Closed Sid-Bhatia-0 closed 4 years ago
Could you also add methods like RLBase.get_state
before merging this PR?
And adding a simple test case like run(RandomPolicy(env), env)
would be much appreciated.
Sure.
For the RLBase.get_state
method, should I add a trait that specifies the format of the state. For example, two of the common formats for gridworlds are:
And adding a simple test case like run(RandomPolicy(env), env) would be much appreciated.
Will do.
Implementing the get_state(env)
would be enough at present I think.
Merging #54 into master will increase coverage by
33.61%
. The diff coverage is88.88%
.
@@ Coverage Diff @@
## master #54 +/- ##
===========================================
+ Coverage 30.88% 64.50% +33.61%
===========================================
Files 14 14
Lines 395 400 +5
===========================================
+ Hits 122 258 +136
+ Misses 273 142 -131
Impacted Files | Coverage Δ | |
---|---|---|
src/GridWorlds.jl | 100.00% <ø> (ø) |
|
src/render_in_terminal.jl | 0.00% <0.00%> (ø) |
|
src/abstract_grid_world.jl | 85.00% <100.00%> (+5.00%) |
:arrow_up: |
src/envs/collectgems.jl | 100.00% <100.00%> (+100.00%) |
:arrow_up: |
src/grid_world_base.jl | 69.76% <0.00%> (+2.32%) |
:arrow_up: |
src/objects.jl | 22.58% <0.00%> (+14.38%) |
:arrow_up: |
src/envs/doorkey.jl | 97.22% <0.00%> (+97.22%) |
:arrow_up: |
src/envs/dynamicobstacles.jl | 100.00% <0.00%> (+100.00%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update f46e78c...c302269. Read the comment docs.
RLBase
dependency as per #48AbstractGridWorld
a subtype ofAbstractEnv
RLBase
methods forCollectGems
environment.show
method so that it does not get overriden by the default one given inRLBase