Flux constraints JuMP #14
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…jl into ar/flux_constraints
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## main #14 +/- ##
===========================================
- Coverage 97.54% 60.89% -36.65%
===========================================
Files 7 10 +3
Lines 285 468 +183
===========================================
+ Hits 278 285 +7
- Misses 7 183 +176 ☔ View full report in Codecov by Sentry. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
(WIP: WORK IN PROGRESS)
Description:
This pull request introduces several key enhancements focusing on enabling the use of neural networks within JuMP constraints and enhancing the handling of two-stage Mixed-Integer Linear Programming (MILP) problems.
Neural Network Integration:
Added a new function dispatch to facilitate the use of neural networks within JuMP constraints.
Specifically adapted the "relu" layer to ensure compatibility with the requirements.
https://github.com/andrewrosemberg/L2O.jl/blob/bf6384cfe56cf1e40462a1065bf51ae63c0b36e5/examples/unitcommitment/unitcommitment.jl#L297-L302
Two-Stage MILP Separation:
Implemented code to effectively separate any MILP problem into a two-stage problem.
Introduced two JuMP objects to represent the two stages of the MILP problem, providing greater flexibility in problem formulation.
https://github.com/andrewrosemberg/L2O.jl/blob/bf6384cfe56cf1e40462a1065bf51ae63c0b36e5/src/cutting_planes.jl#L88-L94
https://github.com/andrewrosemberg/L2O.jl/blob/ar/flux_constraints/src/cutting_planes.jl
Cutting Planes Approach:
Developed code to implement a cutting planes approach for solving the two-stage MILP.
This approach aims to enhance the optimization process by iteratively adding valid cutting planes to tighten the feasible region.
https://github.com/andrewrosemberg/L2O.jl/blob/bf6384cfe56cf1e40462a1065bf51ae63c0b36e5/src/cutting_planes.jl#L177
https://github.com/andrewrosemberg/L2O.jl/blob/ar/flux_constraints/src/cutting_planes.jl
ICNN Fitting:
Introduced code to fit an ICNN to the second stage of a problem created using UnitCommitment.jl.
https://github.com/andrewrosemberg/L2O.jl/blob/bf6384cfe56cf1e40462a1065bf51ae63c0b36e5/examples/unitcommitment/unitcommitment.jl#L217-L232