rearrange codebase (#173)

* remove objects.jl

* move directions.jl into actions.jl

* rename actions.jl to navigation.jl

* move sampling methods from envs.jl to abstract_grid_world.jl

* better organize SingleRoom envs

* reorganize GridRoom envs

* reorganize SequentialRoom envs

* reorganize Maze envs

* reorganize GoToTarget envs

* reorganize DoorKey envs

* reorganize CollectGems envs

* reorganize DynamicObstacles envs

* reorganize Sokoban envs

* reorganize Catcher

* reorganize Snake

* reorganize Transport envs

* move RLBase API into respective env modules

* minor cleanup

Author: Sid-Bhatia-0

src/GridWorlds.jl

@@ -5,12 +5,10 @@ import Random
 import REPL
 import ReinforcementLearningBase as RLBase
 
-include("directions.jl")
-include("actions.jl")
-include("objects.jl")
+include("navigation.jl")
 include("abstract_grid_world.jl")
 include("play.jl")
-include("envs/envs.jl")
 include("rlbase.jl")
+include("envs/envs.jl")
 
 end

src/abstract_grid_world.jl

@@ -59,7 +59,6 @@ function get_sub_tile_map_pretty_repr(env::AbstractGridWorldGame, window_size)
     return str
 end
 
-
 #####
 ##### Sub tile map
 #####
@@ -155,3 +154,35 @@ function get_sub_tile_map!(sub_tile_map, tile_map, position, window_size, direct
 
     return nothing
 end
+
+#####
+##### sampling tile map positions
+#####
+
+function sample_empty_position(rng, tile_map, max_tries = 1024)
+    _, height, width = size(tile_map)
+    position = CartesianIndex(rand(rng, 1:height), rand(rng, 1:width))
+
+    for i in 1:1000
+        if any(@view tile_map[:, position])
+            position = CartesianIndex(rand(rng, 1:height), rand(rng, 1:width))
+        else
+            return position
+        end
+    end
+
+    @warn "Returning non-empty position: $(position)"
+
+    return position
+end
+
+function sample_two_positions_without_replacement(rng, region)
+    position1 = rand(rng, region)
+    position2 = rand(rng, region)
+
+    while position1 == position2
+        position2 = rand(rng, region)
+    end
+
+    return position1, position2
+end

src/directions.jl

@@ -2,6 +2,16 @@ module CatcherModule
 
 import ..GridWorlds as GW
 import Random
+import ReinforcementLearningBase as RLBase
+
+#####
+##### game logic
+#####
+
+const NUM_OBJECTS = 2
+const AGENT = 1
+const GEM = 2
+const NUM_ACTIONS = 3
 
 mutable struct Catcher{R, RNG} <: GW.AbstractGridWorldGame
     tile_map::BitArray{3}
@@ -14,28 +24,6 @@ mutable struct Catcher{R, RNG} <: GW.AbstractGridWorldGame
     terminal_penalty::R
 end
 
-const NUM_OBJECTS = 2
-const AGENT = 1
-const GEM = 2
-
-CHARACTERS = ('☻', '♦', '⋅')
-
-GW.get_tile_map_height(env::Catcher) = size(env.tile_map, 2)
-GW.get_tile_map_width(env::Catcher) = size(env.tile_map, 3)
-
-function GW.get_tile_pretty_repr(env::Catcher, i::Integer, j::Integer)
-    object = findfirst(@view env.tile_map[:, i, j])
-    if isnothing(object)
-        return CHARACTERS[end]
-    else
-        return CHARACTERS[object]
-    end
-end
-
-const NUM_ACTIONS = 3
-GW.get_action_keys(env::Catcher) = ('a', 'd', 's')
-GW.get_action_names(env::Catcher) = (:MOVE_LEFT, :MOVE_RIGHT, :NO_MOVE)
-
 function Catcher(; R = Float32, height = 8, width = 8, rng = Random.GLOBAL_RNG)
     tile_map = falses(NUM_OBJECTS, height, width)
 
@@ -133,11 +121,48 @@ function GW.act!(env::Catcher, action)
     return nothing
 end
 
