feat!: Better reproducibility management of dynamic problems with SeededEnvironment#73
Open
BatyLeo wants to merge 6 commits into
Open
feat!: Better reproducibility management of dynamic problems with SeededEnvironment#73BatyLeo wants to merge 6 commits into
SeededEnvironment#73BatyLeo wants to merge 6 commits into
Conversation
…valuation and various dynamic environments
BatyLeo
changed the title
SeededEnvironmentSeededEnvironment
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Introduce a
SeededEnvironmentwrapper that is put around all concrete environments bygenerate_environments.Concrete environments now shouldn't manage their own seed or RNG, the AbstractEnv interface is updated accordingly:
reset!andstep!now require anrngargument, the defaultget_seedmethod was removed, and the default RNG is switched fromMersenneTwistertoXoshiro.SeededEnvironment{E,R}(src/Utils/environment/seeded_environment.jl) wraps anyAbstractEnvironmentwith aseed+rng.reset_to_initial!(env)re-seed to the stored seed and resets.rollout!(policy, env, rng)runs an episode from the current state, never resets.generate_environmentsnow returnsVector{SeededEnvironment}, seeding each env from an internal seed stream, same forgenerate_environment.build_environmentinstead of the oldgenerate_environment.evaluate_policy!(policy, env::SeededEnvironment; seed=nothing)resets to the initial seed (or aseedoverride) then rolls out.