ELF>@ @8 @  !! AT  $$Std Ptdd d d QtdRtd P P GNUGNUeDp1 4o# nJ;% B `"p @H 0 ,@@D:`@nAHdC1(0b8Kk}LJ f|8\5M3ndy'IK|U=vgl #2ח*X3I#K霳N yME>=g!ni焽dMD.ۗEڋFy$vWJ@IoS:rKff,_̩*1޺KFmK4bs\rb(zK|M&E`61 0nc} 8E &L # ^U[1/ q =8 *   w|E]/ c  p("} |p N X  Z |6+ OW = )pG T GfMUq,  w: F"joC   ` |  0 ]  p -   W   ]R  P  W    -  rc       0 /  P    ] ҅7  :Y  (*  P   /  p W  `   0 # 0K  p (  p I  =    0 (T  `   x  +q  A   R   [  ` *>  p U  P  ` r  1  P t  3i      CX  `   +k  Q a    *4  }k   ?  R     0 |  R  ` __gmon_start___ITM_deregisterTMCloneTable_ITM_registerTMCloneTable__cxa_finalizePyCapsule_GetPointerPyNumber_Subtract_Py_NoneStruct_PyDict_GetItem_KnownHashPyMethod_TypePyFunction_TypePyCFunction_TypePyTuple_NewPyThreadState_Get_Py_CheckRecursionLimitPyEval_SaveThreadPyEval_RestoreThreadPyDict_SizePyExc_TypeErrorPyErr_FormatPyLong_AsUnsignedLongPyErr_OccurredPyExc_OverflowErrorPyErr_SetStringPyLong_FromLongPyLong_TypePyExc_NameErrorPyObject_GetAttrPyObject_Call_Py_CheckRecursiveCallPyExc_SystemError__stack_chk_failPyLong_AsLongPyLong_FromUnsignedLong_PyType_LookupPyFloat_FromDoublePyExc_AttributeErrorPyErr_SetObjectPyDict_NewPyDict_SetItemPyObject_RichCompare_Py_TrueStruct_Py_FalseStructPyObject_IsTrue_PyThreadState_UncheckedGetPyTuple_TypePyNumber_AddPySequence_TuplePyMem_MallocPyCapsule_NewPyNumber_LongPyNumber_OrPyTuple_PackPyBaseObject_TypePyList_TypePyObject_SizePyFloat_TypePyFloat_AsDoublePyLong_FromSsize_tPyObject_GetItemPyObject_GetIterPyExc_ValueErrorPySequence_ContainsPyUnicode_FormatPyObject_IsInstancePyNumber_InPlaceTrueDividePyObject_SetItemPySlice_NewPyNumber_InPlaceAddPyObject_SetAttrPyBytes_FromStringAndSizePyList_NewPySequence_ListPyNumber_MultiplyPyList_AsTuplePyList_AppendPyGILState_EnsurePyExc_ZeroDivisionErrorPyGILState_ReleasePyObject_GC_UnTrackPyErr_FetchPyMem_FreePyErr_RestorePyObject_CallFinalizerFromDeallocPyInit_mtrandPyOS_snprintfPy_GetVersionPyErr_WarnExPyUnicode_FromStringAndSizePyModule_Create2PyModule_GetDictPyImport_AddModulePyObject_SetAttrStringPyUnicode_InternFromStringPyObject_HashPyUnicode_DecodePyLong_FromString__pyx_module_is_main_mtrandPyExc_ImportErrorPyImport_GetModuleDictPyDict_GetItemStringPyDict_SetItemStringPyCode_NewPyType_ReadyPyCFunction_NewExPyImport_ImportModulePyObject_GetAttrStringPyCapsule_TypePyExc_RuntimeErrorPyExc_ExceptionPyErr_Print_PyDict_NewPresizedPyType_Modifiedmemcpy_Py_EllipsisObjectPyFrame_NewPyEval_EvalFrameExlogPyErr_NormalizeExceptionPyException_SetTracebackPyUnicode_Typememcmp_PyUnicode_ReadyPyExc_StopIterationPyErr_GivenExceptionMatchesPyErr_ExceptionMatchesPyTraceBack_Here_PyObject_GetDictPtrPyObject_NotPyUnicode_FromStringPyUnicode_FromFormatPyMem_ReallocPyErr_ClearPyExc_DeprecationWarningPyErr_WarnFormatrk_seedrk_randomrk_doublerk_gausssqrtrk_normalrk_lognormalexprk_standard_cauchyrk_standard_exponentialrk_exponentialrk_paretork_weibullpowrk_powerrk_uniformrk_standard_gammark_gammark_chisquarerk_frk_standard_trk_betark_vonmisesacosfmodrk_laplacerk_gumbelrk_logisticrk_rayleighrk_waldrk_zipfrk_triangularrk_logseriesrk_ulongrk_random_uint64rk_random_uint32rk_random_uint16rk_random_uint8memsetrk_random_boolrk_longrk_intervalrk_fillrk_devfillfopen64freadfcloserk_randomseedgettimeofdaygetpidclockrk_altfillinit_by_arrayrk_binomial_btperk_binomial_inversionrk_binomialrk_poisson_multrk_poisson_ptrsrk_poissonrk_noncentral_chisquarerk_noncentral_frk_negative_binomialrk_geometric_searchrk_geometric_inversionrk_geometricrk_hypergeometric_hyprk_hypergeometric_hruark_hypergeometricPyEval_EvalCodeExPyImport_ImportModuleLevelObjectPyImport_ImportPyExc_IndexErrorPyNumber_IndexPyLong_AsSsize_tPyDict_NextPyUnicode_AsUnicodePyUnicode_Comparerk_strerrorlibpython3.7m.so.1.0libc.so.6GLIBC_2.14GLIBC_2.4GLIBC_2.2.5/opt/alt/python37/lib64D N ii Y ui c L PL         @ H P X             @ H P X             0 @ P X 0           ( 0 @ H P `                0 8 ` 8h   8 x  8    x  ( 0 @ H ` h               ( 0 @ 0H P ` h xp x     0   H @ 8   0 8 P X p x  8  0 X       @Y $v ȵ7صM Z P ; ( @ HYX`{` hxv [l `kІ ȶ@ض` PU6 @Q N (8`J@T HX@?`~ h4x;w  `.  ) ȷ 7ط" k   Q`  (Pn8 @8 HМX` ` hx@ Њ `  `  ȸ?ظ`  p  B ϋ (@Z8 @ H0nX ` hx   8 v R ȹpعj n PZ  0Q  (098G @ H@_X > `݌ hx 0  @$ x   Ⱥ<غ  P  0o  (r8 @5 HX `I h 5 x w @M  E ~  g Ȼk ػ  ^ (0] PX] x] ] ȼм] ]  ] @H`] hp0] ] \ H\ p\ 08P\ X `0\ \ `[ оؾ[ p[ P(@[ H@P[ pxZ Z ȿ@Z Z Y 8@Y `PhPY 0Y PY X X (0X PXpX xPX p X W 0W ` W @H@W h@pW 0V `V @V  V 08U X`U U  `U @U  U (U HPT pxT T (`T T S 8@S `hS S S 8S S (0pS PXPS x S R R R  R @`HR hpR 8R R X R R 0 8xR X`pR dR x_R ]R xVR (QR HHPHR px`D ?D :D 0D +D 8 @%D `hB B B @`B @B (0`: P( XS: x/  / / /  / @H/ hPp`/ 0 /   / / 0/ 0 8. X8`. H. . `. . h(. HXP. px. x% % %  % 8@% `p h% x% `% @y% p% (0`% PX^% x@ h8 h0 H  0  @H h(p@ (  ` X 08 Xx` h    x (@ HP1 p` x+ x    x 8p@ `hh    X (0 PPX x    @   @H hp  ` X x 08 X`  H   0( HpP pXx` hP H   8@б `ẖ   @  P  (0 PXx xh H@ X x   @8 H hp   @  08 X0`   ( @ ( HXP px @  H x 8p@` `xhQ PH 0 } | (0q P Xq xX`d Hd 8@d x0d   'd @(Hd hpc c (c c c 08c X`c hc c c `c (xc HP`c pxKc `T HT XP 8P 8@H `hG G ; @;  ; (0; PXX; x0; : : 0 : 0  : @H6 h p6  6  6 ( 6  6 0 8p6 X ``6  P6 H6 06  -  (- H P, px% P%  % x ` 8@ `h  ` 8` (P (0H PX@ x0      @H hp      08 X`  X  p H(p HP Ph phx     0 ( 8@ `h    p  (0 PX xH  8 H h  @HX hp    x hh 0(8P X`F A H8  (  ( HPP px` C X< P5 H. 8`@( ` h     (0 PXX x( 8 @ ذ   @Hޣ h0p٣ ` s | p pj j @k .   Ā  `w  w ؀ (8@ HX@` hxe bS L% 0_ 8_     ( 0 8  @ H  P  X  ` h p x      '  . / Ⱦ о 0ؾ 1      I K L Q( 0 8 W@ ^H _P X ` gh hp jx m p q  z }  ~  ȿ п ؿ      ( 0 8 @ H P X  ` h p x          ȹ й ع          ( 0 8  @ !H "P X #` $h %p &x (  ) * + ,  -  2Ⱥ 3к 4غ   5 6 7 8 9 :  ( 0 ;8 <@ =H >P ?X @` Ah Bp x C D E F G H   J Ȼ Mл Nػ O P R  S  T U V X( Y0 Z8 @ [H P \X ` ]h `p ax b c d  e f  i k lȼ nм oؼ  r s t u v w x y {( |0 8 @ H P X ` h p x          Ƚ н ؽ     HH  HtH5 % hhhhhhhhqhah Qh Ah 1h !h hhhhhhhhhhqhahQhAh1h!hhhh h!h"h#h$h%h&h'qh(ah)Qh*Ah+1h,!h-h.h/h0h1h2h3h4h5h6h7qh8ah9Qh:Ah;1h<!h=h>h?h@hAhBhChDhEhFhGqhHahIQhJAhK1hL!hMhNhOhPhQhRhShThUhVhWqhXahYQhZAh[1h\!h]h^h_h`hahbhchdhehfhgqhhahiQhjAhk1hl!hmhnhohphqhrhshthuhvhwqhxahyQhzAh{1h|!h}h~hhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D%} D%u D%m D%e D%] D%U D%M D%E D%= D%5 D%- D%% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D%} D%u D%m D%e D%] D%U D%M D%E D%= D%5 D%- D%% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D%} D%u D%m D%e D%] D%U D%M D%E D%= D%5 D%- D%% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D%} D%u D%m D%e D%] D%U D%M D%E D%= D%5 D%- D%% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D% D%} D%u D%m D%e D%] D%U D%M D%E D%= D%5 D%- DAWAVAUATUSHHH->dH%(H$81HD$0HD$(HD$ HD$HD$HD$Ht HE2Ll$hA1H LH\$l-H HH1D$l8D$hu D$n8D$jthLd$p1IH LLe Lh1Ly-H E1E13H331qH4IHu,Hӻ E113H33eL-\ 1LEH=IHu,H E11L3HI3731L0IHu,HI E113H 32H=1H<IHu,H E112H22&HHa2IHu,H E112H|2j2(LHH= H2IHu,Hp E1152H22 2*H=v H;IHu,H) E111H11+H1H=n;H50 H yiHع E1E11H11/iC C!H{t[{"t4IEHH8H<H.H(L+Mu0HCHSHpHt 1%HCHpWH4H H5Ho H:H1H1HH3HmwHxAHS]H<H9CHt<H )H@HHAHH@HpH3HH@HHX@H11H= HX@Hc11H= H0HC11H= H?H#11H= H?H11H= wH?HHH7HHǀH?HHH?HHHs?Hv11H=; HC?HVH 8sH?H=7H5 `QHi E11..H+..6H|$0Ht HuHGP0H|$Ht HuHGP0H|$Ht HuHGP0HtHMu HEHP0MtI $u ID$LP0MtIMu IELP0H=77tEH=M-tH -z-1H= H= 7HuHGP0H6 kHuH H5 H8PH-6}HE E1E1 -H-,1HHu-H E1E1,H,,:L5 HL+IHuBH<6LHy,H E11s,Hp,^,<@H=b6'}HV=HH=>6 }H,HH=6|HH=5|HI-Ht}H=5|H0HteH=5|H!0HtMH=5|H!6Ht5H=}5z|Ht$H571PH7Hu1UH E11e+Hb+P+A2H- 1HHHH+HHHH1H+HHHH1Hn+HHHH1HA+HHHH1cH+HiH561@H*HFHHH1H*H!HHH1H*HHHH1HZ*HH5J61H*HHHH1H)HHHH1dH)HjHHH1?H)HEH551H*H"HHH1H*HHHH1H{,HHHH1HN,HHHH1H/HHHH1cH/HiH541@H1HFHHH1H1H!HHH1H/HHHH1H.HH5J41HW1HHHH1H*1HHHH1dH/HjHHH1?H/HEH531H0H"HHH1H0HHHH1H(HHHH1Hf(HH-F 1HHHH2(HH521^H(HdHHH19H'H?HHH1Hu'HH521H:'HHHH1H 'HH5=21H&HH521H&HHHH1aHR&HgHHH1H_%HDH5-1Ht&H!H5,1HI&HH5,1H&HH5,1H%HH5{,1H%HH5p,1lH%HrH55,1IHr%HOH5*,1&HG%H,H5+1H %H H5+1H$HH5+1H$HH5v+1H$HH5K+1wHP%H}H50+1TH%%HZH5*11H$H7H5*1H$HH5*1H$HH5*1Hq$HH5i*1H>$HH5F*1H$HH5#*1_H#HeH5*1 H5a#1H6H H5&#1HH H5"1HH H5"1HH H5"1HrH H5"1fHGHl H5_"1CHHI H54"1 HH& H5 "1HH H5!1HH H5!1HpH H5!1H%H H5m!1qHHw H5B!1NHHT H5!1+HH1 H5 !1HyH H5 1HNH H5 1H#H H5 1HH H5` 1|HH H55 1YHH_ H5 16H_H< H51H,H H51H9HH51HHH5n1HHH-C HHHHHH51bHkHhHH518HH>H51HHHHH'HHH1HHH5]1HrHHHH1H=HHHH1oHHuHHH1JH+%HPHHH1%H$H+HHH1H$HASL G 15jLH 5NH59H55555H@HH11L IARAAQj5#5RRPRR JHPHSAPL  15LHH I5HT5H5O55k5]5OH@HH>L WA11AQj85E#5RRPRR HPHVL  15LH 5H5H55h5551H@H:HL QA11AQjk5"5NRRPRR HPHPL 21 5ULH 59H5$H555 55H@HH11L 4AWAAQh5!5RRPRR 2HPH;AVL | 15L0H 15H<5nH5755S5E57H@HH&11L ~ATAAQh5!5RRPRR |HPHUL  15L{H |5H5H55L555H@HHq11L SAAQh5U 5/RRPRR HPHASL  155LH 5H5H55555`H@HiH11L ARAAQh755yRRPRR HPHAQL$ 15L HH 5cH5NH55535%5H@HH11L ^APAAQhj55RRPRR \HPHeWL 1 5L[ H \ 5Hg 5H5b 5,5~5p5bH@HHQL VA11AQh55RRPRR HPHH  H-1H5 zHH L ;QA11AQh655RRPRR1HPHt>H51HHtH51HHu,HK E11 H  CL%ɢ `H5G LQ]HHu-H E1E1 WH  dH5l L]Hs HHu-H E1E1p YHm [ e=H H5* L\HHu-Hn E1E1' [H$  f0H5 Lo\HHu-H% E1E1 aH  gH-HH/HHHqy,H E11 QH n jPH=11HL%lHHtBH5HLuHExHHEuKHEHP0?HHEu HEHP0H& E11 QH  oH=rHH5r y,H E11 QH  pr`H5 H=џ HZHHu-H E1E1Q  HN < H1H=jHD$0HHu-HN E1E1 HXH5IH=Hy,H E11HZHMu HEHP0H1H=:HD$0HHu-H E1E1]8HZHd*H5H=Huy,Hb E118HfHMu HEHP0HV1H=u HD$0HHu-H E1E1kHpH5H=XHy,H E11rkHo]r?HMu HEHP0H1H={HD$0HHu-H_ E1E1H|H5BH=H+y,H E11H~HMu HEHP0H 1H=HD$0HHu-H E1E1nHkY;H5H=Hy,Hs E11(H%HMu HEHP0Hg1H=&1HD$0HHu-H E1E1HH5H=iHy,HΌ E11HnPHMu HEHP0H1H=aHD$0HHu-Hp E1E1$7H!H5;H=HL%s %oH05 WHuHGP0HD$8H|$@Ht HuHGP0HD$@H|$HH@H7HGP0+H L5)HL$`HT$XLHt$PryHykr\H5 H=f1跾HHuHH:,3H HMu HEHP0H HL$HHT$@LHt$88H|$PHt HuHGP0H|$XHt HuHGP0H|$`H PHPHGP0OH=1 HD$0HHu-H E1E1c H` N ͪ0H5* H= H{y,HX E11 H  ϪHMu HEHP0H=L1M HD$0HHu-H E1E1 H  ٪H5H=e Hy,H E11 H| j ۪LHMu HEHP0H=1 HD$0HHu-Hc E1E1' H$  H5~H= H?y,H E11 H  HMu HEHP0HD$0葿HL$HT$ HHt$(INݾHD$0HHH HHHEH HH=j HD$IHHMu HEHP0H5e L%HD$0HHH5E H= H>HMu HEHP0HD$0I $u ID$LP0HD$H|$(Ht HuHGP0HD$(H|$ Ht HuHGP0HD$ H|$Ht HuHGP0HD$  H|$0Hc - H*  Ht HuHGP0HD$0H|$Ht HuHGP0I~XH5 HD$z H  H= 5 4HL$HT$0LHt$}y%H E11 q )躼HHu#H[ E1J < 5]Hp HHHEH_ HH= x IHu H   :HMu HEHP0H5 LHHu H   =H5 H=[ Hy H z l ?HMu HEHP0I $u ID$LP0H|$HuHGP0HD$H|$0HuHGP0HD$0H|$HuHGP0HL$HT$ LHD$Ht$(9!E11HL$HT$ LHt$(H' 1H=f HD$HHu-HŁ E1E1 6H t eVH5H=) Hy,H~ E11C 6H@ . gHMu HEHP0 WHD$IHu/H+ E1E11H  E qH=4 _HD$0HHu-H E1E1 EH  st;HD$IHu,H E11d EHa O u1Hc LLHID$H. HH5L ID$ Il$(HD$0Ty,H1 E11 EH  I $u ID$LP0H=$ WHD$HHu-H E1E1 FH  l3HD$0IHu,H E11\ FHY G )H3LLHID$HHH5 ID$ Il$(HD$Ly,H) E11 EH  I $u ID$LP0H=OHD$0HHu-H~ E1E1 GH  d+HD$IHu,H~ E11T GHQ ? !HLLHID$HHH5 ID$ Il$(HD$0Dy,H!~ E11 EH  I $u ID$LP0H=GHD$HHu-H} E1E1 HH z \#HD$0IHu,H} E11L HHI 7 HLLHID$HHH5 ID$ Il$(HD$ , H5 H= HYy,H6p E11 H  ȹHMu HEHP0H=2 ]HD$HHu-Ho E1E1 H  rH5 HIHu,Ho E11b H_ M /HMu HEHP0H5y H= LHD$ay)H>o 1 H  ӸI $u ID$LP0H=< gIHu,Hn E11 H  Ŭ邸H5, HHD$HHu*Hn E1o Hl Z Ǭ j H5j H蒙IHu,HKd E11 'H  lݭHMu HEHP0H5 H= LHD$y)Hc 1 'H  o遭I $u ID$LP0H= IHu,Hc E11c (H` N y0H5r H袘HD$HHu*HVc E1 (H  {I $u ID$LP0H5 H= H$y,Hc E11 (H  ~铬HMu HEHP0H= (HD$HHu-Hb E1E1p )Hm [ =H5w H诗IHu,Hhb E11- )H*  HMu HEHP0H5$ H= LHD$,y)H b 1 )H  鞫I $u ID$LP0H= 2IHu,Ha E11 *H} k MH5 H迖HD$HHu*Hsa E1: *H7 % I $u ID$LP0H5( H= HAy,Ha E11 *H  鰪HMu HEHP0H= EHD$HHu-H` E1E1 +H x ZH5 H̕IHu,H` E11J +HG 5 HMu HEHP0H51 H= LHD$Iy)H&` 1 +H  黩I $u ID$LP0H=$ OIHu,H_ E11 ,H  jH5 HܔHD$HHu*H_ E1W ,HT B $I $u ID$LP0H55 H= H^y,H;_ E11 ,H  ͨHMu HEHP0H=7 bHD$HHu-H^ E1E1 -H  ĭwH5 HIHu,H^ E11g -Hd R ƭ4HMu HEHP0H5> H= LHD$fy)HC^ 1 -H  ɭاI $u ID$LP0H=A lIHu,H] E11 .H  ӭ釧H5 HHD$HHu*H] E1t .Hq _ խAI $u ID$LP0H5B H= H{y,HX] E11 .H  حHMu HEHP0H=T HD$HHu-H] E1E1 /H  锦H5 HIHu,H\ E11 /H o QHMu HEHP0H5K H= LHD$胝y)H`\ 1( /H%  I $u ID$LP0H=^ 艑IHu,H\ E11 0H  餥H5 HHD$HHu*H[ E1 0H | ^I $u ID$LP0H5O H= H蘜y,Hu[ E11: 0H7 % HMu HEHP0H=q 蜐HD$HHu-H [ E1E1 1H  鱤H5 H#IHu,HZ E11 1H  nHMu HEHP0H5X H=1 LHD$蠛y)H}Z 1E 1HB 0 I $u ID$LP0H={ 規IHu,H/Z E11 2H  H5 H3HD$HHu*HY E1 2H  {I $u ID$LP0H5\ H== H赚y,HY E11W 2HT B $HMu HEHP0H= 蹎HD$HHu-H=Y E1E1 3H  ΢H5 H@IHu,HX E11 3H  鋢HMu HEHP0H5e H=N LHD$轙y)HX 1b 3H_ M #/I $u ID$LP0H= ÍIHu,HLX E11 4H  -ޡH5 HPHD$HHu*HX E1 4H  /阡I $u ID$LP0H5i H=Z HҘy,HW E11t 4Hq _ 2AHMu HEHP0H= ֌HD$HHu-HZW E1E1 5H  <H5 H]IHu,HW E11 5H  >騠HMu HEHP0H5r H=k LHD$ڗy)HV 1 5H| j ALI $u ID$LP0H= IHu,HiV E11. 6H+  KH5 HmHD$HHu*H!V E1 6H  M鵟I $u ID$LP0H5v H=w Hy,HU E11 6H | P^HMu HEHP0H= HD$HHu-HwU E1E1; 7H8 & ZH5 HzIHu,H3U E11 7H  \ŞHMu HEHP0H5 H= LHD$y)HT 1 7H  _iI $u ID$LP0H= IHu,HT E11K 8HH 6 iH5 H芉HD$HHu*H>T E1 8H  kҝI $u ID$LP0H5 H= H y,HS E11 8H  n{HMu HEHP0H= HD$HHu-HS E1E1X 9HU C x%H5 H藈IHu,HPS E11 9H  zHMu HEHP0H5 H= LHD$y)HR 1 9H  }醜I $u ID$LP0H= IHu,HR E11h :He S 5H5 H觇HD$HHu*H[R E1" :H  I $u ID$LP0H5 H= H)y,HR E11 :H  阛HMu HEHP0H= -HD$HHu-HQ E1E1u ;Hr ` BH5 H贆IHu,HmQ E112 ;H/  HMu HEHP0H5 H= LHD$1y)HQ 1 ;H  飚I $u ID$LP0H= 7IHu,HP E11 <H p RH5 HąHD$HHu*HxP E1? <H< * I $u ID$LP0H5 H= HFy,H#P E11 <H  鵙HMu HEHP0H= JHD$HHu-HO E1E1 =H } _H5 HфIHu,HO E11O =HL : HMu HEHP0H5 H= LHD$Ny)H+O 1 =H  I $u ID$LP0H=) TIHu,HN E11 ?H  îoH5 HHD$HHu*HN E1\ ?HY G Ů)I $u ID$LP0H5 H= Hcy,H@N E11 ?H  ȮҗHMu HEHP0H=< gHD$HHu-HM E1E1 @H  Ү|H5 HIHu,HM E11l @Hi W Ԯ9HMu HEHP0H5 H= LHD$ky)HHM 1 @H  ׮ݖI $u ID$LP0H=F qIHu,HL E11 AH  錖H5 HHD$HHu*HL E1y AHv d FI $u ID$LP0H5 H= H耍y,H]L E11" AH  HMu HEHP0H=Y 脁HD$HHu-HL E1E1 BH  陕H5 H IHu,HK E11 BH t VHMu HEHP0H5 H= LHD$舌y)HeK 1- BH*  I $u ID$LP0H=c 莀IHu,HK E11 CH  驔H5 HHD$HHu*HJ E1 CH  cI $u ID$LP0H5 H=% H蝋y,HzJ E11? CH< *  HMu HEHP0H=v HD$HHu-H%J E1E1 DH  鶓H5 H(IHu,HI E11 DH  sHMu HEHP0H5 H=6 LHD$襊y)HI 1J DHG 5 I $u ID$LP0H= ~IHu,H4I E11 EH  ƒH5( H8~HD$HHu*HH E1 EH  递I $u ID$LP0H5 H=B H躉y,HH E11\ EHY G ")HMu HEHP0H= }HD$HHu-HBH E1E1 GH  ,ӑH5- HE}IHu,HG E11ÿ GH  .鐑HMu HEHP0H5 H=S LHD$ˆy)HG 1g GHd R 14I $u ID$LP0H= |IHu,HQG E11 HH  ;H55 HU|HD$HHu*H G E1о HH;  =靐I $u ID$LP0H5 H=_ Hׇy,HF E11y HHv d @FHMu HEHP0H= {HD$HHu-H_F E1E1# IH  JH5: Hb{IHu,HF E11 IHݽ ˽ L魏HMu HEHP0H5 H=p LHD$߆y)HE 1 IH o OQI $u ID$LP0H= zIHu,HnE E113 KH0  XH5 HrzHD$HHu*H&E E1 KH ؼ Z麎I $u ID$LP0H5C H=| Hy,HD E11 KH  ]cHMu HEHP0H= yHD$HHu-H|D E1E1@ LH= + e H5׽ HyIHu,H8D E11 LH  gʍHMu HEHP0H5 H= LHD$y)HC 1 LH  jnI $u ID$LP0+贅IHu,HC E11R HO = rHa H5R Hjy)HGC 1 H  t܌H H5 L'y)HC 1̺ Hɺ  u陌H H5 Ly)HB 1 H t vVHh H5Y L衃y)H~B 1F HC 1 wH H5 L^y)H;B 1 H  xЋH H5 Ly)HA 1 H  y鍋Ho H5` L؂y)HA 1} Hz h zJH H5 L蕂y)HrA 1: H7 % {H H5 LRy)H/A 1 H  |ĊHv H5g Ly)H@ 1 H  }遊H# H5 Ĺy)H@ 1q Hn \ ~>H H5 L艁y)Hf@ 1. H+  H} H5n LFy)H#@ 1 H ַ 鸉H* H5 Ly)H? 1 H  uH H5 Ly)H? 1e Hb P 2H H5u L}y)HZ? 1" H  H1 H5" L:y)H? 1߶ Hܶ ʶ 鬈H H5 Ly)H> 1 H  iH H5| Ly)H> 1Y HV D &H8 H5) Lqy)HN> 1 H  H H5 L.y)H > 1ӵ Hе  頇H H5 L~y)H= 1 H { ]H? H50 L~y)H= 1M HJ 8 H H5 Le~y)HB= 1 H  ׆H H5 L"~y)H< 1Ǵ HĴ  锆HF H57 L}y)H< 1 H o QH H5 L}y)Hy< 1A H> , H H5 LY}y)H6< 1 H  ˅HM H5> L}y)H; 1 H  鈅H H5 L|y)H; 1x Hu c EH H5 L|y)Hm; 15 H2  HT H5E LM|y)H*; 1 H ݲ 鿄H H5 L |y)H: 1 H  |H H5 L{y)H: 1l Hi W 9H[ H5L L{y)Ha: 1) H&  H H5 LA{y)H: 1 H ѱ 鳃H H5 Lzy)H9 1 H  pHb H5S Lzy)H9 1` H] K -H H5 Lxzy)HU9 1 H  H H5 L5zy)H9 1ڰ Hװ Ű 駂Hi H5Z Lyy)H8 1 H  dH H5 Lyy)H8 1T HQ ? !Hü H5 Llyy)HI8 1 H  ށH5p H= L)yy)H8 1ί H˯  雁I $ID$LP0rH  L15 E1xH7 m Hj X Ī:H$8dH+%(t#sHHH[]A\A]A^A_AUIATLQH=Ю HGHHtI,yIMuHu LH5F6 H81xLZA\A]HGpHHGxHHHHHtHHHtHHHtHAWIAVIAUATUSHHdH%(H$1 IPLLD$qLD$HHl HHEH5% HH HIM HmID$1E1AH;ro  H;p H;p IcLL$LD$T$pT$LD$HLL$H MtLHHcƒIEHcLlLDID$LM ?pH QH o P ; HL$1HLAIpH QHL$P H= @9oM Hmu HEHP0I,$u ID$LP0I}u IELP0H AuI} IMIHL$qHL$ILHAA)rHjI}IELP0LiIH5~ LIHVoHD$0HH]@HHEn H f/ H/ SH5o0 L o> AH81qH A/ X H Z ߧ  H=W/ E1SpHD$XdH+%(gHhL[]A\A]A^A_DHEHP0 fHELD$HP0LD$VE1HHoAHHu UnHuHym H5/ H8:k5nAHH7.  *H ܦ L HmHEHP0 f.H_LoAHHu mHuHl H5:/ H8jmAHkH- Q +HN < W E1L;-l :IHL$/AAoL% H= IT$LSmIHHI@H5 HH`LD$LLD$HH}I(u I@LP0|$/kIH?Hj H9EFLHKIH I,$,HmvHEHP0gf.HIH0 HcH>fHF0HD$HHE(HD$@HE LHD$8HEHD$0dfIHH~QHu%H5 LHVlHD$HH] IMLd$0Lt$8Ll$@H|$H@HHuH5r LHVkHD$8H7 IH55 LHVkHD$@H IWfID$@=#LL$0E1H)Ll$8LD$@Ht0ID$LX@ uMT$LL$LD$ HLALL$LD$IM 1H* L H LMIMLIfDI(fMt I)OHt HmMtzI,$usID$LP0ffD;jHeHC*  (H  @ .fDH* Ӣ *HТ  J fDH  E1H=%* 5 kfH@`H HH HIH H i I9GrLpI/AIGLP0@ID$LP0LLL$0H)DLL$Ht0Ll$8LD$@LD$?LD$LL$HIMtI)uIALD$LP0LD$I(I@LP0HFLHD$0bI1 H( U HR D HmHEHP0fHa(  /H  x I(DE1E1fDI@LL$LP0LL$LH5. nIHtsg@$f.gHuGH=G LHGHHHHHSg LH5' H81jH' : 1H7 % hHHjIpLE1f.IALP0LD$gHuQH= LD$HGLHHLD$HHSHf LH5& H81`iH& { 1Hx f /f1 HxiI ML$M.Il$IHEI,$HEIAfH1&  1H ֞ qf1HL3hIHH%  1H  HmI=fDHd H5& H8bCDHd H5& H8bbDSeHuGH=ߝ LHGHH0IMHd LH5T% H81gH% ҝ /Hϝ  v H=& HL$cHL$;DdHHb H5|& H8aLLD$gLD$HfD/{ fLuMLMIIHmuHELL$HP0LL$IAH;c H;c LL$cLL$HILpL` IAHH%LL$XcLL$H QH b P ;1HL$LLLL$IcHL$LL$X SP =H @9:MUI/LIGLP0DID$LL$LILD$P0HELL$ALD$fH#  /H 1 -1HL$0ILL # H5d   f.H " 5[ H \ N fHHJa H k" H5|# jL z1 AH" H81dY^ H@`HHHLHHHa H9Eu%HXiHmAHEHP0HH5v( |hHHuHH` AH5" jL 0 H ! H8H! 1&d_ AXHz! L1 1H. LM HH` L V0 H5H" jAH !! Hm! H81cAY AZbcHRIQBፁLt$01Ld$8LB uIyLL$Ht$0_ALL$IMI.uIFLL$LP0LL$I,$uID$LL$LP0LL$LLD$RcLD$HaH1LALD$LL$IJLL$_LL$@$ cILHt$0LL$Lt$0Ld$8LL$HI7H ML /H 1o _HJHe^ H5 H8n\/d_HH<^ H5y H8E\HQ  /H  I,$I)I.LE1LL$^LL$HMLͺ/ AH= HL$LL$]LL$HL$1LLLL$`LL$HI13[H ML8 /H5 1! 1ɺALL$IHF\ H5 H8[LL$1MLE1ff.@AWAVAUATUHSHXH^dH%(HD$H1HD$ HD$(HD$0HD$8HjHpLfLv Ln(H~019`HH=LL`HH^H@HEHHED$ |$ Hmu HEHP0ID$ID$HHHL\HAH9 AD|$ L;-[ L%Q H=* IT$L\HHq HHEH5 HH HIM HmL5 H=Ȕ IVL\\HH HHEH5 HHa HIM@ HmeID$11H; Z A9 H;*[ H;[ Ic|T$ #[T$ HIK HtHhHcƒIEHcMlMtID$HH= ZL5YZ H QP A; 1LLIZH QP AH= @9M I/u IGLP0I,$u ID$LP0I}u IELP0H^ AuI} MuI\t$ IL|$ HLUHIUI}IELP0LTIH56 LIHVvZHD$ HH]@HHX H  H| SH5' L ') AH81\H  X H Z 8 8H=! E1 [HD$HdH+%(6HXL[]A\A]A^A_DHEHP0;HEHP0D$ EfDHHZD$ HHu$YHuHAX H5J H8VXD$ HH  ]H  HmHEHP0t$ |$ HL$IغSL%M H=& IT$LXIHHIFH5 HH~LIMI.u IFLP0H|$VIHVHWV I9D$\LL蹨IHE I.I,$-ID$LP0AD$A9HV H5 H8TWD$ HH O ^HL : DD$ fDAD$A9널HIH HcH>fHF0HD$8HE(HD$0HE LHD$(HEHD$ DQIHH~QHu%H5 LHVVHD$8H}IMLd$ Lt$(Ll$0H|$8@H*HuH5R LHVVHD$(H IH5 LHVyVHD$0H IWfID$@=nHl$ E1H)Ll$(Lt$0Ht ID$LP@ uML$q HLAIM dH v  Hw LILMII.6Ht Hm6MtI/u IGLP0Mt|I,$uuID$LP0h+UHH3  [H ؍ .fDH  Í ]H  fDH  E1H=' 5 VfH@`H HH HIH HS I9G0L(]I/D$ .IGLP0IFLP0FfLLl$(H)DHl$ Ht Lt$09IHHtHmu HEHP0I.IFLP0fDHFLHD$ NIHSHH C bH@ .  I.|1E1IFLP0LH5 YIHR@$ fRHuGH= LHGHHaHH]HR LH5 H81_UH z dHw e ] HUI&d_ Hw 5 H2 $ HmHEHP0fHEHP0RHuGH= LHGHHHHHQ LH5 H81TH  dH  b \dd +HTIIl$HM|$HEII,$u ID$LP0IGMAHI  dH  1LLSSIHH  dH  I/xdHO H5 H8MDPHuGH= LHGHH4IMH+P LH5 H81RHX  bH   PH= 4OOHH N H5 H8LLRIzLbE1 EM|$MIl$IHEI,$u ID$LP0HEH;N :H;BO NIHLxLp HEHHNL56N H QP A;z1LHINX SP AH= @9MI,$ID$LIP0H s bHp 1\ % 1HL$ ILL  H54P o  fH a 5 H   8nHH M H + H5< jL : AHs H81PY^ H@`HLHH<LHH+HPL%BM L9uaDxHEHHHtVHMHAH9HmHEHP0H5 HSHHHPEA9tHBL H5s H8JHmaHEHP0RfEA9iDHmD$ HE1R0HHK AH5 jL  H H8H 1!O_ AXHUBፁL|$ 1Lt$(LR uH}Ht$ #AIMI/u IGLP0I.u IFLP0IH  dH LI v -HHJ L  H5 jAH H H815NAY AZ'NHuNHjH1LAIJ@$fGNIHt$ HL|$ Lt$(SIHH MĨ bHɃ 1 , GHJHJHHI H5 H8GJHHeI H5 H8nGHz 4 bH1  < I.HmI/XHB`HHHHIHL9`u$LTI.AtIFLP0eH5y H|PIHuIHbB ,H=K Hr1LHKIHFWIHKH/H H5l H88F0HD MI bH 1 5 s1ɺAIH G H5 H8E6AWAVAUATUHSHhH^dH%(HD$X1HD$0HD$8HD$@HD$HHHLfLv Ln(H~01KHHLLKHHH@HEHHEAƩ-A9Hmu HEHP0ID$%ID$HHXAD$AǩuAL;-F L%G H= IT$LGHH' HHEH5 HH HIM' HmL H= IPLLD$MGLD$HHl HHEH5] HH HIM HmID$1E1AH;D  H;F H;pF IcLL$LD$T$FT$LD$HLL$H MtLHHcƒIEHcLlLDID$LM EH QH *E P ; HL$1HLAIEH QHL$P H= @9oM Hmu HEHP0I,$u ID$LP0I}u IELP0H+~ AuI} IMIHL$TGHL$ILHAA@H@I}IELP0Lx?IH5} LIHV>EHD$0HH]@HHC H  HV SH5 L  AH81`GH  X&H |} Zm} _} &H= E1EHD$XdH+%(gHhL[]A\A]A^A_DHEHP0 fHELD$HP0LD$VE1HH?EAHHu CHuHB H5 H8@CAHH q| )Hn| \| \HmHEHP0 f.H_LDAHHu 5CHuHYB H5b H8@CAHkH { *H{ { gE1L;-B :IHL$/AA=L%g{ H=@{ IT$LBIHHI@H5z HH`LD$LLD$HH}I(u I@LP0|$/AIH?Hh@ H9EFLH˒IH I,$,HmvHEHP0gf.HIH HcH>fHF0HD$HHE(HD$@HE LHD$8HEHD$0;IHH~QHu%H5,z LHVAHD$HH] IMLd$0Lt$8Ll$@H|$H@HHuH5y LHV>AHD$8H7 IH5y LHVAHD$@H IWfID$@=#LL$0E1H)Ll$8LD$@Ht0ID$LX@ uMT$LL$LD$ HLALL$LD$IMy 0HL Lx Hy LMIMLIfDI(fMt I)OHt HmMtzI,$usID$LP0ffD?HeH}x 'Hzx hx P.fDHSx )HPx >x ZfDH 9x /x E1H=5x @fH@`H HH HIH H> I9GrLGI/AIGLP0@ID$LP0LLL$0H)DLL$Ht0Ll$8LD$@LD$迍LD$LL$HIMtI)uIALD$LP0LD$I(I@LP0HFLHD$08I0Hv Hv v HmHEHP0fHv .Hv v I(DE1E1fDI@LL$LP0LL$LH5* 0DIHt<@$f.;=HuGH=u LHGHHHHH< LH5<H81?Hu 0Hu u hH?IpLE1f.IALP0LD$HAt 0Ht t /f0H>I ML$M.Il$IHEI,$HEIAfHkt 0Hht Vt qf1HL=IHHat 0Ht t HmI=fDHA: H5H88CDH!: H5H87D:HuGH=_s LHGHH0IMHk: LH5H817=HRs .HOs =s H=HL$o9HL$;D3:HHC8 H5H87LLD$ =LD$HfD.fLuMLMIIHmuHELL$HP0LL$IAH; 9 H;d9 LL$8LL$HILpL` IAHH%LL$8LL$H QH >8 P ;1HL$LLLL$I8HL$LL$X SP =H @9:MUI/LIGLP0DID$LL$LILD$P0HELL$ALD$fH;q .H8q 1$q -1HL$0ILL H58 7f.H !5p H p p fHH6 H H5jL  AHEH81d:Y^H@`HHHLHHH7 H9Eu%H8@HmAHEHP0HH5=HHuHH 6 AH5?jL = H H8H19_ AXHLo 0Ho LMo HH5 L  H5jAH HH8109AYAZb}9HRIQBፁLt$01Ld$8LB uIyLL$Ht$0_ALL$IMI.uIFLL$LP0LL$I,$uID$LL$LP0LL$LLD$8LD$HaH1LALD$LL$IJLL$.5LL$@$8ILHt$0LL$Lt$0Ld$8蒄LL$HI7HRMLn .Hn 1m  5HJH3 H5"H81/4HH3 H5H81Hm .Hm vm I,$I)I.LE1LL$j4LL$HMLͺ.AH=HL$LL$Q3LL$HL$1LLLL$[6LL$HI10HMLl .Hl 1l 1ɺALL$IH1 H5H80LL$1MLE1ff.@AWAVAUATUHSHXH^dH%(HD$H1HD$ HD$(HD$0HD$8HzHpLfLv Ln(H~015HHMLLr6HHnH@HEHHEfD$f|$ Hmu HEHP0ID$ID$HHHLb2HAH9* fAfD|$ L;-j1 L%j H=j IT$L92HH} HHEH5j HH HIM HmL5kj H=Dj IVL1HH HHEH5j HHm HIML HmaID$11H;/ AE H;0 H;1 IcT$ 0T$ HIW HtHhHcƒIEHcMlMtID$HHI d0L5/ H QP A; 1LLI90H QP AH= @9M I/u IGLP0I,$u ID$LP0I}u IELP0Hh AuI} MuI2t$IL|$ HL},H*I}IELP0L0*IH5h LIHV/HD$ HH]@HH}. H H!SH5L AH812H yXO H 4h Z%h kh O kH=E10HD$HdH+%(JHXL[]A\A]A^A_DHEHP0?HEHP0E1fDD$CfHH0fD$HHu.HufDH- H5JH8+}.AfDL$HHy3g H0g g HmHEHP0@t$|$ HL$Iغ*L%f H=f IT$L1.IHHIFH5^f HHLIMI.u IFLP0H|$j,IH^H+ I9D$dLL1~IHM I.I,$%ID$LP0AD$A9HV, H5H8*-ft$ HHe He e 1fL$ @AD$A9yyfDHIHHcH>fHF0HD$8HE(HD$0HE LHD$(HEHD$ &IHH~QHu%H5d LHVh,HD$8H}IMLd$ Lt$(Ll$0H|$8@HHuH5d LHV,HD$(H IH5d LHV+HD$0H IWfID$@=bHl$ E1H)Ll$(Lt$0Ht ID$LP@ uML$u HLAIMc H0c Hc LILMII.6Ht Hm6MtI/u IGLP0Mt|I,$uuID$LP0h*HH]c HZc Hc z .fDHy3c H0c c fDH c c E1H=5b ~+fH@`H HH HIH Hk) I9G0LH5I/fD$IGLP0f.IFLP06LLl$(H)DHl$ Ht Lt$0xIHHtHmu HEHP0I.IFLP0fDHFLHD$ w#IH(HHa Ha a I.|1E1IFLP0LH5LR/IH(@$fk(HuGH=` LHGHHeHHQH( LH5lH81*H0` H` ` H*I H` H` ` HmHEHP0fHEHP0'HuGH=` LHGHHHHH#' LH5H81)HP ` H` _ \+H*IIl$HM|$HEII,$u ID$LP0IGMAvHs_ Hp_ ^_ #1LL(IHHq+_ H(_ _ .I/xdHQ% H5H8#D&HuGH=^ LHGHH8IMH% LH5H81g(H^ H^ m^ PH=I$k%HH{# H54H8L"LH(IrL底E1丵 EM|$MIl$IHEI,$u ID$LP0HEH;W$ >H;$ R$IHLxLp HEHH5$L5# H QP A;~1LHI $X SP AH= @9M"I,$ID$LIP0H)\ H\ 1\ 1HL$ ILL uH5d% > fH 5\ H \ ~\ k^HHz" H5L jAH HH81&AZ: A[H@`HNHH>LHH-HPL%" L9u_|HEHHHtVH#HAH9HmHEHP0H5jHm)HHHPEA9tH! H5kH8sHmQHEHP0BfEA9iDHmf|$ HE1R0HH H5+L +jAH HH81$[4 ]HUBፁL|$ 1Lt$(LR uH}Ht$ #AIMI/u IGLP0I.u IFLP0IHh"Z HZ LIZ )HH H5@L @jAH HH81#A\. A]##H#HfH1LAIQ @$b#IHt$ HL|$ Lt$(oIHHMI8Y H5Y 1!Y CH5 H" HH H57H8|HH H5H8HX HX X I.HmI/THB`HHHHIHL9`u$L3,I.ApIFLP0aH5H%IHuH (H=n1LH+!IHHGH H5H8,HMIdW HaW 1MW o1ɺAIHw H50H8H6AWAVAUATUHSHhH^dH%(HD$X1HD$0HD$8HD$@HD$HHHLfLv Ln(H~01 HHLLB!HHH@HEHHEAƩ-fA8Hmu HEHP0ID$$ID$HHWAD$AǩtfAL;-R L%U H=U IT$L!HH% HHEH5vU HH HIM% HmLSU H=,U IPLLD$LD$HHj HHEH5T HH HIM HmID$1E1AH;`  H; H; IcLL$LD$T$rT$LD$HLL$H MtLHHcƒIEHcLlLDID$LM -H QH  P ; HL$1HLAIH QHL$P H= @9mM Hmu HEHP0I,$u ID$LP0I}u IELP0HS AuI} IMIHL$HL$ILHAA7HI}IELP0DLIH5nS LIHVHD$0HH]@HH5 H VHSH5_L _AH81H 1XH R ZR 7R 7H=E1CHD$XdH+%(gHhL[]A\A]A^A_DHEHP0fHELD$HP0LD$XE1HHAHHu EHuHi H5H8*%AHH'Q \HQ Q HmHEHP0 f.H_LAHHu HuH H5RH8AHmHAQ ]H>Q ,Q E1L;- fHF0HD$HHE(HD$@HE LHD$8HEHD$0TIHH~QHu%H5O LHVHD$HH] IMLd$0Lt$8Ll$@H|$H@HHuH5bO LHVHD$8H7 IH5%O LHVHD$@H IWfID$@=%LL$0E1H)Ll$8LD$@Ht0ID$LX@ uMT$LL$LD$ HLALL$LD$IMN cHLoN HpN LMIMLIfDI(fMt I)OHt HmMtzI,$usID$LP0ffD+HeH3M ZHM M .fDH M \HM M fDH M M E1H=_5M fH@`H HH HIH H I9GrLI/AIGLP0@ID$LP0LLL$0H)DLL$Ht0Ll$8LD$@LD$/cLD$LL$HIMtI)uIALD$LP0LD$I(I@LP0HFLHD$0 IcnHEL HBL 4L HmHEHP0fHQ L aHL K $I(DE1E1fDI@LL$LP0LL$LH5IHtc@$f.HuGH=7K LHGHHHHHC LH5H81Hp*K cH'K K lhH8IrLE1f.IALP0LD$HuQH=J LD$HGLHHLD$HHUH LH5H81PHkJ cHhJ VJ q/fcsHhI ML$M0Il$IHEI,$HEIAfH!I cHI I qf1HL#IHHI cHI vI HmI=fDH H5 H8r CDH H5H8R DCHuGH=H LHGHH0IMH LH5DH81HH aHH H "H=HL$HL$=DHH H5lH8 LLD${LD$HfDa'fLuMLMIIHmuHELL$HP0LL$IAH;y H; LL$oLL$HILpL` IAHH%LL$HLL$H QH P ;1HL$LLLL$IHL$LL$X SP =H @9:MUI/LIGLP0DID$LL$LILD$P0HELL$ALD$fHF aHF 1F 4-1HL$0ILL PH5, f.H 5KF H LF >F 7fHH: H [H5ljL jAHH81Y^H@`HHHLHHHv H9Eu%HXHmAHEHP0HH5flHHuHH| AH5jL H H8H1_AXHjL!E cHE LME HH L FH58jAH HH81AYAZbHRIQBፁLt$01Ld$8LB uIyLL$Ht$0_ALL$IMI.uIFLL$LP0LL$I,$uID$LL$LP0LL$LLD$BLD$HaH1LALD$LL$IJLL$ LL$@$ ILHt$0LL$Lt$0Ld$8ZLL$HI7HMLvC aHsC 1_C ;} HJHU H5H8^/T HH, H5iH85HAB aHB B KI,$I)I.LE1LL$ LL$HMLͺaQAH=pHL$LL$LL$HL$1LLLL$ LL$HI1#HtML(B aH%B 1B D1ɺALL$IH6 H5H8LL$1MLE1ff.@AWAVAUATUHSHhH^dH%(HD$X1HD$0HD$8HD$@HD$HHHLvLn Lf(H~01) HHLL HHH@HEH?HH^ HV AH9AHmu HEHP0IFqIFHHHHcH>fLIH5~@ LIHVHD$0H_H]@HHE H fHSH5oL oAH81 H AXH ? Z? ? H=E1SHD$XdH+%(HhL[]A\A]A^A_DE1DL;%  SL5l? H=E? IVLHH HHEH5.? HH HIMHmu HEHP0L? H=> IPLLD$mLD$HH HHEH5> HH} HIM\ HmuHELD$HP0LD$IF1E1H;  AE H;& HH; IcLL$LD$T$T$LD$HLL$H; MtLHHcƒI$HcLdLDIFLM6 H QH < P ; HL$1HLAIH QHL$P H= @9iM Hmu HEHP0I.u IFLP0I<$u ID$LP0H>= At$I|$ IL$IHL$dHL$ILHDDH#I<$UID$LP0EE~AfAVAHBH< H< < .cAFAVHH HHcAH9tH H5%H8E~wAFAVHH HcAH9RE1zIHL$,DDxL%< H=; IT$LmIHyHI@H5; HH* LD$LLD$HHI(u I@LP0Hc|$,IH H H9E LHeSIHy I.HmHEHP0DEUHH ЉAH9]H~ H5H8?:AH=H<: H: : #HmHEHP0wDmHfIHHcH>f.HF0HD$HHE(HD$@HE LHD$8HEHD$0IHH~QHu%H5,: LHVHD$HH IMF Lt$0Ll$8Ld$@H|$H@HHuH59 LHV>HD$8H^ IH59 LHVHD$@H IWfIF@=LL$0E1H)Ld$8LD$@Ht0IFLX@ uMVLL$LD$ HLALL$LD$IM MtI)uIALD$LP0LD$I(`I@LP0QL(HcAH9H4H&@HmHۿ8 H8 8 .fDHk8 Hh8 V8 !fDH Q8 G8 E1H=538 ^H@`H HH HIH H I9G|L8 I/AIGLP0@IFLP0JfHFLHD$0I@H H5H8VLH5@FIHj6D @$fLLL$0H)DLL$Ht0Ld$8LD$@LD$oMLD$LL$HI6 H L6 H6 LMIMLIfI(Mt I)HtHmtVM:I.0IFLP0!fHU6 HR6 D6 HmuHEHP0fHa6 H6 6 OI(E1E1fDI@LL$LP0LL$)HuGH=5 LHGHHHH9H LH5H81WHr5 Ho5 ]5 HI H;5 H85 &5 vI.I)QLE1@IALP0ZLD$HuQH=4 LD$HGLHHHLD$HHH LH5 H81pHѻ4 H4 v4 fHI{MNMInIHEI.HEIADHA3 H3 3 )f1HLCIHH3 H3 3 HmIfDHuGH=/3 LHGHHVIMUH; LH5H81Hh"3 H3  3 MH=HL$?HL$-DHH H5̻H8LLD$LD$HfDR#fLmMLMIEIHmuHELL$HP0LL$IAH; =H;3 LL$LL$HILhLp IAHHLL$LL$H QH P ;1HL$LLLL$InHL$LL$X SP =H @9_M.I/LIGLP0@IFLL$LILD$P0HELL$ALD$H1HH10 H0 10 _ 1HL$0ILL H5, f.H ѷ50 H 0 ~0 HHz H H5jL AH/H81Y^H@`HHHLHHqH H9Eu%HHmAHEHP0HH5HHuHH AH5jL H H8Hg1O_AXHHo L H5jAH zH$H81 AYAZVHsIQBፁLl$01Lt$8LB uIyLL$Ht$0XALL$IMImuIELL$LP0LL$I.uIFLL$LP0LL$LLD$LD$HHLW. HT. LM<. qH1LALD$LL$ILL$LL$@$5ILHt$0LL$Ll$0Lt$8IH I,$HmHEHP0DEUHH ЉAH9H H5H8OJAHHL& H& % HmHEHP0DuAD$AT$HH ЉAH9H H5H8AHHj% Hg% U% @E|$ifDH&IHHcH>fHF0HD$HHE(HD$@HE LHD$8HEHD$0tIHH~QHu%H5$ LHV(HD$HH IMV Ld$0Lt$8Ll$@H|$H@HJHuH5$ LHVHD$8Hg IH5E$ LHVHD$@H IWfID$@=qLL$0E1H)Ll$8LD$@Ht0ID$LX@ uMT$LL$LD$ HLALL$LD$IM MtI)uIALD$LP0LD$I(I@LP0DHHU# HR# @# .fDHq+# H(# # fDH # # E1H=5" vVH@`H HH HIH Hk I9G|LI/ASIGLP0D@ID$LP0HFLHD$0IpH H5zH8ZLH5IHjD@$fLLL$0H)DLL$Ht0Ll$8LD$@LD$/8LD$LL$HI! HL! !H! LMIMLIfI(Mt I)HtHmtVM:I,$/ID$LP0 HW! H! ! HmuHEHP0fH! H  I(E1E1fDI@LL$LP0LL$)HuGH=? LHGHHHHHK LH5H81Hx2 H/  HHIHA H  I,$ I)LE1IALP0ZLD$HuQH=b LD$HGLHHLD$HHQHd LH5ͦH810HK HH 6 fHHI ML$M@LIH5 LIHVHD$ HoH]@HHU H vHGSH5L AH81H QXH  Z  H=E1cHD$HdH+%(HXL[]A\A]A^A_DH$L;% L=| H=U IWLHHHHEH5> HHn HIM Hmu HEHP0L5 H= IVLHHV HHEH5 HH HIM Hmu HEHP0IG11H;.  H;L H; HcL$EL$HI HtHhHcI$MdAHMtIGHH LL$L5x LL$H QA;P i 1LL$LLILL$H QP A=H @9 MH I)u IALP0I/u IGLP0I<$u ID$LP0Ho At$I|$ Mt$IH<$ILHLLHZI<$|ID$LP0lEwAMcITL;% L4$H<$HL$IL:L% H= IT$LOIHHIFH5l HHd LIMI.u IFLP0H|$IH= H I9GD LLQ)IH5 I.#I/nIGLP0_fEwAGII IfAGH$cEwAGII L4$HE1DmDmEII DHVIHDHcH>f.HF0HD$8HE(HD$0HE LHD$(HEHD$ IHH~QHu%H5, LHVHD$8H} IM L|$ Ll$(Ld$0H|$8@HrHuH5 LHV>HD$(H/ IH5 LHVHD$0HC IWfIG@=HcHl$ E1Ld$(H)Lt$0IGHt LP@ uMO; HLAIM HtHmu HEHP0I.IIFLP0:LBIf. HH H  .fDH H  fDH   E1H=z5k LH54:IHiIH`Hj$ H!  HmuHEHP0qHuH H55H8H$HH H  H@`H HH HIH H I9FLI.IhIFLP0Y@IFLP0fHFLHD$ IHI H5H8 DLL$LL$@$HcLLd$(H)Hl$ Ht Lt$0"IHT Hj MHk LILMI@I.Ht Hm^MtI)u IALP0MI/IGLP0wH1 H  I.DE11IFL $LP0L $cHuGH=W LHGHH.HHHc LH5̒H81/HJ HG 5 46HW H  HmHEHP0fHIHEL $HP0L $fHuGH=g LHGHHKHHxHs LH5ܑH81?HZ HW E 9(; HXI1IoHXIWHEHI/uIGHT$LP0HT$HBI׹fDH H  ]Qf1LLLL$LL$HIDHc H` N hI),[HuGH= LHGHHIM+H LH5\H81H  H  8H=LL$LL$y%DL $L $H H H5tH8L $LILE1MOMIoIHEI/uIGL $LP0L $HEH; H; L $L $HIpLHLp HEHHiL5 H QP A;|1LHI>X SP AH= @9ZM{I/IGLIP0DHa H 1 1HL$ ILL H5 xf.H 5 H   HH H ˍH5܎jL ڜAHH81DY^tH@`H]HHMLHHDuEII IHmHMHQ0DuAMcDuEII HmtkDuH5gHjHHHHUHIHHe AH5jL H pH8H>1_nAXRHUBፁ"LL$ 1Lt$(LR uH}L $Ht$ AL $IMI)u IALP0I.u IFLP0IFHH L ޚH5ЌjAH HzH818AYhAZHxHHq+ H( LI VH1LAI@$Ht$ HL $LL$ Lt$(2L $HIHMI H 1 .IpH'Hv H5H8 HE HB 0 I.fHmCI)H=Hp1LHTIHHuH H5H8HA`HtyHHtmHIHt`L9pu$L4I/IIGLP0H56H9IHuHmfHEHP0WMHuH) H5fH82HAMI H 1 |1ɺAL $Iff.AWAVAUATUHSHhH^dH%(HD$X1HD$0HD$8HD$@HD$HHHLfLv Ln(H~01)HHELLHHfH@eHEHHHHVIIHmu HEHP0ID$ID$HHH> LIIL;- lL%E H= IT$LHH HHEH5 HH& HIM Hmu HEHP0L H= IPLLD$ELD$HH HHEH5= HH HIMd HmuHELD$HP0LD$ID$1E1AH; L H; H;X IcLL$LD$T$T$LD$HLL$HB MtLHHcƒIEHcLlLDID$LM< H QH  P ; HL$1HLAIwH QHL$P H= @9M Hmu HEHP0I,$u ID$LP0I}u IELP0H AuI} IMIHL$fHF0HD$HHE(HD$@HE LHD$8HEHD$0IHH~QHu%H5< LHVHD$HH IM Ld$0Lt$8Ll$@H|$H@H"HuH5 LHVNHD$8H IH5 LHV)HD$@H IWfID$@=KLL$0E1H)Ll$8LD$@Ht0ID$LX@ uMT$LL$LD$ HLALL$LD$IM MtI)uIALD$LP0LD$I(I@LP0DHH H  $.fDH H  .fDH  E1H=5s .LH5<BIHqIHhHr, H)  0HmuHEHP0qHH5ԆHH& IH`H  H  ;fH@`H HH HIH H+ I9GLI/IxIGLP0i@ID$LP0 HFLHD$0蟹IHY H5‚H8D諾@$SfLLL$0H)DLL$Ht0Ll$8LD$@LD$LD$LL$HId H~Ls Ht LMIMLIfI(Mt I)gHtHmtVMI,$ID$LP0H7~ H  HmuHEHP0fH~ H  \I( E1E1fDI@LL$LP0LL$)蓽HuGH= LHGHH:HHH+ LH5}H81HX} H  hH(ILE1f.IALP0LD$HuQH=r LD$HGLHH.LD$HHHt LH5|H81@H|[ HX F f+HXIsML$MIl$IHEI,$HEIAvfH| H  Yf1HLIHH{{ Hx f Hm1yfDsHuGH= LHGHHIMeH LH5t{H81׽H8{ H  ZHH=|HL$HL$-DӺHH H5|H8贷LLD$諽LD$HfD_SfH H5~H8ZGDLuMLMIIHmuHELL$HP0LL$IAH; H; LL$LL$HILpL` IAHHLL$XLL$H QH P ;1HL$LLLL$IHL$LL$X SP =H @9&MDI/LIGLP0DID$LL$LILD$P0HELL$ALD$ffHy H 1 l91HL$0ILL yH5 lRf.H x5[ H \ N HHJ H kxH5|yjL zAH9yH81Y^H@`HHHLHHH H9E@H HmI/HEHP0 HH AH5xjL цH wH8Hx1:_AXHHZ L H5xjAH ewHIxH81AYAZAHIQBፁLt$01Ld$8LB uIyLL$Ht$0cALL$IMaI.uIFLL$LP0LL$I,$uID$LL$LP0LL$LBLD$薹LD$HH1LALD$LL$IHmvL$ H! LM LL$µLL$@$LHt$0LL$Lt$0Ld$83LL$HIHuML H 1 s%軸IX螵HHv H5uH8uH^HM H5uH8VCHbu H  I,$I) I.XE1LLL$LL$HMLͺ1LLLL$LL$HIlH=lvHL$LL$轳LL$HL$D1ɺALL$IH~tML2 H/ 1 |MLE1xHI H5vH8LL$#AWAVAUIATIUSHHHt$dH%(HD$81H;, FL= H=h HIWLIH5 HIFH5N HH LIM I.L5, H= IVL虳HH HHCH5 HH~HIM H+u HCHP0IG11H;F A H;c H; Ic5T$\T$HI4 HtHXHcƒHEHcIlMtIGHH LL$H LL$H Q;P  1LL$LLHLL$H QP =H9CH I)I/H}H uH} IUL}Hh HD$HHT$H裲IH H@HT$HHLLIH IUH HHHT$OIH H@HT$HHRLLIH H@H;s MiMIQIEHI)HLHT$HT$HH1 ImIuIEHT$LP0LL$I)u IALP0H+u HCHP0荲1H|$I~HD$LAHH9\$uL=IFL% LM H^ H QP ;K 1LLAI¯H QP H= @9gM I.uIFLP0M I,$u ID$LP0HEIHPHUHEHHEHP0IFLP0L= LqLLZHHH@HHHLLHHL= MuLL HH H@HHHLLHHH@H;3 ULkMHLsIEIH+u HCHP0LLpIH ImLu IELP0H+u HCHP0I/u IGLP0[LIHD$LfI~!HEL- LM, ѭHB H QP ;1LHAI覭H QP H= @9cM HmuHEHP0MI,$u ID$LP0fInDIHHD$8dH+%(\ HHL[]A\A]A^A_f.HEHP0VIGLP0H}@fIALP0I/f. @9 IAHT$LP0HT$LL$vLL$@$HEsfDH踯IzIG@=H\$ E1H)Hl$(Lt$0Ht IGLH@ uMWHLAHHH=m H  H@LLL${LL$HH(H l H  I.uIFLL$LP0LL$I)u IALP0IDHIHH xlB H C 5 Hmu HEHP0H+Au HCHP0MMUI.K11IuILHl$(H)DH\$ Ht Lt$0HH&Ht H+I.IFLP0}DHsHuHH= LHGHHHtIH LH5sjH81֬qH k H 5 H=nVM]ILE1HHj H  sI.u IFLP0H  u M5h fDL萬IkHuHH= LHGHH7HtHH LH5kiH81ΫHj H  vI/uVE1IGLP0MtI.u>IFLE1P0HH+u HCHP0H  5} E13@$f.@$fHqi; H8 & xH+I/tMj1JfD1.HCHP0kI_HIWHHI/uIGHT$LP0HT$HBIAfDHY HH8ΣIp@HhS HP > HI/3[fDH HH8fH7h H  I.IFLIP0R1LLLL$.LL$HHDHgL H { fDHA LH8趢E1X@H=)hLL$LL$LL$FLL$HpHQ H5 hH8"LL$PH LH8.Hf H  Hmu HEHP0E1HfE1 H s ZfD1LLӧI.IlH(I.u IFLP0I%H=fUHmu HEHP0CE11LHMHmI&H=fn̤HuH H5fH8象LIѸML HeZ HW E HcH1LAHHT H5 fH8%۠ImMeff.AWAVAUIATIUSHHH4$D$dH%(HD$81H; QL=z H=S HIWLIH@ HIFH59 HH LIM I.L5 H= IVL脣HH HHCH5 HHHIM H+u HCHP0IG11H;1 A H;N H; Ic@T$GT$HI? HtHXHcƒHEHcIlMtIGHH LL$Hy LL$H Q;P  1LL$LLHԡLL$H QP =H&9NH I)I/H}H uH} IUL}HS HD$HHT$H莢IH H@HT$HHLLIH IUH HHHT$:IH& H@HT$HH]LLIH H@H;^ (MiMIQIEHI) HLHT$HT$HH4 ImIuIEHT$LP0LL$I)u IALP0H+u HCHP0x1H|$I~#D$H$LAHH9\$uL IFL% LM ПHA H QP ;F 1LLAI襟H QP H= @9jM I.uIFLP0M I,$u ID$LP0HEIHPHUHEH"HEHP0IFLP0L= LqLL:HHH@HHHLLHHL= MuLLHH H@HHHLLHHH@H; ULkMHLsIEIH+u HCHP0LLPIH ImLu IELP0H+u HCHP0I/u IGLP0;LD$IH$LfI~HEL- LM 謝H H QP ;1LHAI聝H QP H= @9^MHmuHEHP0MI,$u ID$LP0fInIHHD$8dH+%(T HHL[]A\A]A^A_f.HEHP0KIGLP0H}5fIALP0I/ f. @9 IAHT$LP0HT$LL$VLL$@$HEsfDH蘟IoIG@=H\$ E1H)Hl$(Lt$0Ht IGLH@ uMWHLAHHH] H  H@LLL$[LL$HHbH \ H  I.uIFL $LP0L $I)u IALP0IHIHH X\" H #  Hmu HEHP0H+Au HCHP0MMMI.C11IjILHl$(H)DH\$ Ht Lt$0HHHt H+I.IFLP0rDHSHuHH= LHGHHHtIH LH5SZH81趜H Z H 5 H=^6M]ILE1HHZ H n I.u IFLP0H _ U M5H fDLpIKHuHH= LHGHH7HtHH LH5KYH81讛HY H  I/uVE1IGLP0MtI.u>IFLE1P0HH+u HCHP0H q g 5] E1@$f.@$fHQY H  H+I/tMb1JfD1.HCHP0kI_HIWHHI/uIGHT$LP0HT$HBIAfDH9 HH8讓IPp@HiX3 H0  HI/3[fDHі HH8FHX H  RI.IFLIP0R1LLLL$LL$HHDHWLp Hm [ +fDH! LH8薒E1X@H= XLL$_LL$L $'L $HrH3 H5WH8L $SH LH8HV H  Hmu HEHP0E1%HVE1p Hm [ fD1LL軗I.ITH(I.u IFLP0I-H=V=!Hmu HEHP0E11LH5HmI3$H=Vn贔HuHȒ H5VH8虑LIѸ_ML HxUB H? - H#KH1LAHH< H5UH8 ÐIuMmAWAVIAUATIUSHXHt$LD$dH%(HD$H1H; )L=j H=C HIWLԓHHh HHCH5) HHHIMH+uHCLT$HP0LT$L= H= LT$IWL_LT$HIHICH5w HHLT$ LL\$L\$LT$ IMI+IB11H; AeH; H H;q IcLT$ T$ T$LT$ HIUHtHXHcƒHEHcIlM|IBHHzLT$ƑL=7 LT$H QA;P  1LT$LLH葑LT$H QP A=Hr9xH Im.I*H}H B L}L1HHHH8H LsuH} I9HD$L- HPLHHT$ IHCH@HT$HHLHt$IH&HD$L-m HPLHHT$豑IH=H@HT$HH LHt$IH H@H;ӎ  MjM IRIEHI*uIBHT$LP0HT$HLHT$HT$HIImIb I*uIBL\$LP0L\$I+u ICLP0E1HD$HCHD$H2H(H0H8H@(IL9l$tjH8LH0HD$CH8H@H8Pt8H(IHR8HcR H0L9l$uH|$IFL%l LM?跎L=( H QP A;+ 1LLAI苎H QP AH= @9 MI.uIFLP0M I,$AC3H) qE1E1HHQ HxjjjHH H H8u H@HP0H uH} L}LHD$H IH H8HD$H| HPHHHT$ 踎IH\ H@HT$ HHLHt$IH? HD$H5 HPHt$ HHT$([Ht$ HH* H@HT$(HHHHt$HH H@H;x :LCM-HSIHH+WLHLD$ HT$HT$LD$ HI I(HuI@LT$LP0LT$H+uHCLT$HP0LT$I*u IBLP0耎1H|$HD$'|I(IE(I0HH9\$tWI0LHD$AAUIEtA8I(HH@8Hc@ I0H9\$uH|$ɆIFL%F HH yL= H QP A; 1LLINH QP AH= @9M I.uIFLP0M I,$u ID$LP0HEIHPHUDH@LP01fICLT$LP0LT$GHEHP0PIBLP0H}:fIB@=3H\$0E1H)Hl$8L|$@Ht0IBL@@ uMZLT$W HLALT$HHOHcK- H*  H E11fIH+E11Mt I*MtI/uIGL\$LP0L\$MCI+9ICLP0*fDH@HP0LfHCHT$ HLD$P0LD$HT$ DIELT$LP0LT$I*uD @9LT$LT$@$pf.I6HuHH= LHGHH HtHeH LH5IH81~E1E1H I H | H } s H=uM5b Mt!I$LE1HHEHu HEHP0MtImu IELP0HtH+u HCHP0HD$HdH+%( HXL[]A\A]A^A_fDusIU0I0I;0HI0IU0I0k'HP0H;0HHP0H8H0H0@#HcI0ID-H+(H@(HI0Hx(H;(}HcHI(I|(I0DI(I+0IE0IE(I0HhIHLH G H  I.IE1w@y:H@(H8H(H)0H8HcHHx(H;(}HHx(H8H(H0uH@0H8H@(H8H(H+0H0=IJLLT$SLT$HIGH F H { I.uIFLT$LL\$P0L\$LT$IE1M|E1K@H LHl$8H)DLT$Ht0H\$0L|$@dLT$HHHtH+tpI/HIGLT$LP0LT$/DIEk@$8fIEL\$LHT$P0L\$LT${DHCLT$HP0LT$wH=8FLT$莃LT$YMH ZEE1E11H #   IHI3E1IBL\$LP0L\$LT$LT$HuH H5EH8݀LT$rH=EI.u IFLP0IE1f.HD[ HX F H+E1E1E1HCLT$HP0LT$fH H8aH2D H  <I.E11IFLT$LP0HLT$E1HIgE1 H C HE1H y k fDLT$ HD$yL\$LT$ HufH= HGHH LL\$LT$ HtI)H LLT$H5`BL\$H81辄LT$L\$f.I1ۺ*@LLT$ L\$քLT$ L\$I@HB{ Hx f I+sE1E1E11[@H HH8}I:1f.IZHMjHIEI*u IBLP0IEMAZ蓀@$#f.IE1 L% L- ID$LM@L= H QP A;1LLAIH QP AH= @9MhLI,$u ID$LP0I"E1ZHA۸ Hظ Ƹ HE1H LH8{I5E11LLLT$LT$HH-fDH1 LH8{7E1E1aHI.u IFLP0I1v@1LLsI.IE1E1LH=@~/MHӸI1LL I.I~HH| H5@H8{MIҸD ~@$'Hx?B H? - [PLLT$ L\$DLT$ L\$H1LALT$HL1LPIHIHݸ,}HuH{ H5?H8zDH=q?|H{ H5t?H8z/BzH [ Q I1ۋ5B H=HB~I*uE1E1(1E1E1I/u$E1E1&IE1E1E1E1H  E1H=A5̵ O~f.AWAVAUIATIUSHXHt$D$L$dH%(HD$H1H;{ ZL=S H=, HIWL|IHI HIFH5 HH LIM I.L5 H=ɴ IVL]|HH HHCH5z HHHIM H+u HCHP0IG11H; z A H;'{ H;{ IcIT$ {T$ HIH HtHXHcƒHEHcIlMtIGHH LL$ zHRz LL$ H Q;P  1LL$ LLHzLL$ H QP =H/9WH I)I/H}H` uH} IUL}H, HD$ HHT$(Hg{IH H@HT$(HHLLIH IUH˳ HHHT$({IH/ H@HT$(HHfLLIH H@H;7x 1MiM$IQIEHI)HLHT$(uHT$(HHE ImIuIEHT$(LP0LL$(I)u IALP0H+u HCHP0Q{1H|$ I~+@L$D$LHD$AHH9\$ uLsIFL%F LM xHx H QP ;O 1LLAIvxH QP H= @9kM I.uIFLP0M I,$u ID$LP0HEIHPHUHEH+HEHP0@IFLP0 L=ɱ LqLL yHHH@HHHLLHHL=t MuLLxHH H@HHHLLHHH@H;u ULkMHLsIEIH+u HCHP0LL IH ImLu IELP0H+u HCHP0I/u IGLP0 yLL$D$IHD$LfI~qHEL-" LM uvHu H QP ;1LHAIJvH QP H= @9WMHmuHEHP0MI,$u ID$LP0fIntIH@HD$HdH+%(S HXL[]A\A]A^A_fHEHP0BIGLP0H},fIALP0I/f. @9 IAHT$(LP0HT$(LL$ &uLL$ @$HEsfDHhxIfIG@=H\$0E1Hl$8H)Lt$@IWHt0HBR uMOHLHHH5 H  HLLL$(+LL$(HHH 5` H a S I.uIFLL$LP0LL$I)u IALP0IDHIHH (5 H  Hmu HEHP0H+Au HCHP0MMUI.K11IaILHl$8H)DH\$0Ht0Lt$@HH&Ht H+I.xIFLP0iDH#sHuHH= LHGHHHtIHr LH5#3H81uwH 3 H 5 H=7tMeILE1HH3S HP > yI.u IFLP0H / % M5 fDL@uIrHuHH= LHGHH7HtHHq LH52H81~tH2 H  |I/uVE1IGLP0MtI.u>IFLE1P0HH+u HCHP0H A 7 5- E1p@$f.p@$fH!2 H ֩ ~H+I/tMj1JfD1.HCHP0ksI_HIWHHI/uIGHT$ LP0HT$ HBIAfDH p HH8~lIp@H91 H  HI/3[fDHo HH8lH0 H  I.IFLIP0R1LLLL$ qLL$ HHDHy0L@ H= + fDHn LH8fkE1X@H=0LL$ /nLL$ LL$nLL$HpHm H50H8kLL$PHin LH8jH/y Hv d Hmu HEHP0E1Hq/E18 H5 # `fD1LLpI.I}nH'I.u IFLP0I"%H=/m Hmu HEHP0IE11LHoHmI2#H=S/ln|mHuHk H5I/H8ajLIѸMLH@. H  HpH1LHHk H5.H8iiInMfff.AWAVAUATIUSHHhL-Q H=* H4$HL$IULLD$LL$dH%(HD$X1L;%k lIHHIGH5 HHrLHHqI/0kIHiI$L`$mIHc L% H=j IT$LkIHHI@H5 HHLD$ LLD$ IMI( H5~ LL{mI.y HELMjLAj p VP A;L\$ LLHAIjL\$ p VP AH= 9 MHmS I/) Im I<$ L-\ H=5 IULjIH-HICH5 HHL\$ LL\$ IMI+ IGAE1E1H5jh Ht$ H9H;i  H;i Ic L\$(}iL\$(HHMtLXHt$IcHHtHt$AEHHHtAEHI$LdIGLM%iLh p VP A;L\$1HLAIhL\$p VP AH= @9MHmu HEHP0I/u IGLP0I>u IFLP0IFH5 LHHIMID$H5١ LHH-IMLLeHHI/u IGLP0Imu IELP0H;-g H;-jg u H;-g  DHmu HEHP0EfID$L- HD$HD$LLxLhHHH@HHy HHt$LHHHD$H56 LxHt$L}hHt$HI H@HHELHt$LIHH@H;D$  MEM IUIHImuIEHT$LLD$P0LD$HT$LHLD$HT$轷HT$LD$HI'I(I Imu IELP0I/u IGLP0hM~E1HD$MI8HH0I0H0 H$HD$1BAFIF ,wH(H0H0H@(A9v~MHcI H0H@H0Pt8gH(HR8HcR H0A9vIM93H|$`HEL- HHAeLd p VP A;L$1LHIeL$X SP AH=N 9 M7HmHEHP0MH&& H ۝ !H ܝ ҝ H=)5 DfIGLP0fdIH HICH5ȝ HH L\$ LL\$ IMF I++ID$1E1H5b Ht$ H9 H;c  H;d HcL\$(cL\$(HI1MtLXHt$HcHItHt$EHHItID$HHhcLb p VP A;L\$1LLI8cL\$p VP AH= @9 MzImu IELP0I,$u ID$LP0I>u IFLP0L% H=ś IT$LXcIHI HIGH5 HHSLIM+I/uIGLD$LP0LD$IFLD$LH5ϛ HHLD$IMLD$ bLD$HILLxcLD$HIL% H= IT$LwbLD$HIMHICH5 HHLLD$LL\$L\$LD$HHI+H5 HLLD$cLD$HmI@HHTLD$+aL` LD$p VA;P L\$LLLLD$I`L\$LD$p VP A=p9 MI(ImI/I<$ID$LP0I@LP0dIFLP0xID$LP0 IELP0I<$fIGLP0ImfHEHP0I/f. @9kICLP0ICLP0_@$HD$IcLL\$0L\$(HD$8HD$Ld$HHD$@H)DHt0L\$(HIp M~I+tICLP0eIGLP0kIELP0RI@LP084@I@LP0UH& H  I+- E1E1E11fDICLP0+|Y@$dH H  I.ME1E1YHuJH=# LHGHH? HIHH,Y LH5H81[E1E1 IEX@$HD$LL\$0L\$(HD$8HD$HD$@H)Ht0(L\$(HI@H H  ME1E11M E1LLD$ [LD$ I HuC H@ 2 I(uE1I@L$LP0L$MuE1|fLLHrZIHH E1E1WHuGH= LHGHHIM HW LH5H81`ZHE1x Hu c !W@$,H=1L\$ VL\$ C?DKWHuGH=׏ LHGHH5IMHV LH5LH81YHE1Ǐ Hď  VHHT H5H8SLL\$ YL\$ I " ^LL\$ YL\$ IM\$MIl$IHEI,$HEILD$ULD$@$E1E1I/M_MIoIHEI/HEIAAUHuGH=J LHGHHIMHVU LH5H81"XHs= H: ( sHT LH8iQH: H  Hݍ Hڍ ȍ b WIHME1 H  E11E1:LWI1 d WIHT H8PH^( H%  Hm-E1E1H& H ی B )1LLAVIHME11E1k1E1CVLD$I f 61HLUIH[ P E1E1H=!L\$wRL\$CSHuHWQ H5H8(PfD1H=L\$RL\$BRHHP H5H8O1>H- L- HEHH\+RLQ H‹@ B A;L$1HLHQL$X SP AH= @90HH萨Hmu HEHP0Hڊ H׊ Ŋ x eLD$QLD$HuTH=V HGHHLLD$IMrH]Q LH5L$H81%TL$1ID$L\$(LIP0HEL\$(H6 H  I(I+IE1HIIGL\$(LAIP0HEL\$(ALLD$L\$SLD$L\$H1LHRHmIN?H=0L$OL$Hm+HEHP0LLLLD$~RLD$HI1PHuHN H5H8LH=L\$LD$NLD$L\$UHIM H p HEHHE'1qL$vOL$HLHM H5;H8SLL$-URIE1E@RHH) H އ 4 ,N@$QIQIH H  H1LL\$IQIH1LL\$IwH=0L$ML$H\& H#  t 1H1LwPHHf NHuHL H5H8JLLD$PLD$I7E1E1~I/AE11jIm1nSJAWAVAUIATIUSHXH4$D$L$T$dH%(HD$H1H;L eL= H=ׅ HIWLhMIHd HIFH5 HH5 LIM I.L5 H=t IVLMHH HHCH5% HHHIM H+u HCHP0IG11H;J A! H;K ,H;-L IcdT$ KT$ HIc HtHXHcƒHEHcIlMtIGHH LL$ KHJ LL$ H Q;P  1LL$ LLHXKLL$ H QP =HJ9rH I)I/H}H uH} IUL}Hׄ HD$ HHT$(HLIH6 H@HT$(HHLLIH IUHv HHHT$(KIHJ H@HT$(HHLLIH/ H@H;H LMiM?IQIEHI)0HLHT$( HT$(HHX ImIuIEHT$(LP0LL$(I)u IALP0H+u HCHP0K1H|$ I~3T$L$LD$H$AHH9\$ uLDIFL% LM DIHH H QP ;Z 1LLAIIH QP H= @9~M I.uIFLP0M I,$u ID$LP0HEIHPHUHEH6HEHP0'IFLP0L=i LqLLIHHH@HHHLLHHL= MuLL]IHH H@HHHLLHH H@H;F eLkMXLsIEIH+u HCHP0LLIH ImLu IELP0H+u HCHP0I/u IGLP0ILT$L$ID$H$LfI~`BHEL- LM# GHF H QP ;1LHAIFH QP H= @9bM HmuHEHP0MI,$u ID$LP0fInEIHHD$HdH+%(S HXL[]A\A]A^A_f.HEHP0'IGLP0H}fIALP0I/f. @9 IAHT$(LP0HT$(LL$ ELL$ @$vHEcfDHHIKIG@=H\$0E1Hl$8H)Lt$@IWHt0HBR uMOHLHHH|F~ %HC~ 1~ "HLLL$(軘LL$(HHK"H &} H } } (I.uIFL $LP0L $I)u IALP0IHHIH}!H } H } u} !Hmu HEHP0H+Au HCHP0MMMI.C11IFILHl$8H)DH\$0Ht0Lt$@AHHHt H+I.]IFLP0NDH}CHuHH=?| LHGHHHtIiHJC LH5H81F!%H ]+| H (| 5| H=\DM]ILE1HH{ %H{ { !I.u IFLP0H { { M5{ fDLEIBHuHH=7{ LHGHH7HtHHBB LH5H81EH_){ %H&{ { !I/uVE1IGLP0MtI.u>IFLE1P0HH+u HCHP0H z z 5z E1sA@$tf.[A@$fH{z %Hxz fz !H+I/tMb1JfD1.HCHP0k\DI_HIWHHI/uIGHT$ LP0HT$ HBIAfDH@ HH8=I9"(p@Hy %Hy ~y "HI/3[fDH1@ HH8y ,y ;"I.IFLIP0R1LLLL$ nBLL$ HHDH Lx %Hx x "fDH? LH8;k!!E1X@H=iLL$ >LL$ L $?L $HrH= H5LH8dH'I.u IFLP0I"(-H=B=Hmu HEHP0!!E11LH@HmI/ H=F=n>HuH(< H5H8:LIѸH"MLz! Hv %Hv v !H#@H1LHH; H5VH8n:$:IvMn@AWAVAUATUSHHhLfdH%(HD$X1HD$@HD$HHIHnL~ HD$(H;-:< HD$0HD$8b7L`pLhxHLMtI$MtIEMtIL v H=0u IQLLL$HqE18c H5c #c 5Hmu HEHP0MtI,$u ID$LP0H b b 5b LLLzIHHE1b Hb b 5y@[)@$wfHHHHFHHD$#H|Ll$:fDLH5rc MkHt$(L=`( Ht$X SA;P U1HA@H!a Ha a 5f(HuGH=wa LHGHHHHH( LH5H81O+5H da H aa 5Sa fD (@$gfHa+a H(a a 6HmI/fDIGLP0+HA1HL$IHL cH59* \H {5` H ` ` {53HEHP0q*I5Ht$HLl$L|$vIH9HY#` H ` ` 5Hm{I/@'HH#% H5H8#H=%e_H !IH4H5` HHV&HHD$IEf.1LH(IHuHa+_ H(_ _ 6fH=D%f. &HuH$ H5H8"1H4(IcH=Ht$$Ht$IHH^ H~^ l^ 5YHZ^ HW^ E^ 521ɺAI."AWAVAUATIUSHHxL-] H=] H$Ht$HL$IULLD$ LL$(HD$dH%(HD$h1L;%;$  %IHSHIGH5u] HHLHHI/R $IHI$L`%IH L%#] H=\ IT$L$IHHI@H5] HHtLD$0LLD$0IMsI( H5] LL &I. HELMb#L" p VP A;!L\$0LLHAI0#L\$0p VP AH= 9 M{Hm I/k Im@ I<$% L-[ H=[ IUL[#IHHICH5(\ HH]L\$0LL\$0IM3I+K IGAE1E1H5 Ht$0H9H;" H;r" Ic L\$8"L\$8HH9MtLXHt$IcHHtHt$ AEHHHtHt$(AEHHHtAEHI$LdIGLMb!L! p VP A;rL\$1HLAIr!L\$p VP AH= @9XMHHmu HEHP0I/u IGLP0I>u IFLP0IFH5Z LHHVIM!ID$H5WZ LHHIM|LLHHEI/u IGLP0Imu IELP0H;-B H;- u H;- M DHmu HEHP0EID$L-Z HD$HD$LLxLQ!HH H@HH HHt$LHHHD$H5Y LxHt$ L Ht$ HIH@HHLHt$LIHpH@H;D$0 MEM IUIHImuIEHT$ LLD$P0LD$HT$ LHLD$ HT$;pHT$LD$ HII(I Imu IELP0I/u IGLP0"!M~E1HD$MDI@HH0I8H0I0H0 HD$HD$1BAFIF )t@H(H0H0H@(A9v~MHcI H0H@H0Pt8H(HR8HcR H0A9vIM9#H|$HEL-X HHsL" p VP A;uL\$1LHIL\$X SP AH= 9$MHmHEHP0MH^V CH[V IV ~$H JV @V H=5/V DIGLP0fkIH HICH58V HHDL\$0LL\$0IM I+KID$1E1H5  Ht$0H9>H;( g H; Hc7L\$8 L\$8HIMtLXHt$HcHItHt$ EHHItHt$(EHHItID$HHL6 p VP A;L\$1LLIL\$p VP AH= @9 MImu IELP0I,$u ID$LP0I>u IFLP0L%IT H="T IT$LIH HIGH5 T HHLIMxI/uIGLD$LP0LD$IFLD$LH5,T HH5LD$IMLD$jLD$HILxLD$HIL%hS H=AS IT$LLD$HIHICH5S HHLD$ LL\$L\$LD$ HHI+H5FS HLLD$>LD$HmI@HHLD$L LD$p VA;P L\$ LLLLD$IML\$ LD$p VP A=p9E MI(ImI/I<$ID$LP0fDI@LP0*fIFLP06ID$LP0IELP0I<$fIGLP0ImfHEHP0I/af. @9)ICLP0ICLP0 @$ID$@=HD$L\$@E1HD$HHD$ HD$PHD$(HD$XIT$H)Ht@HBR uMD$L\$=HLL\$IMM"I+ICLP0 E1E1<r#HwP HtP fP I/u IGLP0MtImu IELP0HtHmu HEHP0H +P !P H=5P MtI,$c1ID$LIP0MtI.u IFLP0HD$hdH+%(JHxL[]A\A]A^A_u[HP0H;0HHP0H0H0H0ImHI<$L^lf.y:H@(H0H(H)0H0HcHLP(L;(}ILP(H0H(H0]H@0H0H@(H0H(H+0H0%E1IG@=HD$IcL\$@E1Ld$`HD$HHD$ HD$PHD$(HD$XH)IWHt@HBR uMGL\$HLL\$IMoH.M =HM M #M6E1E11f.I+MOZLHhIHdHM CHM rM $HmHEH1P0|HuJH=M LHGHH IMMxH LH5zH81H.E1L <HL L h#H L L E1H=m5L FLH$E1E1E1<j#E1<m#H{L MHuL gL "HEHICLD$LP0LD$2H/AŅM>#IELP0 @iHEHP0LD$HD$IcLL\$@L\$8HD$HHD$ Ld$`HD$PHD$(HD$XH)DHt@bL\$8HI M&I+ICLP0 IGLP04IELP0I@LP04@I@LP09"HLK HK  K I+K E1E1E11@ICLP0#@$/HJ <HJ J y#I.ME1E1HuJH=KJ LHGHH` HIHLHT LH5H81 E1E1<t#IEx@$HD$LL\$@L\$8HD$HHD$ HD$PHD$(HD$XH)Ht@F`L\$8HIHI 9HI I #ME1E11M E1LLD$0LD$0I<v#H[I HXI JI I(uE1I@L\$LP0L\$MuE1bLLHIH<{#E1E1i HuGH=H LHGHHIMH LH5 H81pHE1H 9HH sH "1@$H=AL\$0L\$0?D[HuGH=G LHGHH>IMH LH5\H81HE1G =HG G #tHH H5H8 LL\$0L\$0I=#HLL\$0L\$0IM\$MIl$IHEI,$HEILD$ LD$@$E1E1I/M_M\IoIHEI/HEIAA- HuGH=ZF LHGHH IM8Hf LH5H812HMF :HJF 8F 6#SH LH8y HJF CHF E $H#E >HE E #IHME1E 9HE E #E11E1:8#LIX1>#I]H% H8Hn8E CH5E #E $HmE1E1H6E =HD D #1LLQIH2M9$#E11E131E1:;#;LD$I>#1HL IH=#E1E1H=1L\$ L\$pS HuHg H5 H88fD1:=#H=L\$. L\$ HH H5H81:B#6H-D L-,E HEHHg; L H‹@ B A;L\$1HLH L\$X SP AH= @97HH`Hmu HEHP0HB ?HB B #:I#[LD$ LD$HuVH=dB HGHHLLD$IM#Hk LLD$H5H812 LD$1:D#ID$L\$8LIP0HEL\$8HB B :H B A F#I(I+IE1HIpzIGL\$8LAIP0HEL\$8A!LLD$ L\$ LD$ L\$H[1LH HmIH=<L\$L\$mHmHEHP0LLLLD$ LD$HI:K#1HuH% H5H8H=L\$ LD$LD$L\$ HIM@ CH@ z@ $HEHHE1kLD$LD$HJH H5CH8[LD$*\ IE1$G HH0? =H? ? #@$ I IH? 9H? ? #H1LL\$I IH1LL\$IHH=7L\$L\$Ha+? ?H(? ? #H1L|HHfDHuH' H5H8LLD$LD$I1E1E1>I/AjE11*Im61.[ff.AWAVIAUATIUSHHH$dH%(HD$81H; 3L= > H== HIWLvIHJ HIEH5= HH LIM ImIG1E1H;'  H;E WH; HcL$>L$HI& MtLhHcHEH IlAHHITIGHH' LL$LL$H QH [ P ;x 1HL$LLLL$HLL$H QHL$P =H99?HX I)I/u IGLP0H}u HEHP0H^< uH} IVL}H5*= HD$HHT$Ht$cHt$HIB H@HT$HHLLIH$ IVH5< HHt$HT$Ht$HI' H@HT$HHLLIH H@H;' MPMMpIII(&LLLT$fSLT$HIU I*MuIBLL$LLt$P0LD$LL$I(uI@LL$LP0LL$I)u IALP0/E1H|$HD$~ $LKIL9t$uH|$IEL%Z< HHZ H QH P ;] H $1LLIWH $X SP H= @9@M ImuIELP0M I,$u ID$LP0HEIHPHUHEH?HEHP00IELP02L=: LiLLHHH@HHHLLHHM~H5P: LHt$Ht$HIsH@HHLLLIHWH@H; MuMIUIHImuIEHT$LP0HT$HLHT$PHT$HIg I.Iu IFLP0Imu IELP0I/u IGLP0L$ILIHEL%%: HHHH QH P ;`H $1LHIH $X SP H= @9MvHmuHEHP0M-I,$u ID$LP0LIHf.HD$8dH+%(HHL[]A\A]A^A_f.IG@=HD Ll$ E1Hl$(HD$0HcH)IGHt L@@ uMOHLAHH,Hc-7 H*7 7 0MtImDI/Au IGLP0H 6 6 56 H=SMIELE1HIALP0 @9LL$HLL$@$f.I@LT$LP0LT$H LHl$(Ll$ HD$0HcH)Ht LHHMOImDIELP05HE3fDLLD$sPLD$HIt?1H޽5 H5 5 ImI(BIMfDI)|IALP0mIE)fDLOIHNx0HP5 H5  5 Hmu HEHP0ImM IIEkfDHuGH=g4 LHGHHHIMHs LH5ܻH81?0H H=M4 E1H G4 594 W@$fLHI-H13 H3 3 0Imu IELP0H 3 3 E1H=u53 ;fDMoMIWIEHI/t.HBI׹f.+@$/IGHT$LP0HT$@I@L $LP0L $MIDIELP0H H8QI-1H 2 H 2 52 DH H8H2 H2 2 /1Imu IELP0H 2 }2 5s2 H=#E1HEH3HyC2 H@2 .2 0MI/F"fD1LLLL$vLL$HH-DHA1 H1 1 1I/FIGL $L1P0L $xDH=yHL$LL$LL$HL$`fDL $L $HjH H5LH8dL $KH LH8yf0E1#H H8XH)0 H0 0 h0HmHEHP0@HImu IELP0H ʸ10 H 0 u0 11LLImIH=)H $H $u0,MM<1SH=H $@H $Hmu HEHP00o1LH7HmIHuH H5H8MIոu0SHm/ Hj/ X/ 0MWJf{IH1LAHI6Ha H5H82SIELL$LL$P0LL$L$SAWAVIAUMATUSHHXHt$dH%(HD$H1H; L=|. H=U. HIWLIH HID$H5:. HHRLIM I,$u ID$LP0IG1E1H; AcH; & H; IcT$T$HIMMtL`HcHEIlHu HcIDHIGHH5L\$[L\$H QH P ;1HL$LLL\$H"L\$H QHL$P =HO9wHI+I/H}H , L}L1HIH H8H, Mt$uH} I9iIUH5^- HHt$HT$Ht$HIH@HT$HHLLIHIUH5, HHt$HT$AHt$HI(H@HT$HH LLIH H@H;`  MjM IRIEHI*uIBHT$LP0HT$HLHT$CHT$HIImIuIEL\$LHT$P0L\$LT$I*uIBL\$LP0L\$I+u ICLP0VE1HD$ID$HD$H5f.H(H0I$8H@(IL9l$tkI$8HH0HD$KI$8H@I$8Pt8HH(IHR8HcR H0L9l$uH|$uIFL-+ HH%H QH P ;HL$1LLIHL$X SP H= @9 MI.uIFLP0MImA;,fDH) qE1E1HHQ HxjjjHH H H8u H@HP0HD) uH} LHD$HEHD$H) IH H8IUL%) HLHT$ IH H@HT$ HHLLIH IUH5) HHt$ HT$(Ht$ HI H@HT$(HHLLIH H@H; Ml$M}IT$IEHI,$tHLHT$ $@HT$ HI ImIuIELT$ LP0LT$ I,$uID$LT$ LP0LT$ I*u IBLP0E1H|$I$|@I(IF(I0IL9d$tZI0HHD$HL$JAVIFtA8I(IH@8Hc@ I0L9d$uL@IGL%( HHE H QH [ P ;G HL$1LLIHL$X SP H= @9iM I/uIGLP0M ImAu IELP0HEIHPHUHEHHEHP0vDH@LP0&HEHP0BIGLP0H},fIG@=H Ld$0E1Hl$8HD$@H)Ht0IGLH@ uMW HLAHHHME1E1H% % % 2f.Mt I,$ZM E1E1I*uIBL\$LP0L\$MtI+u ICLP0H ;% 1% 5'% f.H@LP06ICLP0I/ @9 ID$HT$ LP0HT$ rfDL\$vL\$@$qI9HuHH=7$ LHGHH HtIHB LH5H81E1E12H O$ H $ 5 $ H=ҰMMtI,$u ID$LP0MtI.u IFLP0HD$HdH+%( HXL[]A\A]A^A_DIV0I0I;0gHI0IV0I0@HP0H;0HHP0I$8H0H0&ILE1HfDKHcI0I.H+(H@(HI0 Hx(H;(}HcHI(I|(I0Dy<H@(I$8H(H)0[I$8HcHHx(H;(}HHx(I$8H(H0I(I+0IF0IF(I0HHHHH?L { HLHL@HH SHH5ٝH81WH6m _ 6HD$8H[ XZHL$(H= E1H$dH+%(1HĸL[]A\A]A^A_@Hy HD$"HEHP0fHGP0!