Buckets:
352 kB
55 files
Updated 3 months ago
Ctrl+K
| Name | Size | Uploaded | Xet hash |
|---|---|---|---|
| claim1_3 | 2 items | ||
| claim2 | 1 items | ||
| claim4_5 | 1 items | ||
| README.md | 788 Bytes xet | f4e35e52 | |
| __init__.py | 66 Bytes xet | fb7b22b5 | |
| datasets.py | 3.96 kB xet | 958e4a46 | |
| models.py | 5.03 kB xet | 0770f916 | |
| run_claim1_3.py | 5.87 kB xet | a0e23600 | |
| run_claim2.py | 3.16 kB xet | dd9fc360 | |
| run_claim4_5.py | 6.7 kB xet | d91084c4 | |
| smoke_test.py | 763 Bytes xet | 6af29db6 | |
| train.py | 5.28 kB xet | c46a6c87 | |
| verdicts.json | 1.94 kB xet | 42c61dc9 |
Neural Thickets tinygrad reproduction
Pure tinygrad port of the official RandOpt 1D suite (simple_1D_signals_expts) for ICML 2026 Agent Repro (paper arXiv 2603.12228 / OpenReview 92oF5bU4cU).
Setup
cd tinygrad_repro
uv venv .venv --python 3.12
uv pip install --python .venv/bin/python tinygrad numpy trackio huggingface_hub
Runs
.venv/bin/python run_claim1_3.py # Claims 1 & 3 (needle/thicket/plateau)
.venv/bin/python run_claim2.py # Claim 2 width/depth proxy
.venv/bin/python run_claim4_5.py # Claims 4 & 5 (RandOpt vs ES/SGD + population)
Results land in outputs/claim*/results.json.
Notes
- HF Jobs remain 402 (insufficient credits) — no Qwen2.5 / PPO / GRPO GPU runs.
- Metrics use AR rollout MSE (matches official eval).
- Total size
- 352 kB
- Files
- 55
- Last updated
- Jul 15
- Pre-warmed CDN
- US EU US EU