ELF>@g@8 @p p ئ ئ+ئ+AHT  ++888$$P P P StdP P P Ptdc c c  QtdRtdئ ئ+ئ+( ( GNUq6nB:4% B `"p @H 0 ,@@D:`@nIHdC1 (0b8Kk}LJ f|8\5M3ndy'IK|U=vgl #2ח*X3I#K霳N yME>=g!ni焽dMD.ۗqX|EڋFy$vWJ@IoS:rKffBE,_̩*1޺KFmK4bs\rb(zK|M&E[b-G  $Z?)[Oi 0 {  J  % Q  6 p  2 1 2v= g  {hNO\ 6 z ( Bs( =x ]{ 2 N>U{ L dx  (Cb ] w lf  `\ ,  @ F" nN /C T m  @W   `Q 9  s W  g \W  O O  Y WW H  i -  I ~T  q   W " K  T /g J  t `@  S͇  F 2  S ,  M   l   E W=  PT x } k `- P IF `G a  p A0  X   pT ,  u   pM ~  @o +  { i p-u  - pL R  k [ pk *  t 5 @q ^  `g |% H  pt  k 3 r "  k b p-  o C -p K   pJ +  `Y L  q  po *w  o q.  { m  pp V  j ! L  0d :Q pf __gmon_start___ITM_deregisterTMCloneTable_ITM_registerTMCloneTable__cxa_finalize_Py_NoneStructPyBaseObject_TypePyExc_DeprecationWarningPyErr_WarnFormatPyExc_TypeErrorPyErr_FormatPyLong_TypePyErr_OccurredPyErr_SetString_PyType_LookupPyExc_AttributeErrorPyErr_SetObjectrk_intervalmemcpyPyObject_GC_UnTrackPyErr_FetchPyMem_FreePyErr_RestorePyObject_CallFinalizerFromDealloc__stack_chk_failPyObject_GetAttrPyErr_NormalizeExceptionPyException_SetTracebackPyThreadState_Get_Py_CheckRecursionLimitPyObject_Call_Py_CheckRecursiveCallPyExc_SystemError_PyThreadState_UncheckedGetPyFrame_NewPyEval_EvalFrameEx_Py_FalseStructPyFloat_Type_Py_TrueStructPyObject_RichComparePyUnicode_TypememcmpPyObject_IsTrue_PyUnicode_ReadyPyUnicode_FromStringPyImport_ImportPyOS_snprintfPyErr_WarnExPyExc_ValueErrorPyList_TypePyLong_FromSsize_tPyObject_SetItemPyExc_OverflowErrorPyErr_ExceptionMatchesPyErr_ClearPyModule_GetDictPyDict_NewPyImport_ImportModuleLevelObjectPyList_NewPyDict_NextPyUnicode_AsUnicodePyUnicode_ComparePyTuple_TypePyObject_GetItemPyTuple_NewPyEval_EvalCodeExPySlice_NewPyExc_StopIterationPyErr_GivenExceptionMatchesPyExc_NameError_PyDict_GetItem_KnownHashPyExc_ImportErrorPyLong_AsLongPyNumber_IndexPyLong_AsSsize_tPyExc_IndexErrorPyTraceBack_Here_PyObject_GetDictPtrPyObject_SetAttrPyObject_NotPyUnicode_FromFormatPyCode_NewPyMem_ReallocPyMem_MallocPyFunction_TypePyCFunction_TypePyLong_AsUnsignedLongPyMethod_TypePyDict_SizePyCapsule_NewPyBytes_FromStringAndSizePyEval_SaveThreadrk_fillPyEval_RestoreThreadPyObject_SizePyFloat_AsDoublePyObject_GetIterPyTuple_PackPyNumber_AddPySequence_TuplePyDict_SetItemPyObject_IsInstancePyFloat_FromDoublePyNumber_InPlaceTrueDividePyNumber_SubtractPyNumber_InPlaceAddPyCapsule_GetPointerrk_random_uint16PyLong_FromLongrk_random_uint32rk_random_uint8rk_random_uint64PyLong_FromUnsignedLongPySequence_ListPyNumber_MultiplyPyList_AsTuplePyList_Appendrk_standard_gammaPyGILState_EnsurePyExc_ZeroDivisionErrorPyGILState_ReleasePySequence_ContainsPyNumber_LongPyUnicode_Formatrk_binomial_Py_EllipsisObjectrk_longrk_logseriesrk_geometricrk_poissonrk_negative_binomialrk_zipfrk_standard_cauchyrk_standard_exponentialrk_gaussrk_doublerk_rayleighrk_powerrk_weibullrk_paretork_standard_trk_chisquarerk_exponentialrk_waldrk_lognormalrk_logisticrk_gumbelrk_laplacerk_vonmisesrk_noncentral_chisquarerk_frk_gammark_betark_normalrk_uniformrk_triangularrk_noncentral_frk_random_boolrk_randomseedrk_seedPyNumber_Orinit_by_arrayrk_hypergeometricPyInit_mtrandPy_GetVersionPyUnicode_FromStringAndSizePyModule_Create2PyImport_AddModulePyObject_SetAttrStringPyUnicode_InternFromStringPyObject_HashPyUnicode_DecodePyLong_FromString__pyx_module_is_main_mtrandPyImport_GetModuleDictPyDict_GetItemStringPyDict_SetItemStringPyType_ReadyPyCFunction_NewExPyImport_ImportModulePyObject_GetAttrStringPyCapsule_TypePyExc_RuntimeErrorPyExc_ExceptionPyErr_Print_PyDict_NewPresizedPyType_Modifiedrk_randomrk_ulongrk_devfillfopen64freadfclosegettimeofdaygetpidclockrk_altfilllogsqrtrk_strerrorpowexprk_binomial_btpefloorrk_binomial_inversionrk_poisson_multrk_poisson_ptrsacosfmodrk_geometric_searchrk_geometric_inversionceilrk_hypergeometric_hyprk_hypergeometric_hrualibpython3.7m.so.1.0libpthread.so.0libc.so.6_edata__bss_start_endGLIBC_2.14GLIBC_2.4GLIBC_2.2.5/opt/alt/python37/lib64X z ii  ui  ئ+>+>++`--p-----------------p--- --(--0--@--H--P--`-----(----------(-Ȟ-------- --(-(-0--@--H--P--`--h--p--x--------ȟ--П-------- -(-(-P-0--@--H--`--h---------(-----Ƞ-(-Р-----(----p----- -(-(--@-p-H-(-P--`--h-H-p-P-x----------ȡ--С---8------- --(--@--H--P-P-X--p--x---p--8----------h----(--- --@--H--P--X-@--------@---ȣ--У--أ-@--------@-@--H--P--X-@--------@---Ȥ--Ф--ؤ-@--------@-@--H--P--X-@-p--x---x--` ȥ-p-Х-W -h-- -`- -Y @-X-H-Ha h-P-p-` -H--^ -@--X -8--R -0--U 0-(-8- X- -`- T ---W --- Ч--ا-`V ---T --(- H--P-T p--x-@ ---S --Ȩ-@R --- --- 8--@- `--h- --- ---` ة--- ---S (--0-S P--X- x---`S -x--T Ȫ-p-Ъ-U -h--S -`- -T @-X-H- h-P-p-` -H--R -@--@S -8-- -0-- 0-(-8-@ X- -`- ---` --- S Ь--ج- ---S --(-Q H--P-`U p--x-R ---@U --ȭ- U ---X --- 8--@-Fa `--h- --- ---[ خ--- ---Z (--0-Da P--X-(a x---_ -x--_ ȯ-p-Я-`T -h--@V -`- - @-X-H-U h-P-p-@W -H--$a -@--U -8-- a -0--\ 0-(-8-_ X- -`-a ---^ ---_ б--ر-Z ---^ --(-^ H--P-` p--x-Ba ---a --Ȳ-` ---\ --- 8--@-` `--h-^ ---[ ---a س---_ ---@ (--0-^ P--X-Y x---@ -x--[ ȴ-p-д-` -h--^ -`- - @-X-H-W h-P-p- a -H--` -@--X -8-- a -0-- 0-(-8- X- -`-^ ---^ ---@a ж--ض-` ---5a --(-` H--P-_ p--x-] ---_ --ȷ-] ---] ---[ 8--@-I `--h-a ---_ ---X ظ---_ ---Z (--0-_ P--X-a x---_ -x--` ȹ-p-й-Y -h-->a -`- - @-X-H-_ h-P-p-] -H--\ -@--^ -8--_ -0-- 0-(-8-[ X- -`- ---[ ---] л--ػ-X ---^ --(- H--P-` p--x-@T ---X --ȼ-` ---^ ---_ 8--@-^ `--h-~_ ---` ---x_ ؽ---r_ ---l_ (--0-` P--X-] x---}` -x--\ Ⱦ-p-о-f_ -h--Z -`- -x` @-X-H-\ h-P-p-`_ -H--] -@--- -- r--ɓ -0-n--đ --h--{ -p - (-Ј@-` H-`-t h-`*x-@c-- --]--t -0 -T-- --@S--( -p-H-- - -=- -\ (- 8-9-@-C H- X-@2-`- h-x- '--d -'-#--r --@--Z ----X -- --O -%-- -= (- 8-@,@-A H- X-`,`-, h-Хx-@,-ȗ -r- ,-3 -p-@,- --,-? -P"-@,- --, - (-t8-,@- H-}X-,`- h- hx-,-Ǖ -0S-`|,- -Y-o,-Ɩ -0?-],- -$-R,- - -A, - (-=8-`9,@-u H-PX-`/,`- h-pJx-&,- --,- -- ,-u -`(-`,-' --+-X --+ -, (-B 8-`+@-; H-X-+`-ݒ h-x-`+-| -;-+- --+- -U-+- -P-+ -ݑ (-@-В H-jX-t-`- h-Cx-x--] -0-@}-- -0W--- ----q -- -- -0/-- - (-8-`-@-ܗ H- X--`-a h-a +++ +(+0+8+@+ H+P+ X+ `+ h+p+x+++++++(++/+0Ȯ+Ю+خ+1+2++++++I+K +L(+Q0+8+@+WH+^P+_X+`+h+gp+hx+j+m+p+q++z+}++~+ȯ+Я+د++++++ +(+0+8+@+H+P+X+ `+h+p+x++++++++++ȩ+Щ+ة+++++++++ +(+0+ 8+@+!H+"P+#X+`+$h+%p+&x+'+)++*+++,+-++.+Ȫ+3Ъ+4ت+5+++6+7+8+9+:+ +(+0+;8+<@+=H+>P+?X+@`+Ah+Bp+x+C+D+E+F+G+H+++J+ȫ+MЫ+Nث+O+P+R++S++T+U+V +X(+Y0+Z8+@+[H+P+\X+`+]h+`p+ax+b+c+d++e+f++i+k+lȬ+nЬ+oج++r+s+t+u+v+w+x+y +{(+|0+8+@+H+P+X+`+h+p+x++++++++++ȭ+Э+ح++++++HHA+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!h%-*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%-*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*DAWIAVIAUIATUSHdH%(H$1HHHH Hu HCHP0HLgIHHHI $Hu ID$LP0HMu HEHP0HHCu H_*LLH5H81{LC M9v9HPMLAWH1HMHZ11YHyL9tHH*MLLH5H81HMu HEHP01H uHCH1P0H$dH3 %(Ht_H[]A\A]A^A_AVAUATUSHPH-D-dH%(H$H1H$HD$HD$HD$HD$ HD$(Ht HE2Ld$xA1H_L'Ll$|HGLH1D$|8D$xu D$~8D$ztjH$1MLHHL1Hvy,HE11pC-H]C-[C-1HHUC-Hu,HE11.C-HC-C-p1H=lHHC-Hu,HRE11B-HB-B-'1H=#HHu,HE11B-HB-B-H=&-xHHB-Hu,HE11WB-HDB-BB-&HHHSB-Hu,HE11B-HB-A-(VHH=T{HHB-Hu,H7E11A-HA-A-* H=4HHA-Hu,HE11A-HpA-nA-+HA-H=A-H5H ,(yhHE112A-HA-A-/tC C!H{tc{"t4I$HH8HEH7H(L#Mu-H5MH%11=-H=-=-6H<$Ht HuHGP0H|$ Ht HuHGP0H|$(Ht HuHGP0HtHMu HEHP0HtH u HCHP0MtI $u ID$LP0H=L=-tEH=:=-tH =-=-1H=pH==-HuHGP0H=- HuH*H53H8H-<-HE11<-H<-<-1(HHu,HE11^<-HK<-I<-:H5HHHuFHV<-H5HWy,HnE11<-H;-;-<CH=`;-軞H;-HH=|9-蟞Hx;-HH=`2-胞HH=[9-nH?;-Ht}H=:-VH;-HteH=:->H:-HtMH=;9-&H:-Ht5H=1-Ht$H5z3-1H.-Hu0SH`11:-H:-:-A6H;*1HHH^H'.-HHHH޿19H--HHHH޿1H--HHHH޿1H--HHHH޿1Hs--HhH5/-1HH--HEHHH޿1H--H HHH޿1]H,-HHHH޿18H,-HH51/-1H,-HHHH޿1HY,-HHHH޿1H,,-HiHHH޿1H+-HDH5.-1H+-H!HHH޿1^H+-HHHH޿19Hz+-HHHH޿1HM+-HHHH޿1H +-HHHH޿1H*-HhH5--1H*-HEHHH޿1H*-H HHH޿1]Hn*-HHHH޿18HA*-HH51--1H*-HHHH޿1H)-HHHH޿1H)-HiHHH޿1H)-HDH5,-1Hd)-H!HHH޿1^H7)-HHHH޿19H )-HHHH޿1H(-HH*1HHHH(-HH5+-1H~(-HcHHH޿1HQ(-H>HHH޿1{H$(-HH53-1XH'-HHHH޿13H'-HH5d3-1H'-HH54-1Hv'-HHHH޿1HI'-HfHHH޿1H'-HAH5L2-1H&-HH5(-H=(-HzH'-HH5(-H=(-HTH'-HH5)-1Hz&-HHHH޿1HM&-HH5--1H"&-Hg1HHH޿HIHHH޿1H%-H$HHH޿1aH%-HH5*1-1>H%-HH50-1HT%-HH50-1H)%-HH50-1H$-HsH5+-1H$-HPH50-1H$-H-H5x*-1lH}$-H H5M*-1IHR$-HH5*2-1&H'$-HH51-1H#-HH5t&-1H#-H~H592-1H#-H[H5/-1H{#-H8H5/-1wHP#-HH5(-1TH%#-HH5'-11H"-HH5"/-1H"-HH5W.-1H"-HH5.-1Hy"-HfH5.-1HN"-HCH5.'-1H#"-H H5 '-1_H!-HH5&-1!-1H[-H` H53 -1H0-H= H5-1|H-H H5"-1YH-H H5!-16H-H H5!-1H-H H5!-1HY-H H5$-1H.-Hk H5$-1H-HH H5-1H-H% H5-1dH-H H5-1AH-H H5-1HW-H H5/-1H,-H H5<-1H-Hv H5-1H-HS H5-1H-H0 H5-1oH-H H5-1LHU-H H5-1)H*-H H52-1H-H H5g-1H-H H5,-1H-H^ H5!-1H~-H; H5-1zHS-H H53-1WH(-HH5 -14H-HH5-1H-HH*HHH H-HH5 -1H-HgH#-H5S -1HP-H=H50 -1|H%-HHHH~HHHڿ1CH-HH5-1 H-HHHH޿1H-HHHH޿1H_-HtHHH޿1H2-HOHHH޿1H-H*HHH޿1gH-HUL - 15-L-H L-5-H-5-H52-5-5-50-5-H@HHY"-11L A"-SAAQj5-5*-RRPRR HPHTASL E- 15-LI-H -5$-H-5?-H5-5-5-5~-5-KH@HH!-11L !-ARAAQj85-5w-RRPRR HPHAPL - 15e-L-H -5q-H2-5-H5-5?-5A-5-5e-H@H9H -L -WA11AQjk5-5-RRPRR NHPHVL - 15-L-H 6-5-H-5-H5-5-5-5-5-H@HHC -L 4 -QA11AQh5W-5-RRPRR HPH;AVL ,- 15-L0-H -5 -H-5&-H5g-5-5-5e-5-2H@HH-11L v-AUAAQh5-5[-RRPRR HPHUL w- 15J-L{-H -5V-H-5q-H5-5$-5&-5-5J-}H@HH-11L -SAAQh5-5-RRPRR 0HPHASL - 15-L-H -5-Hb-5-H5-5o-5q-5-5-H@HiH$-11L -ARAAQh75-5-RRPRR zHPHAQL$- 15-L -H a-5-H-5-H5G-5-5-5E-5-H@HHn-11L V-APAAQhj5a-5;-RRPRR HPHeWL W-1 5*-L[-H -56-H-5Q-H5-5-5-5-5*-]H@HH-L -VA11AQh5-5-RRPRR HPHH d-H-1H5-HHB-L 3-QA11AQh65-5X-RRPRRHPHt>H5I-1}H -HtH5*-1^H -Hu+H 11-H--C`H5H=cOHHu,H=E11U-WHB-@-dH5H=HH-Hu,HE11-YH--eDH H5ѸH=ĸHHu,HE11-[H--f0H5H=xdHHu,HRE11j-aHW-U-gHI-H=b,H+-HcH--y+H11-QH--jUH=-11H;,H-,GHHt?H5H-HHHxHHuIHCHP0=HHu HCHP0H11-QHl-j-oH=-Hw,H5߷+y+H116-QH#-!-pxH5,`H5H=fH-HHu,HE11- H--H-1H=-HH$Hu,HE11-H|-z-XH5-H=-Hoy+H11J-H7-5-ZH<$HuHGP0Hq-1H=P-H$HH$Hu,HRE11-8H--d'H5-H=-Hy+H 11-8H--fH<$HuHGP0H-1H=-H$iHH$Hu,HE11;-kH(-&-p}H5j-H=;-Hy+Hb11-kH--r8H<$HuHGP0H-1H=-H$HH$Hu,HE11-H~-|-|H5 -H=-Hqy+H11L-H9-7-~H<$HuHGP0Hs-1H=-H$HH$Hu,HTE11-H--)H5 -H=-Hy+H11-H--H<$HuHGP0H-1H=(-H$kHH$Hu,HE11=-H*-(-H54 -H==-Hy+Hd11-H--:H<$HuHGP0H-1H=^-H$HH$Hu,HE11-7H-~-H5 -H=-Hsy+H11N-7H;-9-H<$HuHGP0Hu -1H=-H$HH$Hu,HVE11-jH--+H5 -H=-Hy+H11-jH--H<$HuHGP0H -1H=,H$mHH$Hu,HE11?-H,-*-H5> -H=?-Hy+Hf11-H--GHHD$ Hu,H̒E11_,~HL,J,L顾H|$(HuHGP0HD$(H=,o_HH$Hu,HnE11,~H,,OCH5,HFHHu,H+E11,~H,,QH<$HuHGP0H=,H$^HH$Hu*HϑE1d,~HQ,O,T馽H5+,HEHHu*HE1#,~H,,VeH<$HuHGP0H5,1HH$GH$Hu*H2E1,~H,,Y HMu HEHP0H5.,H<$MEHHu*HE1u,~Hb,`,\鷼H<$HuHGP0H$H*H9Cu/HCH$Ht"LkHIEH uHCHLP0LH$HuCHHӀHD$(HH@E1,~H,,jHSH;@*uTHt$`HHD$`Hl$hVHD$(HHߏE1t,~Ha,_,q鶻H;K*HSR⍁Ht$`HHD$`Hl$hK>HD$(Hu*HlE1,~H,,zCH<$Ht HuHGP0H$HMHEHP0'IHu*HE1,~H|,z,ѺH$Im 1LHH$IEEHHD$(Hu*HM7,~H$,",yIMu IELP0H u HCHP0H5,H|$(]HHu,H@E11,~H,,H|$(HuHGP0H|$ HHD$(HHD$(Hu*HE1u,~Hb,`,鷹H|$ HuHGP0HD$ H u HCHP0H,HT$(H5,H#y+Hj11,~H,,@H|$(HuHGP0HD$(H=,~H_*H=,HH&,hHHD$(Hu,HE11,Ho,m,ĸH5,H=,Hby+H11=,H*,(,H|$(HuHGP0HD$(H=j,MYHHD$(Hu,HKE11,H,, H5,Hu@HHu,HE11,H,,ݷH|$(HuHGP0H5,H=,HHD$(ay)H1>,H+,),逷H u HCHP0H=v,YXHHu,H\E11,H,,Ŭ1H5,H?HHD$(Hu*HE1,H,,ǬH u HCHP0HT$(H5,H=,xy+H11S,H@,>,ʬ镶H|$(HuHGP0HD$(H=,cWHHD$(Hu,HaE11,H,,Ԭ6H5,H>HHu,HE11,H,,֬H|$(HuHGP0H5,H=,HHD$(wy)H1T,HA,?,٬閵H u HCHP0H=,oVHHu,HrE11,H,,GH5,H=HHD$(Hu*H*E1,H,,H u HCHP0HT$(H5R,H=,莪y+HՈ11i,HV,T,髴H|$(HuHGP0HD$(H=,yUHHD$(Hu,HwE11 ,H,,LH5q,H<HHu,H4E11,H,, H|$(HuHGP0H5,H=,HHD$(荩y)Hԇ1j,HW,U,鬳H u HCHP0H=,THHu,HE11, H,,]H5 ,H;HHD$(Hu*H@E1, H,,H u HCHP0HT$(H5,H=,褨y+H11, Hl,j,H|$(HuHGP0HD$(H=,SHHD$(Hu,HE11 ,!H , ,bH5,H:HHu,HJE11,!H,,H|$(HuHGP0H5k,H=,HHD$(裧y)H1,!Hm,k,±H u HCHP0H=,RHHu,HE111,"H,,sH5,H9HHD$(Hu*HVE1,"H,,!-H u HCHP0HT$(H5n,H=,躦y+H11,"H,,$װH|$(HuHGP0HD$(H=,QHHD$(Hu,HE116,#H#,!,.xH5,H8HHu,H`E11,#H,,05H|$(HuHGP0H51,H=,HHD$(蹥y)H1,#H,,3دH u HCHP0H=,PHHu,HE11G,$H4,2,=鉯H5,H7HHD$(Hu*HlE1,$H,,?CH u HCHP0HT$(H5,H=,Фy+H11,$H,,BH|$(HuHGP0HD$(H=,OHHD$(Hu,HE11L,%H9,7,L鎮H5,H6HHu,HvE11 ,%H,,NKH|$(HuHGP0H5,H=,HHD$(ϣy)H1,%H,,QH u HCHP0H=,NHHu,HʁE11],&HJ,H,[韭H5,H5HHD$(Hu*HE1,&H,,]YH u HCHP0HT$(H5J,H=,y+H-11,&H,,`H|$(HuHGP0HD$(H=,MHHD$(Hu,HπE11b,'HO,M,j餬H5,H4HHu,HE11,'H , ,laH|$(HuHGP0H5u,H=,HHD$(y)H,1,'H,,oH u HCHP0H=,LHHu,HE11s,(H`,^,y鵫H5J,H 4HHD$(Hu*HE1-,(H,,{oH u HCHP0HT$(H5,H=,y+HC11,(H,,~H|$(HuHGP0HD$(H=,KHHD$(Hu,H~E11x,)He,c,麪H5,H3HHu,H~E115,)H", ,wH|$(HuHGP0H5,H=$,HHD$(y)HB~1,)H,,H u HCHP0H=,JHHu,H}E11,*Hv,t,˩H5,H 2HHD$(Hu*H}E1C,*H0,.,酩H u HCHP0HT$(H5,H=/,y+HY}11,*H,,/H|$(HuHGP0HD$(H=,IHHD$(Hu,H|E11,+H{,y,ШH5,H%1HHu,H|E11K,+H8,6,鍨H|$(HuHGP0H5,H=:,HHD$(y)HX|1,+H,,0H u HCHP0H=&, IHHu,H |E11,,H,,H5,H60HHD$(Hu*H{E1Y,,HF,D,雧H u HCHP0HT$(H5|,H=E,(y+Ho{11,,H,,EH|$(HuHGP0HD$(H=0,HHHD$(Hu,H{E11,-H,,ĭH5 ,H;/HHu,HzE11a,-HN,L,ƭ飦H|$(HuHGP0H5,H=P,HHD$('y)Hnz1,-H,,ɭFH u HCHP0H=<,GHHu,H"zE11,.H,,ӭH5,,HL.HHD$(Hu*HyE1o,.H\,Z,խ鱥H u HCHP0HT$(H5,H=[,>y+Hy11,.H,,ح[H|$(HuHGP0HD$(H=F,)FHHD$(Hu,H'yE11,/H,,H51,HQ-HHu,HxE11w,/Hd,b,鹤H|$(HuHGP0H5,H=f,HHD$(=y)Hx1,/H,,\H u HCHP0H=R,5EHHu,H8xE11,0H,, H5R,Hb,HHD$(Hu*HwE1,0Hr,p,ǣH u HCHP0HT$(H5,H=q,Ty+Hw11/,0H,,qH|$(HuHGP0HD$(H=\,?DHHD$(Hu,H=wE11,1H,,H5G,Hg+HHu,HvE11,1Hz,x,ϢH|$(HuHGP0H5,H=|,HHD$(Sy)Hv10,1H,,rH u HCHP0H=h,KCHHu,HNvE11,2H,,#H5,Hx*HHD$(Hu*HvE1,2H,,ݡH u HCHP0HT$(H5,H=,jy+Hu11E,2H2,0,釡H|$(HuHGP0HD$(H=r,UBHHD$(Hu,HSuE11,3H,,(H5],H})HHu,HuE11,3H,, H|$(HuHGP0H5 ,H=,HHD$(iy)Ht1F,3H3,1,#鈠H u HCHP0H=~,aAHHu,HdtE11,4H,,-9H5,H(HHD$(Hu*HtE1,4H,,/H u HCHP0HT$(H5d,H=,耕y+Hs11[,4HH,F,2靟H|$(HuHGP0HD$(H=,k@HHD$(Hu,HisE11,5H,,<>H5C,H'HHu,H&sE11,5H,,>H|$(HuHGP0H5,H=,HHD$(y)Hr1\,5HI,G,A鞞H u HCHP0H=,w?HHu,HzrE11 ,6H,,KOH5,H&HHD$(Hu*H2rE1,6H,,M H u HCHP0HT$(H5B,H=,薓y+Hq11q,6H^,\,P鳝H|$(HuHGP0HD$(H=,>HHD$(Hu,HqE11,7H,,ZTH5),H%HHu,H,;H+,),這H5,H!HHu,HhmE11,;H,,=H|$(HuHGP0H5q,H=,HHD$(y)Hm1,;H,,H u HCHP0H=,9HHu,HlE11O,<H<,:,鑘H5,H HHD$(Hu*HtlE1 ,<H,,KH u HCHP0HT$(H5\,H=,؍y+Hl11,<H,,H|$(HuHGP0HD$(H=,8HHD$(Hu,HkE11T,=HA,?,閗H5#,HHHu,H~kE11,=H,,SH|$(HuHGP0H5,H=,HHD$(׌y)Hk1,=H,,H u HCHP0H=,7HHu,HjE11e,?HR,P,î駖H54,HHHD$(Hu*HjE1,?H , ,ŮaH u HCHP0HT$(H5,H= ,y+H5j11,?H,,Ȯ H|$(HuHGP0HD$(H=,6HHD$(Hu,HiE11j,@HW,U,Ү鬕H5),HHHu,HiE11',@H,,ԮiH|$(HuHGP0H5,H=,HHD$(y)H4i1,@H,,׮ H u HCHP0H=,5HHu,HhE11{,AHh,f,齔H5",HHHD$(Hu*HhE15,AH", ,wH u HCHP0HT$(H5,H=!,y+HKh11,AH,,!H|$(HuHGP0HD$(H= ,4HHD$(Hu,HgE11,BHm,k,“H5,HHHu,HgE11=,BH*,(,H|$(HuHGP0H5[,H=,,HHD$(y)HJg1,BH,,"H u HCHP0H=,3HHu,HfE11,CH~,|,ӒH5,H(HHD$(Hu*HfE1K,CH8,6,鍒H u HCHP0HT$(H5^,H=7,y+Haf11,CH,,7H|$(HuHGP0HD$(H=",3HHD$(Hu,HfE11,DH,,ؑH5u,H-HHu,HeE11S,DH@,>,镑H|$(HuHGP0H5!,H=B,HHD$(y)H`e1,DH,,8H u HCHP0H=.,2HHu,HeE11,EH,,H5V,H>HHD$(Hu*HdE1a,EHN,L,飐H u HCHP0HT$(H5,H=M,0y+Hwd11 ,EH,,"MH|$(HuHGP0HD$(H=8,1HHD$(Hu,HdE11,GH,,,H5,HCHHu,HcE11i,GHV,T,.髏H|$(HuHGP0H5g,H=X,HHD$(/y)Hvc1 ,GH,,1NH u HCHP0H=D,'0HHu,H*cE11,HH,,;H5,HTHHD$(Hu*HbE1w,HHd,b,=鹎H u HCHP0HT$(H5,H=c,Fy+Hb11!,HH, ,@cH|$(HuHGP0HD$(H=N,1/HHD$(Hu,H/bE11,IH,,JH5,HYHHu,HaE11,IHl,j,LH|$(HuHGP0H55,H=n,HHD$(Ey)Ha1",IH, ,OdH u HCHP0H=Z,=.HHu,H@aE11,KH,,XH5,HjHHD$(Hu*H`E1,KHz,x,ZόH u HCHP0HT$(H5H,H=y,\y+H`117,KH$,",]yH|$(HuHGP0HD$(H=d,G-HHD$(Hu,HE`E11,LH,,eH5,HoHHu,H`E11,LH,,g׋H|$(HuHGP0H5;,H=,HHD$([y)H_18,LH%,#,jzH u HCHP0+HHu,HX_E11,H,,r-H,H53,Hˀy)H_1,H,,tH,H5,H舀y)H^1e,HR,P,u駊HD,H5,HEy)H^1",H, ,vdH,H5J,Hy)HI^1,H,,w!H>,H5,Hy)H^1,H,,xމH+,H5,H|y)H]1Y,HF,D,y雉H,H5,H9y)H]1,H,,zXHݽ,H5n,H~y)H=]1,H,,{H,H5#,H~y)H\1,H},{,|҈Hϻ,H5,Hp~y)H\1M,H:,8,}鏈H,H5,H-~y)Ht\1 ,H,,~LHi,H52,H}y)H1\1,H,, H,H5,H}y)H[1,Hq,o,ƇH,H5|,Hd}y)H[1A,H.,,,郇H,H5A,H!}y)Hh[1,H,,@Hm,H5,H|y)H%[1,H,,H*,H5,H|y)HZ1x,He,c,麆H,H5,HX|y)HZ15,H", ,wH,H5e,H|y)H\Z1,H,,4H!,H5,H{y)HZ1,H,,H,H5,H{y)HY1l,HY,W,鮅H+,H5,HL{y)HY1),H,,kHh,H5 ,H {y)HPY1,H,,(Hm,H5V,Hzy)H Y1,H,,H,H5{,Hzy)HX1`,HM,K,预HW,H5H,H@zy)HX1,H ,,_H,H5,Hyy)HDX1,H,,H,H5,Hyy)HX1,H,,كH,H5׿,Hwyy)HW1T,HA,?,閃H,H5<,H4yy)H{W1,H,,SHص,H5,Hxy)H8W1,H,,H,H5,Hxy)HV1,Hx,v,͂HB,H53,Hkxy)HV1H,H5,3,銂H,H5,H(xy)HoV1,H,,GH,,H5ݽ,Hwy)H,V1¿,H,,H,H5,Hwy)HU1,Hl,j,HV,H5_,H_wy)HU1<,H),',~H,H5,Hwy)HcU1,H,,;H,H5,Hvy)H U1,H,,Hͷ,H5n,Hvy)HT1s,H`,^,鵀H2,H5{,HSvy)HT10,H,,rH,H5h,Hvy)HWT1,Hڽ,ؽ,/H,H5m,Huy)HT1,H,,H5,H=,Huy)HS1g,HT,R,H HCHP0退3,5),11H ,H=y[*HmS,H,,ĪGH$HdH3 %(HtoHP[]A\A]A^H=I,HB,H9tHp*Ht H=,H5,H)HHH?HHtHUq*HtfD=,u+UH=:q*Ht H=6h*ud,]wHu710HtH˻,HPH8p*HHP HP(HHQp*1H5,8ATIUHSHH Ht HՅuH{(1Ht[LH]A\fD[]A\ff.UHSHHo*H HH] HHHt H/t?HH}(HH](HHtH/tH1[]fHGP0H1[]HGP0HGE1LH@ uLGuLA1LAHGpHHGxHHHHHtHHHtHHHtHUSHHGpHoxHwpHHWxHHtH(tIHtHmt-HtH+t H[]HCHH@0H[]DHEHP0@HPHR0ff.USHHGXHo`HwXH_hHW`HOhHtH(tOHtHmt3HtH+tH[]fHCHH@0H[]DHEHP0@HPHR0ff.AAxgIcȉH9T|PE~S1}9pA9~'D)HcHL9~AA99@O1fSHGHL@t+Hl*LH;H81qu%H[fHyl*HHH54<H81qH+t 1H[DHCH1P0fHGt HHHH@`Ht+HHtHtH l*H9HuHfDmHt1H5sPHHHk*H5MHD$H:jHD$ff.fATIUSHHHoHVnHt9HPHHtHHHH[]A\fDHH[]A\H l*LHD$H:aiHD$f.AWL~AVAUATUSHM~xLL $IHHLD$ILfIvLlL9t9IHL$H<$HL,LSlHHLElH4$HH6lL)IuHj*HH[]A\A]A^A_SHH dH%(HD$1HGHoHT$HHt$mHH{HPHHtgHHCHH<$HT$HHt$|fH{ HtHC H/tMH{(HtHC(H/t&HCH@HD$dH3%(uPH [HGP0HGP0H'Gi hff.HGHHtfD#nAUIATIUHSHH(dH%(HD$1HGXHT$Ht$HGXH$HG`HG`HD$HGhHGhHHD$gH{XHt$Ht H|$eHD$HtHH$HtHHD$Ht HHD$H $HT$IMI$HUH{pLcxHHKpHCxHHtH/tcMtI,$tGHtHmt+1HL$dH3 %(H([]A\A]f.HEHP0@ID$LP0HGP0H<$IEI$HEHt H/uHGP0H|$Ht H/uHGP0H|$Ht H/uHGP0AfDAVAUATUSHGLM|HIHhL-?g*p NH A;MhLHHAH|hH QP AEH=~9|HtTH[]A\A]A^ @9}Ah@$[]A\A]A^jH=6gt1@chHHuHe*H5u6H8UeAVIAUATIUHSHHb1LHHIcHtiHM~ 1HHHHhHI9u1HieAE HmtAm H[]A\A]A^HUHD$HR0HD$1ff.HHH?H9t}HXHtOHJH~H;rtb1 @H;ttSHH9uHQe*HNH55HWH81i1+@HDHH9tHuH;5e*uH@H1d*H5FH8c1fDH9tfHGH;xe*u>HOHAHw`Ht{GHHHHDH9t0Hd*HfH;d*uOfH*f.GzuHd*HfHt2HuGOHH D1@bfDGOHH HiATUSHH9Hd*H9GHH9F ~ HSH9VHCHNH9@H@t H{ DF D8@ "H{HA |HN0HHA@HEȃpwDA9uEHHc1u)1 L%Qc*L9ut1H[]A\L9uuH߉EaHHH;*c*H;b*uL9u<H+uHSD$HR0D$1H[]A\HefDHK0HH@HHDfHt$F^Ht$>'HHt$^Ht$@HvHD@Dff.USHHHGH;b*t*HhhH}HE(HttHHH[]HOHHHIHxLH9}GHOHHHH?H9}lthHDHH[]HHHH?t3u/H>HHH?H9},t(HSHHH[]DHHHYHHthHH]XHm{HUHD$HR0HD$bfHUHHt$Ht$HxHHEfD1"HX*Ht$H8XHt$tHt$YHEHt$@HHH@@AT1USHYIHti1HHl\I,$Hu ID$LP0HtBHM@teHH*Vf.HmuHE[H]A\H@0[]A\UHW*H5)H8Vf.HW*HH5O)H81%\AWAVIAUATAUSHHLLoHo Xp VP HxW*;AGu A CHK0HS(1HtHEHHE1ELQLLR1PUjfYH0HzXx WP H W*=!@9|HH[]A\A]A^A_D29}4X@$fDH=&WD1Ht+HEHEu IcWH9t4HK0HS(BE9guIcLLLHBLHLH,ff.AVAUATUSHLWDd$@IZpHHCHHMtI0H[]A\A]A^EL-V*E1HH1HV*LRZIMtI.u IFLP0MtQLHSImIu IELP0HL[]A\A]A^H9U*IRH5v'H81YHE1[L]A\A]A^ÐEuCHlU*Hb@HHL$H$gUH$HL$HII"DH@UIHt;HU*LHfYIMtI.t=I,$ID$LP0MPI.FIFLP0IFLP0ff.@AUATUSHPHhXHu1H[]A\A]HHT*H0H9urLk`LchHCXHC`HChHmt/MtImt3MtI,$uID$LP0f.HEHP0@IELP0HEtM@tDHXHtJHJH~%H;rY1H;tGHH9uH"R$HHH9Hu1H;5S*fH9AUATUSHHHteHFHHG@HXHt7HJH~H;rtf1 @H;ltWHH9u1H[]A\A]HH9t2Hu1H;-R*H9QuIM9tJtH9ueH[]A\A]fHHH[]A\A]PfLfMsHvH9t1 fDH;\tHI9uE1HCk@^HXHtfDHHH+tHH[]DHCHP0HH[]@kH+HcukCHH H+uH+kuDkCHH HH+uHQH+HlvfDH@`HtOHHtCHHt9H@H;nO*HH5N3HHu3HDPHuHN*H5`0H8xMH@ff.fAUATIUHSHOHHEHHhID$HHOL-@N*p VP A;U1HLH{OH QP AEH=~D9|JHH+tHH[]A\A]HCHP0HH[]A\A] @9}O@$H1[H]A\A]1HLQHfDH=9LN7@1bf OHHuHDL*H5H8K1ATUSHHWHBhH_HHHRHHHeM*H9FSHFHpHweHuHtsH;M*H;K*aHjhH,HMHHHH[]A\fH6uEHH fHH3HSH; M*tNH;K*uHKHH1HHH?H9HDHH[]A\HKHH1HHH?H9}XtTHSHHH[]A\HqK*HRH5H81OH1[]A\Hf.HKHHHHaJHmNHUHD$HR0HD$5DHKHJLHKHCH;K*H;:J*HhhHVHMHHSHEH0Ht$HHt$HNHHM fHHOIHlH$KI,$HHD$ID$LP0Ht$@HKH%HtHJHf.uEHH ;fDHKHH;J*tw1H;I*BHOHƺHI*HH2eHt)\JHEH5HPHI*H81M@1yHKHHO1HaI*Ht$H8HHt$tHt$IHMHt$HAff.HGH@pHtH@Ht ff.AUATUSHH8EHhXHHu1H[]A\A]H9I*H0H9Lk`LchHCXHC`HChHmtnMtImtrMtI,$uID$LP0fH/uHGHt$P0Ht$HG*HH5H81nLJ@HEHP0@IELP0HEtM@tDHXHtJHJH~%H;r1H;tHH9uH2FHHH9Hu1H;5G*fAWAVIAUIATAUSH(CHÅEE1L ,MYD,DLDtA9;HHIE9y(I)HEH,H1HlCHHD`lHFHmt=H+uHCHH@0H([]A\A]A^A_@H([]A\A]A^A_fHEHP0H+ufDLxXH@`HCXHC`H=,HD$HChHChHD$^IH5?,HHVH8GHHE*H5,1H=,mGHt$H{XL{XHS`HChHs`Ht$HshHt!H/uHOHD$HT$Q0HD$HT$HtH*uHJHD$HQ0HD$HtH(u HPHR0"AA%fDH= ,HGHHHHHHT$"EHT$H6E*H*uHJHD$HQ0HD$@HDEfLHIH$[LH&1H=&:@IMNH7,HE11L ,111AQATAVAUPPPPPBHPHIEHIEHu IELP0I.u IFLP0HL ,E$MkD,DLDVLcE9IcHLD9~D9S,AJDHcHLHcoHHAA9|AD~H.D,HELFIHE1LFIfImIELH@0yDD9,Er@LIcHBIHIcDj,Hk,HD5\,HE93HmHEHH@0DGHH@H,H5,DxH(HEt@IK4)H!B*6B1H>H.H/:HGP0.FHHGH;5B*AUATUSHHH;B*HOQE1 La}BH-A*p VP ;U1LAHBp VP EH=~!9|'HHH[]A\A]fD @9}B@$LozLL-ڌ,MLBH-@*p VP ;U1LHAT11kBHHt&1X@H=YlA 19H~?*H5WH87?fHLH1[]A\A]YDfH=AY1DSHHGt~HGHtHfH@`HtuHHtiHHt_H?*H9Cu/@H8H+hHSHD$HR0HD$OHH5#6HHuH,@HuH>*H5 H8=SHHGt~HGHt*H5jH8B=HfH@`HtuHHtiHHt_H>*H9Cu/@H8H+hHSHD$HR0HD$OHH5c"HHuH,?HuH=*H5zH8HuHq<*H5JH8b;HGtkHGHHHwUH52'HcH>@GHËGWHH fG@GWHH H=SHH@`HtrHHtfHHt\He<*H9Cu/HHH+uHSHD$HR0HD$H[HH5 HHuHS=HuHW;*H50H8H:fDSHHGHW1HtHtHx\O>Hu7H[ÐWЁtH4;*H5H89fDHu1H[WGHH HcʉH9tH3*H5H81@GH[WGHH HHcʉH9uH[GH[D{4HcH9aHuD4HrCH@`HtiHHt]HHtSH2*H9Cu-@HH+HSD$ HR0D$ HH5uHHu3HuH1*H5H80ZDSHHGHGHHHHcH>1H[WGHH HcʉH9tHz1*H5H8#0@GH[WGHH HHcʉH9uH[GH[D2HcH9aHu2HrCH@`HtiHHt]HHtSH"1*H9Cu-@HH+HSD$ HR0D$ HH5xHHu2HuH0*H5H8 /ZDHGtkHGHHHwUH5HcH>@GHËGWHH fG@GWHH H1USHH@`H"HHHHH@H;/*fDHCHHHH5^HcH>HH5AHHHH+uHSHD$HR0HD$H[]CHCSHH HfDCCSHH HH5HHt-H@)fHh0c>0HHcH@`HHHHHHH.*H9EHEtHEHPHwlH 0HcH>HHmHUHD$HR0HD$EHӋEUHH HE뼋EUHH H{/T/HfHT-*H5-H8E,KH9-*H5H8*,DAUATUSHHGH;-*Ht$H;-*HWBH\$HjE1 td_.L%,*H QP A;$HLH5.p VP A$H=~.9|4HHH[]A\A]Lof. @9}-@$Ht$HHH[]A\A]DHt$H@H= -?@1hf-HHuH+*H5H8*7LJE1 uLGHt$uLAH1ɺLAHAVAUATUSHH5r,dH%(HD$1HGHH}HH'HCH;|**HkHLcHEI$H+ID$H;u+*H,$kH;+*~IT$BHZE1 o:,L5**H QP A;HLI,p VP A=:@97MHmLHEHP0fDHCHP00H;**CH;+*^HKQNE1 Hiq+L5 **H QP A;1LII+p VP AH= @9M<H+u HCHP0HL$dH3 %(LGH[]A\A]A^LcaMl$fD29*@$*@$mfH [ ;=%t,%H t,t,;=H=QE1@@-H{HL%t,H3*L5(*H QP A;*1LHI)H QfDHLIMI,$H Ht, t,%t,H=t0Hmu HEHP0H s,s,5s,ID$LP011HDIM,H+H  H s,s,%s,K=K=%DHLI;)HuHw&*H5PH80&{LB1 uI|$HAIfDH=''H='+~(HiH%*H5H8%NHCHP0H gr,mr,5cr,1LH*IH=:M'1ɺAI$fAVAUIATUSH0H5h,dH%(HD$(1HGHHHHHCH;%*SLcMFHkI$HEH+HEH;&* H;n&*&HHL`IELh HELM&L%]%*H QP A;$1HHAI&H QP A$H=<9BMH+Hmu HEHP0Imu IELP0H%*HHL$(dH3 %(H0[]A\A]A^HCHP0HEH;$*H;$*Ll$H;?%*HSBHjE1 %L%G$*H QP A;$LLI%H QP A$=\29HM%HHo,(Ho,o,=H+HCHP0@HUBፁLd$1Ll$LB uH}Ht$ AIMHo,(Ho,o,=DHmu HEHP0I,$u ID$LP0H n,n,5n,DHCHP0 @9-$@$@Lsd@DHt$HLl$Ld$IHI,$lID$LP0\D#'HNH s=(n,(H m,m,=H=11D[#@$fLHUIHMHm,(Hm,~m,=yf1HH%IH|HmHH5m,7m,()m,=u HEHP0H+1fDH= "f"HuH *H5H8{Ht$H6IfDLB1 uH{Ht$AIDH=at!XHJl,(H7l,5l,=01ɺAIHO*H5(H8L1ɺAI?ff.AWAVAUATIUSHHtI.uIFLP0HEHP0I.u HEH g,+H g,g,=HEE11HHHLL-f,ML=*H QP A;1LHAH=@StHE1(f,+Hf,HEf,>HHEHfLLIMH4Le,+He,HEe,>Hf.11HIMWKHLISH"H*H5dH8D@LB1 uI}HAI%H=H=|HH*H5H81LHI1ɺAI+ff.AWAVAUATUHSHHL%*H^dH%(HD$81Ld$ HH HHHH HIHHQH?L GHLIL@HH*SHvH5H81VH5c,c,5Hc,XZH H=eHL$8dH3 %(Q HH[]A\A]A^A_DIHH#HFHHD$ qHLd$ V11HEHfHHH}(H/|L-Eb,H](H=b,IULHHHHCH;*H;&*HH;*HKQE1 LyL-*H QP A;U1LAIH QP AEH= @9/MH+u HCHP0H} H/uHGP0HEL} HH5W,HHHHHEH; *LmMH]IEHHm?HCH;*H;_*1HH{LhI$L` HCLMjL-N*H QP A;U1HHAIH QP AEH= @9CMzHmu HEHP0H+u HCHP0I$HP1I$H"IT$D$LR0D$ f.LfWHGP0x@LsHEHP0HSBፁLl$ 1Ld$(LB uH{Ht$ AIMImIELP0 DLkMHSIEHH+uHCHT$HP0HT$HLHT$\HT$HIImHWIELP0HfD11H輻IM*H1!_,H_, _,5H+uE1HCHP0Mt ImYHtHmu HEHP0H ^,^,5^,:H;A*Ld$~H;*HUBHZE1 uLuL-*H QP A;ULLIp VP AE=29HMdHHjH1],H],],5k@$fH+H'H],],],6ImfIELP0LL5],ML-*H QP A;U1LHAf.H#],H],],56fHuLοHH4H=5\,H\,\,5H )f.@dD@$HT$ HLLH5,DDH{5J\,H7\,5\,{5+HeH{5 \,H[,[,59Ht$ HLd$(Ll$ LIHH[,H[,[,5H+f.kHBH*H5|H8\'@$fH=1DHh IHH5Q,LHV)HnHD$ ID$LHuIHM 1HHIHHZ,HZ,~Z,6mfH=atQf.;HuHw *H5PH80 fDHt$H薶IHW *H50H8 1ɺAI"@AWIAVAUATUSH8dH%(HD$(1HF LfIT$HH HcH>fDH+u HE1R0L1 IHjI_ L5S,HSLHH$HHH@H$HHHHHHfMw HS,IVHHD$HH$HHH@H$LMHLAHHH@H; *'LsMLCIIH+I@H; *Lt$ >H;_ *IPBrHz1H|$ L$ H _ *L$x W;P HL$LHHD$LD$H$ H$LD$H qHL$p =@9hHoI.LIFH$LP0H$HCL$HP0L$H; *H;d *HKQvHAE1 H$ H f *x WP ;HL$1LH$H$ H qp HL$H$=H9HH+uHCH$HP0H$H*u HBHP0 IWLI} HHQHEL%vG,HH H *x WP ;yH $1LHH H QH $P H=9HuHmtnH+u HCHP0IULIUHu IELP0HL$(dH3 %(HcH8[]A\A]A^A_fD1LH HmHu HEHP0HDHCvT,HcT,aT,C@HE:fDH @9 @$@ @9(H$ @$H$@DfFII ]DDfODfFII II- IHH B1H=6H oS,qS,cS,BNufDfAMc@LsIX2HH IIH [B1H=H R,R,R,B趿H*LH8H BR,HR,R,BH H=,1]IEHPiHT$L$HT$L$@$xH*Ht$H8\HmHHR,R,R,BH Q,Q,5Q,eLH5Q,MH4$PH *p V;H4$P HL$1HAH$p }LLL$L$HHHLQQ,H>Q,DcAMcH+HSHR0DDcCII IDDcH+DcCII H5HaHHhHP<@HIj1H.H}H=HL$H4$H4$HL$[HH*H5H8HB`H HHHHHH*H9EHEt$LeID$HHHcH>H*IHm~HEHP0oDeAMcDeEII IDeDeEII HH5VHHZH+HCHP0@HIi1AL$H%PHuHT*H5-H8EAWAVAUATUHSHxH<$dH%(HD$h1HFH;$*HD$PHD$XHD$`KH;d)>H@hH H@H 1HHH\$PH HD$PL%K,I9Hl*H9CI9D${ A|$ 'HSI;T$HCIL$H9@H@t H{ At$ 8@ 6@HK0LCHIDH@  IL$0IH@HIDȃ DA9<Ht'E1H_AfEHUHBpH H@H H5=,HHD$PH HPL%)L93 HPH HPH@ HT$XHD$`HHD$`HH|$PH/uHGP0HD$PH|$`HGHGHH H\HcH>f.H~H^HL-)L9H5y;,H=I,1HD$PHH轤H|$PH/^HD$PHHI,I,I,;E11E1E1HD$H|$XHt H/uHGP0H|$`Ht H/uHGP0MtI/u IGLP0H H,H,E1H=_5H,sM#DM9uLHIHtDH;)L;%r)u M9:I,$DID$LP0EHbG,G,;HG,HD$PE1E11E1HD$HH(HPHR0f.H~HFHHDD$H/uHGP0HD$XHHD$`HD$XHD$H HHUHBpHH@HH5.:,HHD$PHHPL9&HPH(HPH@ HT$`HD$XHHD$XHH|$PH/uHGP0HD$PHl$`Ld$XHD$`HD$X-LXpLPxHD$(LMtIMtIMtIHF,LT$ L\$L(hE1ɹAAHH|$AL\$LT$ HIHD$P HD$XHH|$PH/uHGLT$ L\$P0LT$ L\$HD$PLl$XHD$XMtI+uICLT$LP0LT$MtI*u IBLP0MtI.u IFLP0IE H8pH$H5@,Lx Ht$ IWHHT$IHH@HT$HHLLIHxH$H5@,Lx Ht$ IWHHT$HHHHT$LM LHAHD$PHHHHD$XH; )HPHT$XHH@HHH|$PHD$PH/uHGP0Ht$XH|$PHHD$`HH|$XH/uHGP0HD$XH|$PH/uHGP0HD$PH|$`H/uHGP0HD$`{LxpIL@xLMtIMtIMtIIuH$HHJHHyHxHHxH)H)΁HHBt$HEHEHHCHHcH>fGWHH HcЉD$H9uK|$tSH|$`G؉D$DGWHH HHcЉD$H9tH)H5H8HH|$`D$fDGD$t@1fDH$H@H)I9D$HLLL$ LD$LT$LL$ LD$LT$f. H$H@Mt"I/uIGLL$LL$P0LL$L$MtI(uI@L $LP0L $MtI)u IALP0H53,1LZI.Iu IFLP0M~I/u IGLP0L5)IImu IELP0HtH+u HCHP0HL$HtHH$HHu HAHP0HtHmu HEHP0MtI,$u ID$LP0H\$hdH3%(LHx[]A\A]A^A_fD19IHHHI,$HID$LP0@H-4,L%4,HEI$HY)HRH5H81HD$PHD$HH@,@,@,/;E11E1E1AD$@IHD$PH21?, H?,?,;HD$HHD$PfH|$PֱHD$`HrHk?,HX?,V?,q<fDI.u IFLP0HD$PE1g;HcЉD$H9HUHGXfDH; );HIH!H|$PH/uHGP0HD$PIGLLAHD$XHLAHD$`HALAվH֩I/IGLP0rDHI)HRH5H81HD$PH>,H >, >,;@HvHx3HHHH @HEH)H5H81cHD$PH5=,H5=,=,7;H=,Hp=,n=,l;HD$PE1E11E1fULL$ LD$LT$LT$LD$HLL$ Hi=, HD$0H<,<,<H|$XHt)H/u#HGLL$LD$L$P0LL$LD$L$HD$XH|$PHt)H/u#HGLL$LD$L$P0LL$LD$L$HD$PH|$`Ht)H/u#HGLL$LD$L$P0LL$LD$L$:<,H '<,L$H=5<,LL$LD$HD$`L$HL$XHT$PHt$`LˆL$LD$LL$HL$XHT$P1L$Ht$`LL$LD$LL$LD$HLL$1HLLL$ LD$LT$H$5I.H $HLT$LD$LL$ u:HD$IFLLL$(LD$ HL$L$P0LL$(LD$ HT$HL$L$H)u0HALL$ HLD$HT$L$P0LL$ LD$HT$L$H9H;a)AH;)D H;#) HLL$ LD$LT$H$ZH$LT$ALD$LL$ H*u&HBLL$HLD$L$P0LL$LD$L$E^Ht$`EH.u&HFLL$HLD$L$P0LL$LD$L$HD$`H|$PH/u#HGLL$LD$L$P0LL$LD$L$HD$PH|$XH/u#HGLL$LD$L$P0LL$LD$L$LLLLHD$X-EUHH HHcȉH9LL$ H)H5LD$LT$H8=LT$LD$LL$ fUEUHH HcȉH9LI,$ADHGP0@H{HHPHH@HH@HT$XHD$`}HY)H5-H81HD$PfH|$`H>8,HD$0H&8,$8,;Ht>H/EH|$PHD$`Ht!H/uHGLT$ L\$P0LT$ L\$HD$PH|$XHt!H/uHGLT$ L\$P0LT$ L\$HD$(L-Y7,HD$XHPXI9HtfIEHB9@,HXHtKHQH~"L;it]1L;ltJHH9uH|$(LLLE1|HD$P4HI9tHu1L;-)tH 6,6,H=LT$ 56,L\$覣H|$(HL$`HT$PHt$X}L\$LT$ H6,LT$ L\$L(hE1ɹAAHH|$AL\$LT$ HIH8uH@LT$ LL\$P0LT$ L\$H|$XH/uHGLT$ L\$P0LT$ L\$HD$XH|$PH/uHGLT$ L\$P0LT$ L\$HD$PH|$`H/uHGLT$ L\$P0LT$ L\$H|$(LLLHD$`{H5q',H=5,1IHD$`H;H躐H|$`H/uHGP0HE1HD$`H5,HD$P5,4,L<$HLL$ LD$LT$LT$LD$HcȉLL$ H9tHPLT$LD$HLL$ 3QHQ)Ht$ H8HE1l4,HY4,HD$PR4,_<DHp,LXH)Ht$ H84HD$PHV3,H3,3,a<@It$HfDH;)H{IHH|$PH/uHGP0HD$PIGLHHD$`HLHD$XH&LվH衞I/IGLP0LT$LL$(LD$ D$D$LD$ HLL$(LT$H=2,!HD$0H2,2,<fDH6Hx3HHH HEHd)H5H81HD$PH5E1E11H5P2,E1O2,A2,;p@D@H2,H2,2,n<H[1,H1,HD$P1,<@HGLT$ L\$P0H|$PLT$ L\$D@HPHH@HH@HT$`HD$XHI)H5H81HD$PfH31,H 1,1,;Ht 1,H0,HD$P0,_; DH@`HHHIHH^)I9Eu&LhImD$IELP0LH5wIHufDH[0,HH0,F0,T;HD$PE11E1HD$_LHLT$ L\$LT$ L\$$Hb/,H/,HD$P/,H</,HD$0/,<H/,H/,H/,/,L;IIMHbIuH91 I;THH9uE1ILMIHHLMHC@HXHL_M~H9wt+1H;tt HI9uHI9ItH9uMHILMD1I/u IGLP0iu1HHHH H5HEH^)H81Hg.,HT.,HD$PM.,\;h냐H@`HtHHdLL$ HLD$LT$LT$LD$HHLL$ 5H)H9AunHLL$(LD$ LT$HL$脮HL$LT$LD$ LL$(H)\D$HAHLT$P0LT$T$LD$ LL$(1HH5tLT$LD$HHLL$ o f.:-,HD$0'-,<H-,HvE11E1H,,HD$P,,,,;H*EVH/E11E1H,,HD$P,,,,;1I/u IGLP0Iu1HHHH H5qHEH>)H81HG,,H4,,2,,; ,,HD$0 ,,<H+,LLLLqE1HD$PH4H)H5H8LL$ LD$LT$LT$LD$HLL$ TH5lH);HT$PLt$XLL$(LD$ LT$Ht$H$H$Ht$LHuqHD$0LT$HD$`HD$PLD$ HD$XLL$(H+, +,*,<*,HD$0*,<H*,*,HD$0*,<H*,*,HD$0*,<H*,|HHH9Hu1H;5):HLL$HLD$@HL$8HT$ LT$+LL$HLD$@HL$8HT$ LT$*HHAH5#,HHL$HHHL$HD$8HZH)H|$8HGH;()H;K)Hl$HP H;)HWB Hr1Ht$ uHOHL$HL$p VP H);HHHD$HD$HL$p VP H^)=@9Hr HL$0H|$8H/uHGP0HD$8`HD$8HHT$0HPILx HD$0HD$8MtI,$uIT$HD$LR0HD$MtImuIUHD$LR0HD$MI.IVHD$LR0HD$HkfDHH-)H#H5UL yAH H81H[H4&,6&,H4H&,XZH 26H=诓1H\$hdH3%( Hx[]A\A]A^A_f.HD$0H(ILxHD$0H8uHPHHD$R0HD$HAHP0af2 IH#HHHHH5,LIHVIHD$PHH5!,LHVHD$XH IEHK Hl$PL|$XHL$HL$@$eH=atHL$HD$0H:%,;HD$H"%, %,4H|$@Ht H/uHGP0HD$@H|$0Ht H/uHGP0HD$0H|$8Ht H/uHGP0Hz$,HKXHD$8H9HtkHBHA@HXHxHNH~#H;V1 H;T}HH9u11H{pHSxLcpLLkxLHt!H/uHOHD$HT$Q0HD$HT$HtH*uHJHD$HQ0HD$MtI/uIWHD$LR0HD$H|$0HtH/uHWHD$R0HD$HtHmuHUHD$HR0HD$H|$8HtH/uHWHD$R0HD$H|$@HtH/uHWHD$R0HD$HtH(u HPHR0H #,#,H=5#,1;HH9tHu1H;v)H ",",H=5",虏HL$@HT$0HHt$8rmjH)H9EHEHIHHLxHD$HT$HHmuHMHD$HHT$Q0HD$HT$H*uHJHD$HQ0HD$H|$8H/uHWHD$R0HD$HD$8H|$0H/uHWHD$R0HD$HD$0H|$@H/uHWHD$R0HD$H{pL{xHD$@HLcpLkxLHtH/uHWHD$R0HD$MtI/uIWHD$LR0HD$HHmHUHD$HR0HD$H蘆HD$0HDLD$LD$HuL褃HHH ,;HD$H , ,4l@H{ ,;HD$Hc ,a ,4H)7HAHP0(;HL$AHOHaHGHHH|$8HD$8H/uHGHL$P0HL$H|$8HGH;v)BH;)HL$DHL$HHD$@pHH1HHEH|$8Hh UlHD$0HrH|$@H/uHGP0HD$@bHt$H{T@H;,;HD$H#,!,4@HF HHD$XHFHD$PNNfHFHHD$P/ILJE1 uLGHt$HLA@H:H)H5H8tHHMH5C,8H50,.,y4fHD$1,<,4H,1LBMHrH91 fH;LHI9uE1HA@HXH"LWM~"H9w1 H9tHI9uIM9JtH9}HG@=XHt$PHL$PHL$Hl$X=bHL$HHD$0(H)HAHP0HHHY,=HD$,5H,1n,=HD$,5H~,1Gy,=HD$f, 5HW,HHT$PHLLBH5+rH;4,6H , ,;4.HH)H H5jL EAHӲH81lH'Y^H,74,6,74Ht$PHL$PHL$Hl$XxHL$HHD$0sHȱ`,;HD$HH,F,4H4,;HD$H,,4Hp,;HD$H,,41ɺLAH-,;HD$H,,4EHLL$(LD$ HT$HL$zLL$(LD$ HT$HL$HHH9Hu1H;5)AWAVMAUATUSHHhH-,H<$H=B,Ht$HUHHL$LL$dH%(HD$X1H;U)IH HIEH5,LHH HIEHH IEHIH HHXIH& L-,H=|,IULxHHHHCH5,HHHAIHHM9HHiH5,LLLT$ LT$ I*[HEHHGBL-)p VP A;ULLHHp VP AEp=z9HHm>I,$I/H;H-,H=I,HUHEIHIHIEH5,LHHIIEHMIEHID$E1E1H5 )Ht$ H9H;() H;)Hca LT$(LT$(HHNMtLPHt$Ic1LHHHtAEHHILtAEHHH\cIHHmu HEHP0I,$u ID$LP0I}IEu LP0IEHH5 ,LHIMHCH5 ,HHHnIM/LLHH{I,$u ID$LP0I/u IGLP0H;-)H;-o)u H;-) DHmu HEHP0EHCL5,HD$HD$LL`LHHH@HHo HLHt$HHHD$L5L,L`LL]IHH@HH LLHt$IHtH@H;D$ MwMIWIHI/uIGHT$LP0HT$HLHT$HT$HIbI.I I/u IGLP0I,$u ID$LP0Ld$IIEMiIHCHE,H2,0,GH, H , ,b IHHD$H$H>H=+HDŽ$[HD$H$H?H5+H|$BHD$H$Hv?H$H/kH=|+HDŽ$C[HHSH5X+HHD$BHL$HIFH)u HAHP0H=#+ZHH5GH5+HHD$FBHL$HHD$iHH)u HAHP0H)HD$hI9D$eIHt$L~HD$H$HJHT$HHD$HHu HBHP0I,$u ID$LP0H5+H$AIHMH$H/uHGP0H$HD$hHDŽ$H9GNLw}HD$H$H{I,$u ID$LP0H$H/uHGP0H$H=+HDŽ$HDŽ$HD$yYHD$H$HQH5+H|$@HD$H$HRH$H/uHGP0H$HHDŽ$˨Aă+TH$H/uHGP0HDŽ$Ex=H+ L(hE1ɹAAHHAHD$H$Hz^HD$HH$H/uHGP0H+HDŽ$u HCHP0HD$H;\)_`H|$H5+H_?HHq_H5+HHD$ DHL$ HIH$`H)u HAHP0H$H;=)H;=);H;=Ȩ);AąaH$H/uHGP0HDŽ$EeH5+H|$>IH$HhHH{HHmH$H/uHD$ HGP0HL$ H; A)HDŽ$H; ۧ)CH; )CHHL$ >HL$ ApH)u HAHP0EoH=+VIH$HsH5&+H=HD$ H1sH$H/uHGP0H5+H|$ HDŽ$=IH$H{Ht$ HHD$ HHu HFHP0H5+H|$1/IH{H$HD$hHDŽ$H9GTnHGH$H?nHWHHH$H$H/uHGP0H$H$HmHWH;)tH;)^qIHIH$Mx 1LH$LD$ I@HDŽ$r>LD$ HH&I(uHD$ I@LP0HL$ H$H/uHGHL$ P0HL$ H; )HDŽ$AH; )DOH; )OHHL$ HL$ A H)u HAHP0EAH|$tH!H:HSHfDH\f(Xf(\f(\H9u\W f(fT+W HHYHt$HHD$ HL$ HIH$iH)u HAHP0H$H;=)AH;=X)D|[H;=u)o[¦AąH$H/uHGP0HDŽ$H\$EH5q+H=z+1s3H5l+H|$IHP7H$H/uHD$HGP0LD$H5x+HLD$ HDŽ$OLD$ HHD$H$9H=+LD$LD$HIBH5+HLD$ LD$ HHD$H$BI,$uID$LD$LP0LD$I@E1E1H;D$hxFH;vk)4H;k)HcLD$FlLD$HIGMtL`IcLLLD$ H$HITAEH$HIT1HDŽ$HDŽ$LD$ HHD$H$SI/uIGLD$LP0LD$I(u I@LP0H$H;=j)AH;=4j)DH;=Qj)lAąEH$H/uHGP0HDŽ$E$H=+IH [H5*+HHD$=LD$HHD$ TZI(u I@LP0H=˭+HD$H$HcH5+H|$HD$H$HbH$H/uHGP0H5Ű+HHDŽ$HD$H$HbH$HD$hH9GOcHt$;HD$H$fH$H/uHGP0HDŽ$H$H/uHGP0H5+H|$HDŽ$ IH$H"eHt$HHD$HHu HFHP0HD$ HT$hH9PdH$H|$ :HD$H$HFcH$H/uHGP0HD$ HDŽ$HD$pHT$pHHD$HHu HBHP0HD$HH$HfHD$H$HoHt$H;5g)AH;5{g)DSH;5g)SHiAąpH$H/uHGP0HDŽ$E"TH$HL$HT$HHD$HHu HBHP0H$H/uHGP0L$I<$u ID$LP0Ht$L$HHD$HHu HFHP0HDŽ$Ld$Ǽ@MLLD$ $LD$ HIH$HzHM1E1H+H$E1++GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HGME1E1Hu+1E1E1o+~H$Y+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ H^GME1E1H+H$E1۰+fͰ+;EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$eHFME1E1HF+H$E1=+f/+>EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$DPHF+WHD$H++ DDDHD$ H@@=HD$H|$ H$H$H$sHD$ HXH$Ht H/uHGP0HDŽ$HeEME1E1H+H$E1+fԮ+@EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$uHDME11HW+E1E1H$K+y=+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ IL$HMl$HIEI,$uID$HL$LP0HL$IEH;4b)5H;b) HL$cHL$HI-CHHHD$1LLLD$I@ LD$HHD$H$dBI(M!I@LP0HvCME1E1H+E1E1H$+y+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 錾HBE11z+fHg+e+NEHD$(ME1Lt$HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$!H5̣+H|$0IHs=HDŽ$H+_)I9GS0IGH$H>0MOHII/uIGLL$ LP0LL$ H$H 0IQH;_)2H;U`)iLL$ `LL$ HIEH$1LLHDŽ$I@HEIh LD$(LL$ LL$ LD$(HIH$AI(uI@LL$ LP0LL$ I)u IALP0H$H11jH$E1E1HD$8ZYH@H$H/uHGP0HT$H;^)HDŽ$H|$(H5K+HGHH>PЅPHD$(E1HD$@HD$HHD$0HD$8HD$PHD$`HD$XHD$ H/@ME11H+E1E1H$+f+qEHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$4H?ME1E1H+H$E1 +y+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 騺HG@=H$L$LD$H$LD$HH$iRI(>I@LP0/DH>ME1E1H+H$E1 +y+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 騹HGH$HHWHHH$H$H/uHGP0H$H$HHGH;[)H;1\)t}\IHVH$M` 1LH$LD$I@HDŽ$LD$HHD$H$% I(ZI@LP0KHG@=nH$H$L$HD$H$H GH$Ht H/uHGP0HDŽ$H$H$L$HD$H$HuHi<ME11H+E1E1H$+f֥+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$wH;ME1E1HX+H$E1O+yA+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HF;ME1ؤ+yHŤ+E1H$+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ bH:ME11HD+E1E1E1=+gH$'+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ѵH,:ME1MH+H$E1+y+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0NH9ME11H0+E1E1E1)+gH$+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 齴Ht$H|$ )HD$ HiH8ME11H+E1E1H$v+h+$HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0Hv8ME11H+E1E1H$++GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 鍳EpH7ME1E1He+1E1E1_+gH$I+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$E۠+YHD$Ƞ+yDH+tH7ME1+yH+E1H$+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 3HD$H|$ H$H$H$HD$ HHT6ME11H۟+E1E1H$ϟ++*HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0t@HD$0MHD$(HIkH5ME11H5+E1ME1.+H$+HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0˰H&5M1E1H+H$E1+f+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$.H4ME1+H+H$E1+WHHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0鮯H 4M1E1H+E1E1H$+fv+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$0HD$(H=~+QHH1H5ޕ+HHD$0HL$0HIH$@1H)u HAHP0H$H|$(OAăk1H$H/uHGP0HDŽ$EJH5+H|$(IH$HC&H5܍+1HHD$0H%H$H/uHGP0Ht$(HDŽ$HHD$(HHHFHP0H=2ME1E1HÛ+E1E1H$++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ SH1M1E1H5+E1E1H$)++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ŬHt$0LMHD$(IHH0E11E1H+H$|+n+ILLd$(d@H0ME11HB+E1E1M;+H$%+ HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ϫH*0ME11H+E1E1H$++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0JH/M1E1H,+H$E1#++GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 鿪I@@=H$HLLD$ L$H$H$H$IcHkHLD$ HHD$H$HWR⍁-H$H$L$6HHPH$HtH/uHL$ HGP0HL$ HDŽ$I/9IGHL$ LP0HL$ H\$E1HD$@HD$HHD$8HD$PHD$`HD$XHD$ NH(ME1E1H\$1E1E1H/+H$)+n+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$鼣H(M1E1H+H$E1++NHHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0:HHL$ }HL$ IH$IH+Hl'ME1E1H+H$E1+ې+UIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 酢IE@=HD$LHL$ H$H$H$qHL$ HHD$H$RH)u HAHP0HT$HHD$HHu HBHP0MIQR⍁LϺH$LL$ H$H$LL$ HIH$GH$HtH/uLL$ HGP0LL$ HDŽ$H%ME11HB+E1E1E1;+aH$%+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$齠H%M1E1H+H$E1++PHHD$(HD$@HD$HHD$8HD$PHD$`HD$XDH5+H=+1IH$Hi2HH$H/uHGP0HY$ME1E1H\$1E1E1Hҍ+H$HDŽ$+o+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$SH#M1E1H5+E1E1H$)++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ŞH #ME11H+E1E1E1+[H$+DHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD$"H}"M1E1H+H$E1++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 闝H!ME1+Hq+E1H$d+UHHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0Hr!ME11H\$E1E1H++pH$ڊ+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD${H ME1E1H\+H$E1S+E+8HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HJ ME1E1HЉ+HD$hE1ʉ+H$+sHHD$0HD$(HD$@HD$HHD$8HD$PHD$`HD$XkHME1E1HL+H$E1C+5+>HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HCME11Hʈ+E1E1H$++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ZHME11H<+E1E1H$0+"+HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ̙H'ME1MH+1E1H$++K鏙HME11Hq+H$E1h+Z+KHD$(LHM1E1H.+H$E1%++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HME1E1H\$E1E1H++pH$+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$$H+WHD$H++ADHSME11Hڅ+E1E1H$΅++HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ jHME1E1HK+1E1H$@+2+KHD$($HME1+H+E1H$+XHHD$(HD$@HD$HHD$8HD$PHD$`HD$X魖HME11H+E1E1E1+hH$r+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ L HH>HcME1MH\$E1E1Hރ++rH$ʃ+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$kH5$u+H=+1IH/HI/u IGLP0HME1E1H\$1E1E1H+H$+q+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$铔HME1E1H\$E1E1Hi+k+pH$U+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$H$LLD$ L$H$H$H$IcHkH2LD$ HHD$H$OHM1E1Hy+H$E1p+hb+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ @H^M1E1H+E1E1H$ـ+ˀ+HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$ ~HM1E1H`+E1E1H$T+hF+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HBME11H\$E1E1E1H+H$+r+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$QHME11H\$E1E1H(+*+rH$+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$鵐LGMHGIHH$H$H/uHGLD$ P0LD$ H$HGH;2) )H;M3)(LD$(3LD$(HHD$  8Ht$ H$1HDŽ$H$LFHF IH>7HT$ HHD$ HHHBHP0tH#M1E1H}+E1E1H$}+h}+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ :H$H$L$HHHgME1MH\$E1E1H|+|+rH$|+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$oHME1E1HP|+HD$hE1J|+H$4|+iHHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HFME11H{+HD$hE1E1{+H${+gHHD$(HD$@HD$HHD$8HD$PHD$`HD$0nMLHLL$ ;LL$ HIH${HME11H${+E1MH${+ {+PKHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 鴌HME11Hz+E1E1H$z+f|z+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HxLME1Hy+E1E1E1y+H$y+IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0锋HME11Hvy+E1E1H$jy+\y+HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ Hax+WHD$Hx+x+GDbLϺH$LL$ H$H$0LL$ HIH$HME11Hyx+E1MH$mx+_x+VKHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$LHL$ H$H$H$ZHL$ HHD$H$Hw+fHw+w+UEE1M-H ME11H\$E1E1Hiw+kw+rH$Uw+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HQ ME11H\$Lt$ E1Hv+v+rH$v+GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$HD$ XH ME11H:v+HD$(E1E11v+H$v+KHD$0HD$(Hc ME11Hu+HD$(E1u+H$u+KHD$0HD$(黇v(H ME1E1Hu+HD$hE1u+H${u+qHHD$(HD$@HD$HHD$8HD$PHD$`HD$0;H ME1E1Hu+HD$hE1u+H$u+lHHD$(HD$@HD$HHD$8HD$PHD$`HD$0H M1E1Ht+H$E1t+ht+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$,H M1E1Ht+E1 t+hs+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ υH ME1E1Hs+E1E1H$~s+ps+EIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ H5d+H=r+1ſIH$H+HiH$H/uHGP0H/ ME1E1Hr+1E1E1HDŽ$H$r+r+)IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 7HME1E1Hr+1E1E1r+H$q+IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ 馃HME1E1Hq+HD$hE1q+H$kq+tHHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0"LHHT$ bHT$ HHD$`-HZLE1MHp+1E1E1p+IH$p+IHD$(HD$@HD$HHD$8HD$PHD$0HD$ }M`M{MhI$IEI(u I@LP0IEMAFHME1E1Hp+1E1E1p+hH$o+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$陁HM1o+hHto+E1H$go+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HcME1E1H\$E1E1Hn+n+rH$n+7GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$kHME11HMn+E1E1H$An+3n+CKHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ H8LE1MHm+E1E1E1m+H$m+eIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ KHLE1MH,m+1E1E1&m+H$m+cIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ~H5k^+H=ll+1eIHN/HI/u IGLP0HME1E1H\$1E1E1HUl+H$Ol+sAl+FGHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$}H=ME1E1Hk+HD$(E1k+H$HD$0H )k+KHD$(HD$}HME1IHik+HD$(1dk+H$HD$0H)Bk+KHD$(HD$/}HME1E1Hk+HD$(1E1k+H$HD$0HX)j+KHD$(HD$|H.M1MHj+E1E1j+fj+kEHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$j|HME1MH%j+Lt$"j+fH$ j+eEHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$HD${HME11Hi+E1E1E1i+H$ii+pKHD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ {HwM1E1Hh+H$Mh+h+jKHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ zHLE1MHrh+Lt$ E1lh+H$Vh+tIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ zH[LE1MHg+Lt$ E1g+H$g+qIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ oyHLE1MHPg+Lt$ E1Jg+H$4g+oIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ xH9LE1MHf+E1E1E1f+H$f+mIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0UxHLE1MH6f+E1E1H$*f+f+hIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ wH'uH M1E1He+E1MH$e+we+dKHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ !wH|ME1E1H\$E1E1Hd+d+tH$d+YGHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$vHE@=H$Ht$ HL$H$uIH&I.u IFLP0H$H/uHGP0HDŽ$Ld$P LLIHf#H$H/uHGP0HDŽ$LLd$PۘHLME1Hc+E1IH$c+c+RIHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ *uHM1MH c+E1E1H$c+hb+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ tHM1MH\$E1E1Hsb+ub+rH$_b++GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$tH[ME11Ha+E1E1H$a+a+KsHME11Lt$(E1Ha+a+H$a+LHD$({sHME1E1H\$E1HTa+Va+rH$@a+1GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$rH<ME1E1H\$E1H`+`+tH$`+[GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$GrHME11H)`+E1E1H$`+`+K rHeME1E1H\$1E1E1H_+H$_+t_+]GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$kqEKHME11HD_+E1E1H$8_+f*_+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$pHD$HIHLE1MH^+1E1I^+H$^+IHD$(HD$@HD$HHD$8HD$PHD$`HD$0HD$ 2pLl$pIL|$(1L$HvE1H\$PHl$xLH]+MH$E1]+]+IHD$(HD$0HD$ oHM1E1H]+H$E1]+]+)LoL$I1E1Ll$pHL|$(Z]+Hl$xHB]+LME17]+IH$HD$(HD$0HD$ oLl$pLHD$ E1L$HRL|$(1H\$PHl$xLH\+MH$I\+\+IHD$(HD$0nLl$pHD$0E11L$HL|$(E1H\$PHl$xLHZ\+MH$E1N\+@\+IHD$(HD$ )nL$MLl$pHtL|$(H\$P1E1LHl$xMH[+[+H$E1[+IHD$(HD$0HD$ mH ME11H\$E1E1H[+[+oH$u[+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$mHqM1E1HZ+E1E1H$Z+iZ+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$lHM1E1HjZ+E1E1H$^Z+iPZ+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$kHT$H|$(鰯H8ME11HY+E1E1H$Y+Y+HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ OkHME1E1H0Y+1E1E1HD$(H$Y+Y+KHD$0HD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$ jHLE1MHX+Lt$ E1E1X+H$X+vIHD$(HD$@HD$HHD$8HD$PHD$`HD$0HD$ 2jHME11HX+H$E1 X+W+LHD$(iHJME11HW+H$E1W+W+LHD$(iHME1MHW+1E1H$W+tW+LoiHM1E1HQW+H$E1HW+:W+'L5iHME1MHW+1E1H$ W+V+LhHSM1MHV+H$E1V+V+LHD$(hHM1MHV+H$E1V+V+LHD$(rhHME11HTV+E1E1H$HV+:V+2HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0gHHU+WHD$HU+U+:DHG@=]H$H$L$LD$ H$譚LD$ HII(I@LP0ރH$H$L$LD$ H$褱LD$ HIuHrM1E1HT+H$E1T+T+HHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ fH$Ht$ HL$H$IHLl$pLHHD$ L$HL|$(E1H\$PHl$xLHT+MH$IT+S+IHD$(HD$0eH=LE1MHS+1E1E1S+H$S+IHD$(HD$@HD$HHD$8HD$PHD$`HD$0HD$ ZeHSHME11H\$E1E1H)S++S+qH$S+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$dHR+iHR+R+FME1Lt$ 1HD$(E1HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ dHME11Lt$ E1E1HR+R+iH$Q+ FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ cHQ+iH{Q+yQ+ FHD$(HD$0sLgMHGI$HH$H$H/uHGP0H$HGH;)H;)}~HHL`H$1HH$HL$8HA HDŽ$HL$8HHD$H)'HAHP0HME11Lt$ E1E1HjP+lP+iH$VP+MFHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ bL@HMH@IHD$pHHHD$HHuHFLD$HP0LD$HD$pH@H;:)H;)LD$LD$HH#L@H|$p1HH$HL$ HDŽ$HA HL$ HHD$H$GH)tHAHP0eHnM1E1LD$Lt$ HN+N+iH$N+?FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$HD$ x`HME11Lt$ E1E1HON+QN+iH$;N+FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ _H@ME1E1HM+E1H$M+iLt$pM+jFHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ T_HME1E1H5M+H$/M+iLt$pM+dFHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$^HD$pH@@= H|$pH$H$L$LD$ H$蟑LD$ HHD$H$t6I(u I@LP0H$H/uHGP0HDŽ$tH}M1E1HL+E1H$K+iLt$pK+]FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ ]H|$pH$H$L$LD$ H$ާLD$ HHD$H$HM1E1H%K+E1H$K+iLt$p K+TFHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ \HME1E1Lt$ E1HJ+J+iH$xJ+9FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ "\H}ME1E1Lt$ HI+J+iH$I+3FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$HD$ [HG@=nH$H$L$H$|HD$HI,$ID$LP0釔H$H$L$H$yHD$HuHJME11Lt$ E1HH+H+iH$H+#FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ _ZHME11Lt$ E1H9H+;H+iH$%H+,FHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ YH*ME11HG+E1MH$G+G+^KHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ AYHME11H#G+E1E1H$G+ G+%IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ XHME11Lt$ E1HF+F+H$yF+IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ #XH~M1E1HF+E1E1H$E+E+ IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0WHM1E1HE+H$E1wE+iE+IHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ WLl$pLHHD$ L$HVL|$(E1H\$PHl$xLHD+MH$ID+D+IHD$(HD$0Vf(SLl$pLHHD$ L$HL|$(E1H\$PHl$xLHSD+MH$IGD+9D+IHD$(HD$0"VLl$pHHD$ E1L$HeL|$(1H\$PHl$xLHC+MH$IC+C+IHD$(HD$0UHD$XLl$pH\$PL|$(Hl$xHD$ L$HLHD$0eHME11HVC+E1E1H$JC+iLt$7C+rFHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ TH<ME1MH\$E1E1HB+B+rH$B+$GHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$DTHME11H&B+Lt$E1E1B+iH$B+sFHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$SHME11H\$E1E1HA+A+uH$kA+hGHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$ SHgM1E1H@+H$E1@+h@+EHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ RLl$pLHD$ E1L$HL|$(1H\$PHl$xLH?@+MH$I3@+%@+IHD$(HD$0RH$LH1L$Ll$pHD$ HGL|$(H\$HE1LHl$xMH?+H$Ht$@I?+?+ KHD$(HD$0QH$HE11L$Ll$pHD$0HL|$(H\$HE1LHl$xMH;?+H$HT$@E1-?+?+KHD$(HD$ QLd$ L|$(Hl$0H$L$Ll$pHl$xH:E11E1H>+HD$0H\$HH$LM>+Lt$ >+KHD$@HD$(HD$0HD$ sPH|$8HH Hl$0HDHPL|$(H5vH?(Ld$ H$L$H81Ll$pHl$x2Hl>+fH=+=+^EMHEME1E1H\$1E1E1H=+H$=+t=+XGHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$KOHME11H\$E1E1H"=+$=+sH$=+BGHD$(HD$@HD$HHD$8HD$PHD$`HD$XHD$0HD$ HD$NL$M1E1Ll$pHL|$(<+Hl$xHp<+LME1e<+IH$HD$(HD$0HD$ =NH$HE11Ll$pL|$(HD$ HyL$H\$HHl$xLH;+MHt$@E1H$;+;+KHD$(HD$0MH$HE11Ll$pL|$(HD$ HL$H\$HHl$xLHs;+MHT$@E1H$b;+T;+KHD$(HD$0=MLLl$pIHHD$ L|$(E11L$Hl$xHn;+H$H:+H$HT$@H\$HLMM:+JHt$8HD$(HD$0LLIL$1MLl$pHHD$ L|$(Hl$xH{:+H$H`:+H$HT$@H\$HLMMC:+JHt$8HD$(HD$0'LHC@=tuHt$ HL$H$ IH$HImuIELP0uHt$ HL$H$ IH$HuLIL$1MLl$pHHD$ L|$(Hl$xHY9+H$H>9+H$HT$@H\$HLMM!9+JHt$8HD$(HD$0KLIL$1MLl$pHHD$ L|$(Hl$xH68+H$H8+H$HT$@H\$HLMM8+JHt$8HD$(HD$0zJHLOIH$HL'tLH$H1Ll$pHD$0E1E1L|$(Hl$xH\$HHL$LH8+H$HT$@MHt$8E17+7+JHD$(HD$ ILH$M1Ll$pHHD$ E1L|$(Hl$xH\$HHL$LHx7+H$HT$@MHt$8Mb7+T7+JHD$(HD$0=IHD$ L|$(M1HHl$0H$Ll$pHl$xH7+HD$0Lt$(L$H\$HLHD$@HD$(MH$I6+6+JHD$8HD$(HD$0HLIH$HsLBqLH$H1Ll$pHD$0E1E1L|$(Hl$xH\$HHL$LH36+H$HT$@MHt$8E16+6+JHD$(HD$ GL|$(Hl$xLt$(Ll$pL$HHx.HH$H HEH(H5H81?HLE1MH5+E11E1HD$(H$m5+_5+JHD$8HD$(HD$0HD$ :GH(H5H81yLH$M1Ll$pHHD$ E1L|$(Hl$xH\$HHDL$LH4+H$HT$@MHt$8M4+4+JHD$(HD$0FH;(KIH$H>H$H/uHGP0H$HDŽ$HGHIHH$HHH$ӾHXx.H$H/uHGP0HDŽ$L$mLML$1Ll$pHL|$(E1HH\$8Hl$xLMIH}3+3+H$i3+JHD$(HD$0HD$ IEL|$(MLt$(Hl$xL$Ll$pH$H/uHGLD$ P0LD$ LD$ HDŽ$谒LD$ u6HH>HH 3H5ӼHEH(H81VLD$ H L1MH2+HD$(E1E12+H$t2+JHD$8HD$(HD$0HD$ ODL|$(Hl$xI1Lt$(Ll$pL$LLl$pHD$ E1L$HlL|$(1H\$8Hl$xLH1+MH$E11+1+JHD$(HD$0CHWHHGHHhH$1kLML$1Ll$pLHD$ HL|$(H\$8Hl$xLMHH1+J1+IH$11+|JHD$(HD$0CLML$1Ll$pLH\$8E1HUL|$(Hl$x0+LH0+MH$0+{JIHD$(HD$0HD$ BLLl$pLIL|$(H\$81E1L$HLHl$xHZ0+H$T0+MI@0+yJHD$(HD$ HD$0 BLLl$pHD$0E1L|$(H\$81E1L$HTLHl$xH/+H$/+ME1/+tJHD$(HD$ ALLl$pLHD$ L|$(H\$8E11L$HLHl$xHe/+H$_/+MIK/+qJHD$(HD$04ALLl$pHD$0E1L|$(H\$81E1L$HhLHl$xH.+H$.+ME1.+oJHD$(HD$ @L$M1E1Ll$pL|$(HD$(HHl$xH.+LH$Mx.+j.+`JE1HD$0HD$ P@L$M1E1Ll$pHL|$().+Hl$xH.+LME1.+_JH$HD$(HD$0HD$ ?MLl$pHD$ 1L$H"L|$(-+LHl$xH-+ME1-+]JH$HD$(HD$0s?L$I1E1Ll$pHL|$(L-+Hl$xH4-+LME1)-+XJH$HD$(HD$0HD$ ?Ll$pHD$ E11L$HEL|$(,+LHl$xH,+ME1,+VJH$HD$(HD$0>L$ML1Ll$pHD$0IE1HHl$xd,+LHN,+MH$A,+IJE1HD$(HD$ '>L$IM1Ll$pHl$xE1HeH++H$++LME1++GJHD$(HD$0HD$ =Ll$pHD$ E11L$HL|$(++LHl$xHs++ME1k++JH$HD$(HD$0L=H_HaHGHHH$H$H/uHGP0H$HGH;(H;(txjIH$H+HXHD$p1HH$H$HF nwIHH$H/uHGP0HDŽ$aHG@=sHD$pHt$ H$H$soIHH+`HCHP0`HD$pHt$ H$H$zIHuLl$pIHD$ E1L$H51L|$(LHl$xH)+MH$M)+)+#JHD$(HD$0;Ll$pIHD$ E1L$Hǿ1L|$(LHl$xHG)+MH$M;)+-)++JHD$(HD$0;Ll$pHD$ E11L$HZL|$((+LHl$xH(+ME1(+7JH$HD$(HD$0:Ll$pIHD$ E1L$H1L|$(LHl$xHn(+MH$Mb(+T(+1JHD$(HD$0=:L$LE11Ll$pL|$(HD$ HyHl$xH(+LH$M'+'+JIHD$(HD$09Ll$pHD$0E11L$HE1L|$(LHl$xH'+MH$E1'+x'+ JHD$(HD$ a9L$M1E1Ll$pHL|$(:'+Hl$xH"'+LME1'+IH$HD$(HD$0HD$ 8L$I1E1Ll$pH5L|$(&+Hl$xH&+LME1&+IH$HD$(HD$0HD$ }8ff.fAVAUIATUHSH0L(H (dH%(HD$(1H^H$LD$HL$LD$HHHt~ HtHLF0HM(HU fLHuoHuL%H|$(dH3<%(oH0[]A\A]A^f.LIH5N#+LIHVH$HH]DHHH HOHL LOLD@HH;(SH{H5H81H}C%+%+CH$+XZH TH=ё1f.HNIHHcH>fHF0HD$HE(HD$HE LHD$HEH$IHt~[HHu*M~.H5+LHVHHD$IMH4$HT$HL$LD$HLHLH5*lyMH߹wCr#+H_#+]#+wC_AWAVAUMATUSHHxH-+H$H|$HUH=E#+Ht$HHL$ LD$(HD$dH%(HD$h1H;W(IHHID$H5+LHHFHI$HH=I$HBIHLHHXIHb L%+H=}"+IT$LxHHHHCH5+HHHIHHMHHH5+LLLT$0LT$0<I*HEHHBL%(p VP A;$LLHHp VP A$p=9HZHmnI.DI/H;H-+H=J!+HUHFIHHID$H5+LHHHII$HMI$H$IFE1E1H5(Ht$0H9RH;)( H;(Hc LT$8LT$8HHMtLPHt$ Ic1LHHHtHt$(AD$HHHtAD$HHIELlAD$HHH\lIH8Hmu HEHP0I.u IFLP0I<$ID$u LP0ID$HH5+LHIMHCH5+HHHIMuLLHHI.u IFLP0I/u IGLP0H;-(H;-Z(u H;-|(# DHmu HEHP0EHCL-}+HD$ HD$LLxLHH&H@HH HLHt$HH HD$L-7+LpLLHIHH@HH LLHt$IHH@H;D$0 MoM IWIEHI/uIGHT$LP0HT$HLHT$̤HT$HIImI I/u IGLP0I.u IFLP0Lt$ IID$MIm|I/cI.wH;HCHP0uDHCLT$0HP0LT$0NIBLP0fHCHP0IGLP0H;f.IFLP0I/f.HEHP0I.f.4@9P ID$LP0ID$LP0@$ HC@=HD$ HHLT$@LT$8HD$HHD$(Ll$XHD$PHcHHtHd]LT$8HIoMI*IBLP0H\E11+<H+I+r#HIHu IFLP0MtI/u IGLP0HtHmu HEHP0H ++H=ή5y+dHtH+}1HCHHP0MtI,$u ID$LP0HH\$hdH3%( Hx[]A\A]A^A_fu;HP0H;0HHP0H0H0H0LcJH++#IELP0:HCHP0HD$ IcLLT$@HLT$8HD$HHD$(HtHLl$XHD$PH\$`oLT$8HIa MTI*JIBLP0;IGLP0IELP0uIFLP0zIELP0H5DM+9H5++"I$HQ 1E1E1E1@IBLP0cI*HH++<s+y#IE11HIHI 2HuLuH HE1+<H +I+t#H+HD$ HHLT$@HtHLT$8HD$HHD$(Ll$XHD$PHnLT$8HIH+9H++#MfI1E1E119tIWH5ǧIE1Y+<H5F+D+v#IEH=E1IELT$LP0LT$Mu1@LLHBHHHQE11+<H+I+{#HHuHsHH+9H++"H=rrKHuH.sHRH8+=H%+#+#H ++H=^5 +|HH"(H5H8HH5$M+=H5++#ILSMHkIHEH+GHEHE11MVMInIHEI.yHEIAsHuHqIHHP+:H++6#9H)+>H++#IH[(LH8H{+CHh+f+$>H1E1O+:H<+18+8#c.IpHI1+9H+ +#HU1 +>H +I +#HIEH +=H + +#;HW(LH8贿HmH ڣH i +k +C] +$E1E1lHE1< +:H) +IE# +;#HIE1HE1E1IHSI1 +9H +H1 +$#HHE1 +=H +I +#HH +>Hr +p +#IHH1U +:HB +IE< +=#HH& +:H +IE +B#HH5*H= +1XHHVH[gHmu HEHP0H% +?H + +#HH +:H +IEz +I#H-\HuH?nHTHI +:H6 +IE0 +D#HHCLT$8HHP0HELT$87ImHSH + +: +F#Hm1HIFLT$8LAIP0HELT$8Hy +CHf +d +~$HA +=H + +#dIMH +9H + +#H +?Hm +k +#CE1I}1MC$H=vI1E1E11HI1H1III.A@1E1I/,1AWAVAUATUSHHXHndH%(HD$H1HD$ HD$(HD$0HD$8HHLvLf Hn(H~01IHLLHH&H@eHCHHCAĩ-fA8H+u HCHP0IF!IFHHRAFAƩxfAH;-޻(L=*H=+IWL膽HH HHCH5+HHH IHHM HHH x*H=!+HQHH $H $HH HHCH5*HHHG HHHH HHIG11H;( H;(OH;`(Hcot$H $һH $t$HI HtHXHcƃHEHHcIlILIGHH L$茻H%(L$H Q;P  1LL$LHZL$P P =H @9H I(u I@LP0I/u IGLP0H}u HEHP0H+uH} HMIH $%H $AMHAL諷HH}HEHP0fLHIH5*LIHVHD$ H4Hk@HH(H3H5wUL AH H810H~+7p+Ha+XZH ›7H=ʛ1=qHT$HdH3%(HHX[]A\A]A^A_HCHP0 HCH $HP0H $cfE1HGHAHHu 襹HuH(H5JH8蚶腹AHH+HݚHl+n+\`+HCHP0HLwAHHu HuHQ(H5H8AHiHA+]H++E1H;-(8HL$AAMH*H=+HSH褸HHHHAH51*HH $HHH $HHHHHHu HAHP0|$ͶIHqHZ(H9CLHňHHaI/H+xHCHP0ifDHIHHcH>fHF0HD$8HC(HD$0HC LHD$(HCHD$ 贱IH~WHHu%H5*LHVbHD$8H'IMLt$ Ld$(Hl$0H|$8Hu@IG@=|HHLH $Ht(Hl$(HL$0H\$ EH $HH+cH+H{+LILII@H)VHt H+GMtI(u I@LP0MtuI/uoIGLP0cHHu +ZH**.fDHK*\H**fDH **H=*15*lUfH@`Hs HHc HIHR H(I9GhLxI/AnIGLP0_@IGLP0ZfHLHl$(Ht(HL$0H $H\$ }[H $HHEHtH+uHCH $HP0H $H)f.L8IH5*LIHVHD$0HDHk@HH(H7H5g}UL AH xH81 Hۊn*`*HQ*XZH H=Ί1-`HL$XdH3 %(HHh[]A\A]A^A_D$H;-i(L=l*H=*IWLHH HHCH5*HHH; IHHM6 HHu HCHP0L5*H=*IVL袨HH HHCH5*HHH HHHH HHuHCHL$HP0HL$IG11H;d( H;(H;ݦ(Hc|HL$t$Nt$HL$HI HtHXHcƃHEHHcIlILIGHH H(H QP ;# 1LLHH QP H= @9FH I.u IFLP0I/u IGLP0H}u HEHP0H%*uH} LuI质|$MLHLDܥHtH}lHEHP0]f.EvAfA=H;-פ(Dt$i|$HL$,MDkH*H=e*HSHaHHHHAH5*HHL$HH HL$HHHHHHu HAHP0Hc|$,舤IHH(H9C LHvHH I/H+CHCHP04AFAVHH HHcAH9H(H5xH89$D$HH~*H**.@EvoAFAVHH HcAH9JvDE1zCSHH ЉAH9MH֢(H5wH8jAH-H+H…HQ*S*E*#HCHP0DkH^IHHcH>fHF0HD$HHC(HD$@HC LHD$8HCHD$0$IHK~WHdHu%H5*LHVңHD$HH_IMLt$0Ll$8Hl$@H|$HHu@IG@=oHHLHl$8Ht8HL$@HL$H\$0T2HL$HH HtH+uHCHL$HP0HL$H)HAHP0LHcAH9AHh袢HZk@苢HH*Hr*p*.fDHÃ[*HH*F*!fDH 1*7*H=15$*Yf.H@`H HH HIH H(I9GLkI/AsIGLP0d@IGLP0fHFLHD$0ϛIH5*LHV虡HD$8HIH5*LHVtHD$@HIxDHi(H5sH8LH5>2IH"nDs@$f.HLHl$8Ht8HL$@HL$H\$0,GHL$HH*H*Hq*LILIIf.H)~Ht H+_MtI.u IFLP0MI/IGLP0H=kHH=***OHE1E1@HAHP0sf諟HuLLHHH*H**;sIH À[*H H*F*HHHCHP0I/HH***vI,$oE1LfDHCHP0f軞HuLKHHPH *H**bH {*H h*f*I_HJMwHII/-IFMHk*H**f1LLHHI.HH***kSsHuHVJHHIH~]*HJ*H*MH=1kDg HRHC(H5kH87HL$HfDH3~*H*H*RHdf.LsMLcII$H+u HCHP0ID$H;(H;U(͛IHLp1HLLx 2HHImLIELP0yIGHL$LMP0IFHL$fDH1豛H#4H }*H*1*_HT$0HLL}H5*I;H|O*H<*:*HH*(H |H5njL uAH|H81蜝HW|Y^H***wfDH@`HBHH2LHH!HHL57(L9fDHCHHHzHcH>DsAfH+HCHP0CSHH HHcAH9tHM(H5^mH8H+HCHP0Ds뗋CSHH HcAH9{H+D$HE1Q0#fH5!|H*HHHHuHyLL*H*1~*fdH2Hd(H5=wH8U;HH;(H5wH8,HxML*H*I$*|HHA`HtvHHtjHIHt]L9pu$L"cI/AuIGLP0fH57yH'IHuEI.;E17^HoH^(H57vH8OTHwLL7*H$*1 *oLLE15ff.fAWAVAUATUSHHhHndH%(HD$X1HD$0HD$8HD$@HD$HHHLvLf Hn(H~019IHLLHHH@UHCHHCAĩA)H+u HCHP0IFIFHH3HLHAH9 A/Dt$H;-(L=*H=*IWLÔHH HHCH58*HHH IHHM HHL=*H=^*IWLZHH HHCH5*HHH IHHM HH{IF1E1H;+( H;I(3H;(HcCLD$t$t$LD$HH MtL@HcƃHEHHcHlL|IFHH ҒL=k(H QP A; 1HLH角H QP Ap=4@9Hj H+u HCHP0I.u IFLP0H}u HEHP0H*uH} LuIy|$MLHAL迍H7H}HEHP0DL蠌IH5*LIHVfHD$0H$Hk@HH(HsH5eUL ayAH rH81舔HCs *8* H*XZH s8H=Is1HHL$XdH3 %(HHh[]A\A]A^A_HCHP0%fHCHP0vE1%H/HoAHHu HuHQ(H5cH8AHH+H=rH**]* HCHP0{|$HL$/AM͋H*H=*HSH裐IHHIGH5*LHHHIHHIHu IGLP0H|$/ӎIHH`(H9CLH`HHI.=H+&HCHP0@AFA9H(H5HbH8萌{D$HHpp*^H]*[* &fDD$fDAFA9cf.HIH@yHcH>fHF0HD$HHC(HD$@HC LHD$8HCHD$0$IH~WHHu%H5*LHVҎHD$HHIMLt$0Ld$8Hl$@H|$HHu@IF@=HHLLD$0Ht8LD$Hl$8L|$@TLD$HH *dHdoL* H*LMIMHI@I/vMt I(WHtH+u HCHP0MtuI.uoIFLP0c{HHnu*[Hb*`* .fDHnK*]H8*6* fDH !*'*H=n15*Cef.H@`H, HH HIH Hs(I9GLhSI/A~IGLP0o@IFLP0fHLLD$0Ht8LD$Hl$8L|$@2LD$HHMtI(u I@LP0I/RIGLP0CDHFLHD$0WIH5 *LHV!HD$8HIH5*LHVHD$@HIDH葋HH lIH w*y*bk* H2E1E1@IGLD$LP0LD$qLH5mpIHd%裊@$f.HuL7HH_H=l*dH**] 賍IKH l*dH **_ HHE1LI@LP0f;HuL7HHHk%*dH**b fH ck*dH **d [یIMFMI^IHI.]HCI޺Hj*dHp*n* f1HL蓋HH]H+HjH&*(*d* o[HI(H5R[H8DӈHuH5IHgH%j*bH** sH=V複cGkH2H(H5|VH8\KHHi3*bH *H* HfLcM LCI$IH+uHCLD$HP0LD$I@H;a(H;(LD$/LD$HI)L`1LHLp LD$DLD$HHJImLIELP0fDIFLD$LIP0HCLD$ffDH{h*bH*1*% HT$0HLLhH5~y*&JH,h *8H** HH(H gH5TZjL mAHhH81 HgY^HT* Q*8C* fDH@`HHH}LHHlHPH;(u`DHCHHHtVHօHAH9H+HCHP0H5.hHHHHP@CA9tH(H5 XH8SH+HCHP0f.CA9iDH+D$OHE1R0fHF0HD$HHC(HD$@HC LHD$8HCHD$0tIH~WHHu%H5*LHVzHD$HHIMLt$0Ld$8Hl$@H|$HHu@IF@=HHLLD$0Ht8LD$Hl$8L|$@ LD$HH*H[L*H*LMIMHI@I/vMt I(WHtH+u HCHP0MtuI.uoIFLP0c+yHHZ%*H**z .fDHcZ*H** fDH **H={Z15*/mf.H@`H0 HH HIH H#w(I9GL(;I/A~IGLP0o@IFLP0fHLLD$0Ht8LD$Hl$8L|$@LD$HHMtI(u I@LP0I/UIGLP0FDHFLHD$0rIH5*LHVwHD$8HIH5@*LHVwHD$@HIDHAwHH XIH '*)** H2E1E1@IGLD$LP0LD$qLH5}Y IHd%Sv@$f.vHuL~#HHbHW*Hr*p* cyINH WK*H 8*6* HHE1LI@LP0fuHuL"HHH=Wտ*H¿** fH W*H **[xIMFMI^IHI.]HCI޺HV3*H **#f1HLCwHH`H+HGVH־*ؾ*ʾ*.o[Hr(H5"FH8qDtHuHf!IHgHUm*HZ*X* sH=ABTsfGtH2HSq(H5,BH8 qvHHKU*Hн*H˽* HfLcM LCI$IH+uHCLD$HP0LD$I@H;r(H;lr(LD$rLD$HI-L`1LHLp LD$ LD$HHNImLIELP0fDIFLD$LIP0HCLD$ifDH+Tü*H*1* HT$0HLLKTH5ne*iJHS> o*kH\*Z*> HHJp(L YAjH5EH SH8HS1tHwSAX *kH*AY: *: @H@`HHHLHHpHPH;Wp(u`DHCHHHtVHqHAH9H+HCHP0H5SH~HHHP@CA9tHZo(H5;CH8nH+HCHP0f.CA9iDH+f|$MHE1R09H^IHtH5I/AIGLP0 HHgn(H5(DL WjAH 8QHQH81rHQAZ**kH*A[4 *4 HdQL*H*LM޹*AHHm(H5CL "WjAH PH`QH81@rHP[. H*]*kw*. I@@=LHt$0LD$Ld$0Lt$8lLD$HHI,$uID$LD$LP0LD$I.uIFLD$LP0LD$LhLHt$0LD$Ld$0Lt$8LLD$HHuHPLL*H*1* HmqnH_p^nHH^l(H57NH8Ok5nHH5l(H5NH8&kI.HxOH* ** I(I,$E1LqH0OLM·*H*I* Hk\jHNLL*Hv*1r* MLE1Wff.@AWAVAUATUSHHXHndH%(HD$H1HD$ HD$(HD$0HD$8HHLvLn Hn(H~01oIHLLBpHHH@HCHH HHmAH9 AH+u HCHP0IF-IFHH5H LNmAH9 A9H;-Wj(yL=Z*H=*IWLkHHs HHCH5t*HHH IHHM HHu HCHP0H *H=*HQHH $kH $HH HHCH5 *HHH HHHH HHuHCH $HP0H $IG11H;Lh( H;ji(tH;i(Hct$H $7jH $t$HI HtHXHcƃHEHHcIlILIGHH L$iHh(L$P ;P ] 1LL$LHiL$P P =H @9H? I(u I@LP0I/u IGLP0H}u HEHP0H*uH} HMIH $kH $MLHDDhHJdH}HEHP0LcIH5*LIHV~iHD$ H4Hk@HH-g(HJH5<UL yPAH IH81kH[J^*j*^HѲ*XZH 2JjH=J1HL$HdH3 %(HHX[]A\A]A^A_E1*E1H;-f(HL$MDDxgHɪ*H=r*HSHnhHHHHAH5*HH $HHH $HHHHSHHu HAHP0|$fIH\H%e(H9CLH8HH I/bH+HCHP0CSHH ЉAH9He(H5_:H8Od:gAHH+HHH!*#**GHCHP08DkAFAVHH ЉAH9He(H59H8cfAHHH*H**fEvwH.IHPHcH>fHF0HD$8HC(HD$0HC LHD$(HCHD$ t`IH ~WH$Hu%H5^*LHV"fHD$8HgIMLt$ Ll$(Hl$0H|$8HuO@IG@=HHLH $Ht(Hl$(HL$0H\$ H $HH HtH+uHCH $HP0H $H)OHAHP0@eHMHuF *H**.fDHKF*HЮ*ή*fDH **H=vF15*fH@`H HH HIH Hc(I9GL-I/AIGLP0@IGLP0fHFLHD$ _^IH5*LHV)dHD$(HIH5*LHVdHD$0H6IDHa(H56H8`NLH5+FIH".DL$bL$@$ZfHLHl$(Ht(HL$0H $H\$ H $HH'*H~D*!H*LILII@H)Ht H+/MtI(u I@LP0MI/IGLP0HCHH**{*HE1E1@HAL$HP0L$c;bHuLHHsHC%*H**CeI_H SC*H ث*֫* HHHCHP0LE1DHCL$HP0L$fH $waH $HuHVHH"HB]*HJ*H* H B3*H **Cf dHI_HIWHHI/HBI׾HB*H**1f1LLL$bL$HHpI(HAHN*P*B*<{c`HuH HHHmA*H**#H=-L$^L$_L$_L$HBH\(H5-H8\L$#bH $HtH@c*HP*HK*HlfH](H5 2H8*\hDLCMXLcII$H+uHCL$HP0L$ID$H;r](H;](L$A^L$HI>L@1HLLx \HHfImLIELP0HQ^HIGHT$LH $P0HT$H $HBI׺=fDH;]H->HS?*Hا*1ԧ*HT$ HLL?H5Q*H?M*jH**MHHr[(H Y>H5,1jL DAH?H81_H>Y^H,*I)*j*I?fDH@`H>HH.LHHH[(H9Cu)DH&H+AHCHP0HH5:?HHuHHZ(AH5@0jL CH V=H8H'>1^H=_CH;*AX;*j-*CQHHZ(L xCH5/jAH <H=H81^HJ=AY*jHͥ*AZ=ĥ*=ID$@=Ht$ LL$LD$ L|$(L$HHI(u I@LP0I/u IGLP0LH<D*H1*LI)**Ht$ LL$LD$ L|$(L$HH{HO<LL*HΤ*1ʤ*ZHHX(H5:H8WZH?HX(H5`:H8xW$I/H;HY*[*M*I,$I(jE1VH;ML*H*I$*HVHK;LLݣ*Hʣ*1ƣ*LLE1KAWAVAUATUSHHXHndH%(HD$H1HD$ HD$(HD$0HD$8HHLvLf Hn(H~01[IHLL\HH&H@eHCHHCAĩ-A9H+u HCHP0IF"IFHHSAFAƩyAH;-V(L=*H=*IWLXHH HHCH5*HHH IHHM HHH z*H=#*HQHH $XH $HH HHCH5*HHHI HHHH HHIG11H;T( H;V(QH;bV(Hcqt$H $VH $t$HI HtHXHcƃHEHHcIlILIGHH L$VH'U(L$H Q;P  1LL$LH\VL$P P =H @9H I(u I@LP0I/u IGLP0H}u HEHP0H*uH} HMIH $'XH $AMHALmQHPH}HEHP0LPPIH5*LIHVVHD$ H<Hk@HHS(H7H5)UL =AH 6H818XH6&*x*&Hi*XZH 6H=271E HT$HdH3%(HHX[]A\A]A^A_HCHP0fHCH $HP0H $afE1HGHVAHHu THuHR(H5*'H8QTAHH+H5Hl*n*)`*\HCHP0HLwUAHHu THuHQR(H5&H8PSAHgHA5ٝ**HƝ*ĝ*gE1H;-R(6HL$AAMNH*H=*HSHSHHHHAH5*HH $HHH $HHHHHHu HAHP0|$QIHqHZP(H9CLH#HHaI/H+pHCHP0afDHIH =HcH>fHF0HD$8HC(HD$0HC LHD$(HCHD$ LIH~WHHu%H5*LHVbRHD$8H'IMLt$ Ld$(Hl$0H|$8Hu@IG@=zHHLH $Ht(Hl$(HL$0H\$ H $HH*0H2*H{*LILII@H)VHt H+GMtI(u I@LP0MtuI/uoIGLP0cQHHu2 *'H**P.fDHK2*)HК*Κ*ZfDH **H=215*MfH@`Hs HHc HIHR HO(I9GhLI/AnIGLP0_@IGLP0ZfHLHl$(Ht(HL$0H $H\$ }H $HHEHtH+uHCH $HP0H $H):HAHP0+HFLHD$ IIH5*LHVOHD$(HIH5(*LHVOHD$0HFIDH0HH/*1*.#*HRE1E1@HAL$HP0L$LH51*IH~?fL$WNL$@$f.NHuL~HHJH/*0Hr*p*cQI6H /K*0H 8*6*HHHCHP0LE1DHCL$HP0L$fH $MH $HuHHHH%/*0H**fH .*0H *~*CfkPHI_HIWHHI/HBI׾H{.