mirror of
https://gitlab.rlp.net/mobitar/ReCo.jl.git
synced 2024-12-21 00:51:21 +00:00
state and action ind
This commit is contained in:
parent
3036c5e65a
commit
cfb38c6a19
3 changed files with 89 additions and 34 deletions
|
@ -5,6 +5,7 @@ using Flux: InvDecay
|
||||||
using Intervals
|
using Intervals
|
||||||
using StaticArrays: SVector
|
using StaticArrays: SVector
|
||||||
using Random: Random
|
using Random: Random
|
||||||
|
using ProgressMeter: @showprogress
|
||||||
|
|
||||||
using ..ReCo
|
using ..ReCo
|
||||||
|
|
||||||
|
@ -28,17 +29,19 @@ end
|
||||||
|
|
||||||
mutable struct EnvParams
|
mutable struct EnvParams
|
||||||
action_space::Vector{Tuple{Float64,Float64}}
|
action_space::Vector{Tuple{Float64,Float64}}
|
||||||
|
action_space_ind::Vector{Int64}
|
||||||
distance_state_space::Vector{DistanceState}
|
distance_state_space::Vector{DistanceState}
|
||||||
direction_state_space::Vector{DirectionState}
|
direction_state_space::Vector{DirectionState}
|
||||||
state_space::Vector{Tuple{Union{DistanceState,Nothing},Union{DirectionState,Nothing}}}
|
state_space::Vector{Tuple{Union{DistanceState,Nothing},Union{DirectionState,Nothing}}}
|
||||||
|
state_space_ind::Vector{Int64}
|
||||||
reward::Float64
|
reward::Float64
|
||||||
|
|
||||||
function EnvParams(
|
function EnvParams(
|
||||||
min_distance::Float64,
|
min_distance::Float64,
|
||||||
max_distance::Float64;
|
max_distance::Float64;
|
||||||
n_v_actions::Int64=2,
|
n_v_actions::Int64=5,
|
||||||
n_ω_actions::Int64=3,
|
n_ω_actions::Int64=5,
|
||||||
max_v::Float64=20.0,
|
max_v::Float64=80.0,
|
||||||
max_ω::Float64=π / 1.5,
|
max_ω::Float64=π / 1.5,
|
||||||
n_distance_states::Int64=3,
|
n_distance_states::Int64=3,
|
||||||
n_direction_states::Int64=4,
|
n_direction_states::Int64=4,
|
||||||
|
@ -65,6 +68,8 @@ mutable struct EnvParams
|
||||||
end
|
end
|
||||||
end
|
end
|
||||||
|
|
||||||
|
action_space_ind = collect(1:n_actions)
|
||||||
|
|
||||||
distance_range =
|
distance_range =
|
||||||
min_distance:((max_distance - min_distance) / n_distance_states):max_distance
|
min_distance:((max_distance - min_distance) / n_distance_states):max_distance
|
||||||
|
|
||||||
|
@ -109,8 +114,18 @@ mutable struct EnvParams
|
||||||
end
|
end
|
||||||
state_space[ind] = (nothing, nothing)
|
state_space[ind] = (nothing, nothing)
|
||||||
|
|
||||||
|
state_space_ind = collect(1:n_states)
|
||||||
|
|
||||||
|
initial_reward = 0.0
|
||||||
|
|
||||||
return new(
|
return new(
|
||||||
action_space, distance_state_space, direction_state_space, state_space, 0.0
|
action_space,
|
||||||
|
action_space_ind,
|
||||||
|
distance_state_space,
|
||||||
|
direction_state_space,
|
||||||
|
state_space,
|
||||||
|
state_space_ind,
|
||||||
|
initial_reward,
|
||||||
)
|
)
|
||||||
end
|
end
|
||||||
end
|
end
|
||||||
|
@ -118,14 +133,17 @@ end
|
||||||
mutable struct Env <: AbstractEnv
|
mutable struct Env <: AbstractEnv
|
||||||
params::EnvParams
|
params::EnvParams
|
||||||
particle::ReCo.Particle
|
particle::ReCo.Particle
