ELF>i@ @8 @PYPY```** hA   888$$Ptd0T 0T 0T QtdRtd GNU>JfVij/84\cBgl \ M!_ ( 1y1 <  gb s  w Y~  > 4 uM  i@ s  {   Y' :w V$W Jk  wc!   /~d  v  X az fj8 5 A paP% 0  a A +   J ^F8 Zq    R"B  i @2.  ` |T  ӷ __gmon_start___init_fini_ITM_deregisterTMCloneTable_ITM_registerTMCloneTable__cxa_finalize_Py_NoneStructPyBaseObject_TypePyDict_NewPyExc_TypeErrorPyErr_FormatPyDict_NextPy_EnterRecursiveCallPy_LeaveRecursiveCallPyErr_OccurredPyObject_CallPyExc_SystemErrorPyErr_SetStringmemcpyPyDict_SizePyMethod_NewPyUnicode_InternFromStringPyUnicode_FromFormatPyObject_GetAttrPyExc_DeprecationWarningPyErr_WarnFormat_Py_DeallocPyObject_GC_UnTrackPyObject_GC_IsFinalizedPyObject_CallFinalizerFromDeallocPyErr_NormalizeExceptionPyException_SetTracebackPyImport_ImportModuleLevelObjectPyObject_FreePyObject_ClearWeakRefsPyObject_GC_DelPyObject_GetAttrStringPyDict_GetItemStringPyModule_GetNamePyExc_ImportErrorPyCapsule_IsValidPyCapsule_GetNamePyCapsule_GetPointerPyDict_SetItemStringPyExc_AttributeErrorPyErr_ExceptionMatchesPyErr_ClearPyThreadState_GetPyInterpreterState_GetIDPyModule_NewObjectPyModule_GetDictPyUnicode_FromStringPyUnicode_TypememcmpPyObject_RichCompare_Py_TrueStruct_Py_FalseStructPyObject_IsTrue_PyUnicode_ReadyPyDict_GetItemWithErrorPyExc_KeyErrorPyErr_SetObjectPyTuple_PackPyTuple_GetSlicePyTuple_GetItemPyMem_MallocPyTuple_NewPyMem_FreePyErr_NoMemoryPyExc_RuntimeWarningPyErr_WarnEx_PyObject_GC_NewPyObject_GC_TrackPyExc_ValueErrorPyOS_snprintfPyList_TypePyLong_FromSsize_tPyObject_SetItemPySlice_New_PyType_LookupPyTuple_TypePyObject_GetItemPyExc_OverflowErrorPyUnicode_ComparePyUnicode_DecodePyUnicode_FromStringAndSizePyBytes_FromStringAndSizePyObject_HashPyExc_RuntimeErrorPyLong_TypePyFloat_TypePyErr_GivenExceptionMatchesPyUnicode_ConcatPyImport_GetModulePyLong_AsLongPyCFunction_TypePyObject_VectorcallDictPyList_New_PyThreadState_UncheckedGetPyExc_StopIteration_PyObject_GetDictPtrPyObject_NotPyFrame_NewPyTraceBack_HerePyObject_SetAttrPyCode_NewEmptyPyUnicode_AsUTF8memmovePyMem_ReallocPyObject_GenericGetAttr_PyObject_GenericGetAttrWithDict_PyDict_GetItem_KnownHashPyDict_SetItemPyNumber_AddPyNumber_InPlaceAddPyMethod_TypePyType_ReadyPyNumber_IndexPyLong_AsSsize_tPyExc_IndexErrorPyExc_NameErrorPyLong_FromLongPyFloat_AsDoublePyFloat_FromDoublePyImport_AddModulePyObject_SetAttrStringPy_GetVersionstrrchrPyLong_FromStringPyImport_GetModuleDict_Py_EllipsisObjectPyCode_NewWithPosOnlyArgsPyImport_ImportPyErr_FetchPyErr_RestorePyCapsule_NewmallocfreePyImport_ImportModulePyExc_ModuleNotFoundErrorPyCapsule_TypePyExc_ExceptionPyType_Modified_PyDict_NewPresizedPyDict_CopyPyObject_SizePyObject_IsInstancePyUnicode_New_PyUnicode_FastCopyCharactersPyObject_FormatPyDict_TypePySequence_ContainsPyEval_SaveThreadPyEval_RestoreThreadPySequence_TuplePyNumber_LongPyBool_TypePyUnicode_FormatPyNumber_RemainderPyNumber_MultiplyPyList_AsTuplePySequence_ListPyList_AppendPyObject_GetIterPyNumber_InPlaceTrueDividePyNumber_SubtractPyInit_mtrandPyModuleDef_Initlogexplog1pexpflog1pfpowsqrtpowflogfsqrtfexpm1floor__isnanacosfmodmemsetceillibm.so.6libc.so.6GLIBC_2.14GLIBC_2.2.5 0 ui   ui   P   0 X       @  P z  @ *H pP h 3p px  ; p  E p  N    [  0 e8 @ X n` 0 { 0  @  @  P ( PH P X `p x  ` ή   ݮ й @  h    l   0 `X ` ` 0     ( 8 p     p ( @ ǨH L` zh p  &   x     M  Ϭ Р @ S   O( 8 @ H X `` h x    e      0 @  = @    x (  8 s@ H X `j` h x @_   V  ' F u F ; ~ | /    ! ( B8 @ EH ,X  ` :h x    ` @  O  1   f p5  m  >  ( 0X8 @ H @/X ` h 0Ex `  6 {  = j g  _ , p `T  w I ]( p8 :@ TH @"X +` h  <x !  