*0H**f1LLL$OL$HH3I(H.H**0*_KHJ(H5H8zIDHJ(H5H8ZI[D;LHuHHHH-%*.H**CH=L$KL$ L$KL$HHH(H5H8HL$NH $H^H,*.Hp*Hk*H,fLCMcLcII$H+uHCL$HP0L$ID$H;I(xH; J(L$JL$HIL@1HLLx HH"ImLIELP0IGHT$LH $P0HT$H $HBI׺fDH+k*.HX*1T*HT$ HLL,H5=*H+*H**HHG(H *H5jL =1AH+H81dLH+Y^H***'fDH@`H:HH*LHHHH(H9Cu)DHH+AHCHP0HH5+]HHuHHG(AH5jL Q0H )H8H*1wKH2*_ H*AX** 9H**0H*LI*HHpF(L /H5*jAH A)H)*H81JH)AY3*H *AZ*ID$@=Ht$ LL$LD$ L|$( L$HHXI(u I@LP0I/u IGLP0LHt$ LL$LD$ L|$(L$HHuH(LLe*.HR*1N*4GHAH4E(H5 'H8%D& GHxH E(H5&H8C]I/HN(Hݐ*ߐ*.ѐ*I,$I(E1bH (ML*.H*I$*HC4CH'LLa*.HN*1J*LLE1'AWAVAUATUHSHXH^dH%(HD$H1HD$ HD$(HD$0HD$8HHH^Lf Ln(H~01iHIHLH"IIHH@H/ID$H0HH/ LFHH:fI,$u ID$LP0HCLAHCHHHO H'FIIJL;-;C(}L=>*H=*IWLDHH HHCH5X*HHH IHHM HHu HCHP0H φ*H=x*HQHH $pDH $HH@ HHCH5*HHH HHHHY HHuHCH $HP0H $IG11H;0A(e H;NB(H;B(HchHL$4$C4$HL$HIg HtHXHcƃIEHHcMlILIGLM^ L$BHnA(L$H Q;P  1LL$LAIBL$P P =H @9 M I(u I@LP0I/u IGLP0I}u IELP0H*AuI} IMIH $lDH $MLHHLt@H,=I}IELP0fL(H9C LHoIH I/H+HCHP0Al$AD$HH f.Al$fDDcCII CDDc5HIHx*HcH>f.HF0HD$8HE(HD$0HE LHD$(HEHD$ :IH+~WHDHu%H5*LHV?HD$8H IM; H\$ Ld$(Ll$0H|$8Hu@IG@=HHLH $Ht(Ll$(HL$0H\$ 5H $HI HtH+uHCH $HP0H $H)HAHP0>HH *H**$.fDHs*H`*^*.fDH I*O*E1H=* 5;*&Hr<(H5H8;I/u IGLP0=HHI,$HJHه*ۇ*͇*0oID$LP0_H;(H5H8:H+u HCHP0{=IHHn*H[*Y*;@H@`H HH LIHp HHH;(H$H9f. IGHHHjL>HI/kIOLQ0\IGLP0HFLHD$ 6IH5*LHVIHHHfL$g;L$@$f.HLLl$(Ht(HL$0H $H\$ H $HIHMIz*Hg*e*LMM@H)Ht H+/MtI(u I@LP0MI/IGLP0H[HH**ۄ*\HzE1E1@HAL$HP0L$c:HuL~HHH*Hr*p*c=IH K*H 8*6*HHHCHP0HE1DHCL$HP0L$fH $9H $HuHHHH%*H**H *H *~*CfkHCHHtoHtREH46IH+HSHR0HH5f HH-HPH+HR0DcCII H+DcHHm2(AH5(jL H >H8H616H_H#~*AX#~*~*YHH2(L `H5jAH HH81w6H2AY}*H}*AZ}*HE@=bHt$ HL$LD$ L|$(L$HII(u I@LP0I/u IGLP0HMHMI$}*H}*}*Ht$ HL$LD$ L|$(jL$HII(u I@LP0I/u IGLP0H92HH0(H5vH8/t2HHt0(H5MH8e/I/HHF|*H|*:|*HmI(E1HrMH|*H{*HE{*H.HB`H8HH(HIHH$I9Gu$LI/IIGLP0LH5IHuHA`HtkHHt_LHHtRH$H9Au,HH $=H $HH)[HAHP0LHH5>HHu0HH.(H5H8-H#E1MHHz*z*z*s30HH.(H5\H8t-HMH\z*HIz*Gz*|LHE1l@AWAVAUATUHSHH^H|$dH%(H$1H].(HD$`HD$hHD$pHOn*HD$0HD$xHm*HD$8H$H8Ht$~(HtH`HF8HD$8HE0HD$0L}(L=-(H7HELe H$H$HI$l)HHHYn*HHHEHGn*HH=mq*HHHmH5n*HHL$HL$HI H8H)H-9q*H=x*HUH.HH HHCH5u*HHHIMb H+uHCLL$HP0LL$H+(I9AH4$LLL$LL$HHD$I)u IALP0H$HH$HHu HCHP0H-fp*H=x*HUH .HHHHCH5t*HHHIMH+H*(I9@dLLL$UL$HII(I,$L;=+(SH-o*H=]w*HUHY-HH HHCH5q*HHHK HD$H|$H+u HCHP0IGOH\$HH$HH &HHN%IH@L8IH|$H5l*HGHHIM\LL$,L$HHD$(I(u I@LP0H|$(IEH5l*LHHcIMeLL$y,L$HHxI(u I@LP0H^ H5'd*H=pu*1iHH HH+u HCHP0H xu*Heu*cu*ME1E1E1HD$E11HD$(HD$ HD$afHD$HEHP0HL$HAHP0)H@HL$LP0HL$H) HCL$HP0L$$fID$LP0PI@LP0I,$;f{-IL$IH5m*LIHV~*HD$`HTH]@HH H HMHSH LD@H((H5L XH81,HG s*s*Hs*XZH  H=1H$dH3 %(HS0HĘ[]A\A]A^A_DH6IHHcH>f.HF8H$HE0HD$xHE(HD$pHE LHD$hHEHD$`h#IHHHcH>f.1 #IHSH8 E1E1E1Hr*r*|r*ؑfHD$M1E1HD$(HD$ HD$Hl$LHx%HMHHQH $HUHIfHE1E1E1Hr*H$E11 r*vE1E1q*CHD$HD$(HD$ HD$HD$DMt"I(uI@L\$0LL$P0L\$0L$MtI*uIBL$LP0L$MtI+u ICLP0H `q*fq*H=15Sq*>MtI.u IFLP0MtI/u IGLP0HtH+u HCHP0MtImu IELP0HT$HtHH$HHu HBHP0Ht$ HtHH$HHu HFHP0HL$(HtHH$HHu HAHP0HT$HtHHH$HHuHBP0Ht$HH$HHu H|$HGP0I,$ID$LP0DHE1E1E1H p*H$E11p*vE1E1E1o*HHD$HD$(HD$ HD$HD$HtAHmu:HEL\$HHLT$@LD$8LL$0H $P0L\$HLT$@LD$8LL$0H $Ht6H)u0HAL\$@HLT$8LD$0L $P0L\$@LT$8LD$0L $MpI)fIAL\$8LLT$0L$P0L\$8LT$0L$;H[E1E1E1Hn*H$E11n*vE1E1n*KHD$HD$(HD$ HD$HD${$HuH^HZ5HE1E1E1HVn*H$E1E1Nn*y@n*YHD$HD$(HD$ HD$HD$JfHkE1E1E1Hm*1Hm*ym*[H$E11E1HD$HD$(HD$ HD$HD$fDHCHP0HEP0:@k#HuHNH7HXm*}HEm*Cm*ME1E1E1HD$E1E1HD$(HD$ HD$@HFLHD$`?IH5=i*LHV #HD$hHIM2HD$`Ld$hL|$pH$HD$xHD$0H$HD$80H5Ab*LHV"HtHD$pIM~H5 i*LHV"HtHD$xIM~H59a*LHV]"HtH$IMQHT$`HLLH5*+HHk*Hk*k*Hk*Hk*k*~E1E1E1HD$ME1HD$(HD$ HD$H5aa*LIH011HH$L$HHdI(uH$I@LP0H $H5a*LH $躵H $HI1ҾHHL$H$L$HL$HHI(uI@H $LP0H $HϺHH $!H $HIH)uH$HAHP0L$HmuHEL$HP0L$L;(L;w(L;(LL$ L$I(u I@LP07H|$H5_*HGHH5IM11LL$L$HHI(u I@LP0IEH5_*LHHIM'11LL$+L$HHrI(uH$I@LP0H $HκHH $H $HIcHmuHD$HEHH $P0LD$H $H)uHAL$HP0L$L;U(L;(8L;(+LLD$^LD$$QI(u I@LP0 $fIWHBpHRH@HEH5X*LHLHHD$c LD$HH4I(u I@LP0H5^*H|$H^11HHD$HL$HIHH)uHD$HAHP0LD$HS HCHHH9H9IHSLHHCI(u I@LP0H5\*H|$'HHH(H9ELMMLeII$HmuHELL$HP0LL$ID$H;(H;(NLL$gLL$HILH1HLHHX HD$xLT$HHI*uHD$IBLP0HL$I,$uID$HL$LP0HL$H5\*HHL$HL$HHH)u HAHP0H5=\*H|$HH11HHD$޿HL$HIH)uHD$HAHP0LT$HEH;8(HMHBLeHI$Hm%ID$H;'(H;(Hc+t$ LT$HL$HL$LT$HIt$ HtHHHX*HcLLL$HHITF1LHMTرLL$HII)uHD$IALP0LD$I,$uID$LD$LP0LD$IH&(I9FLLLD$LLD$HHD$HmuHELD$HP0LD$HD$H@H;(HD$HPHHD$HHHL$HH H@(HL$ HD$(HD$HL$HHD$ HHD$(HHHD$HHuLD$HAHP0LD$H|$0LD$H5^*³LD$H|$0LD$H5!X*蔳LD$1t2H|$0LD$H5Z*jLD$&$H=[*LD$@LD$@HHD$I H5`*H|$LD$@LD$@HIHL$HHD$HHuHALD$@HLL$P0LD$@LL$H=U[*LD$@LL$LL$LD$@HIP$H5^*HLD$HLL$@HD$ZLT$LL$@HHLD$H#I*uIBLD$@LLL$P0LD$@LL$H54`*H|$(LD$@LL$LL$LD$@HIH#Ht$ HLD$HLL$@HD$~LT$LL$@HHLD$H%I*u(HD$IBLLD$HLL$@P0LD$HLL$@HL$HEE1E1ҺH;(%H;2( H;(HLT$PLD$HLL$@HL$HL$LL$@HILD$HLT$P$MtLPIc1LHHLD$PILHL$(AD$HLL$HHILL\$@ͭL\$@LL$HHHD$LD$PZ(I+uICLD$HLLL$@P0LD$HLL$@HmuHELD$HHLL$@P0LD$HLL$@LD$HLL$@LL$@LD$HHH%HD$LD$HLL$@HEIELm LL$@LD$HHH%HT$8H5V*HLD$HLL$@HD$3HL$LL$@LD$H* HT$8H5]*HLD$HLL$@HL$HL$LL$@LD$HX HLHLD$HHL$@LL$8uLL$8HL$@HHD$LD$H(I)uIALD$@LHL$8P0LD$@HL$8HmuHELD$@HHL$8P0LD$@HL$8H)uHALD$8HP0LD$8H|$H;=(H;=W(H;=u(LD$8LD$8'!H=YW*LD$0'LD$0HH5Z*HLD$8HD$0pHL$0LD$8HH H)uHALD$0HP0LD$0H=V*LD$0LD$0HIH5S*HLD$8HD$0LL$0LD$8HI]I)uHD$8IALLD$0P0LT$8LD$0Hd(I9B Ht$ LLD$8LT$0LT$0LD$8HII*uIBL\$8LLD$0P0L\$8LD$0ICH5K*H@pHH@HLD$8LL\$0L\$0LD$8HHI+uICLD$8LHL$0P0LD$8HL$0Ht$(HLD$8HL$0HL$0LD$8HIH)uHD$8HAHLD$0P0L\$8LD$0HE1H;=(|H;[(H;(HczL\$@LD$8HL$0!HL$0LD$8HIL\$@ HtHHHc$I1LHLD$0MD@HL$M\L$LD$0HI I*uH$IBLLD$0P0LD$0L $HmuHELD$0HL $P0LD$0L $I(uI@L $LP0L $Ht$LL $ L $HH I)u IALP0HE IHHELL$HH5Q*HHiL$I+u ICLP0MHEIfDIiHIYHEHI)u IALP0HCH;(X H;(nIHHhH$1LHLD$HI@ 耧LD$HHD$/I(I}I@H\$LP0LL$d$DHE1E1E1HQZ*SZ*yEZ*iH$E11E1HD$HD$(HD$ HD$HD$HuHּH#HHE1E1E1HY*E1E1Y*zY*HD$HD$(HD$ HD$IHE1E1IHaY*cY*zUY*HD$E11E1HD$(HD$ HD$@MHMIXIHI(uI@L $LP0L $HCH;t (kH; (0L $CL $HH LH1HHI$L` YIH HmIHEH$HP0L$HE1E11HBX*E1AX*z3X*HD$HD$(HD$ HD$Bf.H[W*HW*W*6E1E1#H|$IYfDHME1E1HW*W*W*81H5E*H=2W*1+HHt HײH+u HCHP0H:W*H'W*%W*H HkHME1HV*E1E1E1V*}1E1E1V*HD$HD$(HD$ 8@HD$IHV*HpV*nV*ZfHME1E1HAV*CV*5V*\HC@=#H$Ht$`HHl$`HD$h+HD$HLHmu HEHP0I@H+U*HU*U*fH|$IHHI`HHHE1E1E1HUU*E1TU*FU*fDHD$ME1HD$(HD$ HD$OH (HRH5FH81 HOT*HT*T*’jDH#T*HT*T*HC@=Ht$`HL $LL$`Ld$h蜙L $HII)u IALP0IHE1E1E1H%T*1%T*T*HD$M1E1HD$(HD$ HD$$ IeH4S*HS*S*E1E1E1HS*HS*S*HELT$HHL$P0ID$HL$LT$ID$@=HF*HLHL$`HL$HD$hHcHLT$pHthLT$:LT$HL$HIpHt$H)uHALT$HLD$P0LT$LD$I*IBLD$LP0LD$ID$@=Ht$`LLL$`LL$H\$h街LL$HH I)IAHL$LP0HL$HME1E1HR* R*R*H5n@*H=Q*1踞HH HdH+u HCHP0H/Q*HQ*Q*JH$Ht$`HHl$`HD$h HD$HHrQ*yH_Q*]Q*oE1E1E1I1jH=Q*~H*Q*(Q*HD$-HpQ*HP*P*dHIE1E1P*HP*P*Ē2HP*HP*P*fHP*HzP*xP*ђHt$`HL $LL$`Ld$hӬL $HIH6P*zH#P*!P*IE1E11HD$E1HD$(HD$ HD$HEO*HO*O*iME1E1E1HD$1E1HD$(HD$ HD$HE1E1{O*HhO*fO*ӒHD$ME1E1HD$(HD$ HD$H'O*HO*O*kELHLD$LD$gHJME1E1HN*E1N*N*֒HD$HD$(HD$ HD$HN*HxN*vN*HdN*HQN*ON*nHt$`LLL$`LL$H\$h詪LL$HHHsE1E1E1HM*M*M*LHA*LHL$`HL$HD$hHcHLT$pHthLT$2LT$HL$HIHE1E1M*HvM*tM*$HME1E1HSM*UM*GM*qHH7(H5L jAH HH81Hd^_HL*L*L*H6E1E1L*yHL*L*}ZH E1E1E1HL*L*L* HE1E1IH_L*aL*ySL* HE1E1E1H/L*E1.L* L*HvE1E1E1HK*K*K*HFK*HK*K*D,1I1HE1E1IHK*K*zK*.1IHIE1E1HUK*E111PK*zE1?K*HD$HD$(HD$ HD$TMNMMInIHEI.uIFLD$LLL$P0LL$LD$HEH;]'H;'LD$LL$&LL$LD$HI@LH1HHIELh LD$ HD$,LT$LD$ HHD$I*IBLD$LP0LD$HE1E11HJ*J*J*H]E1E1I*HI*I*ZE11HD$(MME1HD$ HD$H;'H|$LD$`LD$HHHT$HHD$HHuHBLD$HHL$P0LD$HL$HALD$ HHL$HHL$LD$ HHD$LD$@HHL$LD$@HHD$ HHL$LD$@HHD$(HվH(HL$LD$@H)HE1E1H*HH*H*4HdHx6HHL$H HEH*'H5KH81L$HME1MHH*LD$E1H*H*HD$HD$(HD$ HD$H4E1E1E1HG*G*G*HG*HG*G*ME1ME1HD$E1HE1E11HLG*NG*@G*?NHE@=qHc$HHHL$`HL$@HkLD$hLD$8L\$pHthL\$0 L\$0LD$8HIHL$@Ht,H)u&HAL\$8HLD$0L $P0L\$8LD$0L $I+ICLD$0LL $P0L $LD$0`HE@=HD$(HLD$XLL$PHD$pIcHkLT$`LT$HHL$hHthHL$@XHL$@LT$HHHD$LL$PLD$XgMt.I*u(IBLD$PLLL$HHL$@P0LD$PLL$HHL$@H)HALD$HHLL$@P0LL$@LD$HLLD$8L\$0褬LD$8L\$0HHD$HPHHD$H@HHL$HHH@HL$ HD$(H:E*H'E*%E*ME1ME1E1IHlE*HD*D*RHc$HHL$`HL$@HkLD$hLD$8L\$pHthL\$07L\$0LD$8HIHL$@HD*H|D*zD*ME1MME1HYD*HFD*DD*ǓH52*H=C*1L$L$HHD$'Hl$LD$0H肟HELD$0H$HHEuHEL$HP0L$H4C*HC*C*ݓ+H E1C*HC*C*ME1M1E1HmC*HZC*XC*ME1ME1I|LHME1MHC*LD$E1C*C*HD$.HQB*HB*B*H*ME1IHB*E1B*B*$HE@=Ht$`HLD$ LL$`LL$Ll$h葇LL$LD$ HHD$I)IALD$LP0LD$HLME1ME1E1HA*B*A*+MJMIJIHI*u(IBLD$@LLL$8HL$0P0HL$0LL$8LD$@HAH;+'v H;'c LD$@LL$8HL$0HL$0LL$8HILD$@LHHD$ 1LHLD$@HIB LT$8HL$0HL$0LT$8HILD$@ I*k HD$@IBLLD$8HL$0P0HL$0LD$8L\$@IH-@*H@*@*ME1M1E1HD$(HLD$XLL$PHD$pIcHkLT$`LT$HHL$hHthHL$@HL$@LT$HHHD$LL$PLD$X9HME1MH @*E1@*@* JH'H5L$H81L$H=?*H?*?*7ME1ME1HD$I1H?*yH?*?*wHm?*HZ?*X?*8HMHwLeHI$HmlID$L$FHsME1MH>*E1E1>*>*ڔ H=ME1MH>*1E1>*>*HD$H>*H>*>*HD$ME1ME1'H[>*HH>*F>*HE14>*H!>*>*הME1MME1IL$HXME1E1H=*I=*=*"H|$0LD$8H5,2*蟍LD$8 iH=2*LD$0۠LD$0HIH51*HLD$8HD$0!L\$0LD$8HHI+uHD$0ICLLD$8P0LD$8HL$0H5x+*1HLD$8HL$0̉HL$0LD$8HHH)uHALD$0HP0LD$0HmHELD$0HP0LD$0H<*H<*<*͓ LUMLeII$HmID$LAHMMA<*H.<*E1)<*HD$YHvME1MH;*E1;*;*HD$H:;*H;*;*HME1E1H;*E1M;*;*Hu;*Hb;*`;*HN;*zH;;*9;*HME1MH;*E1E1;*;*5HD$3HPME1MH:*LD$E1:*:*-HD$sHt$`HLD$ LL$`LL$Ll$hLL$LD$ HHD$-HE1E1c:*HP:*N:*`rHD$(H)uHAL$HP0L$L$ߙL$u7HT$(HmH eH5HHEH'H81L$H;ME1MLT$ Hl$E1E1H9*19*9*HD$HD$(HD$ HD$HE1j9*HW9*U9*tvHE11>9*H+9*)9*nHD$FHuHD$ HD$(LT$ LL$(ME1HGHl$M8*H8*E18*HD$HD$(HD$ HD$H8*Hv8*t8*(HELT$PHLLD$HLL$@HL$P0ID$ALT$PLD$HLL$@HL$|HA@=HD$ HHt$`LD$@LL$`LL$8HL$0HD$h}HL$0LL$8HILD$@QI)u(IAL\$@LLD$8HL$0P0L\$@LD$8HL$0IHME1MHp7*LD$E1j7*\7*HD$(HD$ HD$HD$kH&7*H7*7*HME1ME1E15HXME1MH6*IE116*6*9HEL\$@HLLD$8HL$0P0ID$$L\$@LD$8HL$0HE1E1q6*H^6*\6*-jHE1E1E1H86*:6*,6*:HTHD$ HHt$`LD$@LL$`LL$8HL$0HD$hoHL$0LL$8HILD$@H/ME1MH5*E15*5*H逾HME1MHz5*E1y5*k5*_HME1MHG5*E1F5*85*]kH5|#*H=4*1L$ځL$HHtLHL$肐HmL$uHEHP0L$HD4*H4*4*H4*H4*4*vH4*H{4*y4*ScHME1MHU4*E1T4*F4*ŔHE1E1.4*H4*4*h=Ho4*H3*3*jHH3*H3*3*ٓH!3*H3*3*%HME1MH3*E13*q3*HME1MHM3*E1L3*>3*єwHME1MH3*E13* 3*˔DfDAWAVIAUATUHSHH^dH%(H$1H/'HDŽ$H$H Hx HLn H]HDŽ$HHDŽ$HDŽ$HD$HKH2* H(hHE1AAHHHcH$H8uH@HP0H$H-'*HCHDŽ$H=1*HUHHD$IHHHH|$jIH>IG1H;'HDŽ$H;' H;^'HcEt$t$HHH$H$HtHEHDŽ$HcƃIEHHcLlLdIGLM?zL-'H QP A;U1HLAHMH QP AE=@9 HH$H$H/uHGP0HDŽ$I/u IGLP0L-(*H=o0*H$HDŽ$IULWIHKHID$H5c$*LHHH$HeI,$u ID$LP0L-D(*H=/*IULIHHIGH5+*LHH3IIHM.IHu IGLP0L$E1E1IAH;' H;'H;$'Hc H$HMtLxHEIcL$AHMcIlOdH$HALMHL$IL-'H QA;UHL$P *1LHAIH QP AE=W 29 ML$H$H/uHGP0HDŽ$H$H/uHGP0L$HDŽ$IEL$HDŽ$IGAwI HD$HH.*Mf HD$ HZ)*IT$HHD$(HHT$xHD$0HH@HT$HHq LH|$0HD$0HMf H)*IT$HHD$(HHT$H+HHHT$LMj LHAH$HHHH; ]'HDŽ$ HPH$H H@HHH$H$H/uHGP0H$H$Hl oH$HfH$H/uHGP0HDŽ$H$H/uHGP0H$HDŽ$H/uHGP0HDŽ$HHpHpxHD$PHHL$XHt$8HD$@HtHHD$8HtHHD$@HtHHL$E1H|$ HD$`HD$HHD$(Ll$hMLd$HHl$pL|$HH\$xLH|$DK,L|$fI@AH{IIL$$L$AFXL9|$(uff.HD$^ITI f(ʐEHYEH9uIH9T$ [Ll$hL|$HHl$pH\$xH|$`}H|$Xt HL$XHHD$HHu HAHP0HL$8HtHHD$HHu HAHP0Ht$@HtHHD$HHu HFHP0Lt$0H5o*1LvIIHD$HIM,I,$>ID$LP0.H]DHHzH jHOH3L (LOL@HH'SHH5mH81;H)*#{)*Hl)*XZH Ϳ#H=1HH$dH3 %(HH[]A\A]A^A_@L-'@IG@=H$HHLHL$H$L$mHH$H`H(*H(*(*[IH@fD2H۾s(*H`(*^(*MIHIHu IGLP0H$E1E1Ht H/uHGP0MtI,$u ID$LP0H$Ht H/uHGP0H$Ht H/uHGP0H '*'*H=ݿ5'*袔HtH+ugE1HCHLP0MtI/u IGLP0HtHmu HEHP0M ImIELP0f1fDLl$hL|$HHl$pH\$xH5!AHv'H8DL%VoH|$`L%&*&*&*OH$Ht H/H$HDŽ$Ht H/H$HDŽ$Ht H/H ^&*d&*H=yHDŽ$5G&*2H|$PH$H$H$qP H$H$1H$H H|$01HHD$rHt$0HL$IHHD$HHH) M L;5#'L;5'L;5'L1I.uIFT$LP0T$ H$N H/uHGP0H$HDŽ$H/uHGP0H$HDŽ$H/uHGP0HL$@HT$8HDŽ$Ht$XH|$PkjIEMH+LIHcHHHH5!*LIHVIH$HdMH$L$IA@=IHLL$JH$L$%iH$H+ MtI/u IGLP0I,$ID$LP0@;@$ fH$HLL$HL$H$HH$HH$Ht H/BHDŽ$I,$ID$LP0@DHcE11E1H"*1"*|"*H$HH/E1fH E11E1H"*H$E1"*}{"*'HfD[HuH>Ht IHE11E1H2"*H$,"*"*?fHD$0HfHGP0@HGP0T@HGP0!@HGP0W@HFHL$HP0HL$HAHP0H|$0HGP0bI.fDH1Q!*H>!*I9!*AHH$sH$HHm!*H * *Ht$0HHD$HHu H|$0HGP0H$E1fDHH$+@$.fIH$HIoHHEI/u IGLP0HEIILL$JH$L$|H$H H_*H**ME1HuL螂H(HE1E1*H*H$*|RH۵Mp*H]*I$E1T*~HCHfDH+*H**bIHf.HuL΁HpH@E1E1*H*H$*1HLH$H fH1E1~*Hk*If*mHDSIH ;*H (*&*H=$<@HDŽ$gHuH'H5H8fDHF HH$HFH$HKH$HLLH5)^s"Hѳٗd*#HQ*O*ٗf.HFHH$|I@MyMIAIHH$H$H/uHGP0L$AIAfDL%***HL$@HT$8Ht$XH|$P4bHI'Ht$(H8HϲH$E1\*HI*G*fH5*LHV]HH$IEkH'Ht$(H8,HDŽ$HK*H**f1LHH$HOfHE1E1*Hz*x*nH=atHL$HDŽ$f+HuHg'H5@H8 fDHs *H**H$H|$L$HT$bH|$HT$LHH`L%*HDŽ$HDŽ$HDŽ$*{*HѰi*HV*T*RIHL%9*;*-*{L%** *[L%***;HCH$*H**ǙH*H**MInH$E1GIff.fAWAVMAUIATUSHXH|$Ht$dH%(HD$H1H;'=L=*HH=<*IWL8IH| HID$H5*LHHII$HM I$Hu ID$LP0IG1E1H;'H;' H;x'HcWIH MtL`HcHMHHcI\H'HIDIGHHLT$L%B'LT$H QA;$P 1LLT$LHsLT$H QP A$=H9 HjI*I/^H;DH *HkL1H޿IH H8PH*Ml$sH{ I96IVL=*HLHT$IHH@HT$HHnLLIHdIVL=*HLHT$HHH@HT$LM" HLAHHH@H;'$ LqM LyIIH)u HAHP0LL0IHPI.LuIFLT$LL|$ P0LT$HL$ H)uHALT$HP0LT$I*u IBLP0IID$L|H0H(H0I$8H@(HI9tkI$8H|$H0HD$HEI$8H@I$8Pt8H(HHR8HcR H0I9uLH5*1LaImHu IELP0HHm HEH1P0+H*qE1E1HHQ HxjjjHH H H8u H@HP0H*sH{ LIHCHD$ H*HH' H8IVL%*HLHT$ IH> H@HT$HH LLIH# H*IVHHHD$(HT$IHe H@HT$HHLLIHL H@H;'Mt$MIT$IHI,$HLHT$AHT$HH" I.IuIFHL$LP0HL$I,$uID$HL$LP0HL$H)u HAHP0Ld$ HD$O4M#tDH(HE(H0IM9tSH0H|$HD$I$MHEt8H(IH@8Hc@ H0M9uH|$/IEL5*LM L%x'H QP A;$ 1LLAIH QP A$H= @9M ImuIELP0M I.Au IFLP0HIHPHHHu HCHP0MtI,$u ID$LP0HeHmZHEHP0KH@HP0fHCHP0IGLP0H;f.IG@=HcH'HLHH\$8Ht8Ld$0HD$@VHH H.L1E1H***2fMt I,$pH E1E1H)uHALT$HP0LT$MtI*u IBLP0H C*I*H={58*#}MLHE1HaH@LP0IBLP0I/N @9 ID$HT$LP0HT$EfDLT$6LT$@$IEfDkHuLNrIHjH 2K*H=}H 1*/*2|E1H\$HdH3%(L HX[]A\A]A^A_f.u{HM0H0H;0HH0HM0H0HP0H;0HHP0I$8H0H0LcH0JDHx(H;(|ZH@(H+(AH0H+HD(H+(HH0/H|(H;(}LcJDHH(Hx(H0fHcH4LF(L9(kHF(I$8H(H)0+HF(I$8H(H)0I$8HcH4LF(L;(}ILD(I$8H(H0o@H(H+0HE0HE(H0$fL8HHH7 *H * *'2ImI H@0I$8H@(I$8H(H+0H0fHHL$~HL$HI%ImHH' *) *# *3q1IIEfDI$JfDHcH'LH\$8HLd$0Ht8HD$@?hHHMI,$ID$LP0H1@$efH S *H @ *> *2M%I,$L1E1fID$HL$LP0HL$vfDH[I$1H * * *2H1IHE1 *H * *1I17fMgMIoI$HEI/u IGLP0HEIHE12H * * *2H dH=8vH'LH8H92E1H * * *2DH12 H * * z *2lDH1'Ht$(H8茻ImHHA *C *5 *21E1IEHL$LIP0HL$nfDH5a)H=*1UHHrH_dHmu HEHP0H)2"1H**"*2HY'LH8趺H 3#1Hd*f*#X* 3J1LLLT$~LT$HH{DI/HH***2L1E1E1q@H'LH8ImH,H**#* 3E11IEHL$LLT$P0HL$LT$HI1H=aLT$oLT$X%DLT$.LT$HHa'H5:H8LT$HImu IELP0HLp2E1H***p2D1LLImIH=蜻O}I1HIGLE1E1Hm3#H*#H5*3*m3%HLI*H**$2G1E1E1ImHJH**#*3u:MLHL1*H*E1*2I1MHIu2LIHЛh*"HU*S*2IHw'H5PH80)E1E1IE1I/u"L1ME1E11I2H=qff.fAWAVAUATUHSHHHdH%(HD$81HHEHEHHSH5)HHHHHHHCH;'LcMLkI$IEH+Ld$IEH;'kH;'IUB'HZE1 uMu[L='H QP A;LLI2H QP A=+29MI,$LID$LP0fHIMIEH;'`MeMSI]I$HImHCH;'H;'[IH_L`HEHh HCLM9L=Ҷ'H QP A;1LHAI H QP AH= @9)MImIELP0fDH;'H;'HKQE1 LauLk職L='H QP A;x1LAIXH QP AH= @9TM[H+u HCHP0Hmu HEHP0HL$8dH3 %(LHH[]A\A]A^A_@HCHP00IELP0.HSBፁ3Ld$ 1Hl$(LB uH{Ht$ AIMI,$9fH;Q'Hl$H;'IUBHZE1 uMeL='x WP A;HLIH QP A=29#LMH)H|L*H))OHHT$3HT$HFH51HT/E11LH蒲I9H=;~N1H$)H))Oj1ɺAIH)'H5~H8蘬1ɺAIfDAWAVAUATUHSHH(dH%(HD$1H?HEHEHlHSH59)HHHHIMrHHH5)HH4LID$L= )LM苮L5$'H QP A;HLLAI^H QP AH=D9JMI,$<H+u HCHP0Hmu HEHP0HL$dH3 %(LH([]A\A]A^A_ÐHHHYHCH;'LcML{I$IH+IGH;'Ld$&H;߬'IWB*HZE1 uMoNL5'H QP A;ULLI%x WP A=29M:I,$LID$LP0@H;ѫ'H;,'~HKQnE1 LauLk虬L52'x WP A;L1AIpx WP AH= @9dM螬HH֩'H5zH8菩jf.HCHP0m @9@$@HHT$蓦HT$HH5H&KE1fDH kNK)KH ))NH=FE1bHH+)NH))OI,$tH+t*H ))5)ID$LP0H+uHCHP0@D@$f.11HQIMH{)LH))NRf{@$ofH ;NL)LH ))NH NN)NH ~)|)NkH![II,$H H 5)7)N))OteON4LLYIMI/HPH))L)N I,$"ID$LP0fLL-)M$L5'H QP A;1LHAIH QDHLL芫IHfDH+)NH))OcH=wf.ۨHuH'H5vH8ХfDHt$L6OIfDLB1 uIHt$AIsDH=avtl7H=AvT>H)HS'H5,vH8 L1H:IH=uc蔤1ɺAIIGLP0AWAVAUIATUSHH'LfH|$8dH%(H$1H)HDŽ$H$H$HD$(H$H I~0ItIHF0HD$(IE(HD$0Im H\$0HIMeHD$xHDŽ$HDŽ$I$HEH9L-')H=)IUL̦IH` HHD$xI@H5L)LHHI IL$H|$xM> H/H$H'HD$xH9G Ht$(vIHH$H/f.IFH5)LHDŽ$HHVIL$MI.u IFLP0L5)H=)L$HDŽ$IVL袥IHHH$LLAƅH$H/RHDŽ$EeL5)H=7)IVL3IHHI@LLH@pHH@HLD$LD$H$HRI(H$HGH;'HWHaHG LwH$HG(HD$xIH$HHD$xHH$H/}HDŽ$H$L=ʢ'HDŽ$HD$HD$xHD$xHD$ M9|$sL軠HD$H$HHDŽ$L9}H臠HD$H$H!H|$1LHDŽ$HH$H"H;'H;='H9 H$HGP0@HGP0$@I$H+tuL-&)IEI,$tBLM1HGP0@I@LP0HGP0w@ID$LLMP0f.HCHP0|I$Ld$kIH訜IH5)HIHVnH$HMefMHH HOHcL XLOLD@HH'ATHnH5uH81iH$l?))l?H)XZH H=1vXfDHH$dH3%((H[]A\A]A^A_@H\$08fDH/uHWD$DR0D$DHDŽ$H|$Ht$萝HH$HH;m'H;='u H9 H/uHWD$DR0D$DHDŽ$Ht$H|$H$HH;'H;'u H9 t$DH( HDŽ$L$DHD$8H5)Hx {2IHHD$HHD$8H5)Hx U2LD$HHHD$xHDŽ$H /'H9Hu HPH$H` H@HHH|$xHD$xH/H H$HD$xH* HLD$HMpLD$HHH$H$H/uHGLD$HP0LD$HHDŽ$H|$xH/uHGLD$HP0LD$HH$HD$xH/uHGLD$HP0LD$HLD$`HDŽ$肙LD$`HppHHxHD$XHHHt$hHL$HHD$PtHHD$HHtHHD$PHtHHL$H5)HAL9HAHPHHHtyHHWHHELD$`LD$`IL|$xMHL$ E1HHAH;`'H$H;{'H;֜'HcLD$`KLD$`HHMtLxHT$HcD$DLD$`HHHT@HT$xHHTHT$0FHHHTHT$8HL$8HB(HFHR(HHHTH$1HD$x3HL$8LD$`HIH${H)uHALD$8HP0LD$8H$H/uHGLD$8P0LD$8L$HDŽ$HDŽ$H9\$0ILHL$PHT$HLD$(Ht$hH|$X2,LD$(H5~)1L<3LD$(I(uIPHD$(LR0HD$(HH(u HPHR0M/fDHEHl$IHHІJcH>fHF0H$IE(H$IE HH$IEH$II~uIIu-M~1H5m)HHVHH$IMH$L$H$HD$0H$HD$(ZMuf.HD$LD$HLLD$GLD$HHD$x$HS{E1E1E1H)))?HD$HD$HD$ HD$fH$HtH/uHGLD$(P0LD$(H$HtH/uHGLD$(P0LD$(MtI(u I@LP0H 4):)H=|1ۋ5')QMtImu IELP0MtI.u IFLP0HL$HtHHD$HHu HAHP0Ht$ HtHHD$HHu HFHP0HL$HtHHD$HHu HAHP0HT$HtHHD$HHu HBHP0MtI/u IGLP0I,$u ID$LP0HmHEHP0fIHkyE1)H))?HD$HD$HD$ HD$E1E1H fHG@=HD$(H$H$H$&IHH3w)H))?fDH|$xE1E1E1HD$HD$HD$ HD$f@$fHD$(H$H$H$o)H))3@f@HoH|$x)H))p@HD$HnH|$xn)H[)Y)|@@HnH|$x>)H+)))@]@HA@=6HD$HHLD$`L$H$HD$xH$HD$0H$HD$8H@(H$HcD$DHHĘLD$`HH$MtI/uIGLD$8LP0LD$8H|$xH/uLD$8HGP0LD$8HD$xK@辌H$H8IEH$H=)Ln0HD$xHH$H/uHGP0Ht$xH=)HDŽ$]H$HH|$xH/uHGP0H$HD$x2H$H/uHGP0HlH|$xHDŽ$H[)])O)@f.HlH|$x.)H))@M@HklH|$xE1E1H)))@{fDHWHHGL0HPH@H$HD$xAfHD$(LD$0HH=)H$8LD$0HH$H59)HLD$0 LD$0HH/H$H/uHD$0HGLD$8P0LD$8HL$0H$HκLD$8HL$0HDŽ$HL$0LD$8HH$H)uHALD$0HP0LD$0H$H;=3'H;=و' H9 LD$0CLD$0 H$H/uHWLD$8D$0R0LD$8D$0HDŽ$ H=)LD$0z7LD$0HH$9H5)HLD$0LD$0HHH$H/uHD$0HGLD$8P0LD$8HL$0H$HκLD$8HL$0HDŽ$+HL$0LD$8HH$H)uHALD$0HP0LD$0H$H;=߇'H;=' H9 LD$0LD$0H$H/uHWLD$8D$0R0LD$8D$0HDŽ$H=X)LD$0&6LD$0HH$mH5y)HLD$0lLD$0HHH$H/uHD$0HGLD$8P0LD$8HL$0H$HκLD$8HL$0HDŽ$ׄHL$0LD$8HH$H)uHALD$0HP0LD$0H$H;='H;=1' H9 LD$0蛈LD$0H$H/uHWLD$8D$0R0LD$8D$0H$HDŽ$H/uHWLD$8D$0R0LD$8D$0HDŽ$HD$(H5'HH$H9pH|$(LLD$0 XLD$0HH$ H$H/uHGLD$(P0LD$(H$HDŽ$HDŽ$"fFH$Hk IEH$H=)Ln踉HD$xH H$H/uHGP0Ht$xH=))HDŽ$8WH$H H|$xH/uHGP0H$HD$x+H$H/uHGP0HefH|$xHDŽ$H)))@ fH+fH|$x)H))@HeH|$xE1E1H))w)@Hee)HR)P)?HD$HLD$`L$H$HD$xH$HD$0H$HD$8H@(H$HcD$DHHĘn+LD$`HH$CH3e)HD$0H));AMt I/E1H|$xHtH/uHGLD$(P0LD$(H$HD$xHtH/uHGLD$(P0LD$(H$HDŽ$HtH/uHGLD$(P0LD$(H ))H=fLD$(5)HDŽ$:H|$XHL$xH$H$LD$(HL$x1LD$(H$H$?LD$(H1LHLD$(HD$8:LD$(HL$8HI(uHD$8I@LHL$(P0HT$8HL$(H)uHAHT$(HP0HT$(HH;'H;M'H9HHT$(贃HT$(H*uHJD$(HQ0D$(H$H/uHGP0H$HDŽ$H/uHGP0H|$xHDŽ$H/uHGP0HL$PHT$HHD$xHt$hH|$XHH!'H5RH81˄CH*J+H5 )H=)1H$HHf'H$H/uHGP0H,bH|$xHDŽ$H)))@H$LHLdH5=n)X!HaV?^)HK)I)V?HaH|$xE1/)H))AHD$xIGLD$(LE1P0LD$(LHFa)H))ANH;-~'LD$`H|$ LD$`IiH`)HD$0Hn)l))AH$H/HDŽ$HHIG`H|$LD$`PLD$`ILyM HAIHH$H$H/H$D$DHAHDŽ$ )HD$0) BH)H|$XHL$PHT$HHt$hH|$xE1H_])HJ)H)A;H_H|$x1)H))@HD$HD$HD$ HD$,HN_)HD$0H))KAH"_)H))@H^)HD$0H{)y)\AH)HALD$(HP0LD$(H^H|$xE1A)H.),)IBHGLD$`P0H$D$DLD$`HAHS^)H))W@!HL$xH|$8H$HL$(E}H|$8HL$(HHH HD$0HDŽ$HDŽ$HD$xHh)j)\)B5HD$x@HH HHD$x@HH H)HD$0)BH)eHP]H|$x)H))@H$])H))M@)HD$0) BHz)w)HD$0d)BHU)H\K)H8)6)@E1E1H\H|$x)H))@81I(u I@LP0%u1HH;cHH 0\H5OHEHy'H81S~H\)H))_@MiE1H[s)HD$0H[)Y)xAMH[H|$x8)H%)#)@WHo[)HD$0H))zAIHC[)H))@H[)HD$0H))}AHZH|$x)Hp)n)@HZ\)HD$0HD)B)AHZH|$x+)H))@JCHD$@\o*xLD$`IHL$ H6Z)HD$0H))AH Z)HD$0H))DAHYv)HD$0H^)\)AvHYE)HD$0H-)+)AHHHH@HHH$H$H/uHGLD$0HL$(P0HL$(LD$0H$HGH;Yw'uH;w'LD$0HL$("xHL$(LD$0HHD$xHH1IHt$xH$LD$(L~ %LD$(HH$H|$xH/uHGLD$(P0LD$(HD$xHX)HD$0H))AZHTX)HD$0H))ATHG@=H$LD$0H$HL$(L$HL$(LD$0HH$H)QHALD$(HP0LD$(8H$LD$0H$HL$(L$HL$(LD$0HH$uHfW)HD$0H))AfIPH2W)HD$0H))A HW)HD$0H))AHVr)HD$0HZ)X)AHVF)HD$0H.),)AI,H|$xE1E1HD$HD$HD$ HD$HFV޿)HD$0Hƿ)Ŀ)AFI]HV)HD$0H))AHU~)HD$0Hf)d)Aff.@AWAVAUATUHSHH^H|$dH%(HD$x1Hs'HD$`HD$hHD$pH HHHHmUH ]UHMHSHXWL@Hr'H5rHL \H819wHT)y)Hj)XZH TH=H1F+HL$xdH3 %(HHĈ[]A\A]A^A_DHF(HD$pHE H}HD$hH|$`HGHGHPHD$H H ^HcH>H+uHR0HD$Hl$hLl$pH6tHH Hʽ) L(hHE1AAHAHH H8LeHCHD$A$H ID$ITfH\f(Xf(\f(\H9uf/o# L=ڲ)H=)IWLrIHHHwqIH[IF1E1H;o'pH;q'+ H;tq'Hc#LL$0t$(LT$ qLT$ t$(HILL$0MtLPHcƃIEHHcMlMLIFLMMqL1p'H QP A;LT$ 1LLAIgqLT$ H QP Ap=9MgI/I.3L=)H=)IWLqIHHIFH5)LHHIIHMIHIA1E1H;zn'H;o'" H;o'Hc2L$(LL$ dpLL$ L$(HI_MtLpHcIEHHcMlHPo'HIDIALM LL$ pLn'LL$ H QA;P k1LT$(LLLL$ AIoLT$(LL$ H QP A=H @9 MGI/uIGLL$ LP0LL$ I)u IALP0IFAvI~ IHD$@H)H5T)HD$(HD$Ht$0Lx IWHHT$ hpHD$8HrH@HT$ HH LH|$8HD$8HUHD$H5)Lx Ht$0IWHHT$ pIHoH@HT$ LM LLAIHUH@H;Nl'MyMIQIHI)uIAHT$ LP0HT$ HLHT$ ?HT$ HtI/Ij I)uIQHD$ LR0HD$ H(u HPHR0;pE1H|$(LL$@HD$HHHD$0H\$ Hl$@Ll$PMML|$Lt$XMLd$ H|$Hl$ j1"fDL$A\LHH;\$ADIHL$^mH)IHLt$0L9d$(Hl$@Ll$PLt$XH|$H>hHD$8L%)H@HHlLk'H QP A; LT$1LH|$8HlLT$H QP AH=9h H HL$8HHD$HHH+u HCHP0ILHmNIHISImu IELP0I.7IFLP0(fDH@HP0fIFLP0IF@=LT$`HE1Ll$hHthLL$pIVHBR uM^LL$(LT$ HLLL$(LT$ IMH7Lϵ) H))-f.IE1MHIZ E1IFLL$LE1LT$P0MLT$LL$t I/MtI*uIBLL$LP0LL$MI)IALP0IFLL$ LP0LL$ IGLP0I.Q4@9-DIA@=HHIi'HLHthLL$ Lt$`Ll$hHD$pLL$ HI MtI.uIFLL$ LP0LL$ MPDGWHH HHD$H|$g jHD$HPHcJ)H))l@GWHH HD$ GHD$GHHD$t#i@$f.f(L|$81LLkHIHD$HIuf.H|$8HGP0HNHI)H))vl@H59)H=)1IHHWI.u IFLP0H"IE1E1) H))H ))H= =5y)dHmu HEHP0Mt>I.u IFLP0MHD$HHHD$ Il}MImIELP0DhHD$fIH~aHtHHF(HD$pHE LHD$hHEHD$`aIHt^Ht}Ht.MH|$`jHHaIH5u)LIHVugHD$`H H5)LHVTgHD$hH IM~H5)LHV*gHHD$pIYLLL$ ##LL$ H;H G)H))HL$8HHD$HHHE1H|$8LL$HGP0MLL$&E1HLLT$`HthLT$(Ll$hLL$pLL$ LL$ LT$(HIMt I*I)cIALP0THKF)HЯ)ί)ϕH )ů)H=B:1ۋ5)RHE)H)~)ٕ@HD$8Hf.IbLL$ dLL$ @$  @DHT$ IWLHD$0R0LL$ HD$0sDIBLL$ LP0LL$ H:EE1Ϯ) H))D1LLLL$ fLL$ HI~DHDE1x) He)c){I/!E1fDIGLL$LLT$P0LT$LL$DdHuLHHhDE1) H))KHH~b'LLL$ HthLt$`Ll$hHD$p5 LL$ HI8HC) H))bfM?IHIH;E1E1fHCE1@) H-)I()HfMVMM~III.IGMWbHuLHH C) H))NfD#b@$f.{eI-H Bc) H P)N)PfHB3) H ))4ifMqMIQIHI)uIAHT$ LP0HT$ HBIѹ1LLcIHHAME1) Hz)Iu)?H@H=Y/LT$ g`LT$ [+aHuHg^'H5@/H8 ^{H_'Ht$0H84]H_A)H))EDHFHHD$`[IHy_'Ht$0H8\H@E1)H))f.H@c) HP)N)pfE1E1DIFLL$(LMLT$ P0IGLL$(LT$ l@H=-LT$(LL$ ^LL$ LT$(mCLL$_LL$H(H\'H5-H8\LL$H=-LT$^LT$2fHL$8HHD$HH.H|$8HGP00_HuHl\'H5E-H8%\HL$8IHv?)H))HHD$HHYHT$`HLLiAH5P)8H?)H))H>) Hm)k)$fDH@`HHHHHHPL%\'L9HCHHHD$HH5HHcH>CHHD$H+HSHR0CSHH HHD$ЋCHD$H+ʋCKHH HD$H5@HHHMHPCHk]HD$rHHF['H =H51jL DAH?H81_Hs=Y^H)))zH1LLT$ LL$(IH)=) H))j'H]\HqHZ'H5b<H8zYVHB`HHHHHHL9`HEt(HEHD$HHHGHcH>H HD$HmHEHP0EHHD$؋EUHH HHD$EHD$뷋EUHH HD$H5/>HHHFH+ZHCHP0KHr[HD$]X>[HuHBY'H5;H83XIjE1IfAWAVIAUATUHHSHxdH%(HD$h1HD$ HD$(HD$0HD$8HD$@HD$H:[HL%)H=)HIT$LZIHfHIGH5W)LHHILl$ MI/ Ld$ MnI,$sHD$ M9~L%)H=+)IT$L&ZIHJHHD$(IGH5Ɯ)LHHIMmH|$(H/ HD$(LLVAăImk Eu H} H5)IHH} H5)HD$ H HD$(H V'H9HHPHT$(H H@HHH|$ HD$ H/Ht$(HD$ HH)HD$0H#H|$(H/HD$(H|$ H/\HD$ H|$0H/4HD$0&SIHPpH@xI$HT$8HD$@HL$HHt HHD$@HtHHD$HHtHHHHuHFWLHIHD$0H5HLHD$ HHT$0HL'GH|$0H/uHGP0HT$ LLHD$0$H|$ H/uHGP0HD$ HEH|$8Ht H/uHGP0HD$8H|$@Ht H/uHGP0HD$@H|$HHt H/uHGP0H5!)1LHD$HHD$HI/u IGLP0L|$HM=&I/u IGLP0H$U'HwID$LP0HD$ M9IFH5p)LHHIL|$ M2H5{)LnIHH|$ H/uHGP0L%T'H ZT'HD$ M9H $I9 L;=bT' LVAŅI/u IGLP0EH5y)LyIHL9AH;$DL;=S'LAVAąI/u IGLP0ELIFH5)LHHIMIIGH5N)LHH3ILl$ MI/u IGLP0H|$ HG:"HGH|HRHUH4H$H|$ H/uHGP0IFH5)LHD$ HHGIL|$ MIGH;aS' H;Q' H@hHH@H L1IMiH|$ H/uHGP0HD$ L!IHqI/u IGLP0IFH5)LHHIMIGH5 )LHH/ILl$ MI/u IGLP0H|$ R!IH5H|$ H/uHGP0L-)H=M)HD$ IUL@SIH4HHD$ IGH5)LHH5IH|$ M2H/uHGP0HD$ LlQIHD$ H[RIHD$(HHD$ HD$ IESIHD$ HH=ɔ)IHD$0HKH@H5)LHHIM]H|$0H/uHGP0H5Ɨ)H|$ LHD$0SImu IELP0HT$ Ht$(LuHD$HI/u IGLP0H|$(H/uHGP0HD$(H|$ H/uHGP0H5)H|$HD$ IH H5e)HHD$ HImu IELP0H|$ LIH&H|$ H/uHGP0H} H5z)HD$ 4HD$H/H} H5k)IHgHD$(HM'I9EIEHD$(HIUHHImuIEHT$LP0HT$Ht$(HHHT$!HT$HHD$ VH|$(H/uHGHT$P0HT$HD$(H*u HBHP0H|$ H/uHGP0HD$ LkJHL$HHT$@Ht$8HIM~lLIH$HuLNI9t@IH $LLLLLD$OLD$LHLNLLHNL)IuHM'H;H\$ u HCHP0HD$ H|$8Ht H/uHGP0HD$8H|$@Ht H/uHGP0HD$@H|$HHt H/uHGP0Lt$H5ކ)1HD$HLcHD$HIH$HIu IFLP0Ll$HMImu IELP0HD$HL|$HfDIGLP0Ld$ MnI,$uDHGP0n@IELP0EIFH5ڐ)LHHIMH5)LMJHD$(HI/5H|$(L%!L'H K'L9H9H $u H;=K'QDH/HD$(EH5)LIHD$(HL9AH;$D L;-{K'LMADž`Ll$(Imu IELP0HD$(E`H=C)IHIH@H5g)LHHHD$ H|I/u IGLP011L@IH H|$ HI'H9GL&HD$(H~I/u IGLP0H|$ H/uHGP0H} H5)HD$ L|$(HD$(OIH}H} H5)3HD$ H2HD$0H I'H9H[ HPHT$0HI H@HHH|$ HD$ H/uHGP0Ht$0HD$ H H;HD$(HH|$0H/uHGP0HD$0H|$ H/uHGP0HD$ H|$(H/uHGP0HD$(EHHPpHD$HH@xHT$HHD$@HL$8Ht HHD$@HtHHD$8HtHHHHuHILHHD$NHD$(H H5H'HLnDH|$(H/uHGP0HLHD$(HD$(H:Ht$HLmH|$(H/uHGP0LHLHD$(<H+H|$HHt H/uHGP0HD$HH|$@Ht H/uHGP0HD$@H|$8Ht H/uHGP0H5)1LHD$8HD$8Imu IELP0Ll$8MImu IELP0HD$8H|G'DHS)H|$ E1E1Hג)ْ)˒)7Ht H/uHGP0H|$(Ht H/uHGP0H|$0Ht H/uHGP0MtImu IELP0H c)i)H=+1ۋ5V)AMtI/u IGLP0HL$hdH3 %(H6Hx[]A\A]A^A_fDGHuLHHP(H|$ E1)H͑)ˑ)AfDJIdH ()H))CI/u IGLP0H|$ E1IGLP0HGP0@HGP0@HGP0@HGP08@GADž H|$(fDHGP0@@EqIIcH+'E1)H))SfDH$I~aIGL8IpfDoGII L,$fGH$@I~MI#@1DIHuLHJCI,$IID$LP0@HC&H|$ E1ӏ)H))ifHIiEHLjHD$(HIH%H|$ E1g)HT)R)DH%H|$ E13)H ))QfHk%)H))k[fGIGIH%H|$ E1)H))|DHuHB'H5Y$H8qA[DHIH|$ H$H$E1@)H-)+)ypfDGIEHHHD$0H?HE$ݍ)Hʍ)ȍ)I/u IGLP0H|$ E1E1H#H|$ E1E1H))s){fDSCHH#E1J)H7)5)~H#E1 )H ) )NPfDEIMHK#H|$ E1ی)HȌ)ƌ)H#)H))ܜH|$(Ht H/uHGP0HD$(H|$0Ht H/uHGP0HD$0H|$ Ht H/uHGP0H 8)>)H=$HD$ 5$)HL$(HT$0LHt$ HL$(HT$01Ht$ EIH 1HLI/Hu IGLP0Imu IELP0H< H;- @'H;-?'H;-?'H.BHmAu HEHP0E8 Hl$ E: Hmu HEHP0HD$ H|$0H/uHGP0HD$0H|$(H/uHGP0HL$HHT$@LHD$(Ht$8|H=?'H@HmDhP@H !H|$ E1)H))PH s)H`)^)fKCIH 3)H ))ޜ{f?HH] H|$ E1E1H))Չ)?HkLIHD$ HH E1)H))Hk)HX)V)HHD$(HH-)H))Imu IELP0H|$ E1*fH))Ո)HL$HHT$@LE1Ht$8E1XH|$ fDAIHE1)Hu)s)fDH[)HH)F)H3)H ))fMLIH$@HuL=L9tIH$HIHHIHL)IuH)H))LHHT$HT$HHD$ jHI^)HK)I)HL$HH$HHu H|$HGP0H|$ L|$Qf @I)H[H|$ E1)H؆)ֆ) H+Æ)H))!H|$0Ht H/uHGP0HD$0H|$ Ht H/uHGP0HD$ H|$(Ht H/uHGP0H H)N)H=HD$(54)H|$HL$0HT$ Ht$(HL$0HT$ 1Ht$(?IH1HLImHu IELP0I.u IFLP0HL9H;,$H;-9'HCւ)HÂ))-;H )H));IfHq)H^)\)HH|$ E1B)H/)-)`H)H))/SH\H|$ E1)Hف)ׁ)Z H-Ł)H))9HOHPHGHHH|$ HD$ H/uHGHL$P0HL$H|$ HGH;5'H;L6'HL$6HL$HHD$0BHHH|$ 1HLx IHD$(HGH|$0H/uHGP0HD$0H?E1Ԁ)H)),HM)H))HE1)Hm)k)\HH|$ T)HA)?)rHH|$ L|$#)H))ALt$(Ll$05HLLHH)HD$ HD$0HD$())H3'H5H82#H)Hs)q)\HL|$Z)HG)E)H3)H ))YH ) )~)$H~)~)~)4HH&H|$ L|$E1H~)~)~)šHH|$ E1~)Ho~)L|$h~)ΚHT~)V~)H~)nH6~)H#~)!~)КHw~)H})})zLl$0Ld$ v3HLLHH})HD$(HD$ HD$0})})eUHH|$ E1E1Hy}){})m})H[})HH})F})1H2})4})&})]H})})})THD$(wH|)|)|)YH.I|)H|)|)ݚ`fH@`HHHIHH1'I9D$u2LI,$IT$H$LR0H$LH5fIHuHtH|$ E1E1H{){){)nHBH|$ L|${)H{){)_HG@=SHt$PHL$PHL$L|$XHL$HHD$(EH).HAHP0HD$ Ht$PHL$PHL$L|$XHL$HHD$(uHwMI {)Hz)z)a-HEMIz)Hz)z)ϛ/Hz)Hz)z)՛HH|$ E1z)Hnz)lz))HH|$ Uz)HBz)@z)sIIJHMIz)Hz)z)țpff.AWAVAUATUSHHhHndH%(HD$X1HD$0HD$8HD$@HD$HHHL~Lf Hn(H~01)2IHLL2HH>H@LANHCHhHFH-O HO0IIbfH+u HCHP0IGIGHHmHHcH>@L)IH5^r)LIHV.HD$0HlHk@HHu,'H8H5/UL AH @H810H6x)(x)Hx)XZH zH=1HL$XdH3 %(HHh[]A\A]A^A_HD$H;-0,'L=3p)H=w)IWL-HH< HHCH5Ms)HHH IHHMM HHu HCHP0L5o)H=mw)IVLi-HH HHCH5q)HHH HHHH HHuHCHL$HP0HL$IG11H;+*' H;I+'CH;+'HcHL$t$,t$HL$HI HtHXHcHEHIlFHILIGHH +Hl*'H QP ;K 1LLH+H QP H= @9vH- I.u IFLP0I/u IGLP0H}u HEHP0Hu)uH} LuI|-H|$MLHLL)H;&H}kHEHP0\EwAMcIH;-)'Lt$xH|$HL$(ML )Hm)H=:u)HSH6+HHHHAH5o)HHL$HHC HL$HHHH HHu HAHP0H|$(])IH H''H9C@ LHUHH I/H+PHCHP0AfDEwAGII IfAGHD$CfEwAGII Lt$(E1DcCII DDcH.IHHcH>f.HF0HD$HHC(HD$@HC LHD$8HCHD$0#IH;~WHTHu%H5i)LHVb)HD$HHo IM L|$0Ld$8Hl$@H|$H~HuG@IG@=8HcHLHl$8HHL$@Ht8HL$H\$0HL$HH HtH+uHCHL$HP0HL$H)HAHP0Lh(I;(HH 5r)H"r) r).fDHs  r)Hq)q)fDH q)q)H= 15q)H &'H5H8$I.uIFLP0@'IHL#HHrq)tq)fq)pHCHP0a<'HuH@%'H5H81$'HD$HJHt q)Hp)p)fH@`HX HHH HIH7 HPH_%'HD$H9f{ IFH?HcHBL'II.KIVLR0<IGLP0HFLHD$0o IH5%k)LHV9&HD$8H! IH5e)LHV&HD$@HIDH $'H5H8" LH5;޶IHHPf %@$|fHcLHl$8HHL$@Ht8HL$H\$0HL$HHHMI%o)Ho)o)MLIMH)~Ht H+_MtI.u IFLP0MI/IGLP0H HHn)n)n)HE1E1@HAHP0sfK$HuL.HHH5n)H"n) n)4+H s n)H m)m)6HHHCHP0 &IVI/H"Hm)m)m) I,$E1LfDHCHP0f[#HuL>HH)HEm)H2m)0m)9b#&HH s m)H l)l);I_H#MwHII/-IFMH l)Hl)l)]f1LL$HHzI.HHFl)Hl):l)hXf"HuHHHHek)Hk)k)H= g!HRH'H5H87$HL$HfDHkk)HXk)HSk)HTf.LsMLcII$H+u HCHP0ID$H;')H;'m IHLp1HLLx 茷HHnImLSIELP0DIGHL$LMP0IFHL$fDHE1`j)HMj)Kj)6EfI.8I.&LR0EfAFII HT$0HLLH5N)ɿH<xi)Hi)i)xHH'H H5djL AHPH81"HY^Hdi)tai)Si)t/fDH@`HHHLHHHpH'HD$H9&LsIVHH  HcH>H@`HHHLHHHD$H9BHHT$&HT$IH*u HBHP0I/uIGLP0fDH+HsHV0DsAMcDsCII IDsH+tDsCII HH52HHHpHIuH腯IHH@MwIFH4HHcH>EwAMcEwAGII IEwEwAGII HH7'AH5jL H H8H1Hd_nHf)AXf)f)nHH'L *H5jAH HuH81AHAYf)Hf)AZhvf)hRID$@=dHt$0LLt$0L|$8oIHI.u IFLP0I/u IGLP0LLHcMIe)He)e)VHt$0LLt$0L|$8?HH0I.u IFLP0I/u IGLP0L,vHHv'H5OH8gLjIHML?e)H,e)I$&e)H'HB`HHHLHHHD$H9Bu.HHT$HT$IH*_HBHP0PHH5?HHuI.E1HE1LLHdd)fd)Xd)CHLL@d)H-d)+d) HdH'H5H8IH+HCHP0LLE1^HH5dHH(IDHuH'H5zH8AWAVAUIATIUSHXH-[)H=c)dH%(HD$H1HUH~HHHHCH5])HHHIHHMHHID$I.H-O[)H=b)HUHHH HHCH5_)HHH6 HHHH1 HHUH 'HEHL$H9 H='Ld$(H<$H9c H;4'HUB HZE1 L=<'p VP A;J LLIzx WP A=@9]MHrHmHHa)a)a)  HEH1P0H fa)la)H5_a)f.HCHP0ID$GLL17I.H-Y)H=2a)HU2H&HHHHCH5 ^)HHHHHHHHHwH 'HEHL$H9H 'Ld$(H $H9u H;f'HUB HZE1 :L=n'p VP A; LLIH QP A=%29M Hmu HEHP0IEH5U)LHHIMID$H;D$I\$HIl$HHEI,$u ID$LP0HEH;$H;Y'cHHHXILp HEHH H $L=E'H $P A;P  1HH $HIyH $x WP A=p4@9M H)u HAHP0Hmu HEHP0Imu IELP0ILHPIIHu IFLP0HL$HdH3 %(H' HX[]A\A]A^A_ÐIFLP0HCHP0zHCHP0LuLu2@DLt$(H;$H;'IT$BHZ1 uIl$ L='H QP A;LHIH QP A=295LM$H HLL])H])])7fDHE@=Ht$0HH\$0Lt$8IH#H\)H\)H\)HfHHL@cHuHFHHHڝC\)H0\).\)ڝH H=1 @HkH1H[)[)[)ܝHkHCHLP0H Hmu HEHP0H [)[)5[)H=DH'HI1HDsIP@$f.@$Hf@DH $H $@$Yf. IFH [}LH=H Z)Z)Z)}2Ht$0HH\$0Lt$8$IH H+HCHP0LH IMp@LHIMPH12Z)HZ)Z)QQ@$fHuHμHHH=AY)HY)Y)AHuH~HHH{Y)HhY)fY)3[HH5CY)H50Y).Y)HHHCHP0 H@H=[X)H=X)X)CHH]H]HgHMHHHmuHEH $HP0H $H5' 'HAH4$H96H;w 'HL$ HL$HH8HX1HHI$L` HL$HL$HI`HmHEHL$HP0HL$HfHMHeLMHIHmuHEHL$HL $P0L $HL$H=M 'IAH<$H9H; 'HL$LL$ LL$HL$HH{HH1LHI$L` LL$LL$HIH+LGHCHP08fLL赼ILM^Ht$(HnIfDHt$(HNI\fDHV)HV)HV)Hf0LB1 uH}Ht$( AIDLB1 uH}Ht$(AID1HHH $OH $HIGHSLU)HU)HU)HHHHAHP0H= H= P6[ H!H'H5lH8LH=1H $@ H $%'H $ H $H H3'H5 H8H $Ht$(LNIfDLJ1 uI|$Ht$(xAIz@1 H=q )DIEL$HL$H L%='H QP A;$j1LLHxH QP A$H=9H ImuIELP0H;H+XH D)H=K)HSHIHz HIAH5H)LLL$HH LL$IIHM IHHC)H=?K)HSH;IH HIAH5?)LLL$HHH LL$IIHM IHIG1E1H;& H; & H;{&HcZLL$LL$HI MtLHHcÃHEHHcIlMdIGHH L%J&H QP A;$g 1LLHH QP A$p=4@9(HG Imu IELP0I/u IGLP0H;u HCHP0Hmu HEHP0Ic=IH Hc|$'HH D$IH IH^ HH)HHH)IEHI] M}(Im0Me8@H+u HCHP0HL$8dH3 %(L HH[]A\A]A^A_f.ID$LP0#IALP0I,$fIGLP04@9@$@IALP0IALP0HCHP0 @9]IALP0_@$=fIEfDIH;)HHLl$ Ht(Ld$0HD$(:HH^ MtImu IELP0I,$ID$LP0DcHuHFH H 9FG)H=H ,G)*G)90IG@=HHLLL$ Ht(LL$Hl$(Ld$0LL$HHH.F)HF)F)s:MIHH IEH ~F)F)rF)9HuTE11E1IELP0HtHmu HEHP0MtI,$u ID$LP0MtI/u IGLP0H F)F)E1H=5E)HE1fDHHuH螨HHH1E)HE)E)9HmUE1,sIH HMUE)H BE)@E)9IEE11HfLkMwHkIEHEH+HEHIDH&HH8H)9D)HD)D)9H H=jH|HLLL$ Ht(LL$Hl$(Ld$0ܠLL$HHeMtI)u IALP0I,$uID$LP0eHCHP0fImHVHC)C)C)9VMIIME1HIHu IALP0HaH1LAHH=@Imu IELP0H;:UC)HBC)@C);:H&HH8VImH|H C) C)B)9u IELP0HHCB)HB)B)9MRIEHHmD3@$f.1LHHHHH1^B)HKB)IB)9@H=1Df. HuHG&H5 H8fDHuHƤIHlH5HA)HA)A)Q:H E1A)H A)A)S:fD{LL$I-fD1LLImH+HuHIH'H}HA)H@)@)V:HjH&H5H8OH @)H @)@)X:fLL$IfDMOMI_IHI/EHCIߺImHH@)@)@)9>IELL$LE1P0LL$HHC?)H?)