|
||||||
state::Tuple{Union{DistanceState,Nothing},Union{DirectionState,Nothing}}
|
state_ind::Int64
|
||||||
|
|
||||||
function Env(params::EnvParams, particle::ReCo.Particle)
|
function Env(params::EnvParams, particle::ReCo.Particle)
|
||||||
return new(params, particle, (nothing, nothing))
|
# initial_state = (nothing, nothing)
|
||||||
|
initial_state_ind = length(params.state_space_ind)
|
||||||
|
|
||||||
|
return new(params, particle, initial_state_ind)
|
||||||
end
|
end
|
||||||
end
|
end
|
||||||
|
|
||||||
function gen_policy(n_states, n_actions)
|
function gen_policy(n_states::Int64, n_actions::Int64)
|
||||||
return QBasedPolicy(;
|
return QBasedPolicy(;
|
||||||
learner=MonteCarloLearner(;
|
learner=MonteCarloLearner(;
|
||||||
approximator=TabularQApproximator(;
|
approximator=TabularQApproximator(;
|
||||||
|
@ -145,9 +163,13 @@ struct Params{H<:AbstractHook}
|
||||||
n_steps_before_actions_update::Int64
|
n_steps_before_actions_update::Int64
|
||||||
min_distance²::Vector{Float64}
|
min_distance²::Vector{Float64}
|
||||||
r⃗₁₂_to_min_distance_particle::Vector{SVector{2,Float64}}
|
r⃗₁₂_to_min_distance_particle::Vector{SVector{2,Float64}}
|
||||||
|
goal_shape_ratio::Float64
|
||||||
|
|
||||||
function Params{H}(
|
function Params{H}(
|
||||||
n_particles::Int64, env_params::EnvParams, n_steps_before_actions_update::Int64
|
n_particles::Int64,
|
||||||
|
env_params::EnvParams,
|
||||||
|
n_steps_before_actions_update::Int64,
|
||||||
|
goal_shape_ratio::Float64,
|
||||||
) where {H<:AbstractHook}
|
) where {H<:AbstractHook}
|
||||||
policies = [
|
policies = [
|
||||||
gen_policy(length(env_params.state_space), length(env_params.action_space)) for
|
gen_policy(length(env_params.state_space), length(env_params.action_space)) for
|
||||||
|
@ -165,27 +187,32 @@ struct Params{H<:AbstractHook}
|
||||||
n_steps_before_actions_update,
|
n_steps_before_actions_update,
|
||||||
zeros(n_particles),
|
zeros(n_particles),
|
||||||
fill(SVector(0.0, 0.0), n_particles),
|
fill(SVector(0.0, 0.0), n_particles),
|
||||||
|
goal_shape_ratio,
|
||||||
)
|
)
|
||||||
end
|
end
|
||||||
end
|
end
|
||||||
|
|
||||||
RLBase.state_space(env::Env) = env.state_space
|
RLBase.state_space(env::Env) = env.params.state_space_ind
|
||||||
|
|
||||||
RLBase.state(env::Env) = env.state
|
RLBase.state(env::Env) = env.state_ind
|
||||||
|
|
||||||
RLBase.action_space(env::Env) = env.params.action_space
|
RLBase.action_space(env::Env) = env.params.action_space_ind
|
||||||
|
|
||||||
RLBase.reward(env::Env) = env.params.reward
|
RLBase.reward(env::Env) = env.params.reward
|
||||||
|
|
||||||
|
RLBase.is_terminated(::Env) = false
|
||||||
|
|
||||||
function pre_integration_hook!(rl_params::Params, n_particles::Int64)
|
function pre_integration_hook!(rl_params::Params, n_particles::Int64)
|
||||||
for i in 1:n_particles
|
for i in 1:n_particles
|
||||||
env = rl_params.envs[i]
|
env = rl_params.envs[i]
|
||||||
agent = rl_params.agents[i]
|
agent = rl_params.agents[i]
|
||||||