+#####
+##### miscellaneous
+#####
+
+CHARACTERS = ('☻', '♦', '⋅')
+
+GW.get_tile_map_height(env::Catcher) = size(env.tile_map, 2)
+GW.get_tile_map_width(env::Catcher) = size(env.tile_map, 3)
+
+function GW.get_tile_pretty_repr(env::Catcher, i::Integer, j::Integer)
+    object = findfirst(@view env.tile_map[:, i, j])
+    if isnothing(object)
+        return CHARACTERS[end]
+    else
+        return CHARACTERS[object]
+    end
+end
+
+GW.get_action_keys(env::Catcher) = ('a', 'd', 's')
+GW.get_action_names(env::Catcher) = (:MOVE_LEFT, :MOVE_RIGHT, :NO_MOVE)
+
 function Base.show(io::IO, ::MIME"text/plain", env::Catcher)
     str = GW.get_tile_map_pretty_repr(env)
     str = str * "\nreward = $(env.reward)\ndone = $(env.done)"
     print(io, str)
     return nothing
 end
 
+#####
+##### RLBase API
+#####
+
+RLBase.StateStyle(env::GW.RLBaseEnv{E}) where {E <: Catcher} = RLBase.InternalState{Any}()
+RLBase.state_space(env::GW.RLBaseEnv{E}, ::RLBase.InternalState) where {E <: Catcher} = nothing
+RLBase.state(env::GW.RLBaseEnv{E}, ::RLBase.InternalState) where {E <: Catcher} = env.env.tile_map
+
+RLBase.reset!(env::GW.RLBaseEnv{E}) where {E <: Catcher} = GW.reset!(env.env)
+
+RLBase.action_space(env::GW.RLBaseEnv{E}) where {E <: Catcher} = 1:NUM_ACTIONS
+(env::GW.RLBaseEnv{E})(action) where {E <: Catcher} = GW.act!(env.env, action)
+
+RLBase.reward(env::GW.RLBaseEnv{E}) where {E <: Catcher} = env.env.reward
+RLBase.is_terminated(env::GW.RLBaseEnv{E}) where {E <: Catcher} = env.env.done
+
 end # module

src/envs/catcher.jl

@@ -3,37 +3,23 @@ module CollectGemsDirectedModule
 import ..CollectGemsUndirectedModule as CGUM
 import ..GridWorlds as GW
 import Random
+import ReinforcementLearningBase as RLBase
 
-mutable struct CollectGemsDirected{R, RNG} <: GW.AbstractGridWorldGame
-    env::CGUM.CollectGemsUndirected{R, RNG}
-    agent_direction::Int
-end
+#####
+##### game logic
+#####
 
 const NUM_OBJECTS = CGUM.NUM_OBJECTS
 const AGENT = CGUM.AGENT
 const WALL = CGUM.WALL
 const GEM = CGUM.GEM
+const NUM_ACTIONS = 4
 
-CHARACTERS = ('☻', '█', '♦', '→', '↑', '←', '↓', '⋅')
-
-GW.get_tile_map_height(env::CollectGemsDirected) = size(env.env.tile_map, 2)
-GW.get_tile_map_width(env::CollectGemsDirected) = size(env.env.tile_map, 3)
-
-function GW.get_tile_pretty_repr(env::CollectGemsDirected, i::Integer, j::Integer)
-    object = findfirst(@view env.env.tile_map[:, i, j])
-    if isnothing(object)
-        return CHARACTERS[end]
-    elseif object == AGENT
-        return CHARACTERS[NUM_OBJECTS + 1 + env.agent_direction]
-    else
-        return CHARACTERS[object]
-    end
+mutable struct CollectGemsDirected{R, RNG} <: GW.AbstractGridWorldGame
+    env::CGUM.CollectGemsUndirected{R, RNG}
+    agent_direction::Int
 end
 
