Built-in algorithm classes (29)¶
One class per built-in algorithm — 26 uniformly generated (table-driven
from crates/components/src/presets.rs), plus GeneticAlgorithm
(auto-dispatch over 5 space-typed presets) and DifferentialEvolution
(dispatch over 3 mutation variants), both hand-written, plus NSGA2 (a
thin skin over sezgi.mo.nsga2, not a preset). Every class but NSGA2
shares the same shape: __init__(pop_size=..., **preset_kwargs),
run(problem, budget, seed=0, run_id=0, log_dir=None) -> SolveResult.
NSGA2 is the one exception — its own docstring describes the
multi-objective result dictionary its run() returns instead.
These classes are created at runtime (type()), not via class
statements, so this page renders under mkdocstrings'
force_inspection: true handler option, which mkdocstrings requires for
classes it cannot discover through normal static inspection.
builtins ¶
Built-in algorithm wrapper classes (M4-1 Task 5) -- one class per preset
builder in crates/components/src/presets.rs (34 pub fn builders,
verified against sezgi.presets's own SimpleNamespace, which exposes all
34 -- see this task's report for the full reconciliation table: every
preset has exactly one class, no orphan on either side), plus NSGA2 (a
thin sezgi.mo.nsga2 skin, NOT one of the 34 presets -- see its own
docstring).
TABLE-DRIVEN (ruling: "no hand-divergence"): _PRESET_TABLE below is the
single source of truth; _make_preset_class generates every uniformly
-shaped wrapper class from it. GeneticAlgorithm (auto-dispatch over 5
presets) and DifferentialEvolution (variant= over 3 presets) are
hand-written because their dispatch logic genuinely differs from the
uniform "one class, one preset" shape -- but both funnel through the SAME
_run_spec helper every generated class uses, so the actual
spec-building-then-solve()-then-wrap-in-SolveResult machinery has
exactly one call site in this file, not N. The wrap-in-SolveResult step
itself (_wrap_result) is a single shared helper in sezgi.algo
(final-review fix 3), imported here rather than reimplemented -- the SAME
helper sezgi.algorithm.Algorithm.run uses, so that translation has
exactly one implementation across the whole crate, not two.
Every wrapper (GeneticAlgorithm/DifferentialEvolution/NSGA2 included, with
NSGA2's own documented exception): __init__(pop_size=<preset's own
documented/conventional default>, **preset_kwargs),
run(problem, budget, seed=0, run_id=0, log_dir=None) ->
sezgi.algo.SolveResult. run() builds the preset's AlgorithmSpec via
sezgi.presets.<name>(...) (UNCHANGED) and calls sezgi.solve(...)
(UNCHANGED) internally -- literally ruling 1's "solve()/presets.* stay as
compat internals": these wrapper classes add a constructor + a result-shape
translation, nothing else. Each class's docstring names its preset function
and Rust source (presets.rs:<lines>), which carries the algorithm's own
academic citation (unchanged, in the Rust doc comment).
EvolutionStrategy ¶
EvolutionStrategy(pop_size=None, **preset_kwargs)
(mu+lambda)-Evolution Strategy -- gen/step over a caller-selected mutation distribution (dist=, plus its own params: mean=/sigma= for gaussian (default), loc=/scale= for cauchy/laplace, alpha= for levy, nu= for student_t) paired with replace/mu-plus-lambda.
Delegates to sezgi.presets.es_mu_plus_lambda (crates/components/src/presets.rs:56-70).
Constructs a EvolutionStrategy instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.es_mu_plus_lambda, defaults to 20 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.es_mu_plus_lambda(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:56-70).
__doc__
class-attribute
¶
__doc__ = '(mu+lambda)-Evolution Strategy -- gen/step over a caller-selected mutation distribution (dist=, plus its own params: mean=/sigma= for gaussian (default), loc=/scale= for cauchy/laplace, alpha= for levy, nu= for student_t) paired with replace/mu-plus-lambda.\n\nDelegates to sezgi.presets.es_mu_plus_lambda (crates/components/src/presets.rs:56-70).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.es_mu_plus_lambda(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
ParticleSwarm ¶
ParticleSwarm(pop_size=None, **preset_kwargs)
Particle Swarm Optimization (Clerc & Kennedy constriction variant, spec name "pso/clerc-kennedy") -- gen/pso (w=0.7298, c1=c2=1.49618) paired with replace/pso-commit.
Delegates to sezgi.presets.pso (crates/components/src/presets.rs:243-257).
Constructs a ParticleSwarm instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.pso, defaults to 20 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.pso(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:243-257).
__doc__
class-attribute
¶
__doc__ = 'Particle Swarm Optimization (Clerc & Kennedy constriction variant, spec name "pso/clerc-kennedy") -- gen/pso (w=0.7298, c1=c2=1.49618) paired with replace/pso-commit.\n\nDelegates to sezgi.presets.pso (crates/components/src/presets.rs:243-257).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.pso(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
SimulatedAnnealing ¶
SimulatedAnnealing(pop_size=None, **preset_kwargs)
Simulated Annealing (Metropolis acceptance, geometric cooling t0=1.0, alpha=0.999) -- gen/step (gaussian, sigma=0.5) paired with replace/metropolis. Single-trajectory: pop_size is fixed at 1 by the preset itself.
Delegates to sezgi.presets.sa (crates/components/src/presets.rs:743-759).
Constructs a SimulatedAnnealing instance; see the class docstring for the algorithm itself. pop_size: NOT a free parameter -- SimulatedAnnealing is a single-trajectory search, fixed at 1 by sezgi.presets.sa's own Rust signature (presets.rs:743-759); passing anything other than None or 1 raises ValueError. preset_kwargs: forwarded UNCHANGED to sezgi.presets.sa(budget, preset_kwargs) at run() time.
__doc__
class-attribute
¶
__doc__ = 'Simulated Annealing (Metropolis acceptance, geometric cooling t0=1.0, alpha=0.999) -- gen/step (gaussian, sigma=0.5) paired with replace/metropolis. Single-trajectory: pop_size is fixed at 1 by the preset itself.\n\nDelegates to sezgi.presets.sa (crates/components/src/presets.rs:743-759).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.sa(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
SHADE ¶
SHADE(pop_size=None, **preset_kwargs)
SHADE (Success-History-based Adaptive DE, Tanabe & Fukunaga 2013) -- gen/de-shade (h=6, p=0.11) paired with replace/shade and adapter/shade-history.
Delegates to sezgi.presets.shade (crates/components/src/presets.rs:761-774).
Constructs a SHADE instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.shade, defaults to 20 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.shade(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:761-774).
__doc__
class-attribute
¶
__doc__ = 'SHADE (Success-History-based Adaptive DE, Tanabe & Fukunaga 2013) -- gen/de-shade (h=6, p=0.11) paired with replace/shade and adapter/shade-history.\n\nDelegates to sezgi.presets.shade (crates/components/src/presets.rs:761-774).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.shade(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
LSHADE ¶
LSHADE(pop_size=None, **preset_kwargs)
L-SHADE (Linear-population-size-reduction SHADE, Tanabe & Fukunaga 2014) -- same gen/de-shade + replace/shade as SHADE, plus adapter/shade-lshade for the linear population shrink. pop_size is DERIVED as 18*dim (the preset's own formula) -- not a free parameter of this wrapper.
Delegates to sezgi.presets.lshade (crates/components/src/presets.rs:776-791).
Constructs a LSHADE instance; see the class docstring for the algorithm itself. pop_size: NOT a free parameter -- LSHADE DERIVES its population from the problem's own dimensionality at run() time (sezgi.presets.lshade's own formula, presets.rs:776-791 -- see the class docstring for the exact formula); passing anything other than None raises ValueError. preset_kwargs: forwarded UNCHANGED to sezgi.presets.lshade(dim, budget, preset_kwargs) at run() time.
__doc__
class-attribute
¶
__doc__ = "L-SHADE (Linear-population-size-reduction SHADE, Tanabe & Fukunaga 2014) -- same gen/de-shade + replace/shade as SHADE, plus adapter/shade-lshade for the linear population shrink. pop_size is DERIVED as 18*dim (the preset's own formula) -- not a free parameter of this wrapper.\n\nDelegates to sezgi.presets.lshade (crates/components/src/presets.rs:776-791)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.lshade(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
CMAES ¶
CMAES(pop_size=None, **preset_kwargs)
(mu/mu_w,lambda)-CMA-ES (Hansen's tutorial form, positive-weights variant) -- gen/cma paired with replace/cma-update. pop_size is lambda; Hansen's own guideline is 4+floor(3*ln(dim)) (see CMAESIpop, which computes this automatically) -- this class leaves the choice to the caller, matching the Rust preset's own signature.
Delegates to sezgi.presets.cmaes (crates/components/src/presets.rs:798-811).
Constructs a CMAES instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.cmaes, defaults to 20 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.cmaes(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:798-811).
__doc__
class-attribute
¶
__doc__ = "(mu/mu_w,lambda)-CMA-ES (Hansen's tutorial form, positive-weights variant) -- gen/cma paired with replace/cma-update. pop_size is lambda; Hansen's own guideline is 4+floor(3*ln(dim)) (see CMAESIpop, which computes this automatically) -- this class leaves the choice to the caller, matching the Rust preset's own signature.\n\nDelegates to sezgi.presets.cmaes (crates/components/src/presets.rs:798-811)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.cmaes(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
CMAESIpop ¶
CMAESIpop(pop_size=None, **preset_kwargs)
CMA-ES with IPOP-style stagnation restarts (M2b Task 12) -- same gen/cma + replace/cma-update stage as CMAES, plus restart/stagnation (patience=2000, sizing=ipop, factor=2.0, max_pop=512). pop_size is DERIVED as 4+floor(3*ln(dim)) (Hansen's default, computed by the preset itself since IPOP restarts scale from this starting population) -- not a free parameter of this wrapper.
Delegates to sezgi.presets.cmaes_ipop (crates/components/src/presets.rs:813-834).
Constructs a CMAESIpop instance; see the class docstring for the algorithm itself. pop_size: NOT a free parameter -- CMAESIpop DERIVES its population from the problem's own dimensionality at run() time (sezgi.presets.cmaes_ipop's own formula, presets.rs:813-834 -- see the class docstring for the exact formula); passing anything other than None raises ValueError. preset_kwargs: forwarded UNCHANGED to sezgi.presets.cmaes_ipop(dim, budget, preset_kwargs) at run() time.
__doc__
class-attribute
¶
__doc__ = "CMA-ES with IPOP-style stagnation restarts (M2b Task 12) -- same gen/cma + replace/cma-update stage as CMAES, plus restart/stagnation (patience=2000, sizing=ipop, factor=2.0, max_pop=512). pop_size is DERIVED as 4+floor(3*ln(dim)) (Hansen's default, computed by the preset itself since IPOP restarts scale from this starting population) -- not a free parameter of this wrapper.\n\nDelegates to sezgi.presets.cmaes_ipop (crates/components/src/presets.rs:813-834)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.cmaes_ipop(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
NelderMead ¶
NelderMead(pop_size=None, **preset_kwargs)
Nelder-Mead simplex (M2b Task 13) -- gen/nelder-mead paired with replace/nelder-mead. pop_size is DERIVED as dim+1 (the population IS the simplex) -- not a free parameter of this wrapper.
Delegates to sezgi.presets.nelder_mead (crates/components/src/presets.rs:836-853).
Constructs a NelderMead instance; see the class docstring for the algorithm itself. pop_size: NOT a free parameter -- NelderMead DERIVES its population from the problem's own dimensionality at run() time (sezgi.presets.nelder_mead's own formula, presets.rs:836-853 -- see the class docstring for the exact formula); passing anything other than None raises ValueError. preset_kwargs: forwarded UNCHANGED to sezgi.presets.nelder_mead(dim, budget, preset_kwargs) at run() time.
__doc__
class-attribute
¶
__doc__ = 'Nelder-Mead simplex (M2b Task 13) -- gen/nelder-mead paired with replace/nelder-mead. pop_size is DERIVED as dim+1 (the population IS the simplex) -- not a free parameter of this wrapper.\n\nDelegates to sezgi.presets.nelder_mead (crates/components/src/presets.rs:836-853).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.nelder_mead(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
RandomSearch ¶
RandomSearch(pop_size=None, **preset_kwargs)
Uniform random resampling (gen/uniform-resample) paired with replace/mu-plus-lambda -- the baseline every other algorithm should beat.
Delegates to sezgi.presets.random_search (crates/components/src/presets.rs:855-869).
Constructs a RandomSearch instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.random_search, defaults to 20 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.random_search(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:855-869).
__doc__
class-attribute
¶
__doc__ = 'Uniform random resampling (gen/uniform-resample) paired with replace/mu-plus-lambda -- the baseline every other algorithm should beat.\n\nDelegates to sezgi.presets.random_search (crates/components/src/presets.rs:855-869).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.random_search(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
GreyWolfOptimizer ¶
GreyWolfOptimizer(pop_size=None, **preset_kwargs)
Grey Wolf Optimizer (Mirjalili, Mirjalili & Lewis 2014) -- gen/gwo paired with replace/generational (non-elitist by construction). pop_size is the pack size; canonical is 30 per the source paper.
Delegates to sezgi.presets.gwo (crates/components/src/presets.rs:259-279).
Constructs a GreyWolfOptimizer instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.gwo, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.gwo(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:259-279).
__doc__
class-attribute
¶
__doc__ = 'Grey Wolf Optimizer (Mirjalili, Mirjalili & Lewis 2014) -- gen/gwo paired with replace/generational (non-elitist by construction). pop_size is the pack size; canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.gwo (crates/components/src/presets.rs:259-279).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.gwo(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
WhaleOptimization ¶
WhaleOptimization(pop_size=None, **preset_kwargs)
Whale Optimization Algorithm (Mirjalili & Lewis 2016) -- gen/woa paired with replace/generational. pop_size is the school size; canonical is 30 per the source paper.
Delegates to sezgi.presets.woa (crates/components/src/presets.rs:281-302).
Constructs a WhaleOptimization instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.woa, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.woa(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:281-302).
__doc__
class-attribute
¶
__doc__ = 'Whale Optimization Algorithm (Mirjalili & Lewis 2016) -- gen/woa paired with replace/generational. pop_size is the school size; canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.woa (crates/components/src/presets.rs:281-302).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.woa(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
HarmonySearch ¶
HarmonySearch(pop_size=None, **preset_kwargs)
Harmony Search (Geem, Kim & Loganathan 2001) -- gen/hs paired with replace/worst-if-better. pop_size is HMS (Harmony Memory Size); canonical is 30 per the source paper.
Delegates to sezgi.presets.harmony_search (crates/components/src/presets.rs:304-327).
Constructs a HarmonySearch instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.harmony_search, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.harmony_search(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:304-327).
__doc__
class-attribute
¶
__doc__ = 'Harmony Search (Geem, Kim & Loganathan 2001) -- gen/hs paired with replace/worst-if-better. pop_size is HMS (Harmony Memory Size); canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.harmony_search (crates/components/src/presets.rs:304-327).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.harmony_search(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
CuckooSearch ¶
CuckooSearch(pop_size=None, **preset_kwargs)
Cuckoo Search (Yang & Deb 2009) -- gen/cuckoo_levy paired with replace/one-to-one-greedy and adapter/abandon-worst-fraction (pa=0.25). pop_size is the nest count; canonical is 25 per the source paper.
Delegates to sezgi.presets.cuckoo_search (crates/components/src/presets.rs:329-359).
Constructs a CuckooSearch instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.cuckoo_search, defaults to 25 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.cuckoo_search(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:329-359).
__doc__
class-attribute
¶
__doc__ = 'Cuckoo Search (Yang & Deb 2009) -- gen/cuckoo_levy paired with replace/one-to-one-greedy and adapter/abandon-worst-fraction (pa=0.25). pop_size is the nest count; canonical is 25 per the source paper.\n\nDelegates to sezgi.presets.cuckoo_search (crates/components/src/presets.rs:329-359).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.cuckoo_search(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
GrasshopperOptimization ¶
GrasshopperOptimization(pop_size=None, **preset_kwargs)
Grasshopper Optimisation Algorithm (Saremi, Mirjalili & Lewis 2017) -- gen/goa paired with replace/generational. pop_size is the swarm size; canonical is 30 per the source paper.
Delegates to sezgi.presets.goa (crates/components/src/presets.rs:361-386).
Constructs a GrasshopperOptimization instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.goa, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.goa(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:361-386).
__doc__
class-attribute
¶
__doc__ = 'Grasshopper Optimisation Algorithm (Saremi, Mirjalili & Lewis 2017) -- gen/goa paired with replace/generational. pop_size is the swarm size; canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.goa (crates/components/src/presets.rs:361-386).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.goa(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
SineCosineAlgorithm ¶
SineCosineAlgorithm(pop_size=None, **preset_kwargs)
Sine Cosine Algorithm (Mirjalili 2016) -- gen/sca paired with replace/generational. pop_size is the number of search agents; canonical is 30 per the source paper.
Delegates to sezgi.presets.sca (crates/components/src/presets.rs:388-411).
Constructs a SineCosineAlgorithm instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.sca, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.sca(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:388-411).
__doc__
class-attribute
¶
__doc__ = 'Sine Cosine Algorithm (Mirjalili 2016) -- gen/sca paired with replace/generational. pop_size is the number of search agents; canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.sca (crates/components/src/presets.rs:388-411).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.sca(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
JAYA ¶
JAYA(pop_size=None, **preset_kwargs)
JAYA (Rao 2016) -- gen/jaya paired with replace/one-to-one-greedy. pop_size is the candidate count; canonical is 30 per this wave's own convention (the paper itself demonstrates with 5).
Delegates to sezgi.presets.jaya (crates/components/src/presets.rs:413-438).
Constructs a JAYA instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.jaya, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.jaya(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:413-438).
__doc__
class-attribute
¶
__doc__ = "JAYA (Rao 2016) -- gen/jaya paired with replace/one-to-one-greedy. pop_size is the candidate count; canonical is 30 per this wave's own convention (the paper itself demonstrates with 5).\n\nDelegates to sezgi.presets.jaya (crates/components/src/presets.rs:413-438)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.jaya(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
MothFlameOptimization ¶
MothFlameOptimization(pop_size=None, **preset_kwargs)
Moth-Flame Optimization (Mirjalili 2015) -- gen/mfo paired with replace/generational and adapter/mfo-flame-update (the flame memory). pop_size is the number of search agents; canonical is 30 per the source paper.
Delegates to sezgi.presets.mfo (crates/components/src/presets.rs:440-464).
Constructs a MothFlameOptimization instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.mfo, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.mfo(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:440-464).
__doc__
class-attribute
¶
__doc__ = 'Moth-Flame Optimization (Mirjalili 2015) -- gen/mfo paired with replace/generational and adapter/mfo-flame-update (the flame memory). pop_size is the number of search agents; canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.mfo (crates/components/src/presets.rs:440-464).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.mfo(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
SalpSwarm ¶
SalpSwarm(pop_size=None, **preset_kwargs)
Salp Swarm Algorithm (Mirjalili et al. 2017) -- gen/ssa paired with replace/generational. pop_size is the number of salps; canonical is 30 per the source paper.
Delegates to sezgi.presets.ssa (crates/components/src/presets.rs:466-491).
Constructs a SalpSwarm instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.ssa, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.ssa(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:466-491).
__doc__
class-attribute
¶
__doc__ = 'Salp Swarm Algorithm (Mirjalili et al. 2017) -- gen/ssa paired with replace/generational. pop_size is the number of salps; canonical is 30 per the source paper.\n\nDelegates to sezgi.presets.ssa (crates/components/src/presets.rs:466-491).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.ssa(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
FireflyAlgorithm ¶
FireflyAlgorithm(pop_size=None, **preset_kwargs)
Firefly Algorithm (Yang, X.-S., Nature-Inspired Metaheuristic Algorithms, 2nd ed., Luniver Press, 2010) -- gen/fa paired with replace/generational. pop_size is the number of fireflies; canonical is 25 per this wave's own convention (the source's own demo uses 20).
Delegates to sezgi.presets.firefly (crates/components/src/presets.rs:493-521).
Constructs a FireflyAlgorithm instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.firefly, defaults to 25 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.firefly(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:493-521).
__doc__
class-attribute
¶
__doc__ = "Firefly Algorithm (Yang, X.-S., Nature-Inspired Metaheuristic Algorithms, 2nd ed., Luniver Press, 2010) -- gen/fa paired with replace/generational. pop_size is the number of fireflies; canonical is 25 per this wave's own convention (the source's own demo uses 20).\n\nDelegates to sezgi.presets.firefly (crates/components/src/presets.rs:493-521)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.firefly(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
BatAlgorithm ¶
BatAlgorithm(pop_size=None, **preset_kwargs)
Bat Algorithm (Yang, X.-S. 2010, NICSO) -- gen/ba paired with the new replace/bat-loudness-greedy. pop_size is the number of bats; canonical is 30 per this wave's own convention (the source's own demo uses 20).
Delegates to sezgi.presets.bat (crates/components/src/presets.rs:523-550).
Constructs a BatAlgorithm instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.bat, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.bat(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:523-550).
__doc__
class-attribute
¶
__doc__ = "Bat Algorithm (Yang, X.-S. 2010, NICSO) -- gen/ba paired with the new replace/bat-loudness-greedy. pop_size is the number of bats; canonical is 30 per this wave's own convention (the source's own demo uses 20).\n\nDelegates to sezgi.presets.bat (crates/components/src/presets.rs:523-550)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.bat(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
FlowerPollination ¶
FlowerPollination(pop_size=None, **preset_kwargs)
Flower Pollination Algorithm (Yang, X.-S. 2012, UCNC) -- gen/fpa paired with replace/one-to-one-greedy. pop_size is the flower/pollen-gamete count; canonical is 25 per the source's demo.
Delegates to sezgi.presets.fpa (crates/components/src/presets.rs:552-580).
Constructs a FlowerPollination instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.fpa, defaults to 25 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.fpa(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:552-580).
__doc__
class-attribute
¶
__doc__ = "Flower Pollination Algorithm (Yang, X.-S. 2012, UCNC) -- gen/fpa paired with replace/one-to-one-greedy. pop_size is the flower/pollen-gamete count; canonical is 25 per the source's demo.\n\nDelegates to sezgi.presets.fpa (crates/components/src/presets.rs:552-580)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.fpa(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
TLBO ¶
TLBO(pop_size=None, **preset_kwargs)
Teaching-Learning-Based Optimization (Rao, Savsani & Vakharia 2011) -- sezgi's first MULTI-STAGE preset: gen/tlbo-teacher then gen/tlbo-learner, each paired with replace/one-to-one-greedy. pop_size is the class size; canonical is 30 per the source paper. A full generation costs 2*pop_size evaluations.
Delegates to sezgi.presets.tlbo (crates/components/src/presets.rs:582-620).
Constructs a TLBO instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.tlbo, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.tlbo(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:582-620).
__doc__
class-attribute
¶
__doc__ = "Teaching-Learning-Based Optimization (Rao, Savsani & Vakharia 2011) -- sezgi's first MULTI-STAGE preset: gen/tlbo-teacher then gen/tlbo-learner, each paired with replace/one-to-one-greedy. pop_size is the class size; canonical is 30 per the source paper. A full generation costs 2*pop_size evaluations.\n\nDelegates to sezgi.presets.tlbo (crates/components/src/presets.rs:582-620)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.tlbo(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
HarrisHawks ¶
HarrisHawks(pop_size=None, **preset_kwargs)
Harris Hawks Optimization (Heidari, Mirjalili, Faris, Aljarah, Mafarja & Chen 2019) -- gen/hho paired with replace/generational. pop_size is the hawk count; canonical is 30 per the source's own demo.
Delegates to sezgi.presets.hho (crates/components/src/presets.rs:622-648).
Constructs a HarrisHawks instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.hho, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.hho(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:622-648).
__doc__
class-attribute
¶
__doc__ = "Harris Hawks Optimization (Heidari, Mirjalili, Faris, Aljarah, Mafarja & Chen 2019) -- gen/hho paired with replace/generational. pop_size is the hawk count; canonical is 30 per the source's own demo.\n\nDelegates to sezgi.presets.hho (crates/components/src/presets.rs:622-648)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.hho(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
AntLion ¶
AntLion(pop_size=None, **preset_kwargs)
Ant Lion Optimizer (Mirjalili 2015) -- gen/alo paired with replace/mu-plus-lambda. pop_size is the ant/antlion count; canonical is 25 per this wave's own convention.
Delegates to sezgi.presets.alo (crates/components/src/presets.rs:650-675).
Constructs a AntLion instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.alo, defaults to 25 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.alo(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:650-675).
__doc__
class-attribute
¶
__doc__ = "Ant Lion Optimizer (Mirjalili 2015) -- gen/alo paired with replace/mu-plus-lambda. pop_size is the ant/antlion count; canonical is 25 per this wave's own convention.\n\nDelegates to sezgi.presets.alo (crates/components/src/presets.rs:650-675)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.alo(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
ArtificialBeeColony ¶
ArtificialBeeColony(pop_size=None, **preset_kwargs)
Artificial Bee Colony (Karaboga 2005, TR-06 / Karaboga & Basturk 2007) -- gen/abc-employed paired with the new replace/abc-trial-greedy and adapter/abc-onlooker-scout. pop_size IS SN (the food-source count), NOT Karaboga's colony size NP=2SN; canonical is 20 per this module's own resolved convention. A full cycle costs 2pop_size evaluations (+1 when a scout fires).
Delegates to sezgi.presets.abc (crates/components/src/presets.rs:677-708).
Constructs a ArtificialBeeColony instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.abc, defaults to 20 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.abc(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:677-708).
__doc__
class-attribute
¶
__doc__ = "Artificial Bee Colony (Karaboga 2005, TR-06 / Karaboga & Basturk 2007) -- gen/abc-employed paired with the new replace/abc-trial-greedy and adapter/abc-onlooker-scout. pop_size IS SN (the food-source count), NOT Karaboga's colony size NP=2*SN; canonical is 20 per this module's own resolved convention. A full cycle costs 2*pop_size evaluations (+1 when a scout fires).\n\nDelegates to sezgi.presets.abc (crates/components/src/presets.rs:677-708)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.abc(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
GravitationalSearch ¶
GravitationalSearch(pop_size=None, **preset_kwargs)
Gravitational Search Algorithm (Rashedi, Nezamabadi-pour & Saryazdi 2009) -- gen/gsa paired with replace/generational. pop_size is the agent count; canonical is 30 per this wave's own convention.
Delegates to sezgi.presets.gsa (crates/components/src/presets.rs:710-741).
Constructs a GravitationalSearch instance; see the class docstring for the algorithm itself. pop_size: population size for sezgi.presets.gsa, defaults to 30 (see the class docstring for where this default comes from). preset_kwargs: forwarded UNCHANGED to sezgi.presets.gsa(pop_size, budget, preset_kwargs) at run() time -- any keyword that preset's own Rust signature accepts beyond pop_size/budget (presets.rs:710-741).
__doc__
class-attribute
¶
__doc__ = "Gravitational Search Algorithm (Rashedi, Nezamabadi-pour & Saryazdi 2009) -- gen/gsa paired with replace/generational. pop_size is the agent count; canonical is 30 per this wave's own convention.\n\nDelegates to sezgi.presets.gsa (crates/components/src/presets.rs:710-741)."
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs sezgi.presets.gsa(...) via sezgi.solve() and wraps the result in sezgi.algo.SolveResult. See the class docstring for pop_size's own semantics under this preset.
GeneticAlgorithm ¶
GeneticAlgorithm(pop_size=20, representation=None, **preset_kwargs)
Genetic Algorithm (real-coded SBX / permutation OX / binary /
integer / categorical) -- auto-dispatches to one of
sezgi.presets.ga_real / ga_perm / ga_bin / ga_int / ga_cat
(crates/components/src/presets.rs:72-241) based on the problem's own
space block kinds, read via the native Problem handle's .blocks()
accessor (M4-1 Task 5: a minimal read-only introspection accessor added
to PyProblem in py-sezgi/src/lib.rs -- dim()/bounds()/optimum()
alone cannot distinguish an all-Float space from an all-Permutation
one; see this task's report):
all-Float -> presets.ga_real
all-Permutation -> presets.ga_perm
all-Binary -> presets.ga_bin
all-Int -> presets.ga_int
all-Categorical -> presets.ga_cat
A space MIXING block kinds (e.g. Float + Int together) has no ga_* preset in this milestone -- run() raises NotImplementedError naming the gen/compound generator (crates/components/src/compound.rs) via a hand-written AlgorithmSpec dict/TOML passed directly to sezgi.solve() as this milestone's documented workaround (see sezgi.problems.mixed_diagnostic's own doc for a worked mixed-space scaffold problem built for exactly this path).
representation= (one of "real"/"perm"/"bin"/"int"/"cat", a
constructor kwarg) OVERRIDES auto-dispatch entirely -- block-kind
introspection is skipped whenever it is given.
After a run() call, self.dispatched_representation holds the
representation actually engaged (e.g. "real") -- an introspectable
accessor proving which preset ran, independent of representation=
itself (which stays None under auto-dispatch).
pop_size: population size, forwarded to whichever ga_ preset run() ultimately dispatches to (default 20, matching this module's own convention for the ga_ presets -- see _PRESET_TABLE's header comment). representation: overrides auto-dispatch entirely when given (one of "real"/"perm"/"bin"/"int"/"cat"); None (default) means auto-dispatch from the problem's own space at run() time -- see the class docstring for the full dispatch table and the Mixed- space error. preset_kwargs: forwarded UNCHANGED to the dispatched sezgi.presets.ga_*(pop_size, budget, preset_kwargs) at run() time.
__doc__
class-attribute
¶
__doc__ = 'Genetic Algorithm (real-coded SBX / permutation OX / binary /\n integer / categorical) -- auto-dispatches to one of\n sezgi.presets.ga_real / ga_perm / ga_bin / ga_int / ga_cat\n (crates/components/src/presets.rs:72-241) based on the problem\'s own\n space block kinds, read via the native Problem handle\'s `.blocks()`\n accessor (M4-1 Task 5: a minimal read-only introspection accessor added\n to `PyProblem` in py-sezgi/src/lib.rs -- `dim()`/`bounds()`/`optimum()`\n alone cannot distinguish an all-Float space from an all-Permutation\n one; see this task\'s report):\n\n all-Float -> presets.ga_real\n all-Permutation -> presets.ga_perm\n all-Binary -> presets.ga_bin\n all-Int -> presets.ga_int\n all-Categorical -> presets.ga_cat\n\n A space MIXING block kinds (e.g. Float + Int together) has no ga_*\n preset in this milestone -- run() raises NotImplementedError naming the\n gen/compound generator (crates/components/src/compound.rs) via a\n hand-written AlgorithmSpec dict/TOML passed directly to sezgi.solve() as\n this milestone\'s documented workaround (see\n sezgi.problems.mixed_diagnostic\'s own doc for a worked mixed-space\n scaffold problem built for exactly this path).\n\n `representation=` (one of "real"/"perm"/"bin"/"int"/"cat", a\n constructor kwarg) OVERRIDES auto-dispatch entirely -- block-kind\n introspection is skipped whenever it is given.\n\n After a run() call, `self.dispatched_representation` holds the\n representation actually engaged (e.g. "real") -- an introspectable\n accessor proving which preset ran, independent of `representation=`\n itself (which stays None under auto-dispatch).\n '
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Resolves the representation (auto-dispatch from problem.space(), or self.representation if forced), runs the matching sezgi.presets.ga_* preset via sezgi.solve(), and wraps the result in sezgi.algo.SolveResult. Sets self.dispatched_representation as a side effect (see the class docstring). Raises NotImplementedError for a Mixed-kind space unless representation= was given.
DifferentialEvolution ¶
DifferentialEvolution(pop_size=20, variant='rand1', **preset_kwargs)
Differential Evolution -- delegates to sezgi.presets.de_rand_1
(default, DE/rand/1/bin) / de_best_1 (DE/best/1/bin) / jde
(self-adaptive jDE) (crates/components/src/presets.rs:8-54), selected
by variant= ("rand1" | "best1" | "jde") at construction time. All
three presets share the identical (pop_size, budget) signature, so
variant= is the only dispatch axis -- no problem introspection needed
(unlike GeneticAlgorithm's space-driven auto-dispatch).
pop_size: population size, forwarded to whichever de_/jde preset run() dispatches to (default 20, matching this module's own convention -- all three variants share the same (pop_size, budget) signature, presets.rs:8-54). variant: one of "rand1" (default, sezgi.presets.de_rand_1) / "best1" (de_best_1) / "jde" (jde) -- see the class docstring for what each variant means. *preset_kwargs: forwarded UNCHANGED to the dispatched preset at run() time (none of the three currently accept any beyond pop_size/budget).
__doc__
class-attribute
¶
__doc__ = 'Differential Evolution -- delegates to sezgi.presets.de_rand_1\n (default, DE/rand/1/bin) / de_best_1 (DE/best/1/bin) / jde\n (self-adaptive jDE) (crates/components/src/presets.rs:8-54), selected\n by `variant=` ("rand1" | "best1" | "jde") at construction time. All\n three presets share the identical (pop_size, budget) signature, so\n variant= is the only dispatch axis -- no problem introspection needed\n (unlike GeneticAlgorithm\'s space-driven auto-dispatch).'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, budget, seed=0, run_id=0, log_dir=None)
Runs the preset self.variant selected at construction time (sezgi.presets.de_rand_1 / de_best_1 / jde) via sezgi.solve(), and wraps the result in sezgi.algo.SolveResult.
NSGA2 ¶
NSGA2(pop_size, eta_c=20.0, eta_m=20.0, p_c=0.9, p_m=None, p_c_bin=0.9, p_m_bin=None, p_c_cat=0.9, p_m_cat=None)
Class skin over sezgi.mo.nsga2 (crates/components/src/nsga2.rs: 2370-2377, sezgi.mo.nsga2's own docstring in py-sezgi/python/sezgi/init.py for the full parameter contract) -- ZERO new MO capability: run() delegates to mo.nsga2 VERBATIM, same positional/keyword arguments, same return dict. Authoring an NSGA-II variant as an engine-hosted Python callback (the way sezgi.Algorithm lets a scalar algorithm's generate() be authored in Python) is explicitly OUT OF SCOPE this milestone: NSGA-II's own selection/crossover/replacement loop is hard-coded Rust (nsga2_run_float_impl/nsga2_run_binary_impl/nsga2_run_mixed_impl), not routed through the Registry/AlgorithmSpec/Engine/Generator machinery sezgi.Algorithm targets -- see the research doc's §B7 for the full reasoning and what a future milestone would need.
Constructor kwargs mirror mo.nsga2's own "algorithm configuration" parameters -- pop_size plus every VARIATION-OPERATOR knob (eta_c, eta_m, p_c, p_m, p_c_bin, p_m_bin, p_c_cat, p_m_cat), the ones that describe the algorithm instance itself, independent of which problem it is pointed at. run()'s own arguments mirror mo.nsga2's remaining "this particular problem/run" parameters (problem, dim, budget, m, k, l, seed, log_dir, label) -- m/k/l describe the PROBLEM being solved (m = objective count for dtlz/wfg; k/l = WFG's own shape parameters), not the algorithm, so they belong with run() rather than init, exactly mirroring mo.nsga2's own per-problem-family validation (a parameter given where it does not apply, or omitted where required, raises ValueError the same way calling mo.nsga2 directly would). Together, calling NSGA2(pop_size=P, op_kwargs).run(problem, dim, budget, m=M, seed=S, ...) is IDENTICAL to calling mo.nsga2(problem, dim, P, budget, m=M, seed=S, op_kwargs, ...) directly (see this task's anchored-equivalence test).
run()'s return value is mo.nsga2's OWN dict shape (individuals, objectives, front0, evals_used, violations when constrained) -- NOT sezgi.algo.SolveResult: that dataclass's fields (best_x/best_f/f_opt/ gap) assume a single-objective run with one best point, which does not fit NSGA-II's multi-objective Pareto-front result.
The algorithm-configuration half of mo.nsga2's parameters (see the class docstring for why the split is here and not at run()). pop_size: population size -- REQUIRED, no default (mo.nsga2 itself requires >= 4 and a multiple of 4; ValueError at run() time otherwise). eta_c/eta_m/p_c: NSGA-II paper's own pinned experimental settings (Deb et al. 2002, Sec. IV.A) as the defaults. p_m: None (default) resolves on the Rust side to 1/n_variables. p_c_bin/ p_m_bin/p_c_cat/p_m_cat: the Binary/Categorical-genotype counterparts, consulted only when the problem's space contains that block kind -- see mo.nsga2's own docstring for each default's provenance.
__doc__
class-attribute
¶
__doc__ = 'Class skin over sezgi.mo.nsga2 (crates/components/src/nsga2.rs:\n 2370-2377, sezgi.mo.nsga2\'s own docstring in\n py-sezgi/python/sezgi/__init__.py for the full parameter contract) --\n ZERO new MO capability: run() delegates to mo.nsga2 VERBATIM, same\n positional/keyword arguments, same return dict. Authoring an NSGA-II\n variant as an engine-hosted Python callback (the way sezgi.Algorithm\n lets a scalar algorithm\'s generate() be authored in Python) is\n explicitly OUT OF SCOPE this milestone: NSGA-II\'s own\n selection/crossover/replacement loop is hard-coded Rust\n (nsga2_run_float_impl/nsga2_run_binary_impl/nsga2_run_mixed_impl), not\n routed through the Registry/AlgorithmSpec/Engine/Generator machinery\n sezgi.Algorithm targets -- see the research doc\'s §B7 for the full\n reasoning and what a future milestone would need.\n\n Constructor kwargs mirror mo.nsga2\'s own "algorithm configuration"\n parameters -- pop_size plus every VARIATION-OPERATOR knob (eta_c,\n eta_m, p_c, p_m, p_c_bin, p_m_bin, p_c_cat, p_m_cat), the ones that\n describe the algorithm instance itself, independent of which problem\n it is pointed at. run()\'s own arguments mirror mo.nsga2\'s remaining\n "this particular problem/run" parameters (problem, dim, budget, m, k,\n l, seed, log_dir, label) -- m/k/l describe the PROBLEM being solved\n (m = objective count for dtlz/wfg; k/l = WFG\'s own shape parameters),\n not the algorithm, so they belong with run() rather than __init__,\n exactly mirroring mo.nsga2\'s own per-problem-family validation (a\n parameter given where it does not apply, or omitted where required,\n raises ValueError the same way calling mo.nsga2 directly would).\n Together, calling NSGA2(pop_size=P, **op_kwargs).run(problem, dim,\n budget, m=M, seed=S, ...) is IDENTICAL to calling\n mo.nsga2(problem, dim, P, budget, m=M, seed=S, **op_kwargs, ...)\n directly (see this task\'s anchored-equivalence test).\n\n run()\'s return value is mo.nsga2\'s OWN dict shape (individuals,\n objectives, front0, evals_used, violations when constrained) -- NOT\n sezgi.algo.SolveResult: that dataclass\'s fields (best_x/best_f/f_opt/\n gap) assume a single-objective run with one best point, which does not\n fit NSGA-II\'s multi-objective Pareto-front result.\n '
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
__module__
class-attribute
¶
__module__ = 'sezgi.builtins'
str(object='') -> str str(bytes_or_buffer[, encoding[, errors]]) -> str
Create a new string object from the given object. If encoding or errors is specified, then the object must expose a data buffer that will be decoded using the given encoding and error handler. Otherwise, returns the result of object.str() (if defined) or repr(object). encoding defaults to sys.getdefaultencoding(). errors defaults to 'strict'.
run ¶
run(problem, dim, budget, m=None, seed=0, k=None, l=None, log_dir=None, label=None)
Delegates to sezgi.mo.nsga2(problem, dim, self.pop_size, budget, ...) verbatim -- see the class docstring for the full parameter mirroring and the return-value shape (mo.nsga2's own dict, not SolveResult).