PyPI page
Home page
Author:
Daoyuan Li
Summary:
Distill teacher chains-of-thought into a LoRA adapter via a strict boxed-answer format contract and a two-phase Train→Nudge schedule (silver-medal NVIDIA Nemotron reasoning recipe).
Latest version:
0.2.0
Required dependencies:
numpy
|
pandas
|
pyyaml
Optional dependencies:
accelerate
|
build
|
datasets
|
peft
|
pytest
|
pytest-cov
|
ruff
|
torch
|
transformers
|
trl
|
twine
Downloads last day:
0
Downloads last week:
1
Downloads last month:
39
7-day smoothing