bio%dZddlmZddlmZmZmZddlZddl m Z ddl m Z ddl mZddlmZdd lmZdd lmZdd lmZdd lmZdd lmZddlmZddlm Z ddl!m"Z"m#Z#m$Z$ddl%m&Z&ddl'm(Z(edgdZ)dZ*GddeZ+y)a+ Copyright, the CVXPY authors Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. ) namedtuple)AnyListTupleN)problems)maximum)minimum)max)min) Inequality) Parameter)Variable)Minimize)Canonicalization) CANON_METHODS)inversesetstighten) InverseData)Solution BisectionData) feas_problemparam tighten_lower tighten_upperc|g}g}|D]0}t|r|j| |j|2||fSN)callableappend) constraintslazy_constraintsreal_constraintscs ]/home/cdr/jupyterlab/.venv/lib/python3.12/site-packages/cvxpy/reductions/dqcp2dcp/dqcp2dcp.py_get_lazy_and_real_constraintsr%5sN ' A;  # #A &  # #A & ' - --cfeZdZdZd d fd ZdZdZdZdZde e e e e ffdZ d Z xZS) Dqcp2DcpaReduce DQCP problems to a parameterized DCP problem. This reduction takes as input a DQCP problem and returns a parameterized DCP problem that can be solved by bisection. Some of the constraints might be lazy, i.e., callables that return a constraint when called. The problem will only be DCP once the lazy constraints are replaced with actual constraints. Problems emitted by this reduction can be solved with the `cp.bisect` function. returncFtt| t|d|_y)N) canon_methodsproblem)superr(__init__r_bisection_data)selfr, __class__s r$r.zDqcp2Dcp.__init__Ls$ h&' ' :#r&c^t|jtk(xr|jS)z>A problem is accepted if it is (a minimization) DQCP. )type objectiveris_dqcp)r0r,s r$acceptszDqcp2Dcp.acceptsQs&G%%&(2Hw7HHr&c|jDcic] }||jvr||j|"}}|jD] }||vsd||<t|j|j|i|j Scc}w)Ng)id_map primal_varsrstatusopt_valattr)r0solution inverse_datavidpvarss r$invertzDqcp2Dcp.invertVs;G;N;N1C8///h**3//11&& !C% c  !)9)95" ' ' 1s%A>cg}|jD]}||j|z }t|\}}tjj t d|}||_|jj}|jr td}n'|jr td}n t}||j||kz }t|\}}tjj t d|} || _t||gtj|| _| t#|fS)zERecursively canonicalize the objective and every constraint. rT)nonneg)nonpos)r _canonicalize_constraintr%rr,Problemr_lazy_constraintsr4expr is_nonnegr is_nonposrr tighten_fnsr/r) r0r,r constrlazyrealrr4t param_problems r$applyzDqcp2Dcp.apply`s+ )) AF 488@ @K A3K@ d''// TB )- &%%**    &A  "&A At44Y!^DD 3K@ d ((00!dC *. '(5 !)>%11)<)> %k'222r&cj|j|\}}|j||d\}}||z }||fS)NFcanonicalize_params) _canon_argscanonicalize_expr)r0rH canon_argsconstrs canon_exprr#s r$_canonicalize_treezDqcp2Dcp._canonicalize_treezsG"..t4 G..tZUZ.[ A1 7""r&cg}g}|jD]p}|j|\}}t|tr?|j rd|j d<n|j rd|j d<||gz }||z }r||fS)zxCanonicalize arguments of an expression. Like Canonicalization.canonicalize_tree, but preserves signs. TrCrD)argsrZ isinstancerrI attributesrJ)r0rHrWrXarg canon_argr#s r$rUzDqcp2Dcp._canon_argss  99 C2237LIq)X.==?59I((2]]_59I((2 9+ %J qLG 7""r&c|jr|j|d\}}|g|zS|jd}|jd}t|trt j |j}t j |j}t j|t j k(s&t j|t jk(rdgSt j|t jk(s't j|t j k(rdgSt|tsJ|jr|jd|kS|jr|j|dkS|jrP|js?|jsJ|t!j"|rt!j ||}|j%d}|j|} |j'|r|j| |kS|j)|sJ|j| |k\St|t*t,fr5|jD cgc]} |j| |kD]} | c} } S|j/|\} } t1j2|j5| |}|| zS|j7sJ|jsJt!j"|rt!j ||}|j%d}|j|} |j'|r|j|| kS|j)|sJ|j|| k\St|t8t:fr5|jD cgc]} |j|| kD]} | c} } S|j/|\} } t1j<|j5| |}|| zScc} } wcc} } w)ahRecursively canonicalize a constraint. The DQCP grammar has expresions of the form INCR* QCVX DCP and DECR* QCCV DCP ie, zero or more real/scalar increasing (or decreasing) atoms, composed with a quasiconvex (or quasiconcave) atom, composed with DCP expressions. The monotone functions are inverted by applying their inverses to both sides of a constraint. The QCVX (QCCV) atom is lowered by replacing it with its sublevel (superlevel) set. The DCP expressions are canonicalized via graph implementations. FrSrT)rO)is_dcpcanonicalize_treer\r]r nparrayvalueallinfanyis_zerorEis_quasiconvex is_convex is_constantr invertible_non_const_idxis_incris_decrrmax_atomrUrsublevelcopyis_quasiconcaver min_atom superlevel)r0rL canon_constr aux_constrlhsrhslhs_valrhs_validxrHr_r#rWaux_args_constr sublevel_setsuperlevel_sets r$rEz!Dqcp2Dcp._canonicalize_constraints( ==?'+'='=fZ_'='` $L* >J. .kk!nkk!n fj )hhsyy)Ghhsyy)Gvvg"&&()RVVGrvv4E-Fv 266)*bffW5G.Hw&*--- ;;=00c: : ;;=00: :     ??$ )c )$!!#&*gooc*3/((*1-xx};;s#88EE{{3'''44TS[AAC'8!45%(XXLc!%!>!>scz!JLLLL/3.>.>s.C+ O#}}SXXj-ASI #o55""$$$     c "&'//#&s+C$$&q)C88C=D{{344SD[AA;;s# ##00= = gx0 1!$H#!::3#:FHHAH H+/*:*:3*? 'J!__SXXj-ASIN!O3 39L,Hs "Q"Qr)r)N)__name__ __module__ __qualname____doc__r.r6rArQrZrrrrUrE __classcell__)r1s@r$r(r(@sF $ I '34# #5cDI)=#>#$Z4r&r(),r collectionsrtypingrrrnumpyrecvxpyrcvxpy.atoms.elementwise.maximumrcvxpy.atoms.elementwise.minimumr cvxpy.atoms.maxr rscvxpy.atoms.minr rwcvxpy.constraintsr %cvxpy.expressions.constants.parameterr cvxpy.expressions.variablercvxpy.problems.objectiver!cvxpy.reductions.canonicalizationr(cvxpy.reductions.dcp2cone.canonicalizersrcvxpy.reductions.dqcp2dcprrrcvxpy.reductions.inverse_datarcvxpy.reductions.solutionrrr%r(r&r$rsl###33++(;/->B<<5.?A .l4l4r&