?):H?)H?)?):pfH?)Hp?)n?)z:fHS?)H@?)>?):f1LLcHHPHkH1>)H>)>):ImA}nH+M>)H>)>):fH=wkHbH&H5|H8\GHC>)H0>).>)9MFHE11IGLL$LIP0HCLL$OHA=)H=)=)j:zIQII$HHI$lWH+u6HE11.IELP0IH+tI,$uM1E11 1*AWAVAUIATUSHhL%&H^dH%(HD$X1Ld$@H HqHHnI] HMmL9L%G5)H=<)IT$LIHW HIFH5`8)LHH IIHM IHID$1E1H;& H;& H;4&HcL$L$HIR MtLpHcHEHHcIlH&HIDID$HHM `H &P P ; HL$1LLH1HL$p VP H=9H I/HI,$H}H;)uH} L=6)LsHD$HELLHD$IH6 H@HH LLHIH L=6)LsLLHH H@LMMHLHAHHo H@H;&DLAM7LIIIH)uHALL$ HLD$P0LD$LL$ IAH;&LD$8H;"& IQBGLzE1 LL$ LD$LD$LL$ x H &W;P NHL$(LLLL$ LD$AIEHL$(LD$x LL$ WP =@9" MI(LJI@LL$LP0HL$1HHH HIHHH?L HLIL@HH&SH!H5H81HR>8)Z8)>H8)XZH )ZH=1褥H\$XdH3%(HHh[]A\A]A^A_fDLIFLP0HEHP0ID$LP0H}ID$@=H&Lt$@HE1Hl$HHtHHD$PID$LH@ uMD$-HLAHHL-27)XL-7)7)t%MtI.uf.IFLP0MtI,$u ID$LP0H p7)v7)H=^5e7)PH+L-M7)O7)A7)>;>H cDIGLP0I,$fD @9wI$fDk@$WfL5a2)LcLLHH H@HHHLHHHL=,2)LcLL=IH H@HHLLHIHp H@H;&uM~MhIVIHI.uIFHT$LP0HT$HLHT$¼HT$HI I/II.u IFLP0I,$u ID$LP0}LIBLIGHEL5d()LMH &P P ; HL$1LHAIHL$p VP H= @9M Hmt_I,$u ID$LP0LHH H+HCHP0 1LHHmIufHEHP0MuL-*%Ty4)TL-f4)d4)*%@H H=E=HHL$胦HL$HIH)u HAHP0I/u IGLP0Lt$HD$HD$M1)ZHv1)t1)>I-CHuL&IHL-X%X#1)XL-1)1)X%f@$fI[L-;IL-0)0)X0)Z%H3IFLP0$Mt$MBM|$III,$u ID$LP0IGM D@$;f.HIHlH5%)HHVIHOHD$@IF9fHCHP0H /)/)5/)fDH&LH8L-!%/)[L-/)/)%H [H=wHmHEHP0fDLL$ LD$LL$ LD$@$fDH&LH8VI,$L-|L- /) /)[.)%u ID$LP0H .).)H=5.)ÛNfDL-#.)XL-.).)z%M I.fD1LLHHI,$L-L-M.)O.)XA.)%1I/u IGLP0H .) .)H=5.)H@H=HL$HL$"mDHZH&H5̱H8?Ha&LH8L-$Tw-)TL-d-)b-)$DLLLD$ LL$LD$ LL$IML-ML-)[L--)-)%I,$H)u HAHP0M@H&LH8HmL-L-,),)T,)$HEHP0 f.1LLI,$I-3H=QHL$_HL$fI,$u ID$LP0L-%,)[L- ,),)%cfHuH'&H5H8fDL-3VA%+)VLH=L-+)+)A%荘8L-+)[L-p+)n+)%dfLHt$8LD$ LL$ćLL$LD$ IfH=!HL$/HL$fHmuHEHP0ffDHuH&H5H8뾨nLR1 uIyLL$ Ht$8LD$ ALD$LL$ IQH=xHL$(LD$LL$ HL$(-LL$LD$;LD$LL$H Hi&H5BH8"LD$LL$L-nMI*)TL-))))$lL-A))XL-))))l% H1LAHgID$L1P0\1ɺALD$LL$ ICID$HL$LP0HL$eff.@AWAVAUATIUSHHL-&H^H<$dH%(HD$81Ll$(HD$ Ll$0H`HNH,HHHlH \HOH%L LOLDHH&SHH5^H81,HPz()l()PH]()XZH H=E18HL$8dH3 %(LHH[]A\A]A^A_fLl$LMd$I$H=/()HEL9AL-b)IULHHHHCH5C)HHH IHHM HHu HCHP0H=Z%)H5")HGHH HH IH H5!)LHHD$ILD$H55")LHLD$%LD$HCL- ')LM LD$sL &LD$H QA;P  LLT$LHLD$AI7LT$LD$H QP A=p4@9qM H+uHCLD$HP0LD$I(u I@LP0IGE1E1H;& H;& H;&HcMLD$LD$HHMtL@IcAHMcLtHw%)HJDIGLMz=L&H QP A;LT$1HLAI LT$H QP Ap=4@93 MH+u HCHP0I/u IGLP0Imu IELP0H<$H5)HGHH: IM HUH;&H5) HEHHHR H UHzHuH)_IM3  HH I$Lx L`IHw HT$H5 )H[3H)H5 )L= LHLpIH ImH+I/@I,$u ID$LP0HmHEHP0t@H!)HSHIHHIGH5)LHHoHHy I/u IGLP0H=*!)H5)HGHH IM TIHx H5)LHFH5#)LLoIH I/uHD$IGLP0LD$I.uIFLD$LP0LD$HC1E1H;&! H;&, H;u&HcTLD$t$t$LD$HI MtLpHcH!)HHMDFLHHIT1nH I/H+H(sI$ImSH)HI,$%LIfHF(HD$Il$ DLl$HHHtHHF(HD$0ID$ HHD$(ID$HD$ IHHM~.H5)HHVYH; HD$0IM) HD$0Ld$ Hl$(HD$DHH/IH5)HIHVHD$ H M~H5)HHVHZHD$(IGH۶E1p )H] )[ )PI/uIGL$LP0L$Ht!HHHHuHCL$HP0L$MtI.uIFL$LP0L$MtI(u I@LP0H ))E1H=5)蹌@IG@=IHI)HLJt(LD$ LD$Lt$(HD$0dLD$HI MtI(u I@LP0I.IFLP0fDHE1()H))PfDLD$LD$@$wHuL辁HHH-)H))PID$LLIP0yfIELP0HPHR0~HSHD$HR0HD$ZIELP0AIGLP0IHCHP0+HC@=H)HHLD$(LD$HD$0HcHLt$ Ht( cLD$H; Mt$I.uIVLD$LHD$R0LD$HD$I(IPHD$LR0HD$yIWHD$LR0HD$`LLHLD$LD$HI@HE1()H))PfDHI8H CE1E1)H ))PHAH)H))PI/E1E11HH~f.HE1@)H-)+)PfD@$fHk)H))1QImtE1E1I/|IELE1P0E1fDHE1)H))UP3fDIH&)LLD$ Jt(LD$Lt$(HD$0wLD$HIH1=)H*)()PCH{)H))2Q fHK)H))#QfH<$IHuHv}IHH})Hj)h)DPxImHHE)G)9)%QHIELLP0M2E1E1fH;&%HKIWHK)H))FPCfHFHHD$ IH)H)~)'QfHDH$HB`HIMGMI_IHI/HCIߺAfDH=LT$LD$LD$LT$L$L$HH&H5̜H8L$kHE1E1)Hz)Hu)QPH5@cI)H)HLD$(LD$HD$0HcHLt$ Ht(tLD$HPHw)H))qPHHfDImIH;H)))/QfDHE1)H}){)SPfDI/HǭHV)X)J)Q,1HLkIHHsE1E1)H))Q HC)H))VPfH=LT$LT$HzH&H5H8|_LsML{IIH+EIGLH*)H))3Q"IGLD$LAIP0HCLD$HT$ HHLH5,(kHP)H))P0H)Hs)q)PHH.HE1E1R)H?)=)PX}EHH H~EUHH H)dHCLD$HLP0IGLD$MN{XE\I(I\$H1)H))QH)Hq)o)zPHH,ff.AWAVIAUIATIUSHh$dH%(HD$X1H;&H- )HH=.)HUH*IH HIGH5)LHH< HIHH7 IH[HE1E1H;&P H;&H;t&HcsL$L$HI MtLxHcHHHcI\H&HIDHEHHo LD$L=6&LD$p VA;P  1LLD$HHhLD$p VP A=H9H I(~Hmu HEHP0H;u HCHP0H)sH{ IVL= )HD$HCLHHT$HD$HH\ H@HT$HH{HLHHA IVL= )HLHT$HH H@HT$LMHLAHH H@H;&ILqM<LQIIH)IBH;&Lt$8+ H;Y& IRB HrE1Ht$( uMBLD$ LT$LT$LD$ p L=E&VA;P  LT$ LLHD$(HD$zLD$LT$ p VP A=)29NMr I.LuIFLD$LLT$ P0LD$HL$ H)uHALD$HP0LD$I(u I@LP0?L|$HD$HD$M4H~$LIAIGM9uH|$HEL-)LMS L='&p VP A;U 1LHAIbp VP AH= @9M HmuHEHP0M I,$u ID$LP0HHHBHHH0HCHP0!IGLP0L= )HiLHHHCH@HHHHLHH*IVH- )HHHT$IHH@HT$HHLLIHgH@H;&MwMIWIHI/uIGHT$LP0HT$HLHT$+HT$HHU I.Iu IFLP0I/u IGLP0Hmu HEHP0L$HAHI訾HCL-U)HH XL=&p VP A;1LHH-p VP AH= @9yHH+uHCHP0HbHmu HEHP0LHHfHt$XdH34%(H Hh[]A\A]A^A_f.HE@=xH&L|$@HE1H\$HHtHHD$PHEL@@ uLM HLAHH H )Hz )x )0MtI/u IGLP0HtHmuHEHP0DH 9 )? )H=f15, )yfI@LP0s @9QLD$LD$@$9f.HALT$HP0LT$IHa&HHH\$HHtHL|$@HD$PhHHMI/IGLP0@HlHHL$}HL$HIH- )H ) )?1HmH)uHAL$HP0L$M)I(u I@LP0H ) )H=15 )wHHHP:fDHEfDL|HHUHנo )H\ )Z )x0H+HCHP0HI HuHlIHH]0 )H ) )0H 1H=1v@C@$f.@DHH۟IHg )i )[ )0H IGLP0L}MHUIHHmt.HBHչzf.苾@$yHEHT$HP0HT$@H&LH8H!-1)H))-1H H=1uuHHP fLT$ LD$LT$ LD$@$fDH;)H()&)0MI/fDHɼ&LH8&HmHLH)))/1u HEHP0H ))5)1LHLD$ֿLD$HHDHmHѝH`)b)T)1bHEL$HP0L$KH=!LD$/LD$#L$L$HzH+&H5H8L$[H&LH8H$f0)H))f0LLLT$(lLT$IMHҜMLd)HQ)O)<12H &HH8hH+HH) ))h0HCHP0HqHmu HEHP0H91)H))1f1LHHmIH=蔺H ՛0c)H=H I)G)02r LHt$8LT$aLT$IH= VH+u HCHP0HP0)H))01LHH+HJ;萺HuH̷&H5H8腷{LB1 uIzLT$Ht$8ALT$IL$*L$HH^&H57H8L$fDH= LT$LD$ nH@LI)H))u0^H)H))0H1LAHH&H5H8ht1ɺALT$IHEHL$HL$P0HL$L$fDAWAVAUATUHSHHH^H<$dH%(HD$81HT&HD$ HD$(HH[HHF HD$LmH) H(hE111AHLHHH;H@u HP0HCHH5`(HHIMH5G)L躴IHnI/u IGLP0L;5&@L;54&@u L;5U&@I.u IFLP0sLsf.+hff/f/hy H$HT$H5&Lp IHxL%IH I.uIFLP0fH+HCHP0H]HHH HOHsL hLOL@HH&SHGH5H81{H6))H)XZH H=E1mHL$8dH3 %(LHH[]A\A]A^A_ÐHѴ&HD$L-(H=r)IULnIHHIFH5c(LHH IIHMk IHu IFLP0L5Z(H=)IVLIHHID$H5(LHH` IM I,$u ID$LP0IGE1H;Ҳ&7 H;&H;K&Hc*ŴIHI MtL`HcH1I\HcLLHO(LD$HIDKLD$HH I(I/u IGLP0H*&I9E HL蕅IHi Hmu HEHP0Imu IELP0L;=&@L;=&@u L;=β&@I/ L%(H=d(IT$L_IH HIEH5T(LHHB IIEHM IEHL-O(H=(IULIH HIFH5(LHHS IIHM IHuIEE1H;ǰ&& H;&H;@&Hc'躲IHd MtLpHcH1I\HcLLH<(LT$HIDILT$HH= I*u IBLP0Imu IELP0H&I9D$ HL肃IH Hmu HEHP0I,$u ID$LP0L;=&L;=&L;=&rLŅ I/u IGLP0 H$HT$HH55&Lx IHxMIH I/IGLP0@L舲HcHE1E1(H(I(HfDIHu IFLP0MMMtI/uIGL$LP0L$MI(I@LP0IHHiHHXH51(LIHVIHD$ HMHD$(Ll$ HD$H E1H=H (((ugHېs(H`(^(fDH I(O(E1H=5;(&g+IIG@=HD(HHLHt(Ld$ H\$(HD$0?HH MI,$ID$LP0fH (H((I/,IGLP0DIGLP0I@LP0pLHcH0( H((׏fIELP0#IFLP0|5ծH H5m(H=(1EHH; H3THmu HEHP0H(H((#cHuLF[IH,HM( H:(8(rHd(HLLt$ Ht(H\$(HD$0THH M(I.IFLP0H (HLLd$ Ht(H\$(HD$0*THHH( H~(IEx(H@IEE1HMI,$ID$L$LP0L$@HuLYH= HpE1( H(IE(wHIEMHhE1IEL$LP0L$VfDHF HHD$(HFHD$ HHT$ HLL֏H5(OLHŒU(HB(@(HFHHD$ wI?IE@=H4(HHLHt(Lt$ H\$(HD$0 ;HHH)( H((M IHH E1( H m(k(tfD[IHH5(LHVmHHD$(IGfDH5(H=(1AHHHwPHmu HEHP0HA(H((6gHE1(H(I([HHۊE1p( H](IEW(yHIEw@;IMgMIoI$HEI/u IGLP0HEIMeMMuI$IImu IELP0IFH;B&H;&HL`1HLHh HD$2@LD$HII(MI@LP0DH3( H (HE(HHEHE1Im+fDHK( H(IE(HQH5B(H=s(1l?HHHNHmu HEHP0Hz( Hg(e(HMP( H=(IE7(HFHuLTIHHk( H((H DE1( H ((躪I蝧HuLTIHHM( Hq(o( H Ň]( H J(H(>IMuMImIHEImu IELP0HEIM\$M:MD$III,$uID$LD$LL\$P0L\$LD$I@H;9&iH;&LD$L\$L\$LD$HLX1LHHh LD$HD$=LD$LT$HII*MIBLP0Hy( H((]iHRMI( H((:H%( H((H( H((*MMMH5(H="(1MM( H(HE(HI@@=LHt$ L\$ L\$LD$Hl$(2LD$L\$HIAI+uICLD$LP0LD$HmuHELD$HP0LD$MH{( H(IE(HHL( H((H%( H((KHE1M( H}(Ix(яHHʂMM\( HI(HEC(ˏH$LHt$ L\$ L\$LD$Hl$(HLD$L\$HIHY( H((AIM茞H'( H((hMHME1( H(}(WHӁk( HX(V(QuHMM>( H+(HE%(ďHHw( H((JHEHP0E1MMmIAWAVAUATUHSHHH^H<$dH%(HD$81H&HD$ HD$(HH[HHF HD$LmH( H(hE111AHLHHH;H@u HP0HCHH5(HHIMH5(LjIHnI/u IGLP0L;5?&@L;5&@u L;5&@I.u IFLP0sL#f.Off/f/Oy H$HT$H5I&Lp IHxLIH I.uIFLP0fH+HCHP0H]HHjH ZHOH#L LOL@HH&SHH5]rH81+H~xy(k(xH\(XZH ~H=E17UHL$8dH3 %(LHH[]A\A]A^A_ÐH&HD$L-y(H="(IULIHHIFH5(LHH IIHMk IHu IFLP0L5 (H=(IVL话IHHID$H5s(LHH` IM I,$u ID$LP0IGE1H;&7 H;&H;&Hc*uIHI MtL`HcH1I\HcLLH(LD$HIDl3LD$HH I(I/u IGLP0Hڙ&I9E HLEmIHi Hmu HEHP0Imu IELP0L;=&@L;=]&@u L;=~&@I/ L%k(H=(IT$LIH HIEH5(LHHB IIEHM IEHL-(H=(IUL褛IH HIFH5(LHHS IIHM IHuIEE1H;w&& H;&H;&Hc'jIHd MtLpHcH1I\HcLLH(LT$HIDa1LT$HH= I*u IBLP0Imu IELP0Hȗ&I9D$ HL2kIH Hmu HEHP0I,$u ID$LP0L;=&L;=J&L;=h&rL貚Ņ I/u IGLP0 H$HT$HH5͗&Lx IHxMIIH I/IGLP0@L8HcHyE1E1Z(%HG(IB(HfDIHu IFLP0MMMtI/uIGL$LP0L$MI(I@LP0IHHiHHH5(LIHVIΘHD$ HMHD$(Ll$ HD$H x#E1H={H F(H(#:(%OHx#(%H((fDH ((E1H=V{5(NۚIIG@=H(HHLHt(Ld$ H\$(HD$0&HH MI,$ID$LP0fHwS(%H@(>(I/,IGLP0DIGLP0I@LP0pL踗HcHHw(/H((fIELP0#IFLP0|5腖H H5}(H=>(17-HH; H;Hmu HEHP0HvE()H2(0(҈#HuLBIH,Hev(/H((.H(HLLt$ Ht(H\$(HD$0=HL$8dH3 %(HyHH[]A\A]A^A_ÐH&HD$H( H(hE111AHHHHH;H@u HP0HCHH5(HH|IMH5(LZIHI.u IFLP0L;-/&AL;-Ԅ&Du L;-&AĉD$Imu IELP0t$H f.6D$ H5g(1HłIH H;&L;-P&L;-n&L踆Aą Imu IELP0E# H<$H5(HGHH IM LH&IHImu IELP0L;5&L;5&u L;5ƒ&TI.u IFLP0H$D$HT$H5Â&Lp IHxL迺HHsI.)IFLP0HHh@H-Q(H=(HUHIHZ HIFH5(LHH8IIHM IHu IFLP0H-(H=(HUH臄IHK HIGH5L(LHH IM I/u IGLP0IEE1H;]&: H;{&H;ւ&HcuPHH MtLxHcD$HHLH\HH(QHHcHD1GIHHmXImu IELP0H&I9D$ LL"TIH. I.u IFLP0I,$u ID$LP0L;-&@L;-:&@u L;-[&-@ʼnD$Imu IELP0L$ H-9(H=(HUHނIH HID$H5(LHH.HI$HH%I$HL%(H=v(IT$LqIHHIFH5V(LHHhIIHM)IHH<$H5(HGHH_IM'IG1H;&JH;6&PH;&HcHL$HL$HI=HtHHHcD$H1LLI\@HMtIH7Imu IELP0I/u IGLP0Hs~&H9E LHQIHiI,$u ID$LP0Hmu HEHP0L;-P&L;-~&L;-&L^Ņ5Imu IELP0H$HT$HH5~&Lh IEHxMHH^ImlIELP0HHHF HD$HnfDHi~&HD$f.L言D$]H7`(H((Imu IELP0H ((H=c5(z6HH1H/HGP0fDDjIHHHHxyHH\ @HD$(Hl$ HD$@H[_(H((0fHHIDH_(H((H ((H= b5(o5f.kI|IE@=vHl(HLL|$ H\$(HD$0HcD$HHt(M IH MI/IGLP0HC^(H((I1E1E1HIHu.E1IFLP0MtI$HI$Hu ID$LP0MtI/u IGLP0E1MtImu IELP0HtHmu HEHP0MI.IFLP0rHEHP0fL}ŅHI](H((0Lx}D$H](H((DID$LP0IFLP0o[IHcD$LHL$ HL$HH\$(Ht(Lt$0"HL$HI HtH)u HAHP0I.IFLP0DH(LL|$ H\$(HD$0HcD$HHt(%"IHH[1(Hw(I$q(HE{HH[?(H,(*(DHF HHD$(HFHD$ VuHHT$ HLL^H5k(H2[(?H(( HFHHD$ tHHsH5<(LHVzHrHD$(HE[KzHuH.'IHHZ5(H"( ({}HsZ (H(( HH KZ1E1E1H (((} IG@=HcD$HLHL$ HL$HH\$(Ht(Lt$0HL$HIXHYMI[(HH(IC(%Hf.HY#(H(( OfxHuH%Ht H@Y1E1(H(I$(HH5ѱ(H=b(1[HHHHmu HEHP0HXi(HV(T(C{IHX1)(H(I$(HrH[X(H((,0fH<$zIM}MImIHEImu IELP0HEID$@HWc(HP(N(.fIL$HMD$HII,$uID$LD$LHL$P0HL$LD$I@H;u&H;u&kLD$HL$MvHL$LD$HIHH1LHLp LD$] LD$HII/MYIGLP0JH5(H="(1 HH|HHmu HEHP0HV)(H((?qHcV(H((I.A4 @H#V(H((I$HfHU(Hp(n(dfHUE1P(H=(I$7(HfDH59(H=ھ(1 HHHHmu HEHP0HIU(Hξ(̾()H5(LHVtHHD$ HtHuHd!IHHTk(HX(V(LwHH TE1E11(H ((tHuL HHWT(Hܽ(ڽ(H 0TE1Ž(H ((vIHSM(H{(y( H<$kvIIOHMgHI$I/}ID$MD$xLmML}IEIHmu HEHP0IGH;q&yH;q&0RrIHLh1HLL` q IHI.LIFLP0HR|(Hi(g(GE1HRR(H?(=(j~HRMI%(H(I (,HMH_RMM(H޻(ܻ(7sH5(H=(1HHH.Hmu HEHP0HQ(H}({(yI@@=LHt$ HL$ HL$LD$Lt$(kLD$HL$HIH)uHALD$HP0LD$I.uIFLD$LP0LD$MHDQܺ(Hɺ(Ǻ(IGHL$LMP0ID$D$HL$HP(Ht(r(LHt$ HL$ HL$LD$Lt$(LD$HL$HI HPIM(H ( (ąH_P1(H(I$ܹ(H>H.Pƹ(H((;IG@=Ht$ LLl$ Ld$(IHImu IELP0I,$u ID$LP0LMaHO4(H!((qHuOMI(H(I(H/HAO1IMHȸ(IǸ((ԅHIHOLM(H(IE1y(څHHt$ LLl$ Ld$(IHHNL8(H%(#(NHyN(H((uNjHMNL(HϷ(ͷ(^aH#NL(H((dPHMIM(Hx(v(ͅI"I0HMLQ(H>(<(WAWAVIAUIATIUSHh$dH%(HD$X1H;5ek&H-h(H=(HHUH mIH HIGH5(LHHL HIHHG IHcHE1E1H;i&` H;j&H;Tk&HcL$kL$HI" MtLxHcHHHcI\Hj&HIDHEHH LD$}kL=j&LD$p VA;P  1LLD$HHHkLD$p VP A=H9H I(Hmu HEHP0H;u HCHP0H(sH{ IVL=ְ(HD$HCLHHT$HD$kHHl H@HT$HHHLHHQ IVL=(HLHT$kHH H@HT$LMHLAHH H@H;g&YLqMLLQIIH)IBH;h&Lt$8; H;9i& IRB HrE1Ht$( uMBLD$ LT$iLT$LD$ p L=%h&VA;P  LT$ LLHD$(HD$ZiLD$LT$ H QP A=929^M I.LuIFLD$LLT$ P0LD$HL$ H)uHALD$HP0LD$I(u I@LP0kL|$HD$HD$M4H~$$LLIhIGM9uH|$cHEL-(LM^ ihL=g&p VP A;` 1LHAI=hp VP AH= @9M HmuHEHP0M I,$u ID$LP0HHHBHHHCHCHP04@IGLP0fL=(HiLHhHHCH@HHHHLHH*IVH-h(HHHT$xhIHH@HT$HHLLIHgH@H;d&MwMIWIHI/uIGHT$LP0HT$HLHT$7HT$HHU I.Iu IFLP0I/u IGLP0Hmu HEHP0h$LLHfHIsaHCL-(HH #fL=d&p VP A;1LHHeH QP AH= @9tHH+uHCHP0H]Hmu HEHP0LdHHDH\$XdH3%(H Hh[]A\A]A^A_fHE@=hH\d&L|$@HE1H\$HHtHHD$PHEL@@ uLM HLAHH HE](lHJ(H(&MtI/u IGLP0HtHmuHEHP0DH ((H=H15(fI@LP0c @9ALD$XdLD$@$)f.HALT$HP0LT$9H1c&HHH\$HHtHL|$@HD$P HHMI/IGLP0@HlHHL$k HL$HIHeD(oH((9'HmH)uHAL$HP0L$M)I(u I@LP0H ((H=5G15({HHHP/fDHEfDLHHUHC?(hH,(*(r&H+HCHP0HyIbHuHIHH-Cl&(lH((&H CH=5F1@b@$f.@D[eHHBIH7(9(l+(&H IGLP0L}MHUIHHmt.HBHչjf.[a@$~HEHT$HP0HT$@Hi`&LH8]HA''(oHq(o(''H AoH=D1EHHPfLD$ LT$`LD$ LT$@$fDHsA (lH((&MI/fDH_&LH8\HmHAH((o()'u HEHP0H ((5(1LHLD$bLD$HHDHmH@H0(2(l$('bHEL$HP0L$KH=-LD$^LD$L$_L$HzH\&H5-H8\L$[Hl^&LH8[H?h`&(hHo(m(`&LLLT$LT$IMH?ML4(oH!((6'2H]&HH88[H+H_?H((h(b&HCHP0^HqHmu HEHP0H ?y'(oH((y'f1LH`HmIH=Q,d]H >j&3(jH=AH ((&LHt$8LT$vLT$IH=+\[H+u HCHP0H >h&(hH((&1LH_H+HO@`]HuHZ&H5u+H8UZ{LB1 uIzLT$Ht$8ALT$IL$\L$HH.Z&H5+H8YL$fDH=*[LT$LD$ nH=LI(hH((o&^H<{(lHh(f(&H1LAHHY&H5X*H88YtX1ɺALT$IHEHL$HL$P0HL$L$fDAWAVIAUATUSHxH|$(Ht$0HT$HL$LD$ dH%(HD$h1H(H(hE111AHLHH8HH( H(hE111AHH|$HHH8|HEH5)(HHHIMHCH5(HHHIMLƺLLD$OWLD$HIH;/Y&L;-X&AA L;-X&L$A"taImKH5t(LǺLD$VLD$HIH;X&L;-hX&AA ĉL$L;-X&A I/kI(EImut$L%R(H=(IT$LYIH*HIBH5(LLT$8HHLT$8IIHMIHmL%ޛ(H=(IT$LYIHHI@H5G(LLD$8HHLD$8IIHMIHu I@LP0IEE1H;EV&2H;cW&mH;W&HceLD$83XLD$8HIRMtL@HcD$HL$LLHHILH(HcHID1 IHI,$uHD$ID$LP0LT$ImuIELT$LP0LT$HvU&I9FLLLT$(LT$HIkI*u IBLP0I.u IFLP0L;=LV&AL;=U&Du L;=V&AʼnD$I/u IGLP0L$HH=(IH8H@H5(LHHIMI.H=(vIHH@H5(LLT$HH/LT$IIHM'IHIEE1H;#T&)H;AU&H;U&HcLT$VLT$HIMtLPHcD$HL$LLLD$HHILHw(HcHID1LD$HIF#I(u I@LP0Imu IELP0H`S&I9D$LL&IHI.u IFLP0I,$u ID$LP0L;==T&AL;=S&DL;=S&LIVD$hI/u IGLP0T$"H=җ(IH"H5(HIH#I,$uHD$ID$LP0LD$H=(LD$QLD$HI$H5(HLD$LD$HIC$I.uIFLD$LP0LD$IEE1H;R&A$H;!S&/!H;|S&HcrLD$SLD$HI&MtLxHL$HcD$LLLD$HHILHO(HcLT$HID1LT$LD$HI9&I*uIBLD$LP0LD$ImuIELD$LP0LD$H"Q&I9@$LLLD$$LD$HI$I.uIFLD$LP0LD$I(u I@LP0L;=Q&L;=Q&L;=Q&LSAą$I/u IGLP0E$HD$0H=4(L8HD$(IHL%n(HD$HD$ H;MQ&IT$LSIH)HIFH5w(LHH*IIHMw*IHu IFLP0QIH*HD$ HIFSIH*HP&H5 (HHD$>TLT$%IELM,LT$QLT$H QP HP&;-LLT$LLAI]QLT$H QP HO&=p4@9%M;/ImuIELT$LP0LT$I.uIFLT$LP0LT$I*u IBLP0I<$u ID$LP0L-(H=X(IULTQIH5.HIAH5ٗ(LLL$HH<-LL$IIHM,IHu IALP0IF11H;N&.H;4O&)H;O&Hc$t$HL$PHL$t$HIV/HtHHHcHE1LHIlFHHI\FLI$HMdIH?/ImuHD$IELP0LL$I.uIFLL$LP0LL$I9IAuLLL$P0LL$IAHH5(LLL$H.LL$IM00ID$H5(LL$LHH/LL$IM/LֺLLL$LT$KLT$LL$HIi/I.uIFLL$LLT$P0LL$LT$I*uIBLL$LP0LL$L;-rM&AL;-M&D#L;-4M&#LLL$yOLL$A2ImuIELL$LP0LL$E 1L-(MwLL$ID$LLHD$ 0OLL$HHD$-H@HHj&LL$LLH|$LL$HHD$~-L5(MoLL$LLNLL$HIy-H@HHM&LL$LLLLL$HI`-H K&I9C#MsM#MSIII+uICLL$(LLT$P0LT$LL$(LLLL$(LT$0LT$LL$(H2IMHQIHu(IVLL$(LHD$LT$0R0HD$LL$(L\$0I+uISLL$(LHD$R0LL$(HD$H(uHPLL$HR0LL$LL$(NLL$(Ll$ HD$IAMtH H\$(LLl$Hl$ LDH8LH0H0H0H0%L1HEHE 2/ H(H0H0H@(;u HcHLH0H@H0Pt8 H(HR8HcR H0H@HP0GfH@HP0uIGLD$8LP0LD$8I(I@LP0Ew:f.H|$Lf.D$rH|$HKIHH?ff/D$ \$f/xL-k(H=(IULKIH HI@H5U(LLD$HH= LD$IIHM IHu I@LP0D$IIH HG&I9G LLLD$)LD$HII(u I@LP0I/u IGLP0L;-H&L;-?H&u L;-aH&DImu IELP0E=HD$0D$LHt$ LHD$(IH8LLT$vLT$HII*Hm%IELP0t$DLLD$8 JLD$8)fLID$$Hw)ME1 (H((~IE1HIu#E1IBLP0MtImu IELP0MtI.u IFLP0MtI,$u ID$LP0H ((H=6,A5v(aHmu HEHP0HtH+u HCHP0HL$hdH3 %(L 'Hx[]A\A]A^A_DIELD$LP0LD$IBLP0H K(~E1H=+H Ƒ(ȑ((~eH ((H((~f{JIjH'c(HP(N(~f;JIVH'#(H((~I/uIGLD$LP0LD$ME1'fH;'Ӑ(H((~I/uIGLD$LP0LD$IE1E1HIHE1I@LT$LP0LT$MI*z@IE@=HD$HLLD$@LD$8HD$HHJ(HD$PHcD$HHtH;LD$8HIMI(I@LT$LP0LT$fDIBLP0.fLhFD$H%E1(Hy(w(dfIFLP0/LFAƅ^H%ME1:(H'(%(e,IBLP0@IE@=`HD$HLLT$@LT$HD$HH(HD$PHcD$HHtHLT$HI!MI*IBLP0x{DHuL^IH0H$e(HR(P(E1H5}(H=(1IHsHI,$u ID$LP0H`$(H((HfDoHP0H;0NHHP0H0H0H0~@AʼnD$H#{(Hh(f(~HD$LLD$@LD$8HD$HHj(HD$PHcD$HHtHLD$8HI Hu# (H((M(IE1HABHzH#(H((~D{BHuL^IHH"e(HR(P(3BHIH5{(H=(1IHHI,$u ID$LP0HZ"(Hߋ(݋(~BH +"E1(H ((IHE1E1@DLT$8IfDH5z(H=*(1#IHHI,$u ID$LP0H!0(H((~fDHI9IH\$(Hl$ H|$LL$ ;Lt$H5}(1LILL$ IHQHL$HIuH|$LL$HGP0LL$M$ImuIELL$LP0LL$I<$>I)IHIHIGLP0DH E1((H ((wfDCLD$8IfD?HuLIHH- ʼn(H((0MEMMeII$ImID$MD$f:LcJo(H+o()o(gLLLLT$Q'LT$HIkL[E1E1n(Ln(n(W(ImH#MMn(Hn(n(_HA@=4HHt$@HL$L|$@Lt$H蕳HL$HIW I/uIGHL$LP0HL$I.uIFHL$LP0HL$IMH=LT$#LT$LMLE1E1Lm(m(m(k(H)3E1$&LL$IHm(Hm(m(WM^#HuLAIHL~'>m(~LH=LD$Lm(m('LD$LL(LD$H=LD$HD$"LL$HuLLL$LL$H]Ll(Ll(l(i(H l(l(H=NLD$5sl(LL$YLL$LD$5LT$@"LT$HBHs&H5LH8,LT$"INHjMnHIEI.IEM=LCk(Lk(k(,(H)A H k(Hk(k(MVM1MfII$I.yID$MALAk(~L.k(,k('$ILpk(Lj(j((L#LL$I=L7E1j(Lj(j((Hs&LLL$H8LL$Lj(Lvj(tj((H0&LLL$H8LL$LE1Cj(L0j(.j((HHt$@HL$L|$@Lt$HHL$HIHRIi(Hi(Hi(H;L!i(Li(i((Li(Li(}i((s"LL$ILYi(LFi(Di((HI/i(Hi(Ii(HHiIh(Hh(h(6L?h(~Lh(h((MrME1LL$LL$HuLLL$mLL$HI/L(LL$LH=,LD$LHh(Jh(DAWAVAUIATIUSHh$L$dH%(HD$X1H;52&H-5X(H=_(HHUHIHk HIFH5L[(LHH HIHH IHHE1E1H;& H;&`H;!&Hc L$L$HI MtLpHcHHHcI\H&HIDHEHH OL5&p VP A;> 1LHH$p VP AH=j9pH I/?Hmu HEHP0H;u HCHP0Ho^(sH{ L=Y(MuHD$HCLLHD$HH H@HH.HLLHHL=xY(MuLLHH} H@LMIHLLAHHc H@H;&LiMLIIEIH)\IAH;&Ll$8 H;'&A IQBJ HrE1Ht$( uMyLL$ L5%&LL$ p VA;P  LL$ LLHD$(ITLL$ p VP A=29 M ImLuIELL$ LP0HL$ H)u HAHP0I/u IGLP01Lt$IHD$MHEH5D(HHHIMHCH5D(HHH$IM&LLIHH;&L;-q&AA L;-&t$AtWImH5N(LIHH;h&L;-&AA ĉT$L;-)&A I/AI.KENIm T$GL%E(H=M(IT$LIHHIEH5J(LHHIIEHMIEH L%E(H=5M(IT$L0IHHIEH5F(LHHHHImuIEHL$0LP0HL$0HAE1H;%H;&H;t&HcLT$8HL$0HL$0LT$8HIMtLPHcD$HT$HHL$HHITHH@(Dn1McLHKDƘHL$HIeI,$uID$HL$LP0HL$H)u HAHP0H"%I9G8LLIHImuHD$IELP0LT$I/uIGLT$LP0LT$L;%AL;%Du L;%EI*u IBLP0EL%C(H=BK(IT$L=IHHIGH52H(LHHIIHMIHH=/C(IHH@H5D(LHH/HIEHHIEHuHAE1H;%H;%H;0%Hc_HL$HL$HIMtLhH$IcEnLMcHLT$HITH >(1HL$HKD荖HL$LT$HIQI*uIBHL$LP0HL$H)u HAHP0H%I9D$pLLPIHI/u IGLP0I,$u ID$LP0L;-%L;-i%L;-%LAƅImu IELP0EH=]A(0IHH5 F(H腓IHI,$uHD$ID$LP0LT$H=A(LT$ܫLT$HIpH5C(HLT$'LT$HHtI,$uHD$ID$LLT$P0LT$HL$HAE1H;%mH;%H;%HcOLT$HL$kHL$LT$HI!MtLxH$IcHLLT$HITH;(AVHcH $HID1OH $LT$HIJ!ImuIELT$LH $P0LT$H $H)uHAL$HP0L$H%I9BLLL$L$HHI,$uH$ID$LLT$P0H $LT$I*uIBH $LP0H $H; T%H; %H; %HH $^H $AH)u HAHP0EHD$(H=F(L0HD$ IHL%>(H$HD$H;%IT$LkaIH@$HIEH5A(LHH$IIEHM~$IEHuIELT$LP0LT$LT$ALT$HI HD$LT$HIELT$HIq$H)%H5JA(HLT$HD$xLL$LT$IBLM/&LL$LT$LT$LL$P L=O%A;P 1'LLL$LLLT$AILT$LL$P P A=p4@9f M'I*uIBLL$LP0LL$ImuIELL$LP0LL$I)u IALP0I<$u ID$LP0L-<(H=~D(IULzIH&HIGH5@(LHH&IIHMU&IHu IGLP0IEE11H;E%'H;c%#H;%HcHL$3HL$HI*HtHHIcHE1LHLLT$IlAGHHI\AGI$HMdLT$HI)I*u IBLP0Imu IELP0I?IGu LP0IGHH5 9(LH)IM*ID$H58(LHH*IMO*LκLLL$7LL$HI*ImuHD$IELLL$P0LL$LT$I)uIALT$LP0LT$L;%AL;}%DL;%LLT$LT$A+I*u IBLP0E'+IVL-=(ID$LHHT$HD$HD$H(H@HT$HH LH|$HD$Hg(IVL-7=(HLHT$GIH)H@HT$HHb LLIHk)H%I9CEMkM8MKIEII+uICLL$LP0LL$LLLL$LL$Hy+ImMuIUHD$LLL$ R0HD$L\$ I+uISHD$LR0HD$H(u HPHR0dLl$HD$IGITH LH\$ H$MHl$ILII$8HH0I$0H0 J1HEAD$ID$ -H H(H0H0H@(A;t$ HcI H0H@H0Pt8*H(HR8HcR H0fH@HP0~H@HP0IGLP0I.IFLP0E@H<$f.f(H|$ $f.~ $ff/f/f/ [HD$(Ht$L(HD$ IEH8LIH Imu IELP0Hm%/fDIELP0T$-DL ALD$H1E1*>()H>(>( E1fIEHIEHuIELT$LH $P0H $LT$MtI*uIBH $LP0H $HtH)u HAHP0MtI,$u ID$LP0H =(=(H=vA5z=(eHmu HEHP0HtH+u HCHP0HT$hdH3%(L!Hx[]A\A]A^A_fIELP0 IELP0H [&E1H=H <(<(&<(赩qH<('H<(<(fIHs<()H`<(^<(fKIHE11E1H"<(IE1<()<(H@IHu*E1IGLT$LH $P0MH $LT$tI.t{MMt*I/u$E1IGLP0IHMxLL;(;(<v;([fDI.E1E11E1IFLT$LH $P0LT$H $df.H#;()H;(I ;(HIHuE11E1E1HA@=`HD$Lcl$HHLT$@IHD$HH.(JtHLT$0HL$HD$PHL$LT$0HIMI*IBHL$LP0HL${@LLT$LT$AHM#:(6H:(:(fIGLP03IEHL$LP0HL$rHA@=H$HHHL$Ll$@HD$HH-(HD$PIcHHtH~HL$HIoMImIEHL$LP0HL$+HuLH$HE11E1H9(I9(68(HH5)((H=8(1蓅IHH?I,$u ID$LP0H8(2H8(8(zfDGHP0H;0HHP0H0H0H0^@D1H#8()H8(I 8( HfHD$Lcl$HLT$@LT$0IHD$HH,(JtHHL$HD$PFHL$LT$0HI0H E17(6H7(I7(ςHwM$X $H5HM7(*H:7(87('HH5m&(H=6(1ǃIHHsI,$u ID$LP0H<6(.H6(6(: f.HuL~IH6H6(6Hr6(p6(cI"HE11E1H:6(<6(6.6(!fH5i%(H=5(1˂IHHwI,$u ID$LP0H@5(0H5(5(Z$fDHL9cLHl$H\$ MIH|$NH5'((H|$1;IHD$HH$HHu H|$HGP0M+Imu IELP0I<$I/vIHIHvIFLP0gDH[ME14(6H4(IE14(HfHH E114(3H4(IE4(Ht@LQMLaII$H)ID$LD$f.BLcJIH3/(8H/(/(AHALT$0HP0ID$LD$LT$0HE1s/(8H`/(^/(CQTHLiMLqIEIH)u HAHP0IFLAIL$HMt$HII,$uID$HL$LP0HL$IFH;Y%JH;%_ HL$'HL$HIHH1HLLx HD$<{LT$HIgI*M IBLP0H1E1=.(8H*.(I%.(~HHw.(8H-(-(cHPM-(8H-(-([IH%HIBH5{*(LLT$HHdLT$IIHM=IHu IBLP0IEE1E1H;%iH;%( H;/%Hcs IH2MtLxIcHE1LHLLL$IlAD$HHI\yLL$HII)u IALP0Imu IELP0I?u IGLP0L%%(H=,(IT$LIHwHIEH5)((LHHHHIEHHIEHuIEHL$LP0HL$IGH5 "(HL$LHHHL$IMHL$^HL$HI$LhHL$HD$LL$HL$HI"HE%H5f'(HLL$HL$HL$LL$LHLLL$HL$xHL$LL$HI(H)uHALL$HP0LL$I)u IALP0Imu IELP0I<$ID$LP0H$McHLT$IH $HD$HH(JtHL|$@HD$P8H $LT$HI[H*(:H*(*(ۃMIE1E1H^HY*(8HF*(D*(naH5p(H=)(1vIHH薅I,$u ID$LP0H_)(9H)()(CH8)(:H)()(ƒIF@=rHt$@LHL$@HL$Ll$HnHL$HIH)uHALT$HP0LT$ImuIELT$LP0LT$MH%)(:H)()(ă_Hf1((:H((((ǃH=ME1((:H((I$E1((ɃHLyMLaII$H)NID$LAWHɾa((.HN((L((6H:((+H'((%(("IJHMzHII*uIBH $LP0H $IGH;t%7 H;%Y H $CH $HIHH1HLL` ^tHHImMH$IELL|$P0LT$H $xHV'(0HC'(A'(VH/'(2H'('(v{Hp'(:H&(&(HIIE1E1H&(1E1&(:&('H&(:H&(&(H5(H=D&(1=sIHHI,$u ID$LP0HJ&(;H7&(5&(6H#&(:H&(&(IEHHt$@LHL$@HL$Ll$H`HL$HI`H*E1M%(6H%(I%(HL%(L{%(y%(+I*E1E1ImuIELD$LL $P0LD$L $MtI)uIAL$LP0L$'%(H %(L$H= 5 %(ML$9I,$.ID$L$LP0ML$AE1I/L$(L$($(+I*E1E1E1E1IBLD$LL $P0LD$L $MLL$LT$LL$LT$@$xE$IE@=aIcHLHL$@HkHL$Hl$HH\$PHtHLd$XiHL$HIqHH)HAHP0IE@=xIcHLL|$@HkHl$HH\$PHtHhIH/MtI/u IGLP0MIF@=Ht$@LHL$@HL$L|$HQhHL$HI8H)u HAHP0I/u IGLP0MLL\$+L\$HL(M"(L"("(C,HT$HH$HH^ E1H|$LD$L $HGP0L $LD$HĸE1Y"(6HF"(IA"(ׂH0HALT$HAP0ID$LLT$Hg!(7H!(!(&KHuL賄IHL"+!(LH=L!(L$!(+xL$ID$LP0LʷIELU!(W!(I!(+HE1E1E1-I'HD$HLr !(L ( (+wIHBE1M (6H (I (HIcLHL$@HkHL$Hl$HH\$PHtHLd$X}HL$HILжh (LU (S (+Ht-H)u'E1E1HALD$HL $P0L $LD$E1E1IcLL|$@HkHl$HH\$PHtHs|IHsLB(L((E+MtI/< ME1E1E1(H (L$H=5(rL$Ht$@LHL$@HL$L|$H{HL$HI7HME14(8H!(I(H LLLLL$LT$;LT$LL$HI*L@E1E1(L((+I*G&H 1E1MH(I(6(HpIG@=Ht$@LH $HL$@Ld$HscH $HI H)u HAHP0I,$u ID$LP0MLHa(8H((]H=СLL$LT$LT$LL$L LE1E1L(((+H)7E13pIQSHuL6HmL@(L-(+(+)(H (L$H=5 (L$LL$L$L$LL$HH%H5H8LL$L$L+L$H=莉L$HuLcIHLҲ1+`(LH=MLC(L$=(1+(L$*LT$IIMH6M}HIImIGMAM}MMeII$Imu IELP0ID$MATH(9H(}(3Lαf(LS(Q(~+H)AL/(L((3+wLpE1(L((^+LFM(L((S+MDIm ME1EHt$@LH $HL$@Ld$HwH $HIDHҰMg(:HT(R( oHHL$IELE1E1((L((r+ HHuL|IHoL>p+(LH=L(L$(p+蔆L$L(L{(y(+!oIuL¯E1W(LD(B(+H%LH8[L(L ( (1,H_M(8H(I(HI#1H#ME1(8H(I(HL(Lt(r(,LȮ`(LM(K(+AItL,(L((+LmE1E1(L((u+^H%LH8L.E1(L((3,L(L((w+1LݭE1r(L_(](|+HK(:H8(6(ML~(L((,LWE1E1(L((,H5p (H=(1zcIHH&rImu IELP0L(Lu(s(,CHɬa(;HN(L(2IEHL$LMP0IGAHL$BLw(L((+HPME1(:H(IE(!HHM(:H((HE1M(6Hp(Ik(HZHT$LP(L=(;(@,HH$HHImILj(L((,E1IH3ME1(8H(I(HHM(:H((E1E1E1LǪ_(LL(J(+L8(L%(#(M+MZImAE1Mff.AVAUATIUHSH L;%H^dH%(HD$1H$HD$LD$HHHHHH HMHSHsL@HY%H5L H81H/(!(H(XZH sH=1$fLF(IL$ IT$Hu H}UH|$dH3<%(H []A\A]A^@IH~qHtHHF(HD$ID$ LHD$ID$H$IHtlHHt9MH$HL$LD$WfHHoIH55 (LIHV5H$H$H5! (LHVHD$HIM~H5(LHVHt+HD$ISHFHH$IHLHLH5T'OgH§U(HB(@(#HH0%H H5jL {AHH81H]Y^H(((I\$>DAWAVAUIATUHSHXH|$LD$dH%(HD$H1H;%L=(HH=(IWLIH HIFH5 (LHH IIHM%IHuIFLT$LP0LT$L%z(H=#(LT$IT$LLT$HIHIGH59 (LT$LHHQLT$IIHMOIH IBE1E1H;%H;% H;S%Hc2LT$ L\$L\$LT$ HIMMtL`IcAHHMcI\O\IBHHLT$zL=%LT$H QA;P  1LLT$LHELT$H QP A=H9HJ I.{I*1H;HCLH޿H (HD$1IHH8H_(MnsH{ I9HD$L= (L`LLIHIH@HH"LLHt$IH.HD$L=] (L`LLnIHbH@HH^ LLHt$IHGH@H;%L MzM? IRIHI*uIBHT$LP0HT$HLHT$HT$HIhI/I I*u IBLP0I,$u ID$LP0Ld$HD$IFM+H>'H''RHt$ HLd$ Lt$( GIH`HMI'Hn'Ii'ۇHyHS'H@'>'1H,'H''ZHmE1I'H''uH@MI'H'I'HlHMI'H'I'HIAWAVIAUATIUSHHH|$dH%(HD$81H;%jL-'HH=Q'IULMIHY HIGH5'LHH IIHM IHL=?'H='IWLHH HHEH5 'HHHIHEHM` HEHu HEHP0IE11H;% H;̜%H;'%HcFL$蝝L$HI HtHhHcHHHcI\M|IEHH LL$[L=%LL$p VA;P  1LLL$LH&LL$p VP A=H'9OH I)ImH;Hx'sH{ IVH-'IHCHHHT$HD$ܝIH H@HT$HHLLIH IVH-x'HHHT$舝HH H@HT$LMcHLAHH H@H;ә%ULqMHHQIHH) HLHT$mHT$HH I.HuIFHT$LP0HL$H)u HAHP0Hmu HEHP0ƝHl$HD$NtM~fH|$HAEI9uH|$pIEL% 'HH L=%p VP A; 1LLHH QP AH= @9HU ImuIELP0H Hmu HEHP0HIHPHHH$HCHP0DIGLP0L=a'LiLL芛HHH@HHHLLHHL-,'M~LL=HH! H@HHHLLHH H@H;%uLuMhL}IIHmu HEHP0LLjIH| I.Lu IFLP0Hmu HEHP0Imu IELP0芛H|$IALD$QHCL%'HH L=%p VP A; 1LHH֘H QP AH= @9H_ H+uHCHP0HHmu HEHP0D$tIHHt$8dH34%(L HH[]A\A]A^A_f.HCHP0JIELP0H;5f.IALP0Im f @9 HAHT$HP0HT$LL$覗LL$@$HtIbIE@=Hl$ HE1H\$(Ht(L|$0IELH@ uMUHLAHHHwt'Ha'_'HuHmjHEHP0MTMI/~1HHL$[SHL$HHHUw'H''ImH)9HAHP0*@HRIHHv'H|'z'H+HCHE1P0HHmf.IEcfDHHLH\$(Ht(Hl$ L|$0KHuHE1H6fD軓@$f諓@$fHkt'H''xHEHImtM1)f.1fHEHP0ImHPIUHEHImuIEHT$LP0HT$HBIչDH%HH8nHs,'H''H rsH=PwIf.H%HH8ImH4sH'''IELIP0H ''L5',fHrs'H`'^'HeHEHZ1LLLL$nLL$HHDHsrL'H''+fDH%LH8H1r'H''7DH=_LL$蟐LL$eDLL$^LL$HHH%H5j_H8JLL$(H%LH8^HqE1'H 'H'HHU1HMq'H''Z'1LLImHwH=^謏Imu IELP0Hpq'H^'\'@1LH芒H+H4%H=)^<H+u HCHP0HopC'H''CuΏHAH%H5]H8迌&襏HuH%H5]H8蚌qHoLH'Ho'm'HoMLU'HB'@' Ho.'H''H*H1LAHk谋Im1I/@IEHL$LP0HL$E1[I H1ff.AUATIUHSH(H݌%H^dH%(HD$1HT$MHH@HHnH nHIHHnH?L xuHLIL@HH%SHWrH5aH81臐HBnf'M 'fH'XZH nM H=bD1H|$dH3<%(H([]A\A]DHWH-HFLHD$謇HHT$fDH] H5E%HH}HHt8H+zHSHD$HR0HD$aDHVf.H+H GmH '' 'fuHHSHD$HR0H ''5'HD$H=aHD$CHD$fDf HT$HLLpH5[{'6,Hlf<'M H)'''fdfLhIHH5'LHV)HtHD$IE衈AUATIUHSH(H-%H^dH%(HD$1HT$MHH@HHlH lHIHHkH?L rHLIL@HHP%SHoH5 _H81׍Hk X%'' XH'XZH ikH=;`A1H|$dH3<%(H([]A\A]DHWH-HFLHD$HHT$fDH] H5}%HH}H Ht8H+zHSHD$HR0HD$aDHVf.H+H jH &'('':XuHHSHD$HR0H ''5'HD$H=*_HD$@HD$fD:XHT$HLL!nH5w')HiW'Hy'w'WdfL踃IHH5 'LHVyHtHD$IEAUATIUHSH(H}%H^dH%(HD$1HT$MHH@HHjiH ZiHIHH"iH?L pHLIL@HH%SHmH5Y\H81'HhQu'g'QHX'XZH hH=]6?1H|$dH3<%(H([]A\A]DHWH-HFLHD$LHHT$fDH] H5%HH}H]Ht8H+zHSHD$HR0HD$aDHVf.H+H gH v'x' j'QuHHSHD$HR0H J'P'5F'HD$H=\HD$ >HD$fDQ HT$HLLkH5{t'&HIg{Q'H''{QdfLIHH5]'LHVɆHtHD$IEAAUATIUHSH(Ḧ́%H^dH%(HD$1HT$MHH@HHfH fHIHHrfH?L hmHLIL@HH%SH|jH5YH81wH2fo>'.'o>H'XZH f.H=+[<1H|$dH3<%(H([]A\A]DHWH-HFLHD$HHT$fDH] H5%HH}HHt8H+zHSHD$HR0HD$aDHVf.H+H 7eH ''X'>uHHSHD$HR0H ''5'HD$H=ZHD$p;HD$fD>XHT$HLLhH5p'&$Hda>,'.H''a>dfLX~IHH5'LHVHtHD$IE葀AWAVIAUATIUSHHH<$D$dH%(HD$81H;%uL-'HH='IUL訃IHd HIGH5'LHH IIHM IHL='H=C'IWL?HH HHEH5d'HHHIHEHMk HEHu HEHP0IE11H; % H;'%!H;%HcQL$L$HI HtHhHcHHHcI\M|IEHH LL$趁L=O%LL$p VA;P  1LLL$LH聁LL$p VP A=H29ZH I)ImH;H'sH{ IVH-'IHCHHHT$HD$7IH H@HT$HHLLIH IVH-'HHHT$HH H@HT$LMnHLAHH H@H;.~%`LqMSHQIHH)HLHT$uQHT$HH I.HuIFHT$LP0HL$H)u HAHP0Hmu HEHP0!Hl$HD$NtM~ DD$H<$HAEI9uH|$zIEL%P'HH sL= ~%p VP A; 1LLHHH QP AH= @9HX ImuIELP0H Hmu HEHP0HIHPHHH'HCHP0IGLP0L='LiLLHHH@HHHLLHHL-|'M~LLHH! H@HHHLLHH H@H;{%uLuMhL}IIHmu HEHP0LL OIH| I.Lu IFLP0Hmu HEHP0Imu IELP0H<$D$IAL$xHCL%2'HH M}L={%p VP A; 1LHH"}H QP AH= @9H[ H+uHCHP0HHmu HEHP0${IHDHt$8dH34%(L HH[]A\A]A^A_f.HCHP0?IELP0H;*f.IALP0Imf @9 HAHT$HP0HT$LL${LL$@$Ht;IWIE@=Hl$ HE1H\$(Ht(L|$0IELH@ uMUHLAHHH,\'H''HuHmjHEHP0MTMI/~1HHL$7HL$HHH[='H*'('bImH)9HAHP0*@HH7IHHG['H''H+HCHE1P0HHmf.IEXfDHHLH\$(Ht(Hl$ L|$0 HHHt Hm5I/tIGLP0e@HEfDyHuL&IHHEZ'H''H ZH=R^E10HYIH''{'H~IGLP0E1l[|I;yHuL&H6HY('H''Imu-E1IELP0MI/u IGL1P0HHmu HEHP0H ''5'H=H]E1/HuHE1H3fD x@$fw@$fHXS'H@'>'HEHImtM1)f.1fHEHP0ImHEIUHEHImuIEHT$LP0HT$HBIչDHav%HH8sHWP|'Hi'g'PH WH=[?.f.Hv%HH8^sImHWH'''RIELIP0H ''L5',fH+W'H'' HeHEHZ1LLLL$xLL$HHDHVLX'HE'C'++fDHt%LH8VrHV'H''7DH=CLL$tLL$eDL $uL $HJHr%H5CH8rL $+HQt%LH8qHUE1n'H['HV'HHU1HU5'H"' ''1LLKwImHtH=BsImu IELP0H.U'H''@1LHvH+H8)H=yBsH+u HCHP0HTM'H:'8'utHAHVq%H5/BH8q&sHuH1q%H5 BH8pqH@TLHҽ'H''_HTML'H''HS~'Hk'i'H*H1LAHkpIm1I/@IEH $LP0H $E1]I H1AWAVAUIATUSHHH- 'H^dH%(HD$81H!q%Hl$ HD$(HfHtHRHHHSIH SH9WHIHH_p%I?SIH5FH8L YA1tHRw)'b 'wH 'XZH mRb H=V1(HL$8dH3 %(HiHH[]A\A]A^A_fH1p%H$H' H(hE111AHHHHH;H@u HP0HCHH5q'HHIMH5X'LmIHI,$u ID$LP0L;=o%L;=Do%u L;=fo%DI/u IWLR0EHqf.;!D$wH-0'H=ٺ'HUHpIHHID$H5I'LHHHI$HHI$Hu ID$LP0D$nIHMHm%H9ELH@IH] I.u IFLP0Hmu HEHP0L;=mn%L;=n%L;=1n%{L{pŅ1I/u IGLP0 M} D$H$H5l%II}LHH I/uIGLP0H+NHCHP0?H-'H=J'HUHFoIHHIGH5;'LHHHIIHM3IHu IGLP0H-2'H=۸'HUHnIHkHID$H5K'LHHHI$HHI$Hu ID$LP0Hk%H9EHH?IHHmu HEHP0Hnk%I9FLL>HH I/u IGLP0I.u IFLP0H;-Nl%H;-k%u H;-l%DHmu HEHP0E Me H$HH5j%I$I}M触HH I,$ID$LP0HF H$HnjHk%H$LmAąH`ME11' H'޶'=wIHIHu IGLP0IfMI/Au IGLP0MtI.u IFLP0MtI,$u ID$LP0H f'l'H=Q15Y'D#IHH)HHhfHH@HD$(Hl$ H$0DHKL1' Hε'̵'*wH 'õ'H=pP5'"HH1Jf.HKH' Hm'k'9wf[nIHK1A' H.',';wI,$N ID$LIP0NfDHkAąQHPKE1' HҴ'д'5xHmAE1@HEHP0fmImHHJE11n' H['Y'wIHvD-jHuHJH$' H''Hw>f.HF HHD$(HFHD$ 6dHHT$ HLLNH5Z' HJv'b H''vyHFHHD$ cHHcH5'LHViHrHD$(HE[+iHuHIHH}IH' H''w,hHuHIH=H-IH²' H''RwH H1' H ~'|'wH Hc' H P'N'TwI$H~ E1fhHuHH] HpH' H''wRfDHCH۱' Hȱ'Ʊ'WwH HL' H ''wHfDjHLUMOL]IIHmuHEL\$HLT$P0LT$L\$ICH;e%H;"f%>L\$LT$fLT$L\$HI7LP1LHLp L\$L\$HIPI,$LID$LP0H G' H''dwI.HmAfDLUML]IIHmuHEL\$HLT$P0LT$L\$ICH;d%)H; e%L\$LT$xeLT$L\$HILP1LHHHX L\$L\$HIaI,$LID$LP0oDHE' Hp'n'wMII/fMVM?M^III.uIFL\$LLT$P0LT$L\$ICH;c%~H;c%L\$LT$adLT$L\$HILP1LHLx L\$qL\$HHI,$MID$LP0fH5i'H=2'1+HHKH Hmu HEHP0HDH6' H#'!'wP@HsD ' H'I'xHf.H;DE1Э' H'I'wHH5'H=Z'1SHHHHmu HEHP0HCH^' HK'I'Dxx@HCE1L-' H''ixCH5Q'LHV-cHHD$ HrIC@=LHt$ LT$ LT$L\$Lt$(L\$LT$HI"I*uIBL\$LP0L\$I.uIFL\$LP0L\$L\IC@=8LHt$ LT$ LT$L\$H\$(NL\$LT$HII*uIBL\$LP0L\$LIC@=LHt$ LT$ LT$L\$L|$(L\$LT$HHI*uIBL\$LP0L\$I/uIGL\$LP0L\$M|LHt$ LT$ LT$L\$Lt$(L\$LT$HIHxAML ' H''kwbLHt$ LT$ LT$L\$H\$(JL\$LT$HIHALMMH'' 'wXLHt$ LT$ LT$L\$L|$(L\$LT$HHH@ML5' H"'I'xH=Ho@' H''wxHH@Hݩ' Hʩ'ȩ'@xI.MLH@' H''{wI+E1H?MLl' HY'W'wH?ML?' H,'*'xI*H|?H ' ' 'w.LMMHF?MLب' HŨ'I')xHH?M' H''/xI*E1M3[H>ML`' HM'K'twH>LMMH'')' 'wHq>ML' H'I'"xH IFLP0OI M[IE1IAWAVAUIATIUSHHH^dH%(HD$81H[%HD$ HD$(H6H\HHF H$Md$Hv' H(hE111AHLHHKH;H@u HP0HCHH5'HHHHgH5ئ'HKYIHHmu HEHP0L;=[%L;=Z%u L;=Z%yI/u IWLR0nL]f. D$L%'H=\'IT$LW\HHkHHEH5̛'HHHIHEHMHEHu HEHP0D$UZIHHY%I9D$G LL|,IH I.u IFLP0I,$u ID$LP0L;=Y%L;=Y%u L;=Y%aI/u IGLP0 M} D$H$H5X%II}L9IH= I/uIGLP0DH+HCHP0I\$f.HHJ;H :;HOHBL :LOL@HHX%SHc?H5=.H81 ]H:mY'5 K'mH<'XZH :5 H=?E1HL$8dH3 %(LHH[]A\A]A^A_ÐHaX%H$L%a'H= 'IT$LZIHHIGH5'LHHIIHMIHu IGLP0L%'H='IT$LYHHIHHEH5 'HHHIHEHMHEHu HEHP0HgV%I9D$HL)IH}I,$u ID$LP0H,V%I9FLL)IH I/u IGLP0I.u IFLP0L;% W%L;%V%u L;%V%I,$u ID$LP0 Im H$HH5U%HEI}IfIH Hm'HEHP0LXŅxHI8E1E1ۡ' Hȡ'ơ'mIHIHu IGLP0Mf.MI/u IGLP0MtI.u IFLP0HtHmu HEHP0H H'N'H=<5='(E1HIH;HHPHXQH5'LIHVIWHD$ HMOHD$(Ld$ H$@H7E1' H''mH ''H=[;5'l MLL3H6IX' HE'C'm릐;YH[H6E1 ' H ' 'mHm6 HEHMP0NfDLVŅH16E1E1ß' H''nIHfLXVŅ+H5E1~' Hk'i'nI,$E1fDID$LP0%UHH5I' H ''mgfTHuLIHH=5IҞ' H''cnTHuL~HH{H4I' Ho'm'mHF HHD$(HFHD$ NHfHT$ HLL9H5C'@Hr4m'5 H''mHFHHD$ 'NIH #4E1' H ''enfDVIH59'LHVSH/HD$(IGfDH 3K' H 8'6'mHEHE E1!VIRHuLH# HP3' H՜'Ӝ'hnQfDH#3' H''m8H 2M' H }'{'jn@fDkUI1MT$MM\$III,$uID$L\$LLT$P0LT$L\$ICH;P%H;P% L\$LT$mQLT$L\$HHLP1LHLp L\$}L\$HIVHmMHEHP0H1{' Hh'f'mI.iI,$AfDMT$M6M\$III,$uID$L\$LLT$P0LT$L\$ICH;O%H;O%wL\$LT$MPLT$L\$HH'LP1LHHHX L\$YL\$HIHmMHEHP0H0[' HH'F'xnLMI/MVMaM^III.uIFL\$LLT$P0LT$L\$ICH;xN%MH;N%L\$LT$AOLT$L\$HHLP1LHLx L\$QL\$HIHmMHEHP0H5'H='1 HHHHmu HEHP0H/I' H'''na@HS/' Hؘ'IӘ'nH f.H/E1' H'I'LnHH5'H=:'13HHiHHmu HEHP0H.I>' H+')'nH.E1I' H''nIC@=LHt$ LT$ LT$L\$Lt$(L\$LT$HI"I*uIBL\$LP0L\$I.uIFL\$LP0L\$MIC@=tLHt$ LT$ LT$L\$H\$(_L\$LT$HII*uIBL\$LP0L\$MIC@= LHt$ LT$ LT$L\$L|$(L\$LT$HII*uIBL\$LP0L\$I/uIGL\$LP0L\$MLHt$ LT$ LT$L\$Lt$(L\$LT$HIH,MM' H''mLHt$ LT$ LT$L\$H\$([L\$LT$HIH ,MMLH'' '~nLHt$ LT$ LT$L\$L|$(L\$LT$HIH+MMF' H3'I.'nHfH+' H''#nHY+I' H۔'