v0.3.1

RegularizedOptimization v0.3.1

Diff since v0.3.0

Merged pull requests:

• adds an initialization value for the ratio in the solve function (#319) (@BenjaminPINEAU)
• Update ProximalOperators: allow v1.16 (#320) (@MaxenceGollier)

v0.3.0

RegularizedOptimization v0.3.0

Diff since v0.2.0

Breaking changes

• Remove R2NModels by @MaxenceGollier in #311

Merged pull requests:

• Address editor's comments (#288) (@MohamedLaghdafHABIBOULLAH)
• Update RegularizedOptimization dependency in Project.toml of the branch paper (#289) (@MohamedLaghdafHABIBOULLAH)
• Update authors and affiliations in paper.md (#290) (@MohamedLaghdafHABIBOULLAH)
• minor changes for JOSS publication (#291) (@danielskatz)
• Update affiliations in paper.md with full country names (#292) (@MohamedLaghdafHABIBOULLAH)
• Implement SolverCore.reset!(solver) function (#293) (@MaxenceGollier)
• update SolverCore compat and add step_status (#297) (@MaxenceGollier)
• remove the demos workflow (#301) (@MaxenceGollier)
• JSO Benchmarks (#307) (@dpo)
• Remove R2NModels (#311) (@MaxenceGollier)
• fix README and CITATION.bib (#315) (@MaxenceGollier)

Closed issues:

• Documentation (#2)
• Executing R2N twice in a row does not give the same answer (#268)
• JOSS paper: Typos/Phrasing (#284)
• Demos are failing (#298)
• Update README and CITATION.BIB (#300)

v0.2.0

RegularizedOptimization v0.2.0

Diff since v0.1.0

Breaking changes

• Change the callback signature to use reg_nlp instead of nlp by @MaxenceGollier in #283
• Compute hess_op with respect to reg_nlp instead of nlp by @MaxenceGollier in #285

Merged pull requests:

• Update FreeBSD image family to 14-3 (#272) (@amontoison)
• Address reviewer comment (#273) (@MohamedLaghdafHABIBOULLAH)
• R2N: implicitely update the linear operator (#274) (@MaxenceGollier)
• Add a community guideline in the readme and update installation instructions (#276) (@MaxenceGollier)
• Set the dev branch to main in docs (#277) (@MaxenceGollier)
• Add options to recompute gradient and/or obj or not on first iteration (#280) (@MaxenceGollier)
• Change the callback signature to use reg_nlp instead of nlp (#283) (@MaxenceGollier)
• Compute hess_op with respect to reg_nlp instead of nlp (#285) (@MaxenceGollier)

Closed issues:

• Reduce the cost of opnorm of B_k (#133)
• Make R2N more flexible for users (#194)
• Switch to BestieTemplate (#227)
• JOSS Review: Functionality (paper example) (#259)
• Avoid recomputing gradient and objective function on first iteration (#278)
• The callbacks should take reg_nlp as an argument (#282)

v0.1.0

RegularizedOptimization v0.1.0

Diff since v0.0.2

Merged pull requests:

• Fix pre review comments (#249) (@MaxenceGollier)
• Remove NNMF problem and update Benchmark.jl accordingly (#254) (@MohamedLaghdafHABIBOULLAH)
• Add DOI to paper.bib entry (#257) (@MaxenceGollier)
• Update author list (#258) (@MaxenceGollier)
• Fix allocation tests on julia pre (#261) (@MaxenceGollier)
• import SolverCore.reset! explicitly (#263) (@MaxenceGollier)

Closed issues:

• Implement callback with exit criterion (#250)
• CI fail with SolverCore 0.3.9 (#262)

v0.0.2

Version 0.0.2

What's Changed

• DOI badge by @dpo in #69
• Neg tol by @dpo in #73
• update deps, stats API by @dpo in #76
• experiment with absolute + relative stopping tests by @dpo in #74
• replace Arpack with TSVD by @dpo in #79
• run demos with CI by @dpo in #80
• Remove direct access to Delta field in proximal operator by @geoffroyleconte in #83
• TR with bounds by @geoffroyleconte in #84
• R2 with bounds by @geoffroyleconte in #86
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #85
• LMTR with bounds by @geoffroyleconte in #87
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #88
• LM with bounds by @geoffroyleconte in #89
• Set β to a large value by default by @dpo in #90
• TRDH algorithm by @geoffroyleconte in #91
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #92
• Enable subsolver tolerances tuning, fix negative tolerance by @geoffroyleconte in #93
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #94
• Constrained FH tables by @geoffroyleconte in #99
• Change random seed bpdn-cstr after problem update by @geoffroyleconte in #102
• LM and LMTR with TRDH by @geoffroyleconte in #95
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #103
• [TRDH] Change display Dk norm by @geoffroyleconte in #101
• Fix some errors in demos by @geoffroyleconte in #104
• Remove allocations R2 by @geoffroyleconte in #105
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #111
• Bug fix in PG alg for options and iteration count by @AHsu98 in #110
• Add NLSModel to NNMF ouput arguments by @geoffroyleconte in #112
• Restore subsolver options by @geoffroyleconte in #114
• Change random seed FH-table by @geoffroyleconte in #116
• R2 optimality measure by @MaxenceGollier in #115
• added correct (per discussion) gradient nls-checking by @rjbaraldi in #119
• Update criticality measure by @geoffroyleconte in #117
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #121
• TRDH: prevent formatter from butchering line by @dpo in #125
• CirrusCI: test musl and M1 by @dpo in #126
• update types in ROSolverOptions by @dpo in #127
• TR: fix definition of ν by @dpo in #124
• update compat, clean up demo workflow by @dpo in #106
• update CirrusCI config by @dpo in #128
• update TRDH to use diagonal QN model by @dpo in #132
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #137
• Update .cirrus.yml to use FreeBSD 13.3 by @dpo in #140
• fix workflows, demo deps by @dpo in #142
• JSO Compliance: R2 by @MaxenceGollier in #139
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #144
• R2-solve!-0-allocation by @MohamedLaghdafHABIBOULLAH in #147
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #148
• Fix bug in LM by @MohamedLaghdafHABIBOULLAH in #152
• remove redundant return in R2 by @MaxenceGollier in #150
• Update CITATION.bib by @dpo in #154
• Update CITATION.bib by @dpo in #155
• R2: pass kwargs through by @dpo in #160
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #162
• Cirrus: Julia 1 on FreeBSD only by @dpo in #163
• Resolve R2 Allocation test by @MaxenceGollier in #164
• Augmented Lagrangian method by @aldma in #158
• Add R2N, R2DH by @MohamedLaghdafHABIBOULLAH in #153
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #169
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #171
• bump LinearOperators compat to 2.9 by @MaxenceGollier in #173
• use Julia 1.10 on GitHub CI by @dpo in #174
• Reuse Arpack in order to compute norm of Bk by @MohamedLaghdafHABIBOULLAH in #159
• LM: only recompute Jacobian norm if problem is nonlinear by @MohamedLaghdafHABIBOULLAH in #156
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #181
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #183
• ci: upgrade dep workflows by @dpo in #184
• Update the default value of theta in ROSolverOptions and update LM & LMTR accordingly by @MaxenceGollier in #180
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #185
• add step entry in RegularizedExecutionStats by @MaxenceGollier in #179
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #187
• demo workflow: upgrade setup-texlive-action by @dpo in #186
• disable FH demo by @dpo in #188
• Bump LinearOperators.jl compat to 2.10 by @MaxenceGollier in #191
• LMTR: define ν as in TR by @dpo in #190
• JSO Compliance: R2N & R2DH by @MaxenceGollier in #172
• Add v0.21 to NLPModels compatibility by @MaxenceGollier in #197
• Remove dead code by @ma...

