The #1 adoption lever from the competitive analysis: a sklearn-compatible package (python/lawsynth-sklearn) so the scikit-learn ecosystem adopts LawSynth with near-zero switching cost (why gplearn/PySINDy win mindshare). - LawSynthDynamics: fit(X, t=)/predict/simulate/score/equations over the real Study discovery; full get_params/set_params/__sklearn_tags__/NotFittedError. - LawSynthRegressor (RegressorMixin), LawSynthTransformer (feature library + deterministic decorrelation prune), parsimony='auto' (lambda = Cov/Var). - Inherits sklearn mixins when present; faithful standalone fallback otherwise. Demo recovers an oscillator R2=0.999996; 50 tests pass (7 sklearn-integration skip cleanly when sklearn absent). Deterministic, offline. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
35 lines
954 B
TOML
35 lines
954 B
TOML
[build-system]
|
|
requires = ["setuptools>=64"]
|
|
build-backend = "setuptools.build_meta"
|
|
|
|
[project]
|
|
name = "lawsynth-sklearn"
|
|
version = "0.1.0"
|
|
description = "scikit-learn-compatible estimators for LawSynth deterministic governing-equation discovery"
|
|
readme = "README.md"
|
|
requires-python = ">=3.11"
|
|
license = { text = "Apache-2.0" }
|
|
authors = [{ name = "LawSynth" }]
|
|
keywords = [
|
|
"scikit-learn",
|
|
"symbolic-regression",
|
|
"equation-discovery",
|
|
"dynamical-systems",
|
|
"sindy",
|
|
"time-series",
|
|
]
|
|
dependencies = ["lawsynth"]
|
|
|
|
[project.optional-dependencies]
|
|
# The estimators inherit sklearn mixins when sklearn is present, and degrade to
|
|
# a standalone implementation of the same contract when it is absent. numpy is
|
|
# likewise optional: plain Python lists are accepted and returned.
|
|
sklearn = ["scikit-learn>=1.3", "numpy>=1.23"]
|
|
test = ["pytest>=7"]
|
|
|
|
[tool.setuptools.packages.find]
|
|
where = ["src"]
|
|
|
|
[tool.setuptools.package-dir]
|
|
"" = "src"
|