v0.1.3

MathOptSetDistances v0.1.3

Diff since v0.1.2

Merged pull requests:

• updated CRC to 0.10 (#49) (@matbesancon)
• diagonal to identity matrix in reals proj (#51) (@matbesancon)
• bump FillArrays (#57) (@matbesancon)

v0.1.2

MathOptSetDistances v0.1.2

Diff since v0.1.1

Closed issues:

• Error in tests when changing bissection lines (#37)

Merged pull requests:

• small changes on bissect (#38) (@matbesancon)
• added frule for primal exp cone (#39) (@matbesancon)
• Some tests for Nonpositive cone (#41) (@be-apt)
• fix deprecations (#42) (@matbesancon)
• Exp cone heuristics (#43) (@tjdiamandis)
• norm ball and simplex projection (#46) (@matbesancon)
• Static vec (#47) (@matbesancon)

v0.1.1

MathOptSetDistances v0.1.1

Diff since v0.1.0

Closed issues:

• Test projections against AD (#8)
• Should all projection gradient be square matrices? (#9)
• Reduce duplications (#10)
• Register package (#13)
• Leverage special arrays when relevant (#17)
• PSD gradient wrong? (#18)
• Factor of sqrt(2) in unvec_symm and vec_symm? (#21)
• Switch Random to test-only dependency (#33)

Merged pull requests:

• improve workflow, formatting (#15) (@matbesancon)
• CompatHelper: add new compat entry for "FillArrays" at version "0.10" (#19) (@github-actions[bot])
• More tests for SOC and PSD (#20) (@matbesancon)
• consistency projection distance (#22) (@matbesancon)
• distance to PSD cone (#23) (@matbesancon)
• update docstring for matrix vectorization (#24) (@tjdiamandis)
• Implementing ChainRulesCore.rrule for projections (#25) (@matbesancon)
• PSD frule (#28) (@matbesancon)
• Exponential Cone (#29) (@tjdiamandis)
• add frule for orthants (#31) (@matbesancon)
• simplify deps (#34) (@matbesancon)
• CompatHelper: bump compat for "FillArrays" to "0.11" (#36) (@github-actions[bot])

v0.1.0

MathOptSetDistances v0.1.0

Closed issues:

• Add projection api (#4)

Merged pull requests:

• introduce epigraph violation (#1) (@matbesancon)
• normed epigraph distance (#3) (@matbesancon)
• add projection methods (#5) (@Aks1996)
• fix argument order (#7) (@matbesancon)
• Remove SCS specific scaling for PSD constraints (#11) (@Aks1996)