s   `   ]   i   `  ì( {8 @ ݬH X ` ph x @   @ 5   N  @ U   ( p8 @ H @/X ` h x @ p  @ ݬ   ì {   `   i  ( ]8 @ H `X ` h sx    < ! T @" + ] p :  w I , p `T g( 8 _@ H =X j` h 6x {  0E `  @/   0X  m  >  f p5  1( 8 @ H OX `  h `x @ :   E ,    B     ! ~ | / u( F8 ;@ H 'X F` h x V   @_   `j    s    x  = @ ( 08 @@ H X ` h ex        ` O   S   Ϭ Р @ ( M8 @ xH X  ` h &x   z p Ǩ L   sh s       . 6 7( 90 :8 W@ ZH [P ^X `` gp tx { ~        ȿ п ؿ       ( 0 8 @ H P  X  ` h p x                ! " # $ % & '( (0 )8 *@ +H ,P -X /` 0h 1p 2x 3 4 5 8 ; < = > ? @ A B C D E F G H I J K L( M0 N8 O@ PH QP RX S` Th Up Vx X Y [ \ ] _ a b c d e f h i j k l m n o p q( r0 u8 v@ wH xP yX z` |h }p x                      ( 0 8 @ H P X ` h p x          HH=_ HtH5_ %_ @%_ h%_ h%_ h%_ h%_ h%_ h%_ h%_ hp%_ h`%_ h P%_ h @%_ h 0%_ h %z_ h %r_ h%j_ h%b_ h%Z_ h%R_ h%J_ h%B_ h%:_ h%2_ h%*_ hp%"_ h`%_ hP%_ h@% _ h0%_ h %^ h%^ h%^ h%^ h %^ h!%^ h"%^ h#%^ h$%^ h%%^ h&%^ h'p%^ h(`%^ h)P%^ h*@%^ h+0%^ h, %z^ h-%r^ h.%j^ h/%b^ h0%Z^ h1%R^ h2%J^ h3%B^ h4%:^ h5%2^ h6%*^ h7p%"^ h8`%^ h9P%^ h:@% ^ h;0%^ h< %] h=%] h>%] h?%] h@%] hA%] hB%] hC%] hD%] hE%] hF%] hGp%] hH`%] hIP%] hJ@%] hK0%] hL %z] hM%r] hN%j] hO%b] hP%Z] hQ%R] hR%J] hS%B] hT%:] hU%2] hV%*] hWp%"] hX`%] hYP%] hZ@% ] h[0%] h\ %\ h]%\ h^%\ h_%\ h`%\ ha%\ hb%\ hc%\ hd%\ he%\ hf%\ hgp%\ hh`%\ hiP%\ hj@%\ hk0%\ hl %z\ hm%r\ hn%j\ ho%b\ hp%Z\ hq%R\ hr%J\ hs%B\ ht%:\ hu%2\ hv%*\ hwp%"\ hx`%\ hyP%\ hz@% \ h{0%\ h| %[ h}%[ h~%[ h%[ h%[ h%[ h%[ h%[ h%[ h%[ h%[ hp%[ h`%[ hP%[ h@%[ h0%[ h %z[ h%r[ h%j[ h%b[ h%Z[ h%R[ h%J[ h%B[ h%:[ hAWIAVIH5=AUIATUSHAPHLHHUIHu(LLH5HHU H81qLHu9L^LHIMLHHU H5H81)LLHHtHMAuHHMuHAZD[]A\A]A^A_AWIAVIH5<AUIATUSHAPHLHHXIHu(LLH5HHT H81 qLHu9LaLHIMLHHT H5H81)LLHHtHMAuHHMuHAZD[]A\A]A^A_AVIAUIHATUSD Ht5H;S HuE1tHLLAHMu)H5HS AH8tE1[D]A\A]A^AVAUIATUQ(HxHX HuHX Ht#H9tHhS H5H8yE1L% Mt I$H5AL-IHtHIHuLmHtHIHAH :HLH:xkAH :LLH:xHAH |:LLH{:ix%E1H o:LLHk:IxIHMHZL]A\A]A^AWAAVIAUMATUHSLHH=w HT$jHT$HMfInfHnDIflH@(H@p@HtHEIl$ ID$@ID$HHIT$PID$XID$8Ml$`IEHtHWI\$hADŽ$IDŽ$ID$xIDŽ$A$A$AF%t6tKuQHmrID$0l=t#=u5HqID$0PHZID$0BHUqID$04ID$0)HP H5M8H8I $uLE1LHL[]A\A]A^A_AVIHAUIATUDSHH}HIH@u#HO LLH5PH81ML$(ID$ Mt ILLIL9v#HO ILLH5.H81LAuNH9sIHl$RLIPMHzH1H11Y^yI $uL E1HL[]A\A]A^UHu H u SH[HD$HbqHD$Hu HD$0Hj HD$8Hvu HD$XHpHD$`Hfu H$Ho H$HPu H$H1 H$H:u HD$AHD$ fD$(D$*HD$@HD$HfD$PD$RHD$h&HD$pfD$xD$zHDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$H$HoH$Hvt H$H H$H`t H$ H) H$(H$PHAH$pHoH$xHAH$H H$HAH$HHDŽ$>HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$0HDŽ$8fDŽ$@Ƅ$BHDŽ$XHDŽ$`fDŽ$hƄ$jHDŽ$$HDŽ$fDŽ$Ƅ$HDŽ$H$HmH$HA H(H$Hۍ HH$H$HAH$8HI H$@HAH$`H H$hHAH$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$ HDŽ$  HDŽ$(fDŽ$0Ƅ$2HDŽ$HHDŽ$PfDŽ$XƄ$ZHDŽ$pHDŽ$xfDŽ$Ƅ$H$H H$HA H$H H$HA(H$H H$HA0H$H H$HA8H$(H} H$0HA@H$PH H$XHAHHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$ Ƅ$"HDŽ$8!HDŽ$@fDŽ$HƄ$JHDŽ$`H$xH H$HAPH$H` H$HAXH$H H$HA`H$Hj H$HAhH$HO H$ HApHDŽ$hfDŽ$pƄ$rHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$"HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$(HDŽ$0fDŽ$8Ƅ$:H$@H H$HHAxH$hH) H$pHH$Hk H$HH$H hH$HH$H H$HH$H