-const NUM_ACTIONS = 4
-GW.get_action_keys(env::CollectGemsDirected) = ('w', 's', 'a', 'd')
-GW.get_action_names(env::CollectGemsDirected) = (:MOVE_FORWARD, :MOVE_BACKWARD, :TURN_LEFT, :TURN_RIGHT)
-
 function CollectGemsDirected(; R = Float32, height = 8, width = 8, num_gem_init = floor(Int, sqrt(height * width)), rng = Random.GLOBAL_RNG)
     env = CGUM.CollectGemsUndirected(R = R, height = height, width = width, num_gem_init = num_gem_init, rng = rng)
     agent_direction = rand(rng, 0:GW.NUM_DIRECTIONS-1)
@@ -88,11 +74,50 @@ function GW.act!(env::CollectGemsDirected, action)
     return nothing
 end
 
+#####
+##### miscellaneous
+#####
+
+CHARACTERS = ('☻', '█', '♦', '→', '↑', '←', '↓', '⋅')
+
+GW.get_tile_map_height(env::CollectGemsDirected) = size(env.env.tile_map, 2)
+GW.get_tile_map_width(env::CollectGemsDirected) = size(env.env.tile_map, 3)
+
+function GW.get_tile_pretty_repr(env::CollectGemsDirected, i::Integer, j::Integer)
+    object = findfirst(@view env.env.tile_map[:, i, j])
+    if isnothing(object)
+        return CHARACTERS[end]
+    elseif object == AGENT
+        return CHARACTERS[NUM_OBJECTS + 1 + env.agent_direction]
+    else
+        return CHARACTERS[object]
+    end
+end
+
+GW.get_action_keys(env::CollectGemsDirected) = ('w', 's', 'a', 'd')
+GW.get_action_names(env::CollectGemsDirected) = (:MOVE_FORWARD, :MOVE_BACKWARD, :TURN_LEFT, :TURN_RIGHT)
+
 function Base.show(io::IO, ::MIME"text/plain", env::CollectGemsDirected)
     str = GW.get_tile_map_pretty_repr(env)
     str = str * "\nreward = $(env.env.reward)\ndone = $(env.env.done)"
     print(io, str)
     return nothing
 end
 
+#####
+##### RLBase API
+#####
+
+RLBase.StateStyle(env::GW.RLBaseEnv{E}) where {E <: CollectGemsDirected} = RLBase.InternalState{Any}()
+RLBase.state_space(env::GW.RLBaseEnv{E}, ::RLBase.InternalState) where {E <: CollectGemsDirected} = nothing
+RLBase.state(env::GW.RLBaseEnv{E}, ::RLBase.InternalState) where {E <: CollectGemsDirected} = (env.env.env.tile_map, env.env.agent_direction)
+
+RLBase.reset!(env::GW.RLBaseEnv{E}) where {E <: CollectGemsDirected} = GW.reset!(env.env)
+
+RLBase.action_space(env::GW.RLBaseEnv{E}) where {E <: CollectGemsDirected} = 1:NUM_ACTIONS
+(env::GW.RLBaseEnv{E})(action) where {E <: CollectGemsDirected} = GW.act!(env.env, action)
+
+RLBase.reward(env::GW.RLBaseEnv{E}) where {E <: CollectGemsDirected} = env.env.env.reward
+RLBase.is_terminated(env::GW.RLBaseEnv{E}) where {E <: CollectGemsDirected} = env.env.env.done
+
 end # module

src/envs/collect_gems_directed.jl

@@ -2,6 +2,13 @@ module CollectGemsMultiAgentUndirectedModule
 
 import ..GridWorlds as GW
 import Random
+import ReinforcementLearningBase as RLBase
+
+#####
+##### game logic
+#####
+
+const NUM_ACTIONS = 4
 
 mutable struct CollectGemsMultiAgentUndirected{R, RNG} <: GW.AbstractGridWorldGame
     tile_map::BitArray{3}
@@ -16,29 +23,6 @@ mutable struct CollectGemsMultiAgentUndirected{R, RNG} <: GW.AbstractGridWorldGa
     gem_positions::Vector{CartesianIndex{2}}
 end
 