ٔ'n9I.MMH%+' H'' nI+E1@I*H*Hv'x' j'nMMLVH*IMC' H0'.'nH*MM' H'I'nH6HP*IM' Hϓ'͓'nH#*M' H''n!I*=E1MDFH)MMq' H^'\'nH)MMLH8':' ,'nH)MM' H'I'nH4IFLP0M1XMHuM"fAWAVAUIATIUSHHH^dH%(HD$81HF%HD$ HD$(H6H\HHF H$Md$H' H(hE111AHLHHKH;H@u HP0HCHH5'HHHHgH5'H[DIHHmu HEHP0L;=/F%L;=E%u L;=E%yI/u IWLR0nLHf.D$L%É'H=l'IT$LgGHHkHHEH5܆'HHHIHEHMHEHu HEHP0D$eEIHH"D%I9D$G LLIH I.u IFLP0I,$u ID$LP0L;=D%L;=D%u L;=D%aI/u IGLP0 M} D$H$H5C%II}LIIH= I/uIGLP0DH+HCHP0I\$f.HHZ&H J&HOH-L &LOL@HHC%SH*H5MH81HH%ki' ['kHL'XZH % H=,*E1'HL$8dH3 %(LHH[]A\A]A^A_ÐHqC%H$L%q'H='IT$LEIHHIGH5 'LHHIIHMIHu IGLP0L%'H='IT$LDHHIHHEH5'HHHIHEHMHEHu HEHP0HwA%I9D$HLIH}I,$u ID$LP0HH/HD$(IGfDH ['+ H H'F'&lHEHE E1!+AI >HuLH# H`'0 H''lQfDH3ˇ'+ H'')l8H M'0 H ''l@fD{@I1MT$MM\$III,$uID$L\$LLT$P0LT$L\$ICH;;%H;<% L\$LT$}'lHfH(', H''dlHiI'1 H''m9I.MMH5'+ H''MlI+E1@I*HH''0 z'lMMLVHIMS'+ H@'>'SlHMM&'0 H'I'lH6H`IM~'0 H~'~'lH3M~'0 H~'~'m!I*=E1MT1HMM~'+ Hn~'l~'FlHMMLHH~'J~'0 <~'lHMM$~'0 H~'I ~'lH4IFLP0M1XMHuM"fAWAVAUATUHSHHH^H|$dH%(HD$81H2%HD$ HD$(H$HZHHF HD$LmH}' H(hE111AHLHH9H;H@u HP0HCHH5s'HHIM]H5|'Li/IHI/u IGLP0L;5>1%@L;50%@u L;51%F@I.u IFLP0rL"3f.|ff/HD$HT$H50%Lp IHxLaHH I.IFLP0HHH/HGP0H]fDHHrH bHOH+L LOL@HH/%SHH5eH8134H]j{'Y s{']jHd{'XZH Y H=_1@HL$8dH3 %(H HH[]A\A]A^A_f.H/%HD$L-ys'H="{'IUL1IHrHIFH5x'LHHIIHMIHu IFLP0L5 s'H=z'IVL0IHHID$H5kt'LHHIMBI,$u ID$LP0IGE1H;-%gH;.%H;.%Hcu/IHi MtL`HcHI\H n'HcHIDIGLMQ LT$0/H--%LT$H Q;UP  1LLT$LAI.LT$H QP E=H9M I*I/u IGLP0Hy,%I9ELLIH8I.u IFLP0Imu IELP0L;=X-%L;=,%u L;= -%I/5HD$HT$HH5 -%Lx IHxMgHHI/HHL@L/HcH0x' Hx'x'jIHIHE1E1IFLP0MImE1E1IELT$LP0LT$MtI,$uID$LT$LP0LT$MtI/uIGLT$LP0LT$MI*IBLP0IH3HHXH'H5t'LIHVIN-HD$ HMGHD$(Ll$ HD$HK v' Hv'v'jfDH v'v'H=5v'HPH1fDH v' Hpv'nv'jfDH Yv'_v'H=\5Nv'9H173/I5IG@=HLj'HHLHt(Ld$ H\$(HD$0IHMI,$ID$LP0f.H u' Hu'u'jI/IGLP0DIGLP0IBLP0! @DIGLP0HH#L+ŅgHq  u' Ht't'Xka~*HH5-e'H=t'1HH H+Hmu HEHP0H t' Hzt'xt'j [*HuL>IHtH Et' H2t'0t'jcLT$)LT$@$HIh'HLLd$ Ht(H\$(HD$0jIHH9 s' Hs's' k@IEE1HIEfDHF HHD$(HFHD$ #HnHT$ HLL~H5'7HH Mj=s'Y H*s'(s'MjHFHHD$ _#IH [ r' H r'r'jf+IH5qh'LHV(H/HD$(IGfDHr' Hpr'nr'jIHNfK(HuL.HHE15r' H"r' r'jIEMHIEbE1%H[E1q' Hq'IEq'jHIE@*IhMgMIoI$HEI/u IGLP0HEIXMeMLImI$HEImu IELP0HEH;%%H;&%&HvL`1HHLp HD$豽LT$HII*IIBLP0@Hp' Hp'Ip'5kHIHfDHsp' H`p'^p'k1LLLT$(LT$HIHM&p' Hp'p'%kE1[H5U`'H=o'1诼HHH[Hmu HEHP0H%o' Ho'o'gk;H=LT$$LT$JHso' H`o'^o'|kLT$?%LT$H Hr"%H5KH8+"LT$HE@=Ht$ HLd$ Lt$( IHI,$u ID$LP0I.u IFLP0IHn' Hn'n'j+Hn' Hsn'qn'jHt$ HLd$ Lt$(IH`HMI1n' Hn'In'gj' j'gHj'XZH  H=1HL$8dH3 %(H HH[]A\A]A^A_f.H%HD$L-b'H=rj'IULn IHrHIFH5cg'LHHIIHMIHu IFLP0L5Zb'H=j'IVLIHHID$H5c'LHHIMBI,$u ID$LP0IGE1H;%gH;%H;K%HcIHi MtL`HcHI\HZ]'HcHIDIGLMQ LT$H-%LT$H Q;UP  1LLT$LAIJLT$H QP E=H9M I*I/u IGLP0H%I9ELL4IH8I.u IFLP0Imu IELP0L;=%L;=N%u L;=p%I/5HD$HT$HH5%Lx IHxMWHHI/HHL@LXHcHg' Hmg'kg'NgIHIHE1E1IFLP0MImE1E1IELT$LP0LT$MtI,$uID$LT$LP0LT$MtI/uIGLT$LP0LT$MI*IBLP0IH3HHXHH5b'LIHVIHD$ HMGHD$(Ll$ HD$H3f' H f'f';gfDH f'f'H=&5e'HPH1fDH;e' He'e'JgfDH e'e'H=5e'H17I5IG@=HY'HHLHt(Ld$ H\$(HD$0jIHMI,$ID$LP0f.H[d' Hd'd'LgI/IGLP0DIGLP0IBLP0! @DIGLP0HH#L0ŅgHYd' HFd'Dd'ha~HH5T'H=c'1ϰHH H{Hmu HEHP0HEc' Hc'c'qg HuLIHtHc' Hc'c'gcLT$LT$@$HW'HLLd$ Ht(H\$(HD$0躿IHH!c' Hc' c'g@IEE1HIEfDHF HHD$(HFHD$ HnHT$ HLLH5'臸HHgb' Hzb'xb'gHFHHD$ IH Cb' H 0b'.b'gfIH5W'LHV-H/HD$(IGfDH;a' Ha'a'gIHNfHuL~HHE1a' Hra'pa'gIEMHIEbE1%HE1@a' H-a'IE'a'gHIE@ IhMgMIoI$HEI/u IGLP0HEIXMeMLImI$HEImu IELP0HEH;%H;l%HvL`1HHLp HD$LT$HII*IIBLP0@Hk`' H_'I_'gHIHfDH+_' H_'_'g1LLLT$LT$HIHMv_' Hc_'a_'gE1[H5O'H=_'1HHH諺Hmu HEHP0Hu _' H^'^'!h;H=LT$LT$JH+^' H^'^'FhLT$LT$H H%H5H8{LT$HE@=Ht$ HLd$ Lt$(pIHI,$u ID$LP0I.u IFLP0IHe]' H]']'Yg+H>]' H]']'mgHt$ HLd$ Lt$( IH`HMI]' Hn]'Ii]'gHyHS]' H@]'>]'g1H,]' H]']'hZHmE1I\' H\'\' huH@MI\' H\'I\'hHlHMI\' H\'I\'gHIAWAVAUATUHSHHH^H|$dH%(HD$81H%HD$ HD$(H$HZHHF HD$LmH4\' H(hE111AHLHH9H;H@u HP0HCHH5Q'HHIM]H5['L IHI/u IGLP0L;5%@L;5%@u L;5%F@I.u IFLP0rLf.z|ff/HD$HT$H5]%Lp IHxLHH I.IFLP0HHH/HGP0H]fDHHH HOHL LOL@HHL%SHH5H81Hb!Z'Z'bHZ'XZH eH=71HL$8dH3 %(H HH[]A\A]A^A_f.H!%HD$L-R'H=Y'IULIHrHIFH5V'LHHIIHMIHu IFLP0L5Q'H=SY'IVLOIHHID$H5 S'LHHIMBI,$u ID$LP0IGE1H;" %gH;@ %H; %HcIHi MtL`HcHI\HL'HcHIDIGLMQ LT$ H-i %LT$H Q;UP  1LLT$LAI LT$H QP E=H9M I*I/u IGLP0H %I9ELLIH8I.u IFLP0Imu IELP0L;= %L;= %u L;= %I/5HD$HT$HH5 %Lx IHxMUFHHI/HHL@L HcH8V'HV'V'bIHIHE1E1IFLP0MImE1E1IELT$LP0LT$MtI,$uID$LT$LP0LT$MtI/uIGLT$LP0LT$MI*IBLP0IH3HHXH(H5Q'LIHVI HD$ HMGHD$(Ll$ HD$HU'HpU'nU'bfDH YU'_U'H=5NU'9HPH1fDH#U'HU'U'bfDH T'T'H=45T'H17 I5IG@=HH'HHLHt(Ld$ H\$(HD$0躙IHMI,$ID$LP0f.HCT'H0T'.T'bI/IGLP0DIGLP0IBLP0! @DIGLP0HH#L ŅgHS'HS'S'ca~m HH5%D'H=&S'1HH HˮHmu HEHP0H-S'HS'S'c HuL޵IHtHMR'HR'R'DccLT$NLT$@$HF'HLLd$ Ht(H\$(HD$0 IHHqR'H^R'\R']c@IEE1HIEfDHF HHD$(HFHD$ nHnHT$ HLLVH5&קHHJbQ'HQ'Q'bHFHHD$ IH Q'H Q'~Q'Fcfk IH5G'LHV}H/HD$(IGfDH#Q'HQ'Q'-cIHNfHuLγHH@E1P'HP'P'IcIEMHIEbE1%HE1P'H}P'IEwP'KcHIE@[ IhMgMIoI$HEI/u IGLP0HEIXMeMLImI$HEImu IELP0HEH;a%H;%4HvL`1HHLp HD$QLT$HII*IIBLP0@HSO'H@O'I;O'cHIHfDH{O'HO'N'kc1LLLT$'LT$HIH1MN'HN'N'vcE1[H5M?'H=VN'1OHHHHmu HEHP0H]N'HJN'HN'c;H=4LT$BLT$JH{N'HN'M'cLT$LT$H H%H5H8LT$HE@=Ht$ HLd$ Lt$(IHI,$u ID$LP0I.u IFLP0IHMM'H:M'8M'b+H&M'HM'M'cHt$ HLd$ Lt$(pIH`H?MIL'HL'IL'cHyH L'HL'L'ec1H|L'HiL'gL'cZHE1IOL'H'cHX>'V>'rYLMI/MVMaM^III.uIFL\$LLT$P0LT$L\$ICH;$MH;$L\$LT$QLT$L\$HHLP1LHLx L\$aL\$HIHmMHEHP0H5.'H="='1HHHǘHmu HEHP0HI&='_H='='!Ya@Hc<'cH<'I<'YH f.H+E1<'`H<'I<'FYHH5-'H=J<'1CHHiHHmu HEHP0HIN<'dH;<'9<'YHE1I!<'eH<' <'YIC@=LHt$ LT$ LT$L\$Lt$(L\$LT$HI"I*uIBL\$LP0L\$I.uIFL\$LP0L\$MIC@=tLHt$ LT$ LT$L\$H\$(oL\$LT$HII*uIBL\$LP0L\$MIC@= LHt$ LT$ LT$L\$L|$(L\$LT$HII*uIBL\$LP0L\$I/uIGL\$LP0L\$MLHt$ LT$ LT$L\$Lt$(ԖL\$LT$HIHMM+:'^H:':'XLHt$ LT$ LT$L\$H\$(kL\$LT$HIH0MMLH9'9'c9'xYLHt$ LT$ LT$L\$L|$(L\$LT$HIHMMV9'cHC9'I>9'YHfH(9'_H9'9'YHiI8'dH8'8'Y9I.MMH58'^H8'8'YI+E1@I*HH8'8'cz8'YMMLVHIMS8'^H@8'>8' YHMM&8'cH8'I8'YH6H`IM7'cH7'7'YH3M7'cH7'7'Y!I*=E1MTHMM7'^Hn7'l7'XHMMLHH7'J7'c<7'YHMM$7'cH7'I 7'YH4IFLP0M1XMHuM"fAWAVAUIATUSHHH-*'H^dH%(HD$81H$Hl$ HD$(HfHtHRHHHIH HHIHH?$I?SIH5H8L A1HvVV 6'5'VVH5'XZH MH=1ȢHL$8dH3 %(HiHH[]A\A]A^A_fH$H$H5' H(hE111AHHHHH;H@u HP0HCHH5Q+'HHIMH585'LIHI,$u ID$LP0L;=~$L;=$$u L;=F$DI/u IWLR0EHcf.D$wH--'H=4'HUHIHHID$H5)*'LHHHI$HHI$Hu ID$LP0D$IHMHn$H9ELHٺIH] I.u IFLP0Hmu HEHP0L;=M$L;=$L;=${L[Ņ1I/u IGLP0 M} D$H$H5t$II}LeHH I/uIGLP0H+NHCHP0?H-+'H=*3'HUH&IHHIGH50'LHHHIIHM3IHu IGLP0H-+'H=2'HUHIHkHID$H5+('LHHHI$HHI$Hu ID$LP0H$H9EHHIHHmu HEHP0HN$I9FLL蹸HH I/u IGLP0I.u IFLP0H;-.$H;-$u H;-$DHmu HEHP0E Me H$HH5h$I$I}M HH I,$ID$LP0HF H$HnjHi$H$LAąH@E110'H0'0'VIHIHu IGLP0IfMI/Au IGLP0MtI.u IFLP0MtI,$u ID$LP0H F0'L0'H=!1590'$IHH)HHHHH@HD$(Hl$ H$0DH+1/'H/'/'yVH /'/'H=x5/'}HH1Jf.HH`/'HM/'K/'Vf;IH1!/'H/' /'VI,$N ID$LIP0NfDHAąQH0E1.'H.'.'WHmAE1@HEHP0f{IkHHE11N.'H;.'9.'VIHvD HuHoH.'H-'-'V>f.HF HHD$(HFHD$ HHT$ HLL#H5&HEV-'Hr-'p-'EVyHFHHD$ HHcH5"'LHVhHrHD$(HE[ HuHIHH]H,'H,','!W,HuH螏IH=H H,'H,','VH 1q,'H ^,'\,'#WH C,'H 0,'.,'VI$H~ E1fHuHގH] HP+'H+'+'&WRfDH#+'H+'+'VH L+'H }+'{+'(WHfDkHLUMOL]IIHmuHEL\$HLT$P0LT$L\$ICH;$H;$>L\$LT$pLT$L\$HI7LP1LHLp L\$wL\$HIPI,$LID$LP0H*'Hp*'n*'VI.HmAfDLUML]IIHmuHEL\$HLT$P0LT$L\$ICH;$)H;$L\$LT$XLT$L\$HILP1LHHHX L\$dvL\$HIaI,$LID$LP0oDH˿c)'HP)'N)'6WMII/fMVM?M^III.uIFL\$LLT$P0LT$L\$ICH;x$~H;$L\$LT$ALT$L\$HILP1LHLx L\$QuL\$HHI,$MID$LP0fH5'H=('1 uHHKH跃Hmu HEHP0HH('H('('VP@HS''H''I''aWHf.HE1''H''I'' WHH5'H=:''13tHHH߂Hmu HEHP0HH>''H+'')''Wx@H{E1L ''H&'&'WCH51'LHV HHD$ HrIC@=LHt$ LT$ LT$L\$Lt$(kL\$LT$HI"I*uIBL\$LP0L\$I.uIFL\$LP0L\$L\IC@=8LHt$ LT$ LT$L\$H\$(.kL\$LT$HII*uIBL\$LP0L\$LIC@=LHt$ LT$ LT$L\$L|$(jL\$LT$HHI*uIBL\$LP0L\$I/uIGL\$LP0L\$M|LHt$ LT$ LT$L\$Lt$(蓁L\$LT$HIHXML$'H$'$'VbLHt$ LT$ LT$L\$H\$(*L\$LT$HIHLMMHu$'w$'i$'H}HE1H2fD@$f@$fHk'H''~HEHImtM1)f.1fHEHP0ImH۹IH+' H''yI$E1E1E1HH' Hp'n'yI,$ID$LP0DXHFHHD$0IH{' H'&y@Lct$H,&LL\$Ll$0IH\$8Jt8HD$@H\L\$HIHE1&H&I$&yHDReH5 H|$f.eD H5-&H=&1KIH^ HZI,$u ID$LP0Ht & H&&Iyf۴HuLaIHH-&H&&y]IC@=Lcl$H&HLL\$ILt$0Jt8Hl$8HD$@CL\$HI( MI.IFL\$LP0L\$T$T$H ff/H5&H=&1JIH# HIYI,$u ID$LP0H& H&&iyBH E1&H m&k&yIHE1E1@KIL+HuL`H H&H&&yPfDHS& H&I&yH_f.軵I+H ME1&H&I$E1&yHMkMMsIEII+u ICLP0IFMD$M\$MIL$IHI,$uID$L\$LHL$P0HL$L\$HAH;q$TH;̰$ L\$HL$:HL$L\$HIuLX1HHLp HL$JHHL$HI~I/IIGLP0HE1E1J&H7&I$1&yHXHE1&H&I$&yH&Lcl$H*&LL\$Lt$0IHl$8Jt8HD$@FWL\$HIfH&H&&DzMDHېE1p&H]&I$W&yH~HT$0HLLѕH5&PHx& H&&xHH5&H=&1FIHHSUI,$u ID$LP0H&H&&zL腯HuLh\IH{H׏o&H\&Z&+zH$H5>H8?&H E1E1E1H &&&-z:I H X&H &&2zkѱIXHfDHH9HuH;D$efMsMWMkIIEI+u ICLP0IEMD$!Hf.HH9HuH;Ĭ$fHH$AH5jL QH sH8Hs1wH2_xH&AX& &xM_M7IOIHI/uIGHL$LL\$P0L\$HL$HAH;$H;K$6HL$L\$蹬L\$HL$HI[LX1HHL` HL$CHL$HIImIIELP0H6E1E1&H&I$&mzHH&H&&zHڌr&H_&]&RzImHH=&?&1&]zpRH5&H=&1BIHH}QI,$u ID$LP0HF&H&&zvHA@=HHt$0L\$0L\$HL$Lt$8:HL$L\$HII+uICHL$LP0HL$I.uIFHL$LP0HL$IH*& H&&'yHHW& H&&1yH0& H&&Ey`H & H&&ey9HHt$0L\$0L\$HL$Lt$8PHL$L\$HIHE1I8&H%&H &yHGHA@=HHt$0L\$0L\$HL$Ld$8 9HL$L\$HI3I+uICHL$LP0HL$I,$uID$HL$LP0HL$IMHE1x&He&I$_&yHHI&H6&4&zHE1I&H &H&yH+HVE1E1IH&H&&zHHHt$0L\$0L\$HL$Ld$8OHL$L\$HIHIE1u&Hb&I$\&tzHHA&H.&,&zH&H&&LzmImHVH&&&zE1IH#IE1&H&I$&zHHIE1&Hm&I$g&}zHHE1IK&H8&H3&yHZIwIxf.AWAVAUATUSHXL57&L=(&H|$H^dH%(HD$H1H8$Lt$0L|$8HD$@H Hv~,HHHF(H$L~ LvfDH$H$H\H& H(hE111AHLHH H;$ HV& H(hE111AHLHH[ H;L$ HCH5&HHH` IM" HEH5&LT$HHH? LT$HHD HL׺HL$LT$LT$HL$HIH;ͣ$L;%s$AA L;%$@AtgI,$H5&HϺLT$HL$~HL$LT$HI H;Y$L;%$AA L;%$A I*uIBHL$L׉T$P0HL$T$H)uHAT$HP0T$E}I,$ L-&H=t&IULpIHD HID$H5d&LHH II$HM I$Hu ID$LP0L-X&H=&IULHH HHAH5r&HHL$HH HL$IHHM@ HHu HAHP0L-Ǡ$M9l$lHL1tHH%I,$uID$HL$LP0HL$M9n7HLHL$sHL$HI>H) I.uIFLT$LP0LT$L;[$L;$u L;#$e DI*E=HD$H$HIH5؟$LP IHxMLT$/LT$HH_I*H+HHmHEHP0fDH$H$HHIH HHIHH$I?SIH5uH8L /A1bH%u& &%uH&XZH  H= 1oXHT$HdH3%(H]HX[]A\A]A^A_fLf.QD$ Lܡf.QD$L-&H=2&IUL.HHJ HHAH5&HHL$HH HL$IHHM HHu HAHP0D$$HH H$I9D$~ HLHL$FqHL$HI5 H)u HAHP0I,$u ID$LP0L;=$L;=Z$u L;=|$^DI/u IGLP0E HD$L$D$ H$L` H5!$I$HxL0HH$ I,$UH$ID$LP0H $=ID$T$LP0T$ ID$HL$LLT$P0HL$LT$@@H;&U H(&&&huI*uIBH $LP0H $HE1HHHu HAHP0MMoI/uIGH $LP0H $MtI.uIFH $LP0H $HtH)u HAHP0H &&H=5v&aUHtH+u&1HCH $HP0H $HH1@H$H$LHL$ LT$T$ʞHL$ LT$T$L訞…qH9~E1&U H&&ouI,$>E1ID$LP0HH{HtHeHF(HD$@HF HHD$8HFHD$0袗IHHkM~.H5&HHVSHB HD$@IM0 HD$@Lt$0L|$8H$HH/IM~H5`&HHVHtHD$0IM~H5&HHVΜH_HD$8ILH |R Jui&R H=H O&M&Ju8SH&H HxH9HXH HqH~&H;Q1@H;THH9uHݙ$HJH5jHWH81dH|I11H&1&R &Lu~@H{&S Hp&n&WuH _&e&H=5T&?Rf.H&H HxH9EHXH HqH~&H;Q"1@H;THH9uHݘ$HJH5iHWH81dH{I1&S H&1&YufDHz&U Hp&n&duf[IKLT$HfDHz+&U H&&fuI* IBH $LE1P0H $IHD$HAHP0LT$IBLP0fLhAąHyME1&X Hw&u&uH$IBLP0H $fLLT$LT$AHy1$&] H&&qvHHH9HuH;$fHfHy&U H&&kusfVuHHHxo&V H\&Z&zuD25H$Hx/&W H&&uDHFHHD$0OIAۗHuLDIHH-x&] H&&v=蓗HuLvDHHHw}&X Hj&h&uHwS&] H@&>&vI$HE1E1{fDI HL$I=fDHSwE1&X H&&ufDHD$讖HL$HuLHL$CHL$H0 Hv&] Hy&w&vfHvM`&X HM&K&u0fDHv3&] H &&vf HL$InfDMl$MtMT$IEII,$uID$HL$(LLT$P0LT$HL$(IBH;;$H;$xHL$(LT$LT$HL$(HILh1LHHH LT$,LT$HImI.MIFLP0H{uME1 &X H&&uML$MM|$III,$uID$LL$LP0LL$IGH;>$|H;$LL$ LL$HtLH1HLHEHh HD$+LT$HHI*MHD$IBLP0HL$H{tM&] H&&#vI/4@M~MMnIIEI.uIFHL$LP0HL$IEH;9$H;$wHL$HL$HILx1HLHH !*IH$I,$MSHD$ID$LP0LT$9DH5!&H=&1)IHH8I,$u ID$LP0HHs&Y H&&uXfDHsI&] H&&NvfDHr&Z Hp&I$j&uH#H5R&H=&1 )IHH7I,$u ID$LP0Hr&^ H&&vHZr&_ H&&vHT$0HHLswH5߁&1H ru& H&&uH$H5oqH8pH$H5TqH8UHDHH9ZHuH;̏$H=IB@=sLHt$0HL$8HL$(LT$Ll$0 LT$HL$(HIImuIEHL$(LLT$P0HL$(LT$H)uHALT$HP0LT$MCIG@=Ht$0LLL$0LL$Hl$8LL$HHlI)uIAHL$LP0HL$MIE@=tHt$0LHL$8HL$L|$0HL$HI7I/uIGHL$LLT$P0HL$LT$H)uHALT$HP0LT$MLHt$0HL$8HL$(LT$Ll$05LT$HL$(HIHoMF&X H3&1&uHt$0LLL$0LL$Hl$85LL$HHHUoML&] H&I&)vHHt$0LHL$8HL$L|$0%5HL$HIHn&] Ht&r&UvH)MIDHnK&Y H8&6&uHn$&^ H&&|vHenML&] H&I&7vHH1nL&X H&&uHnM&X H&&u\HmLr&] H_&]&=v]HmK&] H8&6&evHmLM&] H & &kvG迉HZmM&X H&&uH0mML&] H&I&1vHHl&] H&&^vI11HAWAVAUATUSHXL5&L=x&H|$H^dH%(HD$H1H$Lt$0L|$8HD$@H Hv~,HHHF(H$L~ LvfDH1$H$H\H& H(hE111AHLHH H;$ H& H(hE111AHLHH[ H;$ HCH5+&HHH` IM" HEH5&LT$HHH? LT$HHD HL׺HL$LT$BLT$HL$HIH;$L;%È$AA L;%$@AtgI,$H5e&HϺLT$HL$ΆHL$LT$HI H;$L;%O$AA L;%n$A I*uIBHL$L׉T$P0HL$T$H)uHAT$HP0T$E}I,$ L-&H=&IULIHD HID$H5&LHH II$HM I$Hu ID$LP0L-&H=Q&IULMHH HHAH5&HHL$HH HL$IHHM@ HHu HAHP0L-$M9l$lHLYHH%I,$uID$HL$LP0HL$M9n7HLHL$?YHL$HI>H) I.uIFLT$LP0LT$L;$L;Q$u L;s$e DI*E=HD$H$HIH50$LP IHxMLT$LT$HH_I*H+HHmHEHP0fDHх$H$HHgIH gHmHIHH6$I?SIH5ZH8L nA1貉Hmg6s& &6sH&XZH Dg H=l1=HT$HdH3%(H]HX[]A\A]A^A_fLHf.7D$ L,f.6D$L-&H=&IUL~HHJ HHAH5&HHL$HH HL$IHHM HHu HAHP0D$tHH H1$I9D$~ HLHL$VHL$HI5 H)u HAHP0I,$u ID$LP0L;=$L;=$u L;=̃$^DI/u IGLP0E HD$L$D$ H$L` H5y$I$HxL耭HH$ I,$UH$ID$LP0H $=ID$T$LP0T$ ID$HL$LLT$P0HL$LT$@@Hd& Hx&v&ysI*uIBH $LP0H $HE1HHHu HAHP0MMoI/uIGH $LP0H $MtI.uIFH $LP0H $HtH)u HAHP0H &&H=i5&:HtH+u&1HCH $HP0H $HH1@H$H$LHL$ LT$T$HL$ LT$T$L…qHcE1& H & &sI,$>E1ID$LP0HH{HtHeHF(HD$@HF HHD$8HFHD$0|IHHkM~.H57&HHV裂HB HD$@IM0 HD$@Lt$0L|$8H$HH|IM~H5 &HHVDHtHD$0IM~H5B&HHVH_HD$8ILH +b [s& H=egH &&[s8Ha&H HxH9HXH HqH~&H;Q1@H;THH9uH-$HJH5jOHWH81贃HoaI11H&1& &]s~@H;a& H&&hsH &&H=af5&7f.Ha&H HxH9EHXH HqH~&H;Q"1@H;THH9uH-~$HJH5jNHWH81贂Ho`I1& H&1&jsfDH;`& H&&usf諂I蛂LT$HfDH_{& Hh&f&wsI* IBH $LE1P0H $IHD$HAHP0LT$IBLP0fLAąHH_ME1& H&&sH$IBLP0H $fLLT$SLT$AH^1t& Ha&_&tHHH9HuH;|$fHfHk^& H&&|ssfV}HHH'^& H&&sD2}H$H]& Hl&j&sDHFHHD$0wIA+}HuL*IHH}]& H&&t=|HuL)HHH5]& H&&sH ]& H&&!tI$HE1E1{fDkI[HL$I=fDH\E18& H%&#&sfDHD${HL$HuLHL$(HL$H0 HD\& H&&$tfH\M& H&&s0fDH[& Hp&n&&tf[~HL$InfDMl$MtMT$IEII,$uID$HL$(LLT$P0LT$HL$(IBH;y$H;y$xHL$(LT$TzLT$HL$(HILh1LHHH LT$dLT$HImI.MIFLP0HZME1]& HJ&H&sML$MM|$III,$uID$LL$LP0LL$IGH;x$|H;x$LL$\yLL$HtLH1HLHEHh HD$oLT$HHI*MHD$IBLP0HL$HYM`& HM&K&4tI/4@M~MMnIIEI.uIFHL$LP0HL$IEH;w$H;w$wHL$WxHL$HILx1HLHH qIH$I,$MSHD$ID$LP0LT$9DH5&H=*&1#IHHI,$u ID$LP0HX0& H&&sXfDHkXI& H&&_tfDH;X& H&I$&tH#H5&H=c&1\IHHI,$u ID$LP0HWi& HV&T&tHWB& H/&-&tHT$0HHL\H5g&H]W"s& H&&"sHt$H5VH8sHs$H5VH8sHDHH9ZHuH;u$H=IB@=sLHt$0HL$8HL$(LT$Ll$0eLT$HL$(HIImuIEHL$(LLT$P0HL$(LT$H)uHALT$HP0LT$MCIG@=Ht$0LLL$0LL$Hl$8LL$HHlI)uIAHL$LP0HL$MIE@=tHt$0LHL$8HL$L|$0iHL$HI7I/uIGHL$LLT$P0HL$LT$H)uHALT$HP0LT$MLHt$0HL$8HL$(LT$Ll$0<LT$HL$(HIHUM& H&&sHt$0LLL$0LL$Hl$8LL$HHHTML7& H$&I&:tHHt$0LHL$8HL$L|$0uHL$HIH?T׽& HĽ&½&ftH)MIDHT& H&&sHSt& Ha&_&tHSMLG& H4&I/&HtHHSL& H&&sHWSM& Hټ&׼&s\H-SL¼& H&&Nt]HS& H&&vtHRLMn& H[&Y&|tGoHRM?& H,&*&sHRML& H&I&BtHHLR& Hѻ&ϻ&otI11HAWAVAUATUSHXL5ׯ&L=ȯ&H|$H^dH%(HD$H1Ho$Lt$0L|$8HD$@H Hv~,HHHF(H$L~ LvfDHo$H$H\H=& H(hE111AHLHH H;3o$ H& H(hE111AHLHH[ H;n$ HCH5{&HHH` IM" HEH5O&LT$HHH? LT$HHD HL׺HL$LT$lLT$HL$HIH;mn$L;%n$AA L;%2n$@AtgI,$H5&HϺLT$HL$lHL$LT$HI H;m$L;%m$AA L;%m$A I*uIBHL$L׉T$P0HL$T$H)uHAT$HP0T$E}I,$ L-k&H=&IULoIHD HID$H5&LHH II$HM I$Hu ID$LP0L-&H=&IULnHH HHAH5&HHL$HH HL$IHHM@ HHu HAHP0L-gk$M9l$lHL>HH%I,$uID$HL$LP0HL$M9n7HLHL$>HL$HI>H) I.uIFLT$LP0LT$L;k$L;k$u L;k$e DI*E=HD$H$HIH5@k$LP IHxMLT$ϜLT$HH_I*H+HHmHEHP0fDH!k$H$HH:MIH 'MH{RHIHHj$I?SIH5?@H8L SA1oHLGqP& B&GqH3&XZH L H=Q1#HT$HdH3%(H]HX[]A\A]A^A_fLlf.PD$ L|lf.4D$L-)&H=ҵ&IULkHHJ HHAH5C&HHL$HH HL$IHHM HHu HAHP0D$iHH Hh$I9D$~ HLHL$;HL$HI5 H)u HAHP0I,$u ID$LP0L;=Ti$L;=h$u L;=i$^DI/u IGLP0E HD$L$D$ H$L` H5h$I$HxLВHH$ I,$UH$ID$LP0H $=ID$T$LP0T$ ID$HL$LLT$P0HL$LT$@@HCJ۳&| Hȳ&Ƴ&qI*uIBH $LP0H $HE1HHHu HAHP0MMoI/uIGH $LP0H $MtI.uIFH $LP0H $HtH)u HAHP0H !&'&H=N5& HtH+u&1HCH $HP0H $HH1@H9g$H$LHL$ LT$T$jiHL$ LT$T$LHi…qHHE1n&| H[&Y&qI,$>E1ID$LP0HH{HtHeHF(HD$@HF HHD$8HFHD$0BbIHHkM~.H5&HHVgHB HD$@IM0 HD$@Lt$0L|$8H$HHaIM~H5p&HHVgHtHD$0IM~H5&HHVngH_HD$8ILH {Gy lq &y H=LH &&lqH&H HxH9HXH HqH~&H;Q1@H;THH9uH}d$HJH54HWH81iHFI11HG&1G&y 9&nq~@HF#&z H&&yqH &&H=K5&f.H&H HxH9EHXH HqH~&H;Q"1@H;THH9uH}c$HJH53HWH81hHEI1R&z H?&1;&{qfDHE#&| H&&qfgIgLT$HfDH3Eˮ&| H&&qI* IBH $LE1P0H $IHD$HAHP0LT$IBLP0fLeAąHDME1*& H&&qH$IBLP0H $fLLT$dLT$AH.D1ĭ& H&&rHHH9HuH;$b$fHfHCS&| H@&>&qsfVcHHHwC&} H&&qD2bH$H7CϬ&~ H&&qDHFHHD$0\IA{bHuL^IHHBe& HR&P&0r=3bHuLHHHB& H &&qH[B& H&ޫ&2rI$HE1E1{fDdIdHL$I=fDHAE1& Hu&s&qfDHD$NaHL$HuLHL$'HL$H0 HA,& H&&5rfHkAM& H&&q0fDH;AӪ& H&&7rfcHL$InfDMl$MtMT$IEII,$uID$HL$(LLT$P0LT$HL$(IBH;^$H;6_$xHL$(LT$_LT$HL$(HILh1LHHH LT$LT$HImI.MIFLP0H@ME1& H&&qML$MM|$III,$uID$LL$LP0LL$IGH;]$|H;9^$LL$^LL$HtLH1HLHEHh HD$LT$HHI*MHD$IBLP0HL$H?M& H&&ErI/4@M~MMnIIEI.uIFHL$LP0HL$IEH;\$H;4]$wHL$]HL$HILx1HLHH IH$I,$MSHD$ID$LP0LT$9DH5&H=z&1sIHHI,$u ID$LP0H=& Hm&k&qXfDH=IP& H=&;&prfDH=#& H&I$ &rH#H5&H=&1IHHXI,$u ID$LP0H!=& H&&rH<& H&}&rHT$0HHLLBH5?L&:H<3q@& H-&+&3qHWY$H5<H8YHH) I.uIFLT$LP0LT$L;KQ$L;P$u L;Q$e DI*E=HD$H$HIH5P$LP IHxMLT$LT$HH_I*H+HHmHEHP0fDHqP$H$HH2IH w2H7HIHHO$I?SIH5%H8L 9A1RTH 2Xo& &XoH&XZH 1 H=c71_HT$HdH3%(H]HX[]A\A]A^A_fLQf.D$ LQf.D$L-y&H="&IULQHHJ HHAH5&HHL$HH HL$IHHM HHu HAHP0D$OHH HM$I9D$~ HLHL$6!HL$HI5 H)u HAHP0I,$u ID$LP0L;=N$L;=JN$u L;=lN$^DI/u IGLP0E HD$L$D$ H$L` H5N$I$HxL xHH$ I,$UH$ID$LP0H $=ID$T$LP0T$ ID$HL$LLT$P0HL$LT$@@H/+& H&&oI*uIBH $LP0H $HE1HHHu HAHP0MMoI/uIGH $LP0H $MtI.uIFH $LP0H $HtH)u HAHP0H q&w&H=Y45f&QHtH+u&1HCH $HP0H $HH1@HL$H$LHL$ LT$T$NHL$ LT$T$LN…qH).E1& H&&oI,$>E1ID$LP0HH{HtHeHF(HD$@HF HHD$8HFHD$0GIHHkM~.H5׌&HHVCMHB HD$@IM0 HD$@Lt$0L|$8H$HHGIM~H5&HHVLHtHD$0IM~H5&HHVLH_HD$8ILH , }oY& H=;2H ?&=&}o(H&H HxH9HXH HqH~&H;Q1@H;THH9uHI$HJH5 HWH81TNH,I11H&1& &o~@H+s& H`&^&oH O&U&H=715D&/f.H&H HxH9EHXH HqH~&H;Q"1@H;THH9uHH$HJH5 HWH81TMH+I1& H&1&ofDH*s& H`&^&ofKMI;MLT$HfDH*& H&&oI* IBH $LE1P0H $IHD$HAHP0LT$IBLP0fLXJAąH)ME1z& Hg&e&oH$IBLP0H $fLLT$ILT$AH~)1& H&&pHHH9HuH;tG$fHfH )& H&&osfVeHHHH(_& HL&J&oD2%HH$H(& H & &oDHFHHD$0?BIAGHuLIHH(& H&&Ap=GHuLfHHH'm& HZ&X&oH'C& H0&.&CpI$HE1E1{fD JIIHL$I=fDHC'E1ؐ& HŐ&Ð&ofDHD$FHL$HuLHL$wHL$H0 H&|& Hi&g&FpfH&MP& H=&;&o0fDH&#& H&&HpfHHL$InfDMl$MtMT$IEII,$uID$HL$(LLT$P0LT$HL$(IBH;+D$H;D$xHL$(LT$DLT$HL$(HILh1LHHH LT$LT$HImI.MIFLP0Hk%ME1& H&&oML$MM|$III,$uID$LL$LP0LL$IGH;.C$|H;C$LL$CLL$HtLH1HLHEHh HD$LT$HHI*MHD$IBLP0HL$Hk$M& H&&VpI/4@M~MMnIIEI.uIFHL$LP0HL$IEH;)B$H;B$wHL$BHL$HILx1HLHH IH$I,$MSHD$ID$LP0LT$9DH5A}&H=ʌ&1IHHoI,$u ID$LP0H8#Ќ& H&&pXfDH #I& H&&pfDH"s& H`&I$Z&*pH#H5r|&H=&1IHHI,$u ID$LP0Hq" & H&&pHJ"& Hϋ&͋&pHT$0HHL'H5o1&H!Do& H}&{&DoH>$H5_!H8`>H>$H5D!H8E>HDHH9ZHuH;?$H=IB@=sLHt$0HL$8HL$(LT$Ll$0LT$HL$(HIImuIEHL$(LLT$P0HL$(LT$H)uHALT$HP0LT$MCIG@=Ht$0LLL$0LL$Hl$8rLL$HHlI)uIAHL$LP0HL$MIE@=tHt$0LHL$8HL$L|$0 HL$HI7I/uIGHL$LLT$P0HL$LT$H)uHALT$HP0LT$MLHt$0HL$8HL$(LT$Ll$0LT$HL$(HIHM6& H#&!&oHt$0LLL$0LL$Hl$8{LL$HHHEML׈& HĈ&I&\pHHt$0LHL$8HL$L|$0HL$HIHw& Hd&b&pH)MIDH;& H(&&&pH|& H&&pHUML& Hԇ&Iχ&jpHH!L& H&&oHM& Hy&w&o\HLb& HO&M&pp]H;& H(&&&pH|LM& H&&pG9HJM߆& H̆&ʆ&oH ML& H&I&dpHH& Hq&o&pI11HAWAVAUATUHSHXH^H|$dH%(HD$H1H:$HD$0HD$8HD$@H;H)HgHF(HD$(HE HD$HEHD$H& H(hE111AHH|$HHH8u H@HP0H& H(hE111AHH|$HHH8u H@HP0HEH51{&HHHfIM(HCH5{&HHH IM LL\7IH H;A9$L;=8$AA L;=9$L$$At\I/u IGLP0H5&L6IH( H;8$L;=8$AA ĉT$$L;=8$A I.JImgEqI/LL$$VL%i|&H=&IT$L :IH HIGH5&LHHo IIHM" IHu IGLP0L%{&H=&IT$L9IHa HIEH5b}&LHH IM ImuIELL$LP0LL$IAL=k6$1L9 H;7$H;7$HcNLL$HL$O8HL$LL$HIn HtHHHcD$$HLLLL$I\HHv&Dj1HMcKD6LL$HID I,$I)M9~ LL HH ImDI.H; 6$H; A6$u H; c6$eDH)'E HD$HT$(IHH55$Lh IEHxMrgIHb ImIELP0H]HH H HMHSHsL@HM5$H5 L H819Hh#& &hH&XZH g H=E1HT$HdH3%(LHX[]A\A]A^A_H)5$HD$(IELP0ELU7D$${I/HHo&q&K c&hEE1E11IGLL$LHL$P0HL$LL$HtH)uHALL$HP0LL$MtI)u IALP0MI,$ID$LP0IFLP0ImfDIGLP0L$$H|$H6f.f(H|$T$&6f.T$f(ff/$HD$HT$(f(H5c3$Lx IHxL]IHI/u IGLP0HmuHEHP0HH+HCHP0L57IHc}HtHHF(HD$@HE LHD$8HEHD$0.IHtzHHtFMHD$0HD$HD$8HD$HD$@HD$(jfDHHo.IH5]w&LIHV54HD$0HkH5`x&LHV4HD$8HIMpH5zs&LHV3HHD$@I=H1E1}&H Hs}&q}&hH b}&h}&H=e5W}&BMOLK@H3}&I H }&}&hfDH }&}&H= 5|&HmAH;I|&K H|&|&hEfD5IIALP0AIA@=Lcl$$Hp&HLHL$0IHL$Jt8LL$H\$8HD$@sLL$HL$HIHH)HALL$HP0LL$IFHL$LP0HL$HD$IELP0HL$HAHP0ID$LL$LP0LL$f/HD$HT$(f(H5$Lx IHxLWGIHkI/u IGLP0HmuHEHP0HH+HCHP0LHIH~yHtH)HF(HD$@HE LHD$8HEHD$0IHtvHHtBM HD$8Lt$0HD$HD$@HD$(HH7IH5c&LIHVHD$0H|H5p`&LHVHD$8H{ IMtH5B]&LHVH HD$@IAH 8 kdE1H=EH 6g&8g&8 *g&kdwH{g&9 Hg&f&zdfDH f&f&H=A5f&HmHf&; Hf&f&d@IcICLP0 fIC@=jHZ&HLL\$Ll$0HD$@HcD$Hl$8HHt8XL\$HIMImIEL\$LP0L\$sf.ID$L\$LP0L\$fDHD$IGLP0L\$mICLP0IFL\$LP0L\$LL\$#L\$D$RHMBe&F H/e&-e&eDIm%IELP0fDID$LP0zHuLIHH5d&H Hd&d&eIHE1E1d&; Hzd&xd&dI$HI$HiE1ID$L\$LP0ML\$HI/>%H#d&; Hd&d&dI,$ID$LP0ImD(HFHHD$0IH c&; Hc&c&d{fHW&LL\$Ll$0HD$@HcD$Hl$8HHt8пL\$HI(HE1/c&F Hc&I$c&6eHDH: H|$f.R H5}S&H=b&1臯IH H3I,$u ID$LP0Hb&@ Hb&b&df.[HuL>IHuHEb&F H2b&0b&e3IC@=DLcd$HGV&HLL\$IL|$0Jt8H\$8HD$@L\$HI MI/wIGL\$LP0L\$^rT$T$Hff/H5 R&H=$a&1IHHɼI,$u ID$LP0H*a&B Ha&a&d IIH [E1`&F H `&`&eIH E1E1@HuLHS HE1E1`&F H`&}`&"eI$HHc`&C HP`&IK`&eHgfHE10`&F H`&I$`&$eHfDIMkMMsIEII+u ICLP0IFMD$Ml$MLIL$IEHI,$uID$HL$LP0HL$HAH;$H;]$HL$HL$HI!Lh1HHLx HL$HL$HI*I.IHD$IFLP0L\$HI^&F H^&I$^&_eHI$E1E1HQj@H^&F H^&I$^&DeH Lcd$HR&LL\$L|$0IH\$8Jt8HD$@ɺL\$HIiH+^&H H^&^&eM.H[E1]&F H]&I$]&OeH^H5_N&H=]&1yIHhH%I,$u ID$LP0H]&G Hs]&q]&etHT$0HLL~H53&.)H3d4]&H!]&]&3dVHuLIHeHH8W&G H%W&#W&e&HyMI W&F HV&HV&veH&HEE1IV&F HV&HV&|eHFq HHt$0L\$0L\$HL$Ld$8 HL$L\$HIHIeV&H HRV&I$LV&eHH6V&I H#V&!V&f$HwV&H HU&U&eI/HLHU&U&H U&fE1MIHIE1U&H HU&I$U&eHHMIsU&F H`U&H[U&oeHIHI:U&H H'U&I$!U&eHITAWAVAUATUHSHXH^H|$dH%(HD$H1HC $HD$0HD$8HD$@H HHHF(HD$(HE LuHD$HT& H(hE111AHLHHG H8u H@HP0HxT& H(hE111AHH|$HH; H8u H@HP0HEH5I&HHH IM_ HCH5I&HHHQ IMS LL#IH H;$L;=$щt$ L;=$шL$'t]I/u IGLP0H5ES&LIH H;$L;=C$ L;=d$L$ ˆT$'I,$Im|$'I/t$L%)K&H=R&IT$LIHq HIGH5O&LHHIIHMIHu IGLP0L5J&H=bR&IVL^IHHIEH5L&LHHHIMImuIEL\$LP0L\$ICE1H;)$H;G$ H;$HcL\$L\$HI.MtLhHcD$HE1LLL\$IlHHE&DiHMcKDL\$HII.I+Hv$I9D$LLIHVI/zI,$OL;`$AL;$Du L;&$AĉD$I+VL$L- I&H=P&IULIHHID$H5M&LHHII$HMI$HvL-H&H=HP&IULDIH`HIGH5J&LHHIIHMIHuIGL\$LP0L\$ICE1H;$H;%$fH;$Hc L\$L\$HIMtLxHcD$HLLL\$I\HHxC&Db1HMcKDL\$HISImuIEL\$LP0L\$I+u ICLP0H>$I9FiLLIHI,$u ID$LP0I.u IFLP0L;-$L;-$L;-$L*AąImu IELP0E&HD$HT$(IHH5^$Lh IEHxM3IHImIELP0xH]HHqH aHMHSH L@H$H5vL H81=H%]M&}M&%]HnM&XZH H=E1IHL$HdH3 %(LHX[]A\A]A^A_H$HD$(IELP0|$'5LD$!HJE1E1L&!HL&L&t]I/u7E1IGL\$LP0L\$MtImuIEL\$LP0L\$MtI+u ICLP0MI.IFLP0ID$LP0ImaDIGLP0t$bLf.zf(HH|$T$f.XT$f(ff/>f/HD$HT$(f(H5#Lx IHxL*IHkI/u IGLP0HmuHEHP0HH+HCHP0LIH~yHtH)HF(HD$@HE LHD$8HEHD$0>IHtvHHtBM HD$8Lt$0HD$HD$@HD$(HHIH5F&LIHVHD$0H|H5xF&LHVHD$8H{ IMtH5?&LHV^H HD$@IAH kK]E1H=.H I&I&I&K]ŶwH+I&HI&I&Z]fDH I&I&H=A5I&sHmHcI&!HPI&NI&i]@CIcICLP0 fIC@=jHD=&HLL\$Ll$0HD$@HcD$Hl$8HHt8L\$HIMImIEL\$LP0L\$sf.ID$L\$LP0L\$fDHD$IGLP0L\$mICLP0IFL\$LP0L\$LL\$L\$D$RH]MG&,HG&G&b^DIm%IELP0fDID$LP0zHuLvIHH}G&.HjG&hG&^[IHE1E1=G&!H*G&(G&k]I$HI$HiE1ID$L\$LP0ML\$HI/>%H;F&!HF&F&m]I,$ID$LP0ImD(HFHHD$0IHSF&!H@F&>F&p]{fHi:&LL\$Ll$0HD$@HcD$Hl$8HHt8耢L\$HI(HJE1E&,HE&I$E&^HDH: H|$Jf. H56&H=>E&17IH HI,$u ID$LP0HDE&&H1E&/E&]f. HuLIHuH]D&,HD&D&]3IC@=DLcd$H8&HLL\$IL|$0Jt8H\$8HD$@L\$HI MI/wIGL\$LP0L\$^rT$/T$Hff/H55&H=C&1͐IHHyI,$u ID$LP0HBC&(HC&C&]IIH E1C&,H C&C&]IH E1E1@[HuL>HS HE1E1BC&,H/C&-C&^I$HH{C&)HC&IB&]HgfHKE1B&,HB&I$B&^HfDIMkMMsIEII+u ICLP0IFMD$Ml$MLIL$IEHI,$uID$HL$LP0HL$HAH;#H; #HL$HL$HI!Lh1HHLx HL$蕎HL$HI*I.IHD$IFLP0L\$HA&,H~A&I$xA&?^HI$E1E1HQj@HKA&,H8A&I$2A&$^H Lcd$H]5&LL\$L|$0IH\$8Jt8HD$@yL\$HIiHC@&.H@&@&^M.H E1@&,H@&I$@&/^H^H5o1&H=0@&1)IHhH՛I,$u ID$LP0H6@&-H#@&!@&q^tHT$0HLLVH5c%ޕ)HQ]?&H?&?&]VHuL蘢IH?&.H +?&)?&^I6M{MnMcII$I+u ICLP0ID$MD$7HH#AH5jL H 6H8H1:H_ ]H~>&AX~>&p>& ]M^MINIHI.uIFL\$LHL$P0HL$L\$HAH;#H;#ZL\$HL$|HL$L\$HI{LX1HHL` HL$茊HL$HII/IIGLP0HE1=&.H|=&I$v=&^HMH`=&.HM=&K=&^H9=&.H&=&$=&^GH5.&H=<&1ʉIH HvI,$u ID$LP0H?<&/H<&<&^ImHH<&<&.<&^ H~<&0Hk<&i<&_HA@=@HHt$0HL$Ll$0L|$8^HL$HIImuIEL\$LHL$P0L\$HL$I/uIGL\$LHL$P0L\$HL$IH';&&H;&;&]HH;&#Hq;&o;&]H];&"HJ;&H;&]H6;&(H#;&!;&]tHHt$0HL$Ll$0L|$8{HL$HIHEMI:&,H:&H:&F^HBHA@=HHt$0L\$0L\$HL$Ld$8HL$L\$HIkI+uICHL$LP0HL$I,$uID$HL$LP0HL$I|HE1:&,H:&I$9&^HHP9&-H9&9&m^&H)MI9&,H9&H9&V^H&HE1I9&,Ht9&Ho9&\^HF!HHt$0L\$0L\$HL$Ld$8軕HL$L\$HIHI9&.H9&I$8&^HHN8&/H8&8&^$H'8&.H8&8&^I/HH8&8&.8&^E1MIHIE1X8&.HE8&I$?8&^HHMI#8&,H8&H 8&O^HIHUI7&.H7&I$7&^HITAWAVAUATUHSHXL-+&H^H<$dH%(HD$H1H#Ll$8HD$0HD$@H9 HHHHHH HOHL LOLDHH#SHgH5H81HP^Z6&h6&^ZH6&XZH 'hH=E1衣HT$HdH3%(L$HX[]A\A]A^A_H#HD$LuH6& H(hE111AHLHH. H8u H@HP0H_6& H(hE111AHLHH$ H8u H@HP0HEH5+&HHH IMh HCH5+&LL$HHH LL$IMj LLϺLD$LL$LL$LD$HHH;#H; y#ADA H; #D@AtgH)H55&LǺLL$LD$LD$LL$HH H;^#H; #ADA H; "#A I)uIAHL$LLD$P0HL$LD$I(uI@HL$LP0HL$EH) EL%,&H=t4&IT$LoIHHIFH5d1&LHH1IIHMIHu IFLP0L%[,&H=4&IT$LIHKHIAH5t)&LLL$HHLL$IIHMIHu IALP0L%#M9eHL4IHImu IELP0M9gLLIHI.| I/Z L;-#L;-)#u L;-K# DImN E?L-6+&H=2&IULIHHIGH5/&LHH}IMHI/ H=*&LL$褕LL$HIZH@H5'&LL$LHHLL$IIHMqIHz M9`9 LHLL$LD$軸LD$LL$HHI(uI@LL$LHL$P0LL$HL$M9aHLHL$LL$bLL$HL$HIH)uHALL$HP0LL$I)u IALP0L;-#L;-i# L;-# LAąaImu IELP0EH$HT$IHH5n#Lh IEHxMIHImuIELP0@Hm,HF(HD$Lm fDH#HD$f.Lf.D$ Lf.D$ L%(&H=B0&IT$L=IH HI@H5%&LLD$ HH LD$ HIHH IHuI@HL$ LP0HL$ D$HL$ $HL$ HI{ L%#L9a LHLD$(HL$ =HL$ LD$(HI I(uI@HL$ LP0HL$ H)u HAHP0L;-#L;-D#u L;-f#DImu IELP0EjL-K'&H=.&IULIHHIAH5e$&LLL$ HHiLL$ IIHMaIH7D$IHM9eSLLLL$ LL$ HI I){ImPL;#L;7#4L;U#'LLD$ LD$ A I(u I@LP0EXH$L$D$HT$Lp H5)#IHxL IH~I.IFLP0@HAHP0EHALL$HLD$P0LL$LD$ DHLL$ LD$HL$LL$ LD$HL$DHHL$HL$A^H&I,&H,&,&ZI.1E1E1IFH $LP0H $HtH)u HAHP0MtImu IELP0MI/IGLP0DIHSHtHHF(HD$@HE LHD$8HEHD$0:IHH[M~.H5!&LHVH9 HD$@IM' HD$@Lt$0Ll$8HD$H,HIH5m!&LIHVHD$0HM~H5!&LHV_HXHD$8IEH kZE1H=CH *&*&*&ZŗH+*&H*&*&ZfDH *&*&H=A5*&sHmu HEHP0HH+HCHP0HC*&H0*&.*&Z@#IZHsE1E1*&H)&I)&ZHIHuIAL$LP0L$LELL$I#fDIGLP0fIFLP0uIELP0L(AƅiHMM)&H:)&8)&[IALP0LAƅ4Hh)&H(&(&\E1{IGLL$LP0LL$bIELD$ LP0LD$ HD$ IALP0LD$ lIFLL$LLD$P0LL$LD$cDHE1E1E(&H2(&I-(&ZHIHuIAHL$LL$P0L$HL$IHIHpE1I@H $LP0MH $RI.H1DDMpMMXIII(uI@LL$LL\$P0L\$LL$ICH;#H;Z#qLL$L\$L\$LL$HLp1LHHHX LL$ L\$HD$sL\$LT$HHLL$ I*M HD$IBLLL$ L\$P0LL$ HL$LD$H E1E1&&H&&I&&ZHS@Hf.DQ5HH/&&H&&&&ZmDHHW%&H%&%&Z-DHFHHD$0IpHuL~IHcH%&Hr%&p%&[SHuL6IHMH=%&H*%&(%&Z{H{%&H%&$&[IH1E1TILD$ HfDHE1$&H$&$&ZIE1HHD$^LL$HuLLL$7LL$HH<$&H)$&'$&[I/LE1MHcE1E1IH#&H#&#&ZHH+ME1#&H#&#&[LL$IfDLqM LIIIH)uHD$ HAHLL$(P0LD$ LL$(IAH;# H;+# LL$(LD$ LD$ LL$(HI]Lp1LHL@ LL$ oLL$ HII/LIGLL$ LP0HL$ fH IE1"&H"&H1"&ZHQf.MMMMEIIImuIELL$LLD$P0LD$LL$I@H;#H;# LL$LD$LD$LL$HIx LH1LHHEHh LD$nLD$HIImIELD$LP0LD$MKHLM}!&Hj!&IEE1a!&[HM_MIOIHI/uIGHL$LL\$P0L\$HL$HAH;#DH;#HL$L\$aL\$HL$HI LX1HHLp HL$qmHL$HI I/HIGHL$LP0HL$Inf.H5&H=" &1mIH H{I,$u ID$LP0H( &H & &[ffDHcLE1&H&I&[HIS7HuL虂IHH&H&&![H5&H=8&11lIH HzI,$u ID$LP0H>&H+&)&\|LL$ IHmE1E1&H&&#[HuL賁IH,H"&H&&%\H&H&~&&[HT$0HLLȺH5%;tHKZA&hH.&,&KZ[H&H&&'\I{MEMMeII$ImuHD$(IELLD$ P0LD$ LL$(ID$H;8#eH;#LL$(LD$ LD$ LL$(HH* L@1HLLH HD$ jHL$ HI8 H)MHD$ HAHP0LD$ HuE1 &H&I&3[HHDM&H&&*\qHM&H&&V[HM&Hr&p&,\mfLL$ICH5r &H= &1iIHnHwI,$u ID$LP0Hy&H&&e[OHR&H&I&[HH$M&H&I&:\HMAMMyIII)uIAHL$LLD$P0LD$HL$IGH;#H;;#_HL$LD$LD$HL$H L@1HLHH HD$gLL$HI I)~IALP0M}H*M&H&&e\(H&H&&\LHt$0LL$0LL$LD$Hl$8vLD$LL$HIA I)uIALD$LP0LD$MH5) &H=&1fIHHwuI,$u ID$LP0H@&H&&\HHt$0L\$0L\$HL$Lt$8vHL$L\$HIN I+uICHL$LP0HL$I.uIFHL$LP0HL$I:IA@=LHt$0LD$8LD$(LL$ Lt$0;^LL$ LD$(HIYI.uIFLL$(LLD$ P0LL$(LD$ I(uI@LL$ LP0LL$ LH &H&&\I@@=_LHt$0LL$0LL$LD$Hl$8}]LD$LL$HII)uIALD$LP0LD$MMHA@=HHt$0L\$0L\$HL$Lt$8]HL$L\$HII+uICHL$LP0HL$I.uIFHL$LP0HL$MIvH].ID$@=Ht$0LLD$0LD$(LL$8LL$ m\LL$ LD$(HHI(uI@LL$(LHL$ P0LL$(HL$ I)uIAHL$ LP0HL$ MIIG@=Ht$0LLD$0LD$HL$8HL$[HL$LD$HII(uI@HL$LP0HL$H)u HAHP0M LHt$0LD$8LD$(LL$ Lt$0rLL$ LD$(HI*HzE1L &H&I&ZHIC@=zLHt$0LL$L\$Lt$0H\$8ZL\$LL$HHI.u(IFLL$ LHL$L\$P0LL$ HL$L\$MHW&HD&B& [Ht$0LLD$0LD$(LL$8LL$ qLL$ LD$(HHH\E1M&H&I&:[HH(&H&&\HE1L&H&I{&ZHIHͪLE1_&HL&IG&[HH1&H&I&[HIHeE1-H\ME1&H&IE1&ZHH%&H&&a[LHt$0LL$L\$Lt$0H\$8oL\$LL$HHH©MW&HD&I?&@\H!Ht$0LLD$0LD$HL$8HL$oHL$LD$HImHU&H&I&l\H9H'LM&H&IE1&[HHMI&Ho&m&[AHè[&HH&F&\HE1M.