Skip to content
This repository has been archived by the owner on Jun 14, 2020. It is now read-only.

readme first sketch #2

Merged
merged 19 commits into from
Mar 11, 2018
Merged

readme first sketch #2

Changes from 6 commits
Commits
Show all changes
19 commits
Select commit Hold shift + click to select a range
2ddf536
readme first sketch
joaquimg Dec 7, 2017
57e8709
Update README.md
joaquimg Dec 7, 2017
6635790
Update README.md
joaquimg Dec 15, 2017
bd6ca68
Merge remote-tracking branch 'origin/master' into jg/readme
joaquimg Mar 10, 2018
210d347
update readme
joaquimg Mar 10, 2018
1fb52ea
improve name
joaquimg Mar 10, 2018
c407096
fixes
joaquimg Mar 10, 2018
1d84e4f
CachingOptimizer
joaquimg Mar 10, 2018
645f9e2
Remove ambiguous sentence
joaquimg Mar 10, 2018
754fdc7
add suggestions
joaquimg Mar 10, 2018
2177186
replace instance by optimizer
joaquimg Mar 10, 2018
3a5a42d
add suggestions
joaquimg Mar 10, 2018
ad51801
try html self ref
joaquimg Mar 10, 2018
17f5ac2
separate mods
joaquimg Mar 10, 2018
03b1ccf
add some
joaquimg Mar 10, 2018
e82075a
missing link
joaquimg Mar 10, 2018
b0765d8
missing brace
joaquimg Mar 10, 2018
c948f35
supports
joaquimg Mar 10, 2018
30f2fc2
fix english
joaquimg Mar 10, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 27 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,2 +1,28 @@
# LinQuadOptInterface.jl
Intermediate wrapper for MOI of GLPK, CPLEX, Gurobi and Xpress

LinQuadOptInterface (LQOI) is designed to make it easier for low-level wrapper designed to bridge low-level Integer Linear and Quadratic solvers to implement the MOI interface. It provides access to many MathOptInterface functionalities mainly related to problem modifications:

1. add constraints/variables 1-by-1 and in batches
2. remove constraints/variables
3. change constraint coefficients
4. change constraint type
5. change constraint rhs
6. change variable bounds and type
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe add a sentence making it explicit that using LQOI is entirely optional and comes with a trade-off of introducing an extra abstraction layer between the solver and MOI.


## LinQuadOptInterface Instance

In LinQuadOptInterface.jl the MOI `AbstractOptimizer` is specified to `LinQuadOptimizer`. In this implementation all the above metioned modifications are caried out by the low-level solver own functionalities and hence it differs from MathOptInterface.Utilities' `AbstractModel`'s generated by the `@model` macro. The latter will keep all problem data in the julia level and typically push to a low-level solver the complete problem at once.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

metioned -> mentioned

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You mention @model but the discussion is closen ti CachingOptimizer

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you have any suggestion about what to say there or should I just remove this part?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You added it when SDOI was keeping a copy but now no optimizer keeps a copy, CachingOptimizer should be used for that. You can say "In this implementation all the above mentioned modifications are carried out by the low-level solver own functionalities and hence the optimizer can be used without a CachingOptimizer"


Here the data in the julia level is minimal, basically only references to constraints and variables.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Instead of this sentence maybe something along the lines of "In contrast, since LinQuadOptimizer incrementally pushes data to the low-level solver, it stores a small subset of the problem data at the Julia level; typically only references to constraints and variables."


## Current uses

This package is currently used to implement `MathOptInterfaceXpress.jl`, `MathOptInterfaceCPLEX.jl`, `MathOptInterfaceGurobi.jl` and `MathOptInterfaceGLPK.jl`. The last one being only a integer linear solver.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be nice to ahh hyperlinks


All these solver have in common a low-level API which supports most of these modifications. Hence, data storage is simplified and duplications are avoided.

## Other possible uses

This package is only recommended if a solver low-level API which supports most of these modifications.

If a solver low-level API does not support most of these modifications, then following the example of `MathOptInterfaceSCS.jl` and `MathOptInterfaceECOS.jl` might be a better idea.