v1.1.0

GraphMatFun v1.1.0

Diff since v1.0.1

Merged pull requests:

• Multiple matrices in linear combination (#75) (@jarlebring)
• Address multiple-parent :lincomb in C code generation (#77) (@mfasi)
• Remove unused code gen_code/gen_julia_code.jl (#79) (@mfasi)

Closed issues:

• DegOpt / Graph refactorization (#73)
• Warning messages in tests (#74)
• New trivial lincomb nodes (#78)

v1.0.1

GraphMatFun v1.0.1

Diff since v1.0.0

v1.0.0

GraphMatFun v1.0.0

Closed issues:

• Move graph generators (#1)
• Code generators (#2)
• Fitting for graphs "linear" in some coefficients (#3)
• Move optimization routines to git (#4)
• expm of Sastre, Ibanez, Defez (#5)
• P-S rule as a general poly recursion call (#6)
• Bug in LangJulia generation (#7)
• Parameter cref in graph compression functions (#8)
• Should eval_graph have an input=:A kwarg? (#9)
• backward error code (#10)
• Add general_poly_recursion version of polynomial versions (#11)
• unit test error (#12)
• recursive -> degopt (#13)
• extraction of degopt coeffs (#14)
• More linear combinations in degopt? (#15)
• export_compgraph bug (#16)
• Code gen LangC identity matrix lincomb (#17)
• JuliaLang code generation slow I-operator (#18)
• get_topo_order_degopt (#19)
• opt over real coeffs (#20)
• Normalization of first degopt row (#21)
• DegOpt data structure (#22)
• compress graph bug (#23)
• get_topo_order_degopt interlace (#24)
• topo order vs priohelp (#25)
• Restructure cpu-time simulations (#26)
• compress_graph talkative (#27)
• Is src/gen_julia_matlab_code.jl obsolete? (#28)
• Julia code gen improvements (#29)
• Sastre code gen (#30)
• Generate code in separate directory (#31)
• Generate code with precomputed nodes (#32)
• Julia expm too many mat-mats (#33)
• get_topo_order identity matrix (#34)
• Degopt and generated julia code (#35)
• Dot fusion and extracting sums (#36)
• Output list manipulation (#37)
• Rename node for :A (#38)
• Degopt display (#39)
• Generator for recent chebyshev result (#40)
• code snippet indentation (#41)
• Degopt uses compress_keys=ture (#42)
• function name with ! in julia generated code (#43)
• gen_degopt_poly(degopt) -> graph_degopt(degopt) (#44)
• Makefile cputiming variable (#45)
• Parsing BigFloat in generated Julia code (#46)
• code_gen julia bug (#47)
• Move simulations to ... (#48)
• graph_sid for lower k (#50)
• Warnings from code generation (#51)
• get degtopt matrices (#52)
• Remove or document add_artificial! (#53)
• Separate functionality in compute_bwd_err (#54)
• References not appearing in documentation (#55)
• Documentation for normalize! does not work online (#56)
• Minor documentation issues (#57)
• Improve test code coverage (#58)
• Improve julia code gen (#59)
• One too many allocations in generated julia code (#60)
• More memory than expected in graph_exp_native_jl: (#61)
• Bug in compress_graph_zero_coeff! (#62)
• Transition of CI (#68)

Merged pull requests:

• Add input kwarg to functions for graph evaluation and optimization (#49) (@mfasi)
• CompatHelper: bump compat for Polynomials to 3, (keep existing compat) (#65) (@github-actions[bot])
• CompatHelper: bump compat for Roots to 2, (keep existing compat) (#66) (@github-actions[bot])
• CompatHelper: bump compat for GenericLinearAlgebra to 0.3, (keep existing compat) (#67) (@github-actions[bot])
• More GitHub actions (#69) (@mfasi)