H$HHDŽ$PHDŽ$XfDŽ$`Ƅ$bHDŽ$x"HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$#HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$H$0H H$8HH$XHw| H$`HH$H H$HH$H H$HH$H H$HHDŽ$ fDŽ$(Ƅ$*HDŽ$@HDŽ$HfDŽ$PƄ$RHDŽ$hHDŽ$pfDŽ$xƄ$zHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$"HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$H$H H$HH$ Hz H$(HH$HH{ H$PHH$pHU H$xHH$H H$HH$Hx H$HHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$0HDŽ$8fDŽ$@Ƅ$BHDŽ$XHDŽ$`fDŽ$hƄ$jHDŽ$'HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$ H$H~ H$HH$H?v H$HH$8H| H$@HH$`H H$hHH$HE| H$H HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$ HDŽ$ !HDŽ$(fDŽ$0Ƅ$2HDŽ$HHDŽ$PfDŽ$XƄ$ZHDŽ$p"HDŽ$xfDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$H$H H$H(H$Hz H$H0H$HaH$H8H$(H H$0H@H$PHYH$XHHH$xHu H$HPHDŽ$"HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$$HDŽ$fDŽ$ Ƅ$"HDŽ$8 HDŽ$@fDŽ$HƄ$JHDŽ$`,HDŽ$hfDŽ$pƄ$rHDŽ$H$H%`H$HXH$H_H$H`H$HWH$HhH$ Hw H$ HpH$@ H-_H$H HxHDŽ$fDŽ$Ƅ$HDŽ$*HDŽ$fDŽ$Ƅ$HDŽ$!HDŽ$fDŽ$Ƅ$HDŽ$ -HDŽ$ fDŽ$ Ƅ$ HDŽ$( HDŽ$0 fDŽ$8 Ƅ$: HDŽ$P %HDŽ$X fDŽ$` Ƅ$b H$h H{ H$p HH$ Hs H$ HH$ Hc{ H$ HH$ Hes H$ HH$ Hv H$ HH$0 HI]H$8 HHDŽ$x HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$( Ƅ$* HDŽ$@ $H$X H-y H$` HH$ HOq H$ HH$ Hz H$ HH$ Hr H$ HH$ H{ H$ HHDŽ$H fDŽ$P Ƅ$R HDŽ$h HDŽ$p fDŽ$x Ƅ$z HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ H$ Hr H$( HH$H Hiw H$P HH$p HKp H$x HH$ H-y H$ HH$ Hq H$ HH$ Hw H$ HHDŽ$0 HDŽ$8 fDŽ$@ Ƅ$B HDŽ$X HDŽ$` fDŽ$h Ƅ$j HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ H$ Ho H$ HH$8 HXH$@ HH$` Hyq H$h HH$ H[XH$ H H$ H=t H$ H(HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$( fDŽ$0 Ƅ$2 HDŽ$H (HDŽ$P fDŽ$X Ƅ$Z HDŽ$p HDŽ$x fDŽ$ Ƅ$ HDŽ$ %HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ H$ HOk H$ H0H$ HQw H$ H8H$( Ho H$0 H@H$P Huq H$X HHH$x Hs H$ HPH$ Hl H$ HXHDŽ$ !HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$" HDŽ$8 HDŽ$@ fDŽ$H Ƅ$J HDŽ$` HDŽ$h fDŽ$p Ƅ$r HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ H$ Hl H$ H`H$ H?UH$ HhH$HTH$ HpH$@HCNH$HHxH$hHem H$pHHDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$(HDŽ$fDŽ$Ƅ$HDŽ$(!HDŽ$0fDŽ$8Ƅ$:HDŽ$P,HDŽ$XfDŽ$`Ƅ$bHDŽ$xHDŽ$fDŽ$Ƅ$H$HW H$HH$Hj H$HH$HSH$HH$Hm H$HH$0HRH$8HH$XHao H$`HHDŽ$'HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$(HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$ fDŽ$(Ƅ$*HDŽ$@#HDŽ$HfDŽ$PƄ$RHDŽ$hH$Hg H$HH$Hl H$HH$H QH$HH$H+o H$HH$ HMg H$(HHDŽ$pfDŽ$xƄ$zHDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$#HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$0 HDŽ$8fDŽ$@Ƅ$BH$HHl H$PHH$pHd H$xHH$Hp H$HH$Hh H$HH$Hgm H$HH$He H$HHDŽ$XHDŽ$`fDŽ$hƄ$jHDŽ$!HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$ HDŽ$  H$8Hmo H$@HH$`HOg H$hHH$Hg H$H H$Hp H$H(H$HEu H$H0HDŽ$(fDŽ$0Ƅ$2HDŽ$HHDŽ$PfDŽ$XƄ$ZHDŽ$pHDŽ$xfDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$H$HgKH$H8H$(HKy H$0H@H$PHJH$XHHH$xH-FH$HPH$H7r H$HXH$HQ H$H`HDŽ$>HDŽ$fDŽ$ Ƅ$"HDŽ$8HDŽ$@fDŽ$HƄ$JHDŽ$`GHDŽ$hfDŽ$pƄ$rHDŽ$LHDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$!H$H-o H$HhH$Ho H$ HpH$@HQw H$HHxH$hHv H$pHH$HAw H$HHDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$( HDŽ$0fDŽ$8Ƅ$:HDŽ$PHDŽ$XfDŽ$`Ƅ$bHDŽ$xHDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$H$H1v H$HH$Hu H$HH$Hu H$HH$0Hu H$8HHH$XHu H$`H$HBH$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$ fDŽ$(Ƅ$*HDŽ$@HDŽ$HfDŽ$PƄ$RHDŽ$hHDŽ$pfDŽ$xƄ$zHDŽ$H$H` H$HBH$H H$HBH$Hb H$HB H$ HEH$(HB(H$HH[ H$PHB0HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$0HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$0&HDŽ$8fDŽ$@Ƅ$BHDŽ$X5HDŽ$`fDŽ$hƄ$jH$pHs H$xHB8H$Hr H$HB@H$HRq H$HBHH$Hr H$HBPH$Hq H$HBXH$8Hm H$@HB`HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$ HDŽ$ HDŽ$(fDŽ$0Ƅ$2HDŽ$H H$`Hq H$hHBhH$Ho H$HBpH$HRm H$HBxH$Hj H$HH$Hi H$HHDŽ$PfDŽ$XƄ$ZHDŽ$pHDŽ$xfDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$ Ƅ$"H$(Ho H$0HH$PHm H$XHH$xHDo H$HH$Ho H$HH$Hm H$HH$H=b H$HHDŽ$8HDŽ$@fDŽ$HƄ$JHDŽ$`HDŽ$hfDŽ$pƄ$rHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$H$Hqj H$ HH$@Hk H$HHH$hH` H$pHH$Hj H$HH$Hqm H$HHDŽ$fDŽ$Ƅ$HDŽ$(HDŽ$0fDŽ$8Ƅ$:HDŽ$PHDŽ$XfDŽ$`Ƅ$bHDŽ$xHDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$H$H9m H$H$HHH$0H1l H$8H$`HBH$Hh H$HBH$Hg H$H(H$XH$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$ fDŽ$(Ƅ$*HDŽ$@HDŽ$HfDŽ$PƄ$RHDŽ$hHDŽ$pfDŽ$xƄ$zHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$ H$HB H$H H$HB(H$ Ha H$(HB0H$HH5a H$PHB8H$pHCh H$xHB@H$HhHDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$0HDŽ$8fDŽ$@Ƅ$BHDŽ$XHDŽ$`fDŽ$hƄ$jHDŽ$HDŽ$fDŽ$Ƅ$H$Hqi H$HBHH$Hi H$HBPH$He H$HBXH`H$HZPHh H$H$@HBH$`H H$hHBH$8HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$ HDŽ$ HDŽ$(fDŽ$0Ƅ$2HDŽ$HHDŽ$PfDŽ$XƄ$ZHDŽ$pH$H9H$HBH$HCd H$H$HB(H$Hd H$HB0H$(Hh H$0HB8H$HHDŽ$xfDŽ$Ƅ$HDŽ$(HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$ Ƅ$"HDŽ$8HDŽ$@fDŽ$HƄ$JH$PH^ H$XHB@H$xH(8H$HBHH$HU` H$H$HBXH$H H$HB`HhH$Hc HH$HðH$ HDŽ$` HDŽ$hfDŽ$pƄ$rHDŽ$4HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$6 HDŽ$fDŽ$Ƅ$HDŽ$(H$HHBH$hHԭ H$pHBH$Hc H$HBH$H[ H$HB H$HW H$HB(H$@HDŽ$0fDŽ$8Ƅ$:HDŽ$PHDŽ$XfDŽ$`Ƅ$bHDŽ$xHDŽ$fDŽ$Ƅ$HDŽ$ HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$H$Hd H$HB0H$0HY H$8HB8H$XHSc H$`HB@H$H?Z H$HBHH$Hc H$HBPH$H4H$HBXHDŽ$HDŽ$ fDŽ$(Ƅ$*HDŽ$@HDŽ$HfDŽ$PƄ$RHDŽ$hHDŽ$pfDŽ$xƄ$zHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$%H$HP3H$HB`H$ H_ H$(HBhH$HHb H$PHBpH$pH|b H$xH$HH$HHDŽ$fDŽ$Ƅ$HDŽ$3HDŽ$fDŽ$Ƅ$HDŽ$0HDŽ$8fDŽ$@Ƅ$BHDŽ$XHDŽ$`fDŽ$hƄ$jHDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$fDŽ$Ƅ$H$H_ H$H$HH$ HX_ H$ H$@ HH$` H,_ H$h HH°H$ H!Z HH$HH$8 HHH$ HDŽ$HDŽ$fDŽ$Ƅ$HDŽ$HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$( fDŽ$0 Ƅ$2 HDŽ$H HDŽ$P fDŽ$X Ƅ$Z HDŽ$p HDŽ$x fDŽ$ Ƅ$ HDŽ$ H$ HBH$ H; H$ HBH$!H@[ H$!HBH$(!H9_ H$0!HB H$P!HV\ H$X!HB(H$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ HDŽ$ fDŽ$ Ƅ$ HDŽ$ b HDŽ$ fDŽ$ Ƅ$ HDŽ$!HDŽ$!fDŽ$ !Ƅ$"!HDŽ$8!HDŽ$@!fDŽ$H!Ƅ$J!HDŽ$`!HDŽ$h!fDŽ$p!Ƅ$r!H$x!HX^ H$!HB0H$!H\\ H$!HB8H$!HY H$!HB@H$!H \ H$!HBHH$"H;V H$ "HBPH$@"HZ H$H"HBXHDŽ$!HDŽ$!fDŽ$!Ƅ$!HDŽ$!HDŽ$!fDŽ$!Ƅ$!HDŽ$!HDŽ$!fDŽ$!Ƅ$!HDŽ$"HDŽ$"fDŽ$"Ƅ$"HDŽ$(" HDŽ$0"fDŽ$8"Ƅ$:"HDŽ$P"H$h"HW H$p"HB`H$"HZ H$"HBhH$"HQ\ H$"HBpH$"HZ H$"HBxH$#H[ H$#HHDŽ$X"fDŽ$`"Ƅ$b"HDŽ$x" HDŽ$"fDŽ$"Ƅ$"HDŽ$" HDŽ$"fDŽ$"Ƅ$"HDŽ$"HDŽ$"fDŽ$"Ƅ$"HDŽ$"HDŽ$"fDŽ$#Ƅ$#HDŽ$# HDŽ$ #fDŽ$(#Ƅ$*#H$0#HR H$8#H$`#HH$#H H$#HH$#H;[ H$#H$#HH$#H[ H$$HH$X#HH$#HHDŽ$@# HDŽ$H#fDŽ$P#Ƅ$R#HDŽ$h# HDŽ$p#fDŽ$x#Ƅ$z#HDŽ$# HDŽ$#fDŽ$#Ƅ$#HDŽ$#HDŽ$#fDŽ$#Ƅ$#HDŽ$#HDŽ$#fDŽ$#Ƅ$#HDŽ$$H$ $H)Q H$($HH$H$HW H$P$HH$p$HW H$x$HH$$HQ H$$HH$$HTW H$$HHDŽ$$fDŽ$$Ƅ$$HDŽ$0$ HDŽ$8$fDŽ$@$Ƅ$B$HDŽ$X$HDŽ$`$fDŽ$h$Ƅ$j$HDŽ$$HDŽ$$fDŽ$$Ƅ$$HDŽ$$ HDŽ$$fDŽ$$Ƅ$$HDŽ$$HDŽ$$fDŽ$$Ƅ$$H$$HS H$$HH$%H"V H$%HH$8%HT H$@%HH$`%HU H$h%HH$%HT H$%HH$%HU H$%HDŽ$$HDŽ$%fDŽ$%Ƅ$ %HDŽ$ %HDŽ$(%fDŽ$0%Ƅ$2%HDŽ$H%HDŽ$P%fDŽ$X%Ƅ$Z%HDŽ$p%HDŽ$x%fDŽ$%Ƅ$%HDŽ$%HDŽ$%fDŽ$%Ƅ$%HDŽ$%H$%HH$&Hz H$&HH$(&HvT H$0&H H$P&H{V H$X&H(H$x&HO H$%HǰH$&HDŽ$%fDŽ$%Ƅ$%HDŽ$%HDŽ$%fDŽ$%Ƅ$%HDŽ$& HDŽ$&fDŽ$ &Ƅ$"&HDŽ$8&HDŽ$@&fDŽ$H&Ƅ$J&HDŽ$`&HDŽ$h&fDŽ$p&Ƅ$r&HDŽ$& HDŽ$&fDŽ$&Ƅ$&H$&H8H$&Hq H$&H@H$&HT H$&HHHPH$'HH H$ 'H$H'HBH$h'HM H$&HxH$@'H$p'HDŽ$& HDŽ$&fDŽ$&Ƅ$&HDŽ$&{HDŽ$&fDŽ$&Ƅ$&HDŽ$'HDŽ$'fDŽ$'Ƅ$'HDŽ$('HDŽ$0'fDŽ$8'Ƅ$:'HDŽ$P'HDŽ$X'fDŽ$`'Ƅ$b'HDŽ$x' H$'HBH$'H#H$'HB H$'HYN H$'HB(H$(HO H$(HB0H$0(H;P H$'HH$8(HDŽ$'fDŽ$'Ƅ$'HDŽ$' HDŽ$'fDŽ$'Ƅ$'HDŽ$'uHDŽ$'fDŽ$'Ƅ$'HDŽ$' HDŽ$'fDŽ$(Ƅ$(HDŽ$(HDŽ$ (fDŽ$((Ƅ$*(HDŽ$@(HDŽ$H(fDŽ$P(Ƅ$R(H$`(HB@H$(HZ H$(HBHH$(H>K H$(HBPH$(H?Q H$(HBXH$(HN H$)HB`H$ )HG H$X(HH$()HDŽ$h(HDŽ$p(fDŽ$x(Ƅ$z(HDŽ$(!HDŽ$(fDŽ$(Ƅ$(HDŽ$( HDŽ$(fDŽ$(Ƅ$(HDŽ$(HDŽ$(fDŽ$(Ƅ$(HDŽ$)HDŽ$)fDŽ$)Ƅ$)HDŽ$0)H$P)HBpH$p)HH H$x)HBxH$)HP H$)HH$)HP H$)HH$)HP H$)HH$H)HHDŽ$8)fDŽ$@)Ƅ$B)HDŽ$X)HDŽ$`)fDŽ$h)Ƅ$j)HDŽ$)oHDŽ$)fDŽ$)Ƅ$)HDŽ$)HDŽ$)fDŽ$)Ƅ$)HDŽ$)HDŽ$)fDŽ$)Ƅ$)HDŽ$)HDŽ$*fDŽ$*Ƅ$ *H$*HL H$*HH$8*HK H$@*HH$`*HO H$h*HH$*HL H$*HH$*HE H$*HH$*HL H$*HHDŽ$ *HDŽ$(*fDŽ$0*Ƅ$2*HDŽ$H* HDŽ$P*fDŽ$X*Ƅ$Z*HDŽ$p*HDŽ$x*fDŽ$*Ƅ$*HDŽ$*HDŽ$*fDŽ$*Ƅ$*HDŽ$*HDŽ$*fDŽ$*Ƅ$*HDŽ$*H$+HK H$+HH$(+HlK H$0+HH$P+HL H$X+HH$x+HL H$+HH$+HI H$+HHDŽ$*fDŽ$*Ƅ$*HDŽ$+HDŽ$+fDŽ$ +Ƅ$"+HDŽ$8+HDŽ$@+fDŽ$H+Ƅ$J+HDŽ$`+HDŽ$h+fDŽ$p+Ƅ$r+HDŽ$+HDŽ$+fDŽ$+Ƅ$+HDŽ$+HDŽ$+fDŽ$+Ƅ$+H$+HAC H$+HH$+H#E H$+HH$,HD H$ ,HH$@,HF H$H,HH$h,HI H$p,HH$,HCF H$,HHDŽ$+HDŽ$+fDŽ$+Ƅ$+HDŽ$, HDŽ$,fDŽ$,Ƅ$,HDŽ$(, HDŽ$0,fDŽ$8,Ƅ$:,HDŽ$P, HDŽ$X,fDŽ$`,Ƅ$b,HDŽ$x,HDŽ$,fDŽ$,Ƅ$,HDŽ$, H$,H7E H$,H H$,HB H$,H(H$-HgJ H$-H0H$0-HrI H$8-H8H$X-HD H$`-H@HDŽ$,fDŽ$,Ƅ$,HDŽ$, HDŽ$,fDŽ$,Ƅ$,HDŽ$, HDŽ$,fDŽ$-Ƅ$-HDŽ$-HDŽ$ -fDŽ$(-Ƅ$*-HDŽ$@-HDŽ$H-fDŽ$P-Ƅ$R-HDŽ$h- HDŽ$p-fDŽ$x-Ƅ$z-H$-HGI H$-HHH$-H'I H$-HPH$-HF H$-H$.H`H$ .HkH H$(.HhH$H.HNE H$P.HpH$-HHDŽ$-HDŽ$-fDŽ$-Ƅ$-HDŽ$-HDŽ$-fDŽ$-Ƅ$-HDŽ$-HDŽ$-fDŽ$-Ƅ$-HDŽ$.HDŽ$.fDŽ$.Ƅ$.HDŽ$0.HDŽ$8.fDŽ$@.Ƅ$B.HDŽ$X.H$p.HG H$x.HxH€H$.HG H$.H$.HBH$.HB H$.H$/HBH$.H$/HHDŽ$`.fDŽ$h.Ƅ$j.HDŽ$.HDŽ$.fDŽ$.Ƅ$.HDŽ$.HDŽ$.fDŽ$.Ƅ$.HDŽ$.HDŽ$.fDŽ$.Ƅ$.HDŽ$.HDŽ$/fDŽ$/Ƅ$ /HDŽ$ /HDŽ$(/fDŽ$0/Ƅ$2/H$8/HT H$@/HB H$`/H)E H$h/HB(H$/H> H$/HB0H$/H= H$/HB8H$/H^B H$/HB@H$0HMH$0HBHHDŽ$H/ HDŽ$P/fDŽ$X/Ƅ$Z/HDŽ$p/HDŽ$x/fDŽ$/Ƅ$/HDŽ$/ HDŽ$/fDŽ$/Ƅ$/HDŽ$/HDŽ$/fDŽ$/Ƅ$/HDŽ$/HDŽ$/fDŽ$/Ƅ$/HDŽ$0TH$(0H: H$00HBPH$P0H(A H$X0HBXH$x0HC H$0HB`H$0Hk< H$0HBhH$0HA H$0HBpHDŽ$0fDŽ$ 0Ƅ$"0HDŽ$80HDŽ$@0fDŽ$H0Ƅ$J0HDŽ$`0HDŽ$h0fDŽ$p0Ƅ$r0HDŽ$0HDŽ$0fDŽ$0Ƅ$0HDŽ$0 HDŽ$0fDŽ$0Ƅ$0HDŽ$0HDŽ$0fDŽ$0Ƅ$0H$0H? H$0HBxH$1H= H$ 1HH$@1HB H$H1H$p1HH$1HA H$1HH$1H: H$1HH$h1HƠHDŽ$1HDŽ$1fDŽ$1Ƅ$1HDŽ$(1 HDŽ$01fDŽ$81Ƅ$:1HDŽ$P1HDŽ$X1fDŽ$`1Ƅ$b1HDŽ$x1HDŽ$1fDŽ$1Ƅ$1HDŽ$1HDŽ$1fDŽ$1Ƅ$1HDŽ$1 H$1H< H$1H$2HH$02H- H$82HH$X2H9: H$`2H$2HH$2HH$2HHDŽ$1fDŽ$1Ƅ$1HDŽ$1 HDŽ$1fDŽ$2Ƅ$2HDŽ$2 HDŽ$ 2fDŽ$(2Ƅ$*2HDŽ$@2q HDŽ$H2fDŽ$P2Ƅ$R2HDŽ$h2 HDŽ$p2fDŽ$x2Ƅ$z2HDŽ$2 HDŽ$2fDŽ$2Ƅ$2H$2H H$2HHH$2H8 H$2H$3HBH$ 3H H$(3HBH$H3H> H$P3H$x3HB H$2H$p3HHDŽ$2HDŽ$2fDŽ$2Ƅ$2HDŽ$2 HDŽ$2fDŽ$2Ƅ$2HDŽ$3 HDŽ$3fDŽ$3Ƅ$3HDŽ$03 HDŽ$83fDŽ$@3Ƅ$B3HDŽ$X3HDŽ$`3fDŽ$h3Ƅ$j3HDŽ$3H$3HR> H$3HB(H$3HK< H$3HB0H$3H= H$3HB8H$4H: H$4HB@H$84H2 H$@4HBHHDŽ$3fDŽ$3Ƅ$3HDŽ$3HDŽ$3fDŽ$3Ƅ$3HDŽ$3HDŽ$3fDŽ$3Ƅ$3HDŽ$3 HDŽ$4fDŽ$4Ƅ$ 4HDŽ$ 4HDŽ$(4fDŽ$04Ƅ$24HDŽ$H4HDŽ$P4fDŽ$X4Ƅ$Z4H$`4H>< H$h4H$4HBXH$4H H$4HB`H$4Hm) H$4HBhH$5H: H$5HBpH$(5H; H$05HBxH$4HHDŽ$p4HDŽ$x4fDŽ$4Ƅ$4HDŽ$4HDŽ$4fDŽ$4Ƅ$4HDŽ$4#HDŽ$4fDŽ$4Ƅ$4HDŽ$4HDŽ$4fDŽ$4Ƅ$4HDŽ$5HDŽ$5fDŽ$ 5Ƅ$"5HDŽ$85H$P5H: H$X5HH$x5H3 H$5HH$5H; H$5HH$5H6 H$5HH$5H@; H$5HDŽ$@5fDŽ$H5Ƅ$J5HDŽ$`5HDŽ$h5fDŽ$p5Ƅ$r5HDŽ$5 HDŽ$5fDŽ$5Ƅ$5HDŽ$5HDŽ$5fDŽ$5Ƅ$5HDŽ$5 