use RLBase instead

.travis.yml

@@ -3,7 +3,7 @@ language: julia
 os:
   - linux
 julia:
-  - 1.2
+  - 1.3
 notifications:
   email: false
 

Dockerfile

@@ -1,4 +1,4 @@
-FROM julia:1.2
+FROM julia:1.3
 
 # install dependencies
 RUN set -eux; \

Project.toml

@@ -4,33 +4,22 @@ authors = ["Jun Tian <[email protected]>"]
 version = "0.1.3"
 
 [deps]
-Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 GR = "28b8d3ca-fb5f-59d9-8090-bfdbd6d07a71"
-LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
-POMDPModels = "355abbd5-f08e-5560-ac9e-8b5f2592a0ca"
-POMDPs = "a93abf59-7444-517b-a68a-c42f96afdd7d"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
-Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
+ReinforcementLearningBase = "e575027e-6cd6-5018-9292-cdc6200d2b44"
 Requires = "ae029012-a4dd-5104-9daa-d747884805df"
-SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
-StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
+
+[compat]
+julia = "1.3"
+GR = "0.46"
+Requires = "1.0"
 
 [extras]
 ArcadeLearningEnvironment = "b7f77d8d-088d-5e02-8ac0-89aab2acc977"
-Hanabi = "705708ad-e62c-5f47-9095-732127600058"
+POMDPModels = "355abbd5-f08e-5560-ac9e-8b5f2592a0ca"
+POMDPs = "a93abf59-7444-517b-a68a-c42f96afdd7d"
 PyCall = "438e738f-606a-5dbb-bf0a-cddfbfd45ab0"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
-ViZDoom = "13bb3beb-38fe-5ca7-9a46-050a216300b1"
 
 [targets]
-test = ["Test", "ArcadeLearningEnvironment", "ViZDoom", "PyCall", "Hanabi"]
-
-[compat]
-julia = "1"
-Distributions = "^0"
-GR = "^0"
-POMDPModels = "^0"
-POMDPs = "^0"
-Reexport = "^0"
-Requires = "^0"
-StatsBase = "^0"
+test = ["Test", "ArcadeLearningEnvironment", "PyCall", "POMDPModels", "POMDPs"]

src/ReinforcementLearningEnvironments.jl

@@ -1,22 +1,20 @@
 module ReinforcementLearningEnvironments
 
+using ReinforcementLearningBase
+
 export RLEnvs
 const RLEnvs = ReinforcementLearningEnvironments
 
-using Reexport, Requires
-
-include("abstractenv.jl")
-include("spaces/spaces.jl")
+using Requires
 
 # built-in environments
 include("environments/classic_control/classic_control.jl")
 
 # dynamic loading environments
 function __init__()
     @require ArcadeLearningEnvironment = "b7f77d8d-088d-5e02-8ac0-89aab2acc977" include("environments/atari.jl")
-    @require ViZDoom = "13bb3beb-38fe-5ca7-9a46-050a216300b1" include("environments/vizdoom.jl")
     @require PyCall = "438e738f-606a-5dbb-bf0a-cddfbfd45ab0" include("environments/gym.jl")
-    @require Hanabi = "705708ad-e62c-5f47-9095-732127600058" include("environments/hanabi.jl")
+    @require POMDPs = "a93abf59-7444-517b-a68a-c42f96afdd7d" include("environments/mdp.jl")
 end
 
 end # module

src/environments/atari.jl

@@ -6,8 +6,8 @@ mutable struct AtariEnv{IsGrayScale, TerminalOnLifeLoss, N, S<:AbstractRNG} <: A
     ale::Ptr{Nothing}
     screens::Tuple{Array{UInt8, N}, Array{UInt8, N}}  # for max-pooling
     actions::Vector{Int64}
-    action_space::DiscreteSpace{Int}
-    observation_space::MultiDiscreteSpace{UInt8, N}
+    action_space::DiscreteSpace{UnitRange{Int}}
+    observation_space::MultiDiscreteSpace{Array{UInt8, N}}
     noopmax::Int
     frame_skip::Int
     reward::Float32
@@ -69,8 +69,8 @@ function AtariEnv(
 
     observation_size = grayscale_obs ? (getScreenWidth(ale), getScreenHeight(ale)) : (3, getScreenWidth(ale), getScreenHeight(ale))  # !!! note the order
     observation_space = MultiDiscreteSpace(
-        fill(typemax(Cuchar), observation_size),
         fill(typemin(Cuchar), observation_size),
+        fill(typemax(Cuchar), observation_size),
     )
 
     actions = full_action_space ? getLegalActionSet(ale) : getMinimalActionSet(ale)
@@ -97,7 +97,7 @@ end
 update_screen!(env::AtariEnv{true}, screen) = ArcadeLearningEnvironment.getScreenGrayscale!(env.ale, vec(screen))
 update_screen!(env::AtariEnv{false}, screen) = ArcadeLearningEnvironment.getScreenRGB!(env.ale, vec(screen))
 
-function interact!(env::AtariEnv{is_gray_scale, is_terminal_on_life_loss}, a) where {is_gray_scale, is_terminal_on_life_loss}
+function (env::AtariEnv{is_gray_scale, is_terminal_on_life_loss})(a) where {is_gray_scale, is_terminal_on_life_loss}
     r = 0.0f0
 
     for i in 1:env.frame_skip
@@ -121,9 +121,9 @@ end
 is_terminal(env::AtariEnv{<:Any, true}) = game_over(env.ale) || (lives(env.ale) < env.lives)
 is_terminal(env::AtariEnv{<:Any, false}) = game_over(env.ale)
 
-observe(env::AtariEnv) = Observation(reward = env.reward, terminal = is_terminal(env), state = env.screens[1])
+RLBase.observe(env::AtariEnv) = (reward = env.reward, terminal = is_terminal(env), state = env.screens[1])
 
-function reset!(env::AtariEnv)
+function RLBase.reset!(env::AtariEnv)
     reset_game(env.ale)
     for _ = 1:rand(env.seed, 0:env.noopmax)
         act(env.ale, Int32(0))
@@ -152,7 +152,7 @@ function imshowcolor(x::Array{UInt8,1}, dims)
     updatews()
 end
 
-function render(env::AtariEnv)
+function RLBase.render(env::AtariEnv)
     x = getScreenRGB(env.ale)
     imshowcolor(x, (Int(getScreenWidth(env.ale)), Int(getScreenHeight(env.ale))))
 end