Spaces:

InosLihka
/

rhythm_env

Sleeping

App Files Files Community

1.88 MB

Ctrl+K

Ctrl+K

3 contributors

History: 57 commits

InosLihka's picture

Clarify documentation: anomaly signal explainer, GRPO scope notes

361aed7 3 months ago

docs
Clarify documentation: anomaly signal explainer, GRPO scope notes 3 months ago
plots
Add SFT v3 + GRPO refine results to README + results.md 4 months ago
scripts
Add SFT v3 + GRPO refine results to README + results.md 4 months ago
server
Refactor grader to use openenv.core.rubrics.WeightedSum + Rubric subclasses 4 months ago
tests
Refactor grader to use openenv.core.rubrics.WeightedSum + Rubric subclasses 4 months ago
training
Clarify documentation: anomaly signal explainer, GRPO scope notes 3 months ago
ui
refactor: rewrite blog around product vision; fix UI for Gradio 6 4 months ago
.dockerignore

92 Bytes
Initial commit: RhythmEnv daily planning RL environment 4 months ago
.env.example

441 Bytes
Algorithm Distillation: grader v2 with belief_accuracy + SFT pipeline 4 months ago
.gitattributes

218 Bytes
Post-deadline: full eval results + bigger plots via Git LFS 4 months ago
.gitignore

211 Bytes
Clarify documentation: anomaly signal explainer, GRPO scope notes 3 months ago
BLOG.md

9.48 kB
Move blog to root as BLOG.md (per Meta mentor guidance) 4 months ago
Dockerfile

1.49 kB
Fix HF Space README rendering + Dockerfile encoding 3 months ago
README.md

23.2 kB
Clarify documentation: anomaly signal explainer, GRPO scope notes 3 months ago
__init__.py

724 Bytes
env: enrich observation with history, anomalies, and discovery bonus 4 months ago
client.py

5.04 kB
client: surface ALL observation fields (was dropping deltas, anomalies, last_action, step_history) 4 months ago
eval_baselines_v2.json

284 Bytes
Algorithm Distillation: grader v2 with belief_accuracy + SFT pipeline 4 months ago
inference.py

13.4 kB
iter4: fix the 'constant belief = free reward' bug + 6 other deep issues 4 months ago
models.py

4.17 kB
Algorithm Distillation: grader v2 with belief_accuracy + SFT pipeline 4 months ago
openenv.yaml

93 Bytes
Initial commit: RhythmEnv daily planning RL environment 4 months ago
pyproject.toml

909 Bytes
Rebuild as Life Simulator: 5 meters, 3 hidden profiles, GRPO training pipeline 4 months ago
uv.lock

576 kB
Initial commit: RhythmEnv daily planning RL environment 4 months ago