jump-dev/Hypatia.jl

Issues

• TagBot trigger issue

  #628 opened 3 years ago by JuliaTagBot
  16

• Hypatia fails to precompile on 1.10

  #835 opened 4 months ago by ranjanan
  2

• Doesn't compile with julia 1.10

  #834 opened 5 months ago by araujoms
  1

• Dual of log-det cone

  #827 opened 10 months ago by blegat
  2

• Can't constrain affine function to be Hermitian PSD

  #823 opened a year ago by araujoms
  7

• `using Hypatia` fails

  #816 opened a year ago by nsajko
  1

• Some questions

  #801 opened a year ago by baggepinnen
  7

• Faster arbitrary precision support for SDPs

  #818 opened a year ago by DanielBrosch
  5

• Add support for MOI.HermitianPositiveSemidefiniteConeTriangle

  #821 opened a year ago by araujoms
  1

• Complex MOI

  #540 opened a year ago by blegat
  4

• are the options of Hypatia.Optimizer documented anywhere?

  #812 opened 2 years ago by kalmarek
  3

• experiment again with SOS inverse Hessian

  #808 opened 2 years ago by lkapelevich
  0

• Numerical issues with Positivstellensatz and interpolant-based polynomial optimization

  #798 opened 2 years ago by projekter
  4

• update termination message for `NearOptimal` status

  #799 opened 2 years ago by lkapelevich
  0

• permute_affine not working with immutable array

  #793 opened 2 years ago by matbesancon
  1

• support string parameters for JuMP 0.21

  #486 opened 3 years ago by lkapelevich
  1

• dispatch update_rhs functions on stepper type

  #773 opened 3 years ago by lkapelevich
  0

• run some JuMP example instances with non-Float64 reals

  #598 opened 3 years ago by chriscoey
  1

• check that SuiteSparse caches are using the maximum number of threads

  #361 opened 3 years ago by lkapelevich
  1

• improve termination criteria for outer 6x6 iterative refinement

  #375 opened 3 years ago by chriscoey
  1

• fix argument types of vec to mat functions

  #392 opened 3 years ago by chriscoey
  0

• allow specifying an initial point for a WSOS dual cone

  #476 opened 4 years ago by lkapelevich
  0

• for WSOS cones, special case P_r for which L_r = 1 (just use log of 1x1 Lambda instead of logdet)

  #533 opened 3 years ago by chriscoey
  0

• try checking hessian nbhd after heuristic nbhd is satisfied for all cones

  #534 opened 3 years ago by chriscoey
  0

• experiment with higher order correctors (again)

  #567 opened 3 years ago by lkapelevich
  0

• optimize LMI cone barrier evals

  #477 opened 3 years ago by chriscoey
  1

• get a more efficient TOA for lmi cone

  #654 opened 3 years ago by lkapelevich
  0

• MOI interface: `set_silent` not respected

  #640 opened 3 years ago by ericphanson
  1

• delete the large raw csv data files in examples

  #511 opened 4 years ago by chriscoey
  0

• describe all options to all examples

  #489 opened 4 years ago by chriscoey
  0

• implement more hessian-related oracles for more cones

  #478 opened 4 years ago by chriscoey
  0

• rename m and n in spectral norm cone definition

  #380 opened 4 years ago by lkapelevich
  1

• investigate central points for epiperexp and hypogeomean cones

  #373 opened 4 years ago by lkapelevich
  0

• more examples for more cones

  #378 opened 4 years ago by lkapelevich
  1

• try 3rd order correction without scaling

  #411 opened 4 years ago by chriscoey
  4

• make it possible to use larger L but not larger U

  #512 opened 4 years ago by lkapelevich
  1

• try using Q factor of P for interpolation based WSOS cone

  #500 opened 4 years ago by lkapelevich
  1

• allow each cone to use a different neighborhood function

  #481 opened 4 years ago by chriscoey
  1

• modify barrier for hypopersumlog to be SC modified barrier from relative entropy

  #469 opened 4 years ago by chriscoey
  1

• reduce dimension of some examples by using cholesky

  #403 opened 4 years ago by chriscoey
  1

• add closed form initial point for geomean cone in special case of equal powers

  #468 opened 4 years ago by chriscoey
  1

• map MOI exponential cone to epiperexp instead of hypoperlog

  #470 opened 4 years ago by chriscoey
  1

• relax type restrictions on parameter matrices in LMI cone

  #433 opened 4 years ago by chriscoey
  1

• improve sparse hessian access for EpiNormInf

  #382 opened 4 years ago by chriscoey
  1

• clean up wsos norm barrier code

  #397 opened 4 years ago by chriscoey
  1

• enable threading of various for loops in barrier functions

  #395 opened 4 years ago by chriscoey
  0

• get MU.interpolate to return single vector of Ps that don't need concatenation

  #390 opened 4 years ago by lkapelevich
  0

• do optional inner iterative refinement in system solver caches

  #374 opened 5 years ago by chriscoey
  0

• improve initial points

  #367 opened 5 years ago by lkapelevich
  0

• simplify spectral norm barrier

  #368 opened 5 years ago by lkapelevich
  1