@I$Ld$xZHD$L-d Hh LLuLIHH@HHLLHIHHD$L- Hh LLuLKHHH@HHLHHHqH@H;t  HGH$HHoHHEH/uHGP0L$MLH+HD$xH!Imu IELP0HDŽ$Hmu HEHP0H|$xH/uHGP0HD$xHD$HxID$L-r HHH QH ` P ;"1HL$0LLHD$(LD$(H QHL$0P =Hq9IM"I,$I(H\$ HtHHD$HHu HCHP0HL$HtHHD$HHu HAHP0M[I/QIGLP0H HIĐHHmHEHP0DH5) 1HH$IH# H;| L;%" u L;%D DI,$7HDŽ$EHD$L-N Lp LIVHHT$(IHZH@HT$(HHLLIH?HD$L- Lp LIVHHT$()HH=H@HT$(HHLHD$xHH)H@HDŽ$H;D$8HGH$HHWHHHT$xH/uHGP0L$H|$xMLb(H$HImu IELP0HDŽ$H|$xH/uHGP0H$HD$xH/uHGP0HDŽ$L@pLHxILMtIMtIMtIHE(HEHHHLT$@LL$8LD$0L\$( HL\$(LD$0LL$8LT$@II HD$LLT$8LL$0HpLD$(LD$(LL$0LT$8Mt$I(uI@LT$0LLL$(P0LT$0LL$(MtI)uIALT$(LP0LT$(MtI*u IBLP0ID$L- LM H QH P ; 1HL$0LLAHD$(RLD$(H QHL$0P =H @9MI,$uID$LD$(LP0LD$(MI(wI@LP0hfDD#Hq HIDHHHHFHH$FHUH$HD$I$QfDI$$fDIT$LR0 @D1LLI,$IufID$LD$(LP0LD$(M]fHAH$1 HD$(H Hʙ !7HD$0Ht H/uHGP0H|$xHDŽ$Ht H/uHGP0H$HD$xHt H/uHGP0H{XH5 HDŽ$|H ] S 5I H|$0H$HT$xHH$ rL% H= IT$LgIH HIEH5 HHLIM[Imu IELP0H HD$8I9D$CHt$Lv#IHzI,$u ID$LP0IEH5e LHHIMlImu IELP0L5 H= IVLIH}HIEH5 HHLIMImu IELP0zIHLpHD$L\$HIH H5y HL\$L\$ID$LML\$L\$H QH ~ P ;HL$@LLLL\$AIL\$H QHL$@P =H @9MI,$uID$L\$LP0L\$I+u ICLP0I.u IFLP0L;- eL H@ I9B6LnIHJLֺLLT$ LT$HII*uHD$IBLP0L\$H5C 1LL\$L\$HILHL\$kL\$HI I+u ICLP0I.u IFLP0ID$H5 LHH/IMI,$u ID$LP0HD$8I9FTMfMGIVI$HI.uIFHT$LP0HT$HLHT$ HT$HIYI,$IuID$LT$LP0LT$I.uIFLT$LP0LT$L;w L; L;; LLT$LT$ABI*u IBLP0EWIEH5| LHHIMLT$LT$HIH  H5" HLT$mLT$IBH5 LMbHt$@LT$LT$Ht$@H QH  P ;HL$@LLLT$AIyLT$H QHL$@P =H @9MI*u IBLP0I.u IFLP0L;% RImu IELP0HD$L5 L@ LIPLD$@HHT$IH H@HT$LD$@HHALLIHHD$L5z L@ LIPLD$@HHT$IHH@HT$LD$@LMLLAIHH@H;D$8mMCM`MsIII+uICLD$LP0LD$LLLD$LD$HIx!I(MuI@LT$LLt$8P0LT$L\$8I+uICLT$LP0LT$I*u IBLP0ID$ HL$It$HHyWIEL5T LM8LD$BLD$H QH P ;HL$1LLAI H QHL$P H= @9MImuIELP0M[I.u IFLP0H$H/uHGP0H|$xHDŽ$H/uHGP0H$HD$xH/uHGP0HD$ H{pHDŽ$LsxLLHCpHD$HCxHt H/uHGP0MtI.u IFLP0MtImu IELP0H MHIImIELP0fDLAŅHH$ HD$(H H 7HD$0H H5H8L\$(LD$0LL$8LT$@LT$@ILL$8LD$0L\$({L\$(LD$0HLL$8LT$@H$HW! HD$(H  7Ht5H/u/HGLT$HLL$@LD$8L\$0P0LT$HLL$@LD$8L\$0HH  LT$P 5 HLL$HLD$@L\$8HD$0HDŽ$H|$8H$HT$xH$OL\$8LD$@LL$HLT$P_HT$xLT$`H$H$LL$XLD$PHD$@H1IL\$HHT$8L\$HLD$PHLL$XLT$`I1HLLT$`LL$XLD$PL\$H-I,$L\$HLD$PLL$XHLT$`u8HD$PID$LLT$hLL$`LD$XP0LT$hLL$`LD$XHT$PL\$HImufDE1H8IH! HD$(H  D7@Il$HID$HEHH$I,$u ID$LP0L$I@H;  H;z LD$(LD$(HH$IHhHD$HID$ I@HHLD$(ܿLD$(H QH B P ;HL$(1LLH訿H QHL$(P =, 29e HHl$xI,$u ID$LP0HDŽ$HH$s HD$(Hk H<R R7HD$0pHqH$3 HD$(H+ H |7HD$00H船IH?H55 HHVIHxH$ID$fDDmNEUHH I2~7~LD$(FLD$(@$LD$(&LD$(@$IHaH$# HD$(H H 7HD$0 DO HuHH= LHGHH HtIH LH5 ~H81nHD$(E1E1E1 E1H| j R8DHD$(E1E1O A 8HB DL\$L\$@$0H LH8^H/~H$ HD$(H H 7HD$0fDH LH8H}H$1 HD$(H H`v 6HD$0@HD$(ME1E1T E1HN < T8L`ITLIH HD$(E1E1  8H hfH LH8HD$x7H LH8H| H | 6I,$HD$(ID$LMT$MM\$III,$uID$LT$HLL\$@P0L\$@LT$HICH;  H; LT$HL\$@蝺L\$@LT$HHIULPHD$HIF ICLM~L\$hL\$H QH ι P ;A1HL$@LLL\$AI.L\$H QHL$@P =H @9MI.MIFLP0@H9HD$(E1E1E1 H  b8afHD$(ME1E1 E1H  8.ӼIF賹HuHH=? LHGHH HtIPHJ LH5yH81HD$(E1E1E1* H'  8L8IHD$(ME1  8E1H Qf77fD6fDHD$(E1E1 y 8Hz DIPBፁHL$H$1H$LJ uIxH$QAHD$xHHm?HEHP00DHD$(E1E1  8H DD1LL#I,$IP`H=yyHL$(϶HL$(I,$u ID$LP0HxH$S HD$(HK H}2 8HD$0P@H=xHL$(THL$(I,$ID$LP0HuH H5xH8H:H H5xH8Ƴ@LLLL\$L\$HIHD$(E1E1Y K 8HL @$H Hs I}H9~HXH HqH~1H;TWHH9uH HJH5|HWH81蜸HD$(ME1E1 E1E1H  8H=iwHL$@L\$躴L\$HL$@-L\$yL\$HH H5=wH8UL\$HD$(E1E1E1 H  8yLL\$L\$HIM޸9HL$(  H ImIELT$LME1P0LT$E1 HD$(ME1r d 8E1Hb @$HD$(E1E16 ( 8H) HD$(E1E1  8H xLT$诳LT$@$UHD$(E1  8H 8HD$H$LH$H$HD$xHrHtl HD$(Hd R X7xDISBፁ7HL$L$1H$LB uI{LT$@H$L\$ ALT$@L\$IMB I*uIBL\$LP0L\$M1+IEHD$(E1E1  8H 觵IIHD$(E1E1E1J HG 5 8H5 H= 1胩IHHHD$L\$I+u ICLP0HD$(E1E1E1 E1H  8CIKHD$(E1E1E1 H  8HD$(E1E1p b 8Hc L\$L\$@$fDL1L裳H$IH,f.˰@$HD$LH$L$LT$@L\$H$.L\$LT$@HIHD$(ME1  h8E1H Hqu HD$(Hm [ f7yHbH H5BrH8ZG`e1LL莲HD$xH\H1qH$1 HD$(H Hu l7HD$0HD$(IE1  8H "LLLT$LT$HIHD$(E1E1i [ 8H\ H=0qHL$0膮LD$(HL$05HD$xAHuHU H5qH8&H HI|$H9HXHHqH~1H;TjHH9uH HJH5tHWH81gHD$(E1E1E1{ Hx f 8LT$LT$HH H5CpH8[LT$H= pHL$HHt$@LT$ULT$Ht$@HL$H*wH@`HHHLT$@HLL$8LD$0L\$(L\$(LD$0HLL$8LT$@IHL I9Fu`LLT$@LL$8LD$0L\$(JI.L\$(LD$0LL$8ILT$@IVLR0LT$@LL$8LD$0L\$(LH5tL\$(LD$0HLL$8LT$@IqHˬ LH8@ HD$( 8H ME1E1E1E1;H LH88E1E1ME1E1 路HD$(ME1i [ v8E1HY 1LL跮ImI.1LLL\$蒮L\$HIHD$(ME1E1 E1H  |8_H=mHL$LD$ LD$HL$SImu IELP0 HD$( *9H L\$蘫L\$HYH H5\mH8tL\$9H=#mHL$@L\$tL\$HL$@ 8HWHH H5mH8< HD$( 7H LLLLH$H H5pH8M¸9H*EHHH9HuH; e,HD$(E1E1E10 H-  8LT$@LL$8LD$0L\$(%L\$(LD$0HLL$8LT$@mH H5&j> HD$( 7H LT$`LL$XLD$PL\$HWHT$8Ht$@LHBHD$(H|$HHDŽ$HL$`HT$XHD$xHt$PHDŽ$HA 3 % 7xH$;HT H5nH8%iH!j HD$(H  `71ɺAHHH9\HuH;N JHD$(ME1 t p8E1Hr 1ɺAL\$LT$@IkC HD$(0 7H1 X HD$( 7H 3AVAUATUHHGH5v HHHHHEH;% u{LeMtrLuI$IHmtNLLsIHI,$L1Hmu HEHP0HL]A\A]A^fHEHP0@H;Ѧ 3H;, HMQE1 LiL%2 H QP A;$1LAI藦H QP A$H= @9M0ĦHHg %H q K=HmHEHP0H ] S 5I kID$LP0Lu!ۥ@$`fH 1g;=% %H  ;=H=-mE1UHL]A\A]A^fDHHf %H  H=I.txI,$u ID$LP0H o e 5[ zfDLL5z Mt~L%q P P A;$u1LHA;f.IFLP0y11HTIMoHH=f,1LH*IH=fwHբ H5fH8覡AVAUIATUH(H5 dH%(HD$1HGHHHHH H9E\LuMOLeII$HmID$H; =H; @膣HHLpIELh ID$LMcL5Ԣ H QP A;"1HLAI7H QP AH=9MHmYI,$u ID$LP0Imu IELP0H HHT$dH+%(7H(]A\A]A^HEHP0ID$H;r LHIHIhfIT$BፁL4$1Ll$LB uI|$HAIMHqc; (H8 & =fDI,$u ID$LP0I.u IFLP0H  5 fDHEHP0f @9m}@$_@HLLl$L4$IHI.@IFLP01f蛤HH*H yb=(9 (H 6 $ =H=h蠢1fH1b (H  =f1HLCIH{Ha (H  =I,$uID$LP0@HmwHEHP0hfDH=Ab蜟fkHuH H58bH8PkHIa (H  =yH"a (H  =1ɺAIo補AVAUIH=n ATL% USIT$LHHHHEH5 HHHIMHmID$H5 LHH HHI,$IEH5 LHH IMHL I9EMuMI]IHImLHIHI.IImu IELP0'H^H HhHHP L`([]A\A]A^fDHEHP0fID$LP0 IELP0[LIHj>H _ H  +HmblfIFLP0˝HuGH=W LHGHHHHHc LH5]H81/H^J +HG 5 =H 6 , H=c5 螞[1]A\A]A^H1^ +H  >HmtI,$uID$LP0HEHP0I,$zDH؟I==H ] H y +Hm5E1E1HEHP0MtImu IELP0McfDHY]# +H  =1;H>[fDIMIݸ>H@AWAVAUIATUSHHL% H=] dH%(HD$81IT$LIHHIGH55 HH LIM I/L% H= IT$LIHS HIGH5 HH LHH I/9IF1E1H;1  H;O H; HcLD$GLD$HI MtL@HN HcՃHcHIDI\IFHH Hw H QP ;6 1LLHܙH QP H=c9iH I/8I.u IFLP0H}u HEHP0Mu L%l I^LH譚IH1 H@HHmLHLIH I] L% LsLL\IHP H@HH,LLHIH7 H@H; DMpM7IXIHI(qLHIH* I.IIFH\$LP0LD$@IGLP0gIGLP0IF@=HL LD$ E1H\$0HD$(H)Ht IFLX@ uMVLD$ HLALD$HH\HHQE1V HS A 9I.?1\L`H<HAQME1E1H   9I(u I@LP0I.u IFLP0HH+HCHP0MFMInIHEI.HEIfDH9 LH8讋I9@LLD$ L)LD$Ht Hl$(H\$0LD$HINMtI(u I@LP0H+HCHP0IFLD$LIP0HELD$fDMIظ9HO H  I/1I(u I@LP0IH1LALD$H$H=Pt @I/u IGLP0I;:fDHɍ LH8>HO H  9I/u IGLP0I@HN H  9M#E1D3@$rIGLD$L1P0LD$1LL諏HH=HINM1 H  9@H=N,fHuH H5NH8fDӌHuGH=_ LHGHHIMHk LH5LH817HMIO HL : Q:TDHYM1! H  S:LLD$+LD$IfD1LLKI/IfۋHuGH=g LHGHH IMsHs LH5KH81?HLIW HT B V:JD[H Hk H5$MH8<H1L H  X:qfLLD$LD$HfDMD$MMt$III,$IFMA9fD:-HKK HH 6 :>fHQK H  z:fH!K H  :f1LL3IHMM清:gHJ H n :|H=IK褈.f.kHuH H58KH8P{HIJ H  9ID$LD$LAMP0IFLD$d*HI H  j:9ċI跋IH1LALD$IBI.uE1uMIIH+VE1uAWAVAUIATIUSHdH%(HD$x1HFH; HD$hHD$pH;c 5H@hH( H@H 1LHHl$`H| H5d HHD$`ւD$ , IT$HBpH" H@H LH5 HD$`HH H@L5 L9 HWH HGHW HD$hHT$pHHD$pHH/uHGP0HD$`L|$pIGIGHH5 H:OHcH>H~HnHEAGD$|$ I/u IGLP0HD$hLHD$pHD$hH$H` HIT$HBpH H@Hw LH5 HD$`HH H@L9mHWH'HGHW HD$pHT$hHHD$hHH/HD$`HD$pH\$hHD$pHD$hHD$OLXpLPxILMtIMtIMtIH LT$ L\$L(hE1H<$AAHƹAL\$LT$ HHD$`I' HD$hH8uH@LT$ LL\$P0LT$ L\$HD$`HD$hMt I+Mt I*Mt I/ID$ H8pM} H5 IWHt$HHT$ LHt$HI[H@HT$ HHLLIH=M} H5 IWHt$HHT$ Ht$HHLH@HT$ HHLHD$`HH5H@HD$hH; HGHD$hHHWHHHT$`H/L|$hH|$`MhL+HD$pH I/u IGLP0HD$hH|$`H/uHGP0HD$`H|$pH/uHGP0HD$p}LxpL@xILMtIMtIMtIIt$IMLL$(LD$ LT$HHyHHHxHxH)H)΁HIEL$H|$~LT$LD$ LL$(IEH H9CHLL$ LD$LT$LL$ LD$LT$f.JzIEMt$I/uIGLL$LLD$P0LL$LD$MtI(uI@LL$LP0LL$MtI)u IALP0IFL= LM?*H QH P ;HL$1LLAIH QHL$P H= @9M I.uIFLP0MImu IELP0L-H IEI,$u ID$LP0HtHmu HEHP0H $HtHH$HHu HAHP0Ht$HtHH$HHu HFHP0HtH+u HCHP0HD$xdH+%(}HĈL[]A\A]A^A_fDH~uHFH(HEa@IGLP0rIBLP0SICLT$LP0LT$*HGP0@HGP0>@AG؉D$!AGAWHH HHcЉD$H9H~ H5hGH8x|sD$HHc@H|$`( H%  _;1HEH$1E1HD$HD$HH/ HGHT$P0HT$AGAWHH HcЉD$H996@1I}IH]HL{I.HIFLP0fDHq HH HD$HHA} HRH5EH81HD$`HH$1H   ;fDH|$`HD$pHq<fDHH|$`f.H=H|$` H  ;lL|HcЉD$H9H}|HL% L5 ID$LMB{H Q{ X SP ;rH $1LLAI{H $X SP =g@9 MQLd$`LCI,$u ID$LP0H<1E11H HBHD$`n ` ;HD$H$HD$xDH;a{ H<$2}H<$HHaH/uH$HGP0H$HBH$HHD$`LAH$HHD$hHHAH$HHD$p|HA׾H2zH$H*HBHP0DHy HRH5BH81<}HD$`H;N HK 9 ;K@HFHx3HHmIH  HEHx H5AH81|H|$`H; H޲ ̲ 7;H: H  l;H|$`1H+@1E1HD$HD$;y@$`f2HWHHWHHRHD$h4Dn<H T: H   I.u IFLP0H?H|$`1HD$DHw H5@H81{H|$`H|$pH9 HD$0H  ;Ht!H/uHGLT$ L\$P0LT$ L\$HD$pH|$hHD$`Ht!H/uHGLT$ L\$P0LT$ L\$I~XH5n LT$ L\$HD$h0L\$LT$ T H>H LT$(߰ 5հ HL\$ HD$KyHL$pHT$`LHt$hpL\$ LT$(n Hc LT$8L\$(H(HL$ hE1H<$AAHƹHD$ L\$(LT$8H H8u(HPLT$(HL\$ HD$R0LT$(L\$ HD$H|$hH/u%HWLT$(L\$ HD$R0LT$(L\$ HD$HD$hH|$`H/u%HWLT$(L\$ HD$R0LT$(L\$ HD$HD$`H|$pH/u%HWLT$(L\$ HD$R0LT$(L\$ HD$I~pMfxHD$pMM^pMVxMHt!H/uHWLD$ HD$R0LD$ HD$Mt&I,$uIT$LD$ LHD$R0LD$ HD$IMsI(iI@LP0ZfL-1 L=b IELM" iuH QH t P ; HL$1LLAI9uH QHL$P =N29;Mg Ll$pLImu IELP0HM6H|$`1HD$pH H;  L<HD$fHt H8!qH5H|$`1 H H>; _<HD$HYt H8pHD$`a<;fDH;t H|$QvH|$HH H/uHD$HGP0HT$HBHT$HHD$`LAHT$HHD$p} HAHT$HHD$h HAԾHKsHT$ H*HBHP0DHHx3HHBH \ HEH\r H5M;H813vH|$`H4I HF 4 ;LT$D$BsLT$T$HLD$ LL$( <H|$hH 4 HL$0H ޫ Ы Ht+H/u%HGLL$ LD$LT$P0LL$ LD$LT$H49H LL$( 5 HLD$ LT$HD$HD$hsH|$HL$hHT$`Ht$p0kLT$LD$ LL$( HD$pHL$hHT$`HD$HH1HL$8HT$@uLT$LD$ HLL$(I8 IFLL$PLD$(LM$ L\$ LT$qLT$L\$ H LD$(LL$PQH p P ;Y 1HL$XLLLL$PLD$(LT$ L\$AI,qL\$LT$ H LD$(LL$PQHL$XP =H @9M I.u2IFLL$PLLD$(L\$ LT$P0LL$PLD$(L\$ LT$I+u(ICLL$(LLD$ LT$P0LL$(LD$ LT$ML;-Ap L;-o L;-p LLL$(LD$ LT$hqImLT$LD$ LL$(Au(IELL$(LLD$ LT$P0LL$(LD$ LT$E\EHL$HHHD$HHu(HALL$ HLD$LT$P0LL$ LD$LT$HD$pHL$@HHD$HHu(HALL$ HLD$LT$P0LL$ LD$LT$HD$`HL$8HHD$HHu(HALL$ HLD$LT$P0LL$ LD$LT$IzpMrxHD$hMMzpMBxMHt H/uHGP0MtI.u IFLP0MDLL$(LD$ LT$D$nLT$D$HLD$ LL$(!<kn@$fH=y0HL$mHL$0fI.u IFLP0H/H|$`1T HQ H48 <HD$BfD1LLpI.I@DnHrH+l H5/H8jWHWH"HWHHRHD$pDH44HD$I~pMnxM^pIMVxMHtH/uHGHT$P0HT$MtImuIEHT$LP0HT$HH*u HBHP0H|$`1)Dl@$Hk H54H81oH|$`L1LFoHD$`IH H-H$H   ;H=o.H $kH $rHD$`lHuHj H5W.H8oiH|$`E1LH@`HHHLHHHk H9Cu*DHH+D$zHCHP0kHH52rHHugH,H|$` H  T;vL1LmHD$pIH)H,H|$`1R HO H16 H<HD$@& HD$0 <H sH=,HL$>jHL$`HD$pujHuHi H5,H8gH+H|$` H  L;E1H*u HBHP0hH+H|$`[ HX F \;1A3 HD$0 <H! H=+H|$` H  ;QImD}IH 9LH  H51HEHh H81{l1H*H|$` H w ;E1H*u HBHP0gH{*E HB 0 ;A봸<HL$0Izp MjxI H MzpMBxMHtH/uHGHT$P0HT$MtImuIEHT$LP0HT$H|$`HiLL$PLD$(L\$ LT$DhLL$PLD$(L\$ LT$@$7hHHZg H5(H8ceL6fLL$(LD$ LT$gHL$8HT$@Ht$HHbLT$LD$ HD$pHD$`LL$(<HD$h<<d<L1LLT$ L\$jL\$LT$ LD$(LL$PIULL$PLD$(L\$ LT$gLT$L\$ HLD$(LL$PItPE1H=)HL$XLL$PLD$(L\$ LT$ZfLT$L\$ LD$(LL$PHL$XaH0e H5(H8dLL$PLD$(L\$ LT$AUATIUSHHHf HndH%(HD$1H$MHHHH+H +HHHH5H?L $ HLHL@HH9e UH,H5j'H81hH 9'Xo>H  Z . o>.H=.E1[gHD$dH+%(pHL[]A\A]fHVHk H5ae HEH{H蠱IHtHHmuHEHP0DHHHFLH$_HdH$fHa&+ XH(  >Hmt'H   H=3-5 xfHEHP01HL +IH5g LvH %a> .H  a>fL_IH4H5e LHVdHtH$IE 2afAWAVAUATUSHHLnH|$8dH%(H$1Hc HDŽ$H$H$HR HD$(H$H I  I I H!c HD$0Lk LsHDŽ$HDŽ$HDŽ$IIEL;-b  L=C H= IWLcHH HHCH$H5e HH=HHH$HH+ L$HSa HDŽ$H9EHt$(L訳HHH$H/ f.HCH5 HHHZ HH$HTH+u HCHP0L= H= IWLbIHfHHLL$_ÅI,$z HDŽ$nL=? H= IWLfHF0H$HC(H$HC HH$HCH$RIIW IIu-MH5 HHVWHtH$IM~q1H$MHL H5Z V yJL CV? L   V?fDMIuMH$L$L$HD$0H$HD$(fH AL %HHU HKH5AUH817YL Xl?L S ZD 6 l?LɺH=E1WH$dH+%(%HL[]A\A]A^A_@HhPIH5 HIHV.VH$HLkMH&H AHNL $EHkLNOD@UHuHH=' LHGHHHtH H2U LH5H81WHOE11E1HDŽ$H   ?HD$HD$HD$ HD$@H$Ht H/uHGP0H$Ht H/uHGP0MtI,$u ID$LP0H ~ t E1H=,5` UHtHmu HEHP0HtH+u HCHP0HL$HtHHD$HHu HAHP0HL$ HtHHD$HHu HAHP0MtI/u IGLP0Ht$HtHHD$HHu HFHP0HT$HtHHD$HHu HBHP0I.u IFLP0ImTIELP0EDE11ۺ?HD$HD$ HD$H5W% E1H$H5  HD$HIH/%HGP0fHVHHEH$HHUHHH$HmH$L$HIGH;R H;kR  RH$IH_HhHD$(HDŽ$HID$ IGHH-QH QH ;Q P ;HL$1LLHQHL$p VP H= @9 H I,$u ID$LP0H$HDŽ$H/{IWBፁH\$(H$1H$LB uIH$AHHH: H  ?HD$E11H$HD$E1HD$ HD$kP@$fHD$(H$LH$H$֟HHHmu HEHP0H$HDŽ$H/6fDHLUI{DH9E1 H  ?H+u HCHP0HD$E11H$HD$HD$ HD$HH|$PH|$HE1q H$Hf T @HD$뢐HGP0@H|$&PH|$D$HHFE1 H$H  @HD$;f.HEHP05HFHH$DH E111HӅ H$  ?HD$HD$HD$ HD$fDH=Ɉ HOH$HHLH= H6IHH+u HCHP0HDŽ$L趢I,$u ID$LP0H5  H  @D1LLKNHHH E111H E1H$  ?HD$HD$HD$ HD$KHuHH= LHGHHHtIeHK LH5 H81MH7  H  (@H= HL$JHL$DH LE1 H  *@JHHH H5t H8GH;yJ LKLIHI/u IGLP0ID$LHDŽ$LAHHFLAH$HLAH$HLA׾HDII,$ID$LP0H Hx.HHH \HEH\H H5MH813LH N HK 9 3@@HD$p@|@HD$@Hgf@IA@=HD$HcT$HE1L$L$H$HD$0H$HD$8H@(H$H)IQHĠHBR uMQLL$HL\$8HLLL$HL\$8H$H MtI+uICLL$8LP0LL$8HL$HHD$HHuHALL$HP0LL$DHDŽ$H$rH$H4A@H5 H5 Ht$@ HHD$HHu H|$@HGP0HD$H$E1FH$IHHEH=n HHhrJH$IHI,$uHD$ID$LP0LD$H=S LLD$辗H$IHLD$I(u I@LP0HDŽ$L1I,$u ID$LP0@HDŽ$>@@(D@IWHIGHPHH@H$fDHD$(HL%~ H=~ H$IT$LTFHHHHGH$H5 HHH|$0H|$0IMH/uHGP0H|$(L:BH$IHI,$uHD$0ID$LP0LL$0L; D L; D L; D LLL$0FLL$0I)uIQD$0LR0D$0HDŽ$ L%} H=} IT$L0EHHHHGH$H5ŀ HHH|$0H|$0IMH/uHGP0H|$(LAH$IH I,$uHD$0ID$LP0LL$0L; C L; eC L; C wLLL$0DLL$0I)uIQD$0LR0D$0HDŽ$g H=| K:H$HHH@H5 H|$0HHH|$0IMH/uHGP0H|$(L@H$IHSI,$uHD$0ID$LP0LL$0L; B L; TB L; rB LLL$0CLL$0_I)uIQD$0LR0D$0HL$(HDŽ$H1HVHt$0HHuH|$(D$0HWR0D$0HDŽ$HD$(H5@ HH$H9pAHt$H|$(,H$HH$H/uHGP0L$HDŽ$HDŽ$LAH$IHHEH=~ HHhEH$IHI,$uHD$ID$LP0LD$H=z LLD$fH$IHLD$I(u I@LP0HDŽ$LٗI,$u ID$LP0@HDŽ$@@fLL$8Ht$@LL$8Ht$@$Hy Hy y ?HD$HcL$HLL$L\$pH$HD$0L$H$HD$8LL$8H@(H$H)HĠ赏LL$8L\$pHH$MtI+uICLL$8LP0LL$8Ht$HHD$HHHFLL$HP0LL$H> H5H81BcL%'y H5{ ID$LM+LD$(Ht$T?H > LD$(p V;Ht$P HL$1LAI?H QHL$P =V29 M]L$L讕I,$u ID$LP0@HDŽ$H> H82;HE1w H$Hw w AHD$LL$(]>LL$(@$A/HN> H8:HDŽ$AAL yCw L @w .w )AH$HDŽ$HtH/uHGLL$(P0LL$(H$HDŽ$HtH/uHGLL$(P0LL$(H v v H=LL$(5v HDŽ$&?H|$`H$H$H$d6LL$(H$L$H$HD$8HL1HT$0ALL$(HIHD$@H@LM LL$pLT$HL\$(HD$Ht$(HHQHL$0HHHDŽ$HHgID$`H|$PHD$YMYMIAIHH$IHD$HHIL$D$HIAE BHt$hr HD$`L r r HxpL@xHppHt$PLHpxHt$XHHtH/uHGLD$(P0LD$(MtI(u I@LP0H$MtI,$ID$LE1P0H$NH=HL$8LL$0Ht$("8Ht$(LL$0HL$8HL$@HHD$(HHu H|$@HGP0Hq Hq q IBH$E1HA;LL$pLT$HL\$(8LL$pLT$H@$L\$(1ɺAHE1@HD$HD$ HD$@L p L p p KAMtTI+uNICLL$(LP0LL$(HD$H$HPH/FHGLL$(P0LL$(0HD$H|Fp HCp 1p @T1LLLL$9LL$HH$HD$L $L o o o \AI,$?ID$LL$(LP0LL$(%H=HL$H5LD$LL$8HL$HHDŽ$w6HuH4 H5kH83͐9H|$@18Ht$@IHHD$(HH5@$H|$ L\$pHGP0L$D$HL\$pIAx6H H4 H5H82Hn Hn n W@LL$(D5HL$8LL$(HxXLX`HHXHL$0HLPhL`hHH`t+H/u%HGLL$8LT$0L\$(P0LL$8LT$0L\$(Mt$I+uICLL$0LLT$(P0LL$0LT$(MtI*uIBLL$(LP0LL$(HDŽ$BHDŽ$HDŽ$ HD$x@HH HHD$x@HH HB@3 BHJm HGm 5m M@XB@H5Em E1H5 m l HD$)H|$@L1LL$HL\$(P6L\$(LL$HIV@E1I,$u ID$LP01t;Hl Hl wl _@HE1~AIHLH .H5*HEH'2 H816H= HL$xLL$pLT$HL\$(S2L\$(LT$HLL$pHL$xE1w 3HuGH=k LHGHHHHH2 LH5 H81n5L HDŽ$L k vk hk xALL$HL\$(|2L\$(LL$HHRHD$HH}0 H56LL$pH8I/LL$pLT$HL\$(@A?L j L j j zA=H|$04H|$0I @L j L j j }A@BL13H$IH4H{E1Bj H$H7j %j @HD$pL @ j L j i ;A9L i L i i A:0H%HDŽ$XH=HL$0/Ht$LD$(HL$0HD$@\~1/HD$}H1LL\$8LL$H!-L b,i L )i i AL`HMH@I$HH$HHD$(HHu HFHP0L$IAH;_/ H;/ MLL$(U/LL$(HH$I~L`HD$1LLLD$(HI@ &LD$(HH$I(u I@LP0HDŽ$L Sh L h h DALH|$012H|$0If/HuGH=g LHGHHHHH. LH5H81s1L HDŽ$L g {g mg AL [g L Xg Fg AL j4g L 1g g AAL C g L g f Ad&1H,L f L f f AL f L f f A 0H|$0IL yf L vf df AIA@=HD$H$LL$H$X$H$HI,$ID$LP0HD$H$LL$H$2|H$HuL e L e e AH* H5H8)L ne L ke Ye AL }Ge L De 2e ATL V e L e  e AwL /d L d d AHD$E1E1H$HD$HD$ HD$.HAWAVIAUIATUSHH|$xHL$8LD$(dH%(H$1HHIL%4d H= d HDŽ$IT$LHDŽ$+IHHIGL$H5f HHeLIMI/ *H$IH<IELh,H$HHH) H5df H,|ID$HH"*H ) p VP ;-HL$HLLHD$)HL$X SP H= @9H|$-I,$u ID$LP0I/u IGLP0HDŽ$Hmu HEHP0ImHDŽ$u IELP0H|$H5de HGHH@,IMJ,H5c 1LLD$LD$HH$I+-I( L;=( L;=h( u L;=( I/HDŽ$ #HhpL`xHLHtHEMtI$MtIEL=c H=pa IWL)IH1HI@H5b HH3LD$LLD$IL$M63I(u I@LP0H|$H5d HGHHDHH*YHy& H$H9GEHGH$H,LOHIH/t$L$M^,LLLL$LT$xLL$HI+JLT$I*N%HDŽ$I)uIALD$LP0LD$H$I9GnEH$LLD$!xLD$HH$kFI(#H$H/#L$HDŽ$HDŽ$Ht Hm$Mt I,$$Mt Im$H5` L[#H$HHFH; & H;-% u H;-% $Hm $HDŽ$IHD$(H;% H&HHH$H$HHaJL%^ H=^ HDŽ$IT$LI&HHJHHCH$H5fa HHbLHHD$HD$H$HKH+$L%e^ H=>^ HDŽ$IT$L%HH$HHCH5` HH8UHIM|TH+u HCHP0L-] H=] IUL_%HHUHHCH5|^ HHVHIMVH+uHCLL$HP0LL$H # H$I9D$XLLLL$_uLL$HH$H(\I)u IALP0I,$u ID$LP0L$H5_ IELHH]IMp\Imu IELP0HD$L$HDŽ$H$H9X^LLtH$HHOI,$u ID$LP0H$H/uHGP0L%g\ H=@\ HDŽ$H$IT$L#HD$HdHHD$H5^ H$H@HHgH|$HD$HD$H$HgHT$HHD$HHu HBHP0Ht$H|$(HDŽ$AăhHt$HHD$HHu HFHP0HDŽ$EEIHH[ L(hH|$(E1AAHƹAHD$H$H+tH|$H?uHGP0Ht$(HDŽ$HHD$HHu HFHP0HD$H;`! uH@H|$HHwHt$(H5] HwIMvH5] LǺLD$ JLD$ HHD$H$6xI(u I@LP0Ht$H;5 H;5 FH;5 wFH)"AąPyHT$HHD$HHu HBHP0HDŽ$E!|H|$H5Y HGHHHD$HD$H$HˀH|$LZIHHT$HHD$HHuHBLD$HP0LD$L; HDŽ$L; PL; PLLD$"!LD$AI(u I@LP0EH=X HD$H$H͐H|$H5;[ HGHHIMڏHt$HHD$HHuHFLL$HP0LL$IALL$ LH5Z HH LL$ HD$HD$H$HI)u IALP0H5KY H|$1IHHD$H$HDŽ$H9p HD$H@H$HHt$HVHHH$HHD$HHuHFLL$HP0LL$L$L$MIEH; H;o {tLL$ LL$HIL`1HLLH HDŽ$HD$L\$HI!I+uHD$ICLP0LD$H$H/uHGLD$P0LD$L;b HDŽ$AL; DwcL; jcLLD$LD$AI(u I@LP0EHEtHEHHHHcH>IGLP0GI@LP0H|$H5X HGHH%IL$MZ&H5X LvH$HH&I/u IGLP0H;-+ HDŽ$H;- u H;- QHmu HEHP0HDŽ$8H|$H5}U HGHH):HH$H#;HEH;( #H; H@hH(8H@H81HIL$M3;Hmu HEHP0L9=U HDŽ$HDŽ$:H5+W H=T 1H$HHZxHqHmu HEHP0HiE1E1E1H1T Ll$11HDŽ$E1H$T aS EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$@IGLP0HDŽ$ZrDLHÅ=HoE1E1E1H7S Ll$11 S TE1E1H$S CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$HLLHD$H@HE1E1E1HaR 11E1LR SH$6R CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$fHt1H/u+HGL$LD$xL\$P0L$LD$xL\$Mt6I,$u/ID$L$LLD$xL\$P0L$LD$xL\$H$Ht1H/u+HGL$LD$xL\$P0L$LD$xL\$Mt$I(uI@LL$xLL\$P0LL$xL\$H$Ht!H/uHGLL$xL\$P0LL$xL\$MtI)uIAL\$LP0L\$MtI+u ICLP0H P P E1H=l5qP MtI/u IGLP0HtHmu HEHP0HtH+u HCHP0HL$HtHHD$HHu HAHP0H\$HtHHD$HHu HCHP0Ht$ HtHHD$HHu HFHP0HT$0HtHHD$HHu HBHP0HL$pHtHHD$HHu HAHP0H\$PHtHHD$HHu HCHP0Ht$hHtHHD$HHu HFHP0HT$@HtHHD$HHu HBHP0HL$HHtHHD$HHu HAHP0H\$XHtHHD$HHu HCHP0Ht$`HtHHD$HHu HFHP0HT$8HtHHD$HHu HBHP0Imu IELP0MtI.u IFLP0HL$(HtHHD$HHu HAHP0H$dH+%(iHL[]A\A]A^A_ÐH!E1E1E1HM 11E1M SH$M CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$D @$:f[HuHH=L LHGHH4HtIH LH5[H81HE1E1E1HDŽ$11HL L SL CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$fHaE1E1E1H)L 11E1L SH$K CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$DLIHE1E1E1HiK 11WK SIK CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$FHE1E1E1HJ 1E1H$J SJ CHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$g11IL;5 +L%I H=I IT$L`HD$H^CHHD$H5"L H$H@HHEH|$IMEHT$HHD$HHuHBLD$HP0LD$LD$*LD$HHD$H$jIHD$ILD$LpLD$HIvJL%I H=H LD$0HD$IT$LrL\$LD$0HHD$ kPHHD$ H5"K H@HHSLD$@H|$ L\$L\$LD$@HD$0HD$0H$H3RHT$ HHD$HHuHBLD$ HL\$P0LD$ L\$HT$0H5H LLD$ L\$L\$LD$ MHT$0HHD$HHuHBLD$ HL\$P0LD$ L\$I@LMULD$ L\$L\$LD$ p H  V;P YLHL$0LHt$L\$ LD$AI^HL$0LD$P L\$ P =229<M6YL$I(uI@L\$LP0L\$Ht$HHD$HHuHFL\$HP0L\$HDŽ$I+u ICLP0I.u IFLP0HDŽ$Ht$8H;5e H;5 XH;5) KHK<H|$(H;= g.HGH5H HHfQHD$H|$PH HT$HDŽ$H$H9B:HBH$H:LZHIHHD$HHuHBL\$HP0L\$L$M9LLL\$]L\$HHD$H$Y]I,$uID$L\$LP0L\$HDŽ$I+u ICLP0L$HLLT$ƹLT$HHD$ H$VHt$ LLT$ LT$HHD$$XHT$ HHD$HHuHBLT$HP0LT$HDŽ$I*u IBLP0H|$xH5F HGHHZHD$ HD$ H$HZHT$ H$HDŽ$H9B3PHBHH$HPHRHHH$HHD$HHu HFHP0H$L$HD$ MOHD$ H@H; RH; *< IHBjL`H|$ 1HILp HDŽ$HD$LD$HHD$ mI(u I@LP0H$H/uHGP0H|$H5E HDŽ$HGHHcIM cL\$ L\$HHD$0H$cHD$ HT$0L\$HHBN L\$HIhHD H5D HL\$8HD$ LD$L\$8RHt$0LLLD$8L\$L\$LD$8HHD$@H$qI+uICLD$LP0LD$HT$0HHD$HHuHBLD$HP0LD$HDŽ$I(u I@LP0H=A HD$PH$HwH|$PH5D HGHHbwIMvHT$PHHD$HHuHBLD$HP0LD$LD$8LD$HHD$PH$HD$@Ht$PLD$HHF LD$HHD$hH$~Ho H|$hLD$H5C Y LD$eHT$hHt$PLLD$ULD$HHD$0zI(u I@LP0HT$PHHD$HHu HBHP0Ht$hHDŽ$HHD$HHu HFHP0HT$@HDŽ$HHD$HHu HBHP0HD$pHD$`HD$XHD$HHD$@HD$hHD$PL;5 LT$0EH|$H5B HGHHkIM8kH5@ 1LLL$5LL$HHD$8H$gkI)u IALP0Ht$8H;5 H;5 7H;5 7HTAą.lHt$8HHD$HHu HFHP0HDŽ$ENL;5 KH|$0H5A HGHHHD$8HD$8H$HՍH5? H|$817IHhHT$8HHD$HHuHBLL$HP0LL$L;  HDŽ$L;  WL;  WLLL$?LL$AI)u IALP0EJH== IHIH5= HHD$ZLL$HH$IўI)u IALP0,IHbH5= H|$HD$ LL$HIH5= LHLL$HD$8LL$L\$8I+uICLL$LP0LL$H5> LLLL$LL$HHD$8I,$uID$LL$LP0LL$HDŽ$I)u IALP0H|$Ht$0HGH@pH(H@HIMȡH5= H|$8LL\$L\$іI+u ICLP0HD$8Lt$MLl$HIDHGLL$P0LL$vf.HGP0%@I@LP0HÅH/E1E1E1H; Ll$11; \E1E1H$; DHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$HEHP0fHD$IBLP0LD$LL$fIELP0RID$LP01HEHP0L-> IEI.IFLP0HÅGHE1E1E1H: Ll$11: ZE1H$o: DHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$6fDHCHP0eH}%L}IL$H.H=s9 LHGHH4mHX.HLL$HHD$I,$uID$LL$LP0LL$HD$HDŽ$HD$hI)u IALP0L$Ll$1HD$`HD$XHD$HHD$L|$(H$L$Ld$ H$HHt$H|$p1IHTH ŅI,$u ID$LP0H5' H|$xuIHҎHt$pH|$H$HHH$I9GMgMMoI$IEI/u IGLP0IEH;6H;1H$IHsL`1HLHh HDŽ$IHȏI/u IGLP0HDŽ$Lt$@Imu IELP0HtH+u HCHP0H5 & H|$pIHЎHÅI,$u ID$LP0HD$hH@HXpHVH{K13IHHHt$pHSI,$Hu ID$LP0HtH|$hHSHmIu HEHP0MLH.% H$LF{I,$u ID$LP0H=# pIHqH5% H%H$IHImu IELP0H$I9FH$H$Y;IH߮H$H/uHGP0H\$HDŽ$HtHHD$HHu HCHP0HL膗IHHLHD$H$H*I,$u ID$LP0I.u IFLP0H5$ H|$H$IHUHIHHD$@HID$H$IHHv$ H5g$ HKLLLHHI.u IFLP0HDŽ$I,$u ID$LP0I/u IGLP0H\$HHDŽ$HtHHD$0HHu HCHP0H=! BIHH5o# HH$IH+Imu IELP0IHjHEHhIHHH5" HlLLLvH$IHII.u IFLP0HDŽ$Imu IELP0I,$u ID$LP0IGH;@IHIGI_ H$H$HHI/u IGLP0HDŽ$HT$XHtHHD$0HHu HBHP0Ht$`HtHHD$0HHu HFHP0H5! HoIHͳH$I9EMeMM}I$IImu IELP0LL7H$HI,$u ID$LP0I/u IGLP0H$H/uHGP0H5$! HHDŽ$IHH$I9GMgMMoI$IEI/u IGLP0IEH;H;|H$IHƐL`1HLHHX H$HI/u IGLP0HDŽ$Imu IELP0H$HmHD$Hu HEHP0H5G H|$HH$IHގH|$pHIHI.u IFLP0Lt$hHDŽ$IFL`pMI|$vHKH|$pLHHHT$HHLAT$HmAu HEHP0E{Imu IELP0H5m H|$HIHH|$pHH$IHImu IELP0HT$pHHD$0HHu HBHP0H$H\$`Lt$pH\$@HDŽ$HD$XyL=  X mDb1iIHHHI,$IID$LP0@H|$HMfDH- L%  HEHH8H rP P ;9HL$1HLHHL$X SP =29WH8H$Hc9Hmu HEHP0HHDŽ$H  ] DHD$pE111HD$8E1E1Ll$HD$`E1HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$fDI,H|$&HfDMOMIGIHH$I/uIGLD$LLL$P0LL$LD$L$IBH;(0H;:LD$ LL$LT$LT$LL$HH$LD$ IHLH1HLL@ H$HLI/u IGLP0HDŽ$ٹf.H _H  DfL= WL=  *DHDŽ$HE1E1E1Hv Ll$11_ _E1E1H$C DHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$ HE1E1E1H Ll$E11 ZH$ DHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$\W H|$xH5 HGHHA;IME;L\$L\$HHD$H$=H HT$L\$HHBILz L\$HHD$ H$GBH5r H|$ LL\$L\$+HT$ Ht$LL\$L\$HHD$0OI+u ICLP0Ht$HHD$HHu HFHP0HT$ HDŽ$HHD$HHu HBHP0HDŽ$HD$pHD$`HD$XHD$HHD$@HD$hHD$PHD$ HD$`DD&HE1E1E1HQ Ll$11: dE1H$! 'EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$H5i H= 1H$HH_=H3Hmu HEHP0HE1E1E1H_ Ll$11HDŽ$E1H$1 [# DHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$@9HМE1E1E1H Ll$E11 dH$j (EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$1HuQH= LHGHH@HtHD铹HLH5H81vHǛE1E1E1H Ll$HDŽ$p fb 4EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$HE1E1E1H Ll$E11 fH$ 6EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$H|HD$閳H= HD$H$H?H|$H5 HGHHaMIMLHT$HHD$HHuHBL\$HP0L\$H|$(L\$H5 HGHHFL\$HD$ HD$ H$HEH=Z L\$L\$HIQH@L\$LH5 HHQL\$HD$HD$H$HPI,$uID$L\$LP0L\$ICE1E1H;$GVH;rf?H;Hcj'L\$jL\$HIXMtL`HT$ IcHt$LLL$ ITAE1HL\$ItLHDŽ$HDŽ$L\$LL$ HHD$H$gI)uIAL\$LP0L\$I+u ICLP0L$L;%AL;%0DL;%MLAŅ[I,$u ID$LP0HDŽ$EH=v !IH3oH5 HHD$L\$HIhnI+uHD$ICLP0LL$H= LL$LL$HHD$H$mH5 H|$LL$ dLL$ HHD$H$lHT$HHD$HHuHBLL$HP0LL$H5 H|$(LL$ LL$HHD$ H$kHD$H$H9POHt$ H$LL$@&LL$HHD$Ht$ HHD$HHuHFLL$HP0LL$HDŽ$H$H/uHGLL$P0LL$H5 H|$LL$ BLL$ HHD$H$܌HT$HHD$HHuHBLL$HP0LL$H$I9AHt$LLL$ Y%LL$ HHD$H$ăHt$HHD$HHuHFLL$HP0LL$HDŽ$MI,$u ID$LP0HL$HLHD$H$HH|$AŅHT$HHD$HHu HBHP0HDŽ$EJnI$MH+u HCHP0I,$u ID$LP0L$I}u IELP0H+u HCHP0HDŽ$L驰H|$H|$@$zL\$LL\$&L\$HHD$H$dH E1E1E1Lt$MH$H Ll$  GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ |@MH*LH5H81HGE1E1E1H Ll$E1 fH$ 9EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$骸LD$ L\$+LD$ L\$@$HyE1E1E1Lt$E1MH$H. Ll$ ~ GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$ݷHΑIE1E1H Ll$1 fH$k ;EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$2H*I HuQH= LHGHHIHtH̪D鐯HLH5H81gHE1E1E1H Ll$m fH$W >EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$L= WL=  DDeHD$ H@@=H|$ H$L$L$HD$ H!kI,$u ID$LP0HDŽ$HtIpH]IE1! fH Ll$1 @EH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$ĴHuJH=; LHGHHPHt HD$mHDLH5H81HaE1E1E1H) Ll$E1HDŽ$H$ y GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ Lt$ųP@$MD$MMl$IIEI,$uID$LL$ LLD$P0LD$LL$ IEH;CH;-&LL$ LD$LD$LL$ HI!VL@1HLLH HD$CL\$HH$HUI+MgICLP0XH|$IHE1E1E1H Ll$r yH$\ GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$Lt$'H|$xH5 HGHHOHD$H|$NHT$H HDŽ$H9B?HBH$H?LJHIHHD$HHuHBLL$HP0LL$L$Mg?IAH;@H;3:$LL$LL$HIVSL`1LHHDŽ$ILx LL$HD$ >LL$L\$ HHD$H$kRI+uICLL$LP0LL$I)u IALP0L$I@LD$HHpH?5HyHL$/5HLHDIH-5HL$H|$HQI,$LD$IuHD$ID$LLD$P0LD$LT$M4I(uI@LT$LP0LT$L;5PHDŽ$7IBLT$LLH5 HHLcLT$XcLT$0HD$pHD$`HD$XHD$HHD$@HD$hHD$PHD$HD$ 5H_E1E1# fH   NEHD$pHD$8Ll$HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$HE1E1E1H Ll$1t fH$^ qEHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$%HE1E1E1H Ll$ yH$ GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ Lt$鉭I3DIB@=H$LL$LL$L$LD$/LD$LL$HH$EbI)IALD$LP0LD$閞HE1E1 yH Ll$ GH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$Lt$OHH@H$H3HSHHH$HHD$HHu HCHP0L$L$MIEH;LH;LD$BLD$HI L@L` IEHDŽ$HHL\$H L\$p V;P 71HL$LLL\$HHL$L\$P P =29HH$I+-ICLP0IE@= H$LL$LD$L$貺LD$HH$HJI(u I@LP0HDŽ$铟@JL\$L\$@$:H$LL$LD$L$[LD$HH$HyHE1E1E1H Ll$ fH$ EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$iLt$Ll$E1E1HJ yH$H  GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$ǨLt$(HD$pHD$`HD$XHD$HHD$@HD$hHD$PHD$ HD$H= LT$LT$HI]DH@LT$0LLL$H5 HHILL$LT$0HD$8HD$8H$HII)uIALT$LP0LT$Ht$8LLT$ԽLT$AFHt$8HHD$HHuHFLT$HP0LT$HDŽ$Eu LT$0EIBLT$LH5 HH8LT$HD$8HD$8H$HA8H5 H|$81LT$QLT$HHD$07Ht$8HHD$HHuHFLT$HP0LT$HDŽ$I*IBLP0醸LD$0L\$蠿L\$LD$0HuhH=" HGHHJLL\$LD$0Ht HD$ BH!LLD$H5L\$H81LD$L\$H*E1E1 yH Ll$ GH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$Lt$頥苾HuJH= LHGHH|QHt HD$H LH5~H81H=E1E1E1H Ll$E1HDŽ$H$ g EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$靤H~LL$ E1P yHM Ll$6 GH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$ HD$Lt$H|$ LD$@L\$LD$@L\$HD$0H}E1E1E1Lt$E1MH$H Ll$m _ GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ 8H|$(.HD$鍮H}E1E1E1H Ll$E1 gH$ EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$遢H|$wHD$$H|$ L HD$ HHC|E1E1E1Lt$E1MH$H Ll$  $HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0鹡H|$Ht$0HGH@pH?H@H?IMfVLt$MLl$HD$8ĢE骜L1蕼HHL=  Y yDE,H{E1E1E1H Ll$E1 gH$ EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$tHt$LLL\$ LD$蚻LD$L\$ HH$ILt$Ll$E1E1Hz yH$H  GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$隟H|$ H$L$L$HD$ HHVyE1E1E1Lt$E1MH$H Ll$  *HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0̞@HD$0Lt$MHD$8Ll$HIHxE1E11H[ Ll$H fH$2 EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$HwE1E1Lt$H MH$ Ll$ WHHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0fHWwE1E1E1Lt$E1MH$H Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0͜踵HvHȳH5wH8虲[@H=IwHL$0LL$ LD$Ht$萴Ht$LD$LL$ HL$0RLD$L\$@L\$LD$H?HDŽ$H=vHL$0+L\$LD$ HL$01LLL\$0L\$HH$HuE1E1E1H Ll$1~ fH$h EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$/*I'HuE1E1E1Lt$E1MH$H Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ LT${HltE1E1E1Lt$MH$H$ Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$әHsE1E1E1Lt$MH$H| Ll$i [ HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$+H=sHL$*L\$HL$HDŽ$L\$ݱL\$HuHH5sH8轮L\$H|$x輴HD$ ٥HrE1E1E1Lt$E1MH$HX Ll$E 7 HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0H$LL$LL$L$LD$LD$LL$HH$L=q WL= u 1DI)uIALD$LP0LD$\IC@=HD$ HLL\$0L$H$HD$H$L)H3L\$0HHD$H$`gMtI,$uID$L\$LP0L\$HT$ HHD$HHuHBL\$HP0L\$Ht$HDŽ$HHD$HHuHFL\$HP0L\$HDŽ$HDŽ$H=kHL$=HL$IE@=pH$LL$LL$L$ߠLL$HI^NI,$uID$LL$ LLD$P0LL$ LD$HDŽ$I)jIALD$LP0LD$QHjjE1E1E1H2 HD$ nLl$ FH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$(HD$ϏLLD$2LD$HD$H$MHHiE1E1E1Lt$MH$HM Ll$: , UIHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HhE1E1E1H HD$E1 nLl$HD$(H$ FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$XHIhE1E1E1Lt$MH$H Ll$  NHHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0H|$踩ICHgE1E1E1Lt$MH$HY Ll$F 8 PHHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$P#IA@=LϺH$LL$L$L$LL$HHD$H$lWI,$uID$LL$LP0LL$HDŽ$IE@=0H$LL$LD$ L$LL$XLL$LD$ HH$HO\I(uI@LL$LP0LL$I)u IALP0M邀H5A H=2 1#HD$H$H4:Ll$L0IEHD$HIEu IELP0HeE1E1E1Hl HD$E1V oLl$HD$(H$HDŽ$* FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$H|$xIHdE1E1E1Lt$MH$H Ll$u g HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$7H(dE1E1E1H Ll$E11 aH$ EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$鉉HzcE1E1E1HB HD$@Lt$MH$ Ll$ sHHD$0HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHbE1E1Lt$H MH$ Ll$} UHHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0_HPbE1E1E1Lt$MH$H Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HaE1E1E1Lt$MH$Hi Ll$V H 8HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ !HaE1E1E1H HD$E1 pLl$ FH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$(HD$tH|$jHD$^~HQ`E1E1E1H Ll$E11 [H$ DHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$鲅H|$訡HD$ 霪艞H}H= LHGHHv4H9HD$'HN_E1E1E1Lt$E1MH$H Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$黄H^E1E1E1Ht HD$a pLl$N FH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(HD$H]E1E1E1Lt$MH$H Ll$  >HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0rHc]E1E1E1Lt$MH$H Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ӂH\E1E1E1Lt$MH$H| Ll$i [ KHD$8酂H|${I_Hd\E1E1E1Lt$MH$H Ll$  K.HLH5P[H81賝H\E1E1E1Lt$E1MH$H Ll$HDŽ$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ e`THL[E1E1E1Lt$E1MH$H Ll$  KHD$8 HZE1E1E1Lt$E1MH$H Ll$  HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$ HD$hHYZE1E1E1H! Ll$E1 hH$ EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$L=Y WL=} k ADHD$ LL\$0L$H$HD$H$L)HL\$0HHD$H$,H5YE1E1 hH Ll$ EH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$~HXE1E1Lt$H^ MH$E Ll$2 XHHD$pHD$8HD$`HD$XHD$HHD$hHD$PHD$0~L$LLLL$LL$HIXHWE1E1 rH HD$ GLl$HD$pH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(HD$D}H5u H=n 1_HD$HW1Ll$LtIEHD$HIEu IELP0HVE1E1E1H HD$E1 qLl$HD$(H$z FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$A|H2VE1E1E1H HD$ pLl$HD$(H$ FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD${L=UI WL=F 4 GD'HPH4H5[H81HE1E1E1Hk HD$E1U sLl$HD$(H$5 FGHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$cXH=ME1E1H H|E1 H$} KLl$HD$8LT$0HD$cH=E1E1E1HQ HD$@Lt$MH$. Ll$ tHHD$pHD$8HD$`HD$XHD$HHD$hHD$PHD$0HD$@cH<E1E1E1Lt$E1MH$H Ll$  CKHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ _bH|$xU~HD$KH<<E1E1E1H Ll$E1 hH$س EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$aH|$)IH~;E1E1E1HF HD$E10 tLl$HD$(H$ ]GHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$`H:ME1E1LT$0Ll$E1E1H Hxk H$U KHD$8HD$z`Hk:E1E1E1H3 Ll$ fH$ EHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$_H9E1E1E1Lt$MH$Hz Ll$g Y LHD$8_Ht9E1E1E1Lt$MH$H, Ll$  KHD$85_H&9E1E1E1Lt$E1MH$H۰ Ll$Ȱ  K^H8ME1E1H HwE1 H$x KLl$LT$0HD$^zLT$0LL$HD$8mHt8E1E1Lt$H: MH$! Ll$ jKHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ ]H7E1E1Lt$H MH$ Ll$r dKHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$B]H37E1E1 fH  kEHD$pH6M1® fH Ll$ eEH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$g\LLD$0L\$UxLD$0L\$ϵHOsH57H8 rL\$LD$H6E1E1֭ rHӭ HD$ 1GLl$HD$pH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(HD$q[Hb5E1E1& rH# HD$ +GLl$HD$pH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(HD$ZH4ME1Lt$Hx ME1H$\ Ll$I RIHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ +ZH|$0!vHD$8lHHOtH3E1E1E1H HD$ tLl$ YGH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(HD$TYHE3Lt$E1MH Ll$H$  qIH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(XH|$tLL$(LD$IhH2E1E1Lt$HI MH$0 Ll$ oIH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$(WH1E1E1Lt$H MH$ Ll$} mIH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(IWH:1E1E1 hH Ll$ EH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$VH0E1E1E1He HD$R tLl$? [GH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$(HD$VH/E1E1E1H H$Lt$MH$ Ll$ hIHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$HD$(_UHP/Lt$E1MH Ll$H$  tIH$HD$(HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$TIE@=Ht$ LL$H$dIHI,$u ID$LP0Hmu HEHP0HDŽ$Lt$@HLIH3Hmu HEHP0HDŽ$MLt$@魀o|H-E1E1E1Lt$MH$H Ll$ z )LSH-E1E1Lt$Hd MH$K Ll$8 vIH$HD$8HD$`HD$XHD$HHD$@HD$hHD$0HD$ HD$HD$( SLL$}H,E1E1E1Lt$E1MH$H Ll$H$  IHD$8HD$`HD$XHD$HHD$@HD$hHD$0HD$ HD$(fRHoiH5(-H8@hH<,E1E1E1Lt$E1MH$H Ll$ޣ У HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ QH+E1E1E1Hb HD$E1L oLl$9 FH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$(HD$PlH*E1E1 iH Ll$ FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$HD$PHN*E1E1E1H Ll$ iH$ FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$OH)E1E1r iHo Ll$X FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$OH)E1E1ݠ iHڠ Ll$à FH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$NHs(E1E1E1H; Ll$( iH$ FHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$MH'E1E1Lt$H MH$w Ll$d LHD$8MH'E1E1E1Lt$MH$H7 Ll$$  LHD$8@MH1'E1E1Lt$H MH$ޞ Ll$˞ LHD$8LH&E1E1E1Lt$MH$H Ll$ } LHD$8LHcH5Y+H8bcgeLT$鬜HI鲑Hb&E1E1E1Lt$E1MH$H Ll$  KLT$0HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$ HD$KH%E1E1E1H HD$E1m qLl$Z FH$HD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$(HD$KH%E1E1E1Lt$MH$HƜ Ll$  ILHD$8JL=$ WL= u :DHHH9AHuH; cmA5Ho$E1E1E1Lt$E1MH$H$ Ll$  2HHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0IHt$ LL$H$IHAHD$ MLl$HL|$(MHT$@E1H#E1L$G HD H$Lt$MH$ IH$HD$8H$HD$0Ll$HD$(!IH#E1E1E1Lt$MH$Hʚ Ll$  'LHH|$I]H"E1E1E1Lt$MH$Hs Ll$` R LHHv"E1E1Lt$H< MH$# Ll$ LHD$8:HID$@=5HD$0LL\$H$L$H$WL\$HHD$P I+u ICLP0HT$0HHD$HHvmHBHP0gmHD$0LL\$H$L$H$誯L\$HHD$0I+u ICLP0Ht$0HHD$HHlHFHP0lIH!HE1Ll$L$HT$@E1HҘ L|$(H$Lt$M Ll$ IH$HD$(H$HD$8H$HD$0HD$ FH E1E1E1HP H$Lt$MH$* Ll$ IHD$8HD$`HD$XHD$HHD$@HD$hHD$0HD$ HD$(EHD$8Ll$HMHE1E1L$H L|$(E1H$Lt$M{ Ll$HT$@H$HD$(H$N IH$HD$0HD$ gEHD$0Ll$HE1L|$(Ht$@E1E1H8E1L$ H H$Lt$MH$ʖ IH$HD$8H$HD$ Ll$HD$(DHHME1Ll$L|$(E1E1L$HT$@Hm H$Lt$MH$G Ll$HD$(H$' IH$HD$8HD$0HD$ 7DIH%E1E1Ll$L$H H$L|$(ȕ Lt$MH$ ILl$HD$(H$HD$8H$HD$0HD$ CHD$ MLl$HL|$(Ht$@E1E1HyE1L$8 H5 H$Lt$MH$ IH$HD$8H$HD$0Ll$HD$(CHE1E1E1Lt$E1MH$H Ll$  %IHD$pHD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ pBHaE1E1% rH" HD$ $GLl$HD$pH$HD$8HD$`HD$XHD$HHD$@HD$hHD$PHD$0HD$ HD$HD$(HD$AHD$ MLl$HL|$(MHt$@E1HE1L$P HM H$Lt$MH$# IH$HD$8H$HD$0Ll$HD$(*AHD$0MLl$HL|$(MHT$@E1HE1L$ H H$Lt$MH$ IH$HD$8H$HD$ Ll$HD$(@HD$ MLl$HL|$(Ht$`E1E1HeE1L$$ H! H$Lt$MH$HD$XLl$H$ KH$H$HD$8HD$0HD$(?IHHE1Ll$L|$(E1E1H H$L$HT$`HD$XH$Lt$MH$Z Ll$G KH$HD$(H$HD$8HD$0HD$ J?ML|$(Ll$H\$H$H$L$HE1E1E1Hِ HD$Ɛ H$HD$`H$Lt$MLl$HD$XH$ KHD$(HD$8HD$0HD$ >H|$HHCH 1Ll$ HDHPHLVH5Ll$L|$(H\$H81H$H$L$YLL$ IH HE1Ll$L|$(E1E1Hۏ H$L$Ht$`HD$XH$Lt$MH$ Ll$ KH$HD$(H$HD$8HD$0HD$ =HD$ Ll$HE1L|$(HT$`E1E1H\E1L$ H H$Lt$MH$HD$XLl$H$܎ KH$H$HD$8HD$0HD$(H5Ll$L|$(H81H$H$L$AMHME1E1Ll$L$E1Hw H$L|$(w Lt$MH$w _JLl$HD$(H$HD$8H$HD$0HD$ %HD$0Ll$ME1H{E1E1L$H>w L|$(H$Lt$Mw Ll$HD$(H$H$H$v ]JHD$8HD$ %IHE1E1Ll$L$Hv H$L|$(v Lt$MH$v XJLl$HD$(H$HD$8H$HD$0HD$ }$HD$ Ll$E1E1H^E1E1L$H!v L|$(H$Lt$Mu Ll$HD$(H$H$H$u VJHD$8HD$0#HLME1Ll$L|$(E1E1L$Ht$Hu H$Lt$MH$eu Ll$Ru IJH$HD$(H$HD$8HD$0HD$ U#IHCLE1Ll$L$HT$E1Ht L|$(H$Lt$Mt Ll$t GJH$HD$(H$HD$8H$HD$0HD$ "HD$ Ll$E1E1HE1E1L$Hbt L|$(H$Lt$M?t Ll$HD$(H$H$H$t JHD$8HD$00"MnMePIFIEHH$I.u IFLP0H$HEH;S:H;:tlR:H$IHLh1LHH$HIG 1IHI/u IGLP0HDŽ$OHE@=uH$Ht$ HL$H$51IHImOIELP0rOH$Ht$ HL$H$IHuHD$ MLl$E1HE1E1L$Hr L|$(H$Lt$M^r Ll$HD$(H$H$H$.r #JHD$8HD$0O HD$ MLl$E1H0E1E1L$Hq L|$(H$Lt$Mq Ll$HD$(H$H$H$q +JHD$8HD$0HD$ Ll$E1E1HE1E1L$Heq L|$(H$Lt$MBq Ll$HD$(H$H$H$q 7JHD$8HD$03HD$0MLl$E1HE1E1L$Hp L|$(H$Lt$Mp Ll$HD$(H$H$H$p 1JHD$8HD$ IE@=oQHt$ LL$H$n.H$HI,$QID$LP0QHt$ LL$H$PH$HuHD$ MLl$HHHME1Ho L|$(E1H$HT$`L$Ht$HHD$XH$Lt$MH$oo Ll$HD$(H$Oo JH$HD$8HD$0hHD$ MLl$HHIHME1Ho L|$(E1H$HT$`L$Ht$HHD$XH$Lt$MH$n Ll$HD$(H$n JH$HD$8HD$0HLbH$HMOHD$0Ll$HHL|$(HT$`E1E1HjE1E1L$H-n H$Lt$H$MHD$XH$m Ll$Ht$HH$HD$(H$m JHD$8HD$ HD$ Ll$HHHME1E1Hm L|$(E1H$HT$`L$Ht$HHD$XH$Lt$MH$Jm Ll$HD$(H$*m JH$HD$8HD$0CHD$ Ll$MMH\$L|$(ME1HHl$8E1L$Hl HD$H$Lt$MHD$`Ll$H$l H$H$l JHD$XHD$8HD$0HD$HH$HD$8HD$(LH$HMLIHTHHLl$L|$(E1E1L$HT$`E1Hk H$Lt$H$MHD$XH$k Ll$Ht$HH$k JH$HD$8HD$0HD$ HD$(Ll$Hl$8L|$(H$H$L$HHxL0HmE1E1E1Lt$E1MH$H"k HD$8k Ll$HD$HH$j JHD$(HD$8HD$0HD$ H0H5oH814dHD$ MLl$HHHE1E1Hj L|$(E1H$HT$`L$Ht$HHD$XH$Lt$MH$fHF0HD$HC(HD$HC HHD$HCH$#IIIIu*MH5@d HHV(HtHD$IM~i1HL -MH5*HKyGH 8wC` H ` ` wCMIuMPH4$HT$HL$LD$H AL HH&HH5AUH81G*H XCH c` ZT` F` CH=/(HD$(dH+%(H01[]A\A]A^f.H!IH5b HIHVN'H$HLkMHJH ;AHNL EHLNOD@@HFHH$!IH5i_ HHV&HtHD$IMH5?b HHV&HHD$I#AWAVAUATIUSHHL-t%H^dH%(HD$81Ll$ H* H H/H^Il$ HEMt$L9L-^ H=e^ IUL%IH- HID$H5M^ HH LIMl I,$qIEE1E1H;# H;$H H;!%Hcx$IH MtLxIcHH$I\AD$HHITIELMG LD$}$H#LD$H Q;P  1LD$LLAIH$LD$H QP =HZ9rMY I(ImI<$iH\ At$I|$ HUM|$H] HD$HHT$H$IH H@HT$HHLHIH HUHa] HHHT$$IH H@HT$HHLHIH H@H;!MQMIQIHI)uIAHT$LLT$P0LT$HT$LHLT$HT$sHT$LT$HID I*IuIBLD$LHT$P0LL$LD$I)uIALD$LP0LD$I(u I@LP0$1H|$HD$~fLh!IHH9\$uH|$_IEL=^ LM\ "H!H QP ; 1LLAI!H QP H= @9 M ImuIELP0M I.u IFLP0I<$BHmHEHP0fHH H HHHHH?L SHLHL@HHh SHH5H8L5o1$L59Z X>%Z ZZZ >LZH=)E1"HD$8dH+%( HHL[]A\A]A^A_ÐID$LP0ID$LP0IELP0I<$rfIE@=sH4 L|$ E1H\$(HD$0IcH)IEHt LH@ uMED HLAIML5SY XL5Y Y t%MtI/nfDImME1IELD$LE1P0LD$MtI(u I@LP0H X X 5X H=!MtI,$u ID$LP0L5X sX eX >Hmu HEHP0H UX KX E1H=_57X .DI@LP0ImVfD @9IE2fDLD$LD$@$vHo HELwL=lX LmLLIHH@HH LLHIHHX LmHL`IHH@HHLLHIHH@H;I_HIWHHI/uIGHT$LP0HT$HHHT$nHT$HIrH+IGI/u IGLP0Imu IELP0LIfLIkID$L=Y LMHH QP ;1LLAIH QP H= @9MI,$tlImu IELP0LIHA%VL5U L5U L5U 1LLI,$Iu@ID$LP0MuT*%L5H=o^U LL5XU 5JU LLL$oLL$HIR%L5>U L5 U T [Im I)MH T T H=5T B#DHHlHHFHHD$ H'H\$ fDLoIH$L5JT L5KT =T TI,$u ID$LP0I/M$H T  T H=5S }pI{HLL|$ H\$(HD$0IcH)Ht jIHdMNI/DIGLP05@I$fDHCHP01HL$ IHL zH5fqL5Z>S ZL5S  S >DIALD$LP0LD$L5 R XL5R R z%MImIELP0E1IGLP0\ID$LP0IB{HuGH=R LHGHH(IMHLH5|H81XX%@$fL5Q XL5Q Q Z%I,$ID$LP0LI;M}M^MeII$Imu IELP0ID$MA(D@$>f.HIHH5Q HHVyHHD$ IFqfHHH8>%[HHH8L5P [L5P P %ImIELP0f.1LLLD$LD$HIDL5yCP XL5@P .P %ImAE1UH=LD$OLD$bLD$LD$HuH%H5H8LD$tHLH8 $TE1HqHH8L5O TL5~O lO $I,$M@H=9Imu IELP0L5\[%O [LH=L5O N %a1LL[ImIHuHH5H8gMIѸ%H=|MI,$ID$LP0HuHH5\H8tII׸$L5c-N XL5*N N l%Mf;IH1LAIIELD$LLL$P0LD$LL$AWAVAUATUSHHHFLnIEH'HYHcH>DnFII fL1&IHHk H5M L}H4$L/H4$HIH@HHLLHIHL{ H5M IWH4$HHT$H4$HH H@HT$HH{HLHHH@H;LEMHUIHHm2LHLD$H$.dH$LD$HI1I(HHmu HEHP0I/u IGLP0HSLI~ HSH ID$L-N HHHH QP ;J1LLH`H QP H=g9mHFI,$tHmu HEHP0IMIHu IFLP0HL[]A\A]A^A_@HEHT$HL$P0L$HT$1LLkI,$Hu ID$LP0Hm@CfDH`eIHBHJ HJ J I,$u ID$LP0Hmu HEHP0MtI/u IGLP0H RJ HJ 5>J hI$BfDHEfD @9@$@DnFII IIIHpH BE1H=H I I I BJDnAMc@DnI@LP05HIbH qBE1H=kH .I  I I BHH8Q BH 5H H H H H=E1JIHffDHH8 HH HH H BI,$#ID$LP0@H=A@I,$mID$LP0]DKHuH_ H5H80 fDMHոBIH@`H+HHHHH HPH;gLoIEHH vHcH>DoH/R0{f.H/R0E1DoAMcH/EHWfDDoGII IDDoGII H5HHHP/H<$H<$IH<$%H<$Iz HH H5H8 f.DAWAVAUATUSHXL-J L5K H|$H^dH%(HD$H1H Ll$0Lt$8HD$@H HfPHHHn H$LnHE H(hE111HALIH$ H8 HsE H(hE111HALHH H8ID$H5E LHHO HH HCH5vE HHH IM LƺHLD$LD$HIGH;x @L;= @щ|$ L;=4 шL$tdI/H5E LǺLD$<LD$HI H; L;=  L;= L$ ˆT$HmuHELD$ HP0LD$ I(u I@LP0|$I/u IGLP0L$L f.f(z LT$ f.T$f(z \f(fTf. HD$H$f(H5 Lx IHxL諍IHw I/H$IGL1E1P0L$I,$SMH+u HCHP0MtImu IELP0HtHmu HEHP0HD$HdH+%(3HXL[]A\A]A^A_fDHHF(Lv H$DH@LP0fH@HP0HH$zL-IB H="B IUL HH HHEH5F HH HIM Hmu HEHP0IFE1H;c@ H;H;HcLD$yLD$HIx MtL@HcD$HI\@HI$MdIFHH 9H QH P ;s HL$1LLH H QHL$P H=l9$H I/AI.u IFLP0HEH@ H0IHp H8u H@LP0L5@ H=@ IVLIH HI@H5+E HH LD$LLD$IMX I(L?@ H=@ IPLLD$LD$HIn HIGH5D HH- LIM I/LH]I9Ak LLLL$WLL$HII)0H!I9FLLLD$WLD$HII($I. L;=L;=L;=L9AƅI/u IGLP0EHD$H$MLH57Lp IHxMkIHI.H$IFLP0L$fDIGLD$LP0LD$ H> H> ~> MHml1E1E1E1I(u I@LP0E1MtI.uIFL$LP0L$MtI/uIGL$LP0L$MtI(u I@LP0H > = H=5= oMtI,$uCE1ID$LL$P0L$$HH!Lv H$@E1LLD$ {LD$ LhD$WHE1E1E1HU= E1D= 6= MLMMHH HIHHI?IAHH H.H5<SL ;H81H XLH < Z< < LH=E1%H A널HHSHH:HFHHD$0IM~TH5R< HHVHtHD$8IM~.H5< HHVxH5 HD$@IM# HD$@Ll$0Lt$8H$DH6HF(HHD$@HF HD$8HFHD$03IfD1E1`fDHE111H; u; g; L\f.HE11F; HC; 1; M&@HQE1; H; ; MfD+HIF@=Lc|$LD$0E1H\$8L)Ld$@Ht0IFLH@ uMVLD$aHLALD$HHMcI(YI@LP0JI-HiE11E1H2: $: : MHmHEL$HMHT$E1P0Hl$L$MxI@LP0kIGLP0 @DIGLL$LP0LL$IALD$LP0LD$IFLP0I@LP0D [HCH=8 LHGHH HH@$fLc|$LLD$0LD$L)H\$8Ht0Ld$@"OLD$HH4HE11E1H8 L8 |8 MaHc8 H`8 N8 MD$eT$HYHWE118 H8 8 &MHLH5\H81HE17 H7 7 MDD$T$L$HHE117 H7 q7 0Mf@HIH_H5e7 HHVHHD$0IfHF HHD$8HFHD$0IfDH-; L5; HELM7 H QH P ; H $1HLAHrH QH $P H= @9+H H THmu HEHP0HE11O6 HL6 :6 QM/DH`IHAE1E1E1H 6 I5 5 MDH E1E1E1H5 5 5 vMzMFMInIHEI.D$HEIH[5 HX5 F5 M;fHaL1E1H*5 5 5 M+HuGH=4 LHGHHIM4HLH5,H81H4 H4 4 MLLD$LD$IfD1LLHH<HqLE11H:4 E1)4 4 MfDH9E14 H3 3 M~H=ǼHL$HL$o{LD$HuRH=j3 LD$HGLHH2LD$HtIPHkLH5ԺH817HE1O3 HL3 :3 MXHHhH5!H89H5I2 H2 2 MLI1HL$0IHL H5H ӺL2 H 2 2 LfDMyMMYIII)uIAL\$LP0L\$ICH;WH;5 L\$L\$HI LxIELh ICLMLT$L\$L\$LT$H QH P ;1HL$(LLLT$ L\$AHD$YLD$L\$H LT$ QHL$(P =H @9QMI*MIBLD$LL\$P0LD$LL$_HJM1 H1 0 MIFLD$LIP0D$HELD$M~MMNIII.uIFLD$LLL$P0LL$LD$IAH;"4H;} LD$LL$LL$LD$HIvLxL@ IALMLT$LL$LL$LT$H QH CP ;#1HL$ LLLT$LL$AILL$LT$H QHL$ P =H @9&MI*MIBLP0Hr/ Ho/ ]/ !NHHE1LHI/ E18/ */ DNL3 L=3 I@LM#L$L$H QH )P ;>H $1LLAIH QH $P H= @92ML'LI.u IFLP0Hr. Ho. ]. TNR@$HsHE17. !H4. L. yNIC@=LHt$0L\$L|$0Ll$8L\$HIgI/uIGLD$LL\$P0LD$L\$MH1LHHHE11|- Hy- g- MM\IA@=LHt$0LD$8LD$ LL$L|$0_LL$LD$ HHD$I/uIGLD$ LLL$P0LD$ LL$I(uI@LL$LP0LL$L|$MHH=H $H $HHH5H8fd1@$HE11E1HR, LA, 3, MH1LALD$HII.LD$ LT$L\$LD$ LT$@$L\$LHt$0L\$L|$0Ll$8$BL\$HIHԳIM+ H+ + MHLT$LL$7LT$LL$@$LD$LD$LHt$0LD$8LD$ LL$L|$0ALL$LD$ HHD$H3M* H* * (Nx1LLKIH0H* H* * PNH=HL$L$L$HL$HuHH5hH81LLLT$L\$L\$LT$HIHRMM* H* * NH%IM) H) ) NHM) H) ) 8N=LT$L $L $LT$HHMME1Hq) c) U) >NLT$L$jL$LT$HHqH5*H8BLT$L$H=HL$ LT$L\$9L\$LT$HL$ %H=HL$ LT$LL$LL$LT$HL$ 1LLLT$LL$LL$LT$HIHIMY( HV( D( N MHE1HHZM!( H(  ( 1NHEL$HE11E1E1P0L$}H&H5߰H8LT$L $AfAWAVAUATUSHHXLfH|$dH%(HD$H1HHD$0HD$8HD$@H IqIHF(HD$ L{ LsH' H(hE111HALHHH8u H@HP0H& H(hE111HALIH H8u H@LP0HEH5& HHH, HH"ID$H5& LHH IM LκHLL$LL$HIS H;@L;-^@щt$ L;-xшL$ tuImuIELL$LP0LL$H5' LϺLL$oLL$HI H;7L;- L;-L$ ˆT$H+\I)|$ImDt$XL5!% H=$ IVLIHHIEH5% HHLHHImu IELP0L5$ H=$ IVL'IHHIAH5( HHLLL$LLL$IMQI)u IALP0IGE1H;XH; H;DHcLL$LL$HIMtLHHcD$HEHIlH% DiMcHKDIGLM3H QH P ;!HL$1LLAIeH QHL$P H=' 9 MI.I/ HH9CLH&;IHImH+L;=L;=Tu L;=vÉD$I/oL$L-" H=" IUL)HHmHHCH5# HHHIMH+L=\" H=5" IWLIHHIEH5F& HHLIMImu IELP0IGE1H;vH;H;Hc IHMtLhHcD$I$HMdH" YHcHIDIGLM!LL$FLL$H QH P ;=1HL$LLLL$AH LL$H QHL$P =H @9HI)u IALP0I/u IGLP0HWI9F;HL8IHH+uHD$HCHP0LL$I.uIFLL$LP0LL$L; #L; 6L; )LLL$TLL$I)u IALP0"HD$HT$ MHH5LH IHxLL$zLL$HII)IALP0fDHXH5" HIHVIHD$0HLc@IH HAHMEIHHH¯H5ATL H81*H {XSH F Z7 w) SwH=`E1HD$HdH+%(HXL[]A\A]A^A_HYHD$ IALP0|$dLD$PHЦE1 H  LTE1E1ImuIELL$LP0LL$MtI)u IALP0MtI/u IGLP0MI.IFLP0}DHCLL$(HP0LL$(I)DLf.f(zLT$f.T$f(zdff/tf/tHD$HT$ f(H5Lh IEHxLgIH Imu IELP0HmuHEHP0MI,$ ID$LP0IELP0t$DLLL$(LL$(fHIqMI HFHHD$0IH5 HHVHD$8HIMHD$@Lt$0L|$8HD$ IHF(HHD$@HF HD$8HFHD$0IM~1HL$0MHL UH5P+ySH 5 H   wH H  2TfDH   H=ǫA5x Hm4%HIGLP0fIG@=H< Lcl$LL$0E1Hl$8HD$@IWL)Ht0HBR uMGLL$HLLL$IMTMPI)FIALP07HCHP0lIELP0RIGLP0IFLP0 @DL8D$CH]M$ H!  :UI)fIALP0WHCHP0^@$SHuGH=y LHGHHIMCHLH5H81QHl Hi W `U$f.{IHaE1E1% H"  CTH+E1HCLL$HP0MLL$jfETH  H   H+HCLL$HP0LL$I)H AfDH5a HHVHAHD$@I.fDHF HHD$8HFHD$0IfDHT3fDH )#TE1H=(H    #TMH Lcl$LLL$0LL$HD$@L)Hl$8Ht0-LL$HIHE1b H_ M T8kHW Lf.2zf L- L=$ IELM H QH .P ;i HL$1LLAIH QHL$P H= @9\MC L*4Imu IELP0Ht Hq _ yTf{HuGH= LHGHH IM HLH5|H81H0 H  THIG@=H Hc\$Ll$0E1Ld$8HD$@IWH)Ht0HBR uMOHLHHM`ImUIELP0F@T$]T$Hx ff/L- L= IELM H QH $P ; HL$1LLAIH QHL$P H= @9`M L 2Imu IELP0Hj Hg U TE1TH d2 H / ! ImE1E1L8HHD$LL$HuWH= HGHH; LLL$HtIHLLL$H5H81jLL$HE1E1z Hw e TPTLLL$sLL$IfDHQE1 H  TfDMOMMoIIEI/AD$IEMkLCMeLKIIH+uHCLL$HLD$P0LD$LL$IAH; H;nLL$LD$LD$LL$HI L@Lh IALM LL$LL$H QH >P ; 1HL$LLLL$AILL$H QHL$P =H @9 M I.LlIFLP0]LL$BLL$@$UH_ H\ J UH+E1E12H^( H%  TH' Hc\$LLl$0Ld$8HD$@H)Ht0F'HHkH H  tUM[E1#1LL IHHE1v Hs a ULL- L=F IELMg H QH hP ;* HL$1LLAIH QHL$P H= @9MLd-Imu IELP0H H  IUH=uHL$HL$HHH5]H8ukHuGH= LHGHHhHHaHLH5lH81H  H  [U8HI2HE1E1 H  ]UIGLL$LMP0D$IELL$bULI @$@$MoM#I_IEHI/u IGLP0D$HCIߺHHAH5 jL H ?H8H 1_SAXM~MINIHI.uIFHL$LP0HL$HAH;H;_CHL$HL$HILx1HHHX HL$HL$HIbImILHD$IELP0LL$3HE1 H  UHޔ H  UH H~ l U_1LLLL$LL$HH6Hm7 H4 " UI)L-| L= IELMH QH P ;uHL$1LLAIH QHL$P H= @9MqL)Imu IELP0He Hb P UHt> H; ) UH=HL$LL$VLL$HL$LL$LL$HH H5ٓH8LL${IA@=JLHt$0LD$0LD$LL$Ll$8LL$LD$HITI(uI@LL$LP0LL$ImuIELL$LP0LL$L @$FL1LIHpH3 H  uTKH  H  WT$HHב H  aTIOH=[HL$HL$yMzH?HH5CH8[$fDL1LIHH8 H  TPH=ɑHL$HL$HuHH5H8LL$dLL$@$LLL$LL$LHt$0LD$0LD$LL$Ll$8LL$LD$HIHdML( H%  UHA@=HHt$0HL$L|$0H\$8HL$HII/uIGLL$LHL$P0LL$HL$H+uHCLL$HHL$P0LL$HL$I:@$HE1Y HV D T/H1LLL$IbKHHC  H   EU[H=ԏHL$*HL$L1L>IHHIHǎML H v .U'H=RHL$LL$LL$HL$VHrLE16 H3 E1 4U 1LLLL$LL$HIVLL$LL$HuH!H5ڎH8LL$wHHt$0HL$L|$0H\$8HL$HIHI~ H{ i UTGH=@HL$HL$mHj4 H1  U=HuHQH5 H8"L1LcIHTH H  |UH1LH4H͌I H  UIEHIE?E1M HIE1O HL : U%HvH5/H8GHCML H  'UH H  AT.HI H  UH A fAWAVAUATIUHSHHHEHEHLIT$H5x LHHHTIMIHEH5 HHC+IEH  LMPL= H QP A;bLHLAInX SP AH=$9*MCImI,$u ID$LP0Hmu HEHP0HL[]A\A]A^A_HIMHI9D$u[Ml$MtQM|$IEII,$t|LLIHImM_IELP0PLXIH4H͉ LH  NDID$LP0t @9@$@HHT$HT$HH5Hf{E1fDH 1NK KH  NH=0E1UjOH H  NImt&I,$t0H  5} IELP0I,$uID$LP0fI!H qNN1 NH .  N;H 9NL LH  N IH NH  OIm IELP0LHLIHfDOfDH=9f.[HuHoH5(H8@fDH9 LH  NI/tImEHIGLP0@AWAVAUATIUHH0dH%(HD$(1HXHEHEHIT$H5 LHHHIMHI9D$Mt$MMl$IIEI,$.LLIHI.MuIFLP0@I,$u ID$LP0Hmu HEHP0HD$(dH+%(H0L]A\A]A^A_HIMHI9EMuMMeII$ImSID$H;H;:TIHNLpHEHh ID$LM\L5(H QP A;1LLAIH QP AH= @9MnImIELP0@LIH|jOH Q H N @ I,$u$E1ID$LP0MtImu IELP0H   5 !DIELP0ID$LP0IT$BፁLt$1Hl$LB uI|$Ht$AIMI.fDHLMJIHnOHHT$軼HT$HH5H^uvE1AfDNOH  H 5 H=HE1U@$fI!O@I3ZOHt$LLt$Hl$IHHyC H@ . OfDI,$u ID$LP0I.IFLP0DH!M H  gOfH H  Oyf1LLIHHk Hh V OI,$ fDH=!|VfKHuH_H5H80H/ H  O1ɺAI?谼AWAVAUATUHSHXL5$H^H|$dH%(HD$H1Lt$8HD$0Lt$@HHHkHHH΄H HNAL H^ELNODHH*HSH5[H81H *XPH Z  PH=E1LHD$HdH+%(HXL[]A\A]A^A_fDLt$MLmIEI$M9L5 H=3 IVLǾHH HHEH5 HH HHHK Hmu HEHP0H=[ H5l HGHH HHJ EIH H5 LHH5 LL HEH5 LMJ Ht$@Ht$H QH P ;HL$ LHAHD$ LT$H QHL$ P =H @9MHmuHELT$HP0LT$I/uIGLT$LP0LT$HCE11H;8AH;UG H;IcgLT$MLT$HIHtHhIcAWMTH HcHIDHCHH H QH vP ;HL$1LHIܻH QHL$P H= @9 MI.u IFLP0H+u HCHP0I/u IGLP0H|$H5 HGHH IM IT$H;1H5 T ID$HHHa HAT$H)HJHHE*HH. ԺHH@ IEHX LhfHHHT$H5 H& H H5 H IGLMgH QH ҹP ;HL$HHLAI6H QHL$P H=9h McI/HmH+Imu IELP0I,$8ID$LP0(@L=Y H= IWL6HHB HHCH5# HHHHH H+u HCHP0H= H5 HGHH@ HH 赺IH1 H5R LHHD$rLT$]HCL=J LM LL$ LT$軸LT$LL$ H QH P ;9LHL$ LHLT$AI{LT$H QHL$ P =H @9M%H+uHCLT$HP0LT$I*u IBLP0HE1E1H;A_ H;зJ H;+IcLT$ T$ķT$LT$ HHiMtLPHcH < L|BHHHLHELMV耷H QH P ;1HL$ HHAHD$NL\$H QHL$ P =H9[MiH+HmI+IEI.H-M HEImEMIOf.HF(HD$Le VfDLt$IH{QH0H]HFHHD$0-IM HD$@Ll$0Ld$8HD$HHF(HHD$@HF HD$8HFHD$0ӰIM~1HL$0ILL H5[jyH HwP H  PPE1H w H  H+uHCLT$HP0LT$HtHmuHELT$HP0LT$MtI*u IBLP0MtI/u IGLP0H f \ E1H=5H ˶&fDHC@=Ht IcHl$0E1LT$8HD$@H)Ht0HCLH@ uL[LT$ HLALT$IM HtHmuHELT$HP0LT$I*IBLP0PE1fDLT$<$@$:kHuGH= LHGHH HHHLH5ltH81϶PH u H  ifDIELMIP0HEL\$HP0L\$ZIFLP0]fICLP06HCHP0AHEHP0IGLP0 @DHE@=Hd LT$0E1L|$8HD$@HcH)HEHt0L@@ uL]LT$ HLALT$IM MtI*uIBL\$LP0L\$I/?IGL\$LP0L\$&fDHCL\$HP0L\$ @DLH襴IHfPE1#fDL\$辱L\$@$HH=HsE1E1 H  PHشHlE1ҺPH r IH | n H+1D苴H`E1ҺPSfD@$6f.1QH H $HHJH5OH8H $H=NLD$ LL$H $H $LL$LD$ LIϸ,Sw1LHLL$H $ H $LL$HH$H=^NLD$(HL$ 诋HL$ LD$(&JLL$H $jH $LL$HvHqH5*NH8BLL$H $RH=MLD$L$?L$LD$eL$L$HOH H5MH8݈L$0菈HLMM rH I| MSHLMILHh Z rL SlHpLMI4 mH1  R$fAUATIUSHHHuHndH%(HD$1H$MHHHH>PH /PHHHHZH?L )HLHL@HHUH7TH5KH81HH KXQH d ZU G QH=+TE1軋HD$dH+%(pHL[]A\A]fHVHk H5HEH{HIHtHHmuHEHP0DHHHFLH$-HdH$fHJ H v QHmt'H p f H=KS5U ؊HEHP01HL RIH5LV=vH ?J{Q H   {QfLhIH4H5 LHV)HtH$IE 蒅fAWAVAUATUSHXL%G HnH|$dH%(HD$H1HLd$0HD$8HTHZH8HHHMIH MH\RHIHH%I?UIH5VIH8L SWA1ɊH IXVVH Z  VVH=4RE1<HD$HdH+%(HXL[]A\A]A^A_fDHHD$HE H(hE111HALHHH8u H@HP0HEH5o HHHIMH57 L誃HHI/u IGLP0H;gH; u H;/DH+u HSHR0ELtf.OD$zL%W H=0 IT$LÆIHHIGH5p HHLIM'I/u IGLP0D$̄IHHaI9E LLHH I.u IFLP0Imu IELP0H;@H;{H;nHvAŅ H+u HCHP0E HD$D$HT$H5cHX HHxHWIHS H+u HCHP0HmQHEHP0BfDL%ѽ H= IT$L=HHHHCH5 HH"HIMqH+u HCHP0L%j H=C IT$LքIHHIGH5 HH[LHH:I/u IGLP0L%L9cHHIH H+u HCHP0M9e LLIH I/u IGLP0Imu IELP0L;59L;5߂u L;5DI.u IFLP0E HD$HT$HH5rLh IEHxMuIHy ImIELP0HF HD$LffDHiHD$fH؃AŅ(HCE1E1» H  VH+u HCHP0LMHH+Au HCHP0MtI.u IFLP0MtI/u IGLP0H H > H=L5- 调E1DIHCHHHh|IH, @HD$8Ld$0HD$@@HBE1 H  yVH  H=K5y ML;HBIP HM ; V릐kIHQBE1 H  VI/1IGLMP0WLЁAąNHA H  WI.ZM1E1IELP0H踃IL訃I@HAE1E1U HR @ VHF HHD$8HFHD$0zHE1HL$0ILL 9JH5b-4H AEV۸ H ظ Ƹ EVfHFHHD$0/zIMH5 LHVHjHD$8ID$RHH@IJ HG 5 VKHuGH=׷ LHGHHT HH-H~LH5L?H81证H@ʷ HǷ  !Wi~HuGH=W LHGHH IM'Hc~LH5>H81/H?IG HD 2 VDHQ?E1 H  #WQfD1ۺVH ? H Ҷ I/M0}HuHH=g LHGHH HtI#Hr}LH5=H81>M&WH >Q H R D VH W>% H "  Imi1jf.(WLHMeMMMI$IImuIELL$ LP0LL$ IAH;8|H;|LL$ .|LL$ HI2 L`Lp IAHH LL$ |LL$ H QH m{P ; 1HL$(LLLL$ H{LL$ H QHL$(P =H @9H I/MIGLP0@H< H  VI. ImAfDLsMJL[IIH+uHCL\$HP0L\$ICH;zH;%{2L\$zL\$HI LpHEHh ICHHLD$ L\$zL\$LD$ H QH yP ; 1HL$(LLLD$ L\$ILzL\$LD$ H QHL$(P =H @9M I(LBI@LP03DHQ; H  6WHMHHsfDI]HMeHI$Imu IELP0ID$H;QyH;yDLyIHeHXLx ID$HHLL$)yLL$H QH xP ;: 1HL$ LLLL$IxLL$H QHL$ P =H @9CM I)M,IALP0fDL%9 L- ID$HHopxH QH wP ;HL$1LLIAxH QHL$P H= @9MMLI,$u ID$LP0HW9I H  Vq@H)9L H ۰ aW)fDH8E1 H  WfDMPE1#L% L- ID$HH(wH QH vP ;HL$1LLIvH QHL$P H= @9*MLI,$u ID$LP0H8I֯ Hӯ  W)@W{H5ٳ LHVvHHD$0IIQBፁLd$01Lt$8LB uIyLL$ Ht$0!ALL$ HHI,$uID$LL$ LP0LL$ I.uIFLL$ LP0LL$ M%ISBፁLt$01Hl$8LJ uI{L\$Ht$0AL\$IMI.uIFL\$LP0L\$Ln+u@$ IT$BፁH\$01L|$8LB uI|$Ht$0AIMH+u HCHP0I/u IGLP0M&t@$LL$ tLL$ @$wI1wHLD$ L\$XtLD$ L\$@$+LHt$0LL$ Ld$0Lt$8LL$ HH(Ht5LM8 H5 # VLHt$0L\$Lt$0Hl$8eL\$HIbH5LM٬ H֬ Ĭ IMH5 LljIH I/u IGLP0L;%)mL;%lu L;%lDI,$u IT$LR0EL4nf.L6D$zL% H= IT$LmIHHIGH50 HH8 LIMI/u IGLP0D$kIH H!kI9F LL脽IH Imu IELP0I.u IFLP0L;%lL;%ku L;%k DI,$u ID$LP0E Lc D$H$H5jI$H{L-IHI,$uID$LP0HmHEHP0fDII+IMMH 1H 0HNH^ L :LNL@HHljAVHs5H5,H81nH k,XXH 6 Z'  XH=5E1lHD$8dH+%(HHL[]A\A]A^A_HIjH$L- H= IULkIHHID$H5r HHJLIM I,$u ID$LP0L-@ H= IULjIHHIGH5Z HHBLIM!I/u IGLP0L-bhM9l$HLĺIH I,$u ID$LP0M9n LL葺IH I/u IGLP0I.u IFLP0L;-iL;-hu L;-h`DImu IELP0E Ls H$HH5hIH{MPIH I."IFLP0LiAƅnH*E1E1ʡ \Hǡ  XI,$u ID$LP0MMDMI,$Au ID$LP0MtImu IELP0MtI/u IGLP0H M C H=252 iE1EDHI)E1 ZH  XH  H=25 diMLfDH(I \H  XfjIH(E1x \Hu c XI/IGLMP0L0hAŅHV(E1E1 ^H  YKLgAąH(؟ cH՟ ß YImlME1E1IFLP0HaH5џ LIHVIfHD$ H M7HD$(Ll$ H$@kfHHc'I* ]H'  X+fHuGH= LHGHH IM4HeLH5,&H81hH& cH  ]YDeHuGH=7 LHGHH% IMGHCeLH5%H81hH`&I' ^H$  XDHF HHD$(HFHD$ v_H1HL$ MLL .H5i^H %X H  XxfHFHHD$ ^IH%E1X cHU C _YfDLhgIH5 LHV}dH'HD$(IGfDE1^XH% Hߜ ќ I/nM'LfIcHuHH=W LHGHHj HtI Ys1LLLT$LL$XLL$LT$HIMM׸YH=HL$LT$L\$)UL\$LT$HL$e_LT$L$UL$LT$H?HSH5H8RLT$L$LT$L $UL $LT$HUHSH5[H8sRLT$L $1H=HL$LT$LL$jTLL$LT$HL$"QE1Q1ɺALD$L\$IHMMʍ ^HǍ  XJ1ɺALT$LL$IHMM cH~ Mi YHRH5^H8vQL $1ɺALL$IpHWMM cH  YIELP0E1sfAUATIUSHHHESHndH%(HD$1H$MHHHHH HHHH"H?L THLHL@HHiRUHH5H81VH iX XH 4 Z%  XH=E1THD$dH+%(pHL[]A\A]fHVHk H5QHEH{HОIHtHHmuHEHP0DHHHFLH$LHdH$fH[ HX F :XHmt'H @ 6 H=5% SHEHP01HL /IH5NVL&vH WԊ H ъ  WfL8LIH4H5 LHVQHtH$IE bNfAWAVAUATUSHHXL- HnH<$dH%(HD$H1HPLl$8HD$0HD$@Hi HHHHHxH iHNAL : HELNODHHOHUH5H81SH X^ZH Z h ^ZhH=1QHD$HdH+%()HXH[]A\A]A^A_ÐHOHD$L{H H(hE111HALIHH,H8u H@LP0H H(hE111HALHH H8u H@HP0ID$H5 LHH= IM+HELD$HH5 HH LD$IM LLǺLL$LD$KLD$LL$HI H;NL;5EN L;5fNtfI.{H5 LϺLD$LL$mKLL$LD$HIg H;0NL;5M L;5M I(uI@T$LLL$P0T$LL$I)uIAT$LP0T$5I.L- H=݆ IULqNIH HI@H5· HH6LD$LLD$HH;I(u I@LP0L- H=m IULNIHHIGH5 HHLIMeI/uIGLL$LP0LL$L-KM9i LLLL$ LL$HIII)uIALD$LP0LD$L9kULHLD$ŝLD$HIDI(H+hL;=ILL;=Ku L;=LI/^L=_ H=8 IWLLIHHIFH5) HH1LHHI.L= H=؄ IWLlLIHRHI@H5 HHLD$LLD$IMI(\M9iLHLL$肜LL$HII)u IALP0L9kLHLIHHI.u IFLP0H+u HCHP0L;=JL;=oJd L;=JW LKÅI/u IGLP0H$HT$ILH5wILx IHxM3HH$I/uIGLP0I,$$ HF(HD$Lk :fDHIHD$f.L8Kf.PD$z LKf.2D$z L- H=ւ IULjJIH HIAH5 HH LL$ LLL$ IMt I)uIALD$ LP0LD$ D$LD$ ZHLD$ HIY L-GM9h LLLD$ HLD$ HI I.uIFLD$ LP0LD$ I(u I@LP0L;=HL;=\Hu L;=~HI/u IWLR0L=Ɓ H= IWL3IIHaHI@H5 HHLD$ LLD$ IMfI(D$8GHHM9obHL7IHqH+I/L;5GL;5fGL;5GLHAŅI.u IFLP0EH$L$D$HT$L@ H5bFIHxLL$L$HHI(I@LP0DIFT$LP0T$;IFLL$LLD$P0LL$LD$bDLT$ LL$LD$HT$ LL$LD$&LG…HME1 H  Z1E1@IHQHHHHFHHD$0 AIMHD$@L|$0Ll$8HD$(HvHF(HHD$@HF HD$8HFHD$0@IM~1HL$0ILL fH5 J;yH (KZ~ hH ~ ~ KZQH~ H~ ~ ZfDH ~ ~ H=5~ GI,$u ID$LP0HHmHEHP0{HIHa1E1E1H*~ ~ ~ ZI(uJE1I@L $LP0L $Mt0I)u*E1IAL$LP0L$MtI(u I@LP0MtI.u IFLP0MtI/u IGLP0HH+HCHP0f.GLD$I fDHCHP0fI@LP0gIGLP0LEÅ H/ME1| H| | [I@LP0NLDÅ&\H | 1H | | IFLP0[IGLP0:HCHP0 I@LL$LP0LL$Ha1E1&| H#| | ZI(f.DMAM8IIIHI)uIAHL$LLD$P0LD$HL$HAH;GBH;BHL$LD$8BLD$HL$HL@1HHHEHh HL$HD$9HL$L\$HII+IICHL$LP0LL$DH11E1z Hz z Z@H HAL HF HHD$8HFHD$0HuHH=_w LHGHHHtIHj>LH5H816AHQw HNw L-Oq L5u IEHH3 7L 6H QP A; L $1LLIX7L $H QP AH= @9M LImu IELP0Ho9p H6p $p \wHH1E1 p H p o #[LLD$ :LD$ IK6HuGH=o LHGHH IMOH6LH5H81`9H{o Hxo fo %\HTo HQo ?o &[E1Һ'\H VILӉo H o  o OL89HMOMMoIIEI/uIGLL$ LP0LL$ IEH;Y5b H;5WLL$ O5LL$ HILHHX IELMLD$ (5L 4LD$ H QA;P 1LL$(LLLD$ AI4LL$(LD$ H QP A=H @9N MI(MI@LP0HI1m Hm m 3[4HuGH=Um LHGHHf IM|Ha4LH5H81-7H~Hm HEm 3m *\HWME1m Hm m V[;,\tLLD$7LD$IL-Am L5q IEHH7 y3L 2H QP A; L $1LLIJ3L $H QP AH= @9M LImu IELP0Ha+l H(l l e[i[H+E1k Hk k :\LKM\LkIIEH+uHCLL$HP0LL$IEH;92A H;2LL$/2LL$HI LH1HLLp HD$)LD$HI> I(LI@LP0\dH9k Hk j e\I.Y,L-Hk L5o IEHHC 1L 0H QP A; L $1LLIQ1L $H QP AH= @9*M LImu IELP0Hh2j H/j j \pIC@=LHt$0LT$0LT$(L\$ Lt$8(L\$ LT$(HIb I*uIBL\$ LP0L\$ I.uIFL\$ LP0L\$ MV\HA@=HHt$0LT$0LT$HL$Ld$8y'HL$LT$HIm I*uIBHL$LP0HL$IM/@$IF@=Ht$0LLL$0LL$LD$8LD$&LD$LL$HI I)uIALD$LP0LD$I(u I@LP0L8/@$L\$ %/L\$ @$IE@=Ht$0LLL$0LL$ H\$8Q&LL$ HI I)u IALP0H+u HCHP0MLHt$0LT$0LT$(L\$ Lt$8"~L\$ LT$(HIHM1MHg Mg wg Zaf.+.@$HL$LD$.HL$LD$@$0k1I@^1I#HHt$0LT$0LT$HL$Ld$8`}HL$LT$HII*uHD$IBLHL$P0HL$LD$ILT$t-LT$@$HA@=HHt$0LD$0LD$HL$Hl$8$HL$LD$HII I(uI@HL$LP0HL$IHt$0LLL$0LL$LD$8LD$m|LD$LL$HIHHe He e [I)A/ IE@= Ht$0LLL$0LL$Lt$8#LL$HI& I)u IALP0I.u IFLP0L+@$L1L.IH?H9e He d [A ,HuH *H5H8(H=L $*L $LD$ g+LD$ @$L1L-IHHkd Hhd Vd \Ht$0LLL$0LL$ H\$8zLL$ HIHHIM d H d 1c :[#.IH=L $*L $!J*H &H:> (> ]KHL> "H> > ]$HH= #H= = ]IOH=HL$HL$yMH?HH5H8$fDL1LIHHxB= (H?= -= ]PH= HL$_HL$+HuH?H5H8LL$LL$@$LLL$LL$LHt$0LD$0LD$LL$Ll$8RLL$LD$HIHMLh< ,He< S< F^HA@=HHt$0HL$L|$0H\$8PHL$HII/uIGLL$LHL$P0LL$HL$H+uHCLL$HHL$P0LL$HL$Iz@$HE1; ,H; ; ^/H1LLL$IbHHM; -HJ; 8; m^[H=HL$jHL$L1L~IH+HIHML: ,H: : V^'H=HL$LL$LL$HL$VHLE1v: ,Hs: E1^: \^ 1LLLL$LL$HIVLL$RLL$HuHaH5H82LL$wHHt$0HL$L|$0H\$8GPHL$HIHI9 .H9 9 ^TH=HL$HL$mHt9 /Hq9 _9 ^}HuHH5JH8bL1LIHTHF9 .H 9 8 ^H1LH4H I8 .H8 8 ^IEHIE?E1M HIE18 .H8 z8 ^%HH5oH8HMLG8 ,HD8 28 O^HV 8 !H8  8 i].H/I7 .H7 7 ^H TA fAWAVAUATUSHHxLfH|$dH%(HD$h1H HD$@HD$HHD$PHD$XHI I HF0HD$(HC(LkHD$HC H$H7 H(hE111HALIHfH8u H@LP0H6 H(hH<$E11HA1HHKH8u H@HP0Hl6 H(hH|$E11HA1HHH8u H@HP0ID$H56 LHHIMbHELL$HH5_6 HHLL$IMLϺLLL$LL$HIH;_@L;@ L;!t$ʈT$  PI(uI@LL$ LLD$P0LL$ LD$HCLL$ HLD$H55 HHLD$LL$ IMLLLL$LL$HI*H;@L;5@ L;U|$ʈT$ @tdI(H55 LLL$YLL$HI3H;!L;D$ L; ЈD$ I/uIGLL$8LLD$0P0LL$8LD$0I) I. |$  I( DL$E L-3 H=3 IULSIHHI@H54 HHL$LL$IMI(uI@L $LP0L $L-p3 H=I3 L $IULL $HIyHI@H5R7 HHLL$LL$L$LL$IMI( IFE1H;w H;H;Hc L $L $HIMtLxHcD$I$L $HMdHX7 QHcHIDIFLM@LL $H QA;P  1L\$LLLL$AH$L\$L$H LL$QP A=He 9MVIm! I. HXI9A(LLLD$L $IL $LD$HIgI( I) L;52L;5e L;5X Lh$nI.u IFLP0D$EL-)1 H=1 IULIHHIAH51 HHULL$LLL$IM I)u IALP0L50 H=0 IVL&IHHI@H54 HHLD$LLD$IMI(u I@LP0IGE1H;H;cH;CHc"LD$LD$HIeMtL@Hc$HEH3 Il@HHITIGHH(H $LH QA;H $P 1L\$LLH$aL\$L $H QP A=H @9MI.uIFL $LP0L $I/uIGL $LP0L $HI9ENLLL $FL $HII)u IALP0Imu IELP0L;5sL;5> L;571 L$I.u IFLP0<${L5l. H=E. IVLIH-HIEH56/ HHlLIM4Imu IELP0L-. H=- IULrIHBHIAH5o3 HHLL$LLL$IMI)u IALP0IFE1H;H;4H;HcLL$,LL$HI MtLHHc$HH1 I\@HHITIFLM !L$LTL$H QA;P !1L\$LLL$AIL\$L$H QP A=H @9M I(u I@LP0I.u IFLP0HI9GLLXDIHImu IELP0I/u IGLP0L;5L;5zL;5L AŅI.u IFLP0EuHD$HLIH5ILp IHxAVHT$8Y^IH I.uIFLP0I,$yHmDHIH5N0 HIHVHD$@H[Lc@IHH HNHATHL@HH5UL UH81H X_H * Z* 3* _3H=E1?HD$hdH+%(!HxL[]A\A]A^A_fHHD$(*IFLD$0LP0|$ LD$04LLD$ f.HF0HD$XHC(HD$PHC HHD$HHCHD$@IIt{I~=IMHD$HLl$@H$HD$PHD$HD$XHD$(MInH5+ HHVBHD$HHIH5q, HHVHD$PHIkfDH y_E1H=ոH ~& p& yb& _HK& {HH& 6& _fI@L $LP0L $fIFLL$LL$P0LL$L$I@LL$LP0LL$ IF@=H\* HcT$L|$@E1Ld$HHD$PH)IVHt@HBR uMFL $6HLL $IMMEI/;IGLL$LL$P0L$LL$IALP0`fI@L $LP0L $6fIELL$LL$P0LL$L$ @DH5$ HHVHtHD$XIML1HL$@MHL JH5s莟$|_H r5<$ H =$ /$ 3[fD‰$DHFHHD$@ILL$L$LL$L$@$H# }H# # _afIXLLL$8LD$0nLL$8LD$0HE1E1E1HQ# C# }5# _I)uIAL$LP0L$M&I( D3LL$IfDE1}_H " H " " I.uIFLL$LL$P0LL$L$E1E1M[n‰$D}_@LL$ LD$I fH& HcL$LL $L|$@HD$PH)Ld$HHt@|8L $HIH-E1! H! ! `f}_HuGH=w! LHGHHfIMHLH5H81OHj! Hg! U! `kHHc-! ~H*! ! `IG@=H$ Hc$LD$@E1Hl$HHD$PH)IWHt@HBR uMOL$HHLL$IM MDI(:I@L $LP0L $#fD$ $HHyC H@ . `D$E $T$HH2 H  `f.L-A L=r% IELM<yLH QP A;L$1LLAIIL$H QP AH= @9 M#L<Imu IELP0H`* H'  .`E1E1E1`H  H   fDLL$L$IL-! L=J$ IELMYLH QP A;L$1LLAI)L$H QP AH= @9 ML;Imu IELP0H@  H  N`LL$H$L$LL$HucH= HGHHLL$LL$HtI9HLH5LL$L$H81JLL$L$HE1E1E1HZ L > `L- L=" IELMLJH QP A;@L$1LLAIL$H QP AH= @9ML@:Imu IELP0H H u n`0LLL$L$LL$L$IfDHiME1- H* E1 `E1E1`M~MMnIIEI.D$IEMMqMMQIII)uIALT$LL$P0L$LT$IBH;H;k LT$L$L$LT$HIyLpL@ IBLM L$LIL$H QA;P L1L\$LLL$AIL\$L$H QP A=H @9 MImMIEL$LP0L $DlL $2L $@$}_sHwE1E1; H8 & `HJE1 H  `aE1E1E1H Hc $LLD$@LD$HD$PH)Hl$HHt@0LD$HIHá H x IaMMaL- L= IELM+LrH QP A;ML$1LLAIL$H QP AH= @9M#Lh6Imu IELP0H H  aX1LLL $HIwHE1E1h He S `H=/L\$L $L $L\$IHuGH= LHGHHkIMHLH5JH81H H  0anL $L $HHH5H8L $IF@=H Hc$LL$@E1H\$HHD$PH)IVHt@HBR uMFL $eHLL $IM&M[I)QIALP0BHE1E1E1H   2ayLLL$LL$IHuGH=C LHGHHJ IMHOLH5H81Hl6 H3 ! 5aaHH!H5ZL ZjAH sHէH81AZr_A[IFL $LMP0IEL $D$m@$CLLD$LD$IE17a{MGMMwIII/$$IFM@$@$`HHH5WL WjAH pHҦH81[l_]MuMIMIHImuIELL$LH $P0H $LL$HAH;- H;g LL$H $H $LL$HI! Lp1HHLH H $H $HI" I/I)IGLP0H6 H  raE1E1L$L$@$aE1E1E1Hۛ H  WaH< Hc $LLL$@LL$HD$PH)H\$HHt@)LL$HIFHo9 H6 $ aE1M*E1H5 H=r 1cIHs H0Imu IELP0H H  ao1LLIH3HMM H n baM}fDHuHH5PH8hfD[HuGH= LHGHH IMHLH5\H81H H  aH=L\$H $H $L\$IB@=#LHt$@LD$HLD$L$Lt$@L$LD$HI I.uIFLT$LL$P0LT$L$I(uI@L$LP0L$MMVH* H  aL I@$HD$LL$HuUH=a HGHH8 LLL$HtI|HgLH5ЗL $H81/L $H|E1C H@ . aIGLD$LMP0$IFLD$H&E1E1 H  aLLL$LL$IMNMMnIIEI.$IEML1LIHHT HQ ? *`mIPHuHdH5H85L1LsIHH H  J`MOM@IGIHI/H$uIGLL$LP0LL$H$H@H;"H;}LL$LL$HPL<$LH1HLh LHD$LD$HI I(I@LP0H=L$BL$VH=ʖL$!L$UHE1 H  a0L$]L$@$HHH5tH8pbE1E1E1lHp: H7 % aL1LIHWH0 H  j`HuHH5ЕH8fDH=L$L$fLLL$L$LL$L$H5 H= 1IHH)Imu IELP0Hh2 H/  *b1LLL$L$HIKaE1H=ƔL\$L$L$L\$L$L$HuHH5H8L$E1E1ɺObLHt$@LD$HLD$L$Lt$@!L$LD$HIHiE1M- H*  `HA@=HHt$@LL$HLL$H $Lt$@H $LL$HII.uIFLL$LH $P0LL$H $I)uIAH $LP0H $IMIFLL$LMP0$IELL$H|E1C H@ . `H1LL $IL1L~IH"H# H  aH=L$ L$HuHH5H8IHHt$@LL$HLL$H $Lt$@H $LL$HIHuI< H9 ' ya7UI1LLL$wL$HINHME1E1H E1  aHE1M H  aZH H m QaH1LL$IH$H@@=H<$Ht$@LL$@LL$Ll$HOLL$HIYI)u IALP0Imu IELP0L<$L$L$HHH5H8L$H=L\$L$L$L\$Ho Hl Z aLLL$LL$qI&HZI! H   aH0IE1 H  alH H  aH1LL $IHƎ H { &b6H<$Ht$@LL$@LL$Ll$HLL$HI=HlL<$2 H/  aL<$E1b H+L<$ H  biHI H  aH֍L<$ H  bHE1Mo Hl Z ` ff.fAWAVAUATUHSHHHLvdH%(HD$81HHD$ HD$(HI|IHF HD$LuH H(hE111HALIHH8u H@LP0ID$H5 LHHlIMH5 L IHI/uHD$IGLP0LL$L; @L; b@u L; u@I)u IALP0Lf.zff/LK HT$H5;IH{LLL$&LL$HH5 I)IALP0LIH/HGP0fIIS IMMHH HNHhL fLNL@HH AVHH5<H81H XbH  Z  bH=E1-HD$8dH+%(HHL[]A\A]A^A_HHD$L5A H= IVLIHHIAH5  HHsLL$LLL$IM(I)u IALP0L= H= IWL>IHHIEH5 HH[ LIM Imu IELP0IGE1H;8 H; sH;dHc[IH MtLhHcI$MdH HcHIDIGLMP LD$H-1LD$H Q;UP  1LLD$LAHD$LL$LD$H QP E=p9|MV I(yI/uIGLL$LP0LL$HI9FfLLLL$.LL$HI} I)u IALP0I.u IFLP0L;=L;=Lu L;=n0I/ L{ HT$LH5PIH{M2HH I/LI'LLL${LL$HcrHfHcQbI) E1E1IALP0M I. E11IFLP0MtImu IELP0MtI/u IGLP0HHmHEHP0HHbfDH H=y5~MLE1DbHMHJ<H =3H=5"LE1f.;IIG@=H Ll$ E1Ld$(HD$0IWH)Ht HBR uMO HLIM MImIELL$LP0LL$bHwEHB4I/IGLP0IGLP0&I@LL$LP0LL$n4@ADIGLP0LISLŅcQHH5 LIHVIHD$ H3MHD$(Lt$ HD$nkHNL-L5 IEHHH-LH QP ;U*1LLIX SP EH= @9MLKImu IELP0c=HuGH=?LHGHHIM$HKLH5H81Hh2H/DcLL$LD$LL$LD$@$bfDHLLl$ Ld$(HD$0H)Ht -IH-HH]cI.e1jHF HHD$(HFHD$ H1HL$ MLL IH52muH VbH bHfHFHHD$ oIwHHFc`fLLL$LL$IfDH5yLHVHHD$(IGfDHKHH6-cfKHuHH=LHGHHHtI+HLH5KH81HE1HIcILHIE1wfDH1H~lKcIHIL8LIMoMIoIEHEI/u IGLP0HEIInHMFHEII.uIFLL$LLD$P0LD$LL$I@H;G+H;LL$LD$8LD$LL$HIHhLH I@LMLD$ H-}LD$H Q;UP /1LD$LLAIֽLD$H QP E=H @9MMImMIELP0H~HcI)^FH~sHp^kcL-L5IEHHcH-kH QP ;Um1LLIϼX SP EH= @9MDLjImu IELP0c\1LLLD$LD$HIH}MH|jvcM 1[H=6~LD$茻LD$]LD$SLD$HuHbH5~H83LD$rfc謻@$I@@=2LHt$ LL$(LL$LD$Hl$ زLD$LL$HIHmuHELL$HLD$P0LL$LD$I)uIALD$LP0LD$M@$RLuL1L藽IHXH<|HcTbImH=| ݺHuHH5|H8·LD$VLD$@$賽xLHt$ LL$(LL$LD$Hl$ LD$LL$HIHc{IM'H$cwH6{HecVH1LI?L1L?IHcVH={蹹HuHͷH5{H8螶HzIMaH^Lc1LLLD$譻LD$HIHHMzME1HcFH=zLD$.LD$LD$LD$HuHH5zH8յLD$艵HyIMHyc1Mo(ff.AWAVAUATUHSHHHLvdH%(HD$81HHD$ HD$(HI|IHF HD$LuH H(hE111HALIHH8u H@LP0ID$H5LHHlIMH5LIHI/uHD$IGLP0LL$L; Ͷ@L; r@u L; u@I)u IALP0Lٷf.zff/LK HT$H5IH{LLL$LL$HH5 I)IALP0LIH/HGP0fIIS IMMH{H {HNHUL vLNL@HHAVH(H5LwH81ʸH wXgH Z g H=πE1=HD$8dH+%(HHL[]A\A]A^A_HHD$L5QH=*IVL辵IHHIAH5HHsLL$LLL$IM(I)u IALP0L=H=IWLNIHHIEH5HH[ LIM Imu IELP0IGE1H;8 H;sH;tHc[IH MtLhHcI$MdHHcHIDIGLMP LD$гH-ALD$H Q;UP  1LLD$LAHD$蘳LL$LD$H QP E=p9|MV I(yI/uIGLL$LP0LL$HI9FfLLLL$>LL$HI} I)u IALP0I.u IFLP0L;=L;=\u L;=~0I/ L{ HT$LH5бIH{MHH I/LI'LLL$苳LL$HcrHsv HsaNgI) E1E1IALP0M I. E11IFLP0MtImu IELP0MtI/u IGLP0HHmHEHP0Hr H;gfDH H=|5MLE1D JgHr]HZLH MCH=J|52赲LE1f.KIIG@=HLl$ E1Ld$(HD$0IWH)Ht HBR uMO HLIM MImIELL$LP0LL$ LgHqUHRDI/IGLP0IGLP0&I@LL$LP0LL$n4@ADIGLP0LISLŅ hQH(H5 LIHVIHD$ H3MHD$(Lt$ HD$n{HNL-L5tIEHHH-\H QP ;U*1LLIX SP EH= @9ML[Imu IELP0 qg=îHuGH=OLHGHHIM$H[LH5nH81'HxoB H?-gLL$LD$LL$LD$@$bfDHLLl$ Ld$(HD$0H)Ht =IH-Hn HgI.e1jHF HHD$(HFHD$ H1HL$ MLL vxH5}aH fng+ H (gHfHFHHD$ IwHn Hg`fLLL$LL$IfDH5LHVHHD$(IGfDHm[ HXFgf[HuHH=LHGHHHtI+HLH5[lH81辮HmE1 HgILHIE1wfDHl1 H|gIHIL8L萮IMoMIoIEHEI/u IGLP0HEIInHMFHEII.uIFLL$LLD$P0LD$LL$I@H;W+H;LL$LD$HLD$LL$HIHhLH I@LMLD$H-LD$H Q;UP /1LD$LLAILD$H QP E=H @9MMImMIELP0Hj HgI)^FHj HngL-L5IEHHc H-{H QP ;Um1LLIߨX SP EH= @9MDLzImu IELP0 !h\1LLLD$(LD$HIHiM HzgM 1[H=FjLD$蜧LD$]LD$cLD$HuHrH5+jH8CLD$rf Fh輧@$I@@=2LHt$ LL$(LL$LD$Hl$ LD$LL$HIHmuHELL$HLD$P0LL$LD$I)uIALD$LP0LD$M@$RLuL1L觩IHXHLh HmgT Yg ImH=hHuHH5hH8ңLD$fLD$@$éxLHt$ LL$(LL$LD$Hl$ LD$LL$HIHsgIM7 H4"gwHFg H gVH1LI?L1LOIH hVH=gɥHuHݣH5gH8订HfIMq Hn\h1LLLD$轧LD$HIHH]fME1! H  hFH=fLD$>LD$LD$LD$HuHH5fH8LD$虡HeIM Hg1Mo(ff.AWAVAUATUSHHXLfH|$dH%(HD$H1HHD$0HD$8HD$@H IqIHF(HD$ L{ LsH H(hE111HALHHH8u H@HP0Hm H(hE111HALIH H8u H@LP0HEH5HHH, HH"ID$H5jLHH IM LκHLL$褟LL$HIS H;l@L;-@щt$ L;-(шL$ tuImuIELL$LP0LL$H5LϺLL$LL$HI H;L;- L;-L$ ˆT$H+\I)|$ImDt$XL5H=IVL>IHHIEH5HHLHHImu IELP0L5jH=CIVLסIHHIAH5THHLLL$LLL$IMQI)u IALP0IGE1H;{XH; H;HcLL$葠LL$HIMtLHHcD$HEHIlH^DiMcHKDIGLM3EH QH P ;!HL$1LLAIH QHL$P H=' 9 MI.I/ HsH9CLHIHImH+L;=^L;=u L;=&ÉD$I/oL$L-lH=EIULٟHHmHHCH56HHHIMH+L= H=IWLyIHHIEH5vHHLIMImu IELP0IGE1H;&H;DH;Hc AIHMtLhHcD$I$HMdHYHcHIDIGLM!LL$LL$H QH \P ;=1HL$LLLL$AH輝LL$H QHL$P =H @9HI)u IALP0I/u IGLP0HI9F;HLjIHH+uHD$HCHP0LL$I.uIFLL$LP0LL$L; ӜL; y6L; )LLL$LL$I)u IALP0"HD$HT$ MHH5fLH IHxLL$00LL$HII)IALP0fDHH5HIHVIΜHD$0HLc@IH aHaAHMEIHH3HLgH5e]ATL ckH81ڞH +]XEdH ZEdH=EgE1MHD$HdH+%(HXL[]A\A]A^A_H HD$ IALP0|$dL[D$PH\E1G; HD2dE1E1ImuIELL$LP0LL$MtI)u IALP0MtI/u IGLP0MI.IFLP0}DHCLL$(HP0LL$(I)DLxf.cf(zLT$Vf.ncT$f(zdff/tf/tHD$HT$ f(H5Lh IEHxLZIH Imu IELP0HmuHEHP0MI,$ ID$LP0IELP0t$DLLL$(胚LL$(fHIqMI HFHHD$0͓IH5HHV藙HD$8HIMHD$@Lt$0L|$8HD$ IHF(HHD$@HF HD$8HFHD$0SIM~1HL$0MHL cH5Ly3dH Y5H HYc9 H`NzdfDH I?H=cA5(諙Hm4%CHIGLP0fIG@=HLcl$LL$0E1Hl$8HD$@IWL)Ht0HBR uMGLL$HLLL$IMTMPI)FIALP07HCHP0lIELP0RIGLP0IFLP0 @DLD$CH XMF HeI)fIALP0WHCHP0^K@$S蝖HuGH=)LHGHHIMCH5LH5VH81HRWH He$f.+IHWE1E1; HdH+E1HCLL$HP0MLL$jfdH VvH wi; H+HCLL$HP0LL$I)H ZAfDH5HHVuHAHD$@I.fDHF HHD$8HFHD$0VIfDd3fDH U8 kdE1H= `H 8 zkdHLcl$LLL$0LL$HD$@L)Hl$8Ht0LL$HIHKUE1F H6e8HW Lʔf.\zf L-;L=IELM sH QH ޒP ;i HL$1LLAICH QHL$P H= @9\MC LImu IELP0HZT$@ H!df+HuGH=LHGHH IM HÒLH5,SH81菕HSF HeHIG@=H$Hc\$Ll$0E1Ld$8HD$@IWH)Ht0HBR uMOHLHHM`ImUIELP0F@T$ T$Hx ff/L-1L= IELM iH QH ԐP ; HL$1LLAI9H QHL$P H= @9`M LImu IELP0HPRB HdE1F eH RH ImE1E1LHHD$ƐLL$HuWH=MHGHH; LLL$HtIHSLLL$H5PH81LL$HfQE1E1*F H'"ePC eLLL$#LL$IfDHQE1F H$efDMOMMoIIEI/AD$IEMkLCMeLKIIH+uHCLL$HLD$P0LD$LL$IAH;Î H;LL$LD$贎LD$LL$HI L@Lh IALM LL$舎LL$H QH P ; 1HL$LLLL$AINLL$H QHL$P =H @9 M I.LlIFLP0]LL$LL$@$UHEOF H _eH+E1E12HOF HDeHoHc\$LLl$0Ld$8HD$@H)Ht0HHkHNuH Hr`eM[E1#1LL躏IHH_NE1&F H#OeLL-uL=FIELMg 譌H QH P ;* HL$1LLAI}H QHL$P H= @9MLImu IELP0HM^G H[IeH=%NHL${HL$DHHTH5 NH8%HuGH=LHGHHhHHaHLH5LH81HLH He8H谎I2HLE1E1]H HZHeIGLL$LMP0D$IELL$H eL7I躊@$謊@$MoM#I_IEHI/u IGLP0D$HCIߺHHAH5KjL YH OH8HU17_)dAXM~MINIHI.uIFHL$LP0HL$HAH;H;CHL$誉HL$HILx1HHHX HL$/HL$HIbImILHD$IELP0LL$3HJE1H H|jeHJXH HUCeHgJ1H H.f_1LLLL$}LL$HH6HJH HeI)L-,L=IELMdH QH χP ;uHL$1LLAI4H QHL$P H= @9MqLImu IELP0HKII HfH$IJ HIHpHG@ HdKHG< Hqd$菆HHGQ= HN<djIOH= HHL$aHL$yM*H?H:H5GH8 $fDL1LCIHHFB HdPH=yGHL$τHL$蛅HuHH5hGH8耂LL$LL$@$LLL$iLL$LHt$0LD$0LD$LL$Ll$8iLL$LD$HIHFMLؽF HսýfeHA@=HHt$0HL$L|$0H\$8{HL$HII/uIGLL$LHL$P0LL$HL$H+uHCLL$HHL$P0LL$HL$I@$HBEE1 F H>e/H1LLL$IbHHDG He[H=EHL$ڂHL$L1LIH蛆H莆IHwDML;F H8&ve'H=EHL$LL$SLL$HL$VH"DLE1F HE1λ|e 1LLLL$/LL$HIVLL$‚LL$HuHрH5DH8LL$wHHt$0HL$L|$0H\$8HL$HIHgCI.H H+eT~H=CHL$FHL$mHCI HϺfHuHH5CH8~L1LIHTHBH H}keH1LH4H}BIDH HA/fIEHIE?E1M H;BIE1H He%H&H5BH8}HAMLF HoeHA; H{d.HAIfH HcQeH EA fAUATIUSHHHHndH%(HD$1H$MHHHH^EH OEHHHH!OH?L HLHL@HH~UHJH5@H81hH @XfH ZuM gfM H=JE1ۀHD$dH+%(pHL[]A\A]fHVHk H51HEH{H IHtHHmuHEHP0DHHHFLH$MyHdH$fH? HfHmt'H H=J5uHEHP01HL IIH5Lv2vH _?f$M H !ffLxIH4H5LHVI~HtH$IE zfAWAVAUATUSHHXLfH|$dH%(HD$H1H }HD$0HD$8HD$@HI1IoHF(HD$L{ LsH H(hE111HALIH\ H8u H@LP0H͵ H(hE111HALHHrH8u H@HP0ID$H5LHHC IM% HELT$HH5µHH LT$IM LL׺LL$ LT$xLT$LL$ HI/ H;{L;-[{ L;-|{L$ ImuIELT$ LLL$P0LT$ LL$H5LϺLT$ LL$`xLL$LT$ HIb H;#{L;-z ÉL$L;-z I*I)2<ImL$L5H=IVL{IHu HIEH5޴HH LIM ImuIELT$ LP0LT$ L5H=|LT$ IVL {LT$ HI HIAH5HHC LT$(LLL$ LL$ LT$(IMF I)uIALT$ LP0LT$ IGE1H;xC H;y H;zHcLT$ yLT$ HHMtLpHcD$HELT$HHlHtDiMcHJDIGLM[yL-xLT$H QA;UP \1LT$ HLAHD$"yLL$LT$ H QP AE=H59MVH+I/HywI9Bc LLLL$ LT$LT$LL$ HIs I)rI*XL;5QxL;5wu L;5xI.V HD$HT$ILH5wLH IHxLL$ LL$HHSI)IALP0HrH5)HIHVInxHD$0HnLc@IH d=HU=AHMEIHHvHCH59ATL GH81zzH 8XhH Z yh H=ECE1xHD$HdH+%(xHXL[]A\A]A^A_HvHD$IALP0LwD$H#81E1E1HޯK ЯhImu IELP0MtI.u IFLP0MtI/u IGLP0HPH+FHCHP07IBLL$ LP0LL$ I)%DL wf.8?f(zLT$vf.?T$f(zff/HD$HT$f(H5uLh IEHxL HHImu IELP0I,$IHm)HEHP0IELP0L$/DI hH w6EH B4H 5+H=@5vI,$1ID$LIP0HX@LLT$(LL$ uLT$(LL$ HI[qMIHFHHD$0oIH5HHVtHD$8H IMHD$@Lt$0L|$8HD$IHF(HHD$@HF HD$8HFHD$0nIM~1HL$0MHL ?H5y#(yhH 55լH ֬Ȭ F1H hH 4H=v?H 5uMcE1vK h@vIIGLT$ LLL$P0LT$ LL$DIG@= HLcl$Lt$0E1Hl$8HD$@IWL)Ht0HBR uMGLT$ HLLT$IMU MI.xIFLT$ LLL$P0LL$LT$ UIBLP0IALT$LP0LT$uIFLP0HCLT$ HLL$P0LT$ LL$D @DLrÅFH3MΪT H˪iI)zIAL1P0DLT$ LL$QqLT$ LL$@$IfDtLT$IfDH21E1E1HJ<K .hI*| IBLL$LP0LL$MM,iH1LLT$IbL1L{iIHU i gHuHeH5(H8cH=(f3H'E1IT HiU1HHHL$hHL$HIH'IE1E1HQE1@T 2iH=(HL$deHL$=HL$+fHL$HuH:dH5'H8 cHL$ybH&1IžT HizMH +AfDAWAVAUATUHSHHHLvdH%(HD$81HdHD$ HD$(HI|IHF HD$LuH H(hE111HALIHH8u H@LP0ID$H5LHHlIMH5מLJaIHI/uHD$IGLP0LL$L; c@L; c@u L; cu@I)u IALP0L ef.!-zff/LK HT$H5cIH{LLL$LL$HH5 I)IALP0LIH/HGP0fIIS IMMH(H (HNH>L 2LNL@HHLbAVH:/H5|$H81eH K$X]jH ZY ]jY H=.E1mdHD$8dH+%(HHL[]A\A]A^A_H)bHD$L5H=ZIVLbIHHIAH5KHHsLL$LLL$IM(I)u IALP0L=H=IWL~bIHHIEH5HH[ LIM Imu IELP0IGE1H;+`8 H;IasH;aHc[FaIH MtLhHcI$MdHHcHIDIGLMP LD$aH-q`LD$H Q;UP  1LLD$LAHD$`LL$LD$H QP E=p9|MV I(yI/uIGLL$LP0LL$H_I9FfLLLL$nLL$HI} I)u IALP0I.u IFLP0L;=_L;=_u L;=_0I/ L{ HT$LH5_IH{M0HH I/LI'LLL$`LL$HcrH  HjI) E1E1IALP0M I. E11IFLP0MtImu IELP0MtI/u IGLP0HHmHEHP0H)  HޗjfDH ٗϗH=*5A`MLE1D jHH|H }sH=\*5b_LE1f.{aIIG@=HԚLl$ E1Ld$(HD$0IWH)Ht HBR uMO HLIM MImIELL$LP0LL$ jHHtI/IGLP0IGLP0&I@LL$LP0LL$n4@ADIGLP0LISL]Ņ XkQHXWH5ɘLIHVI]HD$ H3MHD$(Lt$ HD$n\HNL-L5\IEHH\H-[H QP ;U*1LLI[X SP EH= @9ML苲Imu IELP0 j=[HuGH=LHGHHIM$H[LH5H81W^Hr Ho]jLL$LD$[LL$LD$@$bfDHLLl$ Ld$(HD$0H)Ht mIH-H" Hד kI.e1jHF HHD$(HFHD$ &UH1HL$ MLL &H52`H Mj[Y H XFMjHfHFHHD$ TIwHA  Hj`fLLL$]LL$IfDH5LHVZHHD$(IGfDH HvjfYHuHH=LHGHHHtI+H"YLH5H81[H?E1 HjILHIE1wfDH1 HjIHIL8L[IMoMIoIEHEI/u IGLP0HEIInHMFHEII.uIFLL$LLD$P0LD$LL$I@H;W+H;WLL$LD$xWLD$LL$HIHhLH I@LMLD$LWH-VLD$H Q;UP /1LD$LLAIWLD$H QP E=H @9MMImMIELP0H! H֏5kI)^FH HkL-L5sIEHHc:VH-UH QP ;Um1LLIVX SP EH= @9MDL説Imu IELP0 gk\1LLLD$XXLD$HIHM H%kM 1[H=vLD$TLD$]LD$ULD$HuHSH5[H8sRLD$rf |kT@$I@@=2LHt$ LL$(LL$LD$Hl$ LLD$LL$HIHmuHELL$HLD$P0LL$LD$I)uIALD$LP0LD$MFT@$RLuL1LVIHXH|F HC1jT jPWImH=LSTHuH1RH5H8QLD$SLD$@$VxLHt$ LL$(LL$LD$Hl$ LD$LL$HIHIMg HdRIMH5L\MIH I/u IGLP0L;%PL;%Ou L;%ODI,$u IT$LR0EL$Qf.<D$zL%H=IT$LsPIHHIGH5 HH8 LIMI/u IGLP0D$|NIH HNI9F LLtIH Imu IELP0I.u IFLP0L;%NL;%Nu L;%N DI,$u ID$LP0E Lc D$H$H5MI$H{LIHI,$uID$LP0HmHEHP0fDII+IMMHH HNHNL LNL@HH\MAVHdH5H81 QH [XkH &Z  k H= E1}OHD$8dH+%(HHL[]A\A]A^A_H9MH$L-H=rIULNIHHID$H5bHHJLIM I,$u ID$LP0L-0H= IULMIHHIGH5JHHBLIM!I/u IGLP0L-RKM9l$HL贝IH I,$u ID$LP0M9n LL聝IH I/u IGLP0I.u IFLP0L;-KL;-Ku L;-K`DImu IELP0E Ls H$HH5JIH{M@IH I."IFLP0LLAƅnH E1E1) HlI,$u ID$LP0MMDMI,$Au ID$LP0MtImu IELP0MtI/u IGLP0H =3H=65"LE1EDH9 E1' HkH H=5уTLMLfDH I) H lfMIH E1h) HeS lI/IGLMP0L KAŅHF E1E1 + HYlKLJAąH Ȃ0 HłmImlME1E1IFLP0HDH5yLIHVIIHD$ H M7HD$(Ll$ H$@[IHHS I* HlIHuGH=LHGHH IM4HHLH5 H81KH 0 HlDHHuGH='LHGHH% IMGH3HLH5H81JHP I+ H$lDHF HHD$(HFHD$ fBH1HL$ MLL H5RM^H k H kxfHFHHD$ AIHE1H0 HE3lfDLXJIH5 LHVmGH'HD$(IGfDE1+ &lHHI/nM'LIIFHuHH=GLHGHHj HtILL$LT$HIHMMu+ Huu=l=LHt$ LD$ LD$L\$Hl$(ՋL\$LD$HI'HMMDu0 HAu/ulfLT$LL$;LT$LL$@$9?LHt$ LL$Ld$ L|$(CLL$HI%HMMt0 HtMtlL1L>IHHIqt, Hnt\tdllH=8H $:H $#\;HuHp9H5)H8A8L1L=IHtH'Is1 HssmH=H $ :H $:HuH8H5H87HMMs+ H~slsMlImpI)fI*ME1MMǺ0 lRL $L:L $HMM+ SlH=HL$LL$49LL$HL$1LLLL$>IMH5)pL2IH I/u IGLP0L;%Y5L;%4u L;%!5DI,$u IT$LR0ELd6f.|D$zL%GnH= nIT$L5IHHIGH5`rHH8 LIMI/u IGLP0D$3IH HQ3I9F LL贅IH Imu IELP0I.u IFLP0L;%04L;%3u L;%3 DI,$u ID$LP0E Lc D$H$H53I$H{L]IHI,$uID$LP0HmHEHP0fDII+IMMH=H .HNHL LNL@HH2AVHH5H81J6H XmH flZWl5 Ilm5 H=fE14HD$8dH+%(HHL[]A\A]A^A_Hy2H$L-kH=kIULF3IHHID$H5lHHJLIM I,$u ID$LP0L-pkH=IkIUL2IHHIGH5oHHBLIM!I/u IGLP0L-0M9l$HLIH I,$u ID$LP0M9n LLIH I/u IGLP0I.u IFLP0L;->1L;-0u L;-1`DImu IELP0E Ls H$HH5(0IH{M耝IH I."IFLP0L2AƅnH6E1E1i HiimI,$u ID$LP0MMDMI,$Au ID$LP0MtImu IELP0MtI/u IGLP0H }isiH=5bi1E1EDHyE1@i H=i+imH ,i"iH=@5i1MLfDH!Ih Hhhmf2IHE1h HhhmI/IGLMP0L`0AŅHE1E1Jh HGh5hnKL0AąH>h HhgnImlME1E1IFLP0HH)H5jLIHVI/HD$ H M7HD$(Ll$ H$@.HHIZg HWgEgm[.HuGH=fLHGHH IM4H-LH5\H810Hf HffcnD-HuGH=gfLHGHH% IMGHs-LH5H81?0HIWf HTfBfmDHF HHD$(HFHD$ 'H1HL$ MLL =H5r2-^H me5 H eemxfHFHHD$ /'IHE1e HeseenfDL/IH5IeLHV,H'HD$(IGfDE1些 mHDeHeeI/nM'L/I+HuHH=dLHGHHj HtI H;H;:AA H;YA I)uIAT$LP0T$I/uIGT$LP0T$EjH+PL-qRH=JRIULHHz HHCH5;SHH HIMr H+u HCHP0L5 RH=QIVLxIH HIGH5%VHH LHHL I/u IGLP0L5-L9sHHiIHH+u HCHP0M9uLL_iIHI/ImuIEL $LP0L $L; L; u L; I)^HD$HT$ILH5WLH IHxL $ZL $HIGI)hI,$MfHmu HEHP0HD$HdH+%(HXL[]A\A]A^A_fHHF(HD$HF H$HHD$L f.8D$zH<$f.$z L-OH=OIULRIH HIGH5SHHG LHHF I/u IGLP0$\IH HH9C/ LHTgIHH I/u IGLP0H+u HCHP0L;5L;5wu L;5sI.u IFLP0HD$ $D$HT$HX H5@HHxHIHlH+HCHP0HCT$HP0T$IHCLL$HP0LL$?H[N HXNFNoI)Au IALP0I/u IGLP0IDMI)Au IALP0MtImu IELP0MtI/u IGLP0H MMH=5M7MtI,$u#E1ID$LMP0H@E1HHD$HF H$HLL$ T$?LL$ T$H(…HOE1E1M HMLoH+HCL $HP0L $DH HAHHHH5UL H81xH XXoH LZL wLXo H=E1fDIHHHHFHHD$0 IM~TH5JPLHVnHtHD$8IM~.H5KLHVHHs HD$@IMa HD$8Lt$0H$HD$@HD$HH HIHHI?IAHHF(HHD$@HF HD$8HFHD$0 IuH i }o)K H=`H K K}o\HKH HxH9HXHHqH~1H;T{HH9uHHJH5HWH81lHME11HJE1uJ gJo\f.HKJ HHJ6JoH 7J-JH=d5Jhf.HJH HxH9HXH HqH~1H;THH9uHHJH5HWH81lHIE1I H~I1jIobH[I HXIFIo fkIUHQI HIIoI)IALM1P0LL$I6fDH$IGLP0L $XIALP0LÅ~HLE1E1HHyH kHohfDIALP0fLL $4L $HWE1H HH HpHfHH9HuH;fHG HGGo[fHF HHD$8HFHD$0 I~fDHIHH5KLHVH'HD$0If.KHHC G H GFoHuGH=FLHGHH HHTH LH5H81wHF HF}FApB HfH]F HZFHFo c HuGH=ELHGHHK IM0H LH5dH81HE HEEo CpH EH EEH+E1E1HI5LHHE1XE HUECEofD[ HuHH=DLHGHH HtIH LH5[H81HD HDDFpHID HDDofDH{D HxDfDHp1fLHkLkMLKIEIH+uHCLL$ HP0LL$ IAH; CH; NLL$ LL$ HI LhLx IALM< LL$(L\$ s L\$ LL$(H H Q;P  LL\$(1LLL$ AI4 LL$ L\$(H H =p4@9#Mf I+LICLP0fHAI1C HCBo@LCM(HKIHH+uHCHL$HL$P0L$HL$HAH;B 8H; kHL$L$4 L$HL$HIj L@HEHh HALMM LL$H $ HqLL$H Q;H $P  1LLL$HH $AILL$H QH $P =p4@9Mf I)HIALP0fVpH AIH AA H+uMEMMuIIImuIEL$LP0L$IFH;UH;.yL$L$HIL@Lx IFLM L$HL$H Q;P C 1LT$LLAH$pL $LT$H QP =H @9MW I*MIBL $LP0L $L-@L5EIELMHZH QP ;1LLAIH QP H= @9MLZ]Imu IELP0H? H??pTfHLx? Hu?c?p`fD *pL-?L5DIELM1H[H QP ;v1LLAIH QP H= @9ML[\Imu IELP0H> H>>pUH~> H{>i>p^1HL$0ILL H5 |wH eDo*> H '>>DoHQH5H8"aH6H5H8FHHH9HuH;dIA@=LHt$0LL$ Ll$0L|$8LL$ HIImuIELL$ LP0LL$ I/uIGLL$ LP0LL$ LHA@=HHt$0LD$0LD$H $Hl$8"H $LD$HII(uI@H $LP0H $Hx@$IF@=rHt$0LL$LD$0L|$8L$HII(uI@L $LP0L $I/uIGL $LP0L $M@$HLL$(L\$ LL$(L\$ @$LL$H $LL$H $@$ H ILHt$0LL$ Ll$0L|$8RLL$ HI-HL~; H{;i;oHHt$0LD$0LD$H $Hl$8QH $LD$HIUHSMH; H;;\pDLL$L$LL$L$@$\Ht$0LL$LD$0L|$8 QL$HIFHMLMH:: }:pz`L1LIHH~H: HE:3:pH=j;HuHOH5H8 L1LaIHH9 H99pHuHH5H8H=uv1LLL\$(LL$ LL$ L\$(HIHzMA9 H>9,9o!HPL9 H99oH&MH8 H88jpLL$L$L$LL$H_HH5H8LL$L$;H=mLL$(L\$ L\$ LL$(HMLMHP8B8 48p1H=LL$H $bH $LL$MH˸ppV1LHLL$H $[H $LL$HI1LLL$3L$HI8HԿMM7 H77pH=_L$L$LL$H $zH $LL$H7HH5:H8RLL$H $L$;L$H[HGH5H8L$ H;H;ZAA H;yA I)uIAT$LP0T$I/uIGT$LP0T$EjH+PL-3H=j3IULHHz HHCH5[4HH HIMr H+u HCHP0L5+3H=3IVLIH HIGH5E7HH LHHL I/u IGLP0L5ML9sHHJIHH+u HCHP0M9uLLJIHI/ImuIEL $LP0L $L; L; u L; I)^HD$HT$ILH5?LH IHxL $zL $HIGI)hI,$MfHmu HEHP0HD$HdH+%(HXL[]A\A]A^A_fHHF(HD$HF H$HHD$L@f.XD$zH<$!f.9$z L-1H=0IULrIH HIGH55HHG LHHF I/u IGLP0$|IH HH9C/ LHtHIHH I/u IGLP0H+u HCHP0L;5L;5u L;5sI.u IFLP0HD$ $D$HT$HX H5(HHxH8zIHlH+HCHP0HCT$HP0T$IHCLL$HP0LL$?H{/| Hx/f/qI)Au IALP0I/u IGLP0IDMI)Au IALP0MtImu IELP0MtI/u IGLP0H ..H=75.WMtI,$u#E1ID$LMP0H@E1H HD$HF H$HLL$ T$_LL$ T$HH…HoE1E13.| H0..qH+HCL $HP0L $DH hAHHHIH5"UL !H81H XGqH -Z- -Gq H=E1 fDIHHHHFHHD$0IM~TH5j1LHVHtHD$8IM~.H5-LHVhHs HD$@IMa HD$8Lt$0H$HD$@HD$HH 5H6IHHI?IAHHF(HHD$@HF HD$8HFHD$0IuH y lqI,y H=H ?,-,lq\H-H HxH9HXHHqH~1H;T{HH9uHHJH5HWH81HݳME11H+E1+y +nq\f.Hk+z Hh+V+yqH W+M+H=5<+hf.H ,H HxH9HXH HqH~1H;THH9uHHJH5HWH81HݲIE1*z H*1*{qbH{*| Hx*f*q fIUHq;*| H8*&*qI)IALM1P0;LL$I6fDH$IGLP0L $XIALP0LÅ~H߱LE1E1H)) )qhfDIALP0fLL $TL $HwE1>) H;)))rHfHH9HuH;fH(| H((q[fHF HHD$8HFHD$0&I~fDHIHH5,LHVH'HD$0If.kHHc-(} H*((q3HuGH='LHGHH HHTHLH54H81H' H''0rBHfH}'~ Hz'h'q HuGH='LHGHHK IM0HLH5H81H8' H&&q 2rH &H &&H+E1E1HI5LHHE1x& Hu&c&qfD{HuHH=&LHGHH HtIHLH5{H81H/% H%%5rHI% H%%qfDHѭ% H%%7r1fLHkLkMLKIEIH+uHCLL$ HP0LL$ IAH;CH;$NLL$ LL$ HI LhLx IALM< LL$(L\$ L\$ LL$(H HQ;P  LL\$(1LLL$ AITLL$ L\$(H H =p4@9#Mf I+LICLP0fHaI1&$ H#$$q@LCM(HKIHH+uHCHL$HL$P0L$HL$HAH;b8H;kHL$L$TL$HL$HIj L@HEHh HALMM LL$H $ HLL$H Q;H $P  1LLL$HH $AILL$H QH $P =p4@9Mf I)HIALP0fErH "IH "" H+uMEMMuIIImuIEL$LP0L$IFH;UH;NyL$L$HIL@Lx IFLM L$H6L$H Q;P C 1LT$LLAH$L $LT$H QP =H @9MW I*MIBL $LP0L $L-!L5'IELM HzH QP ;1LLAIH QP H= @9MLz>Imu IELP0H  H  qTfHѨL  H  pr`fD rL- L5&IELM1 H{H QP ;v1LLAIH QP H= @9ML{=Imu IELP0H HrUHԧ Hr^1HL$0ILL H5蜚wH 3qJ H G53qHqH5H8BaHVH5H8'FHHH9HuH;IA@=LHt$0LL$ Ll$0L|$8LL$ HIImuIELL$ LP0LL$ I/uIGLL$ LP0LL$ LHA@=HHt$0LD$0LD$H $Hl$8BH $LD$HII(uI@H $LP0H $H@$IF@=rHt$0LL$LD$0L|$8L$HII(uI@L $LP0L $I/uIGL $LP0L $M @$HLL$(L\$ LL$(L\$ @$LL$H $LL$H $@$ -H ILHt$0LL$ Ll$0L|$8'3LL$ HI-HפL HqHHt$0LD$0LD$H $Hl$82H $LD$HIUHsMH7 H4"KrDLL$L$LL$L$@$\Ht$0LL$LD$0L|$8@2L$HIFHMLMH wrz`L1LIHHh HeSqH=/[HuHoH5(H8@L1LIHH& HrHuH H5ƣH8H=v1LLL\$(LL$ LL$ L\$(HIHMa H^Lq!HpL7 H4"qHFMH  HYrLL$L$ L$LL$H_HH5ʢH8LL$L$;H=LL$(L\$ L\$ LL$(HMLMHpb Tr1H=0LL$H $H $LL$MH˸_rV1LHLL$H ${H $LL$HI1LLL$SL$HI8HMM HrH=L$L$LL$H $H $LL$H7HH5ZH8rLL$H $L$[L$H[HgH5 H88L$ H;H;zAA H;A I)uIAT$LP0T$I/uIGT$LP0T$EjH+PL-H=IULHHz HHCH5{HH HIMr H+u HCHP0L5KH=$IVLIH HIGH5eHH LHHL I/u IGLP0L5mL9sHH+IHH+u HCHP0M9uLL+IHI/ImuIEL $LP0L $L; L; u L; I)^HD$HT$ILH5LH IHxL $mL $HIGI)hI,$MfHmu HEHP0HD$HdH+%(HXL[]A\A]A^A_fHHF(HD$HF H$H HD$L`f.xD$zH<$Af.Y$z L-%H=IULIH HIGH5?HHG LHHF I/u IGLP0$IH H1H9C/ LH)IHH I/u IGLP0H+u HCHP0L;5L;5u L;5sI.u IFLP0HD$ $D$HT$HX H5HHxHX[IHlH+HCHP0HCT$HP0T$IHCLL$HP0LL$?Hј HysI)Au IALP0I/u IGLP0IDMI)Au IALP0MtImu IELP0MtI/u IGLP0H H=q5wMtI,$u#E1ID$LMP0H@E1H)HD$HF H$HLL$ T$LL$ T$Hh…HE1E1S HP>sH+HCL $HP0L $DH AHHHH5BUL AH81H X6sH Z 6s H="E1+fDIHHHHFHHD$0IM~TH5LHVHtHD$8IM~.H5$LHVHs HD$@IMa HD$8Lt$0H$HD$@HD$HH UHVIHHI?IAHHF(HHD$@HF HD$8HFHD$0IuH  [si  H=ՠH _ M [s\H!H HxH9HXHHqH~1H;T{HH9uHHJH5"HWH81HME11H E1   ]s\f.H  H v hsH w m H=ٟ5\ hf.H) H HxH9HXH HqH~1H;THH9uHHJH5"HWH81HIE1  H 1 jsbHѓ  H  us fIUH[  HX F wsI)IALM1P0[LL$I6fDH$IGLP0L $XIALP0LÅ~HLE1E1H    shfDIALP0fLL $tL $HE1^  H[ I tHfHH9HuH;fH1  H  |s[fHF HHD$8HFHD$0FI~fDH0IHH5 LHVH'HD$0If.HHM  HJ 8 sSHuGH=LHGHH HHTHLH5TH81H HtBHfHӐ Hs HuGH=/LHGHHK IM0H;LH5H81HX" H s !tH 'H H+E1E1HI5LHHяE1 HsfDHuHH='LHGHH HtIH2LH5H81HO H$tH!I HsfDH H&t1fLHkLkMLKIEIH+uHCLL$ HP0LL$ IAH;CH;DNLL$ LL$ HI LhLx IALM< LL$(L\$ L\$ LL$(H HQ;P  LL\$(1LLL$ AItLL$ L\$(H H =p4@9#Mf I+LICLP0fHI1F HC1s@LCM(HKIHH+uHCHL$HL$P0L$HL$HAH;8H;kHL$L$tL$HL$HIj L@HEHh HALMM LL$H $@HLL$H Q;H $P  1LLL$HH $AILL$H QH $P =p4@9Mf I)HIALP0f4tH IH  H+uMEMMuIIImuIEL$LP0L$IFH;UH;nyL$ L$HIL@Lx IFLM L$HVL$H Q;P C 1LT$LLAH$L $LT$H QP =H @9MW I*MIBL $LP0L $L-L5IELM)HH QP ;1LLAIH QP H= @9MLImu IELP0H HsTfHL H_t`fD tL-L5IELM1*HH QP ;v1LLAIH QP H= @9MLImu IELP0H HtUH Ht^1HL$0ILL H5{wH "sj H gU"sHH5:H8baHvH5H8GFHHH9HuH;IA@=LHt$0LL$ Ll$0L|$8LL$ HIImuIELL$ LP0LL$ I/uIGLL$ LP0LL$ LHA@=HHt$0LD$0LD$H $Hl$8bH $LD$HII(uI@H $LP0H $H@$IF@=rHt$0LL$LD$0L|$8L$HII(uI@L $LP0L $I/uIGL $LP0L $M-@$HLL$(L\$ LL$(L\$ @$LL$H $LL$H $@$ MH @ILHt$0LL$ Ll$0L|$8GLL$ HI-HL HsHHt$0LD$0LD$H $Hl$8H $LD$HIUHMHW HTB:tDLL$L$LL$L$@$\Ht$0LL$LD$0L|$8`L$HIFHMLMH ftz`L1LIHH HssH=O{HuHH5HH8`L1LIHHF H tHuH-H5H8H=v1LLL\$(LL$ LL$ L\$(HIHM H~ls!HLW HTBsHfMH* H'HtLL$L$*L$LL$H_H1H5H8LL$L$;H=LL$(L\$ L\$ LL$(HȂMLMH tvt1H=PLL$H $H $LL$MH˸NtV1LHLL$H $H $LL$HI1LLL$sL$HI8HMM H|tH=L$L$LL$H $H $LL$H7HH5zH8蒽LL$H $L${L$H[HH5@H8XL$< HKMH H BtHMLMH otHL HsMMME1Pff.AWAVAUATUSHXL5HnH|$dH%(HD$H1HLt$0H$HD$8HHD$@HdH:4HKHHjHD$LvH H(hE111HALIH H; B Hk H(hH<$E11HA1HH H;ؽ ID$H5LHH IMe HELL$HH5bHH LL$IMo LϺLLL$蚺LL$HH H;bH;AA H;$At_H+H5LLL$/LL$HH> H;H;AA H;A I)uIAT$LP0T$I/uIGT$LP0T$EjH+PL-H=IUL>HHz HHCH5HH HIMr H+u HCHP0L5kH=DIVLؼIH HIGH5HH LHHL I/u IGLP0L5L9sHH IHH+u HCHP0M9uLL IHI/ImuIEL $LP0L $L; 9L; ߺu L; I)^HD$HT$ILH5LH IHxL $NL $HIGI)hI,$MfHmu HEHP0HD$HdH+%(HXL[]A\A]A^A_fHHF(HD$HF H$H)HD$L耻f.D$zH<$af.y$z L-EH=IUL貺IH HIGH5_HHG LHHF I/u IGLP0$輸IH HQH9C/ LH IHH I/u IGLP0H+u HCHP0L;51L;5׸u L;5sI.u IFLP0HD$ $D$HT$HX H5HHxHxbMLMH] ^vHbLX HuMMME1Pff.AWAVAUATUSHXL%7HnH|$dH%(HD$H1HןLd$0HD$8HTHZH8HHHeIH eHlHIHHI?UIH5FaH8L CoA1蹢H aXwH Zb wb H=\lE1,HD$HdH+%(HXL[]A\A]A^A_fDHHD$H5 H(hE111HALHHH8u H@HP0HEH5_HHHIMH5'L蚛HHI/u IGLP0H;WH;u H;DH+u HSHR0ELdf.|gD$zL%GH= IT$L賞IHHIGH5`HHLIM'I/u IGLP0D$輜IHHQI9E LLHH I.u IFLP0Imu IELP0H;0H;֜{H;nHfAŅ H+u HCHP0E HD$D$HT$H5HX HHxHGIHS H+u HCHP0HmQHEHP0BfDL%H=IT$L-HHHHCH5HH"HIMqH+u HCHP0L%ZH=3IT$LƜIHHIGH5sHH[LHH:I/u IGLP0L%{L9cHHIH H+u HCHP0M9e LLIH I/u IGLP0Imu IELP0L;5)L;5Ϛu L;5DI.u IFLP0E HD$HT$HH5Lh IEHxMeIHy ImIELP0HF HD$LffDHYHD$fHțAŅ(H[E1E1 H=wH+u HCHP0LMHH+Au HCHP0MtI.u IFLP0MtI/u IGLP0H 8.H=f5蠛E1DIHCHHHXIH, @HD$8Ld$0HD$@@HZE1 H*wH zH=f5iML;HyZI@ H=+9w릐[IHAZE1 H;wI/1IGLMP0WLAąNHY H5xI.ZM1E1IELP0H訛IL蘛I@HYE1E1E HB0wHF HHD$8HFHD$0薒HE1HL$0ILL dH5LH Yvb H vfHFHHD$0IMH5|LHVHjHD$8ID$R{HHsXI: H7%Hw;HuGH=LHGHHT HH-HӖLH5LI HwhHuH"H5LH8봐H=LHL$HL$*L1LIHHKI H|j@xH=FLHL$蜉HL$hHuH|H55LH8MHLKLM H {wI.II)RE1W1LLLD$ EL\$LD$ HI~HJLM HxH=kKHL$(LL$ 輈LL$ HL$(PMM w/1LLLL$ 賋LL$ HHjLM w7HHI*u IBLP0I/u IGLP0He~I9FiHLIHH+u HCHP0I.u IFLP0L;-EL;-~ L;-  L{Å Imu IELP0oHD$HT$ ILH5~Lh IEHxMIHHQ@HzfDImu IELP0H H=K5շXMtI,$`E1ID$LMP0HfHyH5ɽHIHVI~HD$0HLc@IH CHCAHMEIHHC}H KH5u?ATL sMH81H ;?XxH Z x H=JE1]HD$HdH+%(@HXL[]A\A]A^A_H}HD$ qIELT$(LP0|$LT$(LLT$\~LT$D$H|>ME1E1HDE13 %yImfMtI*u IBLP0MtI/u IGLP0MI.IFLP0@HCLT$(HP0ImLT$(C@L}f.Ef(z|LT$^}f.vET$f(z ff/lf/ HD$HT$ f(H5zLh IEHxLbIH^ Imu IELP0I,$5MHmHEHP0fDIBLP0t$m DLLT$(|LT$(fHIqMIHFHHD$0uIH5HHV{HD$8HiIMdHD$@Lt$0L|$8HD$ I&HF(HHD$@HF HD$8HFHD$0cuIM~1HL$0MHL GH5.yxH ;5H  H ; xi H=*GH _Mx{nH!H HxH9LHXHHqH~1H;T&HH9uHxHJH5"@HWH81|H:LE1E1HŲE1E11 E1xfDH+uHCLT$HP0LT$M]ImRIELT$LP0LT$9fH)H>HxH9HXH$HqH~1H;TmHH9uHwHJH5"?HWH81{H9HE1E1HűE1E11 y@{HIGLP0IG@="HLcl$LT$0E1Ld$8HD$@IWL)Ht0HBR uMOLT$HLLT$IMMI*xIBLP0iHCLT$HP0LT$HD$IELP0LT$jIBLP0IFLP0 @DLLT$SxLT$D$KHs8M:H7% z HCHP0v@$uH;uH5<H8 t wHuGH=LHGHHIMJHvLH57H81fyH7H~l0zE1yI Hq7E1E1E1H9+ yyH <7H  H+HCHP0@H H;AfDH5HHVvHHD$@IfDHF HHD$8HFHD$0oI9fDyCfDHLcl$LLT$0LT$HD$@L)Ld$8Ht0LT$HI H36E1HyHH6˭ HȭyH H=nA5vf.tHuGH=7LHGHHGIM7HCtLH54H81wH`5*H'yZ+tH Ltf.<z4 L-KL=IELM sH rX SP ;* HL$1LLAISsHL$X SP H= @9MN LImu IELP0Hj44 H1IydfIG@=HHc\$Ll$0E1Hl$8HD$@IWH)Ht0HBR uMGyHLHH>MImIELP0x@T$rT$H ff/vL-L=IELM qH jqX SP ;a HL$1LLAIqHL$X SP H= @9M L`Imu IELP0H2 HiyE1yH2rHoaImhE1E1LxtH@[qHuHH=LHGHH HtIMHpLH5[1H81sH2E1E1E1HשɩyfD y&fLsIH1ME1uHrE1]yMWMMoIIEI/XD$IEML{MLCIIH+uHCLD$HP0LD$I@H;}on H;oLD$soLD$HI LxLh I@LMm LD$LoH nLD$X S;P  1HL$(LLLD$AHD$oHL$(LT$X LD$SP =H @9 M I.LIFLT$LP0LT$LT$nLT$@$H/E1E1Hy H/E1H}yH)Hc\$LLl$0Hl$8HD$@H)Ht0谽HHHe//H,DzME11LLtpIHtH/E1E1ݦHڦȦy+L-,L=}IELM dmH lX SP ; HL$1LLAI4mHL$X SP H= @9!M LImu IELP0HK.HzEH=.HL$2lHL$lHuGH=LHGHHHHHlLH5,H81_oH-zHwe+zlHbHjH5L.H8diGHxjH5!2H8IiHHE-E1E1E1H -zTHoI@IGLT$LMP0D$IELT$Q2z%ck@$/LnIFHHH9HuH; k]fMoM2I_IEHI/u IGLP0D$HCIߺj@$H@HH9PHuH;j>HHiAH5,jL 9H 0H8Hu71gm_xAXM~MINIHI.uIFHL$LP0HL$HAH;ivH;?j'HL$iHL$HILx1HHHX HL$HD$ZaHL$LT$HII*I IBLP0H*E1E1HmzH*HqRzRH*_H\JzE1LLLT$kLT$HHHK*H]zI*L-ZL=IELMhH QH gP ;HL$1LLAIbhHL$X SP H= @9M9LImu IELP0Hy)CH@.zsLT$GhLT$HHRfH5 *H8#eLT$fH=)HL$LT$gLT$HL$QI@@=LHt$0LD$L|$0Ll$8^LD$HI3I/uIGLT$LLD$P0LT$LD$ImuIELT$LLD$P0LT$LD$Lf@$H>( H'y8L1LYiIH{H'ȟ HşEyfHH' H~1yH=Z(HL$eHL$L1LhIHpHi'3 H0eycLiI/fH)H?dH5'H8cfHuHdH5'H8bfH='HL$dHL$ULT$LD$NeLT$LD$@$]h[LHt$0LD$L|$0Ll$8谴LD$HIH`&E1L$H!yrHA@=HHt$0HL$L|$0H\$8 \HL$HII/uIGHL$LP0HL$H+uHCHL$HP0HL$IJd@$H%E1iHfTyH1LLT$I2[dHHS%HzMH=%HL$:cHL$aL1LNfIHfHfIH$E1LHyd`H=]%HL$LD$bLD$HL$H}$LE1E1HEE14&z1LLLD$eLD$HILD$cLD$HuH)aH5$H8_LD$tHHt$0HL$L|$0H\$8HL$HIH#IE1HntzH#\HYGLz(H1LHSbH|HK#HzEH=#HL$2aHL$+L1LFdIHYH"HzI*E1I|H"IE1vHsazH_H5V#H8n^Hj"E1L.H+y|H=" Hy7IH ]&AH!IE1H}zH^H5"H8]iAWAVAUATUSHHxLfH|$dH%(HD$h1H_HD$@HD$HHD$PHD$XHI I HF0HD$(HC(LkHD$HC H$H H(hE111HALIHVH8u H@LP0Hw H(hH<$E11HA1HH;H8u H@HP0H, H(hH|$E11HA1HHH8u H@HP0ID$H5SLHHIMBHELL$HH5HHLL$IMtLϺLLL$W[LL$HIH;^@L;]@ L;]t$ʈT$  PI(uI@LL$ LLD$P0LL$ LD$HCLL$ HLD$H5XHHLD$LL$ IMLLLL$ZLL$HI H;S]@L;\@ L;]|$ʈT$  tdI(H5LLL$ZLL$HI3H;\L;\D$ L;\ ЈD$ I/uIGLL$8LLD$0P0LL$8LD$0I)z I. |$  I({ DL$Eu L-H=IUL]IHHI@H5pHHL$LL$IMI(uI@L $LP0L $L-0H= L $IUL\L $HIyHI@H5HHLL$LL$L$LL$IMI( IFE1H;7Z H;U[H;[Hc L $N[L $HIMtLxHcD$I$L $Md@HHEIlIFLM [L{ZL $H QA;P 1L\$LLLL$AH$ZL\$L$H LL$QP A=HO 9MXIm I. H"YI9A2LLLD$L $|L $LD$HIqI( I) L;5YL;5YO L;5YB L2[$xI.u IFLP0D$EL-H=̒IUL`ZIHHIAH5HHXLL$LLL$IM I)u IALP0L5H=\IVLYIH!HI@H5HHLD$LLD$IMI(u I@LP0IGE1H;WH;XmH; YHcLD$XLD$HIhMtL@Hc$HEIl@HHI\IGHH+H $gXLWH QA;H $P 1L\$LLH$2XL\$L $H QP A=H @9 MI.uIFL $LP0L $I/uIGL $LP0L $HmVI9EXLLL $̨L $HII)u IALP0Imu IELP0L;5DWL;5V/ L;5W" LzX$I.u IFLP0<$~L5=H=IVLWIH.HIEH5HHmLIM5Imu IELP0L-֏H=IULCWIHCHIAH58HHLL$LLL$IMI)u IALP0IFE1H;TH;V H;`VHcLL$ULL$HI MtLHHc$I$Md@HHI\IFLM!L$UL+UL$H QA;P  !1L\$LLL$AIUL\$L$H QP A=H @9M I(u I@LP0I.u IFLP0HSI9GLL/IHImu IELP0I/u IGLP0L;5TL;5QTL;5oTLUAŅI.u IFLP0E|HD$HLIH5RILp IHxAVHT$8l/Y^IH' I.uIFLP0fDI,$HmDHNIH5HIHVTHD$@HcLc@IHH }HNHATH L@HRH5-L -#H81VH X8{H Z8{H={ E1UHD$hdH+%(!HxL[]A\A]A^A_ÐHRHD$(JIFLD$0LP0|$ LD$0TLLD$ TLD$ D$6H<E1E1NH{I(u,E1f.I@LP0MtI/u IGLP0MtI.u IFLP0MImIELP0qfIALD$0LP0I.LD$0rDL0Sf.Hf(z H|$\$ Sf.$\$f(zH<$T$\$Rf.\$T$f(zf/3 f/I f.z HD$HT$(f(H5OL@ IHxLL$ L$HI I(I@LP0I@LP0DL$E{Ha1)KH&o{@H H=A5sRI,$u ID$LP0HtHmu HEHP0H+H+!HCHP0fLLL$8LD$0~QLL$8LD$0#INHH|JcH>f.HF0HD$XHC(HD$PHC HHD$HHCHD$@JIIt{I~=IMzHD$HLl$@H$HD$PHD$HD$XHD$(MI^H5fHHVPHD$HHIH5HHVOHD$PHIkfDH J`{E1H=1H N@J2`{PHQLH~{fI@L $LP0L $ fIFLL$LL$P0LL$L$I@LL$LP0LL$IF@=HcT$L|$@E1Ld$HHl$PH)IVHt@HBR uMFL $=HLL $IMMRI/HIGLL$LL$P0L$LL$'fIALP0nI@L $LP0L $LfIELL$LL$P0LL$L$ @DH5qHHVMHtHD$XIM\1HL$@MHL 9H5Sn4${H R5H cfD‰$DHFHHD$@gGILL$L$LLL$L$@$H NH{qfOIxLLL$8LD$0NMLL$8LD$0Hi E1E1E1H1#N{I)uIAL$LP0L$M&I( DOLL$IfDE1N{H H I.uIFLL$LL$P0LL$L$E1E1M[n‰$DN{@{NLL$ LD$I@fHcL$LL $L|$@H)Ld$HHt@Hl$PcL $HIH E1ۃ\H؃ƃk|fN{JHuGH=WLHGHHvIMHcJLH5 H81/MH J\HG5R| D$EJ\$HH7 OH{IG@=Hc$LD$@E1Hl$HH\$PH)IWHt@HBR uMOL$OHLL$IMMJI(@I@L $LP0L $)D\$D$WI\$T$HHC  PH {T$$D$I$T$HL$H QH{L-L=2IELM,9HLGH QP A;~L$1LLAI HL$H QP AH= @9 ML蠞Imu IELP0H THՀ{E1E1E1\T|H H fDLL$JL$IL-L= IELM~GLFH QP A;L$1LLAIFL$H QP AH= @9ML耝Imu IELP0HVH{LL$H$FL$LL$HucH=EHGHH LL$LL$HtI9HGFLH5LL$L$H81 ILL$L$HRE1E1E1H \~W|L-aL=IELMEL EH QP A;0L$1LLAIiEL$H QP AH= @9MLImu IELP0HJ~XHG~5~| LLL$L$OHLL$L$IfDH)ME1}\H}E1}Y|E1E1Y;|M~MMnIIEI.D$IEMMqMMQIII)uIALT$LL$P0L$LT$IBH;CH;+D LT$L$CL$LT$HIiLpL@ IBLML$CL CL$H QA;P <1L\$LLL$AI_CL\$L$H QP A=H @9 MImMIEL$LP0L $DUL $BL $@$N{SH7E1E1{\H{{|H E1{\H{{y|\|E1E1E1Hc $LLD$@LD$H)Hl$HHt@H\$PڑLD$HIHT{^HQ{?{|MMHzfDL-{L=IELM#AL:AH QP A;EL$1LLAIAL$H QP AH= @9ML0Imu IELP0Hzz]Hwzez|P1LLCL $HIuHlE1E10z\H-zz|H=L\$L $I@L $L\$AHuGH=yLHGHHcIMH@LH5H81uCHy^Hy{y|fL $@L $HH>H5ZH8r=L $fIF@=Hc$LL$@E1Ld$HH\$PH)IVHt@HBR uMFL $\HLL $IMMKI)AIALP02HE1E1E1Hxx^zx|`LLL$BLL$I~?HuGH= xLHGHHA IMH?LH5H81AH3w^Hww|HHH=H5!L !jAH :H H81AAZ{A[sIFL $LMP0IEL $D$4>@$JLLD$ALD$IE1^|MGMMwIII/$IFM=@$=@$gHH<H5L  jAH 7H H81@[{]rMuMIMIHImuIELL$LH $P0H $LL$HAH;< H;O=^ LL$H $0H Xa~H ZfZKff=fa~fH=3.HD$dH+%(&H 1[]A\A]A^ÐHN(HS HsHD$dH+%(H L[]A\A]A^A@HIkiM IHFHH$&'IH5hHHV,HD$HCIMH4$HT$HL$^fIuzHF(HHD$HF HD$HFH$&IM~1HL 1MH52HByO~H *5dH ddffDH HA4fDH5dHHV,HtHD$ImfH&H5jHIHVI+H$HLcHF HHD$HFH$%IHH<*H H5njL lAHH81-Y^E~'H eAIAWIAVAUATUSHhH|$ Ht$H$HL$dH%(HD$XHjcH(hE111HALHH8IH+c H(hH<$E11HA1HHH8IGH5ZcLHHIMHCH5.cHHHIMLLm&HHAH;:)@H;-(@щt$ H;-(AtZHmH5cL&HHWH;(H;-s( H;-(L$ AI,$}ImE!Hm L$H<$)f.$zTH|$)f.D$zff/D$f/$y4$f/5>HD$ Hh HELhHD$H;'L5+aH=aIVL(IH\HID$H5`HHl LIMI,$uID$LL$LP0LL$IAE1E1H 1&HL$(H9!H;L'NH;'Hc>LL$D'LL$HIMtLpHL$IcAT$HcHILH 'IDHIALM$LL$LD$&LD$LL$H QH O&P ;z%1HL$ LLLD$LL$AI&LL$LD$H QHL$ P =HR9ZMV%I(I)I<$HS_At$I|$ HUL5 `HD$ID$LHHT$ HD$0Q'HD$Hk"H@HT$ HHBH|$HHD$HL"HUL5_HLHT$ &IH"H@HT$ HH"LHIH"H@H;D$(MQMIQIHI)uIAHT$(LLT$ P0LT$ HT$(LHLT$(HT$ =vHT$ LT$(HI)I*IuIBLD$ LHT$(P0LD$ LL$(I)uIALD$ LP0LD$ I(u I@LP0'E1H|$HD$ ~KHl$(Hl$H\$8LLt$0f. $D$L!IHH9uHl$(H\$8H|$ |HD$L5XdH@LMo&'$H QH #P ;&H $1H|$LAI#H QH $P H= @9M'HL$HH$HHuH|$HGP0MJ&Imu IELP0I<$ID$LP0H@LP07H@HP0lID$LP0ImxIELP0EnfL-9\H=\IUL#HHHHEH5]HHKHIMJHm_ L5[H=[IVLE#IH9HIEH5_HHJLIM Imu IELP0H IFE1HL$(H9H;"H;c"HcJ "IHMtLhHcD$HL$HHILH8\HcHIDIFHHLL$!LL$H QH !P ;91HL$LLLL$H}!LL$H QHL$P =H 9bH1I)u IALP0I.u IFLP0HD$(I9D$HL.rIH2Hmu HEHP0I,$u ID$LP0L;5 L;5N u L;5p : ‰D$I.u IFLP0T$L-YH=YIUL!IHHID$H5yZHH|LIM{I,$ L%NYH='YLL$IT$L LL$HHHHEH5^HHLL$HLL$IMjHmO IF1H;D$( H;zLH;Hcl LL$rLL$HI HtHhHcD$H $LL$HHYIL@HHITIFHH` #LL$H QH P ;p"1HL$LLLL$ILL$H QHL$P =H @9MK"ImuIELL$LP0LL$I.uIFLL$LP0LL$HD$(I9ALLLL$oLL$HI2I,$uID$LL$LP0LL$I)u IALP0L;-L;-L;- L"D$VImu IELP0D$FL%VH=VIT$LNIH HIAH5WHH "LL$LLL$IM "I)u IALP0L%qVH=JVIT$LIH"HIAH5\HHF#LL$LLL$IM"I)u IALP0IFE1H;D$(#H;H;HcHH%MtL`H $HcD$HHXHL@HHHTIFLM&YH QH P ;7'1HL$HLAH$(L $H QHL$P =H @9 M 'HmuHEL $HP0L $I.uIFL $LP0L $HD$(I9E#LLL $lL $HI $I)u IALP0Imu IELP0L;5<L;5wL;5jLrŅ$I.u IFLP0%HD$ Ht$HLLh IEHxM[IH%ImIELP0DHEHP0L$DHiHD$HE1S)HSS E1E1fHmuHEL $HP0L $MtImuIEL $LP0L $MtI.uIFL $LP0L $MtI)uIALP0H SSH=|A5R{I/u IGLP0HtH+u HCHP0HD$XdH+%((HhL[]A\A]A^A_HEHP0:HEHP0H &E1H=H nR`R&RRzHq;R'H8R&R fKIH1Q)HQQf IHE1E1Q)HQQI,$1ID$L $LP0HL $fH VQH WQIQ)I,$tImIELP0 ID$LP0Im@IF@=HD$HcT$Ll$@E1HD$HHQHD$PH)IVHt@HBR uMNM#HLHH#MIm IELP0LuH5QLHt$XHt$HIH@HH LLHIHHUH5PHHt$HT$Ht$HIH@HT$HH LHIHH@H; MFM IVIHI.uIFHT$LLD$P0LD$HT$LHLD$HT$>gHT$LD$HIGI(IuI@LL$LP0LL$I.uIFLL$LP0LL$I)u IALP0  $D$LI6LH$ID$L5ULMyH QH P ;EHL$1LLAIIH QHL$P H= @9>M2I,$uID$LP0M7Imu IELP0H<$IHQHmu HEHP0I/ @KDID$LP0>IALP0$LD$HMM6HMM>NfDMqM MaII$I)u IALP0ID$MAfDH=yHL$LL$LL$HL$[IF@=kH$HcT$Ld$@E1HD$HHiAHD$PH)IVHt@HBR uMNfHLIM+MI,$ID$L $LP0L $fHuGH=w=LHGHHIMHLH5H81OHj=8Hg=U=<8L $oL $H*H{H54H8LL $ @$L=I|H&E1E1<8H<<>8HLH8y*aLL$LL$HuTH=T<HGHHLLLL$HH2H[LH5L $H81#L $8AH f4<H 1<#<HLH8dHL$H0;H;;{*H$HHD$HHu H|$HGP0H ;;H=5;-H E18CH $>@$@HLL$LL$IH1LLLD$LL$LL$LD$HI}HOA;H;:T*I)H$IALD$LP0LD$oInHMfHEI$I.0D$ID$M[@$H=kHL$ LL$LD$LD$LL$HL$ T-LL$L$mL$LL$H HtH5-H8ELL$L$@$NMqMLIiIHEI)u IALP0HEH;l H; gHULp1HHL` HD$LD$HII(II@Hl$LP0LL$H~E1E98HB909~L $L $@$HH8^H/8H88)I,$H$H88H88cH88H88c^H$HcL$LLd$@HD$HH;HD$PH)Ht@NIHDHc-8:H*88ۃE1Mt1LLkLL$HIH 78H77nQH-$8L->HELM \H QH P ;c H $1HLAH-H QH $P H= @9 H HTHmu HEHP0HE79H 76LLt$1L[IIH$HIH=H $H $dHL$HH$HHu H|$HGP0H*s6HH=H$Hb6P6**iHuGH=5LHGHH IM:HLH5jH81H5:H55ƒH=HL$LL$LL$HL$hL $L $HHH5H8L $HHH5XH8pI@@=LHt$@LD$Ll$@Hl$HELD$HI ImuIELD$LP0LD$HmuHELD$HP0LD$M*aLLL$LL$I:ăII@$MIѸ*HD$LL$HH= 4HGHH LLL$HtWIAH I,$u ID$LP0)1LLZI,$IHLH5)L $H81L $HջE13:H33ǃIFLL$LMP0D$ID$LL$sH=5HL$HL$!HZME13:H3 3ɃlLLL$/LL$IMfMInI$HEI.u IFLP0D$HEIH1LHHiH2.H}2k26NHY2+HV2D2"'rH/UHuHiH5"H8:H=H $@H $ZMuMMeII$ImuIEL $LP0L $ID$H;HH;L $?L $HHV Lp1HLLH IH HmMHEHP0}H1LHHHI10H10VMIָ)m HuH H5ٹH8H0:H00/H=zH $H $XH1LHHHR02HO0=0v Ha+0:H(00LD$LL$LD$LL$@$HM/:H//'HXHH5H8=@H=yH $H $LHt$@LD$Ll$@Hl$HELD$HI8H]MM!/6H/ /!1HLkIHH :E1H $H55H=,/1HHH9LHmu HEHP0H.;H.n.6QH\.<HY.G.[H=#H $zH $:DH,HTH5 H8%HE@=*Ht$@HLt$@Ld$HIH I.u IFLP0I,$u ID$LP0I&T@$EHE1s-6Hp-^-ׂH1LHH1LHHHW!-7H- -&:I_H=۵H $2H $HuHH5̵H8IHִ,H$H,,C*MNH1LALL$I}HMMN,6HK,9,NLLL$_LL$HHt$@HLt$@Ld$HfBIH7HI+8H++0ID$@=Ht$@LL $LL$HLt$@L $HHI.uIFL $LP0L $I)u IALP0MILD$L $iL $LD$HHXME1E1H ++6+gH=HL$0LD$LL$,LL$LD$HL$01LLLL$LD$,LD$LL$HI=jH²*8H*w*]VH1LLL$Iy~HHv@*9H=*+*YIH=H $QH $H1LfHHHt$@LL $LL$HLt$@#@L $HH)HԱM):H)) HIE1MHr)d)8V)HzIE1>)8H;)))LLL$OLL$HNH5H8H(;H((2H(:H((H1LIHH5H8LD$L $?HMM[(6HX(F([E1M:!H ZH $*HJM(:H('H M':H''cI)IALE1P0YHگI'8H''H|$LL$LD$HGP0LL$LD$HH5\H8tff.@AVAUIATUSHH H LfdH%(HD$1H$HD$HL$HIIIH HH9AHMEIHHHH5ATL H81^H XH z&Zk&]&H=ɺHD$dH+%(&H 1[]A\A]A^ÐHN(HS HsHD$dH+%(H L[]A\A]A^@HIkiM IHFHH$FIH5(HHVHD$HCIMH4$HT$HL$^fIuzHF(HHD$HF HD$HFH$IM~1HL MH5*HbyH J5%H %%fDH hA4fDH5$HHV5HtHD$ImfH8H5+HIHVIH$HLcHF HHD$HFH$IHH\H İH5jL AHsH81Y^H AIAWAVIAUATUSHhH|$(Ht$HT$HL$ dH%(HD$XH#H(hE111HALHbH8IHF# H(hH|$E11HA1HHYH8OIFH5t#LHHaIMcHEH5H#HHHuIMwLLIHH;TL;% L;%L$tWI,$H5#L$IHHH;L;% ÉL$L;% I/RIm\WI,$L$XH|$f.D$zH|$HD$HH|$!ff/D$#\$f/L%x!H=Q!IT$LIHHIEH5(HHi LIMh Imu IELP0D$IH HHD$0I9G!LL8IHc#Imu IELP0I/u IGLP0L;%[L;%u L;%#I,$u ID$LP0%HD$(L@ IH@HD$HD$ H;L-A H= LD$(IULLD$(HI)HID$H5HH=*LD$(LLD$(IM)I,$IEE1H;D$0EH;lH;HcLD$(dLD$(HI-MtLxHL$ HcÃLD$(HcHILH'IDHIEHH.LT$ LT$ LD$(H QH qP ;/1HL$8LLLD$(LT$ ILT$ LD$(H QHL$8P =H @9#M/I*uIBLD$ LP0LD$ ImuIELD$ LP0LD$ I<$uID$LD$ LP0LD$ At$I|$ LD$(H3LD$(M|$HHD$ IPHHHT$8;LD$(HI)H@HT$8HH LD$(LLLD$(HI)IPHLD$(HHHT$8LD$(HI*H@HT$8HH"LD$(LLLD$(HI*H@H;D$0MYMIQIHI)u(IALD$8LHT$0L\$(P0L\$(HT$0LD$8LHLD$8L\$0HT$(4HT$(L\$0HLD$8I 4I+Iu(ICLD$8LLT$(HT$0P0LT$(LL$0LD$8I)uIALD$0LLT$(P0LD$0LT$(I*uIBLD$(LP0LD$(LD$01H|$ LD$0HD$(~]Hl$0Hl$Lt$Lt$LD$Ld$8IH\$ DD$LH7KIL9uHl$0Lt$LD$Ld$8H|$(LD$IELD$L=#HH/LD$LD$H QH P ;D/1HL$LLLD$IdLD$H QHL$P =H @9D$M>1ImuIELD$LP0LD$M/I/uIGLD$LP0LD$I<$#I(u I@LP0I.$  fDH@LP0lH@HP0IGLP0ImIELP0@L%H=bIT$LIHaHI@H5RHHLD$0LLD$0HHI( L%H=IT$LIHHIEH5HHLIMImu IELP0H >ID$E1HL$0H9KH;LH;Hc IIHMtLhHcD$HL$HHILH|HcHIDID$LM#L\$L\$H QH _P ;<1HL$8LLL\$AHD$LD$L\$H QHL$8P =H 9M4 I+uICLD$LP0LD$I,$uID$LD$LP0LD$HD$0H9CLHLD$O0LD$HInI(u I@LP0H+u HCHP0L;=L;=mu L;=9 ‰D$I/u IGLP0T$&L%H=IT$L;HHHHCH5HHgHIMGH+ L%nH=GL\$IT$LL\$HI+HI@H5HHL\$8LLD$LD$L\$8IMI( ID$E1H;D$0 H;AH;Hc LD$8L\$L\$LD$8HI="MtL@HcD$HL$L\$HHIL@HHITID$HH#,L\$H QH P ;$1HL$LLL\$HL\$H QHL$P =H @9Hj$ImuIEL\$LP0L\$I,$uID$L\$LP0L\$HD$0I9C LHL\$-L\$HI H+uHCL\$HP0L\$I+u ICLP0L;=L;= L;= L+D$ I/u IGLP0D$Q!L%H=IT$LXIH"HICH5HH$L\$LL\$IM#I+u ICLP0L%{H=TIT$LIHt%HICH5HH`&L\$LL\$IM_&I+u ICLP0ID$1H;D$0&H;H;HcJIH[*HtHXHL$HcD$HHIL@HHITID$HH*L\$\L\$H QH P ;,1HL$LLL\$HD$!LL$L\$H QHL$P =H @9gM,I+uICLL$LP0LL$I,$uID$LL$LP0LL$HD$0I9E'LLLL$*LL$HI)I)u IALP0Imu IELP0L;=(L;=cL;=VL^Å(I/u IGLP0^*HD$(Ht$ HLLx IHxM0IHU+I/IGLP0ID$LP0L$9@L8LD$HՙM1E1H~E1I(&MtI,$uID$L\$LP0L\$HtH+uHCL\$HP0L\$MtI+u ICLP0H H=A5|I.u IFLP0HtHmu HEHP0HD$XdH+%(0HhL[]A\A]A^A_ID$LP0I@LP0LMhL=5LD$ LLuLD$ HI:"H@HHLD$ LLLLD$ HI!"MhHLD$ HLLD$ HIN"H@HHLD$ LLLLD$ HI5"H@H;D$0iI_H\IWHHI/uIGLD$(LHT$ P0HT$ LD$(HHLD$(HT$ M'HT$ LD$(HI)H+IuHCLD$ HP0LD$ I/uIGLD$ LP0LD$ ImuIELD$ LP0LD$ LD$ D$Ht$H|$ILIID$LD$ L-HLD$H&mLD$H QH P ;(1HL$LLLD$I4LD$H QHL$P =H @9M%I,$uID$LD$LP0LD$M%ImuIELD$LP0LD$LLD$LD$HI&jH H Hى5 H=LD$(MLD$tI,$uID$LD$LP0LD$E1Hn d Z I("1E1I@L\$LP0L\$H Q~E1H=H    ~uH H  ~fIHє H  ~hfI~H U H R D I/!1IGLP0E1fDHI H  ~I/AuIGL\$LP0L\$Im`E1IELD$LL\$P0LD$L\$M,1ID$@=HD$HcT$Ll$@E1HD$HH HD$PH)IT$Ht@HBR uMD$%HLIM%MtImiIELD$LP0LD$Pf. @DLD$WDHCL\$HP0L\$ LHc؅HM1k Hh E1S efDI@L\$LP0L\$!ID$LD$(LP0LD$($fDID$@=HD$HcT$LD$@E1HD$HH HD$PH)IT$Ht@HBR uMT$LD$L\$(HLLD$L\$HHP(MI(I@L\$LP0L\$ukHuGH=LHGHH( IMHLH5lH81H E1H L-1 L%IEHHiH QH P ;fHL$1LLI:H QHL$P H= @9MfL%I,$u ID$LP0HPHHD$ LLD$(L|$@HD$HHHD$PH)Ht@3LD$(HIe#MI/IGLD$ LP0LD$ M}MI]IHImHCIݺ@‰D$@LD$L\$LD$L\$@$YfDHA H~fHD$HcL$LLl$@HD$HHHD$PH)Ht@IHHЎHDE1M@I,${HjHs=H:(~ CHuGH=LHGHHwIMmHLH5DH81HHHL-L% IEHHf;H QH P ;HL$1LLI H QHL$P H= @9 ML"I,$u ID$LP0H"H~f.LLD$0LD$0H^fDE1ۺHČ"L-L% IEHH)H QH P ;HL$1LLIH QHL$P H= @9 ML!I,$u ID$LP0HH~IE@=HD$ L|$@E1HD$HHBHD$PH)Ht@IELH@ uMULD$ "HLALD$ IMHg1lH.&MtI/MDImAuIELD$LP0LD$H 5LLD$0LL$(eLL$(LD$0HI9'HˊHoImX$I)WMfDI*oIBLD$LP0H H>54LD$yfHQE1H%fDL(I HuGH=LHGHHxIMFHLH5 H81oHHu0WMl$MM|$IEII,$u ID$LP0D$IGMnfIALD$LLT$P0LT$LD$M@LIHE1H2IEMMHIE^"E11f5LcMLkI$IEH+uHCLD$HP0LD$IEH;H;3LD$LD$HI=L`L@ IEHH<L\$L\$H QH P ;j1HL$LLL\$InHL$L\$X SP =H @9&MI+LICLP0@MGMAMOIII/uIGLL$8LLD$P0LD$LL$8IAH;H;LL$8LD$LD$LL$8HH-L@Lh IALMLL$mLL$H QH P ;1HL$8LHLL$AI3LL$H QHL$8P =H @9MH+MNHCHP0?L\$L\$@$HE1HHcIHMtLxHcD$I$HMdHYHcHIDHEHH蚟L= H QP A; 1LHHoH QP AH=%9H_ImHmu HEHP0HɝI9F HL,HHH+u HCHP0I.u IFLP0H;-H;-Ou H;-qkÉD$Hmu HEHP0L$7L-H=IULIHHIFH5zHHLIMI.L5PH=)IVL轞HHHHCH5HHHHHH+H|$H5HGHHHHHEE1H;C H;aH;Hc3^IH]MtLxHcD$I$Md@HI\HEHHNL\$"L=L\$H QA;P 1L\$LHIL\$H QP A=H @9 MI+u ICLP0Hmu HEHP0H;I9EGLLHH7I.u IFLP0Imu IELP0H;-H;-H;-ޛHPÅHmu HEHP0OHD$HT$LH5њHh HEHxI4IHjHmHEHP0fDHF HD$Ln2fDH9HD$fH訜D$H \H HmuHEHP0H i_E1H=h5KΜI,$gID$LP0WHHcHHHxIH @HD$(Ll$ HD$0@H [E1H=JhH H[{Hxff苝HHE@=HLHc\$L|$ E1Ld$(HD$0HUH)Ht HBR uLEHLHHM]I/SIGLP0DDH ZE11E1H ff.IELP0 @DHHŅHoZE11E1H8*0H+u"E1HCHP0MtI.u IFLP0MtI/u IGLP01HtHmu HEHP0MtImu IELP0H@H+6HCHP0'HxD$DIFLP0fHCHP0*HhIEHcL$HL|$ Ld$(H)H\$0Ht gIHMtI/u IGLP0H+HCHP0D胗@$-f.RHHc\$HL|$ Ld$(HD$0H)Ht HHHXE1LHI7.f.HF HHD$(HFHD$ 薑H%1HL$ IHL dH5rKH X?H fHFHHD$ IMH5|HHVHjHD$(ID$R{HHsW=H:(CHuGH=LHGHH HHHەLH5DVH81觘HVH{@HVH fHE@=HcT$L|$ E1Ld$(H\$0H)HUHt HBR uLEPHLIMH7VH%H VE11E1H}f 軔HuHH=GLHGHH HtIHRLH5TH81HoU1E14H1fH-L-HEHH 蹓L=*H QP A;P 1HLH莓H QP AH= @9H H)Hmu HEHP0HTsHp^L舖HHqTME15H2 H|$FH fDH)TH,qL}MH]IHHmu HEHP0D$HCHݺKD.fM~MMFIII.uIFLD$LP0LD$I@H;ڑ H;5LD$БLD$HI LxHX I@HH LD$詑L=LD$H QA;P v 1LD$LLHtLD$H QP A=H @9H ImMDIELP05DH-L-HEHH L=jH QP A; 1HLHΐH QP AH= @9H HiHmu HEHP0HQH?1HQHvH+AtE1^f.E1=HqQ;H8&fdcL\$ƏL\$@$1LHSHHSHPE1HfDH-L-JHEHH9L=H QP A;1HLHH QP AH= @9HHHmu HEHP0H)PHqH5HHV-H1HD$ IH=P踎HuGH=DLHGHH4IMHPLH5NH81HmO7H4"@HGHPH5 PH8!,L$IbH O1E1HۍHuHH=gLHGHHHtHHNYHVD}HoH HXNE1H  H )NH H|$ H莌@$L}MLuIIHmu HEHP0D$IFLImHIEHEHImHD$u IELP0HD$H@H;ڋH;5SՋHHHhLp HD$H@HH賋L=$H QP A;1H|$HH膋H QP AH= @9nHH+Ll$HCHP0HLfHcQGjHfL0H-,1LHL\$|L\$HIL۸7H-OL-HEHH臊L=H QP A;&1HLH\H QP AH= @9HHHmu HEHP0HwKAH>,y@$7I@@=LHt$ LD$L|$ H\$(LD$HH5I/uIGLD$LP0LD$H+uHCLD$HP0LD$MnH=hKL\$辈L\$-qL\$肉L\$HYHH5FKH8^L\$9@$QHH1L|HH_H!JHiLD$菈LD$@$6A҈HuHH5JH8跅DH=iJćwLHt$ LD$L|$ H\$(LD$HHkHfILM*H'ąH9IE1HH1LHHD$H@@=H|$Ht$ Hl$ Lt$(~HH.Hmu HEHP0I.u IFLP0Ll$HH1L͉HHpHrH<H9';H=I^/HuHCH5HH8H1LUHHHGĿHBm@$,H=}H؅0詆HuHH5vHH8莃@HGSHP>lI@$HGGHԅH+I(LME1 B1LLLD$6LD$HH3HFM1E1HڅzLD$蜅LD$HuHH5dGH8|LD$H=.GLD$脄LD$lH|$Ht$ Hl$ Lt$(wHHsH,FLl$HܽNm躁H1LIH1L+HH@HEHuH=aF較荄HuHH5ZFH8rHqELl$6H3!^H=EX6H1ELl$HdH|$1HEHH0HuHH5EH8ǀHDLMHuͅLME1OHDLl$PHM;Wff.AWAVAUATUHSHHHLvdH%(HD$81HmHD$ HD$(HIIHF HD$LuH H(hE111HALIHH8u H@LP0ID$H5LHH|IMH5wL~IHI/uHD$IGLP0LL$L; @L; B@u L; c@I)u IALP0L詂f.Jz Jf/LK HT$H5GIH{LLL$|LL$HHA I)IALP0LIH/HGP0f.IIS IMMH}FH nFHNHL 6PLNL@HHAVHNH5 BH81芃H AXH ZH=5NE1HD$8dH+%(HHL[]A\A]A^A_HHD$vL5H=IVL~IHHIAH5۹HHsLL$LLL$IM(I)u IALP0L=H=zIWLIHHIEH5HH[ LIM Imu IELP0IGE1H;}8 H;~sH;4Hc[~IH MtLhHcI$MdHHcHIDIGLMP LD$~H-~LD$H Q;UP  1LLD$LAHD$X~LL$LD$H QP E=p9|MV I(yI/uIGLL$LP0LL$H|I9FfLLLL$LL$HI} I)u IALP0I.u IFLP0L;=v}L;=}u L;=>}0I/ L{ HT$LH5P|IH{M谇HH I/LILLL$K~LL$HcbHl>6H3!3I) E1E1IALP0M I. E11IFLP0MtImu IELP0MtI/u IGLP0HHmHEHP0H=Hn fDH i_H= J5N}MLE1D/HO=H H H=I5u}LE1f. I|IG@=HdLl$ E1Ld$(HD$0IWH)Ht HBR uMO HLIM MImIELL$LP0LL$1HG<HI/IGLP0IGLP0&I@LL$LP0LL$n4@ADIGLP0LIKL{ŅQHtH5YLIHVIzHD$ H3MHD$(Lt$ HD$^;zHNL-sL5IEHHyH-yH QP ;U*1LLIyX SP EH= @9MLImu IELP0V=yHuGH=LHGHHIM$HyLH59H81{H8:HLL$LD$xLL$LD$@$bfDHqLLl$ Ld$(HD$0H)Ht IH-H9|HygI.e1jHF HHD$(HFHD$ rH1HL$ MLL EH5r=,H &9H ְHfHFHHD$ ?rIwH8H`fLLL$zLL$IfDH5ILHVwHHD$(IGfDHQ8H{fwHuHH=LHGHHHtI+HvLH57H81~yH7E1HILHIE1wfDH71QHN<IHIL8LPyIMoMIoIEHEI/u IGLP0HEIInHMFHEII.uIFLL$LLD$P0LD$LL$I@H;u+H;ruLL$LD$uLD$LL$HIHhLH I@LMLD$tH-MtLD$H Q;UP /1LD$LLAItLD$H QP E=H @9MMImMIELP0H5{HxfԇI)^FHy5CH@.L-L5ôIEHHcsH-;sH QP ;Um1LLIsX SP EH= @9MDL:Imu IELP0\1LLLD$uLD$HIH4MOHL:ćM 1[H=5LD$\rLD$]LD$#sLD$HuH2qH54H8pLD$rf|r@$I@@=2LHt$ LL$(LL$LD$Hl$ iLD$LL$HIHmuHELL$HLD$P0LL$LD$I)uIALD$LP0LD$Mq@$RLuL1LgtIHXH 3֪HӪRT>tImH=3pqHuHoH5z3H8nLD$&qLD$@$txLHt$ LL$(LL$LD$Hl$ LD$LL$HIH32IMHۇwH2ЩHͩVH1LI?L1LsIHVH=]2opHuHnH5V2H8nmHm1IM1H.1LLLD$}rLD$HIHH1ME1Hި̨FH=1LD$nLD$LD$oLD$HuHmH51H8lLD$YlH0IM^H[I1Mo(ff.AWAVAUATUHSHHLnH<$dH%(HD$81HlnHD$ HD$(HI{IHF HD$LeH H(hE111HALHH*H8u H@HP0HEH5HHHIM.H5wLjHHI.u IFLP0H;mAH;LmDu H;mmEH+u HCHP0ELnf.6z" ff/" f/6v H$HT$H5lHX HHxHhIHH+uHCHP0fDHmHEHP0fDII IQ MMH2H ~2HNH L F<LNL@HHkAUH:H5.H81oH -XxH ZxH=]:E1 nHD$8dH+%(XHHL[]A\A]A^A_HkHD$L%!H=IT$LlHHHHCH5HH HIMQ H+u HCHP0L5H=IVL'lIH; HIGH5$HH LIM I/u IGLP0IFE1H;i H;j H;NkHcjIHtMtLxIcHEAIlHçMcHKDIFLMZL\$jL\$H QH jP ;/1HL$LLL\$AHojL\$H QHL$P =H9HkI+I.u IFLP0HhI9D$ HL$IH8 H+u HCHP0I,$u ID$LP0L;5iAL;5DiDu L;5ei/EI.E L%H=IT$LjHHAHHCH5zHHHIMH+L5PH=)IVLiHHHHCH5ʨHHHIMH+IF1H;rg/H;hH;hHchIHHtHXIcHEAIlH`McHKDIFLMqKhH QH gP ;VHL$1LLAIhH QHL$P H= @9 M0I/u IGLP0I.u IFLP0HkfI9D$iLL͸IHeImu IELP0I,$u ID$LP0L;5GgL;5f"L;5 gL}hAą-I.u IFLP0EH$HT$HH5nfL` I$HxMaqIHI,$ID$LP0f.HgLcE H(E1E1%Hޟ̟@H+u;E1HCHP0Mt)Imu"E1IELP0MtI/u IGLP0MtI.u IFLP0Mt}I,$uvID$LP0iH '#E1H= 4H >0#"g%H 7'H @H E1H=35ӞVgqhIlIF@=fHTL|$ E1Hl$(HD$0L)Ht IFL@@ uM^dHLAHH(MI/IGLP0%H G&H I. IFLP0IFLP0ICLP0` @9DLeLcE/wDHCHP0fH^H5YLIHVIdHD$ H MHD$(Ld$ HD$oHCHP0 ;dHuGH=ǜLHGHH HHHcLH5<$H81fH$/H.cH L%L-ID$HH$ *cH QH bP ;V H $1LLIbH QH $P H= @9 M L蔹I,$u ID$LP0)҈HLH\$ Hl$(HD$0L)Ht IHHH+HCHP0pf.DL\$&bL\$@$HLL|$ Hl$(HD$0L)Ht 腱HHH:#/HGI,$E19DHF HHD$(HFHD$ >\Hn1HL$ MLL |/H5hFH "hsH p^hHFHHD$ [IIF@=IHԞH\$ E1Hl$(HD$0L)Ht IFLP@ uM^HLAIM)H!1HՉHDH!E1/Hs0fDHcIfH5ILHV`HHD$(IGxfDL%L-ID$HHB _H QH 3_P ; H $1LLI_H QH $P H= @9Mf L2I,$u ID$LP0+k_HuHH='LHGHHG HtIH2_LH5H81aHO E1/H3I,$MMBH E1ؗ,H՗×fDLaIH/H5I,$ E1M~M MnIIEI.u IFLP0IEMAMD$MMl$IIEI,$uID$LD$LP0LD$IEH;]H;] LD${]LD$HI# L@HX IELM Y]H QH \P ; HL$1LLAI)]H QHL$P H= @9"M I/MIGLP0@\@$f.H!M/HӕpHE1HHH/HU1LLL\$^L\$HHHM^/H[I`CL%L-ƜID$HH][H QH O[P ;H $1LLI[H QH $P H= @9mM"LNI,$u ID$LP00H=vHL$L\$ZL\$HL$[HuGH=LHGHHHHH#[LH5H81]1L$'[L$HH3YH5H8XL$eHE1E1IHē1H]IZHuGH=BLHGHHHHHNZLH5H81]Hk51H2 HK]IY@$H&1HےY@$I^HMnHIEI.u IFLP0IEMAMD$MI\$IHI,$uID$LD$LP0LD$HCH;XH;@YLD$XLD$HIL@Lh HCLMXH QH $XP ;HL$1LHAIXH QHL$P H= @9JMI/IIGLP0Hg1HdR1 Hg11H.