action = agent(env)
|
|
||||||
|
action_ind = agent(env)
|
||||||
|
action = rl_params.env_params.action_space[action_ind]
|
||||||
rl_params.actions[i] = action
|
rl_params.actions[i] = action
|
||||||
|
|
||||||
agent(PRE_ACT_STAGE, env, action)
|
agent(PRE_ACT_STAGE, env, action_ind)
|
||||||
rl_params.hooks[i](PRE_ACT_STAGE, agent, env, action)
|
rl_params.hooks[i](PRE_ACT_STAGE, agent, env, action_ind)
|
||||||
end
|
end
|
||||||
|
|
||||||
return nothing
|
return nothing
|
||||||
|
@ -209,17 +236,29 @@ function state_hook(
|
||||||
return nothing
|
return nothing
|
||||||
end
|
end
|
||||||
|
|
||||||
function integration_hook(particle::ReCo.Particle, rl_params::Params, δt::Float64)
|
function integration_hook(
|
||||||
|
particle::ReCo.Particle, rl_params::Params, δt::Float64, si::Float64, co::Float64
|
||||||
|
)
|
||||||
action = rl_params.actions[particle.id]
|
action = rl_params.actions[particle.id]
|
||||||
|
|
||||||
particle.tmp_c += action[1] * δt
|
vδt = action[1] * δt
|
||||||
|
particle.tmp_c += SVector(vδt * co, vδt * si)
|
||||||
particle.φ += action[2] * δt
|
particle.φ += action[2] * δt
|
||||||
|
|
||||||
return nothing
|
return nothing
|
||||||
end
|
end
|
||||||
|
|
||||||
|
function get_state_ind(
|
||||||
|
state::T, states::Vector{T}
|
||||||
|
) where {T<:Tuple{Union{DistanceState,Nothing},Union{DirectionState,Nothing}}}
|
||||||
|
return findfirst(x -> x == state, states)
|
||||||
|
end
|
||||||
|
|
||||||
function post_integration_hook(
|
function post_integration_hook(
|
||||||
rl_params::Params, n_particles::Int64, particles::Vector{ReCo.Particle}
|
rl_params::Params,
|
||||||
|
n_particles::Int64,
|
||||||
|
particles::Vector{ReCo.Particle},
|
||||||
|
half_box_len::Float64,
|
||||||
)
|
)
|
||||||
env_direction_state = rl_params.env_params.direction_state_space[1]
|
env_direction_state = rl_params.env_params.direction_state_space[1]
|
||||||
|
|
||||||
|
@ -239,7 +278,8 @@ function post_integration_hook(
|
||||||
end
|
end
|
||||||
|
|
||||||
if isnothing(env_distance_state)
|
if isnothing(env_distance_state)
|
||||||
env.state = (nothing, nothing)
|
# (nothing, nothing)
|
||||||
|
env.state_ind = length(env.params.state_space)
|
||||||
else
|
else
|
||||||
r⃗₁₂ = rl_params.r⃗₁₂_to_min_distance_particle[i]
|
r⃗₁₂ = rl_params.r⃗₁₂_to_min_distance_particle[i]
|
||||||
si, co = sincos(particles[i].φ)
|
si, co = sincos(particles[i].φ)
|
||||||
|
@ -259,9 +299,17 @@ function post_integration_hook(
|
||||||
end
|
end
|
||||||
end
|
end
|
||||||
|
|
||||||
env.state = (env_distance_state, env_direction_state)
|
state = (env_distance_state, env_direction_state)
|
||||||
|
env.state_ind = get_state_ind(state, env.params.state_space)
|
||||||
end
|
end
|
||||||
|
|
||||||
|
env.params.reward =
|
||||||
|
1 -
|
||||||
|
(
|
||||||
|
ReCo.gyration_tensor_eigvals_ratio(particles, half_box_len) -
|
||||||
|
rl_params.goal_shape_ratio
|
||||||
|
)^2
|
||||||
|
|
||||||
agent(POST_ACT_STAGE, env)
|
agent(POST_ACT_STAGE, env)
|
||||||
