deletions src/generation.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@ struct PolicyGraph{N,E} <: Policy
edges::E
node1::Int64
beliefs::Vector{SparseVector{Float64, Int64}}
node_depth::Vector{Int}
end

##TO DO: Implement function for checking if belief is terminal
Expand All @@ -19,7 +20,7 @@ end
"""
function gen_polgraph end

function gen_polgraph(m::POMDP, s_pomdp::EvalTabularPOMDP, updater::Updater, pol::Policy, b0::SparseVector, depth::Int, action_list, edge_list, b_list, d, j_old, a_old, oo, oa, replace)
function gen_polgraph(m::POMDP, s_pomdp::EvalTabularPOMDP, updater::Updater, pol::Policy, b0::SparseVector, depth::Int, action_list, edge_list, b_list, d, j_old, a_old, oo, oa, replace, depth_list)
if d < depth
d+=1
obs = s_pomdp.O[a_old]
Expand All @@ -40,11 +41,12 @@ function gen_polgraph(m::POMDP, s_pomdp::EvalTabularPOMDP, updater::Updater, pol
a = action_from_vec(m, pol, bp)
push!(action_list, oa[a])
push!(b_list,bp)
push!(depth_list,d)
# @show Vector.(b_list)
j = copy(length(action_list))
push!(edge_list, (j_old, oo[o]) => j)

gen_polgraph(m,s_pomdp,updater,pol,bp,depth,action_list,edge_list,b_list,d,j,a,oo,oa,replace)
gen_polgraph(m,s_pomdp,updater,pol,bp,depth,action_list,edge_list,b_list,d,j,a,oo,oa,replace,depth_list)
end
end
end
Expand All @@ -55,6 +57,7 @@ function gen_polgraph(m::POMDP{S,A}, s_pomdp::EvalTabularPOMDP, updater::Updater
edge_list = Dict{Tuple{Int64,obstype(pol.pomdp)},Int64}()
action_list = A[]
b_list = SparseVector{Float64, Int64}[]
depth_list = Int[]
d = 1
a=if !isempty(replace)
replace[1]
Expand All @@ -63,16 +66,17 @@ function gen_polgraph(m::POMDP{S,A}, s_pomdp::EvalTabularPOMDP, updater::Updater
end::A
push!(action_list, a)
push!(b_list,sparse(b0.b))
push!(depth_list,d)
j = copy(length(action_list))

oo = ordered_observations(m)
oa = ordered_actions(m)

gen_polgraph(m, s_pomdp, updater, pol, sparse(b0.b), depth, action_list, edge_list, b_list, d, j, actionindex(m,a), oo, oa, !isempty(replace))
gen_polgraph(m, s_pomdp, updater, pol, sparse(b0.b), depth, action_list, edge_list, b_list, d, j, actionindex(m,a), oo, oa, !isempty(replace),depth_list)
if !store_beliefs
return PolicyGraph(action_list, edge_list, 1, SparseVector{Float64, Int64}[])
return PolicyGraph(action_list, edge_list, 1, SparseVector{Float64, Int64}[],depth_list)
else
return PolicyGraph(action_list, edge_list, 1, b_list)
return PolicyGraph(action_list, edge_list, 1, b_list, depth_list)
end
end

Expand Down
29 changes: 28 additions & 1 deletion src/recursive_evaluation.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,7 +138,11 @@ function belief_value_recursive(pomdp::POMDP{S,A}, updater::Updater, pol::Policy
d = 1
r_dim = length(rewardfunction(pomdp,ordered_states(pomdp)[1],ordered_actions(pomdp)[1]))
s_pomdp = EvalTabularPOMDP(pomdp;rew_f=rewardfunction,r_len=r_dim)
r = belief_value_recursive(pomdp, s_pomdp, updater, pol, sparse(b.b), depth, d, replace)
if isa(pol,AlphaVectorPolicy)
r = belief_value_recursive(pomdp, s_pomdp, updater, pol, sparse(b.b), depth, d, replace)
else
r = belief_value_recursive_na(pomdp, s_pomdp, updater, pol, sparse(b.b), depth, d, replace)
end
return r
end

Expand All @@ -163,4 +167,27 @@ function belief_value_recursive(pomdp::POMDP{S,A}, s_pomdp::EvalTabularPOMDP, up
end
end
return value
end

function belief_value_recursive_na(pomdp::POMDP{S,A}, s_pomdp::EvalTabularPOMDP, updater::Updater, pol::Policy, b::SparseVector{Float64, Int64}, depth::Int, d::Int, replace::Vector{A}) where {S,A}
a=if d==1 && !isempty(replace)
replace[1]
else
action(pol, b)
end
value = belief_reward(s_pomdp,b,a)
if d<depth
d+=1
obs = s_pomdp.O[a]
pred = s_pomdp.T[a]*b
for o in axes(obs,2)
bp = corrector(s_pomdp, pred, a, o)
po = sum(bp)
if po > 0. && !isterminalbelief(s_pomdp,bp)
bp.nzval ./= po
value += discount(s_pomdp)*po*belief_value_recursive(pomdp, s_pomdp, updater, pol, bp, depth, d, replace)
end
end
end
return value
end