HDŽ$5fDŽ$5Ƅ$5HDŽ$6HDŽ$6fDŽ$6Ƅ$6H$ 6HH$@6H6 H$H6HH$h6H[1 H$p6H$6HH$6HH$6HHH$6H8+ H$6HǐH$6HàH$6HDŽ$(6HDŽ$06fDŽ$86Ƅ$:6HDŽ$P6HDŽ$X6fDŽ$`6Ƅ$b6HDŽ$x6 HDŽ$6fDŽ$6Ƅ$6HDŽ$6 HDŽ$6fDŽ$6Ƅ$6HDŽ$6{ HDŽ$6fDŽ$6Ƅ$6HDŽ$6H$7HBH$07HH$87HBH$X7H8 H$`7H$7HB H$7H6 H$7HB(H$7H$7HHDŽ$6fDŽ$7Ƅ$7HDŽ$7HDŽ$ 7fDŽ$(7Ƅ$*7HDŽ$@7HDŽ$H7fDŽ$P7Ƅ$R7HDŽ$h7HDŽ$p7fDŽ$x7Ƅ$z7HDŽ$7HDŽ$7fDŽ$7Ƅ$7HDŽ$7HDŽ$7fDŽ$7Ƅ$7H$7H1 H$7HB0H$7H3 H$8HB8H$ 8H6 H$(8HB@H$H8H5 H$P8H$x8HBPH$8H5 H$8HBXH$p8HƐHDŽ$7 HDŽ$7fDŽ$7Ƅ$7HDŽ$8HDŽ$8fDŽ$8Ƅ$8HDŽ$08 HDŽ$88fDŽ$@8Ƅ$B8HDŽ$X8HDŽ$`8fDŽ$h8Ƅ$j8HDŽ$8HDŽ$8fDŽ$8Ƅ$8HDŽ$8H$8H}) H$8H$8HBhH$9HH$9HBpH$89H'5 H$@9HBxH$`9Hl, H$h9HH$8HǐHDŽ$8fDŽ$8Ƅ$8HDŽ$8HDŽ$8fDŽ$8Ƅ$8HDŽ$8HDŽ$9fDŽ$9Ƅ$ 9HDŽ$ 9HDŽ$(9fDŽ$09Ƅ$29HDŽ$H9HDŽ$P9fDŽ$X9Ƅ$Z9HDŽ$p9 HDŽ$x9fDŽ$9Ƅ$9H$9H1 H$9H$9HH$9H& H$9HH$:H 4 H$:HH$(:H1 H$0:HH°H$P:H2 H$9HÈH$X:HDŽ$9HDŽ$9fDŽ$9Ƅ$9HDŽ$9HDŽ$9fDŽ$9Ƅ$9HDŽ$9HDŽ$9fDŽ$9Ƅ$9HDŽ$:HDŽ$:fDŽ$ :Ƅ$":HDŽ$8:HDŽ$@:fDŽ$H:Ƅ$J:HDŽ$`:H$:HBH$:H# H$:H$:HBH$:HH$:HB H$;Hj) H$x:H$:HH$ ;HDŽ$h:fDŽ$p:Ƅ$r:HDŽ$:HDŽ$:fDŽ$:Ƅ$:HDŽ$:HDŽ$:fDŽ$:Ƅ$:HDŽ$:HDŽ$:fDŽ$:Ƅ$:HDŽ$; HDŽ$;fDŽ$;Ƅ$;HDŽ$(; HDŽ$0;fDŽ$8;Ƅ$:;H$H;HB0H$h;H(H$p;HB8H$;H- H$;H$;HBHH$;H H$;HBPH$<H0 H$<HBXH$@;HH$;H(HDŽ$P; HDŽ$X;fDŽ$`;Ƅ$b;HDŽ$x;q HDŽ$;fDŽ$;Ƅ$;HDŽ$;HDŽ$;fDŽ$;Ƅ$;HDŽ$;HDŽ$;fDŽ$;Ƅ$;HDŽ$;HDŽ$;fDŽ$<Ƅ$<HDŽ$<H$0<HX+ H$8<H$`<HBhH$<H[- H$<HBpH$<HH$<HBxH$<HH$<HH$X<H8HDŽ$ <fDŽ$(<Ƅ$*<HDŽ$@<HDŽ$H<fDŽ$P<Ƅ$R<HDŽ$h<HDŽ$p<fDŽ$x<Ƅ$z<HDŽ$<HDŽ$<fDŽ$<Ƅ$<HDŽ$<(HDŽ$<fDŽ$<Ƅ$<HDŽ$<#HDŽ$<fDŽ$<Ƅ$<H$<HJ% H$=HH$ =H H$(=HH$H=HF H$P=HH$p=H* H$x=HH$=HJ H$=HH$=HA* H$=HHDŽ$= HDŽ$=fDŽ$=Ƅ$=HDŽ$0=HDŽ$8=fDŽ$@=Ƅ$B=HDŽ$X=HDŽ$`=fDŽ$h=Ƅ$j=HDŽ$=HDŽ$=fDŽ$=Ƅ$=HDŽ$=HDŽ$=fDŽ$=Ƅ$=HDŽ$=H$=Hp* H$=HH$>HL* H$>HH$8>H(* H$@>HH$`>H( H$h>HH$>H& H$>HHDŽ$=fDŽ$=Ƅ$=HDŽ$=HDŽ$>fDŽ$>Ƅ$ >HDŽ$ >HDŽ$(>fDŽ$0>Ƅ$2>HDŽ$H>HDŽ$P>fDŽ$X>Ƅ$Z>HDŽ$p>HDŽ$x>fDŽ$>Ƅ$>HDŽ$> HDŽ$>fDŽ$>Ƅ$>H$>H% H$>HH$>H' H$>HH$?H* H$?HHH$(?H+ H$0?H$X?HBH$x?HT( H$?HBH$P?HDŽ$> HDŽ$>fDŽ$>Ƅ$>HDŽ$>HDŽ$>fDŽ$>Ƅ$>HDŽ$?HDŽ$?fDŽ$ ?Ƅ$"?HDŽ$8?HDŽ$@?fDŽ$H?Ƅ$J?HDŽ$`?HDŽ$h?fDŽ$p?Ƅ$r?HDŽ$?H$?H H$?HBH$?H2' H$?HB H$?H% H$?H$ @HB0H$@@HlH$H@HB8H$@HHDŽ$?fDŽ$?Ƅ$?HDŽ$?HDŽ$?fDŽ$?Ƅ$?HDŽ$?HDŽ$?fDŽ$?Ƅ$?HDŽ$@HDŽ$@fDŽ$@Ƅ$@HDŽ$(@HDŽ$0@fDŽ$8@Ƅ$:@HDŽ$P@HDŽ$X@fDŽ$`@Ƅ$b@H$h@H8' H$p@HB@H$@H H$@H$@HBPH$@Ht H$@HBXH$AHQ$ H$AHB`H$0AH~' H$8AHBhH$@HHDŽ$x@HDŽ$@fDŽ$@Ƅ$@HDŽ$@ HDŽ$@fDŽ$@Ƅ$@HDŽ$@ HDŽ$@fDŽ$@Ƅ$@HDŽ$@HDŽ$@fDŽ$AƄ$AHDŽ$AHDŽ$ AfDŽ$(AƄ$*AHDŽ$@AH$XAHN" H$`AH$AHBxH$AH H$AHH$AHU H$AHH$AH H$BHH$AHHDŽ$HAfDŽ$PAƄ$RAHDŽ$hAHDŽ$pAfDŽ$xAƄ$zAHDŽ$AHDŽ$AfDŽ$AƄ$AHDŽ$A HDŽ$AfDŽ$AƄ$AHDŽ$AHDŽ$AfDŽ$AƄ$AHDŽ$B HDŽ$BfDŽ$BƄ$BH$ BH$ H$(BHH$HBH" H$PBH$xBHH$BHyH$BHH$BHH$BHH$BHZ H$BHH$pBHøHDŽ$0BHDŽ$8BfDŽ$@BƄ$BBHDŽ$XBHDŽ$`BfDŽ$hBƄ$jBHDŽ$BHDŽ$BfDŽ$BƄ$BHDŽ$B HDŽ$BfDŽ$BƄ$BHDŽ$B#HDŽ$BfDŽ$BƄ$BHDŽ$BH$CH H$CHH$8CH" H$@CHH$`CH+# H$hCHHH$CH H$CH$CHBH$CHDŽ$CfDŽ$CƄ$ CHDŽ$ CHDŽ$(CfDŽ$0CƄ$2CHDŽ$HCHDŽ$PCfDŽ$XCƄ$ZCHDŽ$pCHDŽ$xCfDŽ$CƄ$CHDŽ$CHDŽ$CfDŽ$CƄ$CHDŽ$CHDŽ$CfDŽ$CƄ$CH$CH3 H$CHBH$DH0 H$DHBH$(DH H$0DHB H$PDH H$XDH$DHB0H$DH/ H$DHB8H$xDHxHDŽ$CHDŽ$CfDŽ$CƄ$CHDŽ$DHDŽ$DfDŽ$ DƄ$"DHDŽ$8DHDŽ$@DfDŽ$HDƄ$JDHDŽ$`DHDŽ$hDfDŽ$pDƄ$rDHDŽ$DHDŽ$DfDŽ$DƄ$DHDŽ$DH$DH H$DHB@H$DH/ H$DH$ EHBPH$@EH|gH$HEHBXH$hEH H$EHƈH$pEHDŽ$DfDŽ$DƄ$DHDŽ$DqHDŽ$DfDŽ$DƄ$DHDŽ$EHDŽ$EfDŽ$EƄ$EHDŽ$(EHDŽ$0EfDŽ$8EƄ$:EHDŽ$PE HDŽ$XEfDŽ$`EƄ$bEHDŽ$xEHDŽ$EfDŽ$EƄ$EH$EHBhH$EH H$EHBpH$EH H$EH$FHHˆH$0FH H$8FH$`FHBH$EHǐH$FHH$XFHDŽ$EHDŽ$EfDŽ$EƄ$EHDŽ$E( HDŽ$EfDŽ$EƄ$EHDŽ$EHDŽ$EfDŽ$FƄ$FHDŽ$FHDŽ$ FfDŽ$(FƄ$*FHDŽ$@FHDŽ$HFfDŽ$PFƄ$RFHDŽ$hFH$FHYH$FHBH$FH% H$FH$FHB H$FH H$GHB(H$ GH