High-Performance Symbolic Regression in Python and Julia
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.8...v0.16.9
typing_extensions
for compatibility with Python 3.7 by @MilesCranmer in https://github.com/MilesCranmer/PySR/pull/497
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.7...v0.16.8
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.6...v0.16.7
--heap-size-hint
on spawned Julia processes by @MilesCranmer in https://github.com/MilesCranmer/PySR/pull/493
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.5...v0.16.6
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.4...v0.16.5
sr.py
by @MilesCranmer in https://github.com/MilesCranmer/PySR/pull/428
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.3...v0.16.4
Full Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.2...v0.16.3
Backend changes: https://github.com/MilesCranmer/SymbolicRegression.jl/compare/v0.22.2...v0.22.3
Frontend changes: https://github.com/MilesCranmer/PySR/compare/v0.16.1...v0.16.2
Backend Changelog: Diff since v0.22.1 PySR Changelog: https://github.com/MilesCranmer/PySR/compare/v0.16.0...v0.16.1
batching=true
. It results in improved searches on large datasets, as the "winning expression" is not biased towards an expression that landed on a lucky batch.fast_cycle
feature in https://github.com/MilesCranmer/SymbolicRegression.jl/pull/243. Use of this parameter will have no effect.
@eval
-ing new operators.x1 + x2
. All internal search code uses Node()
explicitly to build expressions, so did not rely on method invalidation at any point.Backend Changelog: https://github.com/MilesCranmer/SymbolicRegression.jl/compare/v0.21.5...v0.22.1
PySR Changelog: https://github.com/MilesCranmer/PySR/compare/v0.15.4...v0.16.0