Spaces:

pixxel-phantom
/

orbital-thruster-env

Sleeping

App Files Files Community

orbital-thruster-env / training

3.15 MB

Ctrl+K

Ctrl+K

1 contributor

History: 24 commits

pixxel-phantom's picture

Upload folder using huggingface_hub

2993f56 verified about 1 month ago

data
Add training pipeline: SFT+GRPO notebook, multi-reward verifier, HF job script about 2 months ago
unsloth_compiled_cache
Upload folder using huggingface_hub about 2 months ago
README.md
3.18 kB
Upload folder using huggingface_hub about 2 months ago
common.py
5.97 kB
Upload folder using huggingface_hub about 2 months ago
eval_trained_model.py
3.79 kB
Upload folder using huggingface_hub about 2 months ago
evaluate_baselines.py
3.21 kB
Upload folder using huggingface_hub about 2 months ago
generate_seed_trajectories.py
371 Bytes
Upload folder using huggingface_hub about 2 months ago
hf_job_train.py
4.64 kB
test: try unsloth 2025.10-2026.0 range about 1 month ago
local_train.py
7.36 kB
fix: vanilla GRPO use r=16 to match SFT about 1 month ago
orbital_grpo_train.py
11 kB
Upload folder using huggingface_hub about 2 months ago
plot_results.py
2.78 kB
Upload folder using huggingface_hub about 2 months ago
qwen3_grpo_train.py
5.3 kB
Upload folder using huggingface_hub about 1 month ago
qwen3_smoke_sft.py
3.13 kB
Upload folder using huggingface_hub about 1 month ago
requirements.txt
123 Bytes
Upload folder using huggingface_hub about 2 months ago
rl_utils.py
13.3 kB
Upload folder using huggingface_hub about 1 month ago
train_orbital_grpo.ipynb
7.73 kB
Upload folder using huggingface_hub about 2 months ago