rl_params.hooks[i](POST_ACT_STAGE, agent, env)
|
rl_params.hooks[i](POST_ACT_STAGE, agent, env)
|
||||||
end
|
end
|
||||||
|
@ -270,11 +318,13 @@ function post_integration_hook(
|
||||||
end
|
end
|
||||||
|
|
||||||
function run(;
|
function run(;
|
||||||
|
goal_shape_ratio::Float64,
|
||||||
n_episodes::Int64=100,
|
n_episodes::Int64=100,
|
||||||
episode_duration::Float64=5.0,
|
episode_duration::Float64=100.0,
|
||||||
update_actions_at::Float64=0.1,
|
update_actions_at::Float64=0.1,
|
||||||
n_particles::Int64=10,
|
n_particles::Int64=100,
|
||||||
)
|
)
|
||||||
|
@assert 0.0 <= goal_shape_ratio <= 1.0
|
||||||
@assert n_episodes > 0
|
@assert n_episodes > 0
|
||||||
@assert episode_duration > 0
|
@assert episode_duration > 0
|
||||||
@assert update_actions_at in 0.01:0.01:episode_duration
|
@assert update_actions_at in 0.01:0.01:episode_duration
|
||||||
|
@ -282,7 +332,7 @@ function run(;
|
||||||
|
|
||||||
Random.seed!(42)
|
Random.seed!(42)
|
||||||
|
|
||||||
sim_consts = ReCo.gen_sim_consts(n_particles, 0.0; skin_to_interaction_r_ratio=3.5)
|
sim_consts = ReCo.gen_sim_consts(n_particles, 0.0; skin_to_interaction_r_ratio=4.0)
|
||||||
n_particles = sim_consts.n_particles
|
n_particles = sim_consts.n_particles
|
||||||
|
|
||||||
env_params = EnvParams(sim_consts.particle_radius, sim_consts.skin_r)
|
env_params = EnvParams(sim_consts.particle_radius, sim_consts.skin_r)
|
||||||
|
@ -290,7 +340,7 @@ function run(;
|
||||||
n_steps_before_actions_update = round(Int64, update_actions_at / sim_consts.δt)
|
n_steps_before_actions_update = round(Int64, update_actions_at / sim_consts.δt)
|
||||||
|
|
||||||
rl_params = Params{TotalRewardPerEpisode}(
|
rl_params = Params{TotalRewardPerEpisode}(
|
||||||
n_particles, env_params, n_steps_before_actions_update
|
n_particles, env_params, n_steps_before_actions_update, goal_shape_ratio
|
||||||
)
|
)
|
||||||
|
|
||||||
for i in 1:n_particles
|
for i in 1:n_particles
|
||||||
|
@ -301,7 +351,7 @@ function run(;
|
||||||
agent(PRE_EXPERIMENT_STAGE, env)
|
agent(PRE_EXPERIMENT_STAGE, env)
|
||||||
end
|
end
|
||||||
|
|
||||||
for episode in 1:n_episodes
|
@showprogress 0.6 for episode in 1:n_episodes
|
||||||
dir, particles = ReCo.init_sim_with_sim_consts(sim_consts; parent_dir="RL")
|
dir, particles = ReCo.init_sim_with_sim_consts(sim_consts; parent_dir="RL")
|
||||||
|
|
||||||
for i in 1:n_particles
|
for i in 1:n_particles
|
||||||
|
@ -314,7 +364,7 @@ function run(;
|
||||||
|
|
||||||
for i in 1:n_particles
|
for i in 1:n_particles
|
||||||
rl_params.envs[i].particle = particles[i]
|
rl_params.envs[i].particle = particles[i]
|
||||||
rl_params.envs[i].state = (nothing, nothing)
|
rl_params.envs[i].state_ind = length(rl_params.env_params.state_space)
|
||||||
end
|
end
|
||||||
|
|
||||||
rl_params.env_params.reward = 0.0
|
rl_params.env_params.reward = 0.0
|
||||||
|
@ -339,7 +389,7 @@ function run(;
|
||||||
rl_params.hooks[i](POST_EXPERIMENT_STAGE, agent, env)
|
rl_params.hooks[i](POST_EXPERIMENT_STAGE, agent, env)
|
||||||
end