H$(GHB0H$FHHDŽ$pFfDŽ$xFƄ$zFHDŽ$F HDŽ$FfDŽ$FƄ$FHDŽ$FHDŽ$FfDŽ$FƄ$FHDŽ$FHDŽ$FfDŽ$FƄ$FHDŽ$GHDŽ$GfDŽ$GƄ$GHDŽ$0GHDŽ$8GfDŽ$@GƄ$BGH$HGHU H$PGHB8H$pGH H$xGHB@H$GH H$GH$GHBPH$GH H$GHBXH$HHm H$HHB`H$GHHDŽ$XG HDŽ$`GfDŽ$hGƄ$jGHDŽ$GHDŽ$GfDŽ$GƄ$GHDŽ$GHDŽ$GfDŽ$GƄ$GHDŽ$GHDŽ$GfDŽ$GƄ$GHDŽ$GHDŽ$HfDŽ$HƄ$ HHDŽ$ HH$8HH! H$@HH$hHHBpH$HH~LH$HHBxH$HH H$HHH$HH H$HHH$`HHHDŽ$(HfDŽ$0HƄ$2HHDŽ$HH HDŽ$PHfDŽ$XHƄ$ZHHDŽ$pH HDŽ$xHfDŽ$HƄ$HHDŽ$H HDŽ$HfDŽ$HƄ$HHDŽ$HHDŽ$HfDŽ$HƄ$HHDŽ$H HDŽ$HfDŽ$HƄ$HH$IH H$IHH$(IH H$0IHH$PIH H$XIHH$xIHf H$IHH$IH H$IHH$IH H$IHHDŽ$IHDŽ$IfDŽ$ IƄ$"IHDŽ$8IHDŽ$@IfDŽ$HIƄ$JIHDŽ$`IHDŽ$hIfDŽ$pIƄ$rIHDŽ$I HDŽ$IfDŽ$IƄ$IHDŽ$IHDŽ$IfDŽ$IƄ$IHDŽ$I H$IH H$IHH$JH6 H$ JH$HJHH$hJH H$pJHH$JH H$JHHH$@JHxHDŽ$IfDŽ$IƄ$IHDŽ$J HDŽ$JfDŽ$JƄ$JHDŽ$(JHDŽ$0JfDŽ$8JƄ$:JHDŽ$PJHDŽ$XJfDŽ$`JƄ$bJHDŽ$xJHDŽ$JfDŽ$JƄ$JHDŽ$JHDŽ$JfDŽ$JƄ$JH$JH H$JH$JHBH$KH H$KH$8KHBH$XKH H$`KHB H$KH H$JH$0KHxH$KHDŽ$JHDŽ$JfDŽ$JƄ$JHDŽ$JHDŽ$JfDŽ$KƄ$KHDŽ$KHDŽ$ KfDŽ$(KƄ$*KHDŽ$@KHDŽ$HKfDŽ$PKƄ$RKHDŽ$hK HDŽ$pKfDŽ$xKƄ$zKHDŽ$KH$KHB0H$KH] H$KHB8H$KH H$LHB@H$ LHg H$(LH$PLHBPH$KHH$HLHǨHDŽ$KfDŽ$KƄ$KHDŽ$KHDŽ$KfDŽ$KƄ$KHDŽ$KwHDŽ$KfDŽ$KƄ$KHDŽ$LHDŽ$LfDŽ$LƄ$LHDŽ$0LHDŽ$8LfDŽ$@LƄ$BLHDŽ$XLHDŽ$`LfDŽ$hLƄ$jLH$pLHa H$xLH$LHB`H$LHDH$LHBhH$LHK H$LHBpHxH$MH H$MH$@MHBH$LHðH$8MHDŽ$L HDŽ$LfDŽ$LƄ$LHDŽ$L HDŽ$LfDŽ$LƄ$LHDŽ$L@HDŽ$LfDŽ$LƄ$LHDŽ$L HDŽ$MfDŽ$MƄ$ MHDŽ$ MHDŽ$(MfDŽ$0MƄ$2MHDŽ$HMH$`MHt H$hMH$MHBH$MH H$MHB H$MH H$MHB(H$NH H$MH$NHDŽ$PMfDŽ$XMƄ$ZMHDŽ$pMHDŽ$xMfDŽ$MƄ$MHDŽ$MHDŽ$MfDŽ$MƄ$MHDŽ$MHDŽ$MfDŽ$MƄ$MHDŽ$MHDŽ$MfDŽ$MƄ$MHDŽ$NHDŽ$NfDŽ$ NƄ$"NH$0NHB8H$PNHh H$XNHB@H$xNHS H$NHBHH$NH3 H$NHBPH$NH H$NHBXH$NH H$NHB`H$(NHDŽ$8NHDŽ$@NfDŽ$HNƄ$JNHDŽ$`N HDŽ$hNfDŽ$pNƄ$rNHDŽ$NHDŽ$NfDŽ$NƄ$NHDŽ$NHDŽ$NfDŽ$NƄ$NHDŽ$N HDŽ$NfDŽ$NƄ$NHDŽ$OH$OHC H$ OHBhH$@OH H$HOHBpH$hOH H$pOH$OHH$OH' H$OHH$OHDŽ$OfDŽ$OƄ$OHDŽ$(OHDŽ$0OfDŽ$8OƄ$:OHDŽ$PO HDŽ$XOfDŽ$`OƄ$bOHDŽ$xOHDŽ$OfDŽ$OƄ$OHDŽ$OHDŽ$OfDŽ$OƄ$OHDŽ$OB HDŽ$OfDŽ$OƄ$OH$OH H$OH$PHH$0PH H$8PHH¨HxH$XPH HHƨH$`PH$PHBH$PH1 H$PHBH$PHH$PHDŽ$OHDŽ$OfDŽ$PƄ$PHDŽ$PHDŽ$ PfDŽ$(PƄ$*PHDŽ$@P6HDŽ$HPfDŽ$PPƄ$RPHDŽ$hPHDŽ$pPfDŽ$xPƄ$zPHDŽ$PHDŽ$PfDŽ$PƄ$PHDŽ$P H$PH H$PH$QHB H$ QH~ H$(QHB(H$HQH H$PQH$xQHB8H$PHH$pQHƀHDŽ$PfDŽ$PƄ$PHDŽ$PHDŽ$PfDŽ$PƄ$PHDŽ$QHDŽ$QfDŽ$QƄ$QHDŽ$0QzHDŽ$8QfDŽ$@QƄ$BQHDŽ$XQ HDŽ$`QfDŽ$hQƄ$jQHDŽ$Q HDŽ$QfDŽ$QƄ$QH$QH*H$QHB@H$QH H$QH$QHBPH$RH$ H$RHBXH$8RH)H$@RHB`H$`RH H$hRHBhH$QHǀHDŽ$QHDŽ$QfDŽ$QƄ$QHDŽ$QHDŽ$QfDŽ$QƄ$QHDŽ$QHDŽ$RfDŽ$RƄ$ RHDŽ$ RHDŽ$(RfDŽ$0RƄ$2RHDŽ$HR!HDŽ$PRfDŽ$XRƄ$ZRHDŽ$pRH$RH H$RHBpH$RH H$RHBxH$RH H$RHH$SH H$SHH$(SHT H$0SHHDŽ$xRfDŽ$RƄ$RHDŽ$RHDŽ$RfDŽ$RƄ$RHDŽ$RHDŽ$RfDŽ$RƄ$RHDŽ$R HDŽ$RfDŽ$RƄ$RHDŽ$SHDŽ$SfDŽ$ SƄ$"SHDŽ$8SHDŽ$@SfDŽ$HSƄ$JSH$PSHv&H$XSHH$xSH5 H$SHH$SH H$SHH$SH" H$SHH$SH H$SHH$TH H$ THHDŽ$`SHDŽ$hSfDŽ$pSƄ$rSHDŽ$SHDŽ$SfDŽ$SƄ$SHDŽ$SHDŽ$SfDŽ$SƄ$SHDŽ$SHDŽ$SfDŽ$SƄ$SHDŽ$THDŽ$TfDŽ$TƄ$THDŽ$(T H$@TH H$HTHH$hTH H$pTHH$TH H$THH$THjH$THH$TH H$THHDŽ$0TfDŽ$8TƄ$:THDŽ$PTHDŽ$XTfDŽ$`TƄ$bTHDŽ$xTHDŽ$TfDŽ$TƄ$THDŽ$T HDŽ$TfDŽ$TƄ$THDŽ$THDŽ$TfDŽ$TƄ$THDŽ$THDŽ$TfDŽ$UƄ$UH$UH H$UH$8UHH$XUHxH$`UHH$UH H$UHH$UH H$UHH$UH7 H$UHH$0UHHDŽ$U HDŽ$ UfDŽ$(UƄ$*UHDŽ$@U HDŽ$HUfDŽ$PUƄ$RUHDŽ$hU HDŽ$pUfDŽ$xUƄ$zUHDŽ$UHDŽ$UfDŽ$UƄ$UHDŽ$UHDŽ$UfDŽ$UƄ$UHDŽ$UH$UH` H$VH H$ VH; H$(VH(H$HVH H$PVH0H$pVH H$xVH8H@H$VH H$VHDŽ$UfDŽ$UƄ$UHDŽ$VHDŽ$VfDŽ$VƄ$VHDŽ$0VHDŽ$8VfDŽ$@VƄ$BVHDŽ$XVHDŽ$`VfDŽ$hVƄ$jVHDŽ$VHDŽ$VfDŽ$VƄ$VHDŽ$VHDŽ$VfDŽ$VƄ$VH$VHBH$VHaH$VHBH$WH H$WHBH$8WH H$@WHB H$`WHP H$hWHB(H$WHF H$WHB0H$VHDŽ$VHDŽ$VfDŽ$VƄ$VHDŽ$VHDŽ$WfDŽ$WƄ$ WHDŽ$ WHDŽ$(WfDŽ$0WƄ$2WHDŽ$HWHDŽ$PWfDŽ$XWƄ$ZWHDŽ$pWHDŽ$xWfDŽ$WƄ$WHDŽ$WH$WH\ H$WHB8H$WHA H$WHB@H$XH H$XHBHH$(XH H$0XHBPH$PXHx H$XXHDŽ$WfDŽ$WƄ$WHDŽ$WHDŽ$WfDŽ$WƄ$WHDŽ$WHDŽ$WfDŽ$WƄ$WHDŽ$X HDŽ$XfDŽ$ XƄ$"XHDŽ$8XHDŽ$@XfDŽ$HXƄ$JXHDŽ$`X HDŽ$hXfDŽ$pXƄ$rXH$XHB`H$XHUH$XHBhH$XH" H$XH$XHBxH$YHWs H$ YHH$@YH H$HYH$xXHDŽ$X HDŽ$XfDŽ$XƄ$XHDŽ$X HDŽ$XfDŽ$XƄ$XHDŽ$XHDŽ$XfDŽ$XƄ$XH$XHDŽ$YHDŽ$YfDŽ$YƄ$YHDŽ$(Yk HDŽ$0YfDŽ$8YƄ$:YHDŽ$PYH$pYHH$YH H$YHH$YHG H$YH$YHH$ZHH$ZHH$YH\$HDŽ$XYfDŽ$`YƄ$bYH$hYHDŽ$xYHDŽ$YfDŽ$YƄ$YHDŽ$Y HDŽ$YfDŽ$YƄ$YHDŽ$YHDŽ$YfDŽ$YƄ$YHDŽ$YHDŽ$YfDŽ$ZƄ$ZHDŽ$Z HDŽ$ ZfDŽ$(ZƄ$*ZH$0ZH H$8ZHH$XZH H$`ZHH$ZH"H$ZHH$ZH H$ZHH$ZH H$ZHH$ZHv H$[HHHDŽ$@Z HDŽ$HZfDŽ$PZƄ$RZHDŽ$hZHDŽ$pZfDŽ$xZƄ$zZHDŽ$Z/HDŽ$ZfDŽ$ZƄ$ZHDŽ$ZHDŽ$ZfDŽ$ZƄ$ZHDŽ$ZHDŽ$ZfDŽ$ZƄ$ZHDŽ$[H$ [H H$([H$P[HBH$p[HHDŽ$[fDŽ$[Ƅ$[HDŽ$0[HDŽ$8[fDŽ$@[Ƅ$B[H$H[HDŽ$X[HDŽ$`[fDŽ$h[Ƅ$j[H$x[HDŽ$[ HDŽ$[fDŽ$[Ƅ$[HDŽ$[HDŽ$[HDŽ$[HDŽ$[fDŽ$[Ƅ$[H+HteC C!H{Hst5{"t AWHE0HSHHt 1WHEWHE HoQHEH}Ht-VH(H[1[]ATE1UQH5 xRHtC1HHfWIHu THuHM H5vH8QHMuH0QLZ]A\AWAVAUATIUSHH(LwHI I9t 軭H*ItLLNUIHuELSyHH@u1LMtHD$D$E1 D$E1HUHtBH8Ht:HIT$LH|$TH|$HHRH|$Hu(P|$MtHSHLAIHu:LO0LMH IVLH5hwH81UHH H9EuiH;]ucL}IHMuHOHt$LH\$HD$mIIu_LmOUHHu^OMtIuLLO늺WHt$HH?)D$HMIu HOE1H(L[]A\A]A^A_AVIAUATUVQHH5\ HIHHH5 HnIHuHMu>HN4H; L;%ū u L;%߫ uLRtMtV1I $uL]NHtAHMuHJNIMuLMH=3 Mx)H# LHKxH L- MHMu(HKAE1H 鏚L- MגXyAE1cWLH Ht# LH Ht LH Ht LH} Ht1LHr HwLH` H]11H=mPHH H=HnLH7 H$= u`JHH4H H5 H= 8MyA"AJH5HMHtH= :H HtJ0H H5_HGyA(AH= 9H HuA-A龘H={ 9HW HtH=3 9H7 HtH= v9HtH= e9H HtH= M9H HuH= 19H HYH= 9HDH== 9H H(H= 8He H H5 11OHj HuA/A髗H5 1NH= HtH5 1NH& HtH5 1NH HtH5 1NH HjH H5 1uNH H@H5 1RNH HH5K 1/NH HH50 1 NH HH5 1MHr HH H5 1MHP HH H5i 1MH. H`L%ƣ 1LLLiMH H4H5B 1FMH HHg H5( 1MH HH- H5 1LH HH5 1LH HH5 1LHu HwH5m 1LHZ HTH5B 1fLH? H1H5 1CLH$ HH5 1 LH HH5 1KH HH5 1KH HH5 1KH HH5H 1KH H_H5m 1qKH H H  5 H 5 H5m AH H HWAE11H= H > Qh5 54 RRPRR1Q JAHPH HH Hh 1H5 AH6 HAE11H H Qh55 5 RRPRR1Q@HPH` Hz5 15 56 5 52 5l 5 5@ 5j 5l 5 5` 5 5 5 5 5 L s 5 L^ H o H H5 @HĐH6 HAE11H H QhY5; 5 RRPRR1Q?HPH` HrLLL1?H HMWL 1 5 L H  5Q H: 5 H5 5/ 5y 5# 5= ?H@H] HAE11Hź H ƺ Qh15 5 RRPRR1Q >HPH HH L1H5 >H H\Hҿ H5K 1g>H H2AE11H H  Qh5} 5 RRPRR1Q%>HPH HAE11H¹ Hù Rh5Z 5 PP5$ PPR1=HPH HL H h 1H H5[ ~=H HIAE11H/ H 0 Qh 5 5& RRPRR1Q<=HPH HH5Y 1H=Hk HAE11H H Rh5 5 PP5 PPR1 A-H`HN HAE11H H Qh5 5 RRPRR1Q,HPH H5 L j 1L 5Ƹ H  5 Hʭ 5 H5m 5Ϻ 5ɭ 5 5 57 5i l,HPH H3AE11H H  QhJ5V 5 RRPRR1Q&,HPH HPL 1 5 L H װ 5 H 5 H5 5w 5 +H0Hѽ H{AE11Ha H b Qh5 5X RRPRR1Q n+HPH3 H%LK H 1H H5 "+HK HAE11HӦ H Ԧ Qh\5x 5ʳ RRPRR1Q*HPH HHu H5n 1*HӼ HmAE11HS H T Qh5( 5J RRPRR1Q`*HPH5 HH E111H A1Rh5 5 PPPPPR1*HPH HH5 H5Ϋ 1)H HAE11H H Qh5x 5z RRPRR1Q)HPHu HGHM H5 1R)H HH E111H AQh$5 5 RRPRR1Q)HPH HH E111H ARh+5 5 PP5 PPR1(HPH HrH0 H H HwfHnH fHnHC{H flH1 H ) HH=3 HqL H9HDu2HPH~ H5LA4AH81&vpHu?H t5HHH} LH5LA4AH81&2pHkH=׫ R#HHu*H=ë 'HHuA4AoH5 H IHu$HMuHA4A oH;f} AL;% } EuI9t L8$AI $uLE~KH5 HHHu(HMKHA4A/oHu5y.Au(HMHA4ASnH=2 H ψ !H% EHT$PHt$HH|$@($H5 HHHu u%Ht HuHT$PHt$HH|$@?H|$@Ht HuH|$HHt HuH|$PHt HuHMuHzH= 11H +$HHHH5 H#HEdHHEuHL%š 1M$PM${Mt MHcHcH#"HHHػH$\$ I;]IDH"HHuHM$IL;4$tI>u+IHt$LD$Ht$LD$IFIIH9tIHuPLct$ H5IKDA4AHHI$HPH9z H81"HlM\HdH H5 H= J lH=cf HHuA5ADlApHHcH5cw)Hx H)cHMuHYH=`cHHtA HHdH57c$)H- HbHMuHH= cHH@A HH5cH5b(Hޞ HbHMuHH= cSHHA HHbH5bv(H H(bAH HHbH5bE(Hf HaA0HHbH5}b(H= HaAPHHrbH5Lb'H HaAHHIbH5b'H HdaAHH bH5a'H H3aAHHaH5aP'H HaAHHaH5a'Hp H`AHHaH5Wa&HG H`AHHsaH5&a&H Ho`AHHQaH5`&H H>`AHHaH5`[&H̜ H `AHH`H5`*&H H_AHH`H5b`%Hