+ LV-j ^RIHt^RIHt^RIt^RIt^RIHtH t H t H t ^RI H t ^RIHtHt^RIHtHt^RIHt!R R 4t!R R 4t!R R4t!RR4t!RR4tR#))suppress) signatureN)BoundsLinearConstraintNonlinearConstraintOptimizeResult)PreparedConstraint) PRINT_OPTIONSBARRIER)CallbackSuccessget_arrays_tol)exact_1d_arraycPa]tRt^toRtRtRt]R4t]R4t Rt Vt R#)ObjectiveFunctionz! Real-valued objective function. c V'dHVe\V4'gQh\V\4'gQh\V\4'gQhWnW nW@n^VnR#)a Initialize the objective function. Parameters ---------- fun : {callable, None} Function to evaluate, or None. ``fun(x, *args) -> float`` where ``x`` is an array with shape (n,) and `args` is a tuple. verbose : bool Whether to print the function evaluations. debug : bool Whether to make debugging tests during the execution. *args : tuple Additional arguments to be passed to the function. N)callable isinstancebool_fun_verbose_args_n_eval)selffunverbosedebugargss&&&&*j/Volumes/fast/ai/experiments/ui-tars-smoke/.venv/lib/python3.14/site-packages/scipy/_lib/cobyqa/problem.py__init__ObjectiveFunction.__init__sV& ;(3-- //gt,, ,,eT** **    c  \P!V\R7pVPfRpV#\\P!VP!V.VP O5!44pV;P ^, unVP'dK\P!R/\B;_uu_4\VP RV RV 24RRR4V#V# +'giT#;i)z Evaluate the objective function. Parameters ---------- x : array_like, shape (n,) Point at which the objective function is evaluated. Returns ------- float Function value at `x`. dtypeN() = ) nparrayfloatrsqueezerrr printoptionsr printname)rxfs&& r__call__ObjectiveFunction.__call__6s HHQe $ 99 A bjj1!:tzz!:;????*eT** **XXq!f% XXa[ XXq!f% XXa[ %J&& - z}}=>Kvvk""YY :<< 3L'MN ^^ &MM+6(mmK89  66+&&YY " k\2#{l33  ^^ " {l3# |44 1&B*- "((499%&), "((499%&88DII&$)))<<88DII&YYy!|, YYy) YYy!|, YYy) 7B 7B!cchh - q"''!* -{  s 6P$Pc VP#)z Left-hand side matrix of the linear inequality constraints. Returns ------- `numpy.ndarray`, shape (m, n) Left-hand side matrix of the linear inequality constraints. )rvr6s&rrLinearConstraints.a_ubzzr!c VP#)z Right-hand side vector of the linear inequality constraints. Returns ------- `numpy.ndarray`, shape (m, n) Right-hand side vector of the linear inequality constraints. )rwr6s&rrLinearConstraints.b_ub#rr!c VP#)z Left-hand side matrix of the linear equality constraints. Returns ------- `numpy.ndarray`, shape (m, n) Left-hand side matrix of the linear equality constraints. )rxr6s&rr}LinearConstraints.a_eq/rr!c VP#)z Right-hand side vector of the linear equality constraints. Returns ------- `numpy.ndarray`, shape (m, n) Right-hand side vector of the linear equality constraints. )ryr6s&rrLinearConstraints.b_eq;rr!c .VPP#zl Number of linear inequality constraints. Returns ------- int Number of linear inequality constraints. )rrYr6s&rm_ubLinearConstraints.m_ubGyy~~r!c .VPP#zh Number of linear equality constraints. Returns ------- int Number of linear equality constraints. )rrYr6s&rm_eqLinearConstraints.m_eqSrr!c P\P!VPV4RR7#)rcr%initialr)maxrerfs&&rrgLinearConstraints.maxcv_svvdnnQ'55r!c\VP4'd?\P!VPUu.uFq"P V4NK up4#\P !.4#uupiN)lenrZr)rrer*)rr0pcs&& rreLinearConstraints.violationosK txx==>>TXX"FXr<<?X"FG Gxx|#GsA0)rxrvryrwrZN)r=r@rArBrCrrDrrr}rrrrgrerErFrGs@rrprpsB H            6 r!rpcta]tRtRtoRtRtRt]R4t]R4t ]R4t R R lt R R lt R t VtR#) NonlinearConstraintsiuzA Nonlinear constraints ``c_ub(x) <= 0`` and ``c_eq(x) == b_eq``. c FV'dj\V\4'gQhVFp\V\4'dKQh \V\4'gQh\V\4'gQhWn.VnW nRVnRVnR;Vn Vn R#)z Initialize the nonlinear constraints. Parameters ---------- constraints : list Nonlinear constraints. verbose : bool Whether to print the function evaluations. debug : bool Whether to make debugging tests during the execution. N) rrtrr _constraintsrZr_map_ub_map_eq_m_ub_m_eq)rrrrrs&&&& rrNonlinearConstraints.__init__zs k400 00) !*.ABBBB*gt,, ,,eT** **'   "&& TZr!c  \VP4'g;^;VnVn\P !.4\P !.43#\P !V\ R7p\VP4'Eg.Vn.Vn ^Vn^VnVPEFp\VP4'g3\P!V4pRVn RVn \W14pM \W!4pRVPnVPP#V4\P$!VPP&4pVP(^,VP(^,rv\+Wg4p\P,!Wv, 4V8*p VPP#WY,4VPP#WY(,4V;P\P.!V 4, unV;P\P.!V (4, unEK .p .p \1VP4EFwrVPPV4p VP2'd\P4!R /\6B;_uu_4\9\:4;_uu_4VPV ,PP<p\?V RV RV 24RRR4RRR4VPV ,pVPV ,pV V,p\V4'dVP(^,V,pVP(^,V,pV\P@)8pVV,VV,, pV P#V4V\P@8pVV,VV,, pV P#V4W,p\V4'dJRVP(^,V,VP(^,V,,,pVV,pV P#V4EK VP'd\PB!V 4p M\P !.4p VP'd\PB!V 4p M\P !.4p V PDVnV PDVnW3# +'giEL;i +'giEL+;i) a_ Calculates the residual (slack) for the constraints. Parameters ---------- x : array_like, shape (n,) Point at which the constraints are evaluated. Returns ------- `numpy.ndarray`, shape (m_nonlinear_ub,) Nonlinear inequality constraint slack values. `numpy.ndarray`, shape (m_nonlinear_eq,) Nonlinear equality constraint slack values. r#cV#rr()x0s&r/NonlinearConstraints.__call__..srr!cR#r%r()rvs&&rrrs3r!Tr&r'Nrrr()#rrrrr)r*r+rZrrrjaccopyhessr r f_updatedappendarangerWr[r rzrV enumeraterr-r rr>r=r.rPrrY)rr0rrridxrLrNarr_tolrc_ubc_eqivalfun_nameeq_idxub_idxub_valrQrS finite_xl_v finite_xueq_valmidpoints&& rr2NonlinearConstraints.__call__s 4$$%%&' 'DJ88B<"- - HHQe $488}}DLDLDJDJ"//  // *-A)AE.AF+A1B+J:B$( #ii)1ryy|B(0 ffRWo8  ##C$45 ##C $56 b..{;;  b.. |<< 70:txx(EA&&**Q-C}}}__5}55!.11#'#4#4Q#7#;#;#D#D !A3d3%8926 \\!_F\\!_F[F6{{YYq\&)YYq\&)"&&L  ]VI%66 BK I&I6 B[F6{{"))A,v"61f9M"MN(" KK A)D :::>>$'D88B>$'D88Bs'(vv NN1wN ?  r!c\P!VPUu.uFqDPV4NK up4#uupir)r)rrZre)rr0rrrs&&&& rreNonlinearConstraints.violationVs-~~B2||ABCCBs?)rrrrrrrZNN)r=r@rArBrCrr2rDrrr7rgrerErFrGs@rrrusc':jX**   0DDr!rcTa]tRtRtoRtRtRRlt]R4t]R4t ]R4t ]R4t ]R 4t ]R 4t ]R 4t]R 4t]R 4t]R4t]R4t]R4t]R4t]R4t]R4t]R4tRtRRltRRltRtRtVtR#)ProblemiZz Optimization problem. c  V 'Ed \V\4'gQh\V\4'gQh\V\4'gQh\V\4'gQh\V\ 4'gQh\V\ 4'gQh\V \ 4'gQh\V \4'gQhV 'd V ^8gQh\V \4'gQhV ^8gQh\V \ 4'gQhWnW@n WPn Ve\V4'g \R4hW`n \VR4pVPp VP PV 8wd\#RV R24hVP$P&^,V 8wd\#RV R24h\)VP VP*4pVP VP*8*\,P.!VP VP*, 4V8,VnRVP VP0,VP*VP0,,,Vn\,P4!VP2VP VP0,VP*VP0,4VnW0n\\9VP VP0(,VP*VP0(,44VnVP:P=W P0(,4VnVP@VPBR VP03,VP2,, p\\EVP$R VP0(3,\,PF)VPHVP$R VP03,VP2,, 4\EVPBR VP0(3,W4.VPJV 4Vn T;'dVP:PL;'d\,PN!\,PP!VP:P 44;'d?\,PN!\,PP!VP:P*44pV'Ed=RVP:P*VP:P , ,Vn)RVP:P*VP:P ,,Vn*\\9\,PV!VPJ4)\,PV!VPJ444VnVPP@VPPBVPT,, p\\EVPP$\,PX!VPR4,\,PF)VPPHVPP$VPT,, 4\EVPPB\,PX!VPR4,VV4.VPJV 4Vn VP>VPT, VPR, VnMJ\,PV!VPJ4Vn)\,PZ!VPJ4Vn*Wpn.Wn/.Vn0.Vn1.Vn2Wn3Wn4.Vn5.Vn6.Vn7R#) ai Initialize the nonlinear problem. The problem is preprocessed to remove all the variables that are fixed by the bound constraints. Parameters ---------- obj : ObjectiveFunction Objective function. x0 : array_like, shape (n,) Initial guess. bounds : BoundConstraints Bound constraints. linear : LinearConstraints Linear constraints. nonlinear : NonlinearConstraints Nonlinear constraints. callback : {callable, None} Callback function. feasibility_tol : float Tolerance on the constraint violation. scale : bool Whether to scale the problem according to the bounds. store_history : bool Whether to store the function evaluations. history_size : int Maximum number of function evaluations to store. filter_size : int Maximum number of points in the filter. debug : bool Whether to make debugging tests during the execution. Nz)The callback must be a callable function.z#The initial guess must be a vector.zThe bounds must have z elements.z@The left-hand side matrices of the linear constraints must have z columns.rrrs)8rrrJrprr+rint_obj_linear _nonlinearr TypeError _callbackrrYrQrrshaper rSr)rz _fixed_idx _fixed_valrl _orig_boundsr_boundsrm_x0rr}rrPrrrUrTisfinite_scaling_factor_scaling_shiftrXdiagzeros_feasibility_tol _filter_size _fun_filter _maxcv_filter _x_filter_store_history _history_size _fun_history_maxcv_history _x_history)robjrr[linear nonlinearcallbackfeasibility_tolscale store_history history_size filter_sizerrtolrs&&&&&&&&&&&&& rrProblem.__init___s` 5c#455 55f&677 77f&788 88i)=>> >>ou55 55eT** **mT22 22lC00 00#a'''k3// //? "?eT** **  #  H%% KLL!B E F GG 99>>Q 4QCzBC C ;;  Q 1 $s)%  VYY 2!99 1 FF699vyy( )C /  IIdoo &4??)C C '' OO IIdoo & IIdoo & #' 699doo-. 4??:J0K L  <<''OO+;(<={{V[[DOO);VV3Rl\%VP VP&44F 'dK RM3 RM/!VV3Rl\%VP VP&444pV'EdmVP PS4VP&PS4VP(PV4\+\VP 4^, R R 4EFzp\P!S4'd)\P!VP V,4pM\P!S4'd)\P!VP&V,4pM\P!VP V,4;'gd\P!VP&V,4;'g5SVP V,8*;'dSVP&V,8*pV'gEK)VP PV4VP&PV4VP(PV4EK} \VP 4VP,8dRVP P^4VP&P^4VP(P^4VP.e\1VP.4p VP3V4wrp VPV 4p \5V P64R08Xd!\9V V R 7p VP/V R 7MVP/V 4\P!S4'd\>o\>V\P!V4&\>V\P!V4&\A\CS\>4\>)4o\PD!\PF!V\>4\>)4p\PD!\PF!V\>4\>)4pSWE3# \:d p\<ThRp?ii;i) at Evaluate the objective and nonlinear constraint functions. Parameters ---------- x : array_like, shape (n,) Point at which the functions are evaluated. penalty : float, optional Penalty parameter used to select the point in the filter to forward to the callback function. Returns ------- float Objective function value. `numpy.ndarray`, shape (m_nonlinear_ub,) Nonlinear inequality constraint function values. `numpy.ndarray`, shape (m_nonlinear_eq,) Nonlinear equality constraint function values. Raises ------ `cobyqa.utils.CallbackSuccess` If the callback function raises a ``StopIteration``. r#c3<"TFHwr\P!V4;'dSV8;'g\P!V4xKJ R#5irr)rR).0 fun_filter maxcv_filter maxcv_vals& r #Problem.__call__..7sU 1,J$-- ,**88L)*1#A AAFTc3<"TFHwr\P!V4;'dSV8;'g\P!V4xKJ R#5irr )r rrfun_vals& rrrAsU 1,J&))j(((88J'(1rc3L<"TFwrSV8;'gSV8xK R#5irr()r rrrrs& rrrKs4 1,J*$@@ L(@@1s$$Nintermediate_result)r0r)r)$r)rdr+build_xrrrgrrrrrrrpoprRrrTziprrrangerrr best_evalset parametersr StopIterationr r rminmaximumminimum)rr0penaltyx_fullrr include_pointk remove_pointsigx_bestfun_best_rexcrrs&&& @@rr2Problem.__call__ sN6 JJq &a))F#??62JJq73        $ $W -    & &y 1 OO " "1 %4$$%(:(::!!%%a(##''*##A& 88G  )!4!4 0 01Q6M XXg  C 14$$&&1  CCC 14$$&&1  MXXi C 14$$&&1  CCC 14$$&&1  M C 03$$&&1 CCC 03$$&&1 M =    # #G ,    % %i 0 NN ! !! $ 3t//014b"=88G$$#%88D,<,,>q,A#BL!1!1!!45??88D$6$6q$9:??"d&6&6q&99??%););A)>> !  <$$((+&&**1-NN&&q)>$4##$t'8'88  $$Q'""&&q)""1% >> %DNN+C /&*nnW&=#!f-s~~&+@*AA*8 $+' NN7JNKNN6* 88G  G%,!"%,!"c'7+gX6**RZZ9G8D**RZZ9G8D((! /%3. /sA\#(\## \:.\55\:c .VPP#)zD Number of variables. Returns ------- int Number of variables. )rrYr6s&rr Problem.nsww||r!c .VPP#)z Number of variables in the original problem (with fixed variables). Returns ------- int Number of variables in the original problem (with fixed variables). )rrYr6s&rn_origProblem.n_origs###r!c VP#)zP Initial guess. Returns ------- `numpy.ndarray`, shape (n,) Initial guess. )rr6s&rr Problem.x0r_r!c .VPP#r5)rr7r6s&rr7Problem.n_evalsyyr!c .VPP#)r;)rr/r6s&rrProblem.fun_namerr!c VP#)zM Bound constraints. Returns ------- BoundConstraints Bound constraints. )rr6s&rr[Problem.boundsr9r!c VP#)zP Linear constraints. Returns ------- LinearConstraints Linear constraints. )rr6s&rrProblem.linearr9r!c .VPP#)zT Number of bound constraints. Returns ------- int Number of bound constraints. )r[rWr6s&rm_boundsProblem.m_boundss{{}}r!c .VPP#r)rrr6s&r m_linear_ubProblem.m_linear_ub{{r!c .VPP#r)rrr6s&r m_linear_eqProblem.m_linear_eqrDr!c .VPP#)z Number of nonlinear inequality constraints. Returns ------- int Number of nonlinear inequality constraints. Raises ------ ValueError If the number of nonlinear inequality constraints is not known. )rrr6s&rm_nonlinear_ubProblem.m_nonlinear_ub###r!c .VPP#)z Number of nonlinear equality constraints. Returns ------- int Number of nonlinear equality constraints. Raises ------ ValueError If the number of nonlinear equality constraints is not known. )rrr6s&rm_nonlinear_eqProblem.m_nonlinear_eqrKr!c N\P!VP\R7#)z History of objective function evaluations. Returns ------- `numpy.ndarray`, shape (n_eval,) History of objective function evaluations. r#)r)r*rr+r6s&r fun_historyProblem.fun_history(sxx))77r!c N\P!VP\R7#)z History of maximum constraint violations. Returns ------- `numpy.ndarray`, shape (n_eval,) History of maximum constraint violations. r#)r)r*rr+r6s&r maxcv_historyProblem.maxcv_history4sxx++599r!c VP^8gVP^8dR#VP^8gVP^8dR#VP^8dR#R# \ dR#i;i)z Type of the problem. The problem can be either 'unconstrained', 'bound-constrained', 'linearly constrained', or 'nonlinearly constrained'. Returns ------- str Type of the problem. znonlinearly constrainedzlinearly constrainedzbound-constrained unconstrained)rIrMrBrFr?rr6s&rtype Problem.type@si -""Q&$*=*=*A0!!A%)9)9A)=-"*& - -  -s"A"A A A.-A.c VPR8H#)zw Whether the problem is a feasibility problem. Returns ------- bool Whether the problem is a feasibility problem. r<)rr6s&ris_feasibilityProblem.is_feasibility^s}}""r!c \P!VP4pVPW P&WP ,VP ,W P(&VPPV4#)z Build the full vector of variables from the reduced vector. Parameters ---------- x : array_like, shape (n,) Reduced vector of variables. Returns ------- `numpy.ndarray`, shape (n_orig,) Full vector of variables. ) r)rur2rrrrrrm)rr0r%s&& rrProblem.build_xjsc$++&"&//$%(<(<$<&*&9&9%:   ((00r!Nc VPWVR7p\P!V4'd\P!VRR7#R#r)rer)rVr)rr0rrres&&&& rrg Problem.maxcv~s;(NN1wNG  I & &66)S1 1r!c..pVPP'g-VPPV4pVPV4\ VP P 4'd-VP PV4pVPV4\ VPP 4'd.VPPWV4pVPV4\ V4'd\P!V4#R#r) r[rUrerrrrZrr)r)rr0rrreblcnlcs&&&& rreProblem.violations {{&&& %%a(A   Q  t{{  &&q)B   R t"" # #//++A@C   S ! y>>>>), , r!c \VP4^8XdV!VP4\P!VP4p\P!VP 4p\P!VP 4p\P!V4p\P!V4'EdW0P8*p\P!V4'd\P!\P!W&,44'gVV\P!W&,48*,p\P!V4^8d&Ws\P!W7,48*,p\P!V4R,pEM\P!V4'd \P!V4R,pEM\P !V\P"4p W%,WV,,,W&\P!\P!V 44'd9V\P!V48*p \P!V 4R,pEM0V \P!V 48*p \P!V 4^8d&W\P!W;,48*,p \P!V 4^8d&W\P!W+,48*,p \P!V 4R,pMy\P!\P!V44'g8V\P!V48*p\P!V4R,pM\V4^, pVP$P'WHR3,4W(,W8,3#)aV Return the best point in the filter and the corresponding objective and nonlinear constraint function evaluations. Parameters ---------- penalty : float Penalty parameter Returns ------- `numpy.ndarray`, shape (n,) Best point. float Corresponding objective function value. float Corresponding maximum constraint violation. rsr)rrrr)r*rrrr{rrTrRnanminrVr! flatnonzero full_likenanr[rm) rr$rrx_filter finite_idx feasible_idx fun_min_idxr merit_filter min_maxcv_idx merit_min_idxs && rrProblem.best_evals, t A % MXXd../ xx 2 23 88DNN+[[. 66*  '+@+@@Lvvl##BFF12--+"))J,D"EE ##K014266$14$KNN;/3 %%NN<04 "||J? *WJ7O-OO(66"((<011%1BIIl4K$KM}5b9A%1BIIl4K$KM'' 6:%(7:* '' 6:%rvv&58* }5b9A,-- % *(==K{+B/AJ!#A KK  A / M O  r!)rrrrrrrrrrrrrrrrrrrrrrr)r=r@rArBrCrr2rDrr2rr7rr[rr?rBrFrIrMrPrSrWrZrrgrerrErFrGs@rrrZssiVB)H   $ $                      $$ $$  8 8 : :--: # #1(4- h h r!r) contextlibrinspectrrnumpyr)scipy.optimizerrrrscipy.optimize._constraintsr settingsr r utilsr r rrrJrprrr(r!rryso  ;-2!WWt]G]G@ddNbDbDJv  v  r!