-GW.get_tile_map_height(env::CollectGemsMultiAgentUndirected) = size(env.tile_map, 2)
-GW.get_tile_map_width(env::CollectGemsMultiAgentUndirected) = size(env.tile_map, 3)
-
-function GW.get_tile_pretty_repr(env::CollectGemsMultiAgentUndirected, i::Integer, j::Integer)
-    tile_map = env.tile_map
-    object = findfirst(@view tile_map[:, i, j])
-    num_agents = size(tile_map, 1) - 2
-
-    if isnothing(object)
-        return "⋅"
-    elseif object in 1 : num_agents
-        return "$(object)"
-    elseif object == num_agents + 1
-        return "█"
-    else
-        return "♦"
-    end
-end
-
-const NUM_ACTIONS = 4
-GW.get_action_keys(env::CollectGemsMultiAgentUndirected) = ('w', 's', 'a', 'd')
-GW.get_action_names(env::CollectGemsMultiAgentUndirected) = (:MOVE_UP, :MOVE_DOWN, :MOVE_LEFT, :MOVE_RIGHT)
-
 function CollectGemsMultiAgentUndirected(; R = Float32, height = 8, width = 8, num_gem_init = floor(Int, sqrt(height * width)), num_agents = 4, rng = Random.GLOBAL_RNG)
     tile_map = falses(num_agents + 2, height, width)
     WALL = num_agents + 1
@@ -160,11 +144,53 @@ function GW.act!(env::CollectGemsMultiAgentUndirected, action)
     return nothing
 end
 
+#####
+##### miscellaneous
+#####
+
+GW.get_tile_map_height(env::CollectGemsMultiAgentUndirected) = size(env.tile_map, 2)
+GW.get_tile_map_width(env::CollectGemsMultiAgentUndirected) = size(env.tile_map, 3)
+
+function GW.get_tile_pretty_repr(env::CollectGemsMultiAgentUndirected, i::Integer, j::Integer)
+    tile_map = env.tile_map
+    object = findfirst(@view tile_map[:, i, j])
+    num_agents = size(tile_map, 1) - 2
+
+    if isnothing(object)
+        return "⋅"
+    elseif object in 1 : num_agents
+        return "$(object)"
+    elseif object == num_agents + 1
+        return "█"
+    else
+        return "♦"
+    end
+end
+
+GW.get_action_keys(env::CollectGemsMultiAgentUndirected) = ('w', 's', 'a', 'd')
+GW.get_action_names(env::CollectGemsMultiAgentUndirected) = (:MOVE_UP, :MOVE_DOWN, :MOVE_LEFT, :MOVE_RIGHT)
+
 function Base.show(io::IO, ::MIME"text/plain", env::CollectGemsMultiAgentUndirected)
     str = GW.get_tile_map_pretty_repr(env)
     str = str * "\nreward = $(env.reward)\ndone = $(env.done)\ncurrent_agent = $(env.current_agent)"
     print(io, str)
     return nothing
 end
 
+#####
+##### RLBase API
+#####
+
+RLBase.StateStyle(env::GW.RLBaseEnv{E}) where {E <: CollectGemsMultiAgentUndirected} = RLBase.InternalState{Any}()
+RLBase.state_space(env::GW.RLBaseEnv{E}, ::RLBase.InternalState) where {E <: CollectGemsMultiAgentUndirected} = nothing
+RLBase.state(env::GW.RLBaseEnv{E}, ::RLBase.InternalState) where {E <: CollectGemsMultiAgentUndirected} = env.env.tile_map
+
+RLBase.reset!(env::GW.RLBaseEnv{E}) where {E <: CollectGemsMultiAgentUndirected} = GW.reset!(env.env)
+
+RLBase.action_space(env::GW.RLBaseEnv{E}) where {E <: CollectGemsMultiAgentUndirected} = 1:NUM_ACTIONS
+(env::GW.RLBaseEnv{E})(action) where {E <: CollectGemsMultiAgentUndirected} = GW.act!(env.env, action)
+
+RLBase.reward(env::GW.RLBaseEnv{E}) where {E <: CollectGemsMultiAgentUndirected} = env.env.reward
+RLBase.is_terminated(env::GW.RLBaseEnv{E}) where {E <: CollectGemsMultiAgentUndirected} = env.env.done
+
 end # module