Osrs 5.0 pr - #644
Open
valtterivalo wants to merge 46 commits into
Open
Conversation
replaces the stale mid-july PR 613 payload with the post-recut tree: item-table inventory obs, fused entity encoders in env .cu files, macos openmp and arm simd build fixes, and no python package.
persist denormalized sweep_observations.csv, replay via resume_from, and skip the random phase so a warm surrogate does not re-draw Sobol.
valtterivalo
force-pushed
the
osrs-5.0-pr
branch
from
August 16, 2026 20:30
9eb8a29 to
1b477f3
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
going back to basics here, we're down from 3k+ to 905 and 530 in colo and inferno respectively. we're no longer running sim in the obs and there's nothing in place you could call scaffolding, you could argue current setup is even a bit unfair to RL on.
on a 4090 colo gets 401k SPS, reaching wave 9.7 on avg and a win rate of ~5%.
inferno is sitting at 424k SPS, currently only reaching wave 47. gonna focus on that for the next couple days.
both envs now use puffer's Encoder interface and inventory, equipped gear, and NPCs go through a small shared per record encoder followed by max pooling. i add pooled features to flat projection before entering the net.