&H&I&J[HH)H_H&&&P[uxE1M^H0ME1MH&IM&&T\HrHM1&Hv&Iq&N\HSMuHE1P&H=&I8&\HFH"&H&I &|\HnH\LE1&H&IE1&[HH%&H&I&[HHE1L&Hv&Iq&ZH?HæLE1U&HB&I=&[H HE1$&H&I &[HH^M&H&I&H\HH-&H&I&u\HME1HE1M&Hs&In&C[Hf/HD$HT$(f(H5C#Lx IHxLGIHkI/u IGLP0HmuHEHP0HH+HCHP0L8IH~yHtH)HF(HD$@HE LHD$8HEHD$0~IHtvHHtBM HD$8Lt$0HD$HD$@HD$(HH'IH5}&LIHVHD$0H|H5&LHV̺HD$8H{ IMtH52%LHV螺H HD$@IAH #TE1H=H &&(&&#TqwHk&H&&2TfDH &&H=;A5&pHmH &H&&AT@胼IcICLP0 fIC@=jHl%HLL\$Ll$0HD$@HcD$Hl$8HHt8HHL\$HIMImIEL\$LP0L\$sf.ID$L\$LP0L\$fDHD$IGLP0L\$mICLP0IFL\$LP0L\$LL\$L\$D$RHM2&H&&:UDIm%IELP0fDID$LP0zӷHuLdIHH%&H&&`U蛺IHE1E1}&Hj&h&CTI$HI$HiE1ID$L\$LP0ML\$HI/>%H{&H&&ETI,$ID$LP0ImD(HFHHD$0IH&H&~&HT{fH%LL\$Ll$0HD$@HcD$Hl$8HHt8\L\$HI(HE1&H &I$&THDݵH: H|$芶f.Bf H5-%H=~%1wLIH H#[I,$u ID$LP0H%Hq%o%yTf.KHuL.bIHuH5%H"% %T3IC@=DLcd$H%HLL\$IL|$0Jt8H\$8HD$@DL\$HI MI/wIGL\$LP0L\$^rT$oT$Hff/H5%H=%1 KIHHYI,$u ID$LP0H%H%%TIIH KE1%H %%TIH E1E1@蛳HuL~`HS HE1E1%Ho%m%TI$HHS%H@%I;%THgfHE1 %H %I$%THfDIMkMMsIEII+u ICLP0IFMD$Ml$MLIL$IEHI,$uID$HL$LP0HL$HAH;#H;M#HL$HL$HI!Lh1HHLx HL$HHL$HI*I.IHD$IFLP0L\$H9%H%I$%UHI$E1E1HQj@H%Hx%I$r%TH Lcd$H%LL\$L|$0IH\$8Jt8HD$@WL\$HIiH%H%%tUM.HKE1%H%I$%UH^H5%H=p%1iGIHhHVI,$u ID$LP0Hސv%Hc%a%IUtHT$0HLLĖH5%P)HS$%wH%%SVHuL\IH%THH(%H%%EU&HiMI%H%H%.UH&H5E1I%H%H%4UHFaHHt$0L\$0L\$HL$Ld$8OHL$L\$HIHIU%HB%I$<%UHH&%H%%U$Hg%H%%|UI/H<H%%%UE1MIHIE1%H%I$%UHHшMIc%HP%HK%'UHIHI*%H%I$%UHITAWAVAUATUSHXL5'%L=%H|$H^dH%(HD$H1H(#Lt$0L|$8HD$@H0 H~~,HbHHF(HD$L~ LvDHѥ#HD$H[H% H(hE111AHLHH H8HH% H(hE111AHLHH H8HEH5%HHH IM HCH5%LT$HHH LT$IM L׺LLT$LT$HI+H;̤#L;r#AA L;#@At\I(H5%LLT$胢LT$HI H;c#L; #AA L;(#A I*uIBT$ LLD$P0T$ LD$ImuIET$ LLD$P0T$ LD$EI(uI@T$LP0T$4Lf.UD$( Lإf.UD$t L%%H=.%IT$L)IH HIEH5%LHH[ IIEHM IEHuIELD$ LP0LD$ D$LD$ LD$ HI_ HС#I9@ LLLD$ 6uLD$ HIe ImuIELD$ LP0LD$ I(u I@LP0L;=#L;=A#u L;=c#DI/u IGLP0E HD$L$D$(HT$L@ H5G#IHxLLD$LD$HI1I(wHmuHEHP0HH+HCHP0@H@HP0*H@HP0^H#HD$L%%H=2%IT$L-IHHI@H5"%LLD$HH LD$IIHMIHu I@LP0L%%H=%IT$L賢IH HIFH5(%LHHm IIHMh IHuIFLD$LP0LD$L%}#M9`c LHLD$rLD$HI I(u I@LP0M9e( LLrIH1 I.wImuIELT$LP0LT$L;#L;ß#u L;#DI*IE. HD$HT$IHH5ٞ#LP IHxMLT$LT$HIO I*IBLP0fDHHIH wHwHIHH֞#I?SIH5tH8L A1RH R% %RH%XZH  H=E1^WfDHL$HdH3 %(LHX[]A\A]A^A_fI@LT$LP0LT$H{%jH%%WRI*tIEE1H"IBLD$LE1P0IELD$HIEHuIELD$LP0LD$MMfDM I/Au IGLP0MtImu IELP0MtI.u IFLP0H G%M%H=A56%!VHmH#HD$LljT$(LT$ LD$貟T$(LT$ LD$LLD$苟LD$PHE1%jH%%^RI(E1I@LP0HH HtHUHF(HD$@HF HHD$8HFHD$0肘IHH3M~.H5%HHV3HQ HD$@IM? HD$@Lt$0L|$8HD$tHHIM~H5%HHV̝HtHD$0IM~H5%HHV覝HWHD$8IDH }g5RE1H=H .%0%g"%5R THk}%hH%%DRfH;}%jH%%SRcf諟I蛟LT$IfDH|{%jHh%f%URI*AIBLP0HD$IFLP0LT$pIBLP0I@LP0zL踜AąHH|ME1%mH%%R)LLT$sLT$A0H{E1%rH%~%`Sf:UH,H{O%kH<%:%iRDHuLGIH%Hm{%rH%%RH&f.H+{%jH%%ZRf腚HxHz%lHl%j%sRDHFHHD$0蟔I+HuLGIH9H}z%mH%%}RH Sz%rH %%RIHtE1E15賜LD$IfDH yE1%mH }%{%RfDkIKHuL.FH9 Hy8%rH%%#%SfDHsyM%mH%%RfDIH 3yMM%rH %%SM`M5MPI$II(uI@LT$ LP0LT$ IBH;#kH;T#LT$ ǗLT$ HIL`1LHLh LT$ .LT$ HII.MIFLT$ LP0LD$ DH;xME1%mH%%RMHMMxIII(uI@LL$LP0LL$IGH;#H;]#LL$ЖLL$HLH1HLHHX HD$-LT$HI I*MIBL|$LP0LD$DHCwM%rH%%SI/=)@M}MMeII$Imu IELP0ID$H; #@H;d#ܕHLx1HLLp HD$,LD$HIKI(MdHD$I@LP0LT$KfH5q%H=%1,IHHW;I,$u ID$LP0H v%nH%%RHfDE1HuM%rHm%k%=SfDHuS%oH@%I;%RH\H5%H=%1+IHhH:I,$u ID$LP0HRu%sH%%oSzH+u%tH%%SCHT$0HHL){H50%k4HtQq% H^%\%QIB@=LHt$0LT$ Ld$0Ll$8Q#LT$ HII,$uID$LT$ LP0LT$ ImuIELT$ LP0LT$ MIG@=]Ht$0LLL$0LL$H\$8"LL$HI0I)u IALP0MID$@=Ht$0LL|$0Lt$8p"IHI/uIGLT$LP0LT$I.uIFLT$LP0LT$MALHt$0LT$ Ld$0Ll$8W9LT$ HIH!sM%mH%%RHt$0LLL$0LL$H\$88LL$HIHr]%rHJ%H%SIMMMHjHt$0LL|$0Lt$88IHHfr%rH%%DSI.MMBJ@H+r%nH%%RSHr%sH%%kS,HqMr%mH_%]%RZHqMH%mH5%3%RHq!%rH% %&SHbqM%rH%%,SH8qMM%rH%%ZS4H q%rH%%TSDHpMt%mHa%_%R\HpM%rH:%8% SHp&%rH%%MS#I@AWAVAUATUSHXL5'%H^H|$ dH%(HD$H1H%Lt$0HD$HD$8H#HD$@H[ H~/H]HHF(HD$HF HD$LvHɍ#HD$HH% H(hE111AHLHH H8H@% H(hE111AHH|$IH H8HEH5%HHH+ HH ID$H5%LHH~ HH HκHHL$HL$HIjH;ˌ#L;=q#AA L;=#t$AVt`I/H5%HϺHL$HL$HIN H;_#L;=#AA ʼnT$L;= #A H+uHCHL$(HP0HL$(H)u HAHP0EI/u IGLP0L$KL f.=f( H|$T$f.=T$ f(=\f(fT=f. HD$ HT$f(H5 #Lx IHxLFIH I/IGL1E1P0HmLI,$u ID$LP0HtH+u HCHP0MII/?IGLP00H@HP02H@LP0hH#HD$H%H=2%HSH.IH HIGH5%LHHX HHZ I/u IGLP0HCE1H;# H;"# H;}#Hc IH# MtLpHcD$I$1LHLT$Md@HEHIl!LT$HI I*H+u HCHP0H8%IL0HH/ H8u H@HP0L-H%H=%IULHH! HHAH5%HHL$HH HL$IHHM HHL5%H=~%LT$IVLuLT$HI HIEH5%LT$LHH LT$IM ImHH#I9F HLLT$ZLT$HH I.H #I9B HLHL$LT$mZLT$HL$HI H)I*{L;-#L;-#L;-#LAƅt Imu IELP0E HD$ HT$IHH5#Lh IEHxM衸IH ImIELP0 HHBiIH /iHIoHIHH#I?SIH5G\H8L oA1 HhLX%J%LH;%XZH hH=n1?HT$HdH3%(HHX[]A\A]A^A_ÐIGHL$LP0HL$H;h%H%%MH+)HE11E1E1HHHuHALT$HP0LT$1MtI*uIBHL$LP0HL$MtImuIEHL$LP0HL$HtH)u HAHP0H !%'%H=m5%>HtHmu=E1HEHLP0MLYHA#HD$^1@LHL$(sHL$(fLXD$ Hf1E11x%j%MH[%1IMHMIDHH;HtHmHF(HD$@HF HHD$8HFHD$0RIHHM~.H5%HHVHHD$@IMHD$8Lt$0HD$HD$@HD$Gf.HHIM~H5(%HHV蔅HtHD$0IM~H5Z%HHVnHOHD$8I<E11efDHkeE1E11H%% %LH;eE11% H%%M@H eE1%H%%MVfD{HHC@=HcD$HHLt$0Ld$8HHl$@Ht8dIHa MI.IFLP0fDHzHKdE1E1E1H%%%MH+ HCHL$HE1LT$P0HL$LT$HMI.IFLT$LP0LT$HALT$HP0LT$IBLP0"fIELT$LP0LT$(IFHL$LLT$P0HL$LT$:DIBLP0vfHALT$HP0LT$JaD$菂T$HGHbE11~%Hk%i%&M4@D0;HuH/Hx Hb1&%H%%M@H>~f.HcD$HLt$0Ld$8HHl$@Ht8L(IHHbE11%H%%M@Ha%Hp%n%Mf5D$T$9T$D$HHaE11"%H% %0MHFHHD$0?{IH5a%H=%1HHMH7&H+u HCHP0HaE11%H%%QMKsHH`1E1V%HC%1?%Mf.H`11%H %1%vMLsMZLkIIEH+u HCHP0IELD$$DH`%H%%Ma{HuL^,HHH_e%HR%P%MH_E116%H#%!%MW@Hs_E11%H%%M'@HC_E1%H%%MfD賁HL$I<$~LT$HuLLT$e+LT$HH^1h%HU%S%MILT$I9H^HL)%H%LM%MMFMMnIIEI.uIFLD$LLT$P0LT$LD$IEH;P|#H;|#LD$LT$}LT$LD$H L@1HLHHX LT$HD$#LL$LT$HHI)MHD$IALLT$P0HL$LT$`HT$0HHLcH5Dj%?HR]L%H%%LH&]M%H%%M"MjMMrIEII*uIBHL$LP0HL$IFH;z#3H;M{#HL${HL$HLh1HLHH HD$LD$HII(MI@Lt$LP0LT$tH<\%H%%!NHHH\HE1LH%E1%%DNH5%H="%1+IHH Imu IELP0H[9%H&%$%TNHz[HE1 %!H%L%yN*IE@=-Ht$0LLT$LD$0LD$H\$8 LD$LT$HHI(uI@HL$LLT$P0HL$LT$MHZE1b%HO%M%MMIF@=Ht$0LHL$8HL$Ll$0B HL$HHD$JImuIEHL$LP0HL$H)u HAHP0MLl$ZH"ZE11%H%%MHt$0LLT$LD$0LD$H\$8LD$LT$HHHYR%H?%=%MLHMLMHt$0LHL$8HL$Ll$0HL$HHD$HPYM%H%%(N H&Y%H%%PNtHXL%H%%NHXm%HZ%X% NuHXM>%H+%)%8NeHXLM%H%%>NxHE1UHGX%H%%NIIHXM%H%%1NHCHL$HP0HL$fAWAVIAUATIUSHXH<$D$L$T$dH%(HD$H1H;u#yL-%HH=@%IUL%!H MH=SE1$H{MIH% %%%!H~IGLP0E1loIlHuLH6HM%%H%%!Imu-E1IELP0MI/u IGL1P0HHmu HEHP0H 1%7%5-%H=RE1#HuHE1H'fDk@$f{k@$fH;Lӵ%%H%%!HEHImtM1)f.1fHEHP0ImH1IUHEHImuIEHT$ LP0HT$ HBIչDHi#HH8>gHiK9"%(H%%9"H BK(H=Q!f.Hi#HH8fImHKH%%(%;"IELIP0H e%k%L5^%,fHJC%%H0%.% "HeHEHZ1LLLL$ >lLL$ HHDHCJLس%%Hų%ó%"+fDHyh#LH8eHJk!!%!H|%z%k!7DH=a7LL$ ohLL$ eDL $/iL $HJHcf#H5<7H8fL $+Hg#LH8.eHYIE1%!H۲%Hֲ%m!HHU1HI%#H%%!'1LLjImHwhH=i6|gImu IELP0HH"A%(H.%,%"@1LHZjH+H4%H=5 gH+u HCHP0H?H!!ͱ%!H%%!ugHAHd#H55H8d&ugHuHd#H55H8jdqHGLHR%(H?%=%H"HGML%%!H%%z!HfG%%H%%!H*H1LAHkcIm1I/@IEH $LP0H $E1]I H1AWAVAUATUSHHxHnH|$dH%(HD$h1Hd#HD$@HD$HHD$PHD$XHH HF HF0HD$0HC(HD$HC HD$HCH$H % H(hE111AHH<$HHH8u H@HP0H% H(hE111AHH|$HH H8u H@HP0Hu% L(hE111AHH|$AIH7H8u H@LP0HEH5%HHH IMHCH5Ȥ%HHHIMLLaIHsH;c#L;b#щt$ L;b#шL$( I(uI@LLD$ P0LD$ ID$H5&%LD$ LHHLD$ IMSLLs`IHH;Xb#L;a#щt$ L;b#шL$(tWI(" H5%L`IHH;a#L;a#ЉT$  L;a# ˆT$(I.uIFLD$8LP0LD$8I/! Im.|$(BI( DL$ E L-d%H= %IUL cIHHI@H5%LL$HH'L$IIHMIHu I@LP0L-%H=%IULbIHHI@H5w%LL$HHL$IMI(t IEE1H;f_#_H;`#H;`#Hc~ YaIHMtLpHcD$ HE1LLL $DpIlMcHK\TL $HII) Im H^#I9G LLL$/2L$HII(e I/C L;5_#AL;5Q_#D L;5n_# La$I.u IFLP0D$EH=A%IHH@H5%LHHIIHMIHuIGLL$LP0LL$H=ڢ%LL$ LL$HI~H@H5l%LL$LLD$HHLD$LL$IIHMIHuI@LL$LP0LL$IEE1H;6]#H;T^#H;^#HcLD$LL$_LL$LD$HIFMtL@Hc$H1LLLL$I\@I$HMdLL$HII.uIFL $LP0L $ImuIEL $LP0L $Hi\#I9ALLL $/L $HIFI/uIGL $LP0L $I)u IALP0L;-9]#L;-\# L;-\# LG_$Imu IELP0<$H=Ѡ% H#H5%HHD$LL$HII)u IALP0H=%Z IHH5%HHD$LL$HII)u IALP0IEE1H;[#H;9\#H;\#Hc]IHCMtLxHc$HE1LLL$Il@I$HMd L$HIqI(uH$I@LP0L $ImuIEL $LP0L $HcZ#I9FLLL $-L $HII)u IALP0I.u IFLP0L;-;[#L;-Z#L;-Z#LI]AƅImu IELP0EHD$HHMH5Y#ILh IEHxAUHT$@Y^IHImu IELP0HmH+fDL8VIH5%LIHV[HD$@HtHk@HHA<H 1<HMHUHmBL@HY#H5F/L BH81 ^H;8{[%M%8{H>%XZH ;H=AE1HL$hdH3 %(LHx[]A\A]A^A_HaY#HD$0IELD$8LP0|$(LD$8LLD$(|[LD$(D$ H;E1E1%NH%%{I(ufHF0HD$XHC(HD$PHC LHD$HHCHD$@RIHk~EHHu>MHD$@H$HD$HHD$HD$PHD$HD$XHD$0HM~HT$@HLL->H5G%$yH7${.%H%%${@H k7J`{E1H==H %%Jڠ%`{ H+7à%LH%%~{fI@LP0}IEL$LP0L$ fI@LP0IE@=mHcD$ HLLt$@Hl$HHH\$PHtHTIHMI.IFL$LP0L$tfDIGLP0fI@LP0H$IALP0L$/fH5Q%LHVUH"HD$XIfDAʼn$w@HFLHD$@OIH5%LHViUHD$HHo IH5%LHVDUHD$PHd IXDHS5%NH؞%֞%{)WILLD$8sULD$8 fH4E1E1E1H%I%Nr%{HIHuIGLD$LL $P0LD$L $MIHIHfWIkHk4E1%NH%%{ImtE1E1IHdfIEL$LE1P0E1L$fD‰$)DH3%NH%~%{@sVLD$ IjfDHcD$ LLt$@Hl$HHH\$PHtHIHH3E1 %\H %I%k|HHS3%NH؜%֜%{RHH3%OH%%{DD$oRT$HH2c%PHP%N%{fD$T$RT$D$HHo2%QH%%{EDQHuLIHH%2%\H%%R|IE@=Hc$HLLL$LD$@HLD$HtHH\$HLd$PLD$LL$HIEMI( I@L $LP0L $H5ъ%H=ʚ%1IH HoImu IELP0H91њ%TH%%{H 1E1E1E1H%%\%T|F{SL$IH5%H=%1IH HImu IELP0H0!%VH% %{_H5%H=%1IH HQImu IELP0H0%XH%%|fH$wOL$HuLL$RL$HH/E1E1E1HF%IE%\7%W|HfD#RL$IHk/ME1%\H%IE1%Y|HkfH+/E1E1%YH%I%;|HbMuMIUIHImuIEH$LP0H$HBID$ VMoMIOIEHI/uIGHL$LL$P0L$HL$HAH;L# H;L#HL$L$dML$HL$HI Lh1HHL@ H $HD$qH $LL$HI1 I)IZIALP0KDH-j%NHW%U%{eH-E1E1=%\H*%I%%|HHw-%\H%%|I.HIFLP0E1fDH3-E1Ȗ%\H%I%y|H9Hc$LLL$LD$@HLD$HtHH\$HLd$PLD$LL$HIH,W%^HD%B%|M@H5х%H=%1IH HImu IELP0HQ,%]H֕%ԕ%|'H*,E1%\H%I%|H0H+%^H~%|%|IE@=Hc$HLL|$@Hl$HHLd$PHtHlIH< MI/IGL $LP0L $NIdH T+E1E1E1H ڔ%ܔ%^Δ%|[H$+%^H%%|HHH#H5XL 1jAH +HP1H81 MH*AZZ%HG%A[{>%{H*E1)%^H%%| MLL$LD$IMEM MuIIImIFM$HHG#H5rL 1jAH *Hj0H81#LH)[{Hg%]h%Z%{MiMIIIIEHI)uIAH $LP0H $HAH;G#HH;H#bH $wHH $HILh1HHLx H $H $HII.IIFH $LP0L $H(%^Hz%x%}E1E1H(M]%^HJ%H%=}IH(6%^H#%!%|Hc$LL|$@Hl$HHLd$PHtHyIHHH(%`H͑%ˑ%w}E1MJ\f.H5I%H=b%1[IH>HImu IELP0H'i%_HV%T%L}H'B%^H/%-% }H'%`H%%^}YHA@=HHt$@LD$HLD$H $Ll$@H $LD$HIImuIEHL$LL$P0HL$L$I(uI@H $LP0H $I:H&b%`HO%M%`}.H&E18%`H%%#%c}IELD$LMLL$P0IF$LD$LL$HC&ME1Տ%`H%IE1%e}HCM}MIUIHImuIEH$LP0H$HBI$H%Z%THG%E%{H%3%VH %%{qMnMM~IEII.uIFL $LP0L $IGH;lC#H;C#L $;DL $HiLh1HLLH H$UL$HI I(MI@LP0H$MY%`HF%D%}EH$2%`H%%}MHs$ %XH%%|IHL$%`Hэ%ύ%}qH%$%`H%%}IHbH5-}%H=N%1GIH`HImu IELP0H#U%aHB%@%}H#M+%bH%%}HHt$@LD$HLD$H $Ll$@lH $LD$HI%H2#E1IČ%\H%H%|H5HA@=HHt$@H $Ll$@L|$HH $HIImuIEH $LP0H $I/uIGH $LP0H $IMZH}"E1%\H%I%s|HHL"%]Hы%ϋ%|"HHt$@H $Ll$@L|$H*H $HIH!E1E1IHL%HK%\=%|HH!E1I!%\H%H %|HH[!%^H%ފ%|MRIG@=Ht$@LLL$HLL$Ll$@LL$HH$!ImuIELL$LP0LL$I)u IALP0ML,$H D%_H1%/%H}H I%^H%%1}H[ I%^H݉%ۉ%7}}Ht$@LLL$HLL$Ll$@5LL$HH$HM%`H%~%} HE1Mf%`HS%IN%}H HM5%`H"% %}Hv%aH%%}LHO%`HԈ%҈%}H(I%^H%%*}+IIHE1I%\Hu%Hp%|HHMW%`HD%B%}ff.fAWAVAUATUSHHxHnH|$dH%(HD$h1Hc<#HD$@HD$HHD$PHD$XH"H( Hf HF0HD$0HC(HD$HC HD$HCH$Ḣ% H(hE111AHH<$HHH8u H@HP0H% H(hE111AHH|$HH H8u H@HP0H5% L(hE111AHH|$AIHWH8u H@LP0HEH5|%HHH)IMHCH5|%HHHIMLL8IHH;:#L;j:#щt$ L;:#шL$(< I(uI@LLD$ P0LD$ ID$H5{%LD$ LHHLD$ IMsLL38IHH;:#L;9#щt$ L;9#шL$(tWI(B H5[%L7IHH;9#L;Y9# L;z9#L$  ˆT$(I.uIFLD$8LP0LD$8I/A ImN|$(bI(1 DL$ E; L-$}%H=̈́%IUL:IH HI@H5%LL$HHWL$IIHMIHu I@LP0L-|%H=V%IULR:IHHI@H5~%LL$HH0L$IM6I( IEE1H;&7#H;D8#H;8#Hc 9IHMtLpHcD$ HELLL $IlHHw%Dr1HMcKD L $HII) Im H~6#I9G#LLL$ L$HII({ I/a L;5b7#AL;57#D L;5$7# Ln9$I.u IFLP0D$EH=z%IH8H@H5%LHHIIHMIHuIGLL$LP0LL$H=z%LL$^LL$HIH@H5{%LL$LLD$HH5LD$LL$IIHMIHuI@LL$LP0LL$IEE1H;4#H; 6#-H;e6#Hc,LL$LD$6LD$LL$HIkMtL@Hc$HLLLL$I\HHTu%HHcID1LL$HII.uIFL $LP0L $ImuIEL $LP0L $H4#I9ALLL $|L $HIaI/uIGL $LP0L $I)u IALP0L;-4#L;-4# L;-4# L6$*Imu IELP0<$H=}x%PH?H50}%HHD$LL$HII)u IALP0H=3x%IHH5y%HHD$VLL$HI I)u IALP0IEE1H;2#H;3#H;@4#Hc04IH5MtLxHc$I$LLHOs%L$Md@HHIT1L$HII(uH$I@LP0L $ImuIEL $LP0L $H 2#I9FLLL $pL $HII)u IALP0I.u IFLP0L;-2#L;-2#$L;-2#L4AƅImu IELP0EHD$HHMH52#ILh IEHxAUHT$@ZY^IHmImuIELP0fHmH+fDL-IH5~y%LIHV3HD$@HtHk@HHH HMHUHL@H%1#H5L yH815Hh_|%3|%_H|%XZH ?3H=E1HL$hdH3 %(LHx[]A\A]A^A_H1#HD$0IELD$8LP0|$(LD$8LLD$(3LD$(D$ HE1E18|%}H%|%#|%_I(ufHF0HD$XHC(HD$PHC LHD$HHCHD$@)IHk~EHHu>MHD$@H$HD$HHD$HD$PHD$HD$XHD$0uHM~HT$@HLLH5i%yH;|_x%3Hx%x%|_@H y_E1H=H x%x%yzx%_eHcx%{HPx%Nx%_fI@LP0]IEL$LP0L$fI@LP0IE@=MHl%HLLt$@Hl$HHD$PHcD$ HHtHIHMxI.nIFL$LP0L$WIGLP0I@LP0vH$IALP0L$fH5l%LHV]-H"HD$XIfDAʼn$a@HFLHD$@?'IH5r%LHV -HD$HH~ IH5xo%LHV,HD$PHs IXDH v%}Hxv%vv%_)k/ILLD$8-LD$8fH E1E1E1H!v%I v%}v%_HIHuIGLL$LL$P0LL$L$MIHIHf.IKH E1u%}Hu%u%_ImtE1E1IHdfIEL$LE1P0E1L$fD‰$DH 3u%}H u%u%_@.LD$ IJfDH1i%LLt$@Hl$HHD$PHcD$ HHtHUIHH$ E1t%Ht%It%`HH t%}Hpt%nt%_fE*HH ?t%~H,t%*t%`DD$)L$HH[ s%Hs%s% `fD$)L$D$HH s%Hs%s%`;k)HuLNIHH Us%HBs%@s%`IE@=H\g%HLLL$LD$@HD$PHc$LD$HH\$HHtHLD$LL$HIFMI(I@L $LP0L $@H5c%H=Zr%1SIH HImu IELP0Har%HNr%Lr%.`H E1E1E1H !r%#r%r%`6 +L$IH5b%H=q%1裾IH HOImu IELP0Hq%Hq%q%N`OH$w'L$HuLL$RL$HHE1E1E1HFq%IEq%7q%`H fDH5a%H=p%1ӽIH5 HImu IELP0HIp%Hp%p%n`)L$IHME1p%Hp%IE1zp%`HcHE1E1]p%HJp%IEp%`HbMuMtIUIHImuIEH$LP0H$HBID$ 6MoMIOIEHI/uIGHL$LL$P0L$HL$HAH;:$# H;$#HL$L$%L$HL$HI Lh1HHL@ H $HD$H $LL$HIA I)IDIALP05DHr o%}Hn%n%_eHKE1E1n%Hn%In%`HHn%Hn%n%aI.HIFLP0E1fDHE1hn%HUn%IPn%`H9Hb%LLL$LD$@HD$PHc$LD$HH\$HHtHLD$LL$HI'HXm%Hm%m%IaMDH5^%H=zm%1sIH& HImu IELP0Hm%Hnm%lm%aHE1Wm%HDm%I?m%`H(H)m%Hm%m%0aIE@=H3a%HLL|$@Ld$HHD$PHc$HHtHIH MI/IGL $LP0L $%I?H E1E1E1H kl%ml%_l%2aLHMl%H:l%8l%5ayHH( #H5L | jAH HH81$HUAZk%3Hk%A[r_k%r_H %E1k%H k%k%7a$LL$LD$IMEMMuIIImIFM$HHB#H5L jAH HH81#Ho[l_Hj%]j%3j%l_MiM.IIIEHI)uIAH $LP0H $HAH;9#IH;#cH $ H $HILh1HHLx H $H $HII.IIFH $LP0L $Hj%H j% j%raE1E1HYMi%Hi%i%a:H/i%Hi%i%WaH]%LL|$@Ld$HHD$PHc$HHtHIHHji%HWi%Ui%aE1MCF@H5Z%H=h%1IH>HImu IELP0Hah%Hh%h%aH:h%Hh%h%baHh%Hh%h%aIHA@=HHt$@LD$HLD$H $Ll$@臭H $LD$HIImuIEHL$LL$P0HL$L$I(uI@H $LP0H $IHZg%Hg%g%aH3E1g%Hg%g%aIELL$LMLD$P0IF$LL$LD$HME1eg%HRg%IE1Jg%aH3M}MIUIHImuIEH$LP0H$HBI$HRf%Hf%f%*`MnM$M~IEII.uIFL $LP0L $IGH;##'H;~# L $L $HLh1HLLH H$ L$HI3I(MI@LP0H{f%Hf%e%J`HTe%He%e%a}H-e%He%e%j`cHe%He%e%aHMte%Hae%_e%bHMe%H:e%8e%aIHRH5U%H=d%1ױIHHImu IELP0HMd%Hd%d%*bHHt$@LD$HLD$H $Ll$@&H $LD$HIOHE1I~d%Hkd%Hfd%`HOHMMd%H:d%8d%ObHA@=HHt$@H $Ll$@L|$H.H $HIImuIEH $LP0H $I/uIGH $LP0H $IM>H E1c%Hc%Ic%`HsHtc%Hac%_c%aHHt$@H $Ll$@L|$H躿H $HIfHF0HD$HHC(HD$@HC LHD$8HCHD$04IH~WH4Hu%H5N%LHV HD$HH IMt Lt$0Ld$8Hl$@H|$H.Huo@IG@=HHLLD$0Ht8LD$Hl$8HL$@HL$_HL$LD$HHW%1HjLV% HV%LIILHIfH)Mt I(HtH+u HCHP0MtuI/uoIGLP0c HMH}V%(HjV%hV%@ .fDHSV%*H@V%>V%J fDH )V%/V%H=15V%fH@`H HH HIHu H #H@HT$H9f IGHH<AGAĩTI/IGLP0HLLD$0Ht8LD$Hl$8HL$@HL$DZHL$LD$HHI MtI(uI@HL$LP0HL$H)uHAHP0ffLc&2jHFLHD$0IH5O%LHV HD$8HIH5HJ%LHV HD$@H&IDHHHOT%QT%/CT%x HE1E1@HALD$HP0LD$ALH5HIHRH@3{ @$6f HuL讶HHjHS%1HS%S% c IVH{S%1HhS%fS% HHI/HH;S%=S%//S% H I(o LE1fI@LP0:HL$HL$HuHĵHHH3R%1HR%R% H R%1H R%R% #f{ H@$fMGMI_IHI/MHCIߺH{R%1HR%Q% fLH腷HH|H41Q%/HQ%1Q% fD1HL HHH+HHnQ%pQ%1bQ% H#H5H8:DHq#H5H8 DHuH޳HH~HMP%/HP%P%v H=HL$HL$DH H#H5H8|k HL$HfDHKP%/H8P%H3P%{ Hf.LcMLCI$IH+uHCLD$HP0LD$I@H;q#H;#LD$?LD$HIYL`1LHLx LD$TLD$HHImL(IELP0fDHLAHH/Hw#H50H8 I/lIGLP0]E1nIGLD$LIHL$P0HCLD$HL$1Ht$(H*HLB1 uH{Ht$(AHH=`HL$nHL$s7HeHo#H5HH8(JHT$0HLL H5j$fHX M%HM%M% ZHH#H H5jL AHH818HY^HM% }M%oM% f.H@`HIHH9LHH(H #H@HL$H9uKDcHCHHuHCAƩuaH+HHCHP09HH5`HHH@뙐HHAHtHH#H5H8tH+`HCHP0QE1eHHE#AH5jL H H8H1Hr_ HK%AXK%K% qHCLK%1HK%LIK% HH"L H5gjAH ~HH81HAYpK%H]K%AZ TK% I@@=lLHt$0LD$Ld$0L|$8ILD$HHI,$uID$LD$LP0LD$I/uIGLD$LP0LD$L}LHt$0LD$Ld$0L|$8)LD$HHuHLLJ%/HvJ%1rJ% XHHX"H51H8I@H"H5 H8*Hf"H5H8HH"H5H8H<LMI%/HI%II% HGhH@`H<HH,HIHHD$I9Gu$LwI/AoIGLP0`LH5|IHuH@`HHHLHHt}HD$H9Au.HHL$HL$AH)QHAHP0BHH5ZHHuI,$mE1L1ɺAHsHZHs"H5LH8d?JHHJ"H5#H8;HLL#H%/HH%1 H% HH\mLLE1AWAVAUATUSHH^H|$dH%(H$1H"HD$pH* H1H_HFHD$HD$XHD$`HD$hHH@pLmxLHD$HtHMtIEMtIHD$H;w" L%B?%H=#G%IT$LHH:HHCH5A%HHHpHD$`HpH+nH"H|$`HD$ H9G{Ht$aHHD$XH H|$`H/AH5":%H\$XHD$`HD$XH"H9F"HHHD$XHHHD$`HnH|$XH/uHGP0HD$XH|$`H;=_"H;=" H;=#"pAąH|$`H/uHGP0HD$`EH57%H=3E%1,HD$`Hn"H֠H|$`H/H:E%HD$0H"E% E%7HD$`H|$XHt H/uHGP0HD$XH|$hHt H/uHGP0L%D%L}XHD$hM9+ MtrID$\IGA@IXH HJH~(L;b 1fDL;d HH9u1E1E1E1E1H}pHT$HExHLmxHUpLHt?H/u9HWHt$(HD$ LL$HL$LT$R0Ht$(HD$ LL$HL$LT$Ht8H(u2HPHt$ HLL$HL$LT$R0Ht$ LL$HL$LT$Ht.H.u(HFLL$HHL$LT$P0LL$HL$LT$H|$XHt+H/u%HGLL$HL$LT$P0LL$HL$LT$Mt$I*uIBLL$LHL$P0LL$HL$H|$`Ht!H/uHGLL$HL$P0LL$HL$H|$hHt!H/uHGLL$HL$P0LL$HL$MtI)uIAHL$LP0HL$MtI,$uID$HL$LP0HL$HtH)u HAHP0H OB%UB%H=Q15BB%-MTI/JIGLP0;HHH HIHHZH?L PHLIL@HH"SHH5H81_H6A%A%6HA%XZH H=1lH$dH3%(H$HĘ[]A\A]A^A_H"HD$HCHP0fHGP0@HHt$XfHD$H<%Lx HIWHHT$ IH0H@HT$ HHLLIHHD$HG<%Lx HIWHHT$ OHHHHT$ LM=LHAHHHHD$`H; "!HPHT$`HHXHHH(u HPHR0Ht$`HHHD$XHH|$`H/uHGP0HD$`H+u HCHP0H|$XH/uHGP0HD$XHD$HxH51%1L:I,$Hu ID$LP0H<H+bHCH1P0HL$HtHHD$HHu HAHP0MtImu IELP0M I. IFLP0HO"HHH[H+QHCHP0BfDDH52%H1'HHD$`HH;"AH;"Du H;="EH/QHD$`EHD$L%9%Lx LIWHHT$0HD$(H-H@HT$0HHLH|$(HD$(HHD$L%9%Lx LIWHHT$0HHHHT$0LM^ LHAHD$XHHHHD$hH;L$  HPHT$hH H@HHH|$XHD$XH/uHGP0Ht$hH|$XHd HD$`HH|$hH/uHGP0HD$hH|$XH/uHGP0HD$XH|$`H/uHGP0HD$`fL`pILxxLMtI$MtIMtIHCHCH H H LD$0LL$  H~LL$ LD$0H HL$HLD$ HqMLD$ tI,$uID$LP0LD$ MtI/uIGLD$ LP0LD$ MtI(u I@LP0L|$(H5-%1LiIIHQHL$ IHu H|$(HGP0MI,$,ID$LP0fHHd H2HFHHD$pHHD$pHD$HD$(H6fI$cfDHGP0@1fAąH|$`kfDMM9tMu1L;%F"LH :%:%H=5~:%iHL$`HT$XHHt$hBbL%2%H=t:%IT$LoIH HIGH5D7%LHH IMI/u IGLP0HL"HD$ I9D$Ht$LIHI,$u ID$LP0IGH56%LHHIMI/u IGLP0H 1%H=9%HQHHL$HL$HIHIGH53%LHHsHH-I/uIGHL$LP0HL$HL$lHL$HIHHHD$LT$HHlH/%H5^5%HLT$(HD$HL$LT$(z HLLHL$(LT$2LT$HL$(HII,$uID$HL$(LLT$P0HL$(LT$I*uIBHL$LP0HL$H)u HAHP0L;=m"L +%H"I9A LIHLκLLL$NLL$HII)uHD$IALP0LT$H5+%1LLT$ LT$HHLHLT$HD$(LT$HL$(HII*uIBHL$LP0HL$H)u HAHP0ID$H5K4%LHHHHI,$uID$HL$LP0HL$HD$ H9Al LaM_ HQI$HH)uHAHT$HP0HT$HLHT$[HT$HIlI,$HuID$LL$(LHT$P0HL$LL$(H)uHALL$HP0LL$L; "L; N"L; l"LLL$LL$AI)u IALP0EH52%LyIH]HD$LL$HHoH+*%H542%HLL$(HD$HL$LL$(H5'%HLHL$(LL$LL$HL$(HI[I)uIAHL$LP0HL$H)u HAHP0L;%e"LI/u IGLP0HD$H5s0%Hx 2}IHaHD$HD$H5e0%Hx }LT$HHHD$ H9A LyM HQIHH)uHAHT$ HLT$P0LT$HT$ HLLT$ HT$HT$LT$ HII/Hu(IGLL$(LLT$HT$ P0LT$HL$ LL$(H)uHALL$ HLT$P0LL$ LT$I)uIALT$LP0LT$ID$ HL$LT$It$HyHLT$H5%%1LLT$:LT$II*uHD$IBLP0LL$MI)u IALP0H|$hH/uHGP0HD$hH|$XH/uHGP0HD$XH|$`H/uHGP0Ht$HLLHD$`MxxH9"HHDHGP0mH"H5H8LL$ LD$0LD$0LL$ uLL$ LD$0H=HHIBH5P%L\$ LLT$HHG>LT$L\$ IIHMc=IHuIBLL$ LL\$P0LL$ L\$LL$ L\$蚻L\$LL$ HI<HD$(LL$0L\$ HIBLT$LT$L\$ HILL$0j<Hi"H5%HLL$0L\$(LT$ HD$讽LD$LT$ L\$(LL$0Y8IAHHBLL$0L\$(LT$ LD$LD$LT$ P H q"L\$(LL$0;P ALHL$8LLL\$0LD$(LT$ LL$H荺HL$8LL$P LT$ LD$(L\$0P =H @9b7H}8I)u(IAL\$(LLT$ LD$P0L\$(LT$ LD$I*uIBL\$ LLD$P0L\$ LD$I(uI@L\$LP0L\$H}uHEL\$HP0L\$L%$H=K%L\$IT$LAL\$HIBHIEH5%L\$LHHdBL\$IIEHM>IEHuIEL\$ LLT$P0L\$ LT$E1E1HD$@I9B>IBAUH;D$H<H;`"Hc7HL\$ LT$͸LT$L\$ HIBMtL`IcHHT$LHLL\$(I\AEHIM|AEHHITAE1HEHIlLL$ LT$OLT$LL$ HIL\$(-BI)uIAL\$ LLT$P0L\$ LT$I*uIBL\$LP0L\$I}uIEL\$LP0L\$IEH5X$L\$LHHAL\$IM`AHEH5"$L\$ HLT$HH!ALT$L\$ IM@L׺LL\$ LT$[LT$L\$ HI<I*uHD$ IBLL\$P0LL$ L\$I,$uID$LL$ LL\$P0LL$ L\$L; "L; "5L; "5LL\$ LL$LL$L\$ A?I)uIAL\$LP0L\$EDISHEL\$(L5|$HHD$LHT$ 蟷L\$(HI=H@HT$ HH9LL\$ LL\$ HI=ISL5,$L\$(LHHT$ 7L\$(HH>H@HT$ HH8LL\$ HL\$ HH{>HD$@H9A5LQM5LIIIH)u(HALL$0HL\$(LT$ P0LT$ L\$(LL$0LLL\$0LT$(LL$ 菆LL$ LT$(HIL\$0DI*uIRLL$(LL\$ R0L\$ LL$(I)uIQL\$ LR0L\$ I.uIVL\$ LR0L\$ L\$(L\$(HD$ IEHLt$L\$(IHD$HLMILIH@H|$H0H8H0H0H H0HH0I1HECHC ->fH(H0H0H@(;sHcH H0H@H0Pt8:H(HR8HcR H0@LLL$@LL$@&IFLL$ LP0LL$ LL\$HLL$@螴L\$HLL$@L耴D$ H$H$$)E11E1E1E1SHӓE1E11HZ$E1Y$K$HD$fMtImu IELP0HtHmu HEHP0MtI,$u ID$LP0H $$H=15$iHtH+u{1HCHHP0MtI/u IGLP0H|$t HL$HHD$HHu HAHP0H$dH3%(H-HĨ[]A\A]A^A_1@HE11E1H2$4$&$HD$HkE11E1H$$$۴IMH+1E1$H$$d@蛴I9HE1E1E1Hq$I$o$a$H1I$HID$L\$LE1LL$P0LL$L\$Mt$I.uIFL\$LLL$P0LL$L\$MtI+uICLL$LP0LL$MI)IALP0fE1HE1$H$$I)3>I$1E1E1H@IA@=rH|$HLL\$pL\$H$HcD$ LL$HH\$xHtxP?LL$L\$HI"MI+ICLL$LP0LL$xfDHD$IFLP0L\$ICLP0fIELL$LP0LL$!HEHP0LL\$#L\$HcuHF$H3$1$aI+=ICLE11E1P0IFLP0IE@=HD$(Ld$pE1HD$xHz"H$HcIUHHtxHBR uMU'HLHH]'M7I,$,ID$LP0f.HD$IFLP0LL$+HHE11E1H$$$4HuHZH=H@E11E1H$$$vfD蛭H3HE11E1H$$x$>3[H HE11E1HD$F$8$HH5$H=$1CIH%HRImu IELP0HQE11E1H$$$`H$HLLL$HtxLt$pL|$xH$SLL$HI#MI.IFLL$LP0LL$MqM.IiIHEI)u IALP0HEI1LLLT$(JLT$(HHImHOHD$H$$$-s;1I*u IBLP0H $$H=5$bHtHmu HEHP0z$HD$g$HX$L%$MnLLHHH@HH* HLLHHH$MnHLHD$(IHH@HH LLLIHH@H;";IT$H-IL$HHI,$uID$HL$8LHT$(P0HT$(HL$8HHHT$8HL$(2{HL$(HT$8HIH*Iu HBHP0I,$u ID$LP0Imu IELP0Ht$ HL$HT$H|$0IuLI蚤HEH5g$LMHt$EH ާ"Ht$P ;P HL$1HAIHL$P P H= @9' M?HmuHEHP0MI,$u ID$LP0LާHHNHmb-$HH= HD$H$$b-_MD@HHL$@eHL$@HIXH$HD$H|$z$-HT$(HHD$HH8H)uHALT$HP0LT$MxH .$4$H=H5#$_fHk$H$$VfH|$֪LL$ IfH$LL\$pL\$H$HcD$ LL$HH\$xHtxNLL$L\$HIhH͇E1b$HO$I$I$HH3$H $$ fIA@=H$HHLHtxLL$Lt$pL|$xH$5LL$HI1H1E1$H$$aDH1E1E1Hj$E1i$[$*fDKIH5 $H=$1HHӁE1h$HU$I$O$#HHEHP0sI$x踠@$ HuLMHHFH\y-$HD$H$$y-H +H=W8HE1$H$I$$.H)H57$H=8$117IH;HEImu IELP0HE11E1H.$0$"$p@HP0H;0 HHP0H0H0H0PѢI1H$HEHT$H$$${-HHEHP0IA@=HD$HLL\$LL$HD$xHp$Ld$pH$HcHHtxc.LL$L\$HI!MI,$ID$L\$LLL$P0LL$L\$H@E11E1H$$$vMeMUImI$HEImu IELP0HEI oIH~ME1E1HH$E1G$9$违@$H~E11$H$$H"LH8HC~-$HD$H$$-H ~H=ȄTIAL\$LP0HEID$ L\$dH}E11E1HT$V$H$LT$@ɜLT$@@$H"LH8IHT$(Ho}$HD$H$H$-HD$HHH|$(HGP0H"}$HD$H$$-MI,$mH=xjHL$8LT$(聛LT$(HL$8aLT$@LT$HIHs"H5LjH8,LT$)MMMMuIIImuIELL$LP0LL$IFH;D$HRH;К" LL$CLL$HLH1HLL` HD$[2L\$HII+MVHD$ICLP0LL$=HH;l$_L\$(LMILH|$ L\$H5$1L1I,$L\$IuID$L\$LP0L\$M+I.uIFL\$LP0L\$H}Imc IHIHICLP0H{$H$$ČE1I$1E1HHzE11\$HI$G$HzM2$H$$H٘"LH86Haz-$HD$H$$-HD$HLL\$HtxLL$HD$xH$Ld$pH$@LL$L\$HIeHyu$Hb$`$!M'E1LLHIMHyML%$HD$H $ $-xLcJL\$LL$ HI|LHL\$ HD$ŋLD$L\$ HH#"H5D$HL\$(LD$ HD$mLT$LD$ L\$(QLLLL\$(LT$ LD$ LD$LT$ HHL\$(I,$u)ID$L\$(LLT$ LD$P0L\$(LT$ LD$I(uI@L\$ LLT$P0L\$ LT$I*uIBL\$LP0L\$H}HEL\$HP0L\$HiE11E1H^$E1]$O$IAL\$LP0HEIL\$EH{iE11E1H$$$HLi1E1$H$$HD$LHl$pLl$xH$HcD$LL$HHtx/LL$HIHhE1s$H`$^$ٍ-Hh1E1E1H;$=$/$HhE11$H$$lHYhE1E1$H$I$$HlL]MLmIIEHmIELAXHt$pLLL$pLL$Ld$x-LL$HHHgMK$H8$6$ˌHg$$HD$H $ $-9H1LH\HNgE1I$H$HE$UHaIiHTMaHEI$I)u IALP0ID$MD$Hfl$HY$W$HfE1E1IH3$HE1$#$[HHufE1 $H$$ōHD$H@@=H|$Ht$pLL$pLL$Ll$xLL$HHI)u IALP0ImIELP0HeE11E1HZ$\$N$MNMIFIHI.HD$uIFL\$ LLL$P0LL$L\$ HD$H@H;D$H H;" L\$ LL$RLL$L\$ HILHH|$1HLX eIHI,$Lt$+ID$LP04HdE11b$HO$M$Hd1E16$H#$!$HwdE1 $H$$H5$H=$1IHuHI)Imu IELP0HdE11E1H$$$6IHcMy$Hf$d$'.LL$0L\$(LT$ LD$ւLL$0L\$(LT$ LD$@$hHc$HD$H$$.I1HIE1I*MuIBLL$ LL\$P0L\$LL$ MtI)uIAL\$LP0L\$H $$H=iL\$5$s9HL\$HmHEL\$H1P0ML\$ImLL$(L\$ LT$LD$LD$LT$HL\$ LL$(H`b1$HD$H$$.I)E1IAL\$(LLT$ LD$P0L\$(LT$ LD$MMDrIB@=HD$LL\$ LT$H$IcHkLd$pH\$xL$HtxH$cLT$L\$ HIOMI,$yID$L\$ LLT$P0LT$L\$ UIB@=5HD$LL\$(LL$pH$HcHkLL$ LT$H\$xHtxL$LT$LL$ HIL\$(rMI)IAL\$ LLT$P0LT$L\$ iHEL\$HP0L\$HL\$(HL$ b$.)+L\$MJMeIjIHEI*IBLL$ LL\$P0IL\$LL$ #H;Tӽ$HD$H$$.HD$LL\$(LL$pH$HcHkLL$ LT$H\$xHtxL$LT$LL$ HIL\$(HSJ$HD$H2$0$.MI)u1E1E1EIJIHZS1M$HD$Hռ$Ӽ$.H)S$HD$H$$-/I1E1HIE1HRE1L\$p$H]$[$ZSIAL\$LP0L\$I齫1ɺAIE11H5$H=˻$1L\$L\$HItUHL\$eI,$L\$uID$LP0L\$H$R$HD$H$$l/HQ$HD$Hx$v$h/1E1iE1E15HQE11J$H7$5$2Han"H5:?H8nLL$(L\$ LT$LD$H\Q$HD$Hܺ$ں$.I,$H+QHD$H$$$/I*HPM$Hq$o$ԌMIvIELT$L1P0LT$tHPE1)$H$$MAH\P1E1E1H$Lt$$ҹ$!顾H(P1E1Lt$H$$$pHOE11Lt$H|$~$p$?H|$(HL$ LT$HGP0HL$ LT$HO9$HD$H!$$0xHuO $HD$H$$.ff.AVAUATIUSHH0L +m"HndH%(HD$(1H$HD$HD$LL$HHHHHNH NHMHUHUL@H@l"H5BL UH81pHN$6$H$XZH ZN6H=4D$1"LN0LC(HK HSIt$ I|$H|$(dH3<%(H0[]A\A]A^H#IH,YHcH>fHF0HD$HC(HD$HC LHD$HCH$gIH~-HHu&MyH$HL$LD$LL$IHM~HLMTHLH5T^$ yHFMٶ$6Hƶ$Ķ$H5i$LHVlHtHD$IDHFLH$fIH5v$LHVlHD$HIH5$LHVulHD$Ht?IfLxfIH5$LIHV>lH$HuHkHHi"H uLH5?jL \L$HL$VL$HYff.SHH0 D$ f/L$(f/@HZHD$Z ^L$ D$D$3[ ^L$(D$D$[XD$=~ f/rf/ vV\$H0[^f(fD$ HUHD$D$(UL$H0[X^f(fDD$UZf(D$^L$ L$:ZL$f(^T$(T$ f(_\\$f(L$X\$T$ D$\f(XXD$YL$H0[\f(XHY THXfSHH $D$X $Hf(Yf($X\$$H[Y^f(fAWfAVH*AUIATUHSHD$xt H9x l$xLDžf(\l$Hf/|$@ \$H|$@T$YXL$f($DZL$$YL$@L,fT$Lf(L$`Qf. Y %%-%f(L$@Y \f(fTf.%-Xf(l$ f(D$ff(I*XXXf($\f(L$p |$0^XL$HYfD(D$8f(\f(\f(^YXYf(Yt$@\f(L$X^Yf(XYfA(AXXfA(^L$hYT$D^Xf(D$PfA(XD$H8VL$HY $!V $f/L$f(Df/L$ t$d$8fI*\Yf(^f(XT$(XD$0\X\$ fT^\f/ $VW $T$(H,IM)LH?HL1H)H~#\$`Y\$ fH*\f/IEfd$H^d$@H*YL9Ff/I)t$xf/t$ IGHH[]A\A]A^A_@f/L$PL$($U^D$XXD$0VH,HeL$(\L$$YYL$XDID$H9Tf.ff(H*H^f(\YH9}DL$($`Tl$p^D$h\f(V$L$(H,I9\L$PYYL$hPHCL9ff(H*H^f(\^L9~Df(^|$`LIXf(XYHX^^XT$ YfH*^T$($f(eS$T$(f(f(\f/Xl$(f/HCfEfML*ID$fM)H*fEIGL*IEH)H*fE(D$EYf(f($A^fE(D$YEYfD(,$DYD$$D$D$sRD$$D$fA(^4$f(CR\$@<$D$Y|$H$AYf(^ R=MD<$l$(fA(f(D$D9D$^D$D$D$\f(5^f(\f(%^f(\f($^f(fD(D\fI*XT$ Y$D^$Y$XfI*YfA(Xf(A^E^\f(f(A^DXfA(A^\f(f(A^E^\f(f(A^\f(f(A\\A^f(f(A^A^\f(f(A^\A^\f(f(A^\A^\f(A^\^$A^A^AXA^XXf/$D$Y\f(XQH,f.H,ffUH*f(f(fT\fVf(|$H\l$@f.ztYl$Ld$0$d$8l$pY|$H=l$Xd$hl$t$@d$Pl$L$`|$ f(\$ $(OT$\$ $=fATIUHSH@D$t H9YT$=fLDžI*\f(L$|$ML$YL$ LT$L$ \$D$YfYXWf.QY%Xf/;H,HH'Md$1f/f(w>[Lf\H)HH*YT$YfH*YT$^f/vHPH9}HLL$1f/wH@[]A\f.H\$\$=f(T$8d$0L$(\$ +Md$0Y%T$8L$(\$ Xf/f(T$ LT$ H,fDfH*f/r%Yf/r IfDSKSH\Yf(Hf/rUIH)H[DKH)H[ff.UHS1HfWdJ 4D$@HH $K $Yf/L$wHH[]@ATf(QUHSHPD$ff.4$>D$E1KD$0$YoXof(D$Yef(\ i\5yf(]\%E^ if(d$8Xl$(X5D$@7^\L$HfHJHf(\$I$ D$f(fT\D$(^ $XD$YXD$XK $\$f/ H,r|$Hf/H]=f/v f/Ef(L$ I$D$@IL$ t$8D$D$Y^X]I$XT$HCfH*YL$0\f\L$H*f.AEĄu1f.D„uL$$L$$\f/~HPH[]A\f(iI@f/$sf.zu1ABff.Sf(HH "f/vT\L$f(GHD$DL$ff(f.QwfXf(YXD$H [Y T$f(IBfT$H HH[H*XGfDLFD$f(T$PH\$T$tff.@SHH $f(D$@$$Hf(Yf( $F\$ $H[Y^f(DSH\^f(@H[VAfDSHH?CHD$1CL$H[^f(ff.fSHH D$BL$HY hD$f(L$AL$ff(f.Qw"f.L$YQw5f(H ^[D$f(T$Ff\$T$f(T$L$FT$L$ff.@SHH@D$8L$ f/t$ f/^^D$^Xft$D$0/fDf(^f(oEXD$L$\f/D$s}HEY/D\$0Hf(YXXL$f(L$ ^\d$(YL$DL$\Y\f/T$ZHDD$D$(@ f/L$vfWT$8 Xf(fTkT$Xe?|$T$\Lf/vfWvH@[l$ :=Y|$Yfl$Xf.Qw{XL$t$f(Xf.Qws\t$ f(f(X^f(YXXL$^L$0dsCX\_YH@[L$.DL$oL$(T$DL$(T$kHD$>^D$"B\HDHD$> ^L$HFCfDHD$}>fWA f(^L$H\f(Bff.fHL$$XB `f/vXBYD$X$HÐ \\f(WBYD$$$H\f(ff.@H $D$A\f(BfWA $T$HY\f(f.HL$$xAh\^AL$YX$Hff.@H#;HJ@f.H$A \f(9AYqff.Qw$HYL$AL$ff(f(SHXf(H ^L$\$,$=\$L$Y f(YYYf(YXff.QwW\Y$H\$X$,@\$$f(f(X^f/s Y^f(H f([\$T$d$@\$T$d${UHSH(\f($@f(ؐH\$?HD$t?%d\d$D$-H^,$f(f(?QAf $=(H,H*^T$X?T$YT$f(\ \$^Yf(\ ^f/;H2H(H[]HD$\$>L$$f/vf(fYHXf/wHfH$>>.\f(i>\$D$f(M>L$^f(z8HH,f/r ?D<AWIAVIAUIATIUHSH)HH(H9ffH*t$f/l$I/L$HDHf/L$vKLL$N=fL$H*f(^X>?L$H,fH*\H9uD$\H,M9}I)LH([]A\A]A^A_ffH*l$f/|$@D$\H,D$\H,ff.AWHffAVIffAUH*ATUHSHHXH9Ht$8HNHH9H*IHHT$@H4$LMI)I9HL$0LOH*$I*L)f(\$ ^f(d$HY\XD$fH*HCYYfH*Yf^XQf.d$H$MefHSD$YXLxI*D$fI*YfH*^V=H,fHSH*f.Eф9f. Dф#>f(LfH)H*f.Eф f.Dф T$(T$(f(I)fXʹI*f.Eфf.jDфL$(L$(IXfM)J3H*f.&E„Pf.D„:L$(?L$(XL$(L9,$XT$HD$Y!XD$%f(fTf.v3H,f5fUH*f(f(fT\fV]T$HX9H$K9\SYD$f^$XD$f/wf/D$s+;H,fHCH*f.E„f.D„f(LfH)H*f.E„f.D„T$ T$ f(MfXʺI)ID$H*f.ME„'f.7D„L$ fL$ J3XfH*f.E„f.D„L$ L$ X,$T$(\\Y\ f/s;f($\Yf/{f(7 $Xf/HD$8H9D$@ILH)L9l$0HOHXH[]A\A]A^A_DfZfff(XT$ Dfffdf7ff.fH ~03ff.f(f(H8\\T$L$D$(f(\$^l$ 4$(64$\$L$T$f/r3l$ fd$(YYf.QwHXH8f(@f(\ \fYYf.Qw*\H8f(d$$q6d$$f(T$ $P6T$ $SHH -DD$\f(y5D$H+5f/D$$H5YD$'4 $\f(Yf/rXf(T$5T$$f(5 $^X6H,HjH [f/ʸrHH__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) while calling a Python objectNULL result without error in PyObject_CallCannot convert %.200s to %.200s%.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 %zd%.200s() keywords must be strings%s() got an unexpected keyword argument '%U'%s() got multiple values for keyword argument '%U'calling %R should have returned an instance of BaseException, not %Rraise: exception class must be a subclass of BaseException'%.200s' object is unsliceable'%.200s' object is not subscriptablecannot fit '%.200s' into an index-sized integertoo many values to unpack (expected %zd)can't convert negative value to npy_uint64can't convert negative value to size_tcan't convert negative value to unsigned longcan't convert negative value to npy_uint16value too large to convert to npy_uint16can't convert negative value to npy_boolvalue too large to convert to npy_boolvalue too large to convert to npy_int16can't convert negative value to npy_uint8value too large to convert to npy_uint8value too large to convert to npy_int8can't convert negative value to npy_uint32value too large to convert to npy_uint32value too large to convert to intvalue too large to convert to npy_int32mtrand.RandomState.__getstate__mtrand.RandomState.__setstate__%.200s() takes %.8s %zd positional argument%.1s (%zd given)need more than %zd value%.1s to unpack'%.200s' object does not support slice %.10smtrand.RandomState.multivariate_normalmtrand.RandomState.multinomialmtrand.RandomState.permutationmtrand.RandomState.random_integersmtrand.RandomState.negative_binomialmtrand.RandomState.standard_cauchymtrand.RandomState.standard_exponentialmtrand.RandomState.standard_normalmtrand.RandomState.random_samplemtrand.RandomState.standard_gammamtrand.RandomState.exponentialmtrand.RandomState.noncentral_chisquaremtrand.RandomState.noncentral_fmtrand.RandomState.hypergeometriccompiletime version %s of module '%.100s' does not match runtime version %snumpy.core.multiarray failed to import_ARRAY_API is not PyCapsule objectmodule compiled against ABI version 0x%x but this version of numpy is 0x%xmodule compiled against API version 0x%x but this version of numpy is 0x%xFATAL: module compiled as unknown endianFATAL: module compiled as little endian, but detected different endianness at runtime RandomState(seed=None) Container for the Mersenne Twister pseudo-random number generator. `RandomState` exposes a number of methods for generating random numbers drawn from a variety of probability distributions. In addition to the distribution-specific arguments, each method takes a keyword argument `size` that defaults to ``None``. If `size` is ``None``, then a single value is generated and returned. If `size` is an integer, then a 1-D array filled with generated values is returned. If `size` is a tuple, then an array with that shape is filled and returned. *Compatibility Guarantee* A fixed seed and a fixed series of calls to 'RandomState' methods using the same parameters will always produce the same results up to roundoff error except when the values were incorrect. Incorrect values will be fixed and the NumPy version in which the fix was made will be noted in the relevant docstring. Extension of existing parameter ranges and the addition of new parameters is allowed as long the previous behavior remains unchanged. Parameters ---------- seed : {None, int, array_like}, optional Random seed used to initialize the pseudo-random number generator. Can be any integer between 0 and 2**32 - 1 inclusive, an array (or other sequence) of such integers, or ``None`` (the default). If `seed` is ``None``, then `RandomState` will try to read data from ``/dev/urandom`` (or the Windows analogue) if available or seed from the clock otherwise. Notes ----- The Python stdlib module "random" also contains a Mersenne Twister pseudo-random number generator with a number of methods that are similar to the ones available in `RandomState`. `RandomState`, besides being NumPy-aware, has the advantage that it provides a much larger number of probability distributions to choose from. an integer is requiredMissing type objectname '%U' is not definedcannot import name %Snumpy/random/mtrand/mtrand.c%s (%s:%d)mtrand.pyxmtrand.RandomState.