v0.0.1

Version 0.0.1

What's Changed

• DOI badge by @dpo in #69
• Neg tol by @dpo in #73
• update deps, stats API by @dpo in #76
• experiment with absolute + relative stopping tests by @dpo in #74
• replace Arpack with TSVD by @dpo in #79
• run demos with CI by @dpo in #80
• Remove direct access to Delta field in proximal operator by @geoffroyleconte in #83
• TR with bounds by @geoffroyleconte in #84
• R2 with bounds by @geoffroyleconte in #86
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #85
• LMTR with bounds by @geoffroyleconte in #87
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #88
• LM with bounds by @geoffroyleconte in #89
• Set β to a large value by default by @dpo in #90
• TRDH algorithm by @geoffroyleconte in #91
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #92
• Enable subsolver tolerances tuning, fix negative tolerance by @geoffroyleconte in #93
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #94
• Constrained FH tables by @geoffroyleconte in #99
• Change random seed bpdn-cstr after problem update by @geoffroyleconte in #102
• LM and LMTR with TRDH by @geoffroyleconte in #95
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #103
• [TRDH] Change display Dk norm by @geoffroyleconte in #101
• Fix some errors in demos by @geoffroyleconte in #104
• Remove allocations R2 by @geoffroyleconte in #105
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #111
• Bug fix in PG alg for options and iteration count by @AHsu98 in #110
• Add NLSModel to NNMF ouput arguments by @geoffroyleconte in #112
• Restore subsolver options by @geoffroyleconte in #114
• Change random seed FH-table by @geoffroyleconte in #116
• R2 optimality measure by @MaxenceGollier in #115
• added correct (per discussion) gradient nls-checking by @rjbaraldi in #119
• Update criticality measure by @geoffroyleconte in #117
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #121
• TRDH: prevent formatter from butchering line by @dpo in #125
• CirrusCI: test musl and M1 by @dpo in #126
• update types in ROSolverOptions by @dpo in #127
• TR: fix definition of ν by @dpo in #124
• update compat, clean up demo workflow by @dpo in #106
• update CirrusCI config by @dpo in #128
• update TRDH to use diagonal QN model by @dpo in #132
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #137
• Update .cirrus.yml to use FreeBSD 13.3 by @dpo in #140
• fix workflows, demo deps by @dpo in #142
• JSO Compliance: R2 by @MaxenceGollier in #139
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #144
• R2-solve!-0-allocation by @MohamedLaghdafHABIBOULLAH in #147
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #148
• Fix bug in LM by @MohamedLaghdafHABIBOULLAH in #152
• remove redundant return in R2 by @MaxenceGollier in #150
• Update CITATION.bib by @dpo in #154
• Update CITATION.bib by @dpo in #155
• R2: pass kwargs through by @dpo in #160
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #162
• Cirrus: Julia 1 on FreeBSD only by @dpo in #163
• Resolve R2 Allocation test by @MaxenceGollier in #164
• Augmented Lagrangian method by @aldma in #158
• Add R2N, R2DH by @MohamedLaghdafHABIBOULLAH in #153
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #169
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #171
• bump LinearOperators compat to 2.9 by @MaxenceGollier in #173
• use Julia 1.10 on GitHub CI by @dpo in #174
• Reuse Arpack in order to compute norm of Bk by @MohamedLaghdafHABIBOULLAH in #159
• LM: only recompute Jacobian norm if problem is nonlinear by @MohamedLaghdafHABIBOULLAH in #156
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #181
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #183
• ci: upgrade dep workflows by @dpo in #184
• Update the default value of theta in ROSolverOptions and update LM & LMTR accordingly by @MaxenceGollier in #180
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #185
• add step entry in RegularizedExecutionStats by @MaxenceGollier in #179
• [AUTO] JuliaFormatter.jl run by @github-actions[bot] in #187
• demo workflow: upgrade setup-texlive-action by @dpo in #186
• disable FH demo by @dpo in #188
• Bump LinearOperators.jl compat to 2.10 by @MaxenceGollier in #191
• LMTR: define ν as in TR by @dpo in #190
• JSO Compliance: R2N & R2DH by @MaxenceGollier in #172
• Add v0.21 to NLPModels compatibility by @MaxenceGollier in #197
• remove unused workflows...

Release to obtain DOI only

v0.0.0

:robot: Format .jl files