Skip to content

Releases: lanl-ansi/MathOptSymbolicAD.jl

v0.2.1

17 Sep 03:03
62b478f
Compare
Choose a tag to compare

MathOptSymbolicAD v0.2.1

Diff since v0.2.0

Merged pull requests:

v0.2.0

14 Aug 02:45
1196acb
Compare
Choose a tag to compare

MathOptSymbolicAD v0.2.0

Diff since v0.1.3

Merged pull requests:

  • Update to latest JuMP syntax (#29) (@odow)
  • Relicense under LANL's standard BSD-3 license (#33) (@odow)
  • Tidy package imports (#34) (@odow)
  • Support comparison operators (#35) (@odow)
  • Add support for Symbolics@6 (#36) (@odow)
  • Prep for v0.2.0 (#37) (@odow)

Closed issues:

  • README Maintenance (#28)
  • Use as an AD backend in DifferentiationInterface.jl? (#30)
  • Support Nonlinear.NODE_LOGIC and Nonlinear.NODE_COMPARISON (#31)
  • Update license (#32)

v0.1.3

17 Jan 05:19
0ed013b
Compare
Choose a tag to compare

MathOptSymbolicAD v0.1.3

Diff since v0.1.2

Merged pull requests:

  • Update Symbolics version (#27) (@odow)

Closed issues:

  • Symbolics v5 (#26)

v0.1.2

02 Feb 22:28
058d1b4
Compare
Choose a tag to compare

MathOptSymbolicAD v0.1.2

Diff since v0.1.1

Closed issues:

  • Type error in evaluator subexpression generation (#24)

Merged pull requests:

  • Fix support for constant subexpressions (#25) (@odow)

v0.1.1

13 Jan 03:04
b5ad0e0
Compare
Choose a tag to compare

MathOptSymbolicAD v0.1.1

Diff since v0.1.0

Merged pull requests:

v0.1.0

21 Aug 01:10
3cdae42
Compare
Choose a tag to compare

MathOptSymbolicAD v0.1.0

Closed issues:

  • Speed-up ideas (#4)
  • User-defined functions (#10)
  • Fix operation to function (#11)
  • Implement Hessian sparsity idea discussed with coffrin (#13)

Merged pull requests:

  • Add .JuliaFormatter.toml (#1) (@odow)
  • Refactor into AbstractBackends (#2) (@odow)
  • Add direct JuMP integration bypassing expensive initialize (#3) (@odow)
  • Add timers for objective and gradient (#5) (@odow)
  • Implement a small performance improvement (#6) (@odow)
  • Special case handling of integer exponents (#7) (@odow)
  • Cache an ordered vector of variables instead of using dictionary lookups (#8) (@odow)
  • Add tools to validate the derivatives produced by JuMP (#9) (@odow)
  • Skip Hessian computation if zero dual (#12) (@odow)
  • Implement compacted hessian_lagrangian (#14) (@odow)
  • Add documentation (#16) (@odow)
  • Rebase on MOI.Nonlinear (#17) (@odow)
  • Add license headers (#18) (@odow)
  • Add support for user-defined functions (#19) (@odow)
  • Rename to MathOptSymbolicAD.jl (#20) (@odow)