ۉl1LLZIHH "H ߐ1?L%?L-PID$HHvWH QH VP ;H $1LLIHWH QH $P H= @9&MwLI,$u ID$LP02(M3MH=HL$LVHL$WHH%UH5H8SIE@=Ht$ LLD$ LD$H\$(MLD$HIiI(u I@LP0H+u HCHP0M:V@$Lm+L1LXIH&HS)HΈ&H=H $,UH $fDYHL1L.XIHH+HUHKHSH5oH8R0H=;H $TH $V_UHuHsSH5,H8DRmT@$MUHω1H̉HNH5H8MfAVAUIATUSHH0LOLfdH%(HD$(1H$HD$HD$LD$HIIIHH HNHATHL@HOH5@L @H81RH XH ӈZĈ66H=-QHD$(dH+%(H01[]A\A]A^f.LF0HK(HS HsHD$(dH+%(_H0L[]A\A]A^ IHHHJcH>fHF0HD$HC(HD$HC HHD$HCH$uIIIthI~.IMH4$HT$HL$LD$OMIurH5HHVNHD$HEIH5HHVNHD$HIDH5AHHVNHtHD$IMW1HL MH5(UH@1H $5H 6"HFHH$PHIH8HIH5&HIHVMH$HLcUHHLH H5jL AH>H81PY^@HH@LAH5sjL qH H8H1O_AXIAWAVAUATUHSHHLnH<$dH%(HD$81H,LHD$ HD$(HI{IHF HD$LeHE H(hE111HALHH*H8u H@HP0HEH5oHHHIM.H57LHHHI.u IFLP0H;gKAH; KDu H;-KEH+u HCHP0ELrLf.z" ff/" f/tv H$HT$H5IHX HHxHEIHH+uHCHP0fDHmHEHP0fDII IQ MMHMH >HNHL LNL@HHIAUHH5 H81ZMH XH vZgYH=E1KHD$8dH+%(XHHL[]A\A]A^A_HIHD$L%H=IT$LMJHHHHCH5HH HIMQ H+u HCHP0L5zH=SIVLIIH; HIGH5dHH LIM I/u IGLP0IFE1H;G H;H H;IHcHIHtMtLxIcHEAIlHMcHKDIFLMZL\$iHL\$H QH GP ;/1HL$LLL\$AH/HL\$H QHL$P =H9HkI+I.u IFLP0HFI9D$ HLIH8 H+u HCHP0I,$u ID$LP0L;5_GAL;5GDu L;5%G/EI.E L%qH=JIT$LGHHAHHCH5:HHHIMH+L5H=IVL}GHHHHCH5HHHIMH+IF1H;2E/H;PFH;FHcMFIHHtHXIcHEAIlH McHKDIFLMq FH QH vEP ;VHL$1LLAIEH QHL$P H= @9 M0I/u IGLP0I.u IFLP0H+DI9D$iLL荖IHeImu IELP0I,$u ID$LP0L;5EL;5D"L;5DL=FAą-I.u IFLP0EH$HT$HH5FCL` I$HxM!OIHI,$ID$LP0f.HELcE HE1E1}H}}@H+u;E1HCHP0Mt)Imu"E1IELP0MtI/u IGLP0MtI.u IFLP0Mt}I,$uvID$LP0iH AE1H=5H |||eEH |H ||@H ||E1H=5|EqFIlIF@=fHL|$ E1Hl$(HD$0L)Ht IFL@@ uM^dHLAHH(MI/IGLP0H {H {{I. IFLP0IFLP0ICLP0` @9DLhCLcE ׏wDHCHP0fHHHD$(IGxfDL%QwL-~ID$HHB =H QH <P ; H $1LLIZ=H QH $P H= @9Mf LI,$u ID$LP06k[=HuHH=uLHGHHG HtIH<LH5[H81?HE1u HuuwI,$MMBHE1uHuu[fDL?IH[u HXuFuyI,$ E1M~M MnIIEI.u IFLP0IEMAMD$MMl$IIEI,$uID$LD$LP0LD$IEH;E;H;; LD$;;LD$HI# L@HX IELM ;H QH :P ; HL$1LLAI:H QHL$P H= @9"M I/MIGLP0@:@$f.HMs HssHE1HHHks HhsVs1LLL\$MUHg HgzgJH,H5oH8+fAWAVAUATUHSHxH|$(HT$HL$LD$ dH%(HD$hHfH(hE111HAHIHH8HfH(hH|$E11HA1IHH8H`fH(hH|$E11HA1H$HH<$H?tIFH5fLHHIMIEH5]fLHHIMLֺLLT$)LT$HHH;_,AH;,E H;,DшL$I/uIGLT$0LP0LT$0I*u IBLP0|$ H+u HCHP0EH,IHH|$h,HD$HAH|$O,HD$HhMH|$3"H|$/$HD$LH;D$\%HD$(HX HH@HD$0HD$ H;+L=}dH=VdIWL+HH(HHEH5?dHH*HIMe*HmSH )IG1E1ҺHL$H9E,H;*&H; +HcLT$(*LT$(HIH%MtLPHT$ HcHITHw*EHITIGHHHLL$ [*H )LL$ P ;P .1HL$(LLLL$ H"*HL$(LL$ P P =H @9#H;.I)u IALP0I/u IGLP0H}u HEHP0HbuH} HSL=cHD$ HELHHT$8HD$@*HD$(H+H@HT$8HH H|$(HHD$(H+HSL=cHLHT$8d*IH,,H@HT$8HH_"LHIH,H@H;D$MzMIRIHI*uIBHT$LP0HT$HLHT$yHT$HI3I/IuIGLL$LHT$8P0LL$LT$8I*uIBLL$LP0LL$I)u IALP0*E1H|$ HD$~fH\$8H\$0Ll$HLl$Hl$Hl$Lt$ML|$@LHLH%KIL9t$ uH\$8Ll$HHl$Lt$H|$"HD$(L=hH@LMU0'H &P P ;+1HL$1H|$(LAI['HL$P P H= @9]%M}/HT$(HHD$HHuH|$(HGP0M/I,$u ID$LP0H}k$H+PHCHP0AfH@LP0 H@LP0BHGP0@H+FH<$H5_LT$HGHH"LT$IMLL׺LL$0LT$%#LT$LL$0HHwH;%AH;%E H;%DшL$ H+uHCLT$0HLL$P0LT$0LL$H5#`LϺLT$0LL$"LL$LT$0HH"H;O%AH;$E H;% ˆT$I)IALT$0LP0LT$0DL=I^H="^IWL%HHHHCH5_HH+HIM*H+u HCHP0H-]H=]HUHP%HHHHCH5McHH%HIMH+uHCLT$HP0LT$H "IBE1HL$H9)H; $ H;e$Hc\LT$LL$#LL$LT$HHW!MtLHIcIALtH/^McHJDIBLM"LT$#H #LT$P ;P  $1HL$LHLT$AHs#HL$LT$P P =H9HHl$HmTI*u IBLP0HD$I9GHL*tIH H+I/uIGLL$LP0LL$L; "L; I"u L; k"DI)E L%[H=[IT$L##HH!#HHCH5\HH#HHH#H+L%V[H=/[IT$L"HH$HHCH5`HH$HIM$H+\IB1H;D$H;!qH;!Hc9LT$!LT$HI&HtHXIcIEH[MlAGHHITIBLMV'LT$LL$?!LL$LT$P H ;P (1HL$0LLLL$LT$AI HL$0LT$P LL$P =H @9M(I)uIALT$LP0LT$I*u IBLP0HD$H9E$LHqIH;%I/uHD$IGLP0LL$HmuHELL$HP0LL$L;  L; L;  LLL$5!LL$B%I)u IALP0@&L%XH=XIT$Lc HH 'HHCH5YHH>)HIM(H+uHCLL$HP0LL$L%XH=_XLL$IT$LLL$HH$)HHCH5]HHo*LL$HLL$IM&*H+uHCLT$HLL$P0LT$LL$IBE1H;D$*H;8$H;HcjLT$LL$LL$LT$HH-MtLxH$HcHHTHZEHHHTIBHH.LT$LL$4LL$LT$P H ;P 01HL$0LHLL$LT$HHL$0LT$P LL$P =H @9"HN/H+uHCLT$HLL$P0LT$LL$I*uIBLL$LP0LL$HD$I9AS+LHLL$ynLL$HH+HmuHELL$HP0LL$I)u IALP0H;H;H;HD$_,H+u HCHP0D$]-L%UH=UIT$LIIH.HIAH5VHH/LL$LLL$HHF/I)u IALP0L%lUH=EUIT$LIH0HIAH5ZHH1LL$LLL$IM1I)uIALT$LP0LT$L%TH=TLT$IT$LXLT$HI1HIAH5\HH3LT$0LLL$LL$LT$0IM2I)uIALT$LP0LT$ID$1E1H;D$2H;Q.H;[Hc%LT$LT$HI;6MtLxHcI1LMtELIEHMlLT$0LD$\LD$LT$0HH4I(uI@LT$LP0LT$I,$uID$LT$LP0LT$IBE1H;D$3H;*/H;HcQ&LT$"LT$HI6MtL`HcD$H4$1LLT$Il@HHItLLT$HI%6I/uHD$IGLLT$P0LT$LL$I*uIBLL$LP0LL$HD$H9C3LHLL$jLL$HI3I)uHD$IALP0LT$H+uHCLT$HP0LT$L;L;"L;"LLT$1LT$,5I*u IBLP04HD$(L$LLHt$ LH IHxLL$萩LL$HHb4I)uIALP0I.tImIzHLT$0LT$0fHCLT$HP0LT$HLT$8CLT$8LL$0@H(LcEHMQHQQ)E1E11E1E1>HE1E11HPE1PPH$HtHmu HEHP0MtI,$u ID$LP0MtI/u IGLP0H oPePH=Z5TPMtI.ut1IFLIP0MtImu IELP0H<$tH $HH$HHu HAHP0HD$hdH+%(P3HxL[]A\A]A^A_@E1HE1E11HOOOH$fHE1E11HzOlO^OI"HqE116OH3O!Ol@KIH11E11HNNNE1I/IGLT$LE1LL$P0LL$LT$Ht$H+uHCLT$HLL$P0LL$LT$MtI)uIALT$LP0LT$MI*IBLP0E1E1ɸHY#NH$NNI*5E11E1IB@=HNLL$PE1Lt$XHD$`IRL)HtPHBR uMBLT$LL$(HLLT$LL$HHc(MI)IALT$LP0LT$@HD$HCHP0LL$IALP0:HELT$HP0LT$ @eDHEHP0LLL$LL$AaHLHLLI)4IALE1E11P0fDHCHP0fIG@=HHD$ LT$PE1HD$XHHD$`HcH)HtPIWHBR uMOLT$ *HLLT$ HH*MI*IBLP0HCLT$HP0LT$H`HE1E11H|KnK`K4{HuJH=KLHGHHAILMHLH5yH81H-E1E11HJJJ%DHHE1E11HJJJ>HHE1E11HtJfJXJHH-JL=RHELMH bP P ;wHL$1HLAHHL$P P H= @9HQHXgHmu HEHP0HE1E11HIII`HJIcLLT$H\$PHD$`H)Ll$XHtP_LT$HI%HUH+KHCLT$HP0LT$2IZHMbHI$I*u IBLP0ID$MA1LLLL$ 4LL$ HHL%HL%HH-I/uIGLL$LP0LL$fI)u IALP0H YHOH5EHH=HtHmu HEHP0L%$HHHH-HL{HLIHH@HHj LLHIHH-\HL{LL$HLLL$HIH@HH LL$LLHLL$HIH@H; IjHIRHEHI*uIBLL$ LHT$P0HT$LL$ HHLL$ HT$^HT$LL$ HIHmIuHELL$ HHT$P0LT$LL$ I*uIBLL$LP0LL$I/uIGLL$LP0LL$LL$HL$HT$LH|$0I LHELL$L=YNIALMaLL$ H \ LL$P ;P c1HL$LLLL$AI HL$LL$P P =H @9eMDI)uIALP0M/I,$u ID$LP0Hx HHZb-HaEH^EIH5JED LT$ LT$@$LLT$_LT$HIi-HT$(L%!DL%DDHHD$HHH,I*MJH DDH=5D ZfDNfDH<$LT$If.HqELLL$PLL$HD$`L)LT$HtPLt$XZLT$LL$HHH61CHCC@ fDIBLL$LP0LL$MKVfDIB@=HDIcH\$PE1Ll$XHD$`H)IRHtPHBR uMJLT$&HLLT$IMHXE1E1CHCCQf.H( IHE11E1HBE1BBfDH-CL=JHELMQ H P P ;HL$1HLAH! HL$P P H= @9l HH_Hmu HEHP0H8E1E11HBAA0HD$(HSfHuHH=wAHHGHHHtHHHH5H81N H1E1E1HhAZALAhLLL$ LT$[LT$LL$ HIo-L%C AL%AAI)uIALT$LP0LT$I*xM!I/hIGLP0YfH-AL=HHELMNIH P P ;HL$1HLAHHL$P P H= @9r HH]Hmu HEHP0H0E1E11H???(IHI1?H?1?fH ILL$ FLL$ @$CH-?L=GHELM H zP P ;HL$1HLAHHL$P P H= @9n HbHp\Hmu HEHP0HE1E11H>>>L%>L%>v>-M&I*IBLP0yfMJMIZIHI*HCIںAfDHD$ LLT$PLT$(HD$XHHD$`HcH)HtP8TLT$(HHL%=L%==-MMWM.MgII$I/uIGLT$LP0LT$ID$H;DH;C LT$LT$HHHX LPID$HH;H ,P P ;1HL$HLHD$HL$LL$P P =H @9MHmMRHELL$HP0LL$9DILT$LL$ LT$LL$@$-fD‰D$@HA1E1<H<;> @#fDH1;H;;#HEHP0IIHuGH=F;LHGHHHHHRLH5H81y-L%eH=,;LL%&;5;@$H-i;L=2CHELMH P P ;HL$1HLAHqHL$P P H= @9 HUHXHmu HEHP0HE1E11HQ:C:5:p1LHLT$LT$HHH6E119H99.L% 9L%99{-Hmu HEHP0H 99H=59gfHI)IB@=H$HcL|$PE1HD$XHx<HD$`H)IRHtPHBR uMBLT$LL$HLLT$LL$HHMI/IGLT$LLL$P0LL$LT$H=HL$LT$LT$HL$HuJH==8LHGHHILMHFLH5H81HcE1E11H,888[LT$)LT$HH4H5H8LT$MWMIoIHEI/HEIHHHIE1E1Hz71j7\7xLL$LL$@$bHuJH=6LHGHH1ILMHLH5`H81HE1E16H66HLH8-3IBLL$LAP0HCLL$IںHI;HE1E1E1HS6E676@$SHLH8jHT$(L%66L%55-HHD$HH$H|$(HGP0@$z@$LUMH]IHHmuHELT$HP0LT$HCH; H;eLT$LT$H LP1HHLx HD$LD$HII(HtHD$I@LP0LL$[LL$LL$H&HH5˽H8LL$fH=HL$(LL$ LL$ HL$( >@$HE1]4HZ4H4ČE1E11_LT$LL$LT$LL$@$"HBE1E14H43;,HHH8#-1PH$HcLLL$LT$HD$XH6L|$PHD$`H)HtPILT$LL$HHwH[3HX3F3!1ME11LLLL$LT$LT$LL$HI H,E1E12H22NH5;H=831)HHHEPH+u HCHP0HƺE1E11H22s2H?HLL$H8LL$L%{E2L%B202-I)xIALP0iIGLT$(LIP0HELT$( HuJH=1LHGHHDILMHLH5H81jHE1E11H1v1h1H=DHL$0LT$LL$LL$LT$HL$0)LT$LL$ELL$LT$HHKH5H8LT$LL$HHT$(HHD$HHu H|$(HGP0L%޸.0LH=L%00.ALL|$(1LIIHD$HIID$@=Ht$PLLT$PLT$H\$XGLT$HII*uIBLL$LP0LL$H+uHCLL$HP0LL$MHE1E1HH/// HIH=cHL$HL$&LL$nLL$HuYH=.HGHH=LLL$ILMHLLL$H5]H81LL$H E1E11H.E1.. )t@$"ID$@=CHLLT$H)L|$PHtPLt$XLl$`LT$HHMI/yIGLT$LP0LT$`MIҸ-H=I1.H-1-HLL$LL$I1LLLL$2LL$II)H=HL$LL$LL$HL$uI)u IALP0K- MzMPIZIHI* HCIں$IB@=H$HcT$Ld$PLT$HD$`H)HLHl$XHtPLT$HIMt&I,$uID$LT$LLL$P0LT$LL$HmHELT$HLL$P0LL$LT$LL$LL$HHH5QH8iLL$H1LHHVHGE1E11H,,+\?"IH1LMHHHE1E11H+++|H={HL$HL$kYHKHH5cH8{0MQMIAIHI)HD$uIALT$LP0LT$HD$H@H;6H;Q LT$LT$HILPH|$1HHh $HHJI,$LL$;ID$LP0LL$&HE1E1E1H|*n*`*JH=<HL$HL$oH1LHHHHE1E11H**)@H)H))/IIҸ-IٺmLL$eLL$@$HHH5{H8H=GHL$HL$$H1LHHJHSE1E11H)))K1LHLL$LT$\LT$LL$HHHE1E11H(((:H50H=)1HHHFH+u HCHP0HE1E11HX(J(<(|ZH HjH5#H8;fD;`Ht$PLLT$PLT$H\$XE>LT$HIH1M'H''ELT$LL$LL$LT$HHH5xH8LT$LL$H=9HL$HL$ H=HL$0LT$LL$bLL$LT$HL$00!HuGH=&LHGHH IMHLH5"H81H֮E1E11H&&&HFHH5jH8+H~E1E11HG&E16&(&LLL$NLL$HeLLT$H)L|$PHtPLt$XLl$`H<LT$HHH%H%%1E1MfHC@=,Ht$PHLT$PLT$L|$XLT$HIx I*uIBLL$LP0LL$I/uIGLL$LP0LL$HH>1%H%$ H1LLL$LT$HOHD$LL$HuWH=u$HGHH LLL$HtI*H{LLL$H5߫H81BLL$HE1E11HW$E1F$8$LLL$^LL$IHBLE1E1H $E1##5H$HcL$LLT$Ld$PHD$`H)Hl$XHtP:LT$HIH̫E1#H#~#ٍIBLL$L׽P0HCLL$IںPHxE1E1<#H9#'#qLT$0HD$;LL$LT$0HH="HGHH7 LLL$LT$0HtPIH1L4HH8H٪E1E11H"""lHrLLT$H5֩LL$H814LT$LL$H{E1E11HD"6"("rFHeHVH5H8'JH=۪HL$1HL$'HM1!H!E1!LLT$0LL$LT$0LL$IH I M|$MIl$IHEI,$HEIL%V !L%! !-M.H1LLT$ HIHt$PHLT$PLT$L|$X 7LT$HIVHШH H  ˌ5MbMMzI$II*u IBLP0D$IGMHcE1M' H$  U.H=HL$DHL$HME1E1H1[1HL(IHHD$H@@=H|$Ht$PLT$PLT$Hl$XLT$HHI*u IBLP0HmHEHP0HSME1HōLH&E1E11HLSMHCIHH+HD$uHCLT$HLL$P0LL$LT$HD$H@H;H;weLT$LL$ LL$LT$HILPH|$1HLH IHI/H\$HD$IGLP0LT$kHH HHH5H8HͥE1E11Hz[H5%H=1HH)H:H+u HCHP0HVE1E11H6NMѺ';HE11HHE1HyH1LLT$IBIID$LT$LIP0HELT$HH5<H8TPHPH)H1LLL$LT$HLLL$LL$IH|$Ht$PLT$PLT$Hl$X2LT$HHBHLL$E1E1HwiQHE1E1HHUE1MA3HWHH یHD$H@@=H|$Ht$PLT$PLT$LL$XLL$LL$LT$HHI*uIBLL$LP0LL$I)u IALP0H\$IH|$Ht$PLT$PLT$LL$XLL$0LL$LT$HHuHdH\$E1E1H* XH21MHNHE1E11HxLLT$0LL$LT$0LL$HH}kHLE11HXLL$E1E1?1g{HULL$E1Hau0ILLL$LL$BHHHԌgHؠH\$E1E1HH|$(LT$LL$HGP0LT$LL$I/IGL1P0IBLL$LE11P0LL$E1@HEE1E11H2=HE1HHLL$E1E1HZ E1E1HE11E1HxH\$eW!H{H\$E11HB4&pfAWAVAUATUHSHHLfdH%(H$1HWHD$pHD$xH$H.HD$@H$HHD$HH$HIIBIXHE(HD$HELm HD$HD$HIEIHH_LHIFHMHH=;v2IH I.pH5!LLL$ LL$ HHD$ H|$H?jI)PL%H=rIT$LHH HHEH5jHHHIMI HmuHELT$ HP0LT$ HI9BHt$LLT$ -LT$ HI"I*u IBLP0Ht$HHD$HHu HFHP0L%H=IT$L/HH3HHEH5HH4 HIMHm`HI9D$ LLK,HD$ H!I,$RIm7HD$H;L%H=IT$LzHH^ HHEH5HH#HHD$(H|$()#Hmu HEHP0H|$HGLt$(1LI@HD$HIHD$HIGH5LHH IM LIH I,$u ID$LP0I!H|$ H5NHGHH"IM"LHH#I,$u ID$LP0H# H-L%HEHHe7TH QH P ; HL$1HLH%HL$X SP H= @9-H8H/Hmu HEHP0H<HLl$ E1E1E1HD$811HD$0HD$(HD$ fDINHF8HD$HHE0HD$@HD$IFLP0LL$wIALP0HGLL$ P0LL$ I)HHD$HEHP0fIELP0ID$LP0ImHI^HL$(HHD$HHH|$(HGP0IwzIHJcH>fDHF8H$HE0H$HE(H$HE LHD$xHEHD$pIIHFJcH>LeIH HAHNEOD@HHHPH5,ATL *H81H XH ZH=TE1H$dH+%(*;HĨL[]A\A]A^A_H țAQfD1HD$HHTE1E1E1H|ؑLl$ 11HD$8HD$0HD$(HD$ fDHH&HL$Hl$HH@H!HL|$11HyE1E1E1bvTCHD$8HD$0HD$(HD$ HD$HD$fDMt&I,$uID$L\$@LLD$P0L\$@LD$MtI+uICLD$LP0LD$MtI(u I@LP0H  E1H=b5 H|$t HL$HHD$HHu HAHP0HL$HtHHD$HHu HAHP0HtHmu HEHP0HT$ HtHHD$HHu HBHP0HL$(HtHHD$HHu HAHP0Ht$0HtHHD$HHu HFHP0HT$8HtHHD$HHu HBHP0HtH+u HCHP0I/u IGLP0ImIELP0fHE1E1E1HI L|$112 vE1! HHD$8HD$0HD$(HD$ HD$HD$Mt8I.u2IFL\$PLLL$HLD$@LT$P0L\$PLL$HLD$@LT$Mt.I)u(IAL\$HLLD$@LT$P0L\$HLD$@LT$MZI*PIBL\$@LLD$P0L\$@LD$-DHiE1E1E1H1 L|$11 vE1 KHD$8HD$0HD$(HD$ HD$)HuHH= LHGHHHtH HLH5H81VHE1E1E1Ho L|$1Z yL YHD$8HD$0HD$(HD$ HD$fHAE1E1E1H E1I y [HD$8L|$11HD$0HD$(HD$ HD$@HH=S LHGHH)HCHgLIH5LIHVHD$pHH5yLHVHD$xH?'IMVHD$pLl$xHD$H$HD$H$HD$@H$HD$HyHFLHD$pGIfH5LHV HtH$IM~H5LHVHtH$IMTH5LHVHtH$IM'1HL$pMLL H52MH 15H E5H=HL$HL$IHH~fDH|$ H5HGHH IM ID$H;~H;EH@hHH@H1LIM!I,$uID$LL$LP0LL$H|$ LL$H51HGHH"LL$IM!ID$H;H;9"H@hHcH@HVLL$LLL$IMH"I,$uID$LL$LP0LL$LϺLLL$LL$HI#I)u IALP0I.u IFLP0H-I9L;%;0L;%Y#LAƅ_#I,$u ID$LP0EIGH5LHHnIM9ID$H;H;H@hHH@H1LIMI,$u ID$LP0H|$ H5_HGHHIMID$H;cH;lH@hHH@H1LIMI,$uID$LL$LP0LL$LκLLL$0LL$HII.uIFLL$LP0LL$I)u IALP0I9L;%x%L;%LD$xI,$u ID$LP0L$H|$HWHBpHXH@HKH5 IHRHHHI,$u ID$LP0IGH5LHHZIMIAH;yH;DH@hHH@HLL$(L1LL$(IMNI)u IALP0HU HEHHH9H9I$HUL$HHEI,$u ID$LP0HCH5%HHHIM=HI9FMVMtI^IHI.uIFLT$(LP0LT$(HCH;H;0LT$(LT$(HI$LPHEHh HCLMz'L\$(L\$(H QH P ;'1HL$8LHL\$0AHD$(cLL$(L\$0H QHL$8P =H @9&M'I+uICLL$(LP0LL$(H+uHCLL$(HP0LL$(IALL$(LH5 HH LL$(IMcI)u IALP0IGH5#LHH IM IAH;H;2WH@hHXH@HKLL$(L1LL$(IMe I)uIAL\$(LP0L\$(IFH; MNMU!I^IHI.HCAH;H;HcPL\$0LL$(LL$(L\$0HI+#MtLHHIcHIDAD$HM\HCLM$LT$(HLT$(H QH P ;%1HL$0LHLT$(AHD$P LT$(H QHL$0P =H @9SH|$P$I*u IBLP0H+u HCHP0HD$HOHH9X Lt$Ht$ LHH!I.u IFLP0HCH;(!"H{H"HCHD$(HC HD$0HC(HD$8HD$(HHD$0HHD$8HH+u HCHP0H5H|$@#H5H|$@\)t$H5H|$@<(2w'L%H=IT$L#HHy)HHCH5pHHJ)HIMy*H+uHCLT$XHP0LT$XL%FH=LT$XIT$LLT$XHH-HHCH5HH.LT$XHLT$XIM}.H+uHCLT$XHP0LT$XH|$8LT$XH5HGHH@-LT$XIM,Ht$0LLT$`L\$X)L\$XLT$`HI'/I+uHD$`ICLLT$XP0LL$`LT$XIF1E1H;.H;8(H;0#HL\$hLL$`LT$XLT$XLL$`HL\$hI1MtLXHL$8Hc1LMLCLHHILLT$`LD$XLD$XLT$`HH1I(uI@LT$XLP0LT$XI.uIFLT$XLP0LT$XLT$XLT$XHI}-HXHD$ LT$XHIF LT$XHI$-HT$HH5HLT$XNLT$X*HT$HH5LLT$X(LT$X6*LLLLT$H(LT$HHHS2I*u IBLP0I.u IFLP0I,$u ID$LP0H=1.L%HH=!IT$LIH"HIAH5HH$LL$@LLL$@IMQ$I)u IALP0L%H=IT$LCIH#HIAH5hHH&LL$@LLL$@IM(I)uIAL\$@LP0L\$@HI9C'Ht$0LL\$@@L\$@HI@'I+u ICLP0ID$H5LH@pHi!H@H\!IM*I,$uID$LL$@LP0LL$@Ht$8LLL$@jLL$@HI(I)u IALP0IFE1H;`(H;1#H;HclLL$@)LL$@HI ,MtLHHcD$HL$P1LLL\$HIL@HMd萶L\$HI+I+uHD$ICLP0LT$I.uIFLT$LP0LT$HL$PHHD$HHuHALT$HP0LT$LLLT$LT$HI*I*u IBLP0HXIHm*IFLD$LLH5HH8*LD$%I(u I@LP0Ll$ ILt$ @MrMIjIHEI*u IBLP0HEH;H;޽~IHLpHD$HID$ HELMWH QH ¼P ;HL$ 1LHAI'H QHL$ P H= @9 MI,$IID$Hl$ LP0LT$ fD‰D$@H~L|$11HE1E1E1yiHD$8HD$0HD$(HD$ HD$f諼HuHH=7LHGHHHtHHBLH5|H81H_}E1E1E1H'1z HD$8HD$0HD$(HD$ HD$H|E1E1IHzHD$811HD$0HD$(HD$ HD$H蘾IMT$MIl$IHEI,$uID$LT$LP0LT$HEH;>H;LT$諺LT$HILPIELh HELM脺H QH P ;^HL$1LHAHD$ RH QHL$P H= @9 H|$ I.IIFLP0fHa{E1E11H*1z HD$8HD$0HD$(HD$I|$DID$L0IHzH6E1E1üI7I|$ID$LIHzLl$ E1E1HOA38KHLH5yH81HIFLP0LL$(*1LL$ܬLL$HHHLLL$kHmLL$I~HEHP0LL$jHoHdH|$ I8HnE1E1H{ĒHD$8Ll$ 1HD$0HD$(HD$ 3@$Hfn0H-ђE1E1E1CIH,nLl$ E1E1HfHt$pHLT$pLT$Ll$xLT$HHD$ HmzH{IE1E11HD$81HD$0HD$(HD$IyMYIHVmLl$ 11HE1E1E1iHD$8HD$0HD$(HD$ "H|$ LL$IHlE1E1E1Ll$ E11HӒHD$8HD$0HD$(HD$ 莮@$L\$0LL$(L\$0LL$(@$HJlLl$ 11HE1E1E1kHD$8HD$0HD$(HD$ LT$(肪LT$(@$Ht$pHLT$pLT$(Hl$xLT$(HIHkE1E1gHdRE1IHpk:H7%HIkHn,I_HIcHLL$pLL$0HD$xH)L$HtpL\$(L\$(LL$0HHD$PiHjE1E1E1H|n$HjLl$ E1E1HXJ<qT1LL$(LL$(HILHLL$(rI,$LL$(IHD$0ID$LP0L\$0LL$(jHHݧH5jL xjAH /nHwH81w^_HiE1E1E1Ll$ E11Hzl^ HD$8HD$0HD$(HD$ H^iE1E1"yH }I;LL$(I1LHaIHHiIE1E1HL|$y۪IHhE1E1E1HE1{m苧HuHH5XiH8pfH1L諩HH&HPhHDH)hE1E1E1HLl$ E1HD$81HD$0HD$(HD$ H=hHL$ ٥HL$  販E1LE1ZH=ChHL$虥HL$bH HrH5+hH8CH?gE1E1IHzA詥@$jE1L1LH3HD$ HWHfIE1E1HE1E11z1xHD$8HD$0HD$(HD$ HD$[H='gHL$}HL$vLPHM?LpIIHHD$(HHuHCLT$(HP0LT$(IFH;tH;Ϥ LT$(jLT$(HI&LPHD$ 1LLL\$(HIC L\$(HHI+ICLP0H}eE1E1E1HE7)HMeE1E1HZE1E1E1HD$PLl$ HD$8HD$0HD$(HD$ H HH5eH8ȠHdE1Hv4H;HdIHLH+uHD$(HCHP0LL$(IALL$0LHLL$0HHD$(SLL$8LLL$8HHD$0BLL$XLLL$XHHD$8LӾHJLL$X>I)mIALP0^HHx萠HcIE1E1HyHD$P1dLl$ QHD$ HD$8HD$0HD$(H1L藤HHQfDH1cHHD$PLl$ E1E1E11HD$ 1LHLT$(-LT$(HHD$PHbIE1E1HLl$ E1}o?E11LHL\$(ͣL\$(HIHmbIE1E1H5Ll$ E1LT$-LT$HUH8H5bH8 LT$5H=bHL$0L\$(L\$(HL$0_L\$L\$HGH̞H5bH8蝝L\$'H=IbHL$0LT$(蚟LT$(HL$0[HmHkH5$bH8HRRzHH+RHHRE1H(@HQE1E1H2L脎L\$0LT$8E1E1HQLt$(1Ll$ HUHD$PB4HD$ HD$8HD$0HD$($AUH-QIE1E1HHD$P1Ll$ HD$8HD$0HD$(HD$ 鐺ےI#HPHyHHD$PLl$ E1E1E1HD$ DHPE1E1E1HMHD$P:Ll$ '9HD$ 2IA@=HD$0LHt$pLT$pLT$HLL$@HD$xLL$@LT$HHI:I*uIBLL$@LP0LL$@M^HOE1HtHOE1E1E1Hl1\NnVHrOE1E1E1H:,-IFL\$hLILL$`LT$XP0HCL\$hLL$`LT$XIFLL$@LMP0D$IELL$@LLT$XِLT$XpʐHNH}kŔLLL$@葐LL$@HxNBH?-HQNHٓH*NHNHNE1E1E1HhHME1E1Ll$ HHD$PxjHD$ VHD$0LHt$pLT$pLT$HLL$@HD$xLL$@LT$HHIfHVM H zH/MHєHMH˔GHD$@Hl$hLd$pHމHD$H H L(hE1ɹAAHƺHAHD$ H H|$ H?HD$ HL$LpHAH$AH~9IFITff(H\Xf(\\H9uf/ WL=H=IWL蛈HHvHH|$IHHE1E1H;mAL-L9 H;Ic LT$(T${T$LT$(HIMtLxHcI$MdBHMTHELMLL$8L%LL$H QA;$P 1LL$LHAHD$(LL$H QP A$=H9H|$(I)HmL%ϿH=IT$L;HHHHEH5HHWHIMVHmIAE11H;~L99 H;iHcCLL$LL$HIHtHhHL$(IcH݅HILAD$HHITIAHH LL$0LT$豅LT$LL$0P L%A;$P  1LT$0LLLL$HqLL$LT$0H QP A$=H @9 HkI*uIBLL$LP0LL$I)u IALP0HEHEuH} HD$@HLc H5ҾHD$0M|$Ht$L Ht$HHD$H H@HH H|$HLLHD$HHLc H5fM|$Ht$L謅Ht$HIFH@HH LLLIH*H@H;͂MQMMyIII)uIALT$LP0LT$LLLT$LT$HII*MP I)u IALP0I,$u ID$LP0Ld$@HD$HD$PHD$HH|$0HD$8HHl$XH,$LIHL$HL$HL|$ xLE1 $B\ IL9BI~L $^5I)KMHL$Ld$8H9L$0Hl$XH|$P}HD$HL-mH@HH褂L%H QP A;$E1LH|$HIvH QP A$H=9 MCHT$HHH$HHoImu IELP0HL$ HEIHH$HHHm@HL$(HH$HHu H|$(HGP0HmIHEHP0:DHGP0'@HEHP0HE@=L|$`Hc¹E1Ld$hH)LT$pHUHt`HBR uLMLT$UHLLT$HD$(H|$(HzBD HA/-HmHD$(g1E1fDI*>MtI/uIGL $LP0L $MPI)FIALP07HELL$HP0LL$GWHH HD$hGWHH HHD$H|$G蒀H9GHHD$IALP0Hm?P @9 IA@=HD$(Hl$`E1HD$hHHD$pIQL)Ht`HBR uMQLL$ HLLL$IM HtHmuHELL$HP0LL$LLL$0LL$@$KH-øL-HEHH~L%l~H QP A;$ 1HLH~H QP A$H= @9W H2 HiHmu HEHP0H?1 HHD$(H H=dMA5pHL$ HH$HHu H|$ HGP0HHmu HEHP0H|$(H\$H1LH腀IHH$HHu H|$HHGP0M|H?ضHնöv.fDHD$HIHD$@HL$L|rH|$(HT$(HH$HHH|$(HGP0}HD$II_MI HFHHD$`wIH5LHVj}HD$hH IMH|$`pIHF(HHD$pHF HD$hHFHD$`8wIM~1HL$`MLL KH50yH =5rH seLLL$LL$HIHHT$HH ]='H $HH$HHE1H|$HLL$L$HGP0L$LL$ME1Hc¹DHH)LT$pHt`LT$L|$`Ld$h LT$HHD$(Mt I/I*|IBLP0mH<YHVDϕH E;E1H=J5'|j0HD$H7<HٕHD$HHIoLL$0LT$zLL$0LT$@$7 @IBL|$LP0LL$IGLT$LP0LT$H1L|HHH;1L HI7HD$(1LLLT$0LL$|LL$LT$0HHH%;1 Hز{I*E1IBL $LP0L $yHuHH=`LHGHHHtHWHkyLH59H817|HD$(H:HPB 4HD$(LLL$Hl$`HD$hHxHD$pL)Ht`bLL$HIOH:ܱ HٱDZbHHmE1E1HELL$H1L$P0L$LL$Mv{fH9E1h HeSHD$(HF HHD$hHFHD$`rI6L}M(LmIIEHmIELAw@$QH L=AwHuHH=yLHGHHHtHHwLH57H81PzH8k HhVNHzIHj84 H1P\HC8  H 4IiHuMaHEI$I)u IALP0ID$MA?1LHLL$yLL$HHD$(HD$(H7H|n `?H=<8LL$uLL$L $ZvL $HuHjtH5#8H8;sL $H5LHVbvHIHD$pI6HuH8,rH6ǮHĮH6 HpHpH5ߴLIHVIuHD$`HALmHuH8qE1HELT$HLP0IELT$ALL$L$ uL$LL$HH'sH56H8qLL$L$f.H=6LL$0LT$sLT$LL$0Nt@$H=^6sHL$HHH$HHkH|$HHGP0ZbtHuHvrH5/6H8GqHT$HMH>5HHH$HHfH5ˬ HȬ$v sH>HqH55H8p#fDH=Y5rKHHdrH 8H54jL BAHAH81uY^H1LLT$HD$(H1LLL$IGv@H 4ӫ HЫjoE11H 8Awff.AWAVAUATIUSHHLndH%(H$1HqHDŽ$H$HI I Lv HkHD$xHHDŽ$HDŽ$rHH~H L(hE1ɹAAHƺHAHD$HH|$H|$xH?uHGP0L=HD$HD$xH=2IWLLhqHHHH:pIHHE1H;oAH;p&H;qIcT$pT$HHD$H$/H$HtHL$HDŽ$HAHt$HcILtBHL|HELMOpL=oH QP A;Ht$1HAI!pp VP A=@9<MLt$xHt$HHD$HHu HFHP0HDŽ$Hmu HEHP0L5ʨHD$xHD$xH=IVLHD$$pHHPHHEH5HHHIL$MHmu HEHP0L=HH=!IWLoHHHHEH5ҨHHHIMHmu HEHP0Him11I9FBL$IAH;snH;nHc L$ LL$gnLL$L$ HH$I:HtHhHt$HcHItAHM|IAHH1LL$nL=mLL$p VA;P b1LL$LLHmLL$H QP A= 29HmHl$xI.uIFLL$LP0LL$HDŽ$I)u IALP0Hl$xHDŽ$HEHEuH} HD$xHD$PHJMD$ L5 HD$(MxLLD$LWnHD$8HH@LD$HH H|$8LLHD$8H_M|$ H5IWHt$HHT$ mHt$HHH@HT$ HH`LH$HHuH@H; kHDŽ$ HGH$Ht HWHHH$H/uHGP0L$H$MB LHD$xHI/u IGLP0HDŽ$H$H/uHGP0H|$xHDŽ$H/uHGP0HD$xfHHpHpxHD$XHHL$`Ht$@HD$HHtHHD$@HtHHD$HHtHmH|$(HD$hHLt$PHD$ HD$0Hl$PM<LMfHwfE1BDI}L$hL$CIXL9uff.z13L^fHYBL9uHL$0Ld$ HD$ IIH9D$(jHl$PH|$heH|$`t H\$`HHD$HHu HCHP0H\$@HtHHD$HHu HCHP0HL$HHtHHD$HHu HAHP0HD$8L%}H@HH iL=5iH QP A;1LH|$8IiX SP AH= @9YM&Ht$8HHD$HHuH|$8HGP0MI,$u ID$LP0HL$HEHHHD$HHHl$HIkI MMH.H ~.HNHL F8LNL@HHgAUH7H5*H8H)1kHX#ZHٺ#H=&7 jHD$H$dH+%(HD$Hĸ[]A\A]A^A_fDL5g"@HE@=#L$Hc¹L$L$H)E1L$HEH̐LP@ uL]LL$HLALL$HD$HD$HD$xHfMtI)uIALP0f.HDŽ$I/xIGLP0i2H(KHH6HHmu HEHP0Hl$xL%5Ht Hm1MtI/u IGLP0H$Ht H/uHGP0H$Ht H/uHGP0H L5"hHL$Ht(HHD$HH1H|$HGP0H\$Ht[Hmu HEHP0H|$t"HL$HHD$HHu H|$HGP0HmHEHP0fDH|$HL$HHD$HHfH|$HGP0UfHD$RfHl$PH&?cH5-4AHeH8bD>`H|$hHtfOY`H$Ht H/}H KAL%3HDŽ$5$LfH|$XHt$xH$H$] H$Ll$xL$HD$HL1LhIHt HD$8H@LM: LD$(LL$ ZdLL$ LD$(H L=cQA;P S 1LL$(LH|$8AHD$ dLD$ LL$(H QP A=H9M Ht$8HHD$ HH\I)9M L;cL;(c[L;FcNLLD$ dLD$ I(uIPD$ LR0D$  Y Imu IELP0HD$xI.u IFLP0Ht$HDŽ$HHD$HHu HFHP0HD$XHL$`HDŽ$Ht$@HxpLhxHHpHL$HLHpxHHt H/uHGP0MtImu IELP0MR_@IA@=HD$E1H$L$H$HcH)HԐIQHBR uMQLL$ HLLL$HD$xHQ HtHmuHELL$HP0LL$I/QIGLL$LP0LL$8a@$Hc¹DHL$H)L$H̐L$L$LT$LT$HHD$HD$x MI*IBLP0@fH"H|$xP|1HKL%/2HD$HD$H.H/$E1HGP0HD$H"H}ә'H|$xHH/1E1L%K/`HuGH=\LHGHH HH-Hh`LH5 H814cHD$H|!HM?1?YHD$8HHZH5HIHVId`H$H MgH$L$HGP0wIALD$ LP0LD$ HFLD$(HLL$ P0LD$(LL$  @KI(LD$(LL$ !_LD$(LL$ @$H\$81LHaIHHD$HH1L%-H$蝲HD$xHH ՗H֗H\$8×HHD$HHu H|$8HGP0H|$xL%2-qHD$HHtfA4HH$LL$^LL$@$DHMH$HHUHHHmuHEHT$HP0HT$HBHAHD$LLL$H$H$HcH)L$HLL$HHD$xZHHsH=D]HuGH=LHGHHHH~H]LH5H81_H8H|HE1ؕHՕÕ~H_I%HHbZ1H^HD$xHIHE1_H\JmHD$_\HuGH=LHGHHHHH[LH5`H81^HޔH۔ɔH=[BHD$x3H^IxHHp>[HuHYH5[H8sXHF HH$HFH$UIM;1H$MHL )H5b2Hٗ#Hݓ˓ٗ-fDHFHH$4UIInHIFHEHH$I.4IFLP0yHt$`OHD$XHL$@HF8HxpLpxHppHt$HLHHxHHt H/uHGP0MtI.u IFLP0MtImuIELP0H|$xH|$xHYLH8VHH|$xL%-(H[H5lHHVYHBH$I(HYH8UHDŽ$G1LLLL$[LL$HHD$xHH*L%'HؑH= XLL$H1HD$xL%:'HXHuHVH5nH8U른pX@$OWHL$HxXLH`LhXLxhLp`HHhHtH/uHGLL$P0LL$MtI)u IALP0MtI/u IGLP0HD$xHDŽ$HDŽ$H>HHvRD H|$8L1LL$ YLL$ I(LL$ ZWLL$ HuE1H=LD$(LL$ NVLL$ LD$(%ZHeH=!VHt$8HHD$HHu H|$8HGP0HH|$xL%.%HǙ\VHuHTH5lH8SfH1DYHHj4H1[H1LALL$HD$F3YHLkHݎHڎȎPH1LLL$fHD$(HSH5H8RLD$(LL$ gH|$xHtH/L% $`L%#eARATIUHSHH Ht HՅuH{(1Ht[LH]A\fD[]A\ff.USHHH-lTH HEHk HHEHt H/t=HEH{(HHk(HEHtH/tH1[]DHGP0H1[]HGP0Hu710HtHHPHSHHP HP(HHSH5j18AWIAVIAUATUSHHL-H=ʌdH%(HD$81IULNTIHHID$H5HHLHHAI,$IFHmeL-^H=7IULSIHHID$H5םHHLIMI,$#HQI9XLLLD$LD$HH1 I(IGH5\LHH!IM#I9]MeMtMMI$IImuIELL$LP0LL$IAH; RH;dRLL$QLL$HI L`HEHh IAHH LL$LD$QLD$LL$H QH /QP ;1HL$LLLD$LL$IQHL$LL$X LD$SP =H @9M I(uI@LL$LP0LL$I)u IALP0I.u IFLP0HEIHPHUHEHu HEHP0HD$8dH+%( HHL[]A\A]A^A_fDID$LP0IFWHL1NHmL-H=IUBL QIHHID$H5qHHLIMI,$tYHNI9]XLL-HHImOIELP0@HEHP0BID$LP0ID$LD$LP0LD$fDI@LP0IQBፁZLd$ 1Hl$(LB uIyLL$Ht$ O ALL$IM'H`H]KI,$IM#f.HLIHiM+OHuHH=LHGHHGHtIHNLH5+H81QڝH H 5H=KPMQI$LE1H+fH[HXFܝI,$D E1ID$LLL$IP0Ll$M Imu IELP0H 5TfDLQHLL$LD$MLL$LD$@$fDPII}fDLHt$ LL$Ld$ Hl$(远LL$HII,$ID$LL$LP0LL$fH=HL$LL$LD$ELLD$LL$HL$fDH IMͅHʅI(I@LP0LHH=CLHGHH!HtDID{LHurH=LHGHHHt2IHLLH5{ H81NAKHKLH5Q H81N!DLNIGH ME1HCLNIIH kHhVI,$uID$LD$LP0LD$H ;1Mċ5$MHMM`II$I(uI@LL$LP0LL$ID$H;xJ H;J6LL$nJLL$HILHILp ID$HHGJH QH IP ;'HL$1LLHJH QHL$P H= @9H7ImM(IELd$LP0LD$MEMMMIIImuIELL$LLD$P0LD$LL$IAH;OIwH;ILL$LD$@ILD$LL$HIvL@ILp IAHHLL$ILL$H QH vHP ;s1HL$LLLL$HHLL$H QHL$P =H @9HI,$MID$LP0H E1E1H QH uIHoaM@H KHH6f1LLLD$LL$JLL$LD$HIsfDLL$LD$ HLD$LL$HHFH5 H8DLL$LD$JIT$BፁLL$ 1Lt$(LB uI|$LL$Ht$ ALL$HHI)u IALP0M1IQBፁfLD$ 1Lt$(LR uIyLL$Ht$ LD$ALL$LD$HH/I(uI@LL$LP0LL$MZHHqI,$.$fLL$FLL$@$^F@$ LHt$ LD$ LD$LL$Lt$(tLL$LD$HHAHE1M~H~~WHt$ LLL$ LL$Lt$(LL$HHdH~H~t~I)u IALP0I,$bME1E1~HtH 1ɺALL$IHI~H~} H%M}H}}e1LLLL$8GLL$HHH1}H}}k>1LLFHHNHME1Z}HW}E}&H=!HL$wCHL$LL$>DLL$HYHIBH5H8ALL$9 DHmHBH5H8@RH=HL$LL$BLL$HL$en@HE1Ms|Hp|^|_HL|HI|7|1ɺALL$H1ɺALD$LL$HAME1UHH dH%(HD$HGHFHH|$Ht${DHEH}HPHUHt=HEHEHH|$H$HEHt$*=H} HtHE H/tSH}(HtHE(H/t,HD$dH+%(uwHEHH@H ]HGP0HGP0HGS@HD$dH+%(uH ]>AWL~AVAUATUSHLD$M~vLIHIHMLf.IvL@L9tIHI/u IGLP0L5;;H :HD$XM9HL$I9RL;%:EL]DIGLP0LmI,$#f.IGLP0|ID$LP0EHEH5tHHHIM H5uL5HD$`IH I,$pL5I8H 7M9HL$I9u L;=8hDI/BHD$`EHEH5UqHHHt(ILd$`MR(M9L;d$L;%7L9AŅ(I,$u ID$LP0HD$`EdL-pH=pIUL+8IHc!HID$H5OHH3"LIL|$XM"I,$u ID$LP0HEH;7FH;59H@hH,H@H1HIM1"H5I9G"H|$XLHD$`H#I,$u ID$LP0H|$XH/uHGP0HD$L%pHD$XL|$`HD$`LH LIQLL$ HHT$7IH#H@HT$LL$ HH*LLIHo#HD$L%oLH LIQLL$ HHT$57HH$H@HT$LL$ HH LHD$XHHw$H@HD$hH;I4{HGHD$hHiHWHHHT$XH/uHGP0Ld$hH|$XM=LpHD$`Ho&I,$u ID$LP0HD$hH|$XH/uHGP0HD$XH|$`H/uHGP0HD$` 0HHpHPxHD$(HHL$0HT$HD$ HtHHD$HtHHD$ HtHHHLl$8Lt$@Lt$IvH4HHIHD$`IHH53HL.Imu IELP0HHHD$`IHHLHImu IELP0LHHHD$`H4Ll$8H|$0t H\$0HHD$HHu HCHP0H\$HtHHD$HHu HCHP0H\$ HtHHD$HHu HCHP0H5~1L*ImIu IELP0MI,$uID$LP0DL%2 @DHAH|$XE1E1Hlkk7Ht H/uHGP0H|$`Ht H/uHGP0H|$hHt H/uHGP0MtI,$u ID$LP0H kkE1H=5tk3MtI/u IGLP0H$dH+%(s%HĘL[]A\A]A^A_fDK2HuHH=jLHGHHHtIH1LH5KH814HH|$XE1jHjjAL4IEH{jHxjfjCI/u IGLP0H|$XE1aID$LP0IGLP0En@I.u IFLP0HEH5EjHHH IM IGH5{LHH^ILd$XMI/u IGLP0ID$aIl$HEHB H HcH>DEnAMcI4w0H&HoME13iH0ii~EnAFII I@EnAFII @HGP0@HGP0@HGP08@Al$I,$u ID$LP0L5hH=XhIVL/IHHID$Ld$XH5;hHHLIMZI,$u ID$LP0H .HD$XIH.HD$`IH4L`[0HD$HD$XHpL%gH=gIT$L-/HHHHWH5bgH|$hHHhH|$ H|$ IM)H/uHGP0H5gH|$LHD$h0I,$u ID$LP0IGLM-H QH S-P ;PHL$(HT$LLAHD$ -H QHL$(P H= @9 H|$ jI/u IGLP0I.u IFLP0HD$`HL$HHD$HHu HAHP0H|$ H52xHD$XHGHHIMID$H5wLHHIL|$XMI,$u ID$LP0LCIHI/u IGLP0HD$L%zfHD$XL@ LMxLD$(L-HD$HH@LD$(HHH|$LLHD$HHD$L%fL@ LMxLD$(LC-HHH@LD$(HHLLHHH@H;g*(HGHD$`HALgHI$H/uHGP0L|$`MLL|HD$XHI/u IGLP0HD$`I,$u ID$LP0H|$XH/uHGP0HD$X;&HHpHPxHHL$8HT$(HD$0HtHHD$(HtHHD$0HtHHH H~nMLd$LL|$@It$HC*H9t>IHL$HLLLLD$*H|$HL~*HLLp*M)HuL%*I<$Ld$Xu ID$LP0HD$XH|$8t H\$8HHD$HHu HCHP0H\$(HtHHD$HHu HCHP0H\$0HtHHD$HHu HCHP0HD$H-tH@HHz)H QH )P ;HL$1HH|$Hh)HL$X SP H= @9qHH\$HHD$HHuH|$HGP0HWHmL|$ u HEHP0I$@L*AŅH6H|$XE1E1HaaaAl$HcH(HHH|$XE1E1Haaa@Al$AD$HH HfAl$AD$HH f.IFLP0DR[+I3HAME1aHa`SI~1IGL0IfI~MwI@1a&IHHL$ImIIELP0DHH|$XE1[`HX`F`iYfk*IK'HuHH=_LHGHH<HtI H&LH5KH81)H|$XE1HHD$`H___HE1L&I$f.Hk_Hh_V_kf{)IHaH|$XE1#_H __!L8)I8IEfDHH|$XE1^H^^%HjHH|$XE1E1H^^}^n(I`E1yHR^HO^A^fL|%H^@DH}HEL I$@H|$XxHD$hHH]H]]ImIELE1E1P0H|$XDI|$p]HD$ Il$xHw]M$a]ID$pHD$ID$xHD$I$Ht H/uHGP0HtHmu HEHP0MI/u IGLP0H|$XE1E1HH|$XfHH|$XE1E1H\\\{fDL%#I$5L;IE1N fDHH|$XE1K\HH\6\If[&I."@$GfLl$(Ld$0ܜH|$`H[H[[Ht H/uHGP0HD$`H|$hHt H/uHGP0HD$hH|$XHt H/uHGP0H [{[H=HD$X5a[#HL$`HT$hLHt$X-Lt$`L|$h1Hl$XLLH%IHIELMTLD$(LL$!LL$LD$(H QH !P ;k1HL$0LLLL$(AHD$q!LD$LL$(H QHL$0P =H @9MKImuIELD$(LLL$P0LD$(LL$I)uIALD$LP0LD$ML; L;] "L;{ LLD$!LD$AI(u I@LP0EEHmu HEHP0HD$XI/u IGLP0HD$hI.u IFLP0HD$ I|$pHD$`Ml$xI$ID$pHD$ID$xHD$I$Ht H/uHGP0MtImu IELP0HvHmkHEHP0L%rI$I(D DHH|$XE1XHXXPfHXHXvX f"ILl$(Ld$0ޜK@IEfDL@7HRH}LeI$HuHH=WLHGHHHtIMHLH5H81~!HE1HD$XHWWxWLl$(Ld$0c@H|$XqHD$`HHa+WH,WWImu IELP0H|$XE1HH|$XfHH|$XE1VHVVL I]Ll$(Ld$0@LD$(LL$ILD$(LL$@$fDHD$HfHKVHHV6VfHLL|$Ld$HLIt$HcH9tILHIHUHIHEL)Hu>DHE1UHUU8IL;pHD$XHHsUHtUH\$aUHHD$HHu H|$HGP0H|$XL|$ IfH!1IHHHbImIIELP0D IHH|$XE1THTTLl$8Lt$@!H|$hHvTHsTeTHt H/uHGP0HD$hH|$`Ht H/uHGP0H 4T*TH=HD$`5TH|$(HL$hHT$XHt$`HD$hLd$XHl$`HD$HL1HIH IELM LL$@LT$8ZLT$8LL$@H QH P ; 1HL$HLLLT$@AHD$8LL$8LT$@H QHL$HP =H @9LMf ImuIELL$@LLT$8P0LL$@LT$8I*uIBLL$8LP0LL$8M M9L;L$0L; )#LLL$LL$AI)u IALP0E| E Hmu HEHP0HD$`I,$u ID$LP0HD$XH\$HHD$HHu HCHP0HD$(H\$0HD$hHxpL`xHXpH\$HHXxH\$ HHt H/uHGP0MtI,$u ID$LP0HzHmoHEHP0L%cfI)DDHE1`QH]QKQfDH|$XHZ(QE1H"QQH(GfDPHPH\$0PHD$(PHxpHhxHXpH\$LHXxH\$ HHt H/uHGP0HtHmu HEHP0MI,$nID$LP0^fDHSPHPP>P[HuGH=OLHGHH HHMHLH5\H81HE1HD$hHOOONLl$8Lt$@#HiLH8HH|$XE1E1HuOgOYOlwHuHH=OLHGHHR HtIjHLH5wH81H+H|$XE1NHNNLL$@LT$8LL$@LT$8@$HLH8HD$XLl$8Lt$@- LIH}GNHDN2NH|$ [H|$ ILl$8Lt$@/y0HH|$XE1E1HMMMHH|$XE1MHMMZHMHMpMH|$X~I=Ll$8Lt$@9MoM IGIEHHD$XI/u IGLP0L|$XIGH;H;WHD$hH L` 1HLLhHD$$ HD$`IHcLD$I(u I@LP0HD$hHT$LLHD$ HHE1ZLHWLELHZE1!LHL L\@$HLH8?HH|$XE1KHKKH=HL$ HL$ +DHH|$XL|$ yKHvKdKwH|$ I pHHH59H8QHxXLh`HhXL@hLx`LphHtH/uHGLD$P0LD$MtImuIELD$LP0LD$MtI(u I@LP0HD$XHD$hHD$`HbLH8 HD$X4L|$ I8iPHAHHH|$XL|$ E1H JIIšHLH81 HH|$XL|$ E1HIIIΚL1LLL$ LL$IHMLH8 КE1H=>HL$0LD$(LL$LL$LD$(HL$0cE1LL$ALL$HuHD$(HKH5H8 LD$(LL$w IIHL$HxXLp`HhXLhhL``HHhHt H/uHGP0MtI.u IFLP0MtImu IELP0HD$`eHD$XHD$hBHfH|$XE1E1H,HHH#]TL1LLT$8SLT$8IYݚ0LT$8LT$8HE1zH=qHL$HLL$@LT$8 LT$8LL$@HL$HH\$1HHHHHD$HHH=HL$d HL$VH\$HHD$HHu H|$HGP0HH|$XL|$ E1HFFF_ HuH H5H8 IG@=Ht$pLLl$pLd$xHD$`HImUIELP0FqfgHt$pLLl$pLd$xv\HD$`HuH-MEHEEt  HMEHEEϛ;HD$@H H5H8 LL$@LT$8aHH|$XE1bEH_EME)`՛HbM)EH&EEțIx\HcHA9T|F~J1fD}6p9}%)HcHAL9~؉9|9AH1AVIAUATIUHSHg1LHHIHtkHM~$1DHHHHhHI9u1HAE HmItAm [L]A\A]A^fHEHP0E1ff.@Uf(SH(-Nf/$f(f\H,H*Xf(H YHP^fHYXH9uf(d$L$T$@ T$L$f(5d$^f(\f/YX X\rFH~AfD\L$Hf(T$ L$H9T$\~H(f([]f(1DHGHHt f. ff.AUHIATHHWH=B IHtHHLA\A]fDK HuH=ALHGHHt'IMuHLH5TH81  Iff.HGE1LH@ uLGu LA@1LAAVAUIATIUHSHGLMts6X KHH ; jLLHAIp VP H=~9|MtX[L]A\A]A^@ @9}@$[]A\A]A^k H=,tE1IHuHH5H8ATUHLGpLgxHwpHHWxHMtI(tLMtI,$t0HtHmt H]A\@HEHH@0H]A\@ID$LP0I@LP0@AUIATIUHSHH(dH%(HD$HGXHGXHHt$HD$HG`HG`HD$HGhHGhH|$H$$H{XH4$HtH|$6H$HtHHD$HtHHD$Ht HHD$HL$H$IMI$HUH{pLcxHHKpHCxHHtH/tbMtI,$tFHtHmt*1HT$dH+%(H([]A\A]fHEHP0@ID$LP0HGP0H|$IEI$HEHt H/uHGP0H|$Ht H/uHGP0H<$Ht H/uHGP0B%DATUHLGXLg`HwXHohHW`HOhMtI(tJMtI,$t.HtHmt H]A\fHEHH@0H]A\@ID$LP0I@LP0ff.AUAATUSHH9HwH9GHIH9F A|$ HEI9D$HUIL$H9@H@t Hu A|$ 8@ ,HmH@ nIL$0It$H@HEʃahM>9uGHHH1Au(1fDHH9ut1AH[]A\A]@I9uuHDLHHH;H;-duH9u@HmuHUD$ HR0D$ f.1AH[]A\A]@HfDHM0HH@HE@ B2fL+fDIt$HfDM>@M>ff.H;=H;=ku H;=u ff.HH@HH ĠH5HEHH81ff.ATUSHHxXHu1H[]A\HHnH0H9unLc`HkhHCXHC`HChH/t(MtI,$t,HtHmuHEHP0@HGP0ID$LP0HGtA@t8HXHtEHJ1H~H;tWHH9u&H|$H|$-HHH9HuH;5:f.ATUSHHuJHxXHHu1H[]A\H)H0H9Lc`HkhHCXHC`HChH/tgMtI,$tkHtHmuHEHP0H/uHGHt$P0Ht$HHH5xH81H[]A\HGP0ID$LP0HGtD@t;HXHtHHJ1H~fDH;tHH9uH|$HH|$HHH9HuH;5fAUIATHHGHHtIMtHLA\A]IMuHH8tHfLH54H81ZAWAVIAUIATAUSH(>HEAE1L IMtDHDLDA9VHHIE9yCM!I$H6H1LHHt7XlHDI,$tEHmu)HEHH@0H([]A\A]A^A_I,$H([]A\A]A^A_fDID$LP0Hmu@LxXH@`HEXHE`H=?HD$HEhHEhHD$H5?HHVH8zIHHGH5P?E1H=?HD$H}XL}XLM`LEhHE`HD$HEhHt!H/uHGLD$LL$P0LD$LL$MtI)uIALD$LP0LD$MtI(u I@LP0EEAH=>HGHHHHH|$H|$LH/uHGLD$P0LD$fLDEID$LH@0H([]A\A]A^A_@LIHEZLDHL1H=`IMLH5HE11L =111AQSAVAUPPPPPHPIIEHIEHu IELP0I.u IFLP0MvL EEMJDEDLDAA9HcHLD;~zEA9AJDHcHLDoHcƒHHAA9|AD~L&DBEI$mLE15IHL!IfImIELH@0ZDDA9Dp@LIcHIHIcDDHDHD5DHE91D&HH@HdDH UDDxL I$y@IcHLLL&H/@HGP04ZHfUHGHL@t7HLHQH81wAHEuL]f.HHHH5ĘH81Hmt 1]HEHP01fHmuHUHR01]ff.UHHGHW1HtHtHx`Hu;H]DWЁtHdH5ŹH8%fDHu HtfDH)H5ZH8H@`HtnHHtbHHtXHJH9Eu/@HHmJHUD$HR0D$2HH506HHuDdH6H<H5yH8Eff.UHHGHW1HtHtHx`KHu;H]DWЁtHH5 H8fDHuHtfDHH5H8H@`HtnHHtbHHtXHH9Eu/@HHmJHUD$HR0D$2HH5мHHuDH6HH5H8ff.UHHGHW1HtHtMHtHH9uQH]Wʉ9tHH5ԷH8dDWʉ9uH]fDHu5HtfDH@`HtnHHtbHHtXHH9Eu/@H HmTHUD$HR0D$1H]GH]WGHH HcʉH9tHH5SH8@WGHH HHcʉH9uH]GH]HcH9fHuyHtLf.H@`HtmHHtaHHtWHH9Eu.@HHmHUD$ HR0D$ HH5HHuvHhHH5H8MUHHGHGHtDHt&HtxGH]D1H]GWH]HH DGH]HqH5ڱH82HfH@`HtvHHtjHHt`HH9Eu0@H(Hm_HUHD$HR0HD$FHH5ouHHuH#HuHxH5H8ff.@HGtgHGHPHwQH HcH>G@GWHH fGWHH HGHUHH@`HtsHHtgHHt]H]H9Eu0HHHmuHUHD$HR0HD$H]HH5BHHHuHrHuHNH5H8WDUHHGHGHtDHt&HtxGH]eD1H]GWH]HH DGH]HH5H8HfH@`HtvHHtjHHt`HH9Eu0@H(Hm_HUHD$HR0HD$FHH5HHuH#,HuHH5EH8ff.@UHHGHGHHHHcH>1H]GH]WGHH HcʉH9tHH5cH8s@WGHH HHcʉH9uH]GH]PHcH9fHuHtLf.H@`HtmHHtaHHtWHjH9Eu.@HHmHUD$ HR0D$ HH5QWHHuvHhH]H5H8fMUHHGHGHtDHt&HtxGH]uD1H]GWH]HH DGH]HH5H8HfH@`HtvHHtjHHt`H*H9Eu0@H(Hm_HUHD$HR0HD$FHH5HHuH#DgAMcH/tHLA\fDHGP0HLA\H/Dgu@DgGII H/ufDgGII IH/uH|$]H|$IH/{fH@`HtgHHt[HHtQH@H;H5HHuNI H|$H|$IfHuHH5ĦH8H@DDHHLHH1Hiel HHquHdžpHdždžfHcHP=pt>HHH H1HH%V,H1HHH1HHH1HLHDHHyH%H HǃHHH3` %߰H1HAI9uH` @HH H%H HHHH3J%߰H1HH9uHxH%H HƒHHH3` %߰H1кHx1fDUHSHnHHcHffH*HYH*XYH[]ff.SHH t6f1H [ff.zfudHHXf(\ L$L$Xf(Yf(\جf(YXf/Ĭsf(T$L$\$\$fL$YT$^f.wQYȸYT$L$T$L$H$L$XYD$X$HHHf.UHHHD$L$H]^f(ff.fHCf(\nfWVHÐHD$YD$HfHD$^D$\2HDHD$} ^L$H&fDHD$MfW Ъf(^L$H\f(ff.fH$L$8YD$X$HUHH0f.dD$zF%Nf/d$f.HHD$ t$T$\f/r< D$f(^\$f(f/rH0f(]D$ L$\^D$|$L$D$f(Y\f( {^T$\$ f(\f/H0f(]fd$\%҃f΃Yd$(f.Q=^|$@H0ff(D$YXިf/sf(HL$YYT$`L$Rf(YYY\f/wbL$ eD$D$T%d\d$f(L$ f(XYD$(YYXf/D$&L$(YL$H0]f(H0]n@HL$YD$HfHY@HXfUHH $D$ $Hf(Yf($T$$H]Y^f(fUHHD$yL$HY $f(L$L$ff(f.w!QY $f.w1Q^H]f(D$f(`T$ff(f( $B $f(AVUHSH D$f/L$f/fHHfH~ K^L$fI~fHn^ .^L$$fInAX$= f/rf/ށv\ $H []^A^f(fDD$HH$D$ $H []XA^^f(DfHn~f(fIn^L$ $e $f(^T$T$f(_\\$f(L$\$T$$\f(X$L$H []\A^f(ff.UHH@D$8nzL$ f/>~\$ f/fd$D$^XD$0/fDf(^f(_XD$L$\f/D$s}HY~l$0Hf(YXXL$f(L$ ^\t$(YL$L$~\Y\f/T$ZHD$D$(T$f(/~f(fW%f( o~f(T$8f8Xf(fTT$X~|$T$\~f(fWf(H@f(]f8f(@d$  }f|$YY%d$Xf.wQXL$\$f(Xf.wsQ\|$ f(f(X^f(YXXL$^L$043X\Y'}H@]f(f(sf(L$L$f(sfHL$$f(|f/v&Xf(YD$X$Hf|\\L$Y$H\H $D$Hf(\sfW[}fY$T$H\f(ff.HL$$ 8\^YD$X$HHD$f(\Yx{ff.wQYD$HRf(f(UHXf(H ^L$\$l$\$L$Y ={f(YYYf(YXff.w^Q\YT$H\$XT$\$T$f(f(X^f/s Y^f(H f(]f(\$T$l\$T$f(|ATSHH(\f($izD$fDH0HD$"%r\d$-\D$^,$f(f(tf $=;f: L,(I*^T$X8T$YT$f(\ \$^Yf(\ ޞ^f/@M7H(L[A\@f(f(H8\\T$L$D$(f(d$^l$ 4$4$d$L$T$f/r;l$ f\$(YYf.w\f(QXH8f(f(\Yf(\Yf(ff.w&Q\H8f($W$f($B$ff.SHH -D$\f(iD$Hf/D$$HYD$G $\f(Yf/rXf(T$ T$$f( $^X f: H,HmH [f.f/ʸrff.UHSH^HHSH HH []DAWIAVAUATUSHH$HL$HIIII LHI LHI LHI LHI LH I H<$~BLE1fDI9w>fDHD!I9rHL$LJIL94$uH[]A\A]A^A_HxHHmH H HL!I9rfH~HH@L8HH9uH[]A\A]A^A_DAWIAVAAUATUSHHL$ta  É É É H~)ME1L!9rHL$DBIM9uH[]A\A]A^A_H~HHfD0HH9uH[]A\A]A^A_DAWHAVAUATAUSHHff f щf AfAA H~TLAL,C1ftAD1D!fA9sHAD!fA9rںDHfCL9uH[]A\A]A^A_H~HQffD#HH9uH[]A\A]A^A_@H@AWAV AUAʼnATU SAAHA H~]L$LAH1'H8AD!A8sMADD!A8stADD!A8rDHCL9uH[]A\A]A^A_DDHCL9uH~ff.@tWATMUH,SH1H2#HH9tuLq#HH9u[]A\fH~ @Hff.HHHff.1HATIUHSHHH HHH HHH HHH HHH HH H HH uLH!H9r[]A\L`H!H9sLPH!H9rfDff.@AUIATUHSHHHv>HFHLdfDLH$CHcHHKCI9u؃HuH[]A\A]DLHHCHH9uH[]A\A]f.AUIATIH5lUtFH=lzHHtCLHLHI41E]A\A]fDH=l4H븸AT1ҾUHSH dH%(HD$15uqH MHHDžpHMHHDžDž H HHJH9uHT$dH+%(-H []A\1HVHcHBHH)HH H1HHHH1HHXH H)H$HBHH)HH H1HHHH1HHPH H)HL$HAHH)HH H1H HHH1HHH H)IHI1HI1I1YI1HHHH@HHH)HH H1HHHH1HHH H)I1HL1H1vAVIAUIUHHuuH]A]A^LLHD$ D$ H]A]A^UIIS+H+HHHH1Hiel H HouApH7LJpM9MME11HHHHH1K4Hi fH3LIƉH3HpuHxH7M9LNIuHtHLHH1HieX]I3)HIHpuHxHHu[]HLJHHLJAWAVAUATIUSHH$fH*ƅt H9$Lǃf(\d$Hf/l$@|$Ht$@YXYf(f(f: H,f(D$`ff.H; f(QD\$@mYm=lAYf(|$(\f: Xf(D$ff(H*XXlXf($\f(\$xld$0^f(\XzlfD(\$8\$HYf(f(\^f(YXYf(\f(AY\$h^Yf(XYfA(AXXL$pYf(D$fA(^D^Xf(D$PfA(XD$ MI)IEH$H0YD$ HD$L$f/L$f(f/L$|$l$8fH*5\Y^XXL$0\X\$(fT^\f/kf: L,MI)LHIHH~#L$`YL$(fH*\f/ID$fd$H^d$@H*YI9Jf/$M)f/|$(MGH[]LA\A]A^A_Df/L$PL$XD$d$h^XD$0f: L,ML$X\L$T$YYHEI9Qff(H*H^\YI9}'fL$XD$T$L$Xf(D$x^\$p\f: L,M9\L$PUYYT$pVfDIGH9ff(H*H^\^H9~f(^ hLfX h|$`If(YXX h^HH*^^f(XL$(YL$X\$\$L$Xf(f(\f/XT$Xf/IGffH*fEHEH*$fL*ID$L)H*fD($$DYfD(DYfA(fA(D$^AYfD(t$DYD$D$$D$D$$D$fA(^D$h\$@$$l$Yl$HYf(^35[g$-RgfD(%MgD$f(D$^f(f(D$DY$D$D$D$\^\ffD(f(T$X^D\D^ ff(A\fEM*DXL$(DY$f($A^DfEXfEM*EYfD(EXA^AXfD(E^E\fD(E^E\fD(E^E\fD(E^fD(E\E^E^AXfD(E^A^E\fD(E^\A^E\fD(E^\A^E\fD(E^\A^E\D^\^L$A^E^DXDXfA/$YT$\Xf: L,ff(l$H\|$@S$f.t$Yd$H%dH$T$pt$0Yl$@t$x|$t$8T$ t$hD$`t$Pd$(jD$`T$|$T$L$ff.@ATUHSHH0$t H9b$=fHDžH*\f(T$|$BT$YT$ $T$L$D$YfYX f.Q%DA9ot$Hs0HS(*IuLL讣I;LHcLL蕣I"ff.fAVAUATUHSHHGH;Ht$gH;HWBLt$LbE1 tc膮HH QP ;VLLAI]p VP H=~(9|.M HL[]A\A]A^Lof @9}@$H\$IHHHXHELM)ĭX SH/P ;$1LHAI虭H QP H= @9M Im'IELP0Ht$If苭IHtNE1H=1o茬fLB1 uH}Ht$AIHNH5oH8f.諬@$,f1LHCIH=n@E1軬IHuH̪H5nH8蝩1ɺAIfDAUATUHSHHGH;ܫH;7HOQE1 Lat}ЫHAp VP ;1LAI訫p VP H=~#9|)MHL[]A\A] @9}a@$LozLL-Mtu0X SHP ;|1LHA\@H11[]A\A]DKIHtYE1UH=lL fHLH1[]A\A]YfH=lpHH5lH8ԧff.@AVAUIATUHHtyH="I芭IHɫHHLE1LHLYIHmtHL]A\A]A^HEHP0HL]A\A]A^1YIHtsH=IHtMHHHt@E1LLHLܪI,$IxID$LP0HmisI,$u ID$LP0HE1]LA\A]A^ff.AUATUHHH@@1ߨIHtg1HH芫ImIu IELP0MtAIL$@tsLH8I,$uID$]LA\A]H@0fD]A\A]f.H]A\A]H]H5JA\A]H8饥DHHH5IH815AWAVAUATUSHHhL%$H=HT$HL$IT$LD$dH%(HD$X1H;5 ILaHH]HHEH5HHHIMHm`IHIELhIH H3H5H贩<IFHHH p VP ; HL$ LLLIHL$ p VP H= 9 MI. I,$o I/E I}" L%H=|IT$LHHHHEH5HHHIMHmX ID$1E1H5Ht$ H9%H;եH;0Hc6 LT$(ͥLT$(HI|MtLPHt$HcHItHt$EHHItEHIEMlID$HHvH p VP ;HL$1LLHGHL$p VP H= @9 HI.u IFLP0I,$u ID$LP0H}u HEHP0HEH5XHHHIMzIEH5,LHHIMbLLkIHAI,$u ID$LP0I/u IGLP0L;5L;58 L;5ڣ+ LLAą\ I.u IFLP0EU IEL=HD$HD$LL`LIHjH@HHq LHt$LIHOHD$H5wL`Ht$L辤Ht$HIH@HH- LHt$LIHH@H;D$ k MgM^ IWI$HI/uIGHT$LP0HT$HLHT$ HT$HITI,$I I/uIGLD$LP0LD$I(u I@LP0LeE1HD$Mf.H8HH0H0H0H0蕢Ht$JE1HE (s@H(H0H0H@(;u}MHcHLH0H@H0Pt8H(HR8HcR H0;u|IM9 @ID$LP0oIGLP0UIBLP0< @ID$LD$LP0LD$M캃(.L%DL5UID$HHh {H H‹@ B ;" HL$1LLIIHL$X SP H= @9! M LI,$u ID$LP0(@$_HD$LLT$0LT$(HD$8HD$Ll$HHD$@H)Ht0ILT$(HHhHVH}(MtI*t E11E1IBLP0!LLLIH~HV]HZHW(I.t1E1@$H=WHL$ ]HL$ )HuH=H5VH8wfHuHH=LHGHHHtH HLH5TH81^H Ui(oH=f3H eSi(֕Im E1)^HuJH=LHGHHwIMM"HLH5\TH81迖1'~HIRHTIE11Hk(I*#E11~'HTlHiW,(IE1HILT$LT$@$>LWIZMT$MIl$IHEI,$HEIMt$MFIl$IHEI,$u ID$LP0HEIHgLH8܎(0(!處Ia|HuGH=LHGHHJIMHLH5}RH81(,I HH85HSH(I.+H H=05nDHR{Hxf(HRTHQ?(nHcRE1E1'H$ (L=IH&R~H(ME1=I$1E1HI$I41LLHHHQM1~HE1k (HQYHVD(rLT$IP1LL蘒HHH=QE1H(]HQE1H#($H=QHL$HL$ ÏHuH׍H5QH8訌gHPMkHhV%(H=2QHL$舎HL$:QHHaH5QH82pH.PE1H*(?ID$LT$(LIP0HELT$(t1LLI.I>/H=rPHL$ȍHL$I.IFLP0 LLLLT$运LT$HIVH_O)H&-(2HuHFH5OH8wH=OHL$LT$LT$HL$HNIH(I.$I$HI$LT$蘍LT$H,HH5\OH8tLT$ u@$]rHIN~H'iH1LH IHNH(H1LLT$H+ϏI貌HtX(1H=WNHL$譋HL$fDL1L軎IHE1HnH5'NH8?I,$uE11H H=*5hAWAVAUATUSHHhL%H=}HT$HL$IT$LD$dH%(HD$X1H;5 ILHH]HHEH56HHHIMHmIHIELhvIH HH54H4<IFHH蓊H p VP ; HL$ LLLIcHL$ p VP H= 9 MI. I,$o I/E I}" L%#H=IT$L菊HHHHEH5\HHHIMHmX ID$1E1H5:Ht$ H9%H;UH;Hc6 LT$(MLT$(HI|MtLPHt$HcHItHt$EHHItEHIEMlID$HHH gp VP ;HL$1LLHLjHL$p VP H= @9 HI.u IFLP0I,$u ID$LP0H}u HEHP0HEH5HHHIMzIEH5LHHIMbLLIHAI,$u ID$LP0I/u IGLP0L;5L;5<8 L;5Z+ L̈Aą\ I.u IFLP0EU IEL=]HD$HD$LL`L蔈IHjH@HHq LHt$LIHOHD$H5L`Ht$L>Ht$HIH@HH- LHt$LIHH@H;D$ k MgM^ IWI$HI/uIGHT$LP0HT$HLHT$HT$HITI,$I I/uIGLD$LP0LD$I(u I@LP0oLeE1HD$Mf.H8HH0H0H0 dHt$JE1HE 'rH(H0H0H@(;u}MHcHLH0H@H0Pt8H(HR8HcR H0;u|IM9 @ID$LP0oIGLP0UIBLP0< @ID$LD$LP0LD$M캴,.L%ijL5żID$HHh yH lyH‹@ B ;" HL$1LLIyHL$X SP H= @9! M L`I,$u ID$LP0,jy@$_HD$LLT$0LT$(HD$8HD$Ll$HHD$@H)Ht0LT$(HHhHy:CH@.+MtI*t E11E1IBLP0!LLLn{IH~H:ݱHڱȱ+I.t1E1wx@$H=:HL$ wHL$ xHuHvH5v:H8uwf{xHuHH=LHGHHHtH HxLH5{8H81zH /9+H=H Ӱ+VyIm E1)wHuJH=jLHGHHwIMM"HswLH57H81?z11+HzIRHo8IE11H8*+I*#E113+H"8Hׯ~+IE1HILT$}vLT$@$>LyIZMT$MIl$IHEI,$HEIMt$MFIl$IHEI,$u ID$LP0HEIHuLH8\r1,0+!yIauHuGH=LHGHHJIMHuLH55H81`xp+xI H=uH8qH6PHM;3,I.+H 2(H=35vnDH16H+H 6ԭHѭ+nH5E1E1Hr+LwIH5pHm[S+ME1=I$1E1HI$I41LLvHHH>5M1HE1^+H5٬H֬Ĭ,vLT$IP1LLvHHH4E1Ho+]H4E1ZHWEu+$H=!5HL$wrHL$ CsHuHWqH55H8(pgH$4MH֫w+H=4HL$rHL$:rHHpH54H8opH3E1uHr`|+?ID$LT$(LIP0HELT$(t1LLtI.I>/H=3HL$HqHL$I.IFLP0 LLLLT$?tLT$HIVH2H+qHuHoH53H8nwH=K3HL$LT$pLT$HL$Hi2I0H-@,I.$I$HI$LT$qLT$H,H#oH52H8mLT$ s{p@$srH1H~M+iH1LH sIH1MHJ8+H1LLT$H+OsI2pHtX,1H=1HL$-oHL$fDL1L;rIHE1HmH51H8lI,$uE11glH vH=5epAWAVAUATUSHHhL%$H=HT$HL$IT$LD$L $dH%(HD$X1H;5~n(IL]oHHHHEH5HHHIMHm\nIH/IELhoIH H/nH5HpIFHH nH mp VP ;aHL$ LLLImHL$ p VP H= 9 M:I. I,$ I/q I}N L%H=xIT$L nHHHHEH5ئHHHIMHm ID$1E1H5kHt$ H9qH;l^H;,mHcb LT$(lLT$(HIMtLPHt$HcHItHt$EHHItHt$EHHItEHIEMlID$HH_lH kp VP ;HL$1LLH0lHL$p VP H= @9HI.u IFLP0I,$u ID$LP0H}u HEHP0HEH5AHHHIMIEH5LHHIMLLThIHzI,$u ID$LP0I/u IGLP0L;5jL;5jq L;5jd L5lAą I.u IFLP0E IEL=ƤHD$H$LL`LkIHH@HH LH4$LIHH$H5cL`Ht$LkHt$HIH@HHi LH4$LIHH@H;D$  MgM IWI$HI/uIGH$LP0H$HLH$H$HII,$I@ I/uIGL$LP0L$I(u I@LP0kLeE1H$MfDH@HH0H8H0H0H H0HH0-gHt$JE1HE (s@H(H0H0H@(;u}MHcHLH0H@H0Pt8H(HR8HcR H0;u|IM9,H<$cIFL%(HHX{hH gp VP ;ZH $1LLIMhH $X SP H=m 9 MI.IFLP0Mr0H Y)H=L H 5iImE1HEHP03fL8hIH HIFH5HH LIMZ I.ID$1E1H5eHt$ H9 H;fH;ZgHc=fIHMtLpHt$HcHItHt$EHHItHt$EHHItID$HHfH fp VP ;HL$1LLHwfHL$p VP H= @9 HImu IELP0I,$u ID$LP0H}u HEHP0L-+H=IULfIH HID$H5HHGLIM I,$uID$LT$LP0LT$HELT$HH5 HHLT$IMLT$JeLT$HIVLhLT$fLT$HIHeH5HLT$gLT$ IBLMLT$dH UdLT$p V;P cHL$LLLLT$AIdHL$LT$p VP =H9 MI*I/I,$I}jIELP0[@IELP0fIGLP0I}fID$LP0I/ZfIFLP0I,$0f @9c@$@HEHP0mIFLP0/ID$@=HD$LT$0E1Ll$PHD$8HD$HD$@HD$HD$HIT$H)Ht0HBR uMD$LT$HLLT$HHMI*IBLP0ID$@=HD$Lt$0E1HD$8HD$HD$@HD$HD$HIT$H)Ht0HBR uML$HLHHMI.IFLP0H#E1^H[I.I.1I,$u ID$LP0MtI/u IGLP0H  H=,5cMtImE1HtHmu HEHP0HD$XdH+%(HhL[]A\A]A^A_HD$LLt$0HD$8HD$HD$@HD$HD$HH)Ht0ذHHH"WHTB.1E1MI.AE11IFLP0MucHP0H;0HHP0H0H0H0I,$I}LIELIP0Dy:QH@(H0H(H)0#H0HcHLH(L;(}ILH(H0H(H0H@0H0H@(H0H(H+0H012DH!E1ȘHŘ.eLNIHH Hx/I. I/DH e[H=z5J`Nc_HuHH=LHGHH! HtH$H^LH5cH81a.H H=ԗE1H Η5C`HaIHIE11H|.I,$[E131.HTMHN@I. E1E1 Id]@$IID$LP0> @ID$LP0RIGLP08IBLP0 @ID$L$LP0L$M]/0L%L5ID$HHb -]H \H‹@ B ; H $1LLI\H $X SP H= @9 M L蔳I,$u ID$LP0l/\@$FHD$LLT$0LT$(HD$8HD$Ll$PHD$@HD$HD$HH)Ht0LT$(HHLHmHjX%/MtI*t E1D1E1IBLP0LLL^IH*H;H.I.l1E1[@$nH=HL$ [HL$ [HuHYH5H8Xw[HuHH=7LHGHHHtHHB[LH5H81^H _/H=>H /\Im E1[HuJH=LHGHHrIMMHZLH5 H81o]1.H]IHIE11HhZL/I*E11.HRH.IE1HILT$YLT$@$!L]I*MT$MIl$IHEI,$HEIRMt$MIl$IHEI,$u ID$LP0HEIHYLH8U/V/I\I(,YHuGH=LHGHHEIMHXLH5-H81[.[IHmXH8THH}k/I.+H bXH=w5GY3DHa+H(3/H:HX/nHE1E1אHԐ.LZIHH.ME11I$1E1HI$=(1LLYHHHnM13H0E1.H? HZ/"ZLT$I31LLHYHHHE1HD/QHE1Hu.$H=QHL$UHL$sVHuHTH5@H8XSgHTMH.H=HL$8UHL$VHHTH5H8RpHE1H.?ID$LT$(LIP0HELT$((1LLWI.IH="H $yTH $I.IFLP0LLLLT$qWLT$HI;HۍH؍ƍ.THuHRH5H8QwH=}HL$LT$SLT$HL$uHIbH_M/I.$I$HI$L$KTL$H.HWRH5H8(QL$*WS@$WwHȌHŌ.lH1LHVIHHm-/H1LLT$HVIgSHtUh/H= H $cRH $DL1LsUIHE1H&QH5H8OI,$uE11OH H=5 Tff.ATUHSHHGH;RH;Pt7HXhHHCHt HHH[]A\HHHH?H>HHH?H9}ttpHDHH[]A\fDHHHH?t3u/H>HHH?H9},t(HUHHH[]A\HHH0PIHthHHNI,$sIT$HD$LR0HD$YHHHt$Ht$HxHHC1HRPHt$H8OHt$tHt$PHCHt$@ATUSHHWHBhHHxHQPHHH9FHFHPHw+H*vHt1HH[]A\@HvCHH ƐHHuRPHHtHoOHH2$MtL+OHCH5HPHOH81R%fHNHRH5~H81RH1[]A\fHRIHeHNI,$HIHD$ID$LP0Ht$/@Ht"HjNHf1fvCHH ff.AUATUHSHHGH;$Ot2HXhHHC(HHHH[]A\A]HOHHHIHxTH9}OHOAH 1. The Zipf distribution (also known as the zeta distribution) is a continuous probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. Parameters ---------- a : float or array_like of floats Distribution parameter. Should be greater than 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Zipf distribution. See Also -------- scipy.stats.zipf : probability density function, distribution, or cumulative density function, etc. Notes ----- The probability density for the Zipf distribution is .. math:: p(x) = \frac{x^{-a}}{\zeta(a)}, where :math:`\zeta` is the Riemann Zeta function. It is named for the American linguist George Kingsley Zipf, who noted that the frequency of any word in a sample of a language is inversely proportional to its rank in the frequency table. References ---------- .. [1] Zipf, G. K., "Selected Studies of the Principle of Relative Frequency in Language," Cambridge, MA: Harvard Univ. Press, 1932. Examples -------- Draw samples from the distribution: >>> a = 2. # parameter >>> s = np.random.zipf(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy import special Truncate s values at 50 so plot is interesting: >>> count, bins, ignored = plt.hist(s[s<50], 50, normed=True) >>> x = np.arange(1., 50.) >>> y = x**(-a) / special.zetac(a) >>> plt.plot(x, y/max(y), linewidth=2, color='r') >>> plt.show() zipfzeros weibull(a, size=None) Draw samples from a Weibull distribution. Draw samples from a 1-parameter Weibull distribution with the given shape parameter `a`. .. math:: X = (-ln(U))^{1/a} Here, U is drawn from the uniform distribution over (0,1]. The more common 2-parameter Weibull, including a scale parameter :math:`\lambda` is just :math:`X = \lambda(-ln(U))^{1/a}`. Parameters ---------- a : float or array_like of floats Shape of the distribution. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Weibull distribution. See Also -------- scipy.stats.weibull_max scipy.stats.weibull_min scipy.stats.genextreme gumbel Notes ----- The Weibull (or Type III asymptotic extreme value distribution for smallest values, SEV Type III, or Rosin-Rammler distribution) is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. This class includes the Gumbel and Frechet distributions. The probability density for the Weibull distribution is .. math:: p(x) = \frac{a} {\lambda}(\frac{x}{\lambda})^{a-1}e^{-(x/\lambda)^a}, where :math:`a` is the shape and :math:`\lambda` the scale. The function has its peak (the mode) at :math:`\lambda(\frac{a-1}{a})^{1/a}`. When ``a = 1``, the Weibull distribution reduces to the exponential distribution. References ---------- .. [1] Waloddi Weibull, Royal Technical University, Stockholm, 1939 "A Statistical Theory Of The Strength Of Materials", Ingeniorsvetenskapsakademiens Handlingar Nr 151, 1939, Generalstabens Litografiska Anstalts Forlag, Stockholm. .. [2] Waloddi Weibull, "A Statistical Distribution Function of Wide Applicability", Journal Of Applied Mechanics ASME Paper 1951. .. [3] Wikipedia, "Weibull distribution", http://en.wikipedia.org/wiki/Weibull_distribution Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> s = np.random.weibull(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> x = np.arange(1,100.)/50. >>> def weib(x,n,a): ... return (a / n) * (x / n)**(a - 1) * np.exp(-(x / n)**a) >>> count, bins, ignored = plt.hist(np.random.weibull(5.,1000)) >>> x = np.arange(1,100.)/50. >>> scale = count.max()/weib(x, 1., 5.).max() >>> plt.plot(x, weib(x, 1., 5.)*scale) >>> plt.show() weibullwarningswarn wald(mean, scale, size=None) Draw samples from a Wald, or inverse Gaussian, distribution. As the scale approaches infinity, the distribution becomes more like a Gaussian. Some references claim that the Wald is an inverse Gaussian with mean equal to 1, but this is by no means universal. The inverse Gaussian distribution was first studied in relationship to Brownian motion. In 1956 M.C.K. Tweedie used the name inverse Gaussian because there is an inverse relationship between the time to cover a unit distance and distance covered in unit time. Parameters ---------- mean : float or array_like of floats Distribution mean, should be > 0. scale : float or array_like of floats Scale parameter, should be >= 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(mean, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Wald distribution. Notes ----- The probability density function for the Wald distribution is .. math:: P(x;mean,scale) = \sqrt{\frac{scale}{2\pi x^3}}e^ \frac{-scale(x-mean)^2}{2\cdotp mean^2x} As noted above the inverse Gaussian distribution first arise from attempts to model Brownian motion. It is also a competitor to the Weibull for use in reliability modeling and modeling stock returns and interest rate processes. References ---------- .. [1] Brighton Webs Ltd., Wald Distribution, http://www.brighton-webs.co.uk/distributions/wald.asp .. [2] Chhikara, Raj S., and Folks, J. Leroy, "The Inverse Gaussian Distribution: Theory : Methodology, and Applications", CRC Press, 1988. .. [3] Wikipedia, "Wald distribution" http://en.wikipedia.org/wiki/Wald_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.wald(3, 2, 100000), bins=200, normed=True) >>> plt.show() wald vonmises(mu, kappa, size=None) Draw samples from a von Mises distribution. Samples are drawn from a von Mises distribution with specified mode (mu) and dispersion (kappa), on the interval [-pi, pi]. The von Mises distribution (also known as the circular normal distribution) is a continuous probability distribution on the unit circle. It may be thought of as the circular analogue of the normal distribution. Parameters ---------- mu : float or array_like of floats Mode ("center") of the distribution. kappa : float or array_like of floats Dispersion of the distribution, has to be >=0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mu`` and ``kappa`` are both scalars. Otherwise, ``np.broadcast(mu, kappa).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized von Mises distribution. See Also -------- scipy.stats.vonmises : probability density function, distribution, or cumulative density function, etc. Notes ----- The probability density for the von Mises distribution is .. math:: p(x) = \frac{e^{\kappa cos(x-\mu)}}{2\pi I_0(\kappa)}, where :math:`\mu` is the mode and :math:`\kappa` the dispersion, and :math:`I_0(\kappa)` is the modified Bessel function of order 0. The von Mises is named for Richard Edler von Mises, who was born in Austria-Hungary, in what is now the Ukraine. He fled to the United States in 1939 and became a professor at Harvard. He worked in probability theory, aerodynamics, fluid mechanics, and philosophy of science. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] von Mises, R., "Mathematical Theory of Probability and Statistics", New York: Academic Press, 1964. Examples -------- Draw samples from the distribution: >>> mu, kappa = 0.0, 4.0 # mean and dispersion >>> s = np.random.vonmises(mu, kappa, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import i0 >>> plt.hist(s, 50, normed=True) >>> x = np.linspace(-np.pi, np.pi, num=51) >>> y = np.exp(kappa*np.cos(x-mu))/(2*np.pi*i0(kappa)) >>> plt.plot(x, y, linewidth=2, color='r') >>> plt.show() vonmisesunsafeunique uniform(low=0.0, high=1.0, size=None) Draw samples from a uniform distribution. Samples are uniformly distributed over the half-open interval ``[low, high)`` (includes low, but excludes high). In other words, any value within the given interval is equally likely to be drawn by `uniform`. Parameters ---------- low : float or array_like of floats, optional Lower boundary of the output interval. All values generated will be greater than or equal to low. The default value is 0. high : float or array_like of floats Upper boundary of the output interval. All values generated will be less than high. The default value is 1.0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``low`` and ``high`` are both scalars. Otherwise, ``np.broadcast(low, high).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized uniform distribution. See Also -------- randint : Discrete uniform distribution, yielding integers. random_integers : Discrete uniform distribution over the closed interval ``[low, high]``. random_sample : Floats uniformly distributed over ``[0, 1)``. random : Alias for `random_sample`. rand : Convenience function that accepts dimensions as input, e.g., ``rand(2,2)`` would generate a 2-by-2 array of floats, uniformly distributed over ``[0, 1)``. Notes ----- The probability density function of the uniform distribution is .. math:: p(x) = \frac{1}{b - a} anywhere within the interval ``[a, b)``, and zero elsewhere. When ``high`` == ``low``, values of ``low`` will be returned. If ``high`` < ``low``, the results are officially undefined and may eventually raise an error, i.e. do not rely on this function to behave when passed arguments satisfying that inequality condition. Examples -------- Draw samples from the distribution: >>> s = np.random.uniform(-1,0,1000) All values are within the given interval: >>> np.all(s >= -1) True >>> np.all(s < 0) True Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 15, normed=True) >>> plt.plot(bins, np.ones_like(bins), linewidth=2, color='r') >>> plt.show() uniformuint8uint64uint32uint16uint triangular(left, mode, right, size=None) Draw samples from the triangular distribution over the interval ``[left, right]``. The triangular distribution is a continuous probability distribution with lower limit left, peak at mode, and upper limit right. Unlike the other distributions, these parameters directly define the shape of the pdf. Parameters ---------- left : float or array_like of floats Lower limit. mode : float or array_like of floats The value where the peak of the distribution occurs. The value should fulfill the condition ``left <= mode <= right``. right : float or array_like of floats Upper limit, should be larger than `left`. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``left``, ``mode``, and ``right`` are all scalars. Otherwise, ``np.broadcast(left, mode, right).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized triangular distribution. Notes ----- The probability density function for the triangular distribution is .. math:: P(x;l, m, r) = \begin{cases} \frac{2(x-l)}{(r-l)(m-l)}& \text{for $l \leq x \leq m$},\\ \frac{2(r-x)}{(r-l)(r-m)}& \text{for $m \leq x \leq r$},\\ 0& \text{otherwise}. \end{cases} The triangular distribution is often used in ill-defined problems where the underlying distribution is not known, but some knowledge of the limits and mode exists. Often it is used in simulations. References ---------- .. [1] Wikipedia, "Triangular distribution" http://en.wikipedia.org/wiki/Triangular_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.triangular(-3, 0, 8, 100000), bins=200, ... normed=True) >>> plt.show() triangular tomaxint(size=None) Random integers between 0 and ``sys.maxint``, inclusive. Return a sample of uniformly distributed random integers in the interval [0, ``sys.maxint``]. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray Drawn samples, with shape `size`. See Also -------- randint : Uniform sampling over a given half-open interval of integers. random_integers : Uniform sampling over a given closed interval of integers. Examples -------- >>> RS = np.random.mtrand.RandomState() # need a RandomState object >>> RS.tomaxint((2,2,2)) array([[[1170048599, 1600360186], [ 739731006, 1947757578]], [[1871712945, 752307660], [1601631370, 1479324245]]]) >>> import sys >>> sys.maxint 2147483647 >>> RS.tomaxint((2,2,2)) < sys.maxint array([[[ True, True], [ True, True]], [[ True, True], [ True, True]]], dtype=bool) tolthreading__test__takesvdsum(pvals[:-1]) > 1.0subtractstridesstate must be 624 longsstate standard_t(df, size=None) Draw samples from a standard Student's t distribution with `df` degrees of freedom. A special case of the hyperbolic distribution. As `df` gets large, the result resembles that of the standard normal distribution (`standard_normal`). Parameters ---------- df : int or array_like of ints Degrees of freedom, should be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard Student's t distribution. Notes ----- The probability density function for the t distribution is .. math:: P(x, df) = \frac{\Gamma(\frac{df+1}{2})}{\sqrt{\pi df} \Gamma(\frac{df}{2})}\Bigl( 1+\frac{x^2}{df} \Bigr)^{-(df+1)/2} The t test is based on an assumption that the data come from a Normal distribution. The t test provides a way to test whether the sample mean (that is the mean calculated from the data) is a good estimate of the true mean. The derivation of the t-distribution was first published in 1908 by William Gosset while working for the Guinness Brewery in Dublin. Due to proprietary issues, he had to publish under a pseudonym, and so he used the name Student. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics With R", Springer, 2002. .. [2] Wikipedia, "Student's t-distribution" http://en.wikipedia.org/wiki/Student's_t-distribution Examples -------- From Dalgaard page 83 [1]_, suppose the daily energy intake for 11 women in Kj is: >>> intake = np.array([5260., 5470, 5640, 6180, 6390, 6515, 6805, 7515, \ ... 7515, 8230, 8770]) Does their energy intake deviate systematically from the recommended value of 7725 kJ? We have 10 degrees of freedom, so is the sample mean within 95% of the recommended value? >>> s = np.random.standard_t(10, size=100000) >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 Calculate the t statistic, setting the ddof parameter to the unbiased value so the divisor in the standard deviation will be degrees of freedom, N-1. >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> import matplotlib.pyplot as plt >>> h = plt.hist(s, bins=100, normed=True) For a one-sided t-test, how far out in the distribution does the t statistic appear? >>> np.sum(s>> s = np.random.standard_normal(8000) >>> s array([ 0.6888893 , 0.78096262, -0.89086505, ..., 0.49876311, #random -0.38672696, -0.4685006 ]) #random >>> s.shape (8000,) >>> s = np.random.standard_normal(size=(3, 4, 2)) >>> s.shape (3, 4, 2) standard_normal standard_gamma(shape, size=None) Draw samples from a standard Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, shape (sometimes designated "k") and scale=1. Parameters ---------- shape : float or array_like of floats Parameter, should be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` is a scalar. Otherwise, ``np.array(shape).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", http://en.wikipedia.org/wiki/Gamma_distribution Examples -------- Draw samples from the distribution: >>> shape, scale = 2., 1. # mean and width >>> s = np.random.standard_gamma(shape, 1000000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps >>> count, bins, ignored = plt.hist(s, 50, normed=True) >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ \ ... (sps.gamma(shape) * scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') >>> plt.show() standard_gamma standard_exponential(size=None) Draw samples from the standard exponential distribution. `standard_exponential` is identical to the exponential distribution with a scale parameter of 1. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray Drawn samples. Examples -------- Output a 3x8000 array: >>> n = np.random.standard_exponential((3, 8000)) standard_exponential standard_cauchy(size=None) Draw samples from a standard Cauchy distribution with mode = 0. Also known as the Lorentz distribution. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- samples : ndarray or scalar The drawn samples. Notes ----- The probability density function for the full Cauchy distribution is .. math:: P(x; x_0, \gamma) = \frac{1}{\pi \gamma \bigl[ 1+ (\frac{x-x_0}{\gamma})^2 \bigr] } and the Standard Cauchy distribution just sets :math:`x_0=0` and :math:`\gamma=1` The Cauchy distribution arises in the solution to the driven harmonic oscillator problem, and also describes spectral line broadening. It also describes the distribution of values at which a line tilted at a random angle will cut the x axis. When studying hypothesis tests that assume normality, seeing how the tests perform on data from a Cauchy distribution is a good indicator of their sensitivity to a heavy-tailed distribution, since the Cauchy looks very much like a Gaussian distribution, but with heavier tails. References ---------- .. [1] NIST/SEMATECH e-Handbook of Statistical Methods, "Cauchy Distribution", http://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm .. [2] Weisstein, Eric W. "Cauchy Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/CauchyDistribution.html .. [3] Wikipedia, "Cauchy distribution" http://en.wikipedia.org/wiki/Cauchy_distribution Examples -------- Draw samples and plot the distribution: >>> s = np.random.standard_cauchy(1000000) >>> s = s[(s>-25) & (s<25)] # truncate distribution so it plots well >>> plt.hist(s, bins=100) >>> plt.show() standard_cauchysqrtsortsize is not compatible with inputssizesignbitsigma < 0.0sigma < 0sigmaside shuffle(x) Modify a sequence in-place by shuffling its contents. This function only shuffles the array along the first axis of a multi-dimensional array. The order of sub-arrays is changed but their contents remains the same. Parameters ---------- x : array_like The array or list to be shuffled. Returns ------- None Examples -------- >>> arr = np.arange(10) >>> np.random.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] Multi-dimensional arrays are only shuffled along the first axis: >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.shuffle(arr) >>> arr array([[3, 4, 5], [6, 7, 8], [0, 1, 2]]) shuffle_shape_from_sizeshape < 0shapeset_stateseedsearchsortedscale <= 0scale <= 0.0scale < 0.0scale < 0scalesafertolrngstaterngrightreversedreturn_indexreshapereplacereduce rayleigh(scale=1.0, size=None) Draw samples from a Rayleigh distribution. The :math:`\chi` and Weibull distributions are generalizations of the Rayleigh. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. Should be >= 0. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Rayleigh distribution. Notes ----- The probability density function for the Rayleigh distribution is .. math:: P(x;scale) = \frac{x}{scale^2}e^{\frac{-x^2}{2 \cdotp scale^2}} The Rayleigh distribution would arise, for example, if the East and North components of the wind velocity had identical zero-mean Gaussian distributions. Then the wind speed would have a Rayleigh distribution. References ---------- .. [1] Brighton Webs Ltd., "Rayleigh Distribution," http://www.brighton-webs.co.uk/distributions/rayleigh.asp .. [2] Wikipedia, "Rayleigh distribution" http://en.wikipedia.org/wiki/Rayleigh_distribution Examples -------- Draw values from the distribution and plot the histogram >>> values = hist(np.random.rayleigh(3, 100000), bins=200, normed=True) Wave heights tend to follow a Rayleigh distribution. If the mean wave height is 1 meter, what fraction of waves are likely to be larger than 3 meters? >>> meanvalue = 1 >>> modevalue = np.sqrt(2 / np.pi) * meanvalue >>> s = np.random.rayleigh(modevalue, 1000000) The percentage of waves larger than 3 meters is: >>> 100.*sum(s>3)/1000000. 0.087300000000000003 rayleighravelrange random_sample(size=None) Return random floats in the half-open interval [0.0, 1.0). Results are from the "continuous uniform" distribution over the stated interval. To sample :math:`Unif[a, b), b > a` multiply the output of `random_sample` by `(b-a)` and add `a`:: (b - a) * random_sample() + a Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray of floats Array of random floats of shape `size` (unless ``size=None``, in which case a single float is returned). Examples -------- >>> np.random.random_sample() 0.47108547995356098 >>> type(np.random.random_sample()) >>> np.random.random_sample((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) Three-by-two array of random numbers from [-5, 0): >>> 5 * np.random.random_sample((3, 2)) - 5 array([[-3.99149989, -0.52338984], [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) random_sample random_integers(low, high=None, size=None) Random integers of type np.int between `low` and `high`, inclusive. Return random integers of type np.int from the "discrete uniform" distribution in the closed interval [`low`, `high`]. If `high` is None (the default), then results are from [1, `low`]. The np.int type translates to the C long type used by Python 2 for "short" integers and its precision is platform dependent. This function has been deprecated. Use randint instead. .. deprecated:: 1.11.0 Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution (unless ``high=None``, in which case this parameter is the *highest* such integer). high : int, optional If provided, the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- random.randint : Similar to `random_integers`, only for the half-open interval [`low`, `high`), and 0 is the lowest value if `high` is omitted. Notes ----- To sample from N evenly spaced floating-point numbers between a and b, use:: a + (b - a) * (np.random.random_integers(N) - 1) / (N - 1.) Examples -------- >>> np.random.random_integers(5) 4 >>> type(np.random.random_integers(5)) >>> np.random.random_integers(5, size=(3,2)) array([[5, 4], [3, 3], [4, 5]]) Choose five random numbers from the set of five evenly-spaced numbers between 0 and 2.5, inclusive (*i.e.*, from the set :math:`{0, 5/8, 10/8, 15/8, 20/8}`): >>> 2.5 * (np.random.random_integers(5, size=(5,)) - 1) / 4. array([ 0.625, 1.25 , 0.625, 0.625, 2.5 ]) Roll two six sided dice 1000 times and sum the results: >>> d1 = np.random.random_integers(1, 6, 1000) >>> d2 = np.random.random_integers(1, 6, 1000) >>> dsums = d1 + d2 Display results as a histogram: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(dsums, 11, normed=True) >>> plt.show() random_integersrandom randn(d0, d1, ..., dn) Return a sample (or samples) from the "standard normal" distribution. If positive, int_like or int-convertible arguments are provided, `randn` generates an array of shape ``(d0, d1, ..., dn)``, filled with random floats sampled from a univariate "normal" (Gaussian) distribution of mean 0 and variance 1 (if any of the :math:`d_i` are floats, they are first converted to integers by truncation). A single float randomly sampled from the distribution is returned if no argument is provided. This is a convenience function. If you want an interface that takes a tuple as the first argument, use `numpy.random.standard_normal` instead. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, should be all positive. If no argument is given a single Python float is returned. Returns ------- Z : ndarray or float A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from the standard normal distribution, or a single such float if no parameters were supplied. See Also -------- random.standard_normal : Similar, but takes a tuple as its argument. Notes ----- For random samples from :math:`N(\mu, \sigma^2)`, use: ``sigma * np.random.randn(...) + mu`` Examples -------- >>> np.random.randn() 2.1923875335537315 #random Two-by-four array of samples from N(3, 6.25): >>> 2.5 * np.random.randn(2, 4) + 3 array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], #random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) #random randn_randint_type randint(low, high=None, size=None, dtype='l') Return random integers from `low` (inclusive) to `high` (exclusive). Return random integers from the "discrete uniform" distribution of the specified dtype in the "half-open" interval [`low`, `high`). If `high` is None (the default), then results are from [0, `low`). Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution (unless ``high=None``, in which case this parameter is one above the *highest* such integer). high : int, optional If provided, one above the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result. All dtypes are determined by their name, i.e., 'int64', 'int', etc, so byteorder is not available and a specific precision may have different C types depending on the platform. The default value is 'np.int'. .. versionadded:: 1.11.0 Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- random.random_integers : similar to `randint`, only for the closed interval [`low`, `high`], and 1 is the lowest value if `high` is omitted. In particular, this other one is the one to use to generate uniformly distributed discrete non-integers. Examples -------- >>> np.random.randint(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) >>> np.random.randint(1, size=10) array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0]) Generate a 2 x 4 array of ints between 0 and 4, inclusive: >>> np.random.randint(5, size=(2, 4)) array([[4, 0, 2, 1], [3, 2, 2, 0]]) randint_helpers.pxirandint_rand_uint8_rand_uint64_rand_uint32_rand_uint16_rand_int8_rand_int64_rand_int32_rand_int16 rand(d0, d1, ..., dn) Random values in a given shape. Create an array of the given shape and populate it with random samples from a uniform distribution over ``[0, 1)``. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, should all be positive. If no argument is given a single Python float is returned. Returns ------- out : ndarray, shape ``(d0, d1, ..., dn)`` Random values. See Also -------- random Notes ----- This is a convenience function. If you want an interface that takes a shape-tuple as the first argument, refer to np.random.random_sample . Examples -------- >>> np.random.rand(3,2) array([[ 0.14022471, 0.96360618], #random [ 0.37601032, 0.25528411], #random [ 0.49313049, 0.94909878]]) #random _rand_bool_randrandraise__pyx_vtable__pvalsprodprobabilities do not sum to 1probabilities are not non-negative power(a, size=None) Draws samples in [0, 1] from a power distribution with positive exponent a - 1. Also known as the power function distribution. Parameters ---------- a : float or array_like of floats Parameter of the distribution. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized power distribution. Raises ------ ValueError If a < 1. Notes ----- The probability density function is .. math:: P(x; a) = ax^{a-1}, 0 \le x \le 1, a>0. The power function distribution is just the inverse of the Pareto distribution. It may also be seen as a special case of the Beta distribution. It is used, for example, in modeling the over-reporting of insurance claims. References ---------- .. [1] Christian Kleiber, Samuel Kotz, "Statistical size distributions in economics and actuarial sciences", Wiley, 2003. .. [2] Heckert, N. A. and Filliben, James J. "NIST Handbook 148: Dataplot Reference Manual, Volume 2: Let Subcommands and Library Functions", National Institute of Standards and Technology Handbook Series, June 2003. http://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> samples = 1000 >>> s = np.random.power(a, samples) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=30) >>> x = np.linspace(0, 1, 100) >>> y = a*x**(a-1.) >>> normed_y = samples*np.diff(bins)[0]*y >>> plt.plot(x, normed_y) >>> plt.show() Compare the power function distribution to the inverse of the Pareto. >>> from scipy import stats >>> rvs = np.random.power(5, 1000000) >>> rvsp = np.random.pareto(5, 1000000) >>> xx = np.linspace(0,1,100) >>> powpdf = stats.powerlaw.pdf(xx,5) >>> plt.figure() >>> plt.hist(rvs, bins=50, normed=True) >>> plt.plot(xx,powpdf,'r-') >>> plt.title('np.random.power(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, normed=True) >>> plt.plot(xx,powpdf,'r-') >>> plt.title('inverse of 1 + np.random.pareto(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, normed=True) >>> plt.plot(xx,powpdf,'r-') >>> plt.title('inverse of stats.pareto(5)') powerpoisson_lam_max poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. Parameters ---------- lam : float or array_like of floats Expectation of interval, should be >= 0. A sequence of expectation intervals must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. Notes ----- The Poisson distribution .. math:: f(k; \lambda)=\frac{\lambda^k e^{-\lambda}}{k!} For events with an expected separation :math:`\lambda` the Poisson distribution :math:`f(k; \lambda)` describes the probability of :math:`k` events occurring within the observed interval :math:`\lambda`. Because the output is limited to the range of the C long type, a ValueError is raised when `lam` is within 10 sigma of the maximum representable value. References ---------- .. [1] Weisstein, Eric W. "Poisson Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/PoissonDistribution.html .. [2] Wikipedia, "Poisson distribution", http://en.wikipedia.org/wiki/Poisson_distribution Examples -------- Draw samples from the distribution: >>> import numpy as np >>> s = np.random.poisson(5, 10000) Display histogram of the sample: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 14, normed=True) >>> plt.show() Draw each 100 values for lambda 100 and 500: >>> s = np.random.poisson(lam=(100., 500.), size=(100, 2)) poisson permutation(x) Randomly permute a sequence, or return a permuted range. If `x` is a multi-dimensional array, it is only shuffled along its first index. Parameters ---------- x : int or array_like If `x` is an integer, randomly permute ``np.arange(x)``. If `x` is an array, make a copy and shuffle the elements randomly. Returns ------- out : ndarray Permuted sequence or array range. Examples -------- >>> np.random.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) >>> np.random.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.permutation(arr) array([[6, 7, 8], [0, 1, 2], [3, 4, 5]]) permutation pareto(a, size=None) Draw samples from a Pareto II or Lomax distribution with specified shape. The Lomax or Pareto II distribution is a shifted Pareto distribution. The classical Pareto distribution can be obtained from the Lomax distribution by adding 1 and multiplying by the scale parameter ``m`` (see Notes). The smallest value of the Lomax distribution is zero while for the classical Pareto distribution it is ``mu``, where the standard Pareto distribution has location ``mu = 1``. Lomax can also be considered as a simplified version of the Generalized Pareto distribution (available in SciPy), with the scale set to one and the location set to zero. The Pareto distribution must be greater than zero, and is unbounded above. It is also known as the "80-20 rule". In this distribution, 80 percent of the weights are in the lowest 20 percent of the range, while the other 20 percent fill the remaining 80 percent of the range. Parameters ---------- a : float or array_like of floats Shape of the distribution. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Pareto distribution. See Also -------- scipy.stats.lomax : probability density function, distribution or cumulative density function, etc. scipy.stats.genpareto : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Pareto distribution is .. math:: p(x) = \frac{am^a}{x^{a+1}} where :math:`a` is the shape and :math:`m` the scale. The Pareto distribution, named after the Italian economist Vilfredo Pareto, is a power law probability distribution useful in many real world problems. Outside the field of economics it is generally referred to as the Bradford distribution. Pareto developed the distribution to describe the distribution of wealth in an economy. It has also found use in insurance, web page access statistics, oil field sizes, and many other problems, including the download frequency for projects in Sourceforge [1]_. It is one of the so-called "fat-tailed" distributions. References ---------- .. [1] Francis Hunt and Paul Johnson, On the Pareto Distribution of Sourceforge projects. .. [2] Pareto, V. (1896). Course of Political Economy. Lausanne. .. [3] Reiss, R.D., Thomas, M.(2001), Statistical Analysis of Extreme Values, Birkhauser Verlag, Basel, pp 23-30. .. [4] Wikipedia, "Pareto distribution", http://en.wikipedia.org/wiki/Pareto_distribution Examples -------- Draw samples from the distribution: >>> a, m = 3., 2. # shape and mode >>> s = (np.random.pareto(a, 1000) + 1) * m Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, _ = plt.hist(s, 100, normed=True) >>> fit = a*m**a / bins**(a+1) >>> plt.plot(bins, max(count)*fit/max(fit), linewidth=2, color='r') >>> plt.show() paretop must be 1-dimensionalp is nanp >= 1.0p > 1.0p > 1p <= 0.0p < 0.0p < 0poutoperatoroffnumpy.dualnumpy.core.multiarray failed to importnumpynsample < 1nsamplenp normal(loc=0.0, scale=1.0, size=None) Draw random samples from a normal (Gaussian) distribution. The probability density function of the normal distribution, first derived by De Moivre and 200 years later by both Gauss and Laplace independently [2]_, is often called the bell curve because of its characteristic shape (see the example below). The normal distributions occurs often in nature. For example, it describes the commonly occurring distribution of samples influenced by a large number of tiny, random disturbances, each with its own unique distribution [2]_. Parameters ---------- loc : float or array_like of floats Mean ("centre") of the distribution. scale : float or array_like of floats Standard deviation (spread or "width") of the distribution. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized normal distribution. See Also -------- scipy.stats.norm : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gaussian distribution is .. math:: p(x) = \frac{1}{\sqrt{ 2 \pi \sigma^2 }} e^{ - \frac{ (x - \mu)^2 } {2 \sigma^2} }, where :math:`\mu` is the mean and :math:`\sigma` the standard deviation. The square of the standard deviation, :math:`\sigma^2`, is called the variance. The function has its peak at the mean, and its "spread" increases with the standard deviation (the function reaches 0.607 times its maximum at :math:`x + \sigma` and :math:`x - \sigma` [2]_). This implies that `numpy.random.normal` is more likely to return samples lying close to the mean, rather than those far away. References ---------- .. [1] Wikipedia, "Normal distribution", http://en.wikipedia.org/wiki/Normal_distribution .. [2] P. R. Peebles Jr., "Central Limit Theorem" in "Probability, Random Variables and Random Signal Principles", 4th ed., 2001, pp. 51, 51, 125. Examples -------- Draw samples from the distribution: >>> mu, sigma = 0, 0.1 # mean and standard deviation >>> s = np.random.normal(mu, sigma, 1000) Verify the mean and the variance: >>> abs(mu - np.mean(s)) < 0.01 True >>> abs(sigma - np.std(s, ddof=1)) < 0.01 True Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, normed=True) >>> plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) * ... np.exp( - (bins - mu)**2 / (2 * sigma**2) ), ... linewidth=2, color='r') >>> plt.show() normal noncentral_f(dfnum, dfden, nonc, size=None) Draw samples from the noncentral F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters > 1. `nonc` is the non-centrality parameter. Parameters ---------- dfnum : int or array_like of ints Parameter, should be > 1. dfden : int or array_like of ints Parameter, should be > 1. nonc : float or array_like of floats Parameter, should be >= 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum``, ``dfden``, and ``nonc`` are all scalars. Otherwise, ``np.broadcast(dfnum, dfden, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral Fisher distribution. Notes ----- When calculating the power of an experiment (power = probability of rejecting the null hypothesis when a specific alternative is true) the non-central F statistic becomes important. When the null hypothesis is true, the F statistic follows a central F distribution. When the null hypothesis is not true, then it follows a non-central F statistic. References ---------- .. [1] Weisstein, Eric W. "Noncentral F-Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/NoncentralF-Distribution.html .. [2] Wikipedia, "Noncentral F-distribution", http://en.wikipedia.org/wiki/Noncentral_F-distribution Examples -------- In a study, testing for a specific alternative to the null hypothesis requires use of the Noncentral F distribution. We need to calculate the area in the tail of the distribution that exceeds the value of the F distribution for the null hypothesis. We'll plot the two probability distributions for comparison. >>> dfnum = 3 # between group deg of freedom >>> dfden = 20 # within groups degrees of freedom >>> nonc = 3.0 >>> nc_vals = np.random.noncentral_f(dfnum, dfden, nonc, 1000000) >>> NF = np.histogram(nc_vals, bins=50, normed=True) >>> c_vals = np.random.f(dfnum, dfden, 1000000) >>> F = np.histogram(c_vals, bins=50, normed=True) >>> plt.plot(F[1][1:], F[0]) >>> plt.plot(NF[1][1:], NF[0]) >>> plt.show() noncentral_f noncentral_chisquare(df, nonc, size=None) Draw samples from a noncentral chi-square distribution. The noncentral :math:`\chi^2` distribution is a generalisation of the :math:`\chi^2` distribution. Parameters ---------- df : int or array_like of ints Degrees of freedom, should be > 0 as of NumPy 1.10.0, should be > 1 for earlier versions. nonc : float or array_like of floats Non-centrality, should be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` and ``nonc`` are both scalars. Otherwise, ``np.broadcast(df, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral chi-square distribution. Notes ----- The probability density function for the noncentral Chi-square distribution is .. math:: P(x;df,nonc) = \sum^{\infty}_{i=0} \frac{e^{-nonc/2}(nonc/2)^{i}}{i!} \P_{Y_{df+2i}}(x), where :math:`Y_{q}` is the Chi-square with q degrees of freedom. In Delhi (2007), it is noted that the noncentral chi-square is useful in bombing and coverage problems, the probability of killing the point target given by the noncentral chi-squared distribution. References ---------- .. [1] Delhi, M.S. Holla, "On a noncentral chi-square distribution in the analysis of weapon systems effectiveness", Metrika, Volume 15, Number 1 / December, 1970. .. [2] Wikipedia, "Noncentral chi-square distribution" http://en.wikipedia.org/wiki/Noncentral_chi-square_distribution Examples -------- Draw values from the distribution and plot the histogram >>> import matplotlib.pyplot as plt >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, normed=True) >>> plt.show() Draw values from a noncentral chisquare with very small noncentrality, and compare to a chisquare. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, .0000001, 100000), ... bins=np.arange(0., 25, .1), normed=True) >>> values2 = plt.hist(np.random.chisquare(3, 100000), ... bins=np.arange(0., 25, .1), normed=True) >>> plt.plot(values[1][0:-1], values[0]-values2[0], 'ob') >>> plt.show() Demonstrate how large values of non-centrality lead to a more symmetric distribution. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, normed=True) >>> plt.show() noncentral_chisquarenonc < 0noncngood + nbad < nsamplengood < 0ngood negative_binomial(n, p, size=None) Draw samples from a negative binomial distribution. Samples are drawn from a negative binomial distribution with specified parameters, `n` trials and `p` probability of success where `n` is an integer > 0 and `p` is in the interval [0, 1]. Parameters ---------- n : int or array_like of ints Parameter of the distribution, > 0. Floats are also accepted, but they will be truncated to integers. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized negative binomial distribution, where each sample is equal to N, the number of trials it took to achieve n - 1 successes, N - (n - 1) failures, and a success on the, (N + n)th trial. Notes ----- The probability density for the negative binomial distribution is .. math:: P(N;n,p) = \binom{N+n-1}{n-1}p^{n}(1-p)^{N}, where :math:`n-1` is the number of successes, :math:`p` is the probability of success, and :math:`N+n-1` is the number of trials. The negative binomial distribution gives the probability of n-1 successes and N failures in N+n-1 trials, and success on the (N+n)th trial. If one throws a die repeatedly until the third time a "1" appears, then the probability distribution of the number of non-"1"s that appear before the third "1" is a negative binomial distribution. References ---------- .. [1] Weisstein, Eric W. "Negative Binomial Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/NegativeBinomialDistribution.html .. [2] Wikipedia, "Negative binomial distribution", http://en.wikipedia.org/wiki/Negative_binomial_distribution Examples -------- Draw samples from the distribution: A real world example. A company drills wild-cat oil exploration wells, each with an estimated probability of success of 0.1. What is the probability of having one success for each successive well, that is what is the probability of a single success after drilling 5 wells, after 6 wells, etc.? >>> s = np.random.negative_binomial(1, 0.1, 100000) >>> for i in range(1, 11): ... probability = sum(s>> mean = [0, 0] >>> cov = [[1, 0], [0, 100]] # diagonal covariance Diagonal covariance means that points are oriented along x or y-axis: >>> import matplotlib.pyplot as plt >>> x, y = np.random.multivariate_normal(mean, cov, 5000).T >>> plt.plot(x, y, 'x') >>> plt.axis('equal') >>> plt.show() Note that the covariance matrix must be positive semidefinite (a.k.a. nonnegative-definite). Otherwise, the behavior of this method is undefined and backwards compatibility is not guaranteed. References ---------- .. [1] Papoulis, A., "Probability, Random Variables, and Stochastic Processes," 3rd ed., New York: McGraw-Hill, 1991. .. [2] Duda, R. O., Hart, P. E., and Stork, D. G., "Pattern Classification," 2nd ed., New York: Wiley, 2001. Examples -------- >>> mean = (1, 2) >>> cov = [[1, 0], [0, 1]] >>> x = np.random.multivariate_normal(mean, cov, (3, 3)) >>> x.shape (3, 3, 2) The following is probably true, given that 0.6 is roughly twice the standard deviation: >>> list((x[0,0,:] - mean) < 0.6) [True, True] multivariate_normal multinomial(n, pvals, size=None) Draw samples from a multinomial distribution. The multinomial distribution is a multivariate generalisation of the binomial distribution. Take an experiment with one of ``p`` possible outcomes. An example of such an experiment is throwing a dice, where the outcome can be 1 through 6. Each sample drawn from the distribution represents `n` such experiments. Its values, ``X_i = [X_0, X_1, ..., X_p]``, represent the number of times the outcome was ``i``. Parameters ---------- n : int Number of experiments. pvals : sequence of floats, length p Probabilities of each of the ``p`` different outcomes. These should sum to 1 (however, the last element is always assumed to account for the remaining probability, as long as ``sum(pvals[:-1]) <= 1)``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. Examples -------- Throw a dice 20 times: >>> np.random.multinomial(20, [1/6.]*6, size=1) array([[4, 1, 7, 5, 2, 1]]) It landed 4 times on 1, once on 2, etc. Now, throw the dice 20 times, and 20 times again: >>> np.random.multinomial(20, [1/6.]*6, size=2) array([[3, 4, 3, 3, 4, 3], [2, 4, 3, 4, 0, 7]]) For the first run, we threw 3 times 1, 4 times 2, etc. For the second, we threw 2 times 1, 4 times 2, etc. A loaded die is more likely to land on number 6: >>> np.random.multinomial(100, [1/7.]*5 + [2/7.]) array([11, 16, 14, 17, 16, 26]) The probability inputs should be normalized. As an implementation detail, the value of the last entry is ignored and assumed to take up any leftover probability mass, but this should not be relied on. A biased coin which has twice as much weight on one side as on the other should be sampled like so: >>> np.random.multinomial(100, [1.0 / 3, 2.0 / 3]) # RIGHT array([38, 62]) not like: >>> np.random.multinomial(100, [1.0, 2.0]) # WRONG array([100, 0]) multinomialmumtrand.pyxmtrandmode > rightmodemean must be 1 dimensionalmean and cov must have same lengthmean <= 0.0mean <= 0meanmax__main__low is out of bounds for %slow >= highlowlong logseries(p, size=None) Draw samples from a logarithmic series distribution. Samples are drawn from a log series distribution with specified shape parameter, 0 < ``p`` < 1. Parameters ---------- p : float or array_like of floats Shape parameter for the distribution. Must be in the range (0, 1). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logarithmic series distribution. See Also -------- scipy.stats.logser : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Log Series distribution is .. math:: P(k) = \frac{-p^k}{k \ln(1-p)}, where p = probability. The log series distribution is frequently used to represent species richness and occurrence, first proposed by Fisher, Corbet, and Williams in 1943 [2]. It may also be used to model the numbers of occupants seen in cars [3]. References ---------- .. [1] Buzas, Martin A.; Culver, Stephen J., Understanding regional species diversity through the log series distribution of occurrences: BIODIVERSITY RESEARCH Diversity & Distributions, Volume 5, Number 5, September 1999 , pp. 187-195(9). .. [2] Fisher, R.A,, A.S. Corbet, and C.B. Williams. 1943. The relation between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology, 12:42-58. .. [3] D. J. Hand, F. Daly, D. Lunn, E. Ostrowski, A Handbook of Small Data Sets, CRC Press, 1994. .. [4] Wikipedia, "Logarithmic distribution", http://en.wikipedia.org/wiki/Logarithmic_distribution Examples -------- Draw samples from the distribution: >>> a = .6 >>> s = np.random.logseries(a, 10000) >>> count, bins, ignored = plt.hist(s) # plot against distribution >>> def logseries(k, p): ... return -p**k/(k*log(1-p)) >>> plt.plot(bins, logseries(bins, a)*count.max()/ logseries(bins, a).max(), 'r') >>> plt.show() logseries lognormal(mean=0.0, sigma=1.0, size=None) Draw samples from a log-normal distribution. Draw samples from a log-normal distribution with specified mean, standard deviation, and array shape. Note that the mean and standard deviation are not the values for the distribution itself, but of the underlying normal distribution it is derived from. Parameters ---------- mean : float or array_like of floats, optional Mean value of the underlying normal distribution. Default is 0. sigma : float or array_like of floats, optional Standard deviation of the underlying normal distribution. Should be greater than zero. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``sigma`` are both scalars. Otherwise, ``np.broadcast(mean, sigma).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized log-normal distribution. See Also -------- scipy.stats.lognorm : probability density function, distribution, cumulative density function, etc. Notes ----- A variable `x` has a log-normal distribution if `log(x)` is normally distributed. The probability density function for the log-normal distribution is: .. math:: p(x) = \frac{1}{\sigma x \sqrt{2\pi}} e^{(-\frac{(ln(x)-\mu)^2}{2\sigma^2})} where :math:`\mu` is the mean and :math:`\sigma` is the standard deviation of the normally distributed logarithm of the variable. A log-normal distribution results if a random variable is the *product* of a large number of independent, identically-distributed variables in the same way that a normal distribution results if the variable is the *sum* of a large number of independent, identically-distributed variables. References ---------- .. [1] Limpert, E., Stahel, W. A., and Abbt, M., "Log-normal Distributions across the Sciences: Keys and Clues," BioScience, Vol. 51, No. 5, May, 2001. http://stat.ethz.ch/~stahel/lognormal/bioscience.pdf .. [2] Reiss, R.D. and Thomas, M., "Statistical Analysis of Extreme Values," Basel: Birkhauser Verlag, 2001, pp. 31-32. Examples -------- Draw samples from the distribution: >>> mu, sigma = 3., 1. # mean and standard deviation >>> s = np.random.lognormal(mu, sigma, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 100, normed=True, align='mid') >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, linewidth=2, color='r') >>> plt.axis('tight') >>> plt.show() Demonstrate that taking the products of random samples from a uniform distribution can be fit well by a log-normal probability density function. >>> # Generate a thousand samples: each is the product of 100 random >>> # values, drawn from a normal distribution. >>> b = [] >>> for i in range(1000): ... a = 10. + np.random.random(100) ... b.append(np.product(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, normed=True, align='mid') >>> sigma = np.std(np.log(b)) >>> mu = np.mean(np.log(b)) >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, color='r', linewidth=2) >>> plt.show() lognormal logistic(loc=0.0, scale=1.0, size=None) Draw samples from a logistic distribution. Samples are drawn from a logistic distribution with specified parameters, loc (location or mean, also median), and scale (>0). Parameters ---------- loc : float or array_like of floats, optional Parameter of the distribution. Default is 0. scale : float or array_like of floats, optional Parameter of the distribution. Should be greater than zero. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logistic distribution. See Also -------- scipy.stats.logistic : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Logistic distribution is .. math:: P(x) = P(x) = \frac{e^{-(x-\mu)/s}}{s(1+e^{-(x-\mu)/s})^2}, where :math:`\mu` = location and :math:`s` = scale. The Logistic distribution is used in Extreme Value problems where it can act as a mixture of Gumbel distributions, in Epidemiology, and by the World Chess Federation (FIDE) where it is used in the Elo ranking system, assuming the performance of each player is a logistically distributed random variable. References ---------- .. [1] Reiss, R.-D. and Thomas M. (2001), "Statistical Analysis of Extreme Values, from Insurance, Finance, Hydrology and Other Fields," Birkhauser Verlag, Basel, pp 132-133. .. [2] Weisstein, Eric W. "Logistic Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LogisticDistribution.html .. [3] Wikipedia, "Logistic-distribution", http://en.wikipedia.org/wiki/Logistic_distribution Examples -------- Draw samples from the distribution: >>> loc, scale = 10, 1 >>> s = np.random.logistic(loc, scale, 10000) >>> count, bins, ignored = plt.hist(s, bins=50) # plot against distribution >>> def logist(x, loc, scale): ... return exp((loc-x)/scale)/(scale*(1+exp((loc-x)/scale))**2) >>> plt.plot(bins, logist(bins, loc, scale)*count.max()/\ ... logist(bins, loc, scale).max()) >>> plt.show() logisticlogical_orlocless_equallessleft == rightleft > modeleft laplace(loc=0.0, scale=1.0, size=None) Draw samples from the Laplace or double exponential distribution with specified location (or mean) and scale (decay). The Laplace distribution is similar to the Gaussian/normal distribution, but is sharper at the peak and has fatter tails. It represents the difference between two independent, identically distributed exponential random variables. Parameters ---------- loc : float or array_like of floats, optional The position, :math:`\mu`, of the distribution peak. Default is 0. scale : float or array_like of floats, optional :math:`\lambda`, the exponential decay. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Laplace distribution. Notes ----- It has the probability density function .. math:: f(x; \mu, \lambda) = \frac{1}{2\lambda} \exp\left(-\frac{|x - \mu|}{\lambda}\right). The first law of Laplace, from 1774, states that the frequency of an error can be expressed as an exponential function of the absolute magnitude of the error, which leads to the Laplace distribution. For many problems in economics and health sciences, this distribution seems to model the data better than the standard Gaussian distribution. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] Kotz, Samuel, et. al. "The Laplace Distribution and Generalizations, " Birkhauser, 2001. .. [3] Weisstein, Eric W. "Laplace Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LaplaceDistribution.html .. [4] Wikipedia, "Laplace distribution", http://en.wikipedia.org/wiki/Laplace_distribution Examples -------- Draw samples from the distribution >>> loc, scale = 0., 1. >>> s = np.random.laplace(loc, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, normed=True) >>> x = np.arange(-8., 8., .01) >>> pdf = np.exp(-abs(x-loc)/scale)/(2.*scale) >>> plt.plot(x, pdf) Plot Gaussian for comparison: >>> g = (1/(scale * np.sqrt(2 * np.pi)) * ... np.exp(-(x - loc)**2 / (2 * scale**2))) >>> plt.plot(x,g) laplacelam value too large.lam value too largelam < 0lamlkappa < 0kappaitemsizeitemissubdtypeisnanisfiniteintpintegerint8int64int32int16intindex__import__iinfoignore hypergeometric(ngood, nbad, nsample, size=None) Draw samples from a Hypergeometric distribution. Samples are drawn from a hypergeometric distribution with specified parameters, ngood (ways to make a good selection), nbad (ways to make a bad selection), and nsample = number of items sampled, which is less than or equal to the sum ngood + nbad. Parameters ---------- ngood : int or array_like of ints Number of ways to make a good selection. Must be nonnegative. nbad : int or array_like of ints Number of ways to make a bad selection. Must be nonnegative. nsample : int or array_like of ints Number of items sampled. Must be at least 1 and at most ``ngood + nbad``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``ngood``, ``nbad``, and ``nsample`` are all scalars. Otherwise, ``np.broadcast(ngood, nbad, nsample).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized hypergeometric distribution. See Also -------- scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{m}{n}\binom{N-m}{n-x}}{\binom{N}{n}}, where :math:`0 \le x \le m` and :math:`n+m-N \le x \le n` for P(x) the probability of x successes, n = ngood, m = nbad, and N = number of samples. Consider an urn with black and white marbles in it, ngood of them black and nbad are white. If you draw nsample balls without replacement, then the hypergeometric distribution describes the distribution of black balls in the drawn sample. Note that this distribution is very similar to the binomial distribution, except that in this case, samples are drawn without replacement, whereas in the Binomial case samples are drawn with replacement (or the sample space is infinite). As the sample space becomes large, this distribution approaches the binomial. References ---------- .. [1] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [2] Weisstein, Eric W. "Hypergeometric Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", http://en.wikipedia.org/wiki/Hypergeometric_distribution Examples -------- Draw samples from the distribution: >>> ngood, nbad, nsamp = 100, 2, 10 # number of good, number of bad, and number of samples >>> s = np.random.hypergeometric(ngood, nbad, nsamp, 1000) >>> hist(s) # note that it is very unlikely to grab both bad items Suppose you have an urn with 15 white and 15 black marbles. If you pull 15 marbles at random, how likely is it that 12 or more of them are one color? >>> s = np.random.hypergeometric(15, 15, 15, 100000) >>> sum(s>=12)/100000. + sum(s<=3)/100000. # answer = 0.003 ... pretty unlikely! hypergeometrichigh is out of bounds for %shigh gumbel(loc=0.0, scale=1.0, size=None) Draw samples from a Gumbel distribution. Draw samples from a Gumbel distribution with specified location and scale. For more information on the Gumbel distribution, see Notes and References below. Parameters ---------- loc : float or array_like of floats, optional The location of the mode of the distribution. Default is 0. scale : float or array_like of floats, optional The scale parameter of the distribution. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Gumbel distribution. See Also -------- scipy.stats.gumbel_l scipy.stats.gumbel_r scipy.stats.genextreme weibull Notes ----- The Gumbel (or Smallest Extreme Value (SEV) or the Smallest Extreme Value Type I) distribution is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. The Gumbel is a special case of the Extreme Value Type I distribution for maximums from distributions with "exponential-like" tails. The probability density for the Gumbel distribution is .. math:: p(x) = \frac{e^{-(x - \mu)/ \beta}}{\beta} e^{ -e^{-(x - \mu)/ \beta}}, where :math:`\mu` is the mode, a location parameter, and :math:`\beta` is the scale parameter. The Gumbel (named for German mathematician Emil Julius Gumbel) was used very early in the hydrology literature, for modeling the occurrence of flood events. It is also used for modeling maximum wind speed and rainfall rates. It is a "fat-tailed" distribution - the probability of an event in the tail of the distribution is larger than if one used a Gaussian, hence the surprisingly frequent occurrence of 100-year floods. Floods were initially modeled as a Gaussian process, which underestimated the frequency of extreme events. It is one of a class of extreme value distributions, the Generalized Extreme Value (GEV) distributions, which also includes the Weibull and Frechet. The function has a mean of :math:`\mu + 0.57721\beta` and a variance of :math:`\frac{\pi^2}{6}\beta^2`. References ---------- .. [1] Gumbel, E. J., "Statistics of Extremes," New York: Columbia University Press, 1958. .. [2] Reiss, R.-D. and Thomas, M., "Statistical Analysis of Extreme Values from Insurance, Finance, Hydrology and Other Fields," Basel: Birkhauser Verlag, 2001. Examples -------- Draw samples from the distribution: >>> mu, beta = 0, 0.1 # location and scale >>> s = np.random.gumbel(mu, beta, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, normed=True) >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp( -np.exp( -(bins - mu) /beta) ), ... linewidth=2, color='r') >>> plt.show() Show how an extreme value distribution can arise from a Gaussian process and compare to a Gaussian: >>> means = [] >>> maxima = [] >>> for i in range(0,1000) : ... a = np.random.normal(mu, beta, 1000) ... means.append(a.mean()) ... maxima.append(a.max()) >>> count, bins, ignored = plt.hist(maxima, 30, normed=True) >>> beta = np.std(maxima) * np.sqrt(6) / np.pi >>> mu = np.mean(maxima) - 0.57721*beta >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp(-np.exp(-(bins - mu)/beta)), ... linewidth=2, color='r') >>> plt.plot(bins, 1/(beta * np.sqrt(2 * np.pi)) ... * np.exp(-(bins - mu)**2 / (2 * beta**2)), ... linewidth=2, color='g') >>> plt.show() gumbelgreater_equalgreaterget_state geometric(p, size=None) Draw samples from the geometric distribution. Bernoulli trials are experiments with one of two outcomes: success or failure (an example of such an experiment is flipping a coin). The geometric distribution models the number of trials that must be run in order to achieve success. It is therefore supported on the positive integers, ``k = 1, 2, ...``. The probability mass function of the geometric distribution is .. math:: f(k) = (1 - p)^{k - 1} p where `p` is the probability of success of an individual trial. Parameters ---------- p : float or array_like of floats The probability of success of an individual trial. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized geometric distribution. Examples -------- Draw ten thousand values from the geometric distribution, with the probability of an individual success equal to 0.35: >>> z = np.random.geometric(p=0.35, size=10000) How many trials succeeded after a single run? >>> (z == 1).sum() / 10000. 0.34889999999999999 #random geometric gamma(shape, scale=1.0, size=None) Draw samples from a Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, `shape` (sometimes designated "k") and `scale` (sometimes designated "theta"), where both parameters are > 0. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. Should be greater than zero. scale : float or array_like of floats, optional The scale of the gamma distribution. Should be greater than zero. Default is equal to 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(shape, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", http://en.wikipedia.org/wiki/Gamma_distribution Examples -------- Draw samples from the distribution: >>> shape, scale = 2., 2. # mean=4, std=2*sqrt(2) >>> s = np.random.gamma(shape, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps >>> count, bins, ignored = plt.hist(s, 50, normed=True) >>> y = bins**(shape-1)*(np.exp(-bins/scale) / ... (sps.gamma(shape)*scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') >>> plt.show() gammaformatfloatingfloat64finfo f(dfnum, dfden, size=None) Draw samples from an F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters should be greater than zero. The random variate of the F distribution (also known as the Fisher distribution) is a continuous probability distribution that arises in ANOVA tests, and is the ratio of two chi-square variates. Parameters ---------- dfnum : int or array_like of ints Degrees of freedom in numerator. Should be greater than zero. dfden : int or array_like of ints Degrees of freedom in denominator. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum`` and ``dfden`` are both scalars. Otherwise, ``np.broadcast(dfnum, dfden).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Fisher distribution. See Also -------- scipy.stats.f : probability density function, distribution or cumulative density function, etc. Notes ----- The F statistic is used to compare in-group variances to between-group variances. Calculating the distribution depends on the sampling, and so it is a function of the respective degrees of freedom in the problem. The variable `dfnum` is the number of samples minus one, the between-groups degrees of freedom, while `dfden` is the within-groups degrees of freedom, the sum of the number of samples in each group minus the number of groups. References ---------- .. [1] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [2] Wikipedia, "F-distribution", http://en.wikipedia.org/wiki/F-distribution Examples -------- An example from Glantz[1], pp 47-40: Two groups, children of diabetics (25 people) and children from people without diabetes (25 controls). Fasting blood glucose was measured, case group had a mean value of 86.1, controls had a mean value of 82.2. Standard deviations were 2.09 and 2.49 respectively. Are these data consistent with the null hypothesis that the parents diabetic status does not affect their children's blood glucose levels? Calculating the F statistic from the data gives a value of 36.01. Draw samples from the distribution: >>> dfnum = 1. # between group degrees of freedom >>> dfden = 48. # within groups degrees of freedom >>> s = np.random.f(dfnum, dfden, 1000) The lower bound for the top 1% of the samples is : >>> sort(s)[-10] 7.61988120985 So there is about a 1% chance that the F statistic will exceed 7.62, the measured value is 36, so the null hypothesis is rejected at the 1% level. fexponential__exit__equaleps__enter__empty_likeemptydummy_threadingdtypedot dirichlet(alpha, size=None) Draw samples from the Dirichlet distribution. Draw `size` samples of dimension k from a Dirichlet distribution. A Dirichlet-distributed random variable can be seen as a multivariate generalization of a Beta distribution. Dirichlet pdf is the conjugate prior of a multinomial in Bayesian inference. Parameters ---------- alpha : array Parameter of the distribution (k dimension for sample of dimension k). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- samples : ndarray, The drawn samples, of shape (size, alpha.ndim). Notes ----- .. math:: X \approx \prod_{i=1}^{k}{x^{\alpha_i-1}_i} Uses the following property for computation: for each dimension, draw a random sample y_i from a standard gamma generator of shape `alpha_i`, then :math:`X = \frac{1}{\sum_{i=1}^k{y_i}} (y_1, \ldots, y_n)` is Dirichlet distributed. References ---------- .. [1] David McKay, "Information Theory, Inference and Learning Algorithms," chapter 23, http://www.inference.phy.cam.ac.uk/mackay/ .. [2] Wikipedia, "Dirichlet distribution", http://en.wikipedia.org/wiki/Dirichlet_distribution Examples -------- Taking an example cited in Wikipedia, this distribution can be used if one wanted to cut strings (each of initial length 1.0) into K pieces with different lengths, where each piece had, on average, a designated average length, but allowing some variation in the relative sizes of the pieces. >>> s = np.random.dirichlet((10, 5, 3), 20).transpose() >>> plt.barh(range(20), s[0]) >>> plt.barh(range(20), s[1], left=s[0], color='g') >>> plt.barh(range(20), s[2], left=s[0]+s[1], color='r') >>> plt.title("Lengths of Strings") dirichletdfnum <= 1dfnum <= 0dfnumdfden <= 0dfdendf <= 0dfdatadcumsumctypescovariance is not positive-semidefinite.cov must be 2 dimensional and squarecovcount_nonzerocopycntcline_in_traceback choice(a, size=None, replace=True, p=None) Generates a random sample from a given 1-D array .. versionadded:: 1.7.0 Parameters ----------- a : 1-D array-like or int If an ndarray, a random sample is generated from its elements. If an int, the random sample is generated as if a were np.arange(a) size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. replace : boolean, optional Whether the sample is with or without replacement p : 1-D array-like, optional The probabilities associated with each entry in a. If not given the sample assumes a uniform distribution over all entries in a. Returns -------- samples : single item or ndarray The generated random samples Raises ------- ValueError If a is an int and less than zero, if a or p are not 1-dimensional, if a is an array-like of size 0, if p is not a vector of probabilities, if a and p have different lengths, or if replace=False and the sample size is greater than the population size See Also --------- randint, shuffle, permutation Examples --------- Generate a uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3) array([0, 3, 4]) >>> #This is equivalent to np.random.randint(0,5,3) Generate a non-uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0]) array([3, 3, 0]) Generate a uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False) array([3,1,0]) >>> #This is equivalent to np.random.permutation(np.arange(5))[:3] Generate a non-uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0]) array([2, 3, 0]) Any of the above can be repeated with an arbitrary array-like instead of just integers. For instance: >>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher'] >>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3]) array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'], dtype='|S11') choice chisquare(df, size=None) Draw samples from a chi-square distribution. When `df` independent random variables, each with standard normal distributions (mean 0, variance 1), are squared and summed, the resulting distribution is chi-square (see Notes). This distribution is often used in hypothesis testing. Parameters ---------- df : int or array_like of ints Number of degrees of freedom. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized chi-square distribution. Raises ------ ValueError When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``) is given. Notes ----- The variable obtained by summing the squares of `df` independent, standard normally distributed random variables: .. math:: Q = \sum_{i=0}^{\mathtt{df}} X^2_i is chi-square distributed, denoted .. math:: Q \sim \chi^2_k. The probability density function of the chi-squared distribution is .. math:: p(x) = \frac{(1/2)^{k/2}}{\Gamma(k/2)} x^{k/2 - 1} e^{-x/2}, where :math:`\Gamma` is the gamma function, .. math:: \Gamma(x) = \int_0^{-\infty} t^{x - 1} e^{-t} dt. References ---------- .. [1] NIST "Engineering Statistics Handbook" http://www.itl.nist.gov/div898/handbook/eda/section3/eda3666.htm Examples -------- >>> np.random.chisquare(2,4) array([ 1.89920014, 9.00867716, 3.13710533, 5.62318272]) chisquarecheck_valid must equal 'warn', 'raise', or 'ignore'check_validcasting bytes(length) Return random bytes. Parameters ---------- length : int Number of random bytes. Returns ------- out : str String of length `length`. Examples -------- >>> np.random.bytes(10) ' eh\x85\x022SZ\xbf\xa4' #random bytesbufbroadcastboolbool_ binomial(n, p, size=None) Draw samples from a binomial distribution. Samples are drawn from a binomial distribution with specified parameters, n trials and p probability of success where n an integer >= 0 and p is in the interval [0,1]. (n may be input as a float, but it is truncated to an integer in use) Parameters ---------- n : int or array_like of ints Parameter of the distribution, >= 0. Floats are also accepted, but they will be truncated to integers. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized binomial distribution, where each sample is equal to the number of successes over the n trials. See Also -------- scipy.stats.binom : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the binomial distribution is .. math:: P(N) = \binom{n}{N}p^N(1-p)^{n-N}, where :math:`n` is the number of trials, :math:`p` is the probability of success, and :math:`N` is the number of successes. When estimating the standard error of a proportion in a population by using a random sample, the normal distribution works well unless the product p*n <=5, where p = population proportion estimate, and n = number of samples, in which case the binomial distribution is used instead. For example, a sample of 15 people shows 4 who are left handed, and 11 who are right handed. Then p = 4/15 = 27%. 0.27*15 = 4, so the binomial distribution should be used in this case. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics with R", Springer-Verlag, 2002. .. [2] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [3] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [4] Weisstein, Eric W. "Binomial Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/BinomialDistribution.html .. [5] Wikipedia, "Binomial distribution", http://en.wikipedia.org/wiki/Binomial_distribution Examples -------- Draw samples from the distribution: >>> n, p = 10, .5 # number of trials, probability of each trial >>> s = np.random.binomial(n, p, 1000) # result of flipping a coin 10 times, tested 1000 times. A real world example. A company drills 9 wild-cat oil exploration wells, each with an estimated probability of success of 0.1. All nine wells fail. What is the probability of that happening? Let's do 20,000 trials of the model, and count the number that generate zero positive results. >>> sum(np.random.binomial(9, 0.1, 20000) == 0)/20000. # answer = 0.38885, or 38%. binomialbetab <= 0batolastypeasarrayarray_dataarrayarangeanyalphaallcloseallalgorithm must be 'MT19937'adda must be non-emptya must be greater than 0a must be 1-dimensional or an integera must be 1-dimensionala and p must have same sizea <= 1.0a < 0a <= 0aValueErrorUnsupported dtype "%s" for randintTypeErrorThis function is deprecated. Please call randint({low}, {high} + 1) insteadThis function is deprecated. Please call randint(1, {low} + 1) insteadTSeed must be between 0 and 2**32 - 1RuntimeWarningRange exceeds valid boundsRandomState.zipf (line 3991)RandomState.weibull (line 2759)RandomState.wald (line 3505)RandomState.vonmises (line 2551)RandomState.uniform (line 1210)RandomState.triangular (line 3592)RandomState.tomaxint (line 858)RandomState.standard_t (line 2445)RandomState.standard_normal (line 1514)RandomState.standard_gamma (line 1810)RandomState.standard_exponential (line 1779)RandomState.standard_cauchy (line 2381)RandomState.shuffle (line 4759)RandomState.rayleigh (line 3426)RandomState.random_sample (line 814)RandomState.random_integers (line 1417)RandomState.randn (line 1360)RandomState.randint (line 905)RandomState.rand (line 1316)RandomState.power (line 2869)RandomState.poisson (line 3903)RandomState.permutation (line 4847)RandomState.pareto (line 2649)RandomState.normal (line 1547)RandomState.noncentral_f (line 2099)RandomState.noncentral_chisquare (line 2277)RandomState.negative_binomial (line 3802)RandomState.multivariate_normal (line 4369)RandomState.multinomial (line 4530)RandomState.logseries (line 4272)RandomState.lognormal (line 3302)RandomState.logistic (line 3209)RandomState.laplace (line 2980)RandomState.hypergeometric (line 4150)RandomState.gumbel (line 3078)RandomState.geometric (line 4082)RandomState.gamma (line 1896)RandomState.f (line 1992)RandomState.dirichlet (line 4643)__RandomState_ctorRandomState.choice (line 1028)RandomState.chisquare (line 2196)RandomState.bytes (line 999)RandomState.binomial (line 3686)OverflowErrorMT19937LockLImportErrorFewer non-zero entries in p than sizeDeprecationWarningCannot take a larger sample than population when 'replace=False':0yE>cannot import name %Snumpy/random/mtrand/mtrand.c%s (%s:%d)rb/dev/random/dev/urandommtrand.discnp_arraymtrand.discdd_arraymtrand.discnmN_arrayno errorrandom device unvavailable__int__ returned non-int (type %.200s). The ability to return an instance of a strict subclass of int is deprecated, and may be removed in a future version of Python.__%.4s__ returned non-%.4s (type %.200s)can't convert negative value to size_t%.200s.%.200s is not a type object%s.%s size changed, may indicate binary incompatibility. Expected %zd, got %zd%.200s.%.200s has the wrong size, try recompiling. Expected %zd, got %zdcalling %R should have returned an instance of BaseException, not %Rraise: exception class must be a subclass of BaseException'%.200s' object is not subscriptablecannot fit '%.200s' into an index-sized integer%.200s() keywords must be strings%s() got an unexpected keyword argument '%U'%s() got multiple values for keyword argument '%U'mmPm`mpm@@-DT!@(\@ffffff@.@4@x&?@?UUUUUU?a@X@`@|@@MA$@>@= ףp=@n?[ m?h|?5?333333 @r?$~?B>٬ @r鷯?Q?Q?9v?3?r?q?0@;`K@U0_$$(@T)~@9H@`p'<PR, g P}T P 0 @ P $ x + 5DU0ezHP4PP(pp/1 4PFM8Pl@T@`X`` ` 8@\x l<@p X0$>]x|8pph,00plp4@H p (!,!|!`H""#P### #`p''O(pO( P\(pQ(Q(@R(pR)`Sh)S)U)`V*Xd*Yx*PY*Z*\+\0+a+b,d$,`eH,ft,h,`i,j,pl0-mp-n-0p-q. sT.t.u.Pv.v.w.x /0x$/pxD/x\/xt/x/ y/py/y/ |0@|00`|H0|h0p}0 0 1`,1D1\1Px1P1`1p1` 2422`2@X3 333p(4t44`4585@55@56 `6Pt6@66Т6077,7x7P7p7<88p8990<:::@ 0; t; ;P;@,< |<zRx $PE FJ w?:*3$"DO \QFBB B(A0D8DVfM@ 8D0A(B BBBF _ W_AV`GyW`G\FBB B(A0D8DVfM@ 8D0A(B BBBF W W_ALV`GW`G QFBB B(A0D8DVfM@ 8D0A(B BBBF _ W_AV`GyW`G-FBB B(A0D8DVfM@ 8D0A(B BBBF g W`GJW_FW`GP4QFBB B(A0D8DVfM@ 8D0A(B BBBF _ W_AV`GyW`G@IFBB B(A0D8DgVfM@ 8D0A(B BBBF /W_AV`GIW`G^hFBB B(A0D8DVfM@ 8D0A(B BBBF w W_AV`GIW`GhrFBB B(A0D8DWVfM@ 8D0A(B BBBF W_AV`GW`G`FBB B(A0D8DVfM@ 8D0A(B BBBF  W_AsV`GIW`GLFBB E(D0A8D 8D0A(B BBBK LX FBB E(D0A8D 8D0A(B BBBK `4FBE B(D0A8DdOBBM8 8D0A(B BBBG L$FBB E(D0A8D 8D0A(B BBBC LFBB B(D0A8G 8D0A(B BBBA P8t<FBB E(D0A8D 8D0A(B BBBK l` FBB B(A0A8G 8A0A(B BBBD dWfMh W_AX FBB B(D0A8DpuxH`xMpB 8A0A(B BBBE LX'FBB B(D0A8GR 8D0A(B BBBI LGFBE B(D0A8G 8D0A(B BBBK `DVFBE E(A0A8G OBBM 8D0A(B BBBF \\ k98FBB B(A0A8GHEAw 8D0A(B BBBE L `FBB A(D0 (D BBBJ  (D BBBG 8 FBE A(DPs (A BBBD PH FBL H(A0M (A BBBG + (C BBBD L FBB E(A0A8D 8D0A(B BBBK L h!FBB E(D0A8G 8D0A(B BBBG D< nFBD A(G@tHHP`HM@@ (D ABBJ \ H8FBB B(A0A8J WRMP 8D0A(B BBBE L BBE E(A0A8Gz 8D0A(B BBBB d4 XiFBB D(A0G`v 0D(A BBBI XhWpRhM`} 0C(A BBBK \ `"XFBB B(D0A8DHgPy 8D0A(B BBBB L `31FBB B(A0A8GPi 8D0A(B BBBE \L P:FBB B(A0A8D 8D0A(B BBBG VYMl X.$FBB B(A0A8G)WYM@ 8D0A(B BBBH V`GHp|FBB B(D0D8DP' 8D0A(B BBBD @hFBB B(D0G` 0D(B BBBI \FBB B(A0D8DOYM@ 8D0A(B BBBG \ `FBB B(A0A8D0 8D0A(B BBBA VYMDlnFBD A(G@tHHP`HM@@ (D ABBJ \FBB B(A0A8DzLgM@ 8D0A(B BBBG \XFBB B(A0D8G-I`M@ 8D0A(B BBBH DtnFBD A(G@tHHP`HM@@ (D ABBJ Xw.FBB B(A0A8GOYM 8D0A(B BBBB p!.$FBB B(A0A8G)WYM@ 8D0A(B BBBH V`GDs2FBB B(A0A8G ^KAErM@ 8D0A(B BBBJ pW`GW_F\ vFBB B(A0D8G-I`M@ 8D0A(B BBBH \<FBB B(A0D8G-I`M@ 8D0A(B BBBH l̝.$FBB B(A0A8G)WYM@ 8D0A(B BBBH V`GHPnFBD A(G@tHHP`HM@@ (D ABBJ l:FBB B(A0A8GWYM@ 8D0A(B BBBH  W_F\ FBB B(A0D8G-I`M@ 8D0A(B BBBH \lFBB B(A0D8G-I`M@ 8D0A(B BBBH \p FBB B(A0D8G-I`M@ 8D0A(B BBBH \,#FBB B(A0A8Dw 8D0A(B BBBJ VYM\PBFBB B(A0A8Dw 8D0A(B BBBJ VYM\`FBB B(A0A8Dw 8D0A(B BBBJ VYM\LPFBB B(A0A8Dw 8D0A(B BBBJ VYM\НFBB B(A0A8DzLgM@ 8D0A(B BBBG l p &FBB B(A0A8GIWYM@ 8D0A(B BBBH V`G|c2FBB B(A0A8G ^KAErM@ 8D0A(B BBBB W`GW_Fp=FBE A(A0GPtXW`YXMP} 0C(A BBBB d 0D(A BBBI XW`_XAPL<BEB B(A0A8D 8D0A(B BBBA pN=FBE A(A0GPtXW`YXMP} 0C(A BBBB d 0D(A BBBI XW`_XAPPHP.FFBE B(A0A8D- 8D0A(B BBBA \!FBB B(A0A8DzLgM@ 8D0A(B BBBG \FBB B(A0D8G=I`M@ 8D0A(B BBBH \\FBB B(A0D8D0I`M@ 8D0A(B BBBH |FBE A(A0G`thEprhM`} 0C(A BBBK h 0D(A BBBM hWp_hA`NhVp`hG`\<FBB B(A0D8D0I`M@ 8D0A(B BBBH L 7PFBB B(A0D8Da 8D0A(B BBBE l[jSFBB B(A0D8J WYMF 8D0A(B BBBI -W_Fl\FBB B(A0D8JuWYMC 8D0A(B BBBC -W_A\/FBB B(D0A8J IgPJ 8A0A(B BBBG 4,EFDD a GBH AAB4dyEAG I CAF K CAA `Hx H L49FEE B(A0A8D 8D0A(B BBBG ,$@EG0 AI Y AA H8:҅FBB B(A0A8G!TTMMFFFIbHBFFAAAAANKTTMMFFFIXQBFFAAAAANJTTMMFFFIXQBFFAAAAANJTTMMFFFIbHEFFAAAAANKTTMMFFFIbHEFFAAAAANJTTMMFFFIaHEFFAAAAANKTTMMFFFIbHEFFAAAAANKTTMMFFFIbHEFFAAAAANJTTMMFFFIXQEFFAAAAANBQEFFAAAAAN1]8D0A(B BBBD!FFB B(A0A8DP8A0A(B BBBL!P4FBB B(A0D8G 8D0A(B BBBG "&p<0"L'BEB D(D0k (D BBBJ (p"'KAED@( EAH "(%$">^FED JBB("(FHD ` EBG #H)0L#d)FBE D(D0\ (D BBBE Y (A BBBH h#>4|#)FAD E ABE O ABF 8#H*FED D(GP (A ABBJ 4#+FAD  ABC O ABF L($D,dFEA A(D@< (A ABBE u (A ABBE x$d.5$.50$.6FAA D0T  AABD @$/~FAA D0\  AABD   FABA (%1xFED ^ EBD D%X1^FBE E(D0A8F` 8A0A(B BBBE O 8A0A(B BBBG  8A0A(B BBBF OhTpAxBBAAAAAI`$%6E@ K h H d &6UED x AF 8&7UED x AF (\&9MED z AD @ AG &4:RED x AF (&p;JED z AD @ AG (&>> np.random.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) >>> np.random.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.permutation(arr) array([[6, 7, 8], [0, 1, 2], [3, 4, 5]]) shuffle(x) Modify a sequence in-place by shuffling its contents. This function only shuffles the array along the first axis of a multi-dimensional array. The order of sub-arrays is changed but their contents remains the same. Parameters ---------- x : array_like The array or list to be shuffled. Returns ------- None Examples -------- >>> arr = np.arange(10) >>> np.random.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] Multi-dimensional arrays are only shuffled along the first axis: >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.shuffle(arr) >>> arr array([[3, 4, 5], [6, 7, 8], [0, 1, 2]]) dirichlet(alpha, size=None) Draw samples from the Dirichlet distribution. Draw `size` samples of dimension k from a Dirichlet distribution. A Dirichlet-distributed random variable can be seen as a multivariate generalization of a Beta distribution. Dirichlet pdf is the conjugate prior of a multinomial in Bayesian inference. Parameters ---------- alpha : array Parameter of the distribution (k dimension for sample of dimension k). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- samples : ndarray, The drawn samples, of shape (size, alpha.ndim). Notes ----- .. math:: X \approx \prod_{i=1}^{k}{x^{\alpha_i-1}_i} Uses the following property for computation: for each dimension, draw a random sample y_i from a standard gamma generator of shape `alpha_i`, then :math:`X = \frac{1}{\sum_{i=1}^k{y_i}} (y_1, \ldots, y_n)` is Dirichlet distributed. References ---------- .. [1] David McKay, "Information Theory, Inference and Learning Algorithms," chapter 23, http://www.inference.phy.cam.ac.uk/mackay/ .. [2] Wikipedia, "Dirichlet distribution", http://en.wikipedia.org/wiki/Dirichlet_distribution Examples -------- Taking an example cited in Wikipedia, this distribution can be used if one wanted to cut strings (each of initial length 1.0) into K pieces with different lengths, where each piece had, on average, a designated average length, but allowing some variation in the relative sizes of the pieces. >>> s = np.random.dirichlet((10, 5, 3), 20).transpose() >>> plt.barh(range(20), s[0]) >>> plt.barh(range(20), s[1], left=s[0], color='g') >>> plt.barh(range(20), s[2], left=s[0]+s[1], color='r') >>> plt.title("Lengths of Strings") multinomial(n, pvals, size=None) Draw samples from a multinomial distribution. The multinomial distribution is a multivariate generalisation of the binomial distribution. Take an experiment with one of ``p`` possible outcomes. An example of such an experiment is throwing a dice, where the outcome can be 1 through 6. Each sample drawn from the distribution represents `n` such experiments. Its values, ``X_i = [X_0, X_1, ..., X_p]``, represent the number of times the outcome was ``i``. Parameters ---------- n : int Number of experiments. pvals : sequence of floats, length p Probabilities of each of the ``p`` different outcomes. These should sum to 1 (however, the last element is always assumed to account for the remaining probability, as long as ``sum(pvals[:-1]) <= 1)``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. Examples -------- Throw a dice 20 times: >>> np.random.multinomial(20, [1/6.]*6, size=1) array([[4, 1, 7, 5, 2, 1]]) It landed 4 times on 1, once on 2, etc. Now, throw the dice 20 times, and 20 times again: >>> np.random.multinomial(20, [1/6.]*6, size=2) array([[3, 4, 3, 3, 4, 3], [2, 4, 3, 4, 0, 7]]) For the first run, we threw 3 times 1, 4 times 2, etc. For the second, we threw 2 times 1, 4 times 2, etc. A loaded die is more likely to land on number 6: >>> np.random.multinomial(100, [1/7.]*5 + [2/7.]) array([11, 16, 14, 17, 16, 26]) The probability inputs should be normalized. As an implementation detail, the value of the last entry is ignored and assumed to take up any leftover probability mass, but this should not be relied on. A biased coin which has twice as much weight on one side as on the other should be sampled like so: >>> np.random.multinomial(100, [1.0 / 3, 2.0 / 3]) # RIGHT array([38, 62]) not like: >>> np.random.multinomial(100, [1.0, 2.0]) # WRONG array([100, 0]) multivariate_normal(mean, cov[, size, check_valid, tol]) Draw random samples from a multivariate normal distribution. The multivariate normal, multinormal or Gaussian distribution is a generalization of the one-dimensional normal distribution to higher dimensions. Such a distribution is specified by its mean and covariance matrix. These parameters are analogous to the mean (average or "center") and variance (standard deviation, or "width," squared) of the one-dimensional normal distribution. Parameters ---------- mean : 1-D array_like, of length N Mean of the N-dimensional distribution. cov : 2-D array_like, of shape (N, N) Covariance matrix of the distribution. It must be symmetric and positive-semidefinite for proper sampling. size : int or tuple of ints, optional Given a shape of, for example, ``(m,n,k)``, ``m*n*k`` samples are generated, and packed in an `m`-by-`n`-by-`k` arrangement. Because each sample is `N`-dimensional, the output shape is ``(m,n,k,N)``. If no shape is specified, a single (`N`-D) sample is returned. check_valid : { 'warn', 'raise', 'ignore' }, optional Behavior when the covariance matrix is not positive semidefinite. tol : float, optional Tolerance when checking the singular values in covariance matrix. Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. Notes ----- The mean is a coordinate in N-dimensional space, which represents the location where samples are most likely to be generated. This is analogous to the peak of the bell curve for the one-dimensional or univariate normal distribution. Covariance indicates the level to which two variables vary together. From the multivariate normal distribution, we draw N-dimensional samples, :math:`X = [x_1, x_2, ... x_N]`. The covariance matrix element :math:`C_{ij}` is the covariance of :math:`x_i` and :math:`x_j`. The element :math:`C_{ii}` is the variance of :math:`x_i` (i.e. its "spread"). Instead of specifying the full covariance matrix, popular approximations include: - Spherical covariance (`cov` is a multiple of the identity matrix) - Diagonal covariance (`cov` has non-negative elements, and only on the diagonal) This geometrical property can be seen in two dimensions by plotting generated data-points: >>> mean = [0, 0] >>> cov = [[1, 0], [0, 100]] # diagonal covariance Diagonal covariance means that points are oriented along x or y-axis: >>> import matplotlib.pyplot as plt >>> x, y = np.random.multivariate_normal(mean, cov, 5000).T >>> plt.plot(x, y, 'x') >>> plt.axis('equal') >>> plt.show() Note that the covariance matrix must be positive semidefinite (a.k.a. nonnegative-definite). Otherwise, the behavior of this method is undefined and backwards compatibility is not guaranteed. References ---------- .. [1] Papoulis, A., "Probability, Random Variables, and Stochastic Processes," 3rd ed., New York: McGraw-Hill, 1991. .. [2] Duda, R. O., Hart, P. E., and Stork, D. G., "Pattern Classification," 2nd ed., New York: Wiley, 2001. Examples -------- >>> mean = (1, 2) >>> cov = [[1, 0], [0, 1]] >>> x = np.random.multivariate_normal(mean, cov, (3, 3)) >>> x.shape (3, 3, 2) The following is probably true, given that 0.6 is roughly twice the standard deviation: >>> list((x[0,0,:] - mean) < 0.6) [True, True] logseries(p, size=None) Draw samples from a logarithmic series distribution. Samples are drawn from a log series distribution with specified shape parameter, 0 < ``p`` < 1. Parameters ---------- p : float or array_like of floats Shape parameter for the distribution. Must be in the range (0, 1). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logarithmic series distribution. See Also -------- scipy.stats.logser : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Log Series distribution is .. math:: P(k) = \frac{-p^k}{k \ln(1-p)}, where p = probability. The log series distribution is frequently used to represent species richness and occurrence, first proposed by Fisher, Corbet, and Williams in 1943 [2]. It may also be used to model the numbers of occupants seen in cars [3]. References ---------- .. [1] Buzas, Martin A.; Culver, Stephen J., Understanding regional species diversity through the log series distribution of occurrences: BIODIVERSITY RESEARCH Diversity & Distributions, Volume 5, Number 5, September 1999 , pp. 187-195(9). .. [2] Fisher, R.A,, A.S. Corbet, and C.B. Williams. 1943. The relation between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology, 12:42-58. .. [3] D. J. Hand, F. Daly, D. Lunn, E. Ostrowski, A Handbook of Small Data Sets, CRC Press, 1994. .. [4] Wikipedia, "Logarithmic distribution", http://en.wikipedia.org/wiki/Logarithmic_distribution Examples -------- Draw samples from the distribution: >>> a = .6 >>> s = np.random.logseries(a, 10000) >>> count, bins, ignored = plt.hist(s) # plot against distribution >>> def logseries(k, p): ... return -p**k/(k*log(1-p)) >>> plt.plot(bins, logseries(bins, a)*count.max()/ logseries(bins, a).max(), 'r') >>> plt.show() hypergeometric(ngood, nbad, nsample, size=None) Draw samples from a Hypergeometric distribution. Samples are drawn from a hypergeometric distribution with specified parameters, ngood (ways to make a good selection), nbad (ways to make a bad selection), and nsample = number of items sampled, which is less than or equal to the sum ngood + nbad. Parameters ---------- ngood : int or array_like of ints Number of ways to make a good selection. Must be nonnegative. nbad : int or array_like of ints Number of ways to make a bad selection. Must be nonnegative. nsample : int or array_like of ints Number of items sampled. Must be at least 1 and at most ``ngood + nbad``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``ngood``, ``nbad``, and ``nsample`` are all scalars. Otherwise, ``np.broadcast(ngood, nbad, nsample).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized hypergeometric distribution. See Also -------- scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{m}{n}\binom{N-m}{n-x}}{\binom{N}{n}}, where :math:`0 \le x \le m` and :math:`n+m-N \le x \le n` for P(x) the probability of x successes, n = ngood, m = nbad, and N = number of samples. Consider an urn with black and white marbles in it, ngood of them black and nbad are white. If you draw nsample balls without replacement, then the hypergeometric distribution describes the distribution of black balls in the drawn sample. Note that this distribution is very similar to the binomial distribution, except that in this case, samples are drawn without replacement, whereas in the Binomial case samples are drawn with replacement (or the sample space is infinite). As the sample space becomes large, this distribution approaches the binomial. References ---------- .. [1] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [2] Weisstein, Eric W. "Hypergeometric Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", http://en.wikipedia.org/wiki/Hypergeometric_distribution Examples -------- Draw samples from the distribution: >>> ngood, nbad, nsamp = 100, 2, 10 # number of good, number of bad, and number of samples >>> s = np.random.hypergeometric(ngood, nbad, nsamp, 1000) >>> hist(s) # note that it is very unlikely to grab both bad items Suppose you have an urn with 15 white and 15 black marbles. If you pull 15 marbles at random, how likely is it that 12 or more of them are one color? >>> s = np.random.hypergeometric(15, 15, 15, 100000) >>> sum(s>=12)/100000. + sum(s<=3)/100000. # answer = 0.003 ... pretty unlikely! geometric(p, size=None) Draw samples from the geometric distribution. Bernoulli trials are experiments with one of two outcomes: success or failure (an example of such an experiment is flipping a coin). The geometric distribution models the number of trials that must be run in order to achieve success. It is therefore supported on the positive integers, ``k = 1, 2, ...``. The probability mass function of the geometric distribution is .. math:: f(k) = (1 - p)^{k - 1} p where `p` is the probability of success of an individual trial. Parameters ---------- p : float or array_like of floats The probability of success of an individual trial. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized geometric distribution. Examples -------- Draw ten thousand values from the geometric distribution, with the probability of an individual success equal to 0.35: >>> z = np.random.geometric(p=0.35, size=10000) How many trials succeeded after a single run? >>> (z == 1).sum() / 10000. 0.34889999999999999 #random zipf(a, size=None) Draw samples from a Zipf distribution. Samples are drawn from a Zipf distribution with specified parameter `a` > 1. The Zipf distribution (also known as the zeta distribution) is a continuous probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. Parameters ---------- a : float or array_like of floats Distribution parameter. Should be greater than 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Zipf distribution. See Also -------- scipy.stats.zipf : probability density function, distribution, or cumulative density function, etc. Notes ----- The probability density for the Zipf distribution is .. math:: p(x) = \frac{x^{-a}}{\zeta(a)}, where :math:`\zeta` is the Riemann Zeta function. It is named for the American linguist George Kingsley Zipf, who noted that the frequency of any word in a sample of a language is inversely proportional to its rank in the frequency table. References ---------- .. [1] Zipf, G. K., "Selected Studies of the Principle of Relative Frequency in Language," Cambridge, MA: Harvard Univ. Press, 1932. Examples -------- Draw samples from the distribution: >>> a = 2. # parameter >>> s = np.random.zipf(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy import special Truncate s values at 50 so plot is interesting: >>> count, bins, ignored = plt.hist(s[s<50], 50, normed=True) >>> x = np.arange(1., 50.) >>> y = x**(-a) / special.zetac(a) >>> plt.plot(x, y/max(y), linewidth=2, color='r') >>> plt.show() poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. Parameters ---------- lam : float or array_like of floats Expectation of interval, should be >= 0. A sequence of expectation intervals must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. Notes ----- The Poisson distribution .. math:: f(k; \lambda)=\frac{\lambda^k e^{-\lambda}}{k!} For events with an expected separation :math:`\lambda` the Poisson distribution :math:`f(k; \lambda)` describes the probability of :math:`k` events occurring within the observed interval :math:`\lambda`. Because the output is limited to the range of the C long type, a ValueError is raised when `lam` is within 10 sigma of the maximum representable value. References ---------- .. [1] Weisstein, Eric W. "Poisson Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/PoissonDistribution.html .. [2] Wikipedia, "Poisson distribution", http://en.wikipedia.org/wiki/Poisson_distribution Examples -------- Draw samples from the distribution: >>> import numpy as np >>> s = np.random.poisson(5, 10000) Display histogram of the sample: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 14, normed=True) >>> plt.show() Draw each 100 values for lambda 100 and 500: >>> s = np.random.poisson(lam=(100., 500.), size=(100, 2)) negative_binomial(n, p, size=None) Draw samples from a negative binomial distribution. Samples are drawn from a negative binomial distribution with specified parameters, `n` trials and `p` probability of success where `n` is an integer > 0 and `p` is in the interval [0, 1]. Parameters ---------- n : int or array_like of ints Parameter of the distribution, > 0. Floats are also accepted, but they will be truncated to integers. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized negative binomial distribution, where each sample is equal to N, the number of trials it took to achieve n - 1 successes, N - (n - 1) failures, and a success on the, (N + n)th trial. Notes ----- The probability density for the negative binomial distribution is .. math:: P(N;n,p) = \binom{N+n-1}{n-1}p^{n}(1-p)^{N}, where :math:`n-1` is the number of successes, :math:`p` is the probability of success, and :math:`N+n-1` is the number of trials. The negative binomial distribution gives the probability of n-1 successes and N failures in N+n-1 trials, and success on the (N+n)th trial. If one throws a die repeatedly until the third time a "1" appears, then the probability distribution of the number of non-"1"s that appear before the third "1" is a negative binomial distribution. References ---------- .. [1] Weisstein, Eric W. "Negative Binomial Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/NegativeBinomialDistribution.html .. [2] Wikipedia, "Negative binomial distribution", http://en.wikipedia.org/wiki/Negative_binomial_distribution Examples -------- Draw samples from the distribution: A real world example. A company drills wild-cat oil exploration wells, each with an estimated probability of success of 0.1. What is the probability of having one success for each successive well, that is what is the probability of a single success after drilling 5 wells, after 6 wells, etc.? >>> s = np.random.negative_binomial(1, 0.1, 100000) >>> for i in range(1, 11): ... probability = sum(s>> n, p = 10, .5 # number of trials, probability of each trial >>> s = np.random.binomial(n, p, 1000) # result of flipping a coin 10 times, tested 1000 times. A real world example. A company drills 9 wild-cat oil exploration wells, each with an estimated probability of success of 0.1. All nine wells fail. What is the probability of that happening? Let's do 20,000 trials of the model, and count the number that generate zero positive results. >>> sum(np.random.binomial(9, 0.1, 20000) == 0)/20000. # answer = 0.38885, or 38%. triangular(left, mode, right, size=None) Draw samples from the triangular distribution over the interval ``[left, right]``. The triangular distribution is a continuous probability distribution with lower limit left, peak at mode, and upper limit right. Unlike the other distributions, these parameters directly define the shape of the pdf. Parameters ---------- left : float or array_like of floats Lower limit. mode : float or array_like of floats The value where the peak of the distribution occurs. The value should fulfill the condition ``left <= mode <= right``. right : float or array_like of floats Upper limit, should be larger than `left`. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``left``, ``mode``, and ``right`` are all scalars. Otherwise, ``np.broadcast(left, mode, right).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized triangular distribution. Notes ----- The probability density function for the triangular distribution is .. math:: P(x;l, m, r) = \begin{cases} \frac{2(x-l)}{(r-l)(m-l)}& \text{for $l \leq x \leq m$},\\ \frac{2(r-x)}{(r-l)(r-m)}& \text{for $m \leq x \leq r$},\\ 0& \text{otherwise}. \end{cases} The triangular distribution is often used in ill-defined problems where the underlying distribution is not known, but some knowledge of the limits and mode exists. Often it is used in simulations. References ---------- .. [1] Wikipedia, "Triangular distribution" http://en.wikipedia.org/wiki/Triangular_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.triangular(-3, 0, 8, 100000), bins=200, ... normed=True) >>> plt.show() wald(mean, scale, size=None) Draw samples from a Wald, or inverse Gaussian, distribution. As the scale approaches infinity, the distribution becomes more like a Gaussian. Some references claim that the Wald is an inverse Gaussian with mean equal to 1, but this is by no means universal. The inverse Gaussian distribution was first studied in relationship to Brownian motion. In 1956 M.C.K. Tweedie used the name inverse Gaussian because there is an inverse relationship between the time to cover a unit distance and distance covered in unit time. Parameters ---------- mean : float or array_like of floats Distribution mean, should be > 0. scale : float or array_like of floats Scale parameter, should be >= 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(mean, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Wald distribution. Notes ----- The probability density function for the Wald distribution is .. math:: P(x;mean,scale) = \sqrt{\frac{scale}{2\pi x^3}}e^ \frac{-scale(x-mean)^2}{2\cdotp mean^2x} As noted above the inverse Gaussian distribution first arise from attempts to model Brownian motion. It is also a competitor to the Weibull for use in reliability modeling and modeling stock returns and interest rate processes. References ---------- .. [1] Brighton Webs Ltd., Wald Distribution, http://www.brighton-webs.co.uk/distributions/wald.asp .. [2] Chhikara, Raj S., and Folks, J. Leroy, "The Inverse Gaussian Distribution: Theory : Methodology, and Applications", CRC Press, 1988. .. [3] Wikipedia, "Wald distribution" http://en.wikipedia.org/wiki/Wald_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.wald(3, 2, 100000), bins=200, normed=True) >>> plt.show() rayleigh(scale=1.0, size=None) Draw samples from a Rayleigh distribution. The :math:`\chi` and Weibull distributions are generalizations of the Rayleigh. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. Should be >= 0. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Rayleigh distribution. Notes ----- The probability density function for the Rayleigh distribution is .. math:: P(x;scale) = \frac{x}{scale^2}e^{\frac{-x^2}{2 \cdotp scale^2}} The Rayleigh distribution would arise, for example, if the East and North components of the wind velocity had identical zero-mean Gaussian distributions. Then the wind speed would have a Rayleigh distribution. References ---------- .. [1] Brighton Webs Ltd., "Rayleigh Distribution," http://www.brighton-webs.co.uk/distributions/rayleigh.asp .. [2] Wikipedia, "Rayleigh distribution" http://en.wikipedia.org/wiki/Rayleigh_distribution Examples -------- Draw values from the distribution and plot the histogram >>> values = hist(np.random.rayleigh(3, 100000), bins=200, normed=True) Wave heights tend to follow a Rayleigh distribution. If the mean wave height is 1 meter, what fraction of waves are likely to be larger than 3 meters? >>> meanvalue = 1 >>> modevalue = np.sqrt(2 / np.pi) * meanvalue >>> s = np.random.rayleigh(modevalue, 1000000) The percentage of waves larger than 3 meters is: >>> 100.*sum(s>3)/1000000. 0.087300000000000003 lognormal(mean=0.0, sigma=1.0, size=None) Draw samples from a log-normal distribution. Draw samples from a log-normal distribution with specified mean, standard deviation, and array shape. Note that the mean and standard deviation are not the values for the distribution itself, but of the underlying normal distribution it is derived from. Parameters ---------- mean : float or array_like of floats, optional Mean value of the underlying normal distribution. Default is 0. sigma : float or array_like of floats, optional Standard deviation of the underlying normal distribution. Should be greater than zero. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``sigma`` are both scalars. Otherwise, ``np.broadcast(mean, sigma).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized log-normal distribution. See Also -------- scipy.stats.lognorm : probability density function, distribution, cumulative density function, etc. Notes ----- A variable `x` has a log-normal distribution if `log(x)` is normally distributed. The probability density function for the log-normal distribution is: .. math:: p(x) = \frac{1}{\sigma x \sqrt{2\pi}} e^{(-\frac{(ln(x)-\mu)^2}{2\sigma^2})} where :math:`\mu` is the mean and :math:`\sigma` is the standard deviation of the normally distributed logarithm of the variable. A log-normal distribution results if a random variable is the *product* of a large number of independent, identically-distributed variables in the same way that a normal distribution results if the variable is the *sum* of a large number of independent, identically-distributed variables. References ---------- .. [1] Limpert, E., Stahel, W. A., and Abbt, M., "Log-normal Distributions across the Sciences: Keys and Clues," BioScience, Vol. 51, No. 5, May, 2001. http://stat.ethz.ch/~stahel/lognormal/bioscience.pdf .. [2] Reiss, R.D. and Thomas, M., "Statistical Analysis of Extreme Values," Basel: Birkhauser Verlag, 2001, pp. 31-32. Examples -------- Draw samples from the distribution: >>> mu, sigma = 3., 1. # mean and standard deviation >>> s = np.random.lognormal(mu, sigma, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 100, normed=True, align='mid') >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, linewidth=2, color='r') >>> plt.axis('tight') >>> plt.show() Demonstrate that taking the products of random samples from a uniform distribution can be fit well by a log-normal probability density function. >>> # Generate a thousand samples: each is the product of 100 random >>> # values, drawn from a normal distribution. >>> b = [] >>> for i in range(1000): ... a = 10. + np.random.random(100) ... b.append(np.product(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, normed=True, align='mid') >>> sigma = np.std(np.log(b)) >>> mu = np.mean(np.log(b)) >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, color='r', linewidth=2) >>> plt.show() logistic(loc=0.0, scale=1.0, size=None) Draw samples from a logistic distribution. Samples are drawn from a logistic distribution with specified parameters, loc (location or mean, also median), and scale (>0). Parameters ---------- loc : float or array_like of floats, optional Parameter of the distribution. Default is 0. scale : float or array_like of floats, optional Parameter of the distribution. Should be greater than zero. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logistic distribution. See Also -------- scipy.stats.logistic : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Logistic distribution is .. math:: P(x) = P(x) = \frac{e^{-(x-\mu)/s}}{s(1+e^{-(x-\mu)/s})^2}, where :math:`\mu` = location and :math:`s` = scale. The Logistic distribution is used in Extreme Value problems where it can act as a mixture of Gumbel distributions, in Epidemiology, and by the World Chess Federation (FIDE) where it is used in the Elo ranking system, assuming the performance of each player is a logistically distributed random variable. References ---------- .. [1] Reiss, R.-D. and Thomas M. (2001), "Statistical Analysis of Extreme Values, from Insurance, Finance, Hydrology and Other Fields," Birkhauser Verlag, Basel, pp 132-133. .. [2] Weisstein, Eric W. "Logistic Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LogisticDistribution.html .. [3] Wikipedia, "Logistic-distribution", http://en.wikipedia.org/wiki/Logistic_distribution Examples -------- Draw samples from the distribution: >>> loc, scale = 10, 1 >>> s = np.random.logistic(loc, scale, 10000) >>> count, bins, ignored = plt.hist(s, bins=50) # plot against distribution >>> def logist(x, loc, scale): ... return exp((loc-x)/scale)/(scale*(1+exp((loc-x)/scale))**2) >>> plt.plot(bins, logist(bins, loc, scale)*count.max()/\ ... logist(bins, loc, scale).max()) >>> plt.show() gumbel(loc=0.0, scale=1.0, size=None) Draw samples from a Gumbel distribution. Draw samples from a Gumbel distribution with specified location and scale. For more information on the Gumbel distribution, see Notes and References below. Parameters ---------- loc : float or array_like of floats, optional The location of the mode of the distribution. Default is 0. scale : float or array_like of floats, optional The scale parameter of the distribution. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Gumbel distribution. See Also -------- scipy.stats.gumbel_l scipy.stats.gumbel_r scipy.stats.genextreme weibull Notes ----- The Gumbel (or Smallest Extreme Value (SEV) or the Smallest Extreme Value Type I) distribution is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. The Gumbel is a special case of the Extreme Value Type I distribution for maximums from distributions with "exponential-like" tails. The probability density for the Gumbel distribution is .. math:: p(x) = \frac{e^{-(x - \mu)/ \beta}}{\beta} e^{ -e^{-(x - \mu)/ \beta}}, where :math:`\mu` is the mode, a location parameter, and :math:`\beta` is the scale parameter. The Gumbel (named for German mathematician Emil Julius Gumbel) was used very early in the hydrology literature, for modeling the occurrence of flood events. It is also used for modeling maximum wind speed and rainfall rates. It is a "fat-tailed" distribution - the probability of an event in the tail of the distribution is larger than if one used a Gaussian, hence the surprisingly frequent occurrence of 100-year floods. Floods were initially modeled as a Gaussian process, which underestimated the frequency of extreme events. It is one of a class of extreme value distributions, the Generalized Extreme Value (GEV) distributions, which also includes the Weibull and Frechet. The function has a mean of :math:`\mu + 0.57721\beta` and a variance of :math:`\frac{\pi^2}{6}\beta^2`. References ---------- .. [1] Gumbel, E. J., "Statistics of Extremes," New York: Columbia University Press, 1958. .. [2] Reiss, R.-D. and Thomas, M., "Statistical Analysis of Extreme Values from Insurance, Finance, Hydrology and Other Fields," Basel: Birkhauser Verlag, 2001. Examples -------- Draw samples from the distribution: >>> mu, beta = 0, 0.1 # location and scale >>> s = np.random.gumbel(mu, beta, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, normed=True) >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp( -np.exp( -(bins - mu) /beta) ), ... linewidth=2, color='r') >>> plt.show() Show how an extreme value distribution can arise from a Gaussian process and compare to a Gaussian: >>> means = [] >>> maxima = [] >>> for i in range(0,1000) : ... a = np.random.normal(mu, beta, 1000) ... means.append(a.mean()) ... maxima.append(a.max()) >>> count, bins, ignored = plt.hist(maxima, 30, normed=True) >>> beta = np.std(maxima) * np.sqrt(6) / np.pi >>> mu = np.mean(maxima) - 0.57721*beta >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp(-np.exp(-(bins - mu)/beta)), ... linewidth=2, color='r') >>> plt.plot(bins, 1/(beta * np.sqrt(2 * np.pi)) ... * np.exp(-(bins - mu)**2 / (2 * beta**2)), ... linewidth=2, color='g') >>> plt.show() laplace(loc=0.0, scale=1.0, size=None) Draw samples from the Laplace or double exponential distribution with specified location (or mean) and scale (decay). The Laplace distribution is similar to the Gaussian/normal distribution, but is sharper at the peak and has fatter tails. It represents the difference between two independent, identically distributed exponential random variables. Parameters ---------- loc : float or array_like of floats, optional The position, :math:`\mu`, of the distribution peak. Default is 0. scale : float or array_like of floats, optional :math:`\lambda`, the exponential decay. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Laplace distribution. Notes ----- It has the probability density function .. math:: f(x; \mu, \lambda) = \frac{1}{2\lambda} \exp\left(-\frac{|x - \mu|}{\lambda}\right). The first law of Laplace, from 1774, states that the frequency of an error can be expressed as an exponential function of the absolute magnitude of the error, which leads to the Laplace distribution. For many problems in economics and health sciences, this distribution seems to model the data better than the standard Gaussian distribution. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] Kotz, Samuel, et. al. "The Laplace Distribution and Generalizations, " Birkhauser, 2001. .. [3] Weisstein, Eric W. "Laplace Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LaplaceDistribution.html .. [4] Wikipedia, "Laplace distribution", http://en.wikipedia.org/wiki/Laplace_distribution Examples -------- Draw samples from the distribution >>> loc, scale = 0., 1. >>> s = np.random.laplace(loc, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, normed=True) >>> x = np.arange(-8., 8., .01) >>> pdf = np.exp(-abs(x-loc)/scale)/(2.*scale) >>> plt.plot(x, pdf) Plot Gaussian for comparison: >>> g = (1/(scale * np.sqrt(2 * np.pi)) * ... np.exp(-(x - loc)**2 / (2 * scale**2))) >>> plt.plot(x,g) power(a, size=None) Draws samples in [0, 1] from a power distribution with positive exponent a - 1. Also known as the power function distribution. Parameters ---------- a : float or array_like of floats Parameter of the distribution. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized power distribution. Raises ------ ValueError If a < 1. Notes ----- The probability density function is .. math:: P(x; a) = ax^{a-1}, 0 \le x \le 1, a>0. The power function distribution is just the inverse of the Pareto distribution. It may also be seen as a special case of the Beta distribution. It is used, for example, in modeling the over-reporting of insurance claims. References ---------- .. [1] Christian Kleiber, Samuel Kotz, "Statistical size distributions in economics and actuarial sciences", Wiley, 2003. .. [2] Heckert, N. A. and Filliben, James J. "NIST Handbook 148: Dataplot Reference Manual, Volume 2: Let Subcommands and Library Functions", National Institute of Standards and Technology Handbook Series, June 2003. http://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> samples = 1000 >>> s = np.random.power(a, samples) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=30) >>> x = np.linspace(0, 1, 100) >>> y = a*x**(a-1.) >>> normed_y = samples*np.diff(bins)[0]*y >>> plt.plot(x, normed_y) >>> plt.show() Compare the power function distribution to the inverse of the Pareto. >>> from scipy import stats >>> rvs = np.random.power(5, 1000000) >>> rvsp = np.random.pareto(5, 1000000) >>> xx = np.linspace(0,1,100) >>> powpdf = stats.powerlaw.pdf(xx,5) >>> plt.figure() >>> plt.hist(rvs, bins=50, normed=True) >>> plt.plot(xx,powpdf,'r-') >>> plt.title('np.random.power(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, normed=True) >>> plt.plot(xx,powpdf,'r-') >>> plt.title('inverse of 1 + np.random.pareto(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, normed=True) >>> plt.plot(xx,powpdf,'r-') >>> plt.title('inverse of stats.pareto(5)') weibull(a, size=None) Draw samples from a Weibull distribution. Draw samples from a 1-parameter Weibull distribution with the given shape parameter `a`. .. math:: X = (-ln(U))^{1/a} Here, U is drawn from the uniform distribution over (0,1]. The more common 2-parameter Weibull, including a scale parameter :math:`\lambda` is just :math:`X = \lambda(-ln(U))^{1/a}`. Parameters ---------- a : float or array_like of floats Shape of the distribution. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Weibull distribution. See Also -------- scipy.stats.weibull_max scipy.stats.weibull_min scipy.stats.genextreme gumbel Notes ----- The Weibull (or Type III asymptotic extreme value distribution for smallest values, SEV Type III, or Rosin-Rammler distribution) is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. This class includes the Gumbel and Frechet distributions. The probability density for the Weibull distribution is .. math:: p(x) = \frac{a} {\lambda}(\frac{x}{\lambda})^{a-1}e^{-(x/\lambda)^a}, where :math:`a` is the shape and :math:`\lambda` the scale. The function has its peak (the mode) at :math:`\lambda(\frac{a-1}{a})^{1/a}`. When ``a = 1``, the Weibull distribution reduces to the exponential distribution. References ---------- .. [1] Waloddi Weibull, Royal Technical University, Stockholm, 1939 "A Statistical Theory Of The Strength Of Materials", Ingeniorsvetenskapsakademiens Handlingar Nr 151, 1939, Generalstabens Litografiska Anstalts Forlag, Stockholm. .. [2] Waloddi Weibull, "A Statistical Distribution Function of Wide Applicability", Journal Of Applied Mechanics ASME Paper 1951. .. [3] Wikipedia, "Weibull distribution", http://en.wikipedia.org/wiki/Weibull_distribution Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> s = np.random.weibull(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> x = np.arange(1,100.)/50. >>> def weib(x,n,a): ... return (a / n) * (x / n)**(a - 1) * np.exp(-(x / n)**a) >>> count, bins, ignored = plt.hist(np.random.weibull(5.,1000)) >>> x = np.arange(1,100.)/50. >>> scale = count.max()/weib(x, 1., 5.).max() >>> plt.plot(x, weib(x, 1., 5.)*scale) >>> plt.show() pareto(a, size=None) Draw samples from a Pareto II or Lomax distribution with specified shape. The Lomax or Pareto II distribution is a shifted Pareto distribution. The classical Pareto distribution can be obtained from the Lomax distribution by adding 1 and multiplying by the scale parameter ``m`` (see Notes). The smallest value of the Lomax distribution is zero while for the classical Pareto distribution it is ``mu``, where the standard Pareto distribution has location ``mu = 1``. Lomax can also be considered as a simplified version of the Generalized Pareto distribution (available in SciPy), with the scale set to one and the location set to zero. The Pareto distribution must be greater than zero, and is unbounded above. It is also known as the "80-20 rule". In this distribution, 80 percent of the weights are in the lowest 20 percent of the range, while the other 20 percent fill the remaining 80 percent of the range. Parameters ---------- a : float or array_like of floats Shape of the distribution. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Pareto distribution. See Also -------- scipy.stats.lomax : probability density function, distribution or cumulative density function, etc. scipy.stats.genpareto : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Pareto distribution is .. math:: p(x) = \frac{am^a}{x^{a+1}} where :math:`a` is the shape and :math:`m` the scale. The Pareto distribution, named after the Italian economist Vilfredo Pareto, is a power law probability distribution useful in many real world problems. Outside the field of economics it is generally referred to as the Bradford distribution. Pareto developed the distribution to describe the distribution of wealth in an economy. It has also found use in insurance, web page access statistics, oil field sizes, and many other problems, including the download frequency for projects in Sourceforge [1]_. It is one of the so-called "fat-tailed" distributions. References ---------- .. [1] Francis Hunt and Paul Johnson, On the Pareto Distribution of Sourceforge projects. .. [2] Pareto, V. (1896). Course of Political Economy. Lausanne. .. [3] Reiss, R.D., Thomas, M.(2001), Statistical Analysis of Extreme Values, Birkhauser Verlag, Basel, pp 23-30. .. [4] Wikipedia, "Pareto distribution", http://en.wikipedia.org/wiki/Pareto_distribution Examples -------- Draw samples from the distribution: >>> a, m = 3., 2. # shape and mode >>> s = (np.random.pareto(a, 1000) + 1) * m Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, _ = plt.hist(s, 100, normed=True) >>> fit = a*m**a / bins**(a+1) >>> plt.plot(bins, max(count)*fit/max(fit), linewidth=2, color='r') >>> plt.show() vonmises(mu, kappa, size=None) Draw samples from a von Mises distribution. Samples are drawn from a von Mises distribution with specified mode (mu) and dispersion (kappa), on the interval [-pi, pi]. The von Mises distribution (also known as the circular normal distribution) is a continuous probability distribution on the unit circle. It may be thought of as the circular analogue of the normal distribution. Parameters ---------- mu : float or array_like of floats Mode ("center") of the distribution. kappa : float or array_like of floats Dispersion of the distribution, has to be >=0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mu`` and ``kappa`` are both scalars. Otherwise, ``np.broadcast(mu, kappa).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized von Mises distribution. See Also -------- scipy.stats.vonmises : probability density function, distribution, or cumulative density function, etc. Notes ----- The probability density for the von Mises distribution is .. math:: p(x) = \frac{e^{\kappa cos(x-\mu)}}{2\pi I_0(\kappa)}, where :math:`\mu` is the mode and :math:`\kappa` the dispersion, and :math:`I_0(\kappa)` is the modified Bessel function of order 0. The von Mises is named for Richard Edler von Mises, who was born in Austria-Hungary, in what is now the Ukraine. He fled to the United States in 1939 and became a professor at Harvard. He worked in probability theory, aerodynamics, fluid mechanics, and philosophy of science. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] von Mises, R., "Mathematical Theory of Probability and Statistics", New York: Academic Press, 1964. Examples -------- Draw samples from the distribution: >>> mu, kappa = 0.0, 4.0 # mean and dispersion >>> s = np.random.vonmises(mu, kappa, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import i0 >>> plt.hist(s, 50, normed=True) >>> x = np.linspace(-np.pi, np.pi, num=51) >>> y = np.exp(kappa*np.cos(x-mu))/(2*np.pi*i0(kappa)) >>> plt.plot(x, y, linewidth=2, color='r') >>> plt.show() standard_t(df, size=None) Draw samples from a standard Student's t distribution with `df` degrees of freedom. A special case of the hyperbolic distribution. As `df` gets large, the result resembles that of the standard normal distribution (`standard_normal`). Parameters ---------- df : int or array_like of ints Degrees of freedom, should be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard Student's t distribution. Notes ----- The probability density function for the t distribution is .. math:: P(x, df) = \frac{\Gamma(\frac{df+1}{2})}{\sqrt{\pi df} \Gamma(\frac{df}{2})}\Bigl( 1+\frac{x^2}{df} \Bigr)^{-(df+1)/2} The t test is based on an assumption that the data come from a Normal distribution. The t test provides a way to test whether the sample mean (that is the mean calculated from the data) is a good estimate of the true mean. The derivation of the t-distribution was first published in 1908 by William Gosset while working for the Guinness Brewery in Dublin. Due to proprietary issues, he had to publish under a pseudonym, and so he used the name Student. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics With R", Springer, 2002. .. [2] Wikipedia, "Student's t-distribution" http://en.wikipedia.org/wiki/Student's_t-distribution Examples -------- From Dalgaard page 83 [1]_, suppose the daily energy intake for 11 women in Kj is: >>> intake = np.array([5260., 5470, 5640, 6180, 6390, 6515, 6805, 7515, \ ... 7515, 8230, 8770]) Does their energy intake deviate systematically from the recommended value of 7725 kJ? We have 10 degrees of freedom, so is the sample mean within 95% of the recommended value? >>> s = np.random.standard_t(10, size=100000) >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 Calculate the t statistic, setting the ddof parameter to the unbiased value so the divisor in the standard deviation will be degrees of freedom, N-1. >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> import matplotlib.pyplot as plt >>> h = plt.hist(s, bins=100, normed=True) For a one-sided t-test, how far out in the distribution does the t statistic appear? >>> np.sum(s>> s = np.random.standard_cauchy(1000000) >>> s = s[(s>-25) & (s<25)] # truncate distribution so it plots well >>> plt.hist(s, bins=100) >>> plt.show() noncentral_chisquare(df, nonc, size=None) Draw samples from a noncentral chi-square distribution. The noncentral :math:`\chi^2` distribution is a generalisation of the :math:`\chi^2` distribution. Parameters ---------- df : int or array_like of ints Degrees of freedom, should be > 0 as of NumPy 1.10.0, should be > 1 for earlier versions. nonc : float or array_like of floats Non-centrality, should be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` and ``nonc`` are both scalars. Otherwise, ``np.broadcast(df, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral chi-square distribution. Notes ----- The probability density function for the noncentral Chi-square distribution is .. math:: P(x;df,nonc) = \sum^{\infty}_{i=0} \frac{e^{-nonc/2}(nonc/2)^{i}}{i!} \P_{Y_{df+2i}}(x), where :math:`Y_{q}` is the Chi-square with q degrees of freedom. In Delhi (2007), it is noted that the noncentral chi-square is useful in bombing and coverage problems, the probability of killing the point target given by the noncentral chi-squared distribution. References ---------- .. [1] Delhi, M.S. Holla, "On a noncentral chi-square distribution in the analysis of weapon systems effectiveness", Metrika, Volume 15, Number 1 / December, 1970. .. [2] Wikipedia, "Noncentral chi-square distribution" http://en.wikipedia.org/wiki/Noncentral_chi-square_distribution Examples -------- Draw values from the distribution and plot the histogram >>> import matplotlib.pyplot as plt >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, normed=True) >>> plt.show() Draw values from a noncentral chisquare with very small noncentrality, and compare to a chisquare. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, .0000001, 100000), ... bins=np.arange(0., 25, .1), normed=True) >>> values2 = plt.hist(np.random.chisquare(3, 100000), ... bins=np.arange(0., 25, .1), normed=True) >>> plt.plot(values[1][0:-1], values[0]-values2[0], 'ob') >>> plt.show() Demonstrate how large values of non-centrality lead to a more symmetric distribution. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, normed=True) >>> plt.show() chisquare(df, size=None) Draw samples from a chi-square distribution. When `df` independent random variables, each with standard normal distributions (mean 0, variance 1), are squared and summed, the resulting distribution is chi-square (see Notes). This distribution is often used in hypothesis testing. Parameters ---------- df : int or array_like of ints Number of degrees of freedom. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized chi-square distribution. Raises ------ ValueError When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``) is given. Notes ----- The variable obtained by summing the squares of `df` independent, standard normally distributed random variables: .. math:: Q = \sum_{i=0}^{\mathtt{df}} X^2_i is chi-square distributed, denoted .. math:: Q \sim \chi^2_k. The probability density function of the chi-squared distribution is .. math:: p(x) = \frac{(1/2)^{k/2}}{\Gamma(k/2)} x^{k/2 - 1} e^{-x/2}, where :math:`\Gamma` is the gamma function, .. math:: \Gamma(x) = \int_0^{-\infty} t^{x - 1} e^{-t} dt. References ---------- .. [1] NIST "Engineering Statistics Handbook" http://www.itl.nist.gov/div898/handbook/eda/section3/eda3666.htm Examples -------- >>> np.random.chisquare(2,4) array([ 1.89920014, 9.00867716, 3.13710533, 5.62318272]) noncentral_f(dfnum, dfden, nonc, size=None) Draw samples from the noncentral F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters > 1. `nonc` is the non-centrality parameter. Parameters ---------- dfnum : int or array_like of ints Parameter, should be > 1. dfden : int or array_like of ints Parameter, should be > 1. nonc : float or array_like of floats Parameter, should be >= 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum``, ``dfden``, and ``nonc`` are all scalars. Otherwise, ``np.broadcast(dfnum, dfden, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral Fisher distribution. Notes ----- When calculating the power of an experiment (power = probability of rejecting the null hypothesis when a specific alternative is true) the non-central F statistic becomes important. When the null hypothesis is true, the F statistic follows a central F distribution. When the null hypothesis is not true, then it follows a non-central F statistic. References ---------- .. [1] Weisstein, Eric W. "Noncentral F-Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/NoncentralF-Distribution.html .. [2] Wikipedia, "Noncentral F-distribution", http://en.wikipedia.org/wiki/Noncentral_F-distribution Examples -------- In a study, testing for a specific alternative to the null hypothesis requires use of the Noncentral F distribution. We need to calculate the area in the tail of the distribution that exceeds the value of the F distribution for the null hypothesis. We'll plot the two probability distributions for comparison. >>> dfnum = 3 # between group deg of freedom >>> dfden = 20 # within groups degrees of freedom >>> nonc = 3.0 >>> nc_vals = np.random.noncentral_f(dfnum, dfden, nonc, 1000000) >>> NF = np.histogram(nc_vals, bins=50, normed=True) >>> c_vals = np.random.f(dfnum, dfden, 1000000) >>> F = np.histogram(c_vals, bins=50, normed=True) >>> plt.plot(F[1][1:], F[0]) >>> plt.plot(NF[1][1:], NF[0]) >>> plt.show() f(dfnum, dfden, size=None) Draw samples from an F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters should be greater than zero. The random variate of the F distribution (also known as the Fisher distribution) is a continuous probability distribution that arises in ANOVA tests, and is the ratio of two chi-square variates. Parameters ---------- dfnum : int or array_like of ints Degrees of freedom in numerator. Should be greater than zero. dfden : int or array_like of ints Degrees of freedom in denominator. Should be greater than zero. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum`` and ``dfden`` are both scalars. Otherwise, ``np.broadcast(dfnum, dfden).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Fisher distribution. See Also -------- scipy.stats.f : probability density function, distribution or cumulative density function, etc. Notes ----- The F statistic is used to compare in-group variances to between-group variances. Calculating the distribution depends on the sampling, and so it is a function of the respective degrees of freedom in the problem. The variable `dfnum` is the number of samples minus one, the between-groups degrees of freedom, while `dfden` is the within-groups degrees of freedom, the sum of the number of samples in each group minus the number of groups. References ---------- .. [1] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [2] Wikipedia, "F-distribution", http://en.wikipedia.org/wiki/F-distribution Examples -------- An example from Glantz[1], pp 47-40: Two groups, children of diabetics (25 people) and children from people without diabetes (25 controls). Fasting blood glucose was measured, case group had a mean value of 86.1, controls had a mean value of 82.2. Standard deviations were 2.09 and 2.49 respectively. Are these data consistent with the null hypothesis that the parents diabetic status does not affect their children's blood glucose levels? Calculating the F statistic from the data gives a value of 36.01. Draw samples from the distribution: >>> dfnum = 1. # between group degrees of freedom >>> dfden = 48. # within groups degrees of freedom >>> s = np.random.f(dfnum, dfden, 1000) The lower bound for the top 1% of the samples is : >>> sort(s)[-10] 7.61988120985 So there is about a 1% chance that the F statistic will exceed 7.62, the measured value is 36, so the null hypothesis is rejected at the 1% level. gamma(shape, scale=1.0, size=None) Draw samples from a Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, `shape` (sometimes designated "k") and `scale` (sometimes designated "theta"), where both parameters are > 0. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. Should be greater than zero. scale : float or array_like of floats, optional The scale of the gamma distribution. Should be greater than zero. Default is equal to 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(shape, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", http://en.wikipedia.org/wiki/Gamma_distribution Examples -------- Draw samples from the distribution: >>> shape, scale = 2., 2. # mean=4, std=2*sqrt(2) >>> s = np.random.gamma(shape, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps >>> count, bins, ignored = plt.hist(s, 50, normed=True) >>> y = bins**(shape-1)*(np.exp(-bins/scale) / ... (sps.gamma(shape)*scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') >>> plt.show() standard_gamma(shape, size=None) Draw samples from a standard Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, shape (sometimes designated "k") and scale=1. Parameters ---------- shape : float or array_like of floats Parameter, should be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` is a scalar. Otherwise, ``np.array(shape).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", http://en.wikipedia.org/wiki/Gamma_distribution Examples -------- Draw samples from the distribution: >>> shape, scale = 2., 1. # mean and width >>> s = np.random.standard_gamma(shape, 1000000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps >>> count, bins, ignored = plt.hist(s, 50, normed=True) >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ \ ... (sps.gamma(shape) * scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') >>> plt.show() standard_exponential(size=None) Draw samples from the standard exponential distribution. `standard_exponential` is identical to the exponential distribution with a scale parameter of 1. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray Drawn samples. Examples -------- Output a 3x8000 array: >>> n = np.random.standard_exponential((3, 8000)) exponential(scale=1.0, size=None) Draw samples from an exponential distribution. Its probability density function is .. math:: f(x; \frac{1}{\beta}) = \frac{1}{\beta} \exp(-\frac{x}{\beta}), for ``x > 0`` and 0 elsewhere. :math:`\beta` is the scale parameter, which is the inverse of the rate parameter :math:`\lambda = 1/\beta`. The rate parameter is an alternative, widely used parameterization of the exponential distribution [3]_. The exponential distribution is a continuous analogue of the geometric distribution. It describes many common situations, such as the size of raindrops measured over many rainstorms [1]_, or the time between page requests to Wikipedia [2]_. Parameters ---------- scale : float or array_like of floats The scale parameter, :math:`\beta = 1/\lambda`. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized exponential distribution. References ---------- .. [1] Peyton Z. Peebles Jr., "Probability, Random Variables and Random Signal Principles", 4th ed, 2001, p. 57. .. [2] Wikipedia, "Poisson process", http://en.wikipedia.org/wiki/Poisson_process .. [3] Wikipedia, "Exponential distribution", http://en.wikipedia.org/wiki/Exponential_distribution beta(a, b, size=None) Draw samples from a Beta distribution. The Beta distribution is a special case of the Dirichlet distribution, and is related to the Gamma distribution. It has the probability distribution function .. math:: f(x; a,b) = \frac{1}{B(\alpha, \beta)} x^{\alpha - 1} (1 - x)^{\beta - 1}, where the normalisation, B, is the beta function, .. math:: B(\alpha, \beta) = \int_0^1 t^{\alpha - 1} (1 - t)^{\beta - 1} dt. It is often seen in Bayesian inference and order statistics. Parameters ---------- a : float or array_like of floats Alpha, non-negative. b : float or array_like of floats Beta, non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` and ``b`` are both scalars. Otherwise, ``np.broadcast(a, b).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized beta distribution. normal(loc=0.0, scale=1.0, size=None) Draw random samples from a normal (Gaussian) distribution. The probability density function of the normal distribution, first derived by De Moivre and 200 years later by both Gauss and Laplace independently [2]_, is often called the bell curve because of its characteristic shape (see the example below). The normal distributions occurs often in nature. For example, it describes the commonly occurring distribution of samples influenced by a large number of tiny, random disturbances, each with its own unique distribution [2]_. Parameters ---------- loc : float or array_like of floats Mean ("centre") of the distribution. scale : float or array_like of floats Standard deviation (spread or "width") of the distribution. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized normal distribution. See Also -------- scipy.stats.norm : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gaussian distribution is .. math:: p(x) = \frac{1}{\sqrt{ 2 \pi \sigma^2 }} e^{ - \frac{ (x - \mu)^2 } {2 \sigma^2} }, where :math:`\mu` is the mean and :math:`\sigma` the standard deviation. The square of the standard deviation, :math:`\sigma^2`, is called the variance. The function has its peak at the mean, and its "spread" increases with the standard deviation (the function reaches 0.607 times its maximum at :math:`x + \sigma` and :math:`x - \sigma` [2]_). This implies that `numpy.random.normal` is more likely to return samples lying close to the mean, rather than those far away. References ---------- .. [1] Wikipedia, "Normal distribution", http://en.wikipedia.org/wiki/Normal_distribution .. [2] P. R. Peebles Jr., "Central Limit Theorem" in "Probability, Random Variables and Random Signal Principles", 4th ed., 2001, pp. 51, 51, 125. Examples -------- Draw samples from the distribution: >>> mu, sigma = 0, 0.1 # mean and standard deviation >>> s = np.random.normal(mu, sigma, 1000) Verify the mean and the variance: >>> abs(mu - np.mean(s)) < 0.01 True >>> abs(sigma - np.std(s, ddof=1)) < 0.01 True Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, normed=True) >>> plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) * ... np.exp( - (bins - mu)**2 / (2 * sigma**2) ), ... linewidth=2, color='r') >>> plt.show() standard_normal(size=None) Draw samples from a standard Normal distribution (mean=0, stdev=1). Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray Drawn samples. Examples -------- >>> s = np.random.standard_normal(8000) >>> s array([ 0.6888893 , 0.78096262, -0.89086505, ..., 0.49876311, #random -0.38672696, -0.4685006 ]) #random >>> s.shape (8000,) >>> s = np.random.standard_normal(size=(3, 4, 2)) >>> s.shape (3, 4, 2) random_integers(low, high=None, size=None) Random integers of type np.int between `low` and `high`, inclusive. Return random integers of type np.int from the "discrete uniform" distribution in the closed interval [`low`, `high`]. If `high` is None (the default), then results are from [1, `low`]. The np.int type translates to the C long type used by Python 2 for "short" integers and its precision is platform dependent. This function has been deprecated. Use randint instead. .. deprecated:: 1.11.0 Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution (unless ``high=None``, in which case this parameter is the *highest* such integer). high : int, optional If provided, the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- random.randint : Similar to `random_integers`, only for the half-open interval [`low`, `high`), and 0 is the lowest value if `high` is omitted. Notes ----- To sample from N evenly spaced floating-point numbers between a and b, use:: a + (b - a) * (np.random.random_integers(N) - 1) / (N - 1.) Examples -------- >>> np.random.random_integers(5) 4 >>> type(np.random.random_integers(5)) >>> np.random.random_integers(5, size=(3,2)) array([[5, 4], [3, 3], [4, 5]]) Choose five random numbers from the set of five evenly-spaced numbers between 0 and 2.5, inclusive (*i.e.*, from the set :math:`{0, 5/8, 10/8, 15/8, 20/8}`): >>> 2.5 * (np.random.random_integers(5, size=(5,)) - 1) / 4. array([ 0.625, 1.25 , 0.625, 0.625, 2.5 ]) Roll two six sided dice 1000 times and sum the results: >>> d1 = np.random.random_integers(1, 6, 1000) >>> d2 = np.random.random_integers(1, 6, 1000) >>> dsums = d1 + d2 Display results as a histogram: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(dsums, 11, normed=True) >>> plt.show() randn(d0, d1, ..., dn) Return a sample (or samples) from the "standard normal" distribution. If positive, int_like or int-convertible arguments are provided, `randn` generates an array of shape ``(d0, d1, ..., dn)``, filled with random floats sampled from a univariate "normal" (Gaussian) distribution of mean 0 and variance 1 (if any of the :math:`d_i` are floats, they are first converted to integers by truncation). A single float randomly sampled from the distribution is returned if no argument is provided. This is a convenience function. If you want an interface that takes a tuple as the first argument, use `numpy.random.standard_normal` instead. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, should be all positive. If no argument is given a single Python float is returned. Returns ------- Z : ndarray or float A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from the standard normal distribution, or a single such float if no parameters were supplied. See Also -------- random.standard_normal : Similar, but takes a tuple as its argument. Notes ----- For random samples from :math:`N(\mu, \sigma^2)`, use: ``sigma * np.random.randn(...) + mu`` Examples -------- >>> np.random.randn() 2.1923875335537315 #random Two-by-four array of samples from N(3, 6.25): >>> 2.5 * np.random.randn(2, 4) + 3 array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], #random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) #random rand(d0, d1, ..., dn) Random values in a given shape. Create an array of the given shape and populate it with random samples from a uniform distribution over ``[0, 1)``. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, should all be positive. If no argument is given a single Python float is returned. Returns ------- out : ndarray, shape ``(d0, d1, ..., dn)`` Random values. See Also -------- random Notes ----- This is a convenience function. If you want an interface that takes a shape-tuple as the first argument, refer to np.random.random_sample . Examples -------- >>> np.random.rand(3,2) array([[ 0.14022471, 0.96360618], #random [ 0.37601032, 0.25528411], #random [ 0.49313049, 0.94909878]]) #random uniform(low=0.0, high=1.0, size=None) Draw samples from a uniform distribution. Samples are uniformly distributed over the half-open interval ``[low, high)`` (includes low, but excludes high). In other words, any value within the given interval is equally likely to be drawn by `uniform`. Parameters ---------- low : float or array_like of floats, optional Lower boundary of the output interval. All values generated will be greater than or equal to low. The default value is 0. high : float or array_like of floats Upper boundary of the output interval. All values generated will be less than high. The default value is 1.0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``low`` and ``high`` are both scalars. Otherwise, ``np.broadcast(low, high).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized uniform distribution. See Also -------- randint : Discrete uniform distribution, yielding integers. random_integers : Discrete uniform distribution over the closed interval ``[low, high]``. random_sample : Floats uniformly distributed over ``[0, 1)``. random : Alias for `random_sample`. rand : Convenience function that accepts dimensions as input, e.g., ``rand(2,2)`` would generate a 2-by-2 array of floats, uniformly distributed over ``[0, 1)``. Notes ----- The probability density function of the uniform distribution is .. math:: p(x) = \frac{1}{b - a} anywhere within the interval ``[a, b)``, and zero elsewhere. When ``high`` == ``low``, values of ``low`` will be returned. If ``high`` < ``low``, the results are officially undefined and may eventually raise an error, i.e. do not rely on this function to behave when passed arguments satisfying that inequality condition. Examples -------- Draw samples from the distribution: >>> s = np.random.uniform(-1,0,1000) All values are within the given interval: >>> np.all(s >= -1) True >>> np.all(s < 0) True Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 15, normed=True) >>> plt.plot(bins, np.ones_like(bins), linewidth=2, color='r') >>> plt.show() choice(a, size=None, replace=True, p=None) Generates a random sample from a given 1-D array .. versionadded:: 1.7.0 Parameters ----------- a : 1-D array-like or int If an ndarray, a random sample is generated from its elements. If an int, the random sample is generated as if a were np.arange(a) size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. replace : boolean, optional Whether the sample is with or without replacement p : 1-D array-like, optional The probabilities associated with each entry in a. If not given the sample assumes a uniform distribution over all entries in a. Returns -------- samples : single item or ndarray The generated random samples Raises ------- ValueError If a is an int and less than zero, if a or p are not 1-dimensional, if a is an array-like of size 0, if p is not a vector of probabilities, if a and p have different lengths, or if replace=False and the sample size is greater than the population size See Also --------- randint, shuffle, permutation Examples --------- Generate a uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3) array([0, 3, 4]) >>> #This is equivalent to np.random.randint(0,5,3) Generate a non-uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0]) array([3, 3, 0]) Generate a uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False) array([3,1,0]) >>> #This is equivalent to np.random.permutation(np.arange(5))[:3] Generate a non-uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0]) array([2, 3, 0]) Any of the above can be repeated with an arbitrary array-like instead of just integers. For instance: >>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher'] >>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3]) array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'], dtype='|S11') bytes(length) Return random bytes. Parameters ---------- length : int Number of random bytes. Returns ------- out : str String of length `length`. Examples -------- >>> np.random.bytes(10) ' eh\x85\x022SZ\xbf\xa4' #random randint(low, high=None, size=None, dtype='l') Return random integers from `low` (inclusive) to `high` (exclusive). Return random integers from the "discrete uniform" distribution of the specified dtype in the "half-open" interval [`low`, `high`). If `high` is None (the default), then results are from [0, `low`). Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution (unless ``high=None``, in which case this parameter is one above the *highest* such integer). high : int, optional If provided, one above the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result. All dtypes are determined by their name, i.e., 'int64', 'int', etc, so byteorder is not available and a specific precision may have different C types depending on the platform. The default value is 'np.int'. .. versionadded:: 1.11.0 Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- random.random_integers : similar to `randint`, only for the closed interval [`low`, `high`], and 1 is the lowest value if `high` is omitted. In particular, this other one is the one to use to generate uniformly distributed discrete non-integers. Examples -------- >>> np.random.randint(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) >>> np.random.randint(1, size=10) array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0]) Generate a 2 x 4 array of ints between 0 and 4, inclusive: >>> np.random.randint(5, size=(2, 4)) array([[4, 0, 2, 1], [3, 2, 2, 0]]) tomaxint(size=None) Random integers between 0 and ``sys.maxint``, inclusive. Return a sample of uniformly distributed random integers in the interval [0, ``sys.maxint``]. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray Drawn samples, with shape `size`. See Also -------- randint : Uniform sampling over a given half-open interval of integers. random_integers : Uniform sampling over a given closed interval of integers. Examples -------- >>> RS = np.random.mtrand.RandomState() # need a RandomState object >>> RS.tomaxint((2,2,2)) array([[[1170048599, 1600360186], [ 739731006, 1947757578]], [[1871712945, 752307660], [1601631370, 1479324245]]]) >>> import sys >>> sys.maxint 2147483647 >>> RS.tomaxint((2,2,2)) < sys.maxint array([[[ True, True], [ True, True]], [[ True, True], [ True, True]]], dtype=bool) random_sample(size=None) Return random floats in the half-open interval [0.0, 1.0). Results are from the "continuous uniform" distribution over the stated interval. To sample :math:`Unif[a, b), b > a` multiply the output of `random_sample` by `(b-a)` and add `a`:: (b - a) * random_sample() + a Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray of floats Array of random floats of shape `size` (unless ``size=None``, in which case a single float is returned). Examples -------- >>> np.random.random_sample() 0.47108547995356098 >>> type(np.random.random_sample()) >>> np.random.random_sample((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) Three-by-two array of random numbers from [-5, 0): >>> 5 * np.random.random_sample((3, 2)) - 5 array([[-3.99149989, -0.52338984], [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) set_state(state) Set the internal state of the generator from a tuple. For use if one has reason to manually (re-)set the internal state of the "Mersenne Twister"[1]_ pseudo-random number generating algorithm. Parameters ---------- state : tuple(str, ndarray of 624 uints, int, int, float) The `state` tuple has the following items: 1. the string 'MT19937', specifying the Mersenne Twister algorithm. 2. a 1-D array of 624 unsigned integers ``keys``. 3. an integer ``pos``. 4. an integer ``has_gauss``. 5. a float ``cached_gaussian``. Returns ------- out : None Returns 'None' on success. See Also -------- get_state Notes ----- `set_state` and `get_state` are not needed to work with any of the random distributions in NumPy. If the internal state is manually altered, the user should know exactly what he/she is doing. For backwards compatibility, the form (str, array of 624 uints, int) is also accepted although it is missing some information about the cached Gaussian value: ``state = ('MT19937', keys, pos)``. References ---------- .. [1] M. Matsumoto and T. Nishimura, "Mersenne Twister: A 623-dimensionally equidistributed uniform pseudorandom number generator," *ACM Trans. on Modeling and Computer Simulation*, Vol. 8, No. 1, pp. 3-30, Jan. 1998. get_state() Return a tuple representing the internal state of the generator. For more details, see `set_state`. Returns ------- out : tuple(str, ndarray of 624 uints, int, int, float) The returned tuple has the following items: 1. the string 'MT19937'. 2. a 1-D array of 624 unsigned integer keys. 3. an integer ``pos``. 4. an integer ``has_gauss``. 5. a float ``cached_gaussian``. See Also -------- set_state Notes ----- `set_state` and `get_state` are not needed to work with any of the random distributions in NumPy. If the internal state is manually altered, the user should know exactly what he/she is doing. seed(seed=None) Seed the generator. This method is called when `RandomState` is initialized. It can be called again to re-seed the generator. For details, see `RandomState`. Parameters ---------- seed : int or array_like, optional Seed for `RandomState`. Must be convertible to 32 bit unsigned integers. See Also -------- RandomState _rand_uint64(low, high, size, rngstate) Return random np.uint64 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.uint64 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.uint64 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_uint32(low, high, size, rngstate) Return random np.uint32 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.uint32 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.uint32 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_uint16(low, high, size, rngstate) Return random np.uint16 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.uint16 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.uint16 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_uint8(low, high, size, rngstate) Return random np.uint8 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.uint8 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.uint8 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_int64(low, high, size, rngstate) Return random np.int64 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.int64 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.int64 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_int32(low, high, size, rngstate) Return random np.int32 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.int32 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.int32 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_int16(low, high, size, rngstate) Return random np.int16 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.int16 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.int16 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_int8(low, high, size, rngstate) Return random np.int8 integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.int8 type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.int8 `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. _rand_bool(low, high, size, rngstate) Return random np.bool_ integers between ``low`` and ``high``, inclusive. Return random integers from the "discrete uniform" distribution in the closed interval [``low``, ``high``). On entry the arguments are presumed to have been validated for size and order for the np.bool_ type. Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution. high : int Highest (signed) integer to be drawn from the distribution. size : int or tuple of ints Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. rngstate : encapsulated pointer to rk_state The specific type depends on the python version. In Python 2 it is a PyCObject, in Python 3 a PyCapsule object. Returns ------- out : python integer or ndarray of np.bool_ `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. @Y $v 7M Z P;   Y`{ v [l `kІ @` PU6 @QN `JT @?~ 4;w `.  ) 7" k  Q` Pn 8 М` @ Њ ` ` ?` p B ϋ @Z 0n     8 v R pj n PZ 0Q 09G @_ > ݌  0  @$ x  < P 0o  r 5  I 5  w @M  E ~  g k   ^ A] ] &] ] ] ] ] `] !0] ] "\ H\ p\ "P\  0\ \ "`[ [ 'p[ P@[ !@[ "Z "Z $@Z ,Z *Y -Y %PPY 0Y PY $X X X pX PX p X (W %0W !`W @W (@W -0V '`V (@V # V U #U U ! `U @U  U U T T %T (`T GT LS S #S S S 8S S pS PS  S &R R R R `R R 8R R X R R  xR pR dR x_R ]R xVR QR HHR `D ?D :D 0D +D  %D B eB B @`B 4@B `: ( S: / s  / / / / / P`/ %0 / )  / / 0/  . 8. H. . `. . h. X. . x% % %  % % p % x% `% @y% p% `% ^% @  h8 h0 H  0  (@ (  <` X  x h    x @ 81 ` + x    x p h    X  P     @     ` X x r    H   0 p X` $ hP H   б ̱   @  P   x h H@ #X x  8     @   &0   ( @  X  @  H x p` xQ PH 0 } , | q   q X`d Y Hd 8@d x0d  'd (d 'c c (c c c c c hc c c `c xc `c Kc `T HT XP 8P H G G ; b @; # ; ; X; 0; : : 0 : 0 : 6  6  6  6 ( 6  6  p6  `6  P6 H6 06  -   -  , % 5P%  % x `  _  ` 8` y(P H @ 0               X  p Hp P h h  A   0 (     # p    H  8 H h qX     x hh (P F A H8  (   `P ` M C X< P5 H. `(    @     X ( 8 @ ذ  ޣ 0٣ ` y s 0| Dp pj j @k .   Ā  `w w؀  @ e bS L% _ _ GCC: (GNU) 11.4.1 20231218 (Red Hat 11.4.1-3)AV:4g1231RV:running gcc 11.4.1 20231218BV:annobin gcc 11.4.1 20231218GW:0x3d1056a ltoSP:3SC:-1 ltoCF:8 ltoFL:-2 ltoGA:1PI:2SE:0iS:0GA$3a1KKGA$3a1GA$3a1y y GA$3a1KLGA$3a1y y GA$3a1y y GA$3a1GA$3a1y y  pj E' j yJ @k `kH  k 9X P  | @h8$ 0<H@ U c q x  p  h     7 Y@ h~ 8<O(z@ 2Fr"Nwx&p9hL`_XrPH@80( 0G\r +<M^qxph`X P. HZ @ 0 ( x  h  `    # 3 A O ] k 0   (             2  H ^ X o H  8   ~ 0 8  8 1  P  o ` @       4  V  f  z    `  P  0  (      , F Wf|4I\mxph`XPH#@?8Q0`(t ?j7b.Xxph`,XWPH@80*(U |!Lw$O{#NwxphG`oXPH@8E0o( Ak@f=hx?pjh`XPH" ~ 4L ] l ~p    A  -  [  &   8  <! y A! k!@$ ! 0 ! > M !G "Q yI"Z u"j $ "v " r " b # E# n# # #` # , /$` [$@  $`  $  $` % >%"l% )>%`.Y %;q%@? &`J5H&Qo&U@ &`s &`ke&l 'v`?'`{_o'2'/'@'` y (٣  (ޣ .( V(ذ f( w( ( ( ( ( ( ) ) @ 7) G)( U). d)5 s)< )C )` M ) ) `) )( *8 *A **F 6*P N*h _*x o* * * *X * q* + >+ U+ |+ + + + + + #, (, 8, J, Z,  h, u,  A,h ,p , , , , - - (- <- N- ^- l- y- - - - - -0 -@ -H -P .` y.. ?. M. [. _.` . .% .% .% 5/, /- 1/ -  Y/06 u/H6 /P6 /`6 /p6 /6 /6 /6 /6 06 06 F0: X0: g0: t0: 0; 0; 0; 0 ; 0@; #0; b '1G 51G M1H t1P 1P 1HT 1`T 1Kc 1`c 2xc '2c 72c E2c Q2c a2c o2c {2c 2c 2c 2c 2c 2d '2'd 20d 3@d 3Hd $3`d Y L3q [3q  3| 3} , 30 3H 3Q 3` 4x $4 24 Z4 t4 4 4 4 4 4 4 4 4 &5 "5@ J5 ^5 i5 |5 5 5 5 5@ #5h 6x 6 $6 06 =6 `6 q6̱ }6б 6 6 6 6P 6` $ &7 77 J7 V7 i7 v7 7 7 7 r 7X 7` 7 8 '8 38 =8 M8 [8 l8 y8 8 8 8 8 8 8 8 8 8  9 9 (9x99+ G91 V9@ 8~9x 9 9 9 9 9 : #: 5: <]: o:@ : : :  :0 :8 :@  ;^% ;`% );p% 6;y% D;% P;% ^;% q;% ;% ;% ;% ;% ;. ;. ;. <. <. (<. 6<. C<. N<. [</ e</ t< / < / )<`/ %</ </ </ =/ =/ +=/ s S=S: b=`: =@B =`B 4=B =B =B e>%D >+D +>0D =>:D L>?D Y>`D >HR >QR >VR >]R >_R >dR >pR >xR >R ?R ?R ?R +?R DX bDX DY $D0Y DPY DY %$EY -MEZ *vE@Z ,EZ $EZ "E[ "F@[ !>F[ dF[ 'F[ F\ "F0\ FP\ G\ "EG\ ^G\ G] "G0] G`] !G]  H] H] (H] 2H] FH] &oH] H ^ AHH HI`8IXIwI@II}II I pJ 0 6J Kb PDJ UJ +_J KaJ LtJ PLJ@J J LJ J (J (K ($K (8K (LK@ (`K@ (tK (K (KP K@ K0 K K (K 0%L (8L KL ^L qL L@ L` L0 L L L L M ("M 6M` JM ^M rM M M` M@ M M M M N &N :N` (NN@ bN <N NP N0 N N (Np  O 0!O 5O IO WO 0}OOOO(OP p7P=PbP}PP pPPPQ04Q @m VQxqQhQ QQ MQRRX8R ` PR pP zR8RR`RR@RS85S 0opSHS `SSST .$6TQTplTTxTTThTU ,UGUbUh~U`UUUUU j$V  SV XV VVVp W 0 $#W MWiW8WW WW P"Xx=XXXtXX C$X@XXP YX'YCY 0nwYYYY 7h!Z P ~-ZGZ`Zh{ZZ(ZpZZp [X%[`D[`[{[[ [[  \`<\XY\ P\\\y \ b\ ]=]`]P{]] .F] wQ]^.^ Pnw.`^{^^^@^ ` U^ _p_ [8S_`l__p_ ___`(``B`X^`h{` 09 &`` `Ha 0 A*a 98Zaraa <a ` a(ab @S?b\bxbb bb kn,cGcccx~cc0c cc p Rd.d JRdodhddd `.$e("e -Kekepe8eehe @f f Sf  f(fffpf(g(2g@NgjgpgHgHgggh1h jhhhxhh h(1i =oii iPii 5 .j `j{j UjjHj jkP5kPkpkkk@kkXkPl @iMl@kll ll =l 0 % m$m?m kmm 1m m Am@n0nPLnhn н 5nnn Qn 5o-oKo s2o !o @Zop P@p 7xppphppHqq6q rlq 4nq qq q 0-r 3 }VrPqrrr dr 0 sr t s  ?s`Zsuss I^ssst p $t?t AgtPt t t ,t @M /#u?uZuvuuHu p u $vvPMz0hzzz zzzz {.{`G{ >Jd{0{{{{ ||7|Q|hl||| |@|} QQ}(l}}x}} p }~h#~:~8U~n~~x~(~ ~/K  Pd  o :T@` v P ^ɀ  B:85Q xm ?nŁ݁0-0DH_8{ς @_c25 0 Tn  @̓07S0j HԄ w !@<V jS  М.$ʅp Yn8XSpo  LQچ`! ]xw80LJX&(H( P    y  d j       @*  0 #4 W9K` nz0 P  = P ҈X +8ESb~  ]  lj҉  '> (H ` _r  ` rϊ܊  -   0 2 *=Pb s  { P؋ C ` |0@TY ` *n  0 /ƌߌ-  ? :INbn P  p ( p ֍ P $1L_r `  RŎ 0 (Ў ܎  p 6 BN Rdt ɏݏ x } ` !2 0 :IZm{  ]א + p -59JZi  Q ȑ ֑҅ + 1=OJgs  R’ƒג):O  [_s r+ p Wœ֓0A [j{Ô  Wה"" 35 ` DQ_s__pyx_tp_traverse_6mtrand_RandomState__pyx_tp_clear_6mtrand_RandomState__pyx_tp_new_6mtrand_RandomState__pyx_vtabptr_6mtrand_RandomState__pyx_pw_6mtrand_11RandomState_105permutation__pyx_n_s_arange__pyx_n_s_shuffle__pyx_tp_dealloc_6mtrand_RandomState__pyx_moduledef__pyx_string_tab__pyx_int_2__pyx_int_3__pyx_int_5__pyx_int_10__pyx_int_128__pyx_int_256__pyx_int_32768__pyx_int_65536__pyx_int_2147483648__pyx_int_4294967296__pyx_int_9223372036854775808__pyx_int_18446744073709551616__pyx_int_neg_128__pyx_int_neg_32768__pyx_int_neg_2147483648__pyx_int_neg_9223372036854775808__pyx_n_s_main__pyx_n_s_ImportError__pyx_builtin_ImportError__pyx_n_s_ValueError__pyx_n_s_range__pyx_n_s_TypeError__pyx_n_s_OverflowError__pyx_n_s_DeprecationWarning__pyx_n_s_RuntimeWarning__pyx_builtin_RuntimeWarning__pyx_n_s_reversed__pyx_kp_s_numpy_core_multiarray_failed_to__pyx_tuple___pyx_kp_s_size_is_not_compatible_with_inpu__pyx_kp_s_Seed_must_be_between_0_and_2_32__pyx_n_s_L__pyx_kp_s_algorithm_must_be_MT19937__pyx_kp_s_state_must_be_624_longs__pyx_kp_s_low_high__pyx_kp_s_a_must_be_1_dimensional_or_an_in__pyx_kp_s_a_must_be_greater_than_0__pyx_kp_s_a_must_be_1_dimensional__pyx_kp_s_a_must_be_non_empty__pyx_kp_s_p_must_be_1_dimensional__pyx_kp_s_a_and_p_must_have_same_size__pyx_kp_s_probabilities_are_not_non_negati__pyx_kp_s_probabilities_do_not_sum_to_1__pyx_kp_s_Cannot_take_a_larger_sample_than__pyx_kp_s_Fewer_non_zero_entries_in_p_than__pyx_kp_s_Range_exceeds_valid_bounds__pyx_kp_s_scale_0__pyx_kp_s_a_0__pyx_kp_s_b_0__pyx_kp_s_shape_0__pyx_kp_s_dfnum_0__pyx_kp_s_dfden_0__pyx_kp_s_dfnum_1__pyx_kp_s_nonc_0__pyx_kp_s_df_0__pyx_kp_s_kappa_0__pyx_kp_s_a_0_2__pyx_kp_s_sigma_0__pyx_kp_s_sigma_0_0__pyx_kp_s_scale_0_0__pyx_kp_s_mean_0__pyx_kp_s_scale_0_2__pyx_kp_s_mean_0_0__pyx_kp_s_scale_0_0_2__pyx_kp_s_left_mode__pyx_kp_s_mode_right__pyx_kp_s_left_right__pyx_kp_s_n_0__pyx_kp_s_p_0__pyx_kp_s_p_1__pyx_kp_s_p_is_nan__pyx_kp_s_n_0_2__pyx_kp_s_lam_0__pyx_kp_s_lam_value_too_large__pyx_kp_s_lam_value_too_large_2__pyx_kp_s_a_1_0__pyx_kp_s_p_0_0__pyx_kp_s_p_1_0__pyx_kp_s_ngood_0__pyx_kp_s_nbad_0__pyx_kp_s_nsample_1__pyx_kp_s_ngood_nbad_nsample__pyx_kp_s_p_0_0_2__pyx_kp_s_p_1_0_2__pyx_kp_s_mean_must_be_1_dimensional__pyx_kp_s_cov_must_be_2_dimensional_and_sq__pyx_kp_s_mean_and_cov_must_have_same_leng__pyx_kp_s_check_valid_must_equal_warn_rais__pyx_kp_s_covariance_is_not_positive_semid__pyx_kp_s_sum_pvals_1_1_0__pyx_tuple__174__pyx_tuple__175__pyx_tuple__176__pyx_n_s_off__pyx_n_s_array_data__pyx_n_s_state__pyx_n_s_cnt__pyx_n_s_out__pyx_n_s_buf__pyx_n_s_rng__pyx_n_s_rand_bool__pyx_kp_s_randint_helpers_pxi__pyx_n_s_rand_int8__pyx_n_s_rand_int16__pyx_n_s_rand_int32__pyx_n_s_rand_int64__pyx_n_s_rand_uint8__pyx_n_s_rand_uint16__pyx_n_s_rand_uint32__pyx_n_s_rand_uint64__pyx_kp_s_mtrand_pyx__pyx_tuple__197__pyx_tuple__198__pyx_vtable_6mtrand_RandomState__pyx_type_6mtrand_RandomState__pyx_f_6mtrand_11RandomState__shuffle_raw__pyx_n_s_pyx_vtable__pyx_ptype_6mtrand_RandomState__pyx_n_s_mtrand__pyx_mdef_6mtrand_1_rand_bool__pyx_mdef_6mtrand_3_rand_int8__pyx_mdef_6mtrand_5_rand_int16__pyx_mdef_6mtrand_7_rand_int32__pyx_mdef_6mtrand_9_rand_int64__pyx_mdef_6mtrand_11_rand_uint8__pyx_mdef_6mtrand_13_rand_uint16__pyx_mdef_6mtrand_15_rand_uint32__pyx_mdef_6mtrand_17_rand_uint64__pyx_n_s_numpy__pyx_n_s_threading__pyx_n_s_dummy_threading__pyx_mdef_6mtrand_19_shape_from_size__pyx_n_s_iinfo__pyx_n_s_max__pyx_n_s_rand_2__pyx_n_s_choice__pyx_n_s_bytes__pyx_n_s_uniform__pyx_n_s_randn__pyx_n_s_random_integers__pyx_n_s_normal__pyx_n_s_beta__pyx_n_s_exponential__pyx_n_s_standard_exponential__pyx_n_s_standard_gamma__pyx_n_s_gamma__pyx_n_s_f__pyx_n_s_noncentral_f__pyx_n_s_chisquare__pyx_n_s_noncentral_chisquare__pyx_n_s_standard_cauchy__pyx_n_s_standard_t__pyx_n_s_vonmises__pyx_n_s_pareto__pyx_n_s_weibull__pyx_n_s_power__pyx_n_s_laplace__pyx_n_s_gumbel__pyx_n_s_logistic__pyx_n_s_lognormal__pyx_n_s_rayleigh__pyx_n_s_wald__pyx_n_s_triangular__pyx_n_s_binomial__pyx_n_s_negative_binomial__pyx_n_s_poisson__pyx_n_s_zipf__pyx_n_s_geometric__pyx_n_s_hypergeometric__pyx_n_s_logseries__pyx_n_s_multivariate_normal__pyx_n_s_multinomial__pyx_n_s_dirichlet__pyx_kp_u_random_sample_size_None_Return__pyx_kp_u_RandomState_random_sample_line_8__pyx_kp_u_tomaxint_size_None_Random_integ__pyx_kp_u_RandomState_tomaxint_line_858__pyx_kp_u_randint_low_high_None_size_None__pyx_kp_u_RandomState_randint_line_905__pyx_kp_u_bytes_length_Return_random_byte__pyx_kp_u_RandomState_bytes_line_999__pyx_kp_u_choice_a_size_None_replace_True__pyx_kp_u_RandomState_choice_line_1028__pyx_kp_u_uniform_low_0_0_high_1_0_size_N__pyx_kp_u_RandomState_uniform_line_1210__pyx_kp_u_rand_d0_d1_dn_Random_values_in__pyx_kp_u_RandomState_rand_line_1316__pyx_kp_u_randn_d0_d1_dn_Return_a_sample__pyx_kp_u_RandomState_randn_line_1360__pyx_kp_u_random_integers_low_high_None_s__pyx_kp_u_RandomState_random_integers_line__pyx_kp_u_standard_normal_size_None_Draw__pyx_kp_u_RandomState_standard_normal_line__pyx_kp_u_normal_loc_0_0_scale_1_0_size_N__pyx_kp_u_RandomState_normal_line_1547__pyx_kp_u_standard_exponential_size_None__pyx_kp_u_RandomState_standard_exponential__pyx_kp_u_standard_gamma_shape_size_None__pyx_kp_u_RandomState_standard_gamma_line__pyx_kp_u_gamma_shape_scale_1_0_size_None__pyx_kp_u_RandomState_gamma_line_1896__pyx_kp_u_f_dfnum_dfden_size_None_Draw_sa__pyx_kp_u_RandomState_f_line_1992__pyx_kp_u_noncentral_f_dfnum_dfden_nonc_s__pyx_kp_u_RandomState_noncentral_f_line_20__pyx_kp_u_chisquare_df_size_None_Draw_sam__pyx_kp_u_RandomState_chisquare_line_2196__pyx_kp_u_noncentral_chisquare_df_nonc_si__pyx_kp_u_RandomState_noncentral_chisquare__pyx_kp_u_standard_cauchy_size_None_Draw__pyx_kp_u_RandomState_standard_cauchy_line__pyx_kp_u_standard_t_df_size_None_Draw_sa__pyx_kp_u_RandomState_standard_t_line_2445__pyx_kp_u_vonmises_mu_kappa_size_None_Dra__pyx_kp_u_RandomState_vonmises_line_2551__pyx_kp_u_pareto_a_size_None_Draw_samples__pyx_kp_u_RandomState_pareto_line_2649__pyx_kp_u_weibull_a_size_None_Draw_sample__pyx_kp_u_RandomState_weibull_line_2759__pyx_kp_u_power_a_size_None_Draws_samples__pyx_kp_u_RandomState_power_line_2869__pyx_kp_u_laplace_loc_0_0_scale_1_0_size__pyx_kp_u_RandomState_laplace_line_2980__pyx_kp_u_gumbel_loc_0_0_scale_1_0_size_N__pyx_kp_u_RandomState_gumbel_line_3078__pyx_kp_u_logistic_loc_0_0_scale_1_0_size__pyx_kp_u_RandomState_logistic_line_3209__pyx_kp_u_lognormal_mean_0_0_sigma_1_0_si__pyx_kp_u_RandomState_lognormal_line_3302__pyx_kp_u_rayleigh_scale_1_0_size_None_Dr__pyx_kp_u_RandomState_rayleigh_line_3426__pyx_kp_u_wald_mean_scale_size_None_Draw__pyx_kp_u_RandomState_wald_line_3505__pyx_kp_u_triangular_left_mode_right_size__pyx_kp_u_RandomState_triangular_line_3592__pyx_kp_u_binomial_n_p_size_None_Draw_sam__pyx_kp_u_RandomState_binomial_line_3686__pyx_kp_u_negative_binomial_n_p_size_None__pyx_kp_u_RandomState_negative_binomial_li__pyx_kp_u_poisson_lam_1_0_size_None_Draw__pyx_kp_u_RandomState_poisson_line_3903__pyx_kp_u_zipf_a_size_None_Draw_samples_f__pyx_kp_u_RandomState_zipf_line_3991__pyx_kp_u_geometric_p_size_None_Draw_samp__pyx_kp_u_RandomState_geometric_line_4082__pyx_kp_u_hypergeometric_ngood_nbad_nsamp__pyx_kp_u_RandomState_hypergeometric_line__pyx_kp_u_logseries_p_size_None_Draw_samp__pyx_kp_u_RandomState_logseries_line_4272__pyx_kp_u_multivariate_normal_mean_cov_si__pyx_kp_u_RandomState_multivariate_normal__pyx_kp_u_multinomial_n_pvals_size_None_D__pyx_kp_u_RandomState_multinomial_line_453__pyx_kp_u_dirichlet_alpha_size_None_Draw__pyx_kp_u_RandomState_dirichlet_line_4643__pyx_kp_u_shuffle_x_Modify_a_sequence_in__pyx_kp_u_RandomState_shuffle_line_4759__pyx_kp_u_permutation_x_Randomly_permute__pyx_kp_u_RandomState_permutation_line_484__pyx_n_s_test__pyx_pw_6mtrand_11RandomState_103shuffle__pyx_n_s_ctypes__pyx_n_s_data__pyx_n_s_strides__pyx_n_s_empty_like__pyx_n_s_itemsize__pyx_doc_6mtrand_11RandomState_104permutation__pyx_doc_6mtrand_11RandomState_102shuffle__pyx_doc_6mtrand_11RandomState_100dirichlet__pyx_doc_6mtrand_11RandomState_98multinomial__pyx_doc_6mtrand_11RandomState_96multivariate_normal__pyx_doc_6mtrand_11RandomState_94logseries__pyx_doc_6mtrand_11RandomState_92hypergeometric__pyx_doc_6mtrand_11RandomState_90geometric__pyx_doc_6mtrand_11RandomState_88zipf__pyx_doc_6mtrand_11RandomState_86poisson__pyx_doc_6mtrand_11RandomState_84negative_binomial__pyx_doc_6mtrand_11RandomState_82binomial__pyx_doc_6mtrand_11RandomState_80triangular__pyx_doc_6mtrand_11RandomState_78wald__pyx_doc_6mtrand_11RandomState_76rayleigh__pyx_doc_6mtrand_11RandomState_74lognormal__pyx_doc_6mtrand_11RandomState_72logistic__pyx_doc_6mtrand_11RandomState_70gumbel__pyx_doc_6mtrand_11RandomState_68laplace__pyx_doc_6mtrand_11RandomState_66power__pyx_doc_6mtrand_11RandomState_64weibull__pyx_doc_6mtrand_11RandomState_62pareto__pyx_doc_6mtrand_11RandomState_60vonmises__pyx_doc_6mtrand_11RandomState_58standard_t__pyx_doc_6mtrand_11RandomState_56standard_cauchy__pyx_doc_6mtrand_11RandomState_54noncentral_chisquare__pyx_doc_6mtrand_11RandomState_52chisquare__pyx_doc_6mtrand_11RandomState_50noncentral_f__pyx_doc_6mtrand_11RandomState_48f__pyx_doc_6mtrand_11RandomState_46gamma__pyx_doc_6mtrand_11RandomState_44standard_gamma__pyx_doc_6mtrand_11RandomState_42standard_exponential__pyx_doc_6mtrand_11RandomState_40exponential__pyx_doc_6mtrand_11RandomState_38beta__pyx_doc_6mtrand_11RandomState_36normal__pyx_doc_6mtrand_11RandomState_34standard_normal__pyx_doc_6mtrand_11RandomState_32random_integers__pyx_doc_6mtrand_11RandomState_30randn__pyx_doc_6mtrand_11RandomState_28rand__pyx_doc_6mtrand_11RandomState_26uniform__pyx_doc_6mtrand_11RandomState_24choice__pyx_doc_6mtrand_11RandomState_22bytes__pyx_doc_6mtrand_11RandomState_20randint__pyx_doc_6mtrand_11RandomState_18tomaxint__pyx_doc_6mtrand_11RandomState_16random_sample__pyx_doc_6mtrand_11RandomState_8set_state__pyx_doc_6mtrand_11RandomState_6get_state__pyx_doc_6mtrand_11RandomState_4seed__pyx_k_zipf_a_size_None_Draw_samples_f__pyx_k_zipf__pyx_k_zeros__pyx_k_weibull_a_size_None_Draw_sample__pyx_k_weibull__pyx_k_warnings__pyx_k_warn__pyx_k_wald_mean_scale_size_None_Draw__pyx_k_wald__pyx_k_vonmises_mu_kappa_size_None_Dra__pyx_k_vonmises__pyx_k_unsafe__pyx_k_unique__pyx_k_uniform_low_0_0_high_1_0_size_N__pyx_k_uniform__pyx_k_uint8__pyx_k_uint64__pyx_k_uint32__pyx_k_uint16__pyx_k_uint__pyx_k_triangular_left_mode_right_size__pyx_k_triangular__pyx_k_tomaxint_size_None_Random_integ__pyx_k_tol__pyx_k_threading__pyx_k_test__pyx_k_take__pyx_k_svd__pyx_k_sum_pvals_1_1_0__pyx_k_subtract__pyx_k_strides__pyx_k_state_must_be_624_longs__pyx_k_state__pyx_k_standard_t_df_size_None_Draw_sa__pyx_k_standard_t__pyx_k_standard_normal_size_None_Draw__pyx_k_standard_normal__pyx_k_standard_gamma_shape_size_None__pyx_k_standard_gamma__pyx_k_standard_exponential_size_None__pyx_k_standard_exponential__pyx_k_standard_cauchy_size_None_Draw__pyx_k_standard_cauchy__pyx_k_sqrt__pyx_k_sort__pyx_k_size_is_not_compatible_with_inpu__pyx_k_size__pyx_k_signbit__pyx_k_sigma_0_0__pyx_k_sigma_0__pyx_k_sigma__pyx_k_side__pyx_k_shuffle_x_Modify_a_sequence_in__pyx_k_shuffle__pyx_k_shape_from_size__pyx_k_shape_0__pyx_k_shape__pyx_k_set_state__pyx_k_seed__pyx_k_searchsorted__pyx_k_scale_0_2__pyx_k_scale_0_0_2__pyx_k_scale_0_0__pyx_k_scale_0__pyx_k_scale__pyx_k_safe__pyx_k_rtol__pyx_k_rngstate__pyx_k_rng__pyx_k_right__pyx_k_reversed__pyx_k_return_index__pyx_k_reshape__pyx_k_replace__pyx_k_reduce__pyx_k_rayleigh_scale_1_0_size_None_Dr__pyx_k_rayleigh__pyx_k_ravel__pyx_k_range__pyx_k_random_sample_size_None_Return__pyx_k_random_sample__pyx_k_random_integers_low_high_None_s__pyx_k_random_integers__pyx_k_random__pyx_k_randn_d0_d1_dn_Return_a_sample__pyx_k_randn__pyx_k_randint_type__pyx_k_randint_low_high_None_size_None__pyx_k_randint_helpers_pxi__pyx_k_randint__pyx_k_rand_uint8__pyx_k_rand_uint64__pyx_k_rand_uint32__pyx_k_rand_uint16__pyx_k_rand_int8__pyx_k_rand_int64__pyx_k_rand_int32__pyx_k_rand_int16__pyx_k_rand_d0_d1_dn_Random_values_in__pyx_k_rand_bool__pyx_k_rand_2__pyx_k_rand__pyx_k_raise__pyx_k_pyx_vtable__pyx_k_pvals__pyx_k_prod__pyx_k_probabilities_do_not_sum_to_1__pyx_k_probabilities_are_not_non_negati__pyx_k_power_a_size_None_Draws_samples__pyx_k_power__pyx_k_poisson_lam_max__pyx_k_poisson_lam_1_0_size_None_Draw__pyx_k_poisson__pyx_k_permutation_x_Randomly_permute__pyx_k_permutation__pyx_k_pareto_a_size_None_Draw_samples__pyx_k_pareto__pyx_k_p_must_be_1_dimensional__pyx_k_p_is_nan__pyx_k_p_1_0_2__pyx_k_p_1_0__pyx_k_p_1__pyx_k_p_0_0_2__pyx_k_p_0_0__pyx_k_p_0__pyx_k_p__pyx_k_out__pyx_k_operator__pyx_k_off__pyx_k_numpy_dual__pyx_k_numpy_core_multiarray_failed_to__pyx_k_numpy__pyx_k_nsample_1__pyx_k_nsample__pyx_k_np__pyx_k_normal_loc_0_0_scale_1_0_size_N__pyx_k_normal__pyx_k_noncentral_f_dfnum_dfden_nonc_s__pyx_k_noncentral_f__pyx_k_noncentral_chisquare_df_nonc_si__pyx_k_noncentral_chisquare__pyx_k_nonc_0__pyx_k_nonc__pyx_k_ngood_nbad_nsample__pyx_k_ngood_0__pyx_k_ngood__pyx_k_negative_binomial_n_p_size_None__pyx_k_negative_binomial__pyx_k_ndim__pyx_k_ndarray__pyx_k_nbad_0__pyx_k_nbad__pyx_k_name__pyx_k_n_0_2__pyx_k_n_0__pyx_k_n__pyx_k_multivariate_normal_mean_cov_si__pyx_k_multivariate_normal__pyx_k_multinomial_n_pvals_size_None_D__pyx_k_multinomial__pyx_k_mu__pyx_k_mtrand_pyx__pyx_k_mtrand__pyx_k_mode_right__pyx_k_mode__pyx_k_mean_must_be_1_dimensional__pyx_k_mean_and_cov_must_have_same_leng__pyx_k_mean_0_0__pyx_k_mean_0__pyx_k_mean__pyx_k_max__pyx_k_main__pyx_k_low_is_out_of_bounds_for_s__pyx_k_low_high__pyx_k_low__pyx_k_long__pyx_k_logseries_p_size_None_Draw_samp__pyx_k_logseries__pyx_k_lognormal_mean_0_0_sigma_1_0_si__pyx_k_lognormal__pyx_k_logistic_loc_0_0_scale_1_0_size__pyx_k_logistic__pyx_k_logical_or__pyx_k_loc__pyx_k_less_equal__pyx_k_less__pyx_k_left_right__pyx_k_left_mode__pyx_k_left__pyx_k_laplace_loc_0_0_scale_1_0_size__pyx_k_laplace__pyx_k_lam_value_too_large_2__pyx_k_lam_value_too_large__pyx_k_lam_0__pyx_k_lam__pyx_k_l__pyx_k_kappa_0__pyx_k_kappa__pyx_k_itemsize__pyx_k_item__pyx_k_issubdtype__pyx_k_isnan__pyx_k_isfinite__pyx_k_intp__pyx_k_integer__pyx_k_int8__pyx_k_int64__pyx_k_int32__pyx_k_int16__pyx_k_int__pyx_k_index__pyx_k_import__pyx_n_s_import__pyx_k_iinfo__pyx_k_ignore__pyx_k_hypergeometric_ngood_nbad_nsamp__pyx_k_hypergeometric__pyx_k_high_is_out_of_bounds_for_s__pyx_k_high__pyx_k_gumbel_loc_0_0_scale_1_0_size_N__pyx_k_gumbel__pyx_k_greater_equal__pyx_k_greater__pyx_k_get_state__pyx_k_geometric_p_size_None_Draw_samp__pyx_k_geometric__pyx_k_gamma_shape_scale_1_0_size_None__pyx_k_gamma__pyx_k_format__pyx_k_floating__pyx_k_float64__pyx_k_finfo__pyx_k_f_dfnum_dfden_size_None_Draw_sa__pyx_k_f__pyx_k_exponential__pyx_k_exit__pyx_k_equal__pyx_k_eps__pyx_k_enter__pyx_k_empty_like__pyx_k_empty__pyx_k_dummy_threading__pyx_k_dtype__pyx_k_dot__pyx_k_dirichlet_alpha_size_None_Draw__pyx_k_dirichlet__pyx_k_dfnum_1__pyx_k_dfnum_0__pyx_k_dfnum__pyx_k_dfden_0__pyx_k_dfden__pyx_k_df_0__pyx_k_df__pyx_k_data__pyx_k_d__pyx_k_cumsum__pyx_k_ctypes__pyx_k_covariance_is_not_positive_semid__pyx_k_cov_must_be_2_dimensional_and_sq__pyx_k_cov__pyx_k_count_nonzero__pyx_k_copy__pyx_k_cnt__pyx_k_cline_in_traceback__pyx_k_choice_a_size_None_replace_True__pyx_k_choice__pyx_k_chisquare_df_size_None_Draw_sam__pyx_k_chisquare__pyx_k_check_valid_must_equal_warn_rais__pyx_k_check_valid__pyx_k_casting__pyx_k_bytes_length_Return_random_byte__pyx_k_bytes__pyx_k_buf__pyx_k_broadcast__pyx_k_bool_2__pyx_k_bool__pyx_k_binomial_n_p_size_None_Draw_sam__pyx_k_binomial__pyx_k_beta__pyx_k_b_0__pyx_k_b__pyx_k_atol__pyx_k_astype__pyx_k_asarray__pyx_k_array_data__pyx_k_array__pyx_k_arange__pyx_k_any__pyx_k_alpha__pyx_k_allclose__pyx_k_all__pyx_k_algorithm_must_be_MT19937__pyx_k_add__pyx_k_a_must_be_non_empty__pyx_k_a_must_be_greater_than_0__pyx_k_a_must_be_1_dimensional_or_an_in__pyx_k_a_must_be_1_dimensional__pyx_k_a_and_p_must_have_same_size__pyx_k_a_1_0__pyx_k_a_0_2__pyx_k_a_0__pyx_k_a__pyx_k_ValueError__pyx_k_Unsupported_dtype_s_for_randint__pyx_k_TypeError__pyx_k_This_function_is_deprecated_Plea_2__pyx_k_This_function_is_deprecated_Plea__pyx_k_T__pyx_k_Seed_must_be_between_0_and_2_32__pyx_k_RuntimeWarning__pyx_k_Range_exceeds_valid_bounds__pyx_k_RandomState_zipf_line_3991__pyx_k_RandomState_weibull_line_2759__pyx_k_RandomState_wald_line_3505__pyx_k_RandomState_vonmises_line_2551__pyx_k_RandomState_uniform_line_1210__pyx_k_RandomState_triangular_line_3592__pyx_k_RandomState_tomaxint_line_858__pyx_k_RandomState_standard_t_line_2445__pyx_k_RandomState_standard_normal_line__pyx_k_RandomState_standard_gamma_line__pyx_k_RandomState_standard_exponential__pyx_k_RandomState_standard_cauchy_line__pyx_k_RandomState_shuffle_line_4759__pyx_k_RandomState_rayleigh_line_3426__pyx_k_RandomState_random_sample_line_8__pyx_k_RandomState_random_integers_line__pyx_k_RandomState_randn_line_1360__pyx_k_RandomState_randint_line_905__pyx_k_RandomState_rand_line_1316__pyx_k_RandomState_power_line_2869__pyx_k_RandomState_poisson_line_3903__pyx_k_RandomState_permutation_line_484__pyx_k_RandomState_pareto_line_2649__pyx_k_RandomState_normal_line_1547__pyx_k_RandomState_noncentral_f_line_20__pyx_k_RandomState_noncentral_chisquare__pyx_k_RandomState_negative_binomial_li__pyx_k_RandomState_multivariate_normal__pyx_k_RandomState_multinomial_line_453__pyx_k_RandomState_logseries_line_4272__pyx_k_RandomState_lognormal_line_3302__pyx_k_RandomState_logistic_line_3209__pyx_k_RandomState_laplace_line_2980__pyx_k_RandomState_hypergeometric_line__pyx_k_RandomState_gumbel_line_3078__pyx_k_RandomState_geometric_line_4082__pyx_k_RandomState_gamma_line_1896__pyx_k_RandomState_f_line_1992__pyx_k_RandomState_dirichlet_line_4643__pyx_k_RandomState_ctor__pyx_k_RandomState_choice_line_1028__pyx_k_RandomState_chisquare_line_2196__pyx_k_RandomState_bytes_line_999__pyx_k_RandomState_binomial_line_3686__pyx_k_OverflowError__pyx_k_MT19937__pyx_k_Lock__pyx_k_L__pyx_k_ImportError__pyx_k_Fewer_non_zero_entries_in_p_than__pyx_k_DeprecationWarning__pyx_k_Cannot_take_a_larger_sample_than__pyx_methods_6mtrand_RandomState__pyx_doc_6mtrand_16_rand_uint64__pyx_doc_6mtrand_14_rand_uint32__pyx_doc_6mtrand_12_rand_uint16__pyx_doc_6mtrand_10_rand_uint8__pyx_doc_6mtrand_8_rand_int64__pyx_doc_6mtrand_6_rand_int32__pyx_doc_6mtrand_4_rand_int16__pyx_doc_6mtrand_2_rand_int8__pyx_doc_6mtrand__rand_bool__pyx_methods__pyx_bisect_code_objects__Pyx_PyFunction_FastCallNoKwloggam.part.0__pyx_code_cacherk_uint64deregister_tm_clones__do_global_dtors_auxcompleted.0__do_global_dtors_aux_fini_array_entryframe_dummy__frame_dummy_init_array_entry__pyx_pyargnames.42__pyx_pyargnames.43__pyx_pyargnames.47__pyx_pyargnames.44__pyx_pyargnames.48__pyx_pyargnames.45__pyx_pyargnames.49__pyx_pyargnames.46__pyx_pyargnames.50__pyx_pyargnames.51__pyx_pyargnames.41__pyx_pyargnames.0__pyx_pyargnames.1__pyx_pyargnames.3__pyx_pf_6mtrand_11RandomState_24choice__pyx_pyargnames.4__pyx_pyargnames.2__pyx_pyargnames.5__pyx_pyargnames.9__pyx_pyargnames.6__pyx_pyargnames.8__pyx_pyargnames.7__pyx_pyargnames.10__pyx_pyargnames.12__pyx_pyargnames.11__pyx_pyargnames.13__pyx_pyargnames.14__pyx_pyargnames.15__pyx_pyargnames.16__pyx_pyargnames.19__pyx_pyargnames.17__pyx_pyargnames.18__pyx_pyargnames.20__pyx_pyargnames.21__pyx_pyargnames.22__pyx_pyargnames.23__pyx_pyargnames.24__pyx_pyargnames.25__pyx_pyargnames.26__pyx_pyargnames.27__pyx_pyargnames.28__pyx_pyargnames.29__pyx_pyargnames.30__pyx_pyargnames.31__pyx_pf_6mtrand_11RandomState_84negative_binomial__pyx_pyargnames.32__pyx_pyargnames.33__pyx_pyargnames.34__pyx_pyargnames.35__pyx_pyargnames.36__pyx_pyargnames.37__pyx_pyargnames.38__pyx_pyargnames.39__pyx_pyargnames.40__FRAME_END____Pyx_PyCFunction_FastCall.lto_priv.0__pyx_n_s_mu.lto_priv.0__pyx_tuple__113.lto_priv.0__pyx_tuple__12.lto_priv.0__pyx_tuple__9.lto_priv.0__pyx_tuple__162.lto_priv.0__pyx_pf_6mtrand_11RandomState_92hypergeometric.lto_priv.0__pyx_n_s_poisson_lam_max.lto_priv.0__pyx_tuple__96.lto_priv.0__pyx_n_s_bool.lto_priv.0__pyx_pw_6mtrand_11RandomState_59standard_t.lto_priv.0__pyx_n_s_raise.lto_priv.0__pyx_b.lto_priv.0__pyx_tuple__111.lto_priv.0__pyx_n_s_format.lto_priv.0__Pyx_GetItemInt_Fast.constprop.0__pyx_tuple__56.lto_priv.0__pyx_tuple__146.lto_priv.0__Pyx_PyInt_As_int.lto_priv.0__pyx_n_s_take.lto_priv.0__Pyx_PyInt_As_npy_int8.lto_priv.0__pyx_tuple__78.lto_priv.0__pyx_n_s_zeros.lto_priv.0__pyx_tuple__4.lto_priv.0__Pyx_Raise.constprop.0__pyx_f_6mtrand_discnmN_array.constprop.0__pyx_n_s_df.lto_priv.0__pyx_tuple__55.lto_priv.0__pyx_n_s_uint8.lto_priv.0__pyx_tuple__101.lto_priv.0__pyx_n_s_equal.lto_priv.0__pyx_tuple__80.lto_priv.0__pyx_tuple__35.lto_priv.0__pyx_pw_6mtrand_11RandomState_93hypergeometric.lto_priv.0__pyx_n_s_uint64.lto_priv.0__pyx_f_6mtrand_discd_array_sc.lto_priv.0__pyx_empty_tuple.lto_priv.0__pyx_int_1.lto_priv.0__pyx_tuple__99.lto_priv.0__pyx_pw_6mtrand_11RandomState_39beta.lto_priv.0__pyx_tuple__95.lto_priv.0__pyx_n_s_int32.lto_priv.0__pyx_n_s_MT19937.lto_priv.0__pyx_n_s_atol.lto_priv.0__pyx_tuple__52.lto_priv.0__pyx_tuple__61.lto_priv.0__pyx_tuple__122.lto_priv.0__pyx_tuple__161.lto_priv.0__pyx_tuple__93.lto_priv.0__pyx_tuple__77.lto_priv.0__pyx_tuple__13.lto_priv.0__pyx_tuple__168.lto_priv.0__pyx_tuple__16.lto_priv.0__pyx_tuple__74.lto_priv.0__pyx_n_s_p.lto_priv.0__pyx_n_s_svd.lto_priv.0__pyx_tuple__60.lto_priv.0__Pyx_PyInt_As_npy_uint32.lto_priv.0__pyx_pw_6mtrand_19_shape_from_size.lto_priv.0__pyx_pw_6mtrand_11RandomState_19tomaxint.lto_priv.0__pyx_pw_6mtrand_11RandomState_71gumbel.lto_priv.0__pyx_tuple__19.lto_priv.0__pyx_n_s_side.lto_priv.0__pyx_tuple__2.lto_priv.0__Pyx_Import.constprop.0__Pyx_PyInt_As_npy_intp.part.0.lto_priv.0__pyx_tuple__158.lto_priv.0__pyx_builtin_ValueError.lto_priv.0__pyx_n_s_Lock.lto_priv.0__pyx_pw_6mtrand_9_rand_int64.lto_priv.0__pyx_tuple__156.lto_priv.0__pyx_pw_6mtrand_11RandomState_75lognormal.lto_priv.0__pyx_tuple__65.lto_priv.0__pyx_n_s_dfden.lto_priv.0__pyx_tuple__140.lto_priv.0__pyx_tuple__105.lto_priv.0__pyx_f_6mtrand_cont2_array_sc.lto_priv.0__pyx_tuple__7.lto_priv.0__pyx_float_0_0.lto_priv.0__pyx_tuple__5.lto_priv.0__pyx_n_s_isfinite.lto_priv.0__pyx_n_s_bool_2.lto_priv.0__pyx_pw_6mtrand_11RandomState_65weibull.lto_priv.0__pyx_tuple__138.lto_priv.0__pyx_n_s_return_index.lto_priv.0__pyx_tuple__23.lto_priv.0__pyx_pw_6mtrand_11RandomState_9set_state.lto_priv.0__Pyx_IternextUnpackEndCheck.lto_priv.0__pyx_n_s_long.lto_priv.0__pyx_n_s_loc.lto_priv.0__pyx_tuple__47.lto_priv.0__pyx_int_0.lto_priv.0__pyx_tuple__92.lto_priv.0__pyx_builtin_OverflowError.lto_priv.0__pyx_tuple__43.lto_priv.0__pyx_tuple__46.lto_priv.0__pyx_tuple__31.lto_priv.0__pyx_n_s_get_state.lto_priv.0__pyx_tuple__104.lto_priv.0__pyx_tuple__20.lto_priv.0__pyx_kp_s_Unsupported_dtype_s_for_randint.lto_priv.0__pyx_pw_6mtrand_7_rand_int32.lto_priv.0__pyx_n_s_kappa.lto_priv.0__pyx_pw_6mtrand_17_rand_uint64.lto_priv.0__pyx_tuple__123.lto_priv.0__pyx_n_s_greater.lto_priv.0__pyx_pw_6mtrand_11RandomState_91geometric.lto_priv.0__pyx_tuple__69.lto_priv.0__pyx_tuple__24.lto_priv.0_fini__pyx_pw_6mtrand_3_rand_int8.lto_priv.0__pyx_f_6mtrand_cont0_array.lto_priv.0__pyx_builtin_TypeError.lto_priv.0__pyx_n_s_random_sample.lto_priv.0__pyx_tuple__89.lto_priv.0__pyx_n_s_greater_equal.lto_priv.0__pyx_pf_6mtrand_11RandomState_82binomial.lto_priv.0__pyx_pw_6mtrand_11_rand_uint8.lto_priv.0__pyx_n_s_int8.lto_priv.0__pyx_n_s_T.lto_priv.0__pyx_pw_6mtrand_11RandomState_47gamma.lto_priv.0__pyx_n_s_dtype.lto_priv.0__pyx_tuple__40.lto_priv.0__pyx_n_s_replace.lto_priv.0__pyx_n_s_nbad.lto_priv.0__Pyx_PyInt_As_npy_bool.lto_priv.0__pyx_tuple__100.lto_priv.0__pyx_m.lto_priv.0__pyx_pw_6mtrand_11RandomState_21randint.lto_priv.0__pyx_n_s_eps.lto_priv.0__pyx_n_s_dot.lto_priv.0__pyx_tuple__85.lto_priv.0__pyx_tuple__59.lto_priv.0__pyx_n_s_signbit.lto_priv.0__pyx_tuple__132.lto_priv.0__pyx_n_s_rtol.lto_priv.0__pyx_n_s_size.lto_priv.0__pyx_tuple__3.lto_priv.0__pyx_n_s_uint16.lto_priv.0__pyx_n_s_float64.lto_priv.0__pyx_pw_6mtrand_11RandomState_79wald.lto_priv.0__pyx_tuple__53.lto_priv.0__Pyx_PyInt_As_npy_int64.lto_priv.0__pyx_n_s_less.lto_priv.0__Pyx_PyInt_As_npy_uint64.lto_priv.0__pyx_pw_6mtrand_11RandomState_5seed.lto_priv.0__pyx_lineno.lto_priv.0__pyx_tuple__38.lto_priv.0__pyx_pw_6mtrand_11RandomState_89zipf.lto_priv.0__Pyx_GetModuleGlobalName.lto_priv.0__pyx_n_s_name.lto_priv.0__pyx_int_neg_1.lto_priv.0__pyx_f_6mtrand_cont2_array.lto_priv.0__pyx_n_s_casting.lto_priv.0__pyx_tuple__118.lto_priv.0__pyx_kp_s_low_is_out_of_bounds_for_s.lto_priv.0__pyx_f_6mtrand_discd_array.lto_priv.0__pyx_tuple__37.lto_priv.0__pyx_pw_6mtrand_11RandomState_43standard_exponential.lto_priv.0__pyx_tuple__81.lto_priv.0__pyx_tuple__117.lto_priv.0__pyx_n_s_int16.lto_priv.0__pyx_tuple__82.lto_priv.0__pyx_tuple__173.lto_priv.0__pyx_tuple__151.lto_priv.0__pyx_tuple__136.lto_priv.0__Pyx_PyInt_As_npy_uint16.lto_priv.0__pyx_tuple__137.lto_priv.0__Pyx_PyInt_As_npy_int16.lto_priv.0__pyx_n_s_reshape.lto_priv.0__pyx_n_s_finfo.lto_priv.0__pyx_n_s_broadcast.lto_priv.0__pyx_tuple__119.lto_priv.0__pyx_pw_6mtrand_11RandomState_55noncentral_chisquare.lto_priv.0__pyx_tuple__128.lto_priv.0__pyx_pw_6mtrand_5_rand_int16.lto_priv.0__pyx_n_s_less_equal.lto_priv.0__pyx_tuple__167.lto_priv.0__pyx_tuple__58.lto_priv.0__pyx_n_s_l.lto_priv.0__pyx_tuple__15.lto_priv.0__pyx_pw_6mtrand_11RandomState_29rand.lto_priv.0__pyx_tuple__70.lto_priv.0__pyx_pw_6mtrand_11RandomState_37normal.lto_priv.0__pyx_pw_6mtrand_11RandomState_15__reduce__.lto_priv.0__pyx_n_s_reduce.lto_priv.0__pyx_tuple__41.lto_priv.0__pyx_tuple__68.lto_priv.0__pyx_tuple__145.lto_priv.0__pyx_n_s_operator.lto_priv.0__pyx_tuple__150.lto_priv.0__pyx_tuple__172.lto_priv.0__pyx_tuple__135.lto_priv.0__pyx_n_s_sqrt.lto_priv.0__pyx_n_s_standard_normal.lto_priv.0__pyx_tuple__125.lto_priv.0__pyx_n_s_empty.lto_priv.0__pyx_n_s_shape.lto_priv.0__pyx_n_s_sigma.lto_priv.0__pyx_tuple__83.lto_priv.0__pyx_pw_6mtrand_11RandomState_11__getstate__.lto_priv.0__pyx_tuple__152.lto_priv.0__pyx_n_s_seed.lto_priv.0__pyx_tuple__63.lto_priv.0__pyx_tuple__27.lto_priv.0__Pyx_PyInt_As_npy_int32.lto_priv.0__pyx_kp_s_This_function_is_deprecated_Plea.lto_priv.0__pyx_pw_6mtrand_11RandomState_85negative_binomial.lto_priv.0__pyx_tuple__114.lto_priv.0__Pyx_PyNumber_IntOrLongWrongResultType.lto_priv.0__pyx_tuple__171.lto_priv.0__pyx_tuple__134.lto_priv.0__pyx_pw_6mtrand_11RandomState_99multinomial.lto_priv.0__pyx_pw_6mtrand_11RandomState_31randn.lto_priv.0__pyx_tuple__79.lto_priv.0__Pyx_PyInt_As_npy_uint8.lto_priv.0__pyx_int_4294967295.lto_priv.0__pyx_tuple__22.lto_priv.0__pyx_n_s_randint_type.lto_priv.0__pyx_n_s_asarray.lto_priv.0__pyx_n_s_random.lto_priv.0__pyx_tuple__54.lto_priv.0__pyx_n_s_numpy_dual.lto_priv.0__pyx_tuple__98.lto_priv.0__pyx_n_s_check_valid.lto_priv.0__pyx_tuple__17.lto_priv.0__pyx_clineno.lto_priv.0__pyx_n_s_set_state.lto_priv.0__pyx_tuple__66.lto_priv.0__pyx_pw_6mtrand_11RandomState_25choice.lto_priv.0__pyx_n_s_warnings.lto_priv.0__pyx_tuple__14.lto_priv.0__dso_handle__pyx_tuple__131.lto_priv.0__pyx_pw_6mtrand_11RandomState_83binomial.lto_priv.0__Pyx_PyObject_GetAttrStr.lto_priv.0__pyx_tuple__28.lto_priv.0__pyx_tuple__67.lto_priv.0__pyx_builtin_DeprecationWarning.lto_priv.0__pyx_tuple__51.lto_priv.0__pyx_pw_6mtrand_11RandomState_23bytes.lto_priv.0__Pyx__ExceptionReset.lto_priv.0__Pyx_PyInt_As_size_t.lto_priv.0__pyx_tuple__57.lto_priv.0__pyx_tuple__71.lto_priv.0__pyx_n_s_uint32.lto_priv.0__pyx_tuple__155.lto_priv.0__Pyx_RaiseNeedMoreValuesError.lto_priv.0__pyx_n_s_add.lto_priv.0__pyx_tuple__94.lto_priv.0__pyx_pw_6mtrand_13_rand_uint16.lto_priv.0__Pyx_PyObject_IsTrue.lto_priv.0__pyx_tuple__97.lto_priv.0__pyx_n_s_allclose.lto_priv.0__pyx_pw_6mtrand_11RandomState_51noncentral_f.lto_priv.0__pyx_pw_6mtrand_11RandomState_87poisson.lto_priv.0__pyx_pw_6mtrand_11RandomState_63pareto.lto_priv.0__pyx_tuple__112.lto_priv.0__pyx_pw_6mtrand_11RandomState_13__setstate__.lto_priv.0__pyx_pw_6mtrand_11RandomState_41exponential.lto_priv.0__pyx_n_s_unique.lto_priv.0__pyx_n_s_exit.lto_priv.0__pyx_n_s_int64.lto_priv.0__pyx_n_s_low.lto_priv.0__pyx_n_s_floating.lto_priv.0__pyx_tuple__153.lto_priv.0__pyx_n_s_left.lto_priv.0__pyx_pw_6mtrand_11RandomState_95logseries.lto_priv.0__pyx_pw_6mtrand_11RandomState_35standard_normal.lto_priv.0__pyx_f.lto_priv.0__pyx_n_s_permutation.lto_priv.0__pyx_tuple__110.lto_priv.0__pyx_pw_6mtrand_11RandomState_77rayleigh.lto_priv.0__pyx_f_6mtrand_discdd_array.constprop.0__pyx_tuple__21.lto_priv.0__pyx_n_s_safe.lto_priv.0__pyx_n_s_ravel.lto_priv.0__Pyx_PyUnicode_Equals.lto_priv.0__Pyx_PyFunction_FastCallDict.constprop.0__Pyx_ParseOptionalKeywords.constprop.0__Pyx_PyObject_CallOneArg.lto_priv.0__pyx_tuple__87.lto_priv.0__pyx_tuple__11.lto_priv.0__pyx_n_s_ndim.lto_priv.0__Pyx_GetBuiltinName.lto_priv.0__pyx_tuple__130.lto_priv.0__pyx_tuple__115.lto_priv.0__pyx_tuple__84.lto_priv.0__Pyx_SetItemInt_Fast.constprop.0__pyx_float_1_0.lto_priv.0__Pyx_PyInt_As_unsigned_long.lto_priv.0__pyx_tuple__149.lto_priv.0__Pyx__GetException.lto_priv.0__Pyx_ErrRestoreInState.lto_priv.0__pyx_f_6mtrand_cont1_array.lto_priv.0__pyx_pw_6mtrand_11RandomState_101dirichlet.lto_priv.0__pyx_tuple__116.lto_priv.0__pyx_tuple__36.lto_priv.0__pyx_n_s_ignore.lto_priv.0__pyx_n_s_lam.lto_priv.0__pyx_tuple__6.lto_priv.0__Pyx_PyObject_CallNoArg.lto_priv.0__pyx_pw_6mtrand_11RandomState_7get_state.lto_priv.0__pyx_filename.lto_priv.0__pyx_n_s_issubdtype.lto_priv.0__pyx_tuple__32.lto_priv.0__pyx_n_s_logical_or.lto_priv.0__pyx_n_s_any.lto_priv.0__pyx_pw_6mtrand_11RandomState_73logistic.lto_priv.0__pyx_tuple__26.lto_priv.0__pyx_pw_6mtrand_15_rand_uint32.lto_priv.0__pyx_pw_6mtrand_11RandomState_27uniform.lto_priv.0__pyx_n_s_enter.lto_priv.0__Pyx_IterFinish.lto_priv.0__Pyx_CheckKeywordStrings.constprop.0__pyx_n_s_prod.lto_priv.0__pyx_n_s_ndarray.lto_priv.0__pyx_tuple__159.lto_priv.0__Pyx_PyInt_EqObjC.constprop.0__pyx_n_s_pvals.lto_priv.0__pyx_n_s_nonc.lto_priv.0__pyx_n_s_RandomState_ctor.lto_priv.0__pyx_tuple__42.lto_priv.0__pyx_n_s_array.lto_priv.0__pyx_empty_bytes.lto_priv.0__pyx_n_s_nsample.lto_priv.0__pyx_tuple__170.lto_priv.0__pyx_tuple__133.lto_priv.0__pyx_f_6mtrand_cont3_array_sc.lto_priv.0__pyx_n_s_uint.lto_priv.0__pyx_n_s_shape_from_size.lto_priv.0__pyx_tuple__148.lto_priv.0__pyx_tuple__147.lto_priv.0__pyx_tuple__142.lto_priv.0__pyx_tuple__107.lto_priv.0__pyx_n_s_cline_in_traceback.lto_priv.0__pyx_n_s_isnan.lto_priv.0__Pyx__ExceptionSave.lto_priv.0__pyx_tuple__91.lto_priv.0__pyx_n_s_integer.lto_priv.0__pyx_n_s_a.lto_priv.0_DYNAMIC__pyx_n_s_astype.lto_priv.0__pyx_tuple__76.lto_priv.0__pyx_n_s_randint.lto_priv.0__pyx_n_s_index.lto_priv.0__pyx_n_s_rand.lto_priv.0__pyx_n_s_all.lto_priv.0__Pyx_ImportType.constprop.0__pyx_tuple__127.lto_priv.0__pyx_tuple__139.lto_priv.0__pyx_n_s_b.lto_priv.0__pyx_tuple__154.lto_priv.0__pyx_pw_6mtrand_11RandomState_67power.lto_priv.0__pyx_tuple__49.lto_priv.0__pyx_slice__44.lto_priv.0__pyx_n_s_mean.lto_priv.0__pyx_tuple__86.lto_priv.0__pyx_tuple__164.lto_priv.0__pyx_kp_s_This_function_is_deprecated_Plea_2.lto_priv.0__pyx_tuple__10.lto_priv.0__pyx_tuple__90.lto_priv.0__pyx_float_1eneg_8.lto_priv.0__pyx_pw_6mtrand_11RandomState_45standard_gamma.lto_priv.0__pyx_tuple__30.lto_priv.0__pyx_n_s_count_nonzero.lto_priv.0__pyx_slice__45.lto_priv.0__pyx_kp_s_high_is_out_of_bounds_for_s.lto_priv.0__pyx_f_6mtrand_discnp_array.constprop.0__pyx_n_s_subtract.lto_priv.0PyArray_API.lto_priv.0__pyx_n_s_alpha.lto_priv.0__pyx_int_624.lto_priv.0__pyx_tuple__73.lto_priv.0__pyx_tuple__144.lto_priv.0__pyx_tuple__109.lto_priv.0__pyx_pw_6mtrand_11RandomState_53chisquare.lto_priv.0__pyx_tuple__166.lto_priv.0__pyx_tuple__141.lto_priv.0__pyx_tuple__106.lto_priv.0__pyx_pw_6mtrand_11RandomState_1__init__.lto_priv.0__pyx_tuple__124.lto_priv.0__pyx_n_s_sort.lto_priv.0__GNU_EH_FRAME_HDR__Pyx_PyObject_GetIndex.lto_priv.0__pyx_n_s_right.lto_priv.0__pyx_n_s_tol.lto_priv.0__pyx_tuple__62.lto_priv.0__pyx_n_s_item.lto_priv.0__TMC_END___GLOBAL_OFFSET_TABLE___pyx_tuple__163.lto_priv.0__pyx_n_s_int.lto_priv.0__Pyx_AddTraceback.lto_priv.0__pyx_tuple__129.lto_priv.0__pyx_pw_6mtrand_11RandomState_61vonmises.lto_priv.0__pyx_tuple__18.lto_priv.0__pyx_tuple__120.lto_priv.0__Pyx_ImportFrom.lto_priv.0__pyx_pw_6mtrand_11RandomState_57standard_cauchy.lto_priv.0__pyx_n_s_unsafe.lto_priv.0__pyx_n_s_np.lto_priv.0__pyx_tuple__72.lto_priv.0__pyx_tuple__8.lto_priv.0__pyx_tuple__39.lto_priv.0__pyx_k__48.lto_priv.0__pyx_tuple__33.lto_priv.0__pyx_tuple__126.lto_priv.0__pyx_n_s_cumsum.lto_priv.0__pyx_tuple__157.lto_priv.0__pyx_tuple__143.lto_priv.0__pyx_tuple__108.lto_priv.0__pyx_d.lto_priv.0__pyx_pw_6mtrand_11RandomState_81triangular.lto_priv.0__Pyx_PyObject_Call.lto_priv.0__pyx_n_s_high.lto_priv.0__pyx_f_6mtrand_cont1_array_sc.lto_priv.0__pyx_tuple__25.lto_priv.0__pyx_n_s_intp.lto_priv.0__pyx_n_s_d.lto_priv.0__pyx_n_s_scale.lto_priv.0__pyx_tuple__103.lto_priv.0__pyx_slice__165.lto_priv.0__pyx_tuple__102.lto_priv.0__pyx_n_s_n.lto_priv.0__pyx_n_s_dfnum.lto_priv.0__pyx_pw_6mtrand_11RandomState_69laplace.lto_priv.0__pyx_n_s_ngood.lto_priv.0__pyx_n_s_searchsorted.lto_priv.0__Pyx_PyErr_ExceptionMatchesInState.isra.0__pyx_tuple__34.lto_priv.0__pyx_n_s_copy.lto_priv.0__pyx_pw_6mtrand_11RandomState_97multivariate_normal.lto_priv.0_init__pyx_pw_6mtrand_11RandomState_49f.lto_priv.0__pyx_tuple__64.lto_priv.0__pyx_pw_6mtrand_11RandomState_17random_sample.lto_priv.0__pyx_n_s_cov.lto_priv.0__pyx_tuple__88.lto_priv.0__pyx_tuple__121.lto_priv.0__pyx_tuple__160.lto_priv.0__pyx_f_6mtrand_cont3_array.lto_priv.0__pyx_pw_6mtrand_1_rand_bool.lto_priv.0__pyx_ptype_6mtrand_ndarray.lto_priv.0__pyx_cython_runtime.lto_priv.0__pyx_pw_6mtrand_11RandomState_33random_integers.lto_priv.0__pyx_n_s_mode.lto_priv.0__pyx_n_s_warn.lto_priv.0__pyx_tuple__75.lto_priv.0__pyx_tuple__29.lto_priv.0__pyx_n_s_rngstate.lto_priv.0PyUnicode_FromFormatPyObject_SetItemrk_logseriesPyList_New_PyUnicode_Readyrk_fPyExc_SystemErrorPyDict_SetItemStringrk_triangularPyDict_Sizerk_strerrorrk_rayleighrk_lognormalrk_logisticrk_randomseedPyException_SetTracebackPyMethod_Type__pyx_module_is_main_mtrand_ITM_deregisterTMCloneTablePyGILState_ReleasePyFloat_TypePyTuple_TypePyErr_RestorePyList_AsTuple_PyThreadState_UncheckedGetPyEval_RestoreThreadrk_poisson_multrk_noncentral_chisquarePyFrame_NewPyMem_FreePyCFunction_NewExPyNumber_InPlaceAddfread@GLIBC_2.2.5rk_random_uint8PyNumber_AddPyObject_GetAttrStringrk_normalrk_hypergeometric_hruaPyImport_AddModulePyBytes_FromStringAndSizerk_gammark_poisson_ptrsgetpid@GLIBC_2.2.5PyObject_SetAttrStringPyErr_WarnExrk_poissonrk_zipfclock@GLIBC_2.2.5PyErr_SetObjectrk_fillPyErr_NormalizeExceptionrk_weibullfclose@GLIBC_2.2.5PyNumber_Multiplyrk_random_uint16rk_waldPyObject_RichComparePyCode_Newrk_altfillPyImport_GetModuleDict__stack_chk_fail@GLIBC_2.4rk_powerPyExc_RuntimeErrorPyNumber_Longrk_standard_gammaPyErr_GivenExceptionMatchesPyErr_SetStringPyObject_IsInstancefmodrk_negative_binomialPyObject_GetItemPyExc_ExceptionPyExc_ValueErrorrk_standard_trk_standard_exponentialPyExc_DeprecationWarningPyExc_TypeErrorgettimeofday@GLIBC_2.2.5PyGILState_EnsurePyEval_EvalFrameExrk_hypergeometricrk_randomacosPySequence_ContainsPyErr_Printmemset@GLIBC_2.2.5PyMem_ReallocPyErr_ExceptionMatchesrk_random_uint64rk_gaussrk_hypergeometric_hyprk_geometricPyOS_snprintfPyTraceBack_HerePyObject_CallFinalizerFromDeallocPyObject_NotPyNumber_InPlaceTrueDividePyLong_FromSsize_tPyFloat_FromDoublePyType_Readyrk_longPyLong_FromLongmemcmp@GLIBC_2.2.5rk_gumbelPyLong_AsSsize_trk_uniformrk_vonmisesPyErr_ClearPyList_Append_Py_CheckRecursiveCallrk_random_uint32_Py_CheckRecursionLimitrk_intervalPyNumber_Orrk_binomial_inversion_Py_FalseStruct__gmon_start__PyUnicode_AsUnicodefopen64@GLIBC_2.2.5PyTuple_NewPyThreadState_GetPyExc_OverflowErrormemcpy@GLIBC_2.14rk_devfillrk_laplacePyType_Modifiedrk_binomialPyObject_SetAttrrk_betaPyErr_OccurredPyModule_Create2_Py_EllipsisObjectPyLong_AsLongPyImport_ImportModule_PyDict_GetItem_KnownHashrk_geometric_inversionPyDict_GetItemStringrk_ulongPyEval_EvalCodeExinit_by_arrayPyObject_Size_Py_NoneStructPyExc_ZeroDivisionErrorexpPyFloat_AsDoublePyObject_IsTrue_PyType_LookupPyImport_ImportModuleLevelObjectrk_chisquarePyObject_HashPyUnicode_Comparerk_binomial_btpePyInit_mtrandrk_pareto_Py_TrueStructPyFunction_TypePyDict_NewPyExc_IndexErrorPyLong_AsUnsignedLongPyDict_NextPyBaseObject_TypePyLong_FromUnsignedLongPyLong_TypePyCapsule_Typerk_double_PyObject_GetDictPtrPyErr_FetchPyUnicode_FromStringlogPyObject_GetIterPyEval_SaveThreadPyUnicode_InternFromStringpowPyExc_ImportErrorPyDict_SetItemPySequence_TuplePyExc_AttributeErrorrk_noncentral_fPyExc_StopIterationPySequence_ListPyObject_Callrk_random_boolPyUnicode_TypePyCapsule_Newrk_seedPyUnicode_DecodePyErr_FormatPyCapsule_GetPointerPySlice_NewPyExc_NameErrorPyUnicode_FromStringAndSizePyModule_GetDict_ITM_registerTMCloneTablePyNumber_IndexPyObject_GetAttrPyCFunction_Type_PyDict_NewPresizedPyUnicode_FormatPyLong_FromStringrk_geometric_searchPyMem_MallocPyErr_WarnFormat__cxa_finalize@GLIBC_2.2.5PyNumber_Subtractrk_standard_cauchyrk_exponentialPyTuple_PackPy_GetVersionPyObject_GC_UnTrackPyList_TypePyImport_Import.symtab.strtab.shstrtab.note.gnu.property.note.gnu.build-id.gnu.hash.dynsym.dynstr.gnu.version.gnu.version_r.rela.dyn.rela.plt.init.plt.sec.text.fini.rodata.eh_frame_hdr.eh_frame.init_array.fini_array.data.rel.ro.dynamic.got.data.bss.comment.annobin.notes.gnu.build.attributes .$AoK S [oX&X&ho((@wH(H(jBPP  ͵ y y   d d j j 7   @    @8 @@p 0@. 0n*  Ȉ rw2