google-research / FirstOrderLp.jl

Experimental first-order solvers for linear and quadratic programming.

google-research/FirstOrderLp.jl Issues

Warm starts
Updated 4 months ago1
Diagonal scaling is slow for large instances and may raise an out of memory error
Closed 7 months ago
benchmarking fails with Julia 1.8
Closed a year ago1
CI tests don't test macos nor Julia 1.7
Closed a year ago
Test fails
Closed a year ago2
Small typo in code
Closed a year ago3
Rename csv datafiles (and processing in scripts/analyze_csv_data) to reflect new dataset names.
Updated a year ago2
Default tolerances
Updated 2 years ago7
How to warm start the FirstOrderLp.jl solver?
Closed 2 years ago4
How to access the duals of LP
Closed 2 years ago2
Error when loading unpreprocessed lp benchmark files
Closed 2 years ago4
More complete example of using process_json_to_csv.jl
Closed 3 years ago
Update "MIPLIB 2017" to "MIP Relaxations"
Closed 3 years ago2
lint errors for analyze_csv_data.py
Updated 3 years ago
Duplicated code in mirror_prox.jl and primal_dual_hybrid_gradient.jl
Updated 3 years ago
Malitsky-pock dies with assertion error
Closed 3 years ago1
Norms change on each iteration
Closed 3 years ago
iteration counter includes rejected steps
Closed 3 years ago
Function docstrings don't match julia style
Updated 3 years ago
Docstrings display poorly as markdown
Updated 3 years ago
Investigate replacing ad-hoc logging with the Logging package
Updated 3 years ago
PDHG performs poorly on a Mittelman LP
Updated 3 years ago