All pull requests

Pull requests

• Deep-copy expressions without pickling them

  #140·aethomsen opened Oct 26, 2026Contributor·

  ·

• Speed up symmetry normalization, expansion, and wildcard matching

  #137·ValentinHirschi opened Oct 25, 2026Contributor·

  ·

• Add prepared bracketed roots and shared direct vector evaluation

  #127·ValentinHirschi opened Oct 23, 2026Contributor·

  ·

• Preserve scalar primal domains during dual vectorization

  bug
  #124·benruijl opened Oct 23, 2026Member·

  ·
  1

• Preserve complex modulus derivatives across numeric and native duals

  bug
  #123·benruijl opened Oct 23, 2026Member·

  ·
  1

• Preserve finite negative powers across evaluator backends

  bug
  #121·benruijl opened Oct 23, 2026Member·

  ·
  1

• Rebuild imported symbol identities and deserialize owned bytes

  bug
  #119·benruijl opened Oct 23, 2026Member·

  ·
  1

• Reduce high-precision PSLQ runtime with two-level reduction

  performance improvement
  #116·benruijl opened Oct 22, 2026Member·

  ·
  1

  1

• Track achieved numerical precision through expression evaluation

  #113·ValentinHirschi opened Oct 21, 2026Contributor·

  ·

• Preserve floating coefficient precision through normalization

  #112·ValentinHirschi opened Oct 21, 2026Contributor·

  ·

• Preserve original nested rational poles in exact solutions

  #111·aethomsen opened Oct 21, 2026Contributor·

  ·

• Evaluate restricted sixth-root GPLs using the HPL kernel

  #109·SecretGmG opened Oct 21, 2026Contributor·

  ·

• Add checked series expansion and resource controls

  #103·aethomsen opened Oct 21, 2026Contributor·

  ·
  1

• Expose simultaneous symbol renaming before normalization

  #101·aethomsen opened Oct 20, 2026Contributor·

  ·
  1

• Round Float products once and keep factors of ±1 exact

  #97·ValentinHirschi opened Oct 19, 2026Contributor·

  ·
  1

• Evaluate HPL constants at sixth roots with arbitrary precision

  #95·ValentinHirschi opened Oct 19, 2026Contributor·

  ·
  1

• Evaluate binary64 dilogarithms without MPFR

  performance improvement
  #89·ValentinHirschi opened Oct 17, 2026Contributor·

  ·
  1

• Add allocation-free borrowed Dual shape components

  performance improvement
  #86·ValentinHirschi opened Oct 16, 2026Contributor·

  ·
  1

• Support generic batched evaluation and shared-body Dual transformations

  #82·ValentinHirschi opened Oct 16, 2026Contributor·

  ·
  1

• Handle zero-variable polynomial exponent iteration

  #73·ValentinHirschi opened Oct 11, 2026Contributor·

  ·
  3

• Support tracked evaluation of exact fixed polygamma arguments

  #72·ValentinHirschi opened Oct 11, 2026Contributor·

  ·
  3

• Add caller-stop observations to prepared bracketed root refinement

  #71·ValentinHirschi opened Oct 11, 2026Contributor·

  ·
  2

• Honor registered derivative bodies when inlining is disabled

  #65·ValentinHirschi opened Oct 10, 2026Contributor·

  ·
  2

• Use clone-owned callback environments in JIT evaluators

  #62·ValentinHirschi opened Oct 10, 2026Contributor·

  ·
  2

• Expose shared direct evaluation for expression vectors

  #59·ValentinHirschi opened Oct 10, 2026Contributor·

  ·
  2