keyflux 0.2.0

keyflux 0.2.0

• allotaxonometer — the Dodds (2020) diamond allotaxonograph: a rotated-square rank-rank histogram with iso-divergence contours and a wordshift/balance panel. The two-panel allotaxonograph is unchanged.
• RankedList.from_scores — rank by any numeric score (keyness, log ratio, ...), so keyword rankings and "frequent words vs keywords" are first-class.
• Presidential-speech gallery gains diamond figures + a frequency-vs-keyness diamond; README notes comparing many rankings as a research direction.

🤖 Generated with Claude Code

keyflux 0.1.2

keyflux 0.1.2

Fixes version drift: it is now read from package metadata, so an installed keyflux reports the correct version (0.1.1 reported 0.1.0). No API changes.

🤖 Generated with Claude Code

keyflux 0.1.1

keyflux 0.1.1

• PNG logo so the README renders on PyPI.
• Presidential-speech allotaxonograph gallery (examples/presidential_speeches.py + notebook + saved figures) comparing US presidential speech vocabulary across eras.

🤖 Generated with Claude Code

keyflux 0.1.0

First release of keyflux — corpus keyness, rank-turbulence divergence, and allotaxonographs, in pure Python.

• Keyness: positive/negative keywords, lockwords, and a reproducibility record; measures (log-likelihood, log ratio, Simple Maths, %DIFF, chi-square) validated against Brezina Ch. 3.
• Rank-turbulence divergence: rank-sensitive comparison with an alpha→0 log limit; reproduces the Dodds et al. / jkbren reference value at alpha=1.0.
• Allotaxonograph: two-panel matplotlib figure (diamond rank-rank histogram + contribution balance) — no JavaScript runtime.

Docs: https://keyflux.readthedocs.io

🤖 Generated with Claude Code