boostedprob 0.1.0

Utilities to compute boosted probabilities and identify dominant tokens.

boostedprob

Utilities to compute "dominant tokens" and derive boosted probabilities from model log-probabilities.

Install

• Locally (development editable install):

python -m pip install -e .

• From GitHub (example):

python -m pip install "git+https://github.com/yourusername/boostedprob.git"

Example

import torch
import boostedprob

# log_probs: shape [batch, seq_len, vocab]
# target: shape [batch, seq_len]
# (fill with your model outputs)

log_probs = torch.log_softmax(torch.randn(2, 4, 1000), dim=-1)
target = torch.randint(0, 1000, (2, 4))

scores = boostedprob.calculate_boostedprob(log_probs, target)
print(scores.shape)  # -> (2, 4)

Build & publish

python -m pip install --upgrade build twine
python -m build
python -m twine upload dist/*

Or test first on TestPyPI (recommended).