__reduce__at leastat mostmtrand.RandomState.__init__mtrand.RandomState.bytesmtrand.RandomState.set_stateexactlymtrand._shape_from_sizemtrand.cont2_arrayassignmentdeletionmtrand.RandomState.choicemtrand.cont3_arrayrandint_helpers.pximtrand._rand_uint16mtrand._rand_int32mtrand._rand_int8mtrand._rand_int16mtrand._rand_uint32mtrand._rand_uint8mtrand._rand_uint64multivariate_normalfloat divisionmtrand.RandomState.dirichletmtrand.discd_arraymtrand.RandomState.randnmtrand.RandomState.randmtrand.RandomState.randintmultinomialmtrand.RandomState.shufflemtrand._rand_int64mtrand.RandomState.get_statemtrand.disc0_arraymtrand.RandomState.tomaxintrandom_integersmtrand.discd_array_scmtrand.RandomState.logseriesmtrand.RandomState.geometricmtrand.RandomState.poissonmtrand.discnp_array_scmtrand.RandomState.binomialmtrand.discnp_arraymtrand.discdd_array_scmtrand.discdd_arraynegative_binomialmtrand.cont1_arraymtrand.RandomState.zipfmtrand.cont0_arraystandard_cauchystandard_exponentialstandard_normalrandom_samplemtrand.cont1_array_scmtrand.RandomState.rayleighmtrand.RandomState.powermtrand.RandomState.weibullmtrand.RandomState.paretomtrand.RandomState.standard_tmtrand.RandomState.chisquarestandard_gammamtrand.cont2_array_scmtrand.RandomState.waldmtrand.RandomState.lognormalmtrand.RandomState.logisticmtrand.RandomState.gumbelmtrand.RandomState.laplacemtrand.RandomState.vonmisesnoncentral_chisquaremtrand.RandomState.fmtrand.RandomState.gammamtrand.RandomState.betamtrand.RandomState.normalmtrand.RandomState.uniformmtrand.cont3_array_scmtrand.RandomState.triangularnoncentral_fmtrand._rand_boolmtrand.RandomState.seedmtrand.discnmN_array_scmtrand.discnmN_arrayhypergeometric%d.%d%sbuiltinscython_runtime__builtins__21474836484294967295429496729618446744073709551616-9223372036854775808__name__numpy.pxdndarraynumpyflatiterbroadcasttype.pxdnumpy.core.multiarray_ARRAY_API_ARRAY_API not found_ARRAY_API is NULL pointermtrand.import_arrayinit mtranddtypemtrand.RandomState__getstate____setstate__permutationPx@( $!& D  (@{oUp8&"$'4': <;;;[la^^^pxolq qldtrqq{{'|| |tކՆ̆4f]T ,h6-$dr TyHy 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() 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() 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() 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() 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() 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) 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= 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 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() 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]]) 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)') 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() numpy.core.multiarray failed to import 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() 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_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() 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] 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]) 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() 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() 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! 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() 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 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() 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. 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') 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]) 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 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%. Unsupported dtype "%s" for randintSeed must be between 0 and 2**32 - 1RandomState.standard_gamma (line 1810)RandomState.multivariate_normal (line 4369)RandomState.logseries (line 4272)RandomState.lognormal (line 3302)RandomState.hypergeometric (line 4150)RandomState.geometric (line 4082)RandomState.dirichlet (line 4643)RandomState.chisquare (line 2196) 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() 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) 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)) 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() 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]]) 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]]) 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 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)) 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]]) 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) 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") RandomState.vonmises (line 2551)RandomState.rayleigh (line 3426)RandomState.logistic (line 3209)RandomState.binomial (line 3686)probabilities do not sum to 1RandomState.weibull (line 2759)RandomState.uniform (line 1210)RandomState.tomaxint (line 858)RandomState.shuffle (line 4759)RandomState.poisson (line 3903)RandomState.laplace (line 2980)RandomState.randint (line 905)RandomState.pareto (line 2649)RandomState.normal (line 1547)RandomState.gumbel (line 3078)RandomState.choice (line 1028)high is out of bounds for %sa and p must have same sizeRandomState.randn (line 1360)RandomState.power (line 2869)RandomState.gamma (line 1896)mean must be 1 dimensionallow is out of bounds for %sRange exceeds valid boundsRandomState.zipf (line 3991)RandomState.wald (line 3505)RandomState.rand (line 1316)RandomState.bytes (line 999)algorithm must be 'MT19937'a must be greater than 0state must be 624 longsp must be 1-dimensionala must be 1-dimensionalRandomState.f (line 1992)lam value too large.standard_exponentialnoncentral_chisquarerandint_helpers.pximultivariate_normallam value too largea must be non-emptyngood + nbad < nsamplecline_in_tracebackDeprecationWarningnegative_binomial__RandomState_ctorsum(pvals[:-1]) > 1.0standard_normalstandard_cauchy_shape_from_sizerandom_integerspoisson_lam_maxdummy_threadingstandard_gammahypergeometricRuntimeWarningrandom_samplegreater_equalcount_nonzeroOverflowErrorsearchsortedreturn_index_randint_typenoncentral_fscale <= 0.0_rand_uint64_rand_uint32_rand_uint16permutationmultinomialexponentialcheck_validImportErrortriangularstandard_t_rand_uint8_rand_int64_rand_int32_rand_int16__pyx_vtable__numpy.dualmtrand.pyxmode > rightlogical_orless_equalleft == rightissubdtypeempty_likearray_dataValueErrorthreadingsigma < 0.0set_statescale <= 0scale < 0.0_rand_int8_rand_boolnsample < 1logserieslognormalleft > modeget_stategeometricdirichletchisquarebroadcastTypeErrorwarningsvonmisessubtractrngstatereversedrayleighp is nanoperatormean <= 0.0low >= highlogisticitemsizeisfinitefloatingbinomialallcloseweibulluniformstridessignbitsigma < 0shuffleshape < 0scale < 0reshapereplacerandintpoissonp >= 1.0p <= 0.0nsamplengood < 0ndarraylaplacekappa < 0integergreaterfloat64dfnum <= 1dfnum <= 0dfden <= 0castingasarrayMT19937unsafeuniqueuint64uint32uint16reducerandom_randparetonormalnonc < 0nbad < 0mtrandmean <= 0__import__ignoregumbelformatcumsumctypeschoiceboolastypearangezerosuint8statesigmashapescalerightravelrangerandnraisepvalspowerp > 1.0p < 0.0numpyngoodn <= 0lam < 0kappaisnanint64int32int16indexiinfogammafinfoequal__enter__emptydtypedfnumdfdenbytesarrayalphaa <= 1.0a < 0zipfwarnwalduint__test__takesqrtsortsizesideseedsafertolrandprodnoncndimnbadnamemodemean__main__longlessleftitemintpint8high__exit__df <= 0datacopybool_betaatolLocktolsvdrngp > 1p < 0outoffn < 0maxlowloclamintepsdotcovcntbufb <= 0anyalladda <= 0npmudfpnlfdbaTL??:0yE>rb/dev/random/dev/urandomno errorrandom device unvavailableA<UUUUUU?llfJ?88C$+K?<ٰj_AAz?SˆB8?5gG8?5gG@dg??UUUUUU?"@m{?(\@ffffff@0C.@4@x&?@?UUUUUU?a@X@`@|@@MA$@>@= ףp=@n?[ m?h|?5?333333 @r?$~?@B>٬ @r鷯?Q?Q?9v?-DT! @h㈵>@-DT!@3?r?q?0@;  =(GPP R0hp@p@TPp(p4 px 0 pH  0 l  ` H @ p \`  TP  8 \p`(ph$%h0+2 >XJm|dp(s|````DdH}P P\`x 3\`FYm\ @l ,! |!3!7,"G|" "P,###@$0$$%@`% %`&Pp&@&)0':'PK'@`P(u()p))`0***$`+B+P_@,,-p-- ..P7H/@O/P|<000@4PT4h4@4 5 x555$6<6p666 7P7@777 8P,8D8\8t8@8p880949`9909:4:@H:@x:::::p (; @; X; p; ; ;P ;p ; ;<@<@X<p<<< = 4=0X=zRx $6 FJ w?:*3$"D@ \`Hx H 4xEFDD a GBH AAB4yEDD E CAJ K CAA D(`=4AAD D AAD O AAG 4HAAD ~ AAJ O AAG BBA A(D0T (A ABBD h KBA A(G0p (A ABBD B (F ABBC D(G ABBP0(xdmADG e DAJ 0PAGD ` DAG NDA(pADD ^ DAD 4AAG \ DAF N DAE \<`hBBI D(D0 (D ABBD N (D ABBI \ (C DBBI `pBAA D0  DABK |  AABH t  AABH `  CABA  ,%8 HBBA A(D@` (A ABBH P _BBE E(D0C8D` 8A0A(B BBBF D 8A0A(B BBBJ hTpBxBBAAAAAI`d lSBA A(G0 (D ABBG pP0T (I ABBN <H 0AD h AJ F AA J AM G AH < 0AD h AJ F AA J AM G AH < 0AD h AJ F AA J AM G AH  yD Y AA , IAD x AB P LAD x AB (t IAD z AH @ AG LAD x AB ( 8LAD z AH @ AG ( \iAD z AH  AH < AD A AI  AH ^ AI G AH <\ AD A AI  AH ^ AI G AH ( `yAD0 AAH L  BBA A(D@ (D ABBD ~ (G ABBF @ t WFBB A(A0D@ 0A(A BBBI @\ $FBE A(A0D`q 0A(A BBBD H |uFBB B(D0A8DP 8A0A(B BBBG \ l FBB B(A0D8DuH@Au 8A0A(B BBBF HL) FEB B(A0A8Dp 8A0A(B BBBG L$6?"FBB B(A0D8DZ 8A0A(B BBBG lXFBB B(A0A8GVFAr 8A0A(B BBBK  WfALXdg'BBE B(A0A8G 8A0A(B BBBA pDBBB E(A0A8Js 8A0A(B BBBD 6 MUBsFYAPpZFBE A(D0D` 0A(A BBBK nhHp@hA`Lp<]BBB E(A0A8G 8A0A(B BBBJ wyFBB B(A0A8GVFAu 8A0A(B BBBH ' WfAVgNJWgSTx#FBB B(A0A8GgVFAu 8A0A(B BBBH  WfAVgN]WgSMFBB B(A0A8GVFAu 8A0A(B BBBH  WfAVgNMWgS|вQFBB B(A0A8GVFAu 8A0A(B BBBH  VhSWgSHWfMFBB B(A0A8GVFAu 8A0A(B BBBH  WfAVgN]WgSyFBB B(A0A8GVFAu 8A0A(B BBBH / WfAVgNJWgS8FBB B(A0D8DVFAv 8A0A(B BBBG  WfA$VgN]WgSl*GFBB B(A0D8GDSA{ 8A0A(B BBBF :&WfA\<HFBE B(A0D8G* H@A{ 8A0A(B BBBE `bBBE E(A0A8DxOBBM 8A0A(B BBBK L>+3 H X  Р } ئ++o`@ +X(i oo(ood&o(+ 0@P`pС 0@P`pТ 0@P`pУ 0@P`pФ 0@P`pХ 0@P`pЦ 0@P`pЧ 0@P`pШ 0@P`pЩ 0@P`pЪ 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]]) 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. ---------p----------(-----(------(-----------------(-P---------(---(---(--p---(--p-(---H-P--------8--------P----p-8-----h--(------@----@----@----@----@----@----@----@----@---x-` Ap-W h- &`-Y X-Ha P-` H-^ @-X 8-R !0-U (- " - T -W - "-`V -T - "-T -@ '-S -@R !- "- "- $- ,- *-` -- %-S -S - $-`S x-T p-U h-S `-T X- (P-` %H-R !@-@S 8- (0- -(-@ ' - (-` #- S - #-S -Q !-`U -R -@U - U -X - %-Fa - G- L-[ - #-Z -Da -(a -_ x-_ p-`T h-@V `- &X-U P-@W H-$a @-U 8- a 0-\ (-_  -a -^ -_ -Z -^ -^ -` -Ba -a -` -\ - -` -^ -[ -a -_ -@ e-^ -Y -@ 4x-[ p-` h-^ `- s X-W P- a H-` @-X 8- a 0- %(- ) -^ -^ -@a -` -5a -` -_ -] -_ -] -] -[ -I -a -_ -X -_ -Z -_ -a -_ x-` p-Y h->a `-  X-_ P-] H-\ @-^ 8-_ 0- (-[  - <-[ -] -X -^ - -` -@T -X -` 8-^ -_ -^ -~_ -` -x_ -r_ -l_ -` -] -}` x-\ p-f_ h-Z `-x` X-\ P-`_ H-] @-  r-ɓ 0n-đ h-{ p Ј` t `*@c- ]-t 0 T- @S-( pH- =-\ 9-C @2-  '-d '#-r @-Z -X  -O %-= @,A `,, Х@,ȗ r ,3 p@, ,? P"@, , t, }, h,Ǖ 0S`|, Yo,Ɩ 0?], $R, A, =`9,u P`/, pJ&, ,  ,u `(`,' +X +, B `+; +ݒ `+| ;+ + U+ P+ݑ В jt- Cx-] 0@}- 0W- -q  - 0/- `-ܗ -a a GCC: (GNU) 8.5.0 20210514 (Red Hat 8.5.0-16)GA$3a1Р} GA$3p1067R GA*GA$annobin gcc 8.5.0 20210514GA$plugin name: annobinGA$running gcc 8.5.0 20210514GA*GA*GA! GA*FORTIFYGA+GLIBCXX_ASSERTIONS GA*GOW*GA*cf_protectionGA+omit_frame_pointerGA+stack_clashGA!stack_realign GA$3p1067E } GA*GA$annobin gcc 8.5.0 20210514GA$plugin name: annobinGA$running gcc 8.5.0 20210514GA*GA*GA! GA*FORTIFYGA+GLIBCXX_ASSERTIONS GA*GOW*GA*cf_protectionGA+omit_frame_pointerGA+stack_clashGA!stack_realignGA+GLIBCXX_ASSERTIONSS GA*GOW*SGA*cf_protectionGA+omit_frame_pointerGA+stack_clashGA!stack_realign GA*FORTIFYS GA*SGA*GA*GA! > E ) >? >Y t > > > > > >- @?[ >`|-- @? ? @?E> ?p @ ?y @ 8@ @(0 8@T }@v @@= }@ A @ A A5 AM Av B Aw B B# B~K Bs 3C B 3C C @Cv% C_ mD C mD E( pD$M Ev E E E G E G2 HS Gg H 2I H 2I  I%  @I;  Ie  J  J  J  4M  Jd  6  Sa    4M  N  @Ma  N-  OT  N"n -v  O  P  O  P?  Tu  P  T  \V  T#  \VK  ]Wq  `V  ]W  X  `We' Xc Z X Z [ Z> [T ] [ ] -^ ]m- -^E ^l 0^P- ^ ^ ^p ^  s`9 ^Y s` a `h a f a* fP 5ft f% 5f g @f g  /m@ g_S-d-yX-- /m n 0m n8 p_ n0y p Pq p0 Pq r- Pq0J rr s r s t sI tB  ۚY> +>- > ۚ> [? {=?-M?-]? [? ? `;*@(-;@8-L@0-]@-m@ -~@-@-@-@ @ 2A +eA- |A A A A -A0- B(-B NB pB B-B-B-B p C B"4C pPCH-_CP-nC B"C $C P"D-4D $yD 'D %D- E 'IE _*E 'E`-E _* F -IF `*xFp-F -F =F -G8-G@-&G =_G 0SG =@G-G -G-G- H(-H 0SQH hH 0SH-H -H-H hI }SI h|I`-I0-I(-I }I !J }IJ@-ZJ@-qJ8-J J pJ "K-/Kh-?K-VK`-fK pK K pL-L-*L-:L yL L L-L- M-M YM PM @M-M(-M -M P%N BQN PpN-N-N BN N PvO-.O-?O@- VO-gO-xO O $O P-%P-6P-GP- ^P $P /?P $P-Q-Q-(Q- ?Q /?vQ YQ 0?Q-Q-Q- R YDR tzR YR-R-R- R tS xLS tvS(-SP-SX-SH-S - S xT ХST PT-T-T-Tx-T- Tp-T Х!U QU ХPtU-U-U-U-U-U@- U-U !V UV k(|V-V-V - V-V-V W ;W PaW-mWH-}W@-W8-W- W0-W W 0X XXX-hXP-xX- X X n9X &Y-9Y8-GYx-ZYh-jY-Y - Y`-Y n9Y pJZ p9&Z-6Z-FZ pJZ rZ pJ3(Z-Z-[`-([H-+[-<[-M[-^[-o[-[x-[ r[  \ rc(7\`-(N\-^\-n\-~\-\-\-\ \ ] !]-0]@-(G] {] ] -]-](-]0-^ -^-(^0-7^-I^-Z^-j^-z^`-^-^ ^ A -_ ae_p-u_H-_-_x-_-_-_-_-_`-_- `-`-+`h-;` A z` E ` B `@-(`- aP-a-/a SLa >ga-hwa-8a-a-a-a-a-a-a-a-a- b-b-=b-\b-nbx-bp-bh-b-b`-b-b-c-!c-5c@-Mcp-jc-c-cX-c8-cX-c-d-BdX-Nd@-sd -d-d`-dX-dh-eP-h0-Oh(-nh -hx-h-h-h-h-hH-h-i--i-@iP-fi0-iX-i-i(-j-1j(-?j-Tj(-djP-rj-j-jH-j@-j-j-j0-j(- k -"k-7k-Mk-ck-yk0-k-k-k-k-k`-l-&l8-7l@- Vl - ul- l- l- l- l- m`- :m@- \m@-lm-m-m - m-mx-mH-mh-n-n-"n-2n-Lnh-]n-lnp-nh-nX-n8-nh-nx-nx-o- ox-:o8-Oox-bo-soH-o-o-o-o-o-o-oh-p-p-)p-Ep-Wp0-fp(-zp-p-p-p -p-p-qP-Eq-pq-q-qh-q-r0-=r`-hr -r-r-r8-sp-4s-^s`-s-sX-s@-t -2t`-]t-t`-t0-tP-u(-0u0-[u-u`-u-up-u-'vp-Rv(-}v-v-vp-v8-*w0-Uw-wp-w-w-x-)x@-Tx-}x-xx-x-x-"y-My-uy-y-y-y- z-KzH-uz`-z-z-z-{-G{8-q{-{-{-{-|(-F|-l| -|-|-|-}-C}-n}-}-}-}-~-E~-p~-~@-~-~--(` AQW l &Y Ha ` ^ X R ! U 0 "X T }W  "`V ހT  "*T O@ 'wS @R !ā " " $= ,e *` - %S S * $S`S yT U S T   (2` %[R !@S  (ф -@ '" (K` #t S  #ÅS Q !`U 3R Y@U | U X  %ކFa  G L<[ N #vZ Da (a _ _ `T ߇@V  &(U I@W e$a qU  a \ _ a ʈ^ و_ Z ^  ^ ` &Ba 0a <` I\ Z ` ^ [ a _ ʉ@ e^ Y @ 4?[ Q` y^  s W ˊ a ׊` X  a  %/ )X^ g^ v@a ` 5a ` _ ] Ë_ ы] ] [ I *a 6_ DX \_ jZ }_ a _ ` Y ƌ>a Ќ  _ ] \ '^ 6_ D l[ ~ <[ ] ȍX ލ^  ` "@T FX ]` 8^ _ -^ Ž~_ Ў` ܎x_ r_ l_ ` ] #}` 0\ Af_ OZ bx` o\ `_ ] ^ LV l] |\ X `  ` ` ˝Z ݝ`  `Y $ M L_ Y<^ h5^ w.^ ^ \  @ ̞'^ ۞ ^ \  #_ 0@ W_ d[ u\ @ ^ _ ȟ y - -  r-Fn-/qh-2@c-_̠]-`T- !@S-eIH-s r=-@ 9-á@2-5 '- #-qO@-Y x-> -͢-@, 5`, ]@,  , @,ܣ,, @,E, r, ,Ƥ, `|,b o,r B],kR,$ A,¥`9,y`/, &,M A, l , `,ʦ+y +<`+8N+ z`+&+ ާ+ +A6+et-x-@}-Ȩ-- -&-E`-}c- E  YQ  >ǩ > > >$ >A >b >| > E  F ƪ F ߪ RG  RG  H 4 H T H r H  I  I ̫ nJ  nJ  J  pJ +  J @ K ^ K u K  K  aL  aL ׬ L  L  nM  nM 4 M L M i O  O  P  P έ YQ  `Q  R  >+ >H >f > > >Ʈ > > `Q  R 6 R O } l > > >̯ > > >8 >V >x R  S  R Cb Pư S ߰ T  T  OT B OT ` nT | nT  T  T ϱ W  W  2W  2W 3 X H X d X ~ X  WY  WY IJ .d  .d  jf * jf E f ^ f } \g  \g  i ֳ i  j  j / k T k s kk  kk  k ִ k  k  k 5 l P l k ?o  ?o  ko  ko ε o  o  o  o . ap F ap _ p v p  q  q ƶ 6q  6q  q  q + r @ r W s l s  t  t ַ pt  pt  t 0 t U u x u  { ¸ {  {  {  | : | V } p >r P> >-+ѹ >ݹئ+L  } ++&c 9p-E+[ Р8`@d&(X( Р    } ~ c i P ئ++++++- mav }  Y WǺܺ { `- @q ^ q  p A' M 5N\-x ϻ޻ g \ j 7CN`t H  S ,ż u ܼ  W " `g |"6MZ_p-f i -q r y L  po *Ƚܽ H  q ' O O2Ie o Cn T m; pk *z  k * T /B[k {  F 2ƿ˿߿ J ! P I* t 5@ pt M[l K  pp V0 pT ,; l GSax `G  K  0d : #5I\ pM ~g o qr pf  @W  t `/D pJ +M_ `Q 9m{$ - X $ `Y 5 S͇C @o +M\p-hxǿ pL R&;?Pb} k [Y I ~ ( E W0ANco  . s WBO`"| k 3 PT .annobin_mtrand.c.annobin_mtrand.c_end.annobin_mtrand.c.hot.annobin_mtrand.c_end.hot.annobin_mtrand.c.unlikely.annobin_mtrand.c_end.unlikely.annobin_mtrand.c.startup.annobin_mtrand.c_end.startup.annobin_mtrand.c.exit.annobin_mtrand.c_end.exit.annobin___pyx_tp_new_6mtrand_RandomState.start.annobin___pyx_tp_new_6mtrand_RandomState.end__pyx_tp_new_6mtrand_RandomState__pyx_vtabptr_6mtrand_RandomState__pyx_empty_tuple.annobin___pyx_tp_traverse_6mtrand_RandomState.start.annobin___pyx_tp_traverse_6mtrand_RandomState.end__pyx_tp_traverse_6mtrand_RandomState.annobin___pyx_tp_clear_6mtrand_RandomState.start.annobin___pyx_tp_clear_6mtrand_RandomState.end__pyx_tp_clear_6mtrand_RandomState.annobin___Pyx_PyCFunction_FastCall.start.annobin___Pyx_PyCFunction_FastCall.end__Pyx_PyCFunction_FastCall.annobin___Pyx__ExceptionSave.start.annobin___Pyx__ExceptionSave.end__Pyx__ExceptionSave.annobin___Pyx__ExceptionReset.start.annobin___Pyx__ExceptionReset.end__Pyx__ExceptionReset.annobin___Pyx_ErrRestoreInState.start.annobin___Pyx_ErrRestoreInState.end__Pyx_ErrRestoreInState.annobin___pyx_bisect_code_objects.start.annobin___pyx_bisect_code_objects.end__pyx_bisect_code_objects.annobin___Pyx_PyNumber_IntOrLongWrongResultType.start.annobin___Pyx_PyNumber_IntOrLongWrongResultType.end__Pyx_PyNumber_IntOrLongWrongResultType.annobin___Pyx_PyNumber_IntOrLong.start.annobin___Pyx_PyNumber_IntOrLong.end__Pyx_PyNumber_IntOrLong.annobin___Pyx_PyObject_LookupSpecial.start.annobin___Pyx_PyObject_LookupSpecial.end__Pyx_PyObject_LookupSpecial.annobin___pyx_f_6mtrand_11RandomState__shuffle_raw.start.annobin___pyx_f_6mtrand_11RandomState__shuffle_raw.end__pyx_f_6mtrand_11RandomState__shuffle_raw.annobin___pyx_tp_dealloc_6mtrand_RandomState.start.annobin___pyx_tp_dealloc_6mtrand_RandomState.end__pyx_tp_dealloc_6mtrand_RandomState.annobin___Pyx_PyObject_GetAttrStr.start.annobin___Pyx_PyObject_GetAttrStr.end__Pyx_PyObject_GetAttrStr.annobin___Pyx__GetException.start.annobin___Pyx__GetException.end__Pyx__GetException.annobin___Pyx_PyObject_Call.start.annobin___Pyx_PyObject_Call.end__Pyx_PyObject_Call.annobin___Pyx_PyFunction_FastCallNoKw.start.annobin___Pyx_PyFunction_FastCallNoKw.end__Pyx_PyFunction_FastCallNoKw.annobin___Pyx_TypeTest.isra.1.start.annobin___Pyx_TypeTest.isra.1.end__Pyx_TypeTest.isra.1.annobin___Pyx_PyInt_EqObjC.isra.12.start.annobin___Pyx_PyInt_EqObjC.isra.12.end__Pyx_PyInt_EqObjC.isra.12.annobin___Pyx_PyUnicode_Equals.start.annobin___Pyx_PyUnicode_Equals.end__Pyx_PyUnicode_Equals.annobin___Pyx_ImportType.constprop.77.start.annobin___Pyx_ImportType.constprop.77.end__Pyx_ImportType.constprop.77.annobin___Pyx_SetItemInt_Fast.constprop.79.start.annobin___Pyx_SetItemInt_Fast.constprop.79.end__Pyx_SetItemInt_Fast.constprop.79.annobin___Pyx_Import.constprop.80.start.annobin___Pyx_Import.constprop.80.end__Pyx_Import.constprop.80__pyx_m.annobin___Pyx_CheckKeywordStrings.constprop.82.start.annobin___Pyx_CheckKeywordStrings.constprop.82.end__Pyx_CheckKeywordStrings.constprop.82.annobin___Pyx_ParseOptionalKeywords.constprop.83.start.annobin___Pyx_ParseOptionalKeywords.constprop.83.end__Pyx_ParseOptionalKeywords.constprop.83.annobin___Pyx_GetItemInt_Fast.constprop.85.start.annobin___Pyx_GetItemInt_Fast.constprop.85.end__Pyx_GetItemInt_Fast.constprop.85.annobin___Pyx_Raise.constprop.93.start.annobin___Pyx_Raise.constprop.93.end__Pyx_Raise.constprop.93.annobin___Pyx_PyFunction_FastCallDict.constprop.94.start.annobin___Pyx_PyFunction_FastCallDict.constprop.94.end__Pyx_PyFunction_FastCallDict.constprop.94.annobin___Pyx_PyObject_GetSlice.isra.14.constprop.95.start.annobin___Pyx_PyObject_GetSlice.isra.14.constprop.95.end__Pyx_PyObject_GetSlice.isra.14.constprop.95.annobin___Pyx_IterFinish.start.annobin___Pyx_IterFinish.end__Pyx_IterFinish.annobin___Pyx_PyErr_ExceptionMatchesInState.isra.19.start.annobin___Pyx_PyErr_ExceptionMatchesInState.isra.19.end__Pyx_PyErr_ExceptionMatchesInState.isra.19.annobin___Pyx_GetBuiltinName.start.annobin___Pyx_GetBuiltinName.end__Pyx_GetBuiltinName__pyx_b.annobin___Pyx_GetModuleGlobalName.start.annobin___Pyx_GetModuleGlobalName.end__Pyx_GetModuleGlobalName__pyx_d.annobin___Pyx_ImportFrom.start.annobin___Pyx_ImportFrom.end__Pyx_ImportFrom.annobin___Pyx_PyInt_As_npy_intp.part.10.start.annobin___Pyx_PyInt_As_npy_intp.part.10.end__Pyx_PyInt_As_npy_intp.part.10.annobin___Pyx__PyObject_CallOneArg.start.annobin___Pyx__PyObject_CallOneArg.end__Pyx__PyObject_CallOneArg.annobin___Pyx_PyObject_GetIndex.start.annobin___Pyx_PyObject_GetIndex.end__Pyx_PyObject_GetIndex.annobin___Pyx_PyObject_GetItem.start.annobin___Pyx_PyObject_GetItem.end__Pyx_PyObject_GetItem.annobin___Pyx_IternextUnpackEndCheck.start.annobin___Pyx_IternextUnpackEndCheck.end__Pyx_IternextUnpackEndCheck.annobin___Pyx_AddTraceback.start.annobin___Pyx_AddTraceback.end__Pyx_AddTraceback__pyx_code_cache__pyx_cython_runtime__pyx_n_s_cline_in_traceback__pyx_empty_bytes.annobin___Pyx_PyObject_CallNoArg.start.annobin___Pyx_PyObject_CallNoArg.end__Pyx_PyObject_CallNoArg.annobin___Pyx_PyInt_As_npy_uint64.start.annobin___Pyx_PyInt_As_npy_uint64.end__Pyx_PyInt_As_npy_uint64.annobin___Pyx_PyInt_As_size_t.start.annobin___Pyx_PyInt_As_size_t.end__Pyx_PyInt_As_size_t.annobin___Pyx_PyInt_As_unsigned_long.start.annobin___Pyx_PyInt_As_unsigned_long.end__Pyx_PyInt_As_unsigned_long.annobin___Pyx_PyInt_As_npy_int64.start.annobin___Pyx_PyInt_As_npy_int64.end__Pyx_PyInt_As_npy_int64.annobin___Pyx_PyInt_As_npy_uint16.start.annobin___Pyx_PyInt_As_npy_uint16.end__Pyx_PyInt_As_npy_uint16.annobin___Pyx_PyInt_As_npy_bool.start.annobin___Pyx_PyInt_As_npy_bool.end__Pyx_PyInt_As_npy_bool.annobin___Pyx_PyInt_As_npy_int16.start.annobin___Pyx_PyInt_As_npy_int16.end__Pyx_PyInt_As_npy_int16.annobin___Pyx_PyInt_As_npy_uint8.start.annobin___Pyx_PyInt_As_npy_uint8.end__Pyx_PyInt_As_npy_uint8.annobin___Pyx_PyInt_As_npy_int8.start.annobin___Pyx_PyInt_As_npy_int8.end__Pyx_PyInt_As_npy_int8.annobin___Pyx_PyInt_As_npy_uint32.start.annobin___Pyx_PyInt_As_npy_uint32.end__Pyx_PyInt_As_npy_uint32.annobin___Pyx_PyInt_As_int.start.annobin___Pyx_PyInt_As_int.end__Pyx_PyInt_As_int.annobin___Pyx_PyInt_As_npy_int32.start.annobin___Pyx_PyInt_As_npy_int32.end__Pyx_PyInt_As_npy_int32.annobin___Pyx_PyInt_As_npy_intp.start.annobin___Pyx_PyInt_As_npy_intp.end__Pyx_PyInt_As_npy_intp.annobin___Pyx_PyObject_CallOneArg.start.annobin___Pyx_PyObject_CallOneArg.end__Pyx_PyObject_CallOneArg.annobin___pyx_pw_6mtrand_11RandomState_11__getstate__.start.annobin___pyx_pw_6mtrand_11RandomState_11__getstate__.end__pyx_pw_6mtrand_11RandomState_11__getstate____pyx_n_s_get_state__pyx_lineno__pyx_filename__pyx_clineno.annobin___pyx_pw_6mtrand_11RandomState_13__setstate__.start.annobin___pyx_pw_6mtrand_11RandomState_13__setstate__.end__pyx_pw_6mtrand_11RandomState_13__setstate____pyx_n_s_set_state.annobin___pyx_pw_6mtrand_11RandomState_15__reduce__.start.annobin___pyx_pw_6mtrand_11RandomState_15__reduce__.end__pyx_pw_6mtrand_11RandomState_15__reduce____pyx_n_s_np__pyx_n_s_random__pyx_n_s_RandomState_ctor.annobin___pyx_pw_6mtrand_11RandomState_1__init__.start.annobin___pyx_pw_6mtrand_11RandomState_1__init__.end__pyx_pw_6mtrand_11RandomState_1__init____pyx_n_s_Lock__pyx_n_s_seed__pyx_pyargnames.21379.annobin___pyx_pw_6mtrand_11RandomState_23bytes.start.annobin___pyx_pw_6mtrand_11RandomState_23bytes.end__pyx_pw_6mtrand_11RandomState_23bytes__pyx_n_s_exit__pyx_n_s_enter__pyx_tuple__52.annobin___pyx_pw_6mtrand_11RandomState_9set_state.start.annobin___pyx_pw_6mtrand_11RandomState_9set_state.end__pyx_pw_6mtrand_11RandomState_9set_state__pyx_n_s_MT19937__pyx_slice__44__pyx_tuple__43__pyx_builtin_ValueError__pyx_slice__45PyArray_API__pyx_tuple__47__pyx_int_0__pyx_float_0_0__pyx_builtin_TypeError__pyx_tuple__46.annobin___pyx_pw_6mtrand_19_shape_from_size.start.annobin___pyx_pw_6mtrand_19_shape_from_size.end__pyx_pw_6mtrand_19_shape_from_size__pyx_n_s_operator__pyx_n_s_index__pyx_n_s_size__pyx_n_s_d__pyx_pyargnames.21277.annobin___pyx_f_6mtrand_cont2_array.start.annobin___pyx_f_6mtrand_cont2_array.end__pyx_f_6mtrand_cont2_array__pyx_n_s_empty__pyx_n_s_float64__pyx_n_s_dtype__pyx_n_s_broadcast__pyx_n_s_shape__pyx_tuple__12__pyx_tuple__11.annobin___pyx_pf_6mtrand_11RandomState_24choice.start.annobin___pyx_pf_6mtrand_11RandomState_24choice.end__pyx_pf_6mtrand_11RandomState_24choice__pyx_n_s_array__pyx_n_s_copy__pyx_n_s_ndim__pyx_n_s_item__pyx_tuple__54__pyx_int_1__pyx_n_s_sqrt__pyx_n_s_finfo__pyx_n_s_eps__pyx_n_s_ndarray__pyx_n_s_logical_or__pyx_n_s_reduce__pyx_tuple__60__pyx_n_s_prod__pyx_n_s_intp__pyx_n_s_cumsum__pyx_n_s_random_sample__pyx_n_s_searchsorted__pyx_n_s_right__pyx_n_s_side__pyx_tuple__64__pyx_tuple__53__pyx_n_s_count_nonzero__pyx_n_s_zeros__pyx_n_s_int__pyx_n_s_ravel__pyx_n_s_rand__pyx_n_s_unique__pyx_n_s_return_index__pyx_n_s_sort__pyx_n_s_take__pyx_tuple__55__pyx_n_s_randint__pyx_tuple__56__pyx_n_s_issubdtype__pyx_n_s_floating__pyx_n_s_permutation__pyx_tuple__63__pyx_tuple__61__pyx_ptype_6mtrand_ndarray__pyx_tuple__57__pyx_tuple__58__pyx_tuple__62__pyx_tuple__59.annobin___pyx_pw_6mtrand_11RandomState_25choice.start.annobin___pyx_pw_6mtrand_11RandomState_25choice.end__pyx_pw_6mtrand_11RandomState_25choice__pyx_n_s_a__pyx_n_s_p__pyx_n_s_replace__pyx_pyargnames.22381.annobin___pyx_f_6mtrand_cont3_array.start.annobin___pyx_f_6mtrand_cont3_array.end__pyx_f_6mtrand_cont3_array__pyx_tuple__16__pyx_tuple__15.annobin___pyx_pw_6mtrand_13_rand_uint16.start.annobin___pyx_pw_6mtrand_13_rand_uint16.end__pyx_pw_6mtrand_13_rand_uint16__pyx_n_s_uint16__pyx_n_s_low__pyx_n_s_rngstate__pyx_n_s_high__pyx_pyargnames.19262.annobin___pyx_pw_6mtrand_7_rand_int32.start.annobin___pyx_pw_6mtrand_7_rand_int32.end__pyx_pw_6mtrand_7_rand_int32__pyx_n_s_int32__pyx_pyargnames.18921.annobin___pyx_pw_6mtrand_3_rand_int8.start.annobin___pyx_pw_6mtrand_3_rand_int8.end__pyx_pw_6mtrand_3_rand_int8__pyx_n_s_int8__pyx_pyargnames.18693.annobin___pyx_pw_6mtrand_5_rand_int16.start.annobin___pyx_pw_6mtrand_5_rand_int16.end__pyx_pw_6mtrand_5_rand_int16__pyx_n_s_int16__pyx_pyargnames.18807.annobin___pyx_pw_6mtrand_15_rand_uint32.start.annobin___pyx_pw_6mtrand_15_rand_uint32.end__pyx_pw_6mtrand_15_rand_uint32__pyx_n_s_uint32__pyx_pyargnames.19375.annobin___pyx_pw_6mtrand_11_rand_uint8.start.annobin___pyx_pw_6mtrand_11_rand_uint8.end__pyx_pw_6mtrand_11_rand_uint8__pyx_n_s_uint8__pyx_pyargnames.19149.annobin___pyx_pw_6mtrand_17_rand_uint64.start.annobin___pyx_pw_6mtrand_17_rand_uint64.end__pyx_pw_6mtrand_17_rand_uint64__pyx_n_s_uint64__pyx_pyargnames.19488.annobin___pyx_pw_6mtrand_11RandomState_97multivariate_normal.start.annobin___pyx_pw_6mtrand_11RandomState_97multivariate_normal.end__pyx_pw_6mtrand_11RandomState_97multivariate_normal__pyx_n_s_warn__pyx_float_1eneg_8__pyx_n_s_svd__pyx_n_s_numpy_dual__pyx_n_s_integer__pyx_tuple__163__pyx_n_s_mean__pyx_n_s_cov__pyx_n_s_check_valid__pyx_n_s_tol__pyx_pyargnames.27672__pyx_slice__165__pyx_n_s_standard_normal__pyx_n_s_reshape__pyx_int_neg_1__pyx_n_s_ignore__pyx_n_s_raise__pyx_n_s_allclose__pyx_n_s_dot__pyx_n_s_T__pyx_n_s_rtol__pyx_n_s_atol__pyx_tuple__170__pyx_tuple__162__pyx_tuple__164__pyx_tuple__166__pyx_n_s_warnings__pyx_tuple__167__pyx_tuple__168.annobin___pyx_pw_6mtrand_11RandomState_101dirichlet.start.annobin___pyx_pw_6mtrand_11RandomState_101dirichlet.end__pyx_pw_6mtrand_11RandomState_101dirichlet__pyx_n_s_shape_from_size__pyx_tuple__173__pyx_n_s_alpha__pyx_pyargnames.28085.annobin___pyx_f_6mtrand_discd_array.start.annobin___pyx_f_6mtrand_discd_array.end__pyx_f_6mtrand_discd_array__pyx_tuple__35__pyx_tuple__33__pyx_tuple__34.annobin___pyx_pw_6mtrand_11RandomState_31randn.start.annobin___pyx_pw_6mtrand_11RandomState_31randn.end__pyx_pw_6mtrand_11RandomState_31randn.annobin___pyx_pw_6mtrand_11RandomState_29rand.start.annobin___pyx_pw_6mtrand_11RandomState_29rand.end__pyx_pw_6mtrand_11RandomState_29rand.annobin___pyx_pw_6mtrand_11RandomState_21randint.start.annobin___pyx_pw_6mtrand_11RandomState_21randint.end__pyx_pw_6mtrand_11RandomState_21randint__pyx_k__48__pyx_n_s_name__pyx_n_s_randint_type__pyx_tuple__51__pyx_kp_s_Unsupported_dtype_s_for_randint__pyx_kp_s_low_is_out_of_bounds_for_s__pyx_n_s_bool_2__pyx_n_s_long__pyx_kp_s_high_is_out_of_bounds_for_s__pyx_tuple__49__pyx_pyargnames.22112.annobin___pyx_pw_6mtrand_11RandomState_99multinomial.start.annobin___pyx_pw_6mtrand_11RandomState_99multinomial.end__pyx_pw_6mtrand_11RandomState_99multinomial__pyx_tuple__172__pyx_tuple__171__pyx_n_s_n__pyx_n_s_pvals__pyx_pyargnames.27941.annobin___pyx_pw_6mtrand_11RandomState_103shuffle.start.annobin___pyx_pw_6mtrand_11RandomState_103shuffle.end__pyx_pw_6mtrand_11RandomState_103shuffle__pyx_tuple__176__pyx_n_s_ctypes__pyx_n_s_data__pyx_n_s_strides__pyx_n_s_itemsize__pyx_tuple__174__pyx_n_s_empty_like__pyx_tuple__175.annobin___pyx_pw_6mtrand_9_rand_int64.start.annobin___pyx_pw_6mtrand_9_rand_int64.end__pyx_pw_6mtrand_9_rand_int64__pyx_n_s_int64__pyx_pyargnames.19035.annobin___pyx_pw_6mtrand_11RandomState_105permutation.start.annobin___pyx_pw_6mtrand_11RandomState_105permutation.end__pyx_pw_6mtrand_11RandomState_105permutation__pyx_n_s_arange__pyx_n_s_shuffle.annobin___pyx_pw_6mtrand_11RandomState_7get_state.start.annobin___pyx_pw_6mtrand_11RandomState_7get_state.end__pyx_pw_6mtrand_11RandomState_7get_state__pyx_n_s_uint__pyx_int_624__pyx_tuple__42__pyx_n_s_asarray.annobin___pyx_pw_6mtrand_11RandomState_19tomaxint.start.annobin___pyx_pw_6mtrand_11RandomState_19tomaxint.end__pyx_pw_6mtrand_11RandomState_19tomaxint__pyx_tuple__17__pyx_tuple__18__pyx_pyargnames.22064.annobin___pyx_pw_6mtrand_11RandomState_33random_integers.start.annobin___pyx_pw_6mtrand_11RandomState_33random_integers.end__pyx_pw_6mtrand_11RandomState_33random_integers__pyx_kp_s_This_function_is_deprecated_Plea_2__pyx_n_s_format__pyx_builtin_DeprecationWarning__pyx_n_s_l__pyx_kp_s_This_function_is_deprecated_Plea__pyx_pyargnames.23158.annobin___pyx_f_6mtrand_discd_array_sc.start.annobin___pyx_f_6mtrand_discd_array_sc.end__pyx_f_6mtrand_discd_array_sc__pyx_tuple__32__pyx_tuple__31.annobin___pyx_pw_6mtrand_11RandomState_95logseries.start.annobin___pyx_pw_6mtrand_11RandomState_95logseries.end__pyx_pw_6mtrand_11RandomState_95logseries__pyx_n_s_any__pyx_n_s_less_equal__pyx_n_s_greater_equal__pyx_float_1_0__pyx_tuple__158__pyx_pyargnames.27528__pyx_tuple__159__pyx_tuple__160__pyx_tuple__161.annobin___pyx_pw_6mtrand_11RandomState_91geometric.start.annobin___pyx_pw_6mtrand_11RandomState_91geometric.end__pyx_pw_6mtrand_11RandomState_91geometric__pyx_n_s_less__pyx_n_s_greater__pyx_tuple__146__pyx_pyargnames.27130__pyx_tuple__147__pyx_tuple__148__pyx_tuple__149.annobin___pyx_pw_6mtrand_11RandomState_87poisson.start.annobin___pyx_pw_6mtrand_11RandomState_87poisson.end__pyx_pw_6mtrand_11RandomState_87poisson__pyx_n_s_poisson_lam_max__pyx_pyargnames.26868__pyx_tuple__140__pyx_tuple__141__pyx_tuple__142__pyx_n_s_lam__pyx_tuple__143.annobin___pyx_f_6mtrand_discnp_array_sc.constprop.89.start.annobin___pyx_f_6mtrand_discnp_array_sc.constprop.89.end__pyx_f_6mtrand_discnp_array_sc.constprop.89__pyx_tuple__20__pyx_tuple__19.annobin___pyx_pf_6mtrand_11RandomState_82binomial.isra.43.start.annobin___pyx_pf_6mtrand_11RandomState_82binomial.isra.43.end__pyx_pf_6mtrand_11RandomState_82binomial.isra.43__pyx_n_s_isnan__pyx_tuple__129__pyx_tuple__127__pyx_tuple__128__pyx_tuple__22__pyx_tuple__131__pyx_tuple__130__pyx_tuple__132__pyx_tuple__133__pyx_tuple__21.annobin___pyx_pw_6mtrand_11RandomState_83binomial.start.annobin___pyx_pw_6mtrand_11RandomState_83binomial.end__pyx_pw_6mtrand_11RandomState_83binomial__pyx_pyargnames.26473.annobin___pyx_f_6mtrand_discdd_array_sc.constprop.87.start.annobin___pyx_f_6mtrand_discdd_array_sc.constprop.87.end__pyx_f_6mtrand_discdd_array_sc.constprop.87__pyx_tuple__24__pyx_tuple__23.annobin___pyx_pf_6mtrand_11RandomState_84negative_binomial.isra.42.start.annobin___pyx_pf_6mtrand_11RandomState_84negative_binomial.isra.42.end__pyx_pf_6mtrand_11RandomState_84negative_binomial.isra.42__pyx_tuple__136__pyx_tuple__134__pyx_tuple__135__pyx_tuple__26__pyx_tuple__137__pyx_tuple__138__pyx_tuple__139__pyx_tuple__25.annobin___pyx_pw_6mtrand_11RandomState_85negative_binomial.start.annobin___pyx_pw_6mtrand_11RandomState_85negative_binomial.end__pyx_pw_6mtrand_11RandomState_85negative_binomial__pyx_pyargnames.26680.annobin___pyx_f_6mtrand_cont1_array.start.annobin___pyx_f_6mtrand_cont1_array.end__pyx_f_6mtrand_cont1_array__pyx_tuple__8__pyx_tuple__6__pyx_tuple__7.annobin___pyx_pw_6mtrand_11RandomState_89zipf.start.annobin___pyx_pw_6mtrand_11RandomState_89zipf.end__pyx_pw_6mtrand_11RandomState_89zipf__pyx_tuple__144__pyx_pyargnames.27019__pyx_tuple__145.annobin___pyx_f_6mtrand_cont0_array.start.annobin___pyx_f_6mtrand_cont0_array.end__pyx_f_6mtrand_cont0_array__pyx_tuple__3__pyx_tuple__2.annobin___pyx_pw_6mtrand_11RandomState_57standard_cauchy.start.annobin___pyx_pw_6mtrand_11RandomState_57standard_cauchy.end__pyx_pw_6mtrand_11RandomState_57standard_cauchy__pyx_pyargnames.24747.annobin___pyx_pw_6mtrand_11RandomState_43standard_exponential.start.annobin___pyx_pw_6mtrand_11RandomState_43standard_exponential.end__pyx_pw_6mtrand_11RandomState_43standard_exponential__pyx_pyargnames.23748.annobin___pyx_pw_6mtrand_11RandomState_35standard_normal.start.annobin___pyx_pw_6mtrand_11RandomState_35standard_normal.end__pyx_pw_6mtrand_11RandomState_35standard_normal__pyx_pyargnames.23268.annobin___pyx_pw_6mtrand_11RandomState_17random_sample.start.annobin___pyx_pw_6mtrand_11RandomState_17random_sample.end__pyx_pw_6mtrand_11RandomState_17random_sample__pyx_pyargnames.22019.annobin___pyx_f_6mtrand_cont1_array_sc.start.annobin___pyx_f_6mtrand_cont1_array_sc.end__pyx_f_6mtrand_cont1_array_sc__pyx_tuple__5__pyx_tuple__4.annobin___pyx_pw_6mtrand_11RandomState_77rayleigh.start.annobin___pyx_pw_6mtrand_11RandomState_77rayleigh.end__pyx_pw_6mtrand_11RandomState_77rayleigh__pyx_n_s_signbit__pyx_pyargnames.25980__pyx_tuple__115__pyx_tuple__116__pyx_n_s_scale.annobin___pyx_pw_6mtrand_11RandomState_67power.start.annobin___pyx_pw_6mtrand_11RandomState_67power.end__pyx_pw_6mtrand_11RandomState_67power__pyx_pyargnames.25270__pyx_tuple__105__pyx_tuple__106.annobin___pyx_pw_6mtrand_11RandomState_65weibull.start.annobin___pyx_pw_6mtrand_11RandomState_65weibull.end__pyx_pw_6mtrand_11RandomState_65weibull__pyx_pyargnames.25144__pyx_tuple__103__pyx_tuple__104.annobin___pyx_pw_6mtrand_11RandomState_63pareto.start.annobin___pyx_pw_6mtrand_11RandomState_63pareto.end__pyx_pw_6mtrand_11RandomState_63pareto__pyx_tuple__101__pyx_pyargnames.25033__pyx_tuple__102.annobin___pyx_pw_6mtrand_11RandomState_59standard_t.start.annobin___pyx_pw_6mtrand_11RandomState_59standard_t.end__pyx_pw_6mtrand_11RandomState_59standard_t__pyx_n_s_df__pyx_tuple__97__pyx_pyargnames.24793__pyx_tuple__98.annobin___pyx_pw_6mtrand_11RandomState_53chisquare.start.annobin___pyx_pw_6mtrand_11RandomState_53chisquare.end__pyx_pw_6mtrand_11RandomState_53chisquare__pyx_tuple__91__pyx_pyargnames.24478__pyx_tuple__92.annobin___pyx_pw_6mtrand_11RandomState_45standard_gamma.start.annobin___pyx_pw_6mtrand_11RandomState_45standard_gamma.end__pyx_pw_6mtrand_11RandomState_45standard_gamma__pyx_pyargnames.23794__pyx_tuple__75__pyx_tuple__76.annobin___pyx_pw_6mtrand_11RandomState_41exponential.start.annobin___pyx_pw_6mtrand_11RandomState_41exponential.end__pyx_pw_6mtrand_11RandomState_41exponential__pyx_pyargnames.23621__pyx_tuple__73__pyx_tuple__74.annobin___pyx_f_6mtrand_cont2_array_sc.start.annobin___pyx_f_6mtrand_cont2_array_sc.end__pyx_f_6mtrand_cont2_array_sc__pyx_tuple__10__pyx_tuple__9.annobin___pyx_pw_6mtrand_11RandomState_79wald.start.annobin___pyx_pw_6mtrand_11RandomState_79wald.end__pyx_pw_6mtrand_11RandomState_79wald__pyx_tuple__117__pyx_tuple__118__pyx_pyargnames.26109__pyx_tuple__119__pyx_tuple__120.annobin___pyx_pw_6mtrand_11RandomState_75lognormal.start.annobin___pyx_pw_6mtrand_11RandomState_75lognormal.end__pyx_pw_6mtrand_11RandomState_75lognormal__pyx_n_s_sigma__pyx_tuple__113__pyx_tuple__114__pyx_pyargnames.25835.annobin___pyx_pw_6mtrand_11RandomState_73logistic.start.annobin___pyx_pw_6mtrand_11RandomState_73logistic.end__pyx_pw_6mtrand_11RandomState_73logistic__pyx_n_s_loc__pyx_tuple__111__pyx_tuple__112__pyx_pyargnames.25689.annobin___pyx_pw_6mtrand_11RandomState_71gumbel.start.annobin___pyx_pw_6mtrand_11RandomState_71gumbel.end__pyx_pw_6mtrand_11RandomState_71gumbel__pyx_tuple__109__pyx_tuple__110__pyx_pyargnames.25543.annobin___pyx_pw_6mtrand_11RandomState_69laplace.start.annobin___pyx_pw_6mtrand_11RandomState_69laplace.end__pyx_pw_6mtrand_11RandomState_69laplace__pyx_tuple__107__pyx_tuple__108__pyx_pyargnames.25397.annobin___pyx_pw_6mtrand_11RandomState_61vonmises.start.annobin___pyx_pw_6mtrand_11RandomState_61vonmises.end__pyx_pw_6mtrand_11RandomState_61vonmises__pyx_n_s_mu__pyx_n_s_kappa__pyx_tuple__99__pyx_tuple__100__pyx_pyargnames.24905.annobin___pyx_pw_6mtrand_11RandomState_55noncentral_chisquare.start.annobin___pyx_pw_6mtrand_11RandomState_55noncentral_chisquare.end__pyx_pw_6mtrand_11RandomState_55noncentral_chisquare__pyx_n_s_nonc__pyx_tuple__93__pyx_tuple__94__pyx_tuple__95__pyx_pyargnames.24590__pyx_tuple__96.annobin___pyx_pw_6mtrand_11RandomState_49f.start.annobin___pyx_pw_6mtrand_11RandomState_49f.end__pyx_pw_6mtrand_11RandomState_49f__pyx_n_s_dfnum__pyx_n_s_dfden__pyx_tuple__81__pyx_tuple__82__pyx_tuple__83__pyx_pyargnames.24113__pyx_tuple__84.annobin___pyx_pw_6mtrand_11RandomState_47gamma.start.annobin___pyx_pw_6mtrand_11RandomState_47gamma.end__pyx_pw_6mtrand_11RandomState_47gamma__pyx_tuple__77__pyx_tuple__79__pyx_pyargnames.23921__pyx_tuple__78__pyx_tuple__80.annobin___pyx_pw_6mtrand_11RandomState_39beta.start.annobin___pyx_pw_6mtrand_11RandomState_39beta.end__pyx_pw_6mtrand_11RandomState_39beta__pyx_n_s_b__pyx_tuple__69__pyx_tuple__70__pyx_tuple__71__pyx_pyargnames.23463__pyx_tuple__72.annobin___pyx_pw_6mtrand_11RandomState_37normal.start.annobin___pyx_pw_6mtrand_11RandomState_37normal.end__pyx_pw_6mtrand_11RandomState_37normal__pyx_tuple__67__pyx_tuple__68__pyx_pyargnames.23315.annobin___pyx_pw_6mtrand_11RandomState_27uniform.start.annobin___pyx_pw_6mtrand_11RandomState_27uniform.end__pyx_pw_6mtrand_11RandomState_27uniform__pyx_n_s_subtract__pyx_n_s_all__pyx_n_s_isfinite__pyx_tuple__65__pyx_builtin_OverflowError__pyx_pyargnames.22905__pyx_tuple__66.annobin___pyx_f_6mtrand_cont3_array_sc.start.annobin___pyx_f_6mtrand_cont3_array_sc.end__pyx_f_6mtrand_cont3_array_sc__pyx_tuple__14__pyx_tuple__13.annobin___pyx_pw_6mtrand_11RandomState_81triangular.start.annobin___pyx_pw_6mtrand_11RandomState_81triangular.end__pyx_pw_6mtrand_11RandomState_81triangular__pyx_n_s_equal__pyx_n_s_left__pyx_pyargnames.26267__pyx_n_s_mode__pyx_tuple__121__pyx_tuple__122__pyx_tuple__123__pyx_tuple__124__pyx_tuple__125__pyx_tuple__126.annobin___pyx_pw_6mtrand_11RandomState_51noncentral_f.start.annobin___pyx_pw_6mtrand_11RandomState_51noncentral_f.end__pyx_pw_6mtrand_11RandomState_51noncentral_f__pyx_pyargnames.24273__pyx_tuple__85__pyx_tuple__86__pyx_tuple__87__pyx_tuple__88__pyx_tuple__89__pyx_tuple__90.annobin___pyx_pw_6mtrand_1_rand_bool.start.annobin___pyx_pw_6mtrand_1_rand_bool.end__pyx_pw_6mtrand_1_rand_bool__pyx_n_s_bool__pyx_pyargnames.18580.annobin___pyx_pw_6mtrand_11RandomState_5seed.start.annobin___pyx_pw_6mtrand_11RandomState_5seed.end__pyx_pw_6mtrand_11RandomState_5seed__pyx_int_4294967295__pyx_tuple__37__pyx_tuple__36__pyx_tuple__38__pyx_n_s_astype__pyx_n_s_safe__pyx_n_s_casting__pyx_n_s_unsafe__pyx_tuple__40__pyx_tuple__41__pyx_pyargnames.21455__pyx_tuple__39.annobin___pyx_pf_6mtrand_11RandomState_92hypergeometric.isra.39.start.annobin___pyx_pf_6mtrand_11RandomState_92hypergeometric.isra.39.end__pyx_pf_6mtrand_11RandomState_92hypergeometric.isra.39__pyx_tuple__28__pyx_n_s_add__pyx_tuple__150__pyx_tuple__27__pyx_tuple__151__pyx_tuple__152__pyx_tuple__153__pyx_tuple__154__pyx_tuple__30__pyx_tuple__155__pyx_tuple__156__pyx_tuple__157__pyx_tuple__29.annobin___pyx_pw_6mtrand_11RandomState_93hypergeometric.start.annobin___pyx_pw_6mtrand_11RandomState_93hypergeometric.end__pyx_pw_6mtrand_11RandomState_93hypergeometric__pyx_pyargnames.27273__pyx_n_s_nbad__pyx_n_s_nsample__pyx_n_s_ngood.annobin_PyInit_mtrand.start.annobin_PyInit_mtrand.end__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_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_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_k_Cannot_take_a_larger_sample_than__pyx_k_DeprecationWarning__pyx_k_Fewer_non_zero_entries_in_p_than__pyx_k_ImportError__pyx_k_L__pyx_k_Lock__pyx_k_MT19937__pyx_k_OverflowError__pyx_k_RandomState_binomial_line_3686__pyx_k_RandomState_bytes_line_999__pyx_k_RandomState_chisquare_line_2196__pyx_k_RandomState_choice_line_1028__pyx_k_RandomState_ctor__pyx_k_RandomState_dirichlet_line_4643__pyx_k_RandomState_f_line_1992__pyx_k_RandomState_gamma_line_1896__pyx_k_RandomState_geometric_line_4082__pyx_k_RandomState_gumbel_line_3078__pyx_k_RandomState_hypergeometric_line__pyx_k_RandomState_laplace_line_2980__pyx_k_RandomState_logistic_line_3209__pyx_k_RandomState_lognormal_line_3302__pyx_k_RandomState_logseries_line_4272__pyx_k_RandomState_multinomial_line_453__pyx_k_RandomState_multivariate_normal__pyx_k_RandomState_negative_binomial_li__pyx_k_RandomState_noncentral_chisquare__pyx_k_RandomState_noncentral_f_line_20__pyx_k_RandomState_normal_line_1547__pyx_k_RandomState_pareto_line_2649__pyx_k_RandomState_permutation_line_484__pyx_k_RandomState_poisson_line_3903__pyx_k_RandomState_power_line_2869__pyx_k_RandomState_rand_line_1316__pyx_k_RandomState_randint_line_905__pyx_k_RandomState_randn_line_1360__pyx_k_RandomState_random_integers_line__pyx_k_RandomState_random_sample_line_8__pyx_k_RandomState_rayleigh_line_3426__pyx_k_RandomState_shuffle_line_4759__pyx_k_RandomState_standard_cauchy_line__pyx_k_RandomState_standard_exponential__pyx_k_RandomState_standard_gamma_line__pyx_k_RandomState_standard_normal_line__pyx_k_RandomState_standard_t_line_2445__pyx_k_RandomState_tomaxint_line_858__pyx_k_RandomState_triangular_line_3592__pyx_k_RandomState_uniform_line_1210__pyx_k_RandomState_vonmises_line_2551__pyx_k_RandomState_wald_line_3505__pyx_k_RandomState_weibull_line_2759__pyx_k_RandomState_zipf_line_3991__pyx_k_Range_exceeds_valid_bounds__pyx_k_RuntimeWarning__pyx_k_Seed_must_be_between_0_and_2_32__pyx_k_T__pyx_k_This_function_is_deprecated_Plea__pyx_k_This_function_is_deprecated_Plea_2__pyx_k_TypeError__pyx_k_Unsupported_dtype_s_for_randint__pyx_k_ValueError__pyx_k_a__pyx_k_a_0__pyx_k_a_0_2__pyx_k_a_1_0__pyx_k_a_and_p_must_have_same_size__pyx_k_a_must_be_1_dimensional__pyx_k_a_must_be_1_dimensional_or_an_in__pyx_k_a_must_be_greater_than_0__pyx_k_a_must_be_non_empty__pyx_k_add__pyx_k_algorithm_must_be_MT19937__pyx_k_all__pyx_k_allclose__pyx_k_alpha__pyx_k_any__pyx_k_arange__pyx_k_array__pyx_k_array_data__pyx_k_asarray__pyx_k_astype__pyx_k_atol__pyx_k_b__pyx_k_b_0__pyx_k_beta__pyx_k_binomial__pyx_k_binomial_n_p_size_None_Draw_sam__pyx_k_bool__pyx_k_bool_2__pyx_k_broadcast__pyx_k_buf__pyx_k_bytes__pyx_k_bytes_length_Return_random_byte__pyx_k_casting__pyx_k_check_valid__pyx_k_check_valid_must_equal_warn_rais__pyx_k_chisquare__pyx_k_chisquare_df_size_None_Draw_sam__pyx_k_choice__pyx_k_choice_a_size_None_replace_True__pyx_k_cline_in_traceback__pyx_k_cnt__pyx_k_copy__pyx_k_count_nonzero__pyx_k_cov__pyx_k_cov_must_be_2_dimensional_and_sq__pyx_k_covariance_is_not_positive_semid__pyx_k_ctypes__pyx_k_cumsum__pyx_k_d__pyx_k_data__pyx_k_df__pyx_k_df_0__pyx_k_dfden__pyx_k_dfden_0__pyx_k_dfnum__pyx_k_dfnum_0__pyx_k_dfnum_1__pyx_k_dirichlet__pyx_k_dirichlet_alpha_size_None_Draw__pyx_k_dot__pyx_k_dtype__pyx_k_dummy_threading__pyx_k_empty__pyx_k_empty_like__pyx_k_enter__pyx_k_eps__pyx_k_equal__pyx_k_exit__pyx_k_exponential__pyx_k_f__pyx_k_f_dfnum_dfden_size_None_Draw_sa__pyx_k_finfo__pyx_k_float64__pyx_k_floating__pyx_k_format__pyx_k_gamma__pyx_k_gamma_shape_scale_1_0_size_None__pyx_k_geometric__pyx_k_geometric_p_size_None_Draw_samp__pyx_k_get_state__pyx_k_greater__pyx_k_greater_equal__pyx_k_gumbel__pyx_k_gumbel_loc_0_0_scale_1_0_size_N__pyx_k_high__pyx_k_high_is_out_of_bounds_for_s__pyx_k_hypergeometric__pyx_k_hypergeometric_ngood_nbad_nsamp__pyx_k_ignore__pyx_k_iinfo__pyx_n_s_import__pyx_k_import__pyx_k_index__pyx_k_int__pyx_k_int16__pyx_k_int32__pyx_k_int64__pyx_k_int8__pyx_k_integer__pyx_k_intp__pyx_k_isfinite__pyx_k_isnan__pyx_k_issubdtype__pyx_k_item__pyx_k_itemsize__pyx_k_kappa__pyx_k_kappa_0__pyx_k_l__pyx_k_lam__pyx_k_lam_0__pyx_k_lam_value_too_large__pyx_k_lam_value_too_large_2__pyx_k_laplace__pyx_k_laplace_loc_0_0_scale_1_0_size__pyx_k_left__pyx_k_left_mode__pyx_k_left_right__pyx_k_less__pyx_k_less_equal__pyx_k_loc__pyx_k_logical_or__pyx_k_logistic__pyx_k_logistic_loc_0_0_scale_1_0_size__pyx_k_lognormal__pyx_k_lognormal_mean_0_0_sigma_1_0_si__pyx_k_logseries__pyx_k_logseries_p_size_None_Draw_samp__pyx_k_long__pyx_k_low__pyx_k_low_high__pyx_k_low_is_out_of_bounds_for_s__pyx_k_main__pyx_k_max__pyx_k_mean__pyx_k_mean_0__pyx_k_mean_0_0__pyx_k_mean_and_cov_must_have_same_leng__pyx_k_mean_must_be_1_dimensional__pyx_k_mode__pyx_k_mode_right__pyx_k_mtrand__pyx_k_mtrand_pyx__pyx_k_mu__pyx_k_multinomial__pyx_k_multinomial_n_pvals_size_None_D__pyx_k_multivariate_normal__pyx_k_multivariate_normal_mean_cov_si__pyx_k_n__pyx_k_n_0__pyx_k_n_0_2__pyx_k_name__pyx_k_nbad__pyx_k_nbad_0__pyx_k_ndarray__pyx_k_ndim__pyx_k_negative_binomial__pyx_k_negative_binomial_n_p_size_None__pyx_k_ngood__pyx_k_ngood_0__pyx_k_ngood_nbad_nsample__pyx_k_nonc__pyx_k_nonc_0__pyx_k_noncentral_chisquare__pyx_k_noncentral_chisquare_df_nonc_si__pyx_k_noncentral_f__pyx_k_noncentral_f_dfnum_dfden_nonc_s__pyx_k_normal__pyx_k_normal_loc_0_0_scale_1_0_size_N__pyx_k_np__pyx_k_nsample__pyx_k_nsample_1__pyx_k_numpy__pyx_k_numpy_core_multiarray_failed_to__pyx_k_numpy_dual__pyx_k_off__pyx_k_operator__pyx_k_out__pyx_k_p__pyx_k_p_0__pyx_k_p_0_0__pyx_k_p_0_0_2__pyx_k_p_1__pyx_k_p_1_0__pyx_k_p_1_0_2__pyx_k_p_is_nan__pyx_k_p_must_be_1_dimensional__pyx_k_pareto__pyx_k_pareto_a_size_None_Draw_samples__pyx_k_permutation__pyx_k_permutation_x_Randomly_permute__pyx_k_poisson__pyx_k_poisson_lam_1_0_size_None_Draw__pyx_k_poisson_lam_max__pyx_k_power__pyx_k_power_a_size_None_Draws_samples__pyx_k_probabilities_are_not_non_negati__pyx_k_probabilities_do_not_sum_to_1__pyx_k_prod__pyx_k_pvals__pyx_k_pyx_vtable__pyx_k_raise__pyx_k_rand__pyx_k_rand_2__pyx_k_rand_bool__pyx_k_rand_d0_d1_dn_Random_values_in__pyx_k_rand_int16__pyx_k_rand_int32__pyx_k_rand_int64__pyx_k_rand_int8__pyx_k_rand_uint16__pyx_k_rand_uint32__pyx_k_rand_uint64__pyx_k_rand_uint8__pyx_k_randint__pyx_k_randint_helpers_pxi__pyx_k_randint_low_high_None_size_None__pyx_k_randint_type__pyx_k_randn__pyx_k_randn_d0_d1_dn_Return_a_sample__pyx_k_random__pyx_k_random_integers__pyx_k_random_integers_low_high_None_s__pyx_k_random_sample__pyx_k_random_sample_size_None_Return__pyx_k_range__pyx_k_ravel__pyx_k_rayleigh__pyx_k_rayleigh_scale_1_0_size_None_Dr__pyx_k_reduce__pyx_k_replace__pyx_k_reshape__pyx_k_return_index__pyx_k_reversed__pyx_k_right__pyx_k_rng__pyx_k_rngstate__pyx_k_rtol__pyx_k_safe__pyx_k_scale__pyx_k_scale_0__pyx_k_scale_0_0__pyx_k_scale_0_0_2__pyx_k_scale_0_2__pyx_k_searchsorted__pyx_k_seed__pyx_k_set_state__pyx_k_shape__pyx_k_shape_0__pyx_k_shape_from_size__pyx_k_shuffle__pyx_k_shuffle_x_Modify_a_sequence_in__pyx_k_side__pyx_k_sigma__pyx_k_sigma_0__pyx_k_sigma_0_0__pyx_k_signbit__pyx_k_size__pyx_k_size_is_not_compatible_with_inpu__pyx_k_sort__pyx_k_sqrt__pyx_k_standard_cauchy__pyx_k_standard_cauchy_size_None_Draw__pyx_k_standard_exponential__pyx_k_standard_exponential_size_None__pyx_k_standard_gamma__pyx_k_standard_gamma_shape_size_None__pyx_k_standard_normal__pyx_k_standard_normal_size_None_Draw__pyx_k_standard_t__pyx_k_standard_t_df_size_None_Draw_sa__pyx_k_state__pyx_k_state_must_be_624_longs__pyx_k_strides__pyx_k_subtract__pyx_k_sum_pvals_1_1_0__pyx_k_svd__pyx_k_take__pyx_k_test__pyx_k_threading__pyx_k_tol__pyx_k_tomaxint_size_None_Random_integ__pyx_k_triangular__pyx_k_triangular_left_mode_right_size__pyx_k_uint__pyx_k_uint16__pyx_k_uint32__pyx_k_uint64__pyx_k_uint8__pyx_k_uniform__pyx_k_uniform_low_0_0_high_1_0_size_N__pyx_k_unique__pyx_k_unsafe__pyx_k_vonmises__pyx_k_vonmises_mu_kappa_size_None_Dra__pyx_k_wald__pyx_k_wald_mean_scale_size_None_Draw__pyx_k_warn__pyx_k_warnings__pyx_k_weibull__pyx_k_weibull_a_size_None_Draw_sample__pyx_k_zeros__pyx_k_zipf__pyx_k_zipf_a_size_None_Draw_samples_f__pyx_methods__pyx_methods_6mtrand_RandomState__pyx_doc_6mtrand_11RandomState_4seed__pyx_doc_6mtrand_11RandomState_6get_state__pyx_doc_6mtrand_11RandomState_8set_state__pyx_doc_6mtrand_11RandomState_16random_sample__pyx_doc_6mtrand_11RandomState_18tomaxint__pyx_doc_6mtrand_11RandomState_20randint__pyx_doc_6mtrand_11RandomState_22bytes__pyx_doc_6mtrand_11RandomState_24choice__pyx_doc_6mtrand_11RandomState_26uniform__pyx_doc_6mtrand_11RandomState_28rand__pyx_doc_6mtrand_11RandomState_30randn__pyx_doc_6mtrand_11RandomState_32random_integers__pyx_doc_6mtrand_11RandomState_34standard_normal__pyx_doc_6mtrand_11RandomState_36normal__pyx_doc_6mtrand_11RandomState_38beta__pyx_doc_6mtrand_11RandomState_40exponential__pyx_doc_6mtrand_11RandomState_42standard_exponential__pyx_doc_6mtrand_11RandomState_44standard_gamma__pyx_doc_6mtrand_11RandomState_46gamma__pyx_doc_6mtrand_11RandomState_48f__pyx_doc_6mtrand_11RandomState_50noncentral_f__pyx_doc_6mtrand_11RandomState_52chisquare__pyx_doc_6mtrand_11RandomState_54noncentral_chisquare__pyx_doc_6mtrand_11RandomState_56standard_cauchy__pyx_doc_6mtrand_11RandomState_58standard_t__pyx_doc_6mtrand_11RandomState_60vonmises__pyx_doc_6mtrand_11RandomState_62pareto__pyx_doc_6mtrand_11RandomState_64weibull__pyx_doc_6mtrand_11RandomState_66power__pyx_doc_6mtrand_11RandomState_68laplace__pyx_doc_6mtrand_11RandomState_70gumbel__pyx_doc_6mtrand_11RandomState_72logistic__pyx_doc_6mtrand_11RandomState_74lognormal__pyx_doc_6mtrand_11RandomState_76rayleigh__pyx_doc_6mtrand_11RandomState_78wald__pyx_doc_6mtrand_11RandomState_80triangular__pyx_doc_6mtrand_11RandomState_82binomial__pyx_doc_6mtrand_11RandomState_84negative_binomial__pyx_doc_6mtrand_11RandomState_86poisson__pyx_doc_6mtrand_11RandomState_88zipf__pyx_doc_6mtrand_11RandomState_90geometric__pyx_doc_6mtrand_11RandomState_92hypergeometric__pyx_doc_6mtrand_11RandomState_94logseries__pyx_doc_6mtrand_11RandomState_96multivariate_normal__pyx_doc_6mtrand_11RandomState_98multinomial__pyx_doc_6mtrand_11RandomState_100dirichlet__pyx_doc_6mtrand_11RandomState_102shuffle__pyx_doc_6mtrand_11RandomState_104permutation__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.annobin_randomkit.c.annobin_randomkit.c_end.annobin_randomkit.c.hot.annobin_randomkit.c_end.hot.annobin_randomkit.c.unlikely.annobin_randomkit.c_end.unlikely.annobin_randomkit.c.startup.annobin_randomkit.c_end.startup.annobin_randomkit.c.exit.annobin_randomkit.c_end.exit.annobin_rk_seed.start.annobin_rk_seed.end.annobin_rk_random.start.annobin_rk_random.end.annobin_rk_random_uint32.start.annobin_rk_random_uint32.end.annobin_rk_random_uint16.start.annobin_rk_random_uint16.end.annobin_rk_random_uint8.start.annobin_rk_random_uint8.end.annobin_rk_random_bool.start.annobin_rk_random_bool.end.annobin_rk_ulong.start.annobin_rk_ulong.endrk_uint64.annobin_rk_random_uint64.start.annobin_rk_random_uint64.end.annobin_rk_long.start.annobin_rk_long.end.annobin_rk_interval.start.annobin_rk_interval.end.annobin_rk_double.start.annobin_rk_double.end.annobin_rk_fill.start.annobin_rk_fill.end.annobin_rk_devfill.start.annobin_rk_devfill.end.annobin_rk_randomseed.start.annobin_rk_randomseed.end.annobin_rk_altfill.start.annobin_rk_altfill.end.annobin_rk_gauss.start.annobin_rk_gauss.end.annobin_initarray.c.annobin_initarray.c_end.annobin_initarray.c.hot.annobin_initarray.c_end.hot.annobin_initarray.c.unlikely.annobin_initarray.c_end.unlikely.annobin_initarray.c.startup.annobin_initarray.c_end.startup.annobin_initarray.c.exit.annobin_initarray.c_end.exit.annobin_init_by_array.start.annobin_init_by_array.end.annobin_distributions.c.annobin_distributions.c_end.annobin_distributions.c.hot.annobin_distributions.c_end.hot.annobin_distributions.c.unlikely.annobin_distributions.c_end.unlikely.annobin_distributions.c.startup.annobin_distributions.c_end.startup.annobin_distributions.c.exit.annobin_distributions.c_end.exit.annobin_loggam.part.0.start.annobin_loggam.part.0.endloggam.part.0a.15261.annobin_rk_normal.start.annobin_rk_normal.end.annobin_rk_standard_exponential.start.annobin_rk_standard_exponential.end.annobin_rk_exponential.start.annobin_rk_exponential.end.annobin_rk_uniform.start.annobin_rk_uniform.end.annobin_rk_standard_gamma.start.annobin_rk_standard_gamma.end.annobin_rk_gamma.start.annobin_rk_gamma.end.annobin_rk_beta.start.annobin_rk_beta.end.annobin_rk_chisquare.start.annobin_rk_chisquare.end.annobin_rk_f.start.annobin_rk_f.end.annobin_rk_binomial_btpe.start.annobin_rk_binomial_btpe.end.annobin_rk_binomial_inversion.start.annobin_rk_binomial_inversion.end.annobin_rk_binomial.start.annobin_rk_binomial.end.annobin_rk_poisson_mult.start.annobin_rk_poisson_mult.end.annobin_rk_poisson_ptrs.start.annobin_rk_poisson_ptrs.end.annobin_rk_poisson.start.annobin_rk_poisson.end.annobin_rk_noncentral_chisquare.start.annobin_rk_noncentral_chisquare.end.annobin_rk_noncentral_f.start.annobin_rk_noncentral_f.end.annobin_rk_negative_binomial.start.annobin_rk_negative_binomial.end.annobin_rk_standard_cauchy.start.annobin_rk_standard_cauchy.end.annobin_rk_standard_t.start.annobin_rk_standard_t.end.annobin_rk_vonmises.start.annobin_rk_vonmises.end.annobin_rk_pareto.start.annobin_rk_pareto.end.annobin_rk_weibull.start.annobin_rk_weibull.end.annobin_rk_power.start.annobin_rk_power.end.annobin_rk_laplace.start.annobin_rk_laplace.end.annobin_rk_gumbel.start.annobin_rk_gumbel.end.annobin_rk_logistic.start.annobin_rk_logistic.end.annobin_rk_lognormal.start.annobin_rk_lognormal.end.annobin_rk_rayleigh.start.annobin_rk_rayleigh.end.annobin_rk_wald.start.annobin_rk_wald.end.annobin_rk_zipf.start.annobin_rk_zipf.end.annobin_rk_geometric_search.start.annobin_rk_geometric_search.end.annobin_rk_geometric_inversion.start.annobin_rk_geometric_inversion.end.annobin_rk_geometric.start.annobin_rk_geometric.end.annobin_rk_hypergeometric_hyp.start.annobin_rk_hypergeometric_hyp.end.annobin_rk_hypergeometric_hrua.start.annobin_rk_hypergeometric_hrua.end.annobin_rk_hypergeometric.start.annobin_rk_hypergeometric.end.annobin_rk_triangular.start.annobin_rk_triangular.end.annobin_rk_logseries.start.annobin_rk_logseries.endderegister_tm_clones__do_global_dtors_auxcompleted.7303__do_global_dtors_aux_fini_array_entryframe_dummy__frame_dummy_init_array_entry__FRAME_END___fini__dso_handle_DYNAMIC__GNU_EH_FRAME_HDR__TMC_END___GLOBAL_OFFSET_TABLE__initPyUnicode_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_WarnExceil_edatark_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_binomialPyExc_ExceptionPyExc_ValueErrorrk_standard_trk_standard_exponentialPyExc_DeprecationWarningPyExc_TypeErrorgettimeofday@@GLIBC_2.2.5PyGILState_EnsurePyEval_EvalFrameExrk_hypergeometricrk_randomacosPySequence_ContainsPyErr_PrintPyMem_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_TrueStruct__bss_startPyFunction_TypePyDict_NewPyExc_IndexErrorPyLong_AsUnsignedLongPyDict_NextPyBaseObject_TypePyLong_FromUnsignedLongPyLong_TypePyCapsule_Typerk_double_PyObject_GetDictPtrPyErr_FetchPyUnicode_FromStringlogPyObject_GetIterPyEval_SaveThreadPyUnicode_InternFromStringpowPyExc_ImportErrorPyDict_SetItemPySequence_TuplePyExc_AttributeErrorrk_noncentral_fPyExc_StopIterationPySequence_Listrk_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_exponentialfloorPyTuple_PackPy_GetVersionPyObject_GC_UnTrackPyList_TypePyImport_Import.symtab.strtab.shstrtab.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.note.gnu.property.init_array.fini_array.data.rel.ro.dynamic.got.data.bss.comment.gnu.build.attributes88$.o``8 @@p@ Hod&d&Uo((@dX(X(inBxРРs ~ "} } ~ ~  c c i i p7P P ئ+ئ + + + + + p8 -p  0p -  m ,  f#