|
end
|
||||||
|
|
||||||
return rl_params.hooks
|
return rl_params
|
||||||
end
|
end
|
||||||
|
|
||||||
end # module
|
end # module
|
|
@ -70,16 +70,16 @@ function gen_sim_consts(
|
||||||
max_approach_after_one_integration_step = buffer * (2 * v₀ * δt) / interaction_r
|
max_approach_after_one_integration_step = buffer * (2 * v₀ * δt) / interaction_r
|
||||||
@assert skin_to_interaction_r_ratio >= 1 + max_approach_after_one_integration_step
|
@assert skin_to_interaction_r_ratio >= 1 + max_approach_after_one_integration_step
|
||||||
|
|
||||||
skin_r = skin_to_interaction_r_ratio * interaction_r
|
|
||||||
n_steps_before_verlet_list_update = round(
|
n_steps_before_verlet_list_update = round(
|
||||||
Int64,
|
Int64,
|
||||||
(skin_to_interaction_r_ratio - 1) / max_approach_after_one_integration_step,
|
(skin_to_interaction_r_ratio - 1) / max_approach_after_one_integration_step,
|
||||||
)
|
)
|
||||||
else
|
else
|
||||||
skin_r = 1.5 * interaction_r
|
|
||||||
n_steps_before_verlet_list_update = 100
|
n_steps_before_verlet_list_update = 100
|
||||||
end
|
end
|
||||||
|
|
||||||
|
skin_r = skin_to_interaction_r_ratio * interaction_r
|
||||||
|
|
||||||
grid_n = round(Int64, ceil(sqrt(n_particles)))
|
grid_n = round(Int64, ceil(sqrt(n_particles)))
|
||||||
|
|
||||||
n_particles = grid_n^2
|
n_particles = grid_n^2
|
||||||
|
|
|
@ -41,8 +41,11 @@ function update_verlet_lists!(args, cl)
|
||||||
end
|
end
|
||||||
|
|
||||||
function euler!(
|
function euler!(
|
||||||
args, state_hook::F, integration_hook::F, rl_params::Union{RL.Params,Nothing}
|
args,
|
||||||
) where {F<:Function}
|
state_hook::Function,
|
||||||
|
integration_hook::Function,
|
||||||
|
rl_params::Union{RL.Params,Nothing},
|
||||||
|
)
|
||||||
for id1 in 1:(args.n_particles - 1)
|
for id1 in 1:(args.n_particles - 1)
|
||||||
p1 = args.particles[id1]
|
p1 = args.particles[id1]
|
||||||
p1_c = p1.c
|
p1_c = p1.c
|
||||||
|
@ -74,11 +77,11 @@ function euler!(
|
||||||
args.v₀δt * si + args.c₃ * rand_normal01(),
|
args.v₀δt * si + args.c₃ * rand_normal01(),
|
||||||
)
|
)
|
||||||
|
|
||||||
p.φ += args.c₄ * rand_normal01()
|
|
||||||
|
|
||||||
restrict_coordinates!(p, args.half_box_len)
|
restrict_coordinates!(p, args.half_box_len)
|
||||||
|
|
||||||
integration_hook(p, rl_params, args.δt)
|
integration_hook(p, rl_params, args.δt, si, co)
|
||||||
|
|
||||||
|
p.φ += args.c₄ * rand_normal01()
|
||||||
|
|
||||||
p.c = p.tmp_c
|
p.c = p.tmp_c
|
||||||
end
|
end
|
||||||
|
@ -153,7 +156,9 @@ function simulate(
|
||||||
euler!(args, state_hook, integration_hook, rl_params)
|
euler!(args, state_hook, integration_hook, rl_params)
|
||||||
|
|
||||||
if run_hooks
|
if run_hooks
|
||||||
post_integration_hook(rl_params, args.n_particles, args.particles)
|
post_integration_hook(
|
||||||
|
rl_params, args.n_particles, args.particles, args.half_box_len
|
||||||
|
)
|
||||||
state_hook = empty_hook
|
state_hook = empty_hook
|
||||||
end
|
end
|
||||||
end
|
end
|
||||||
|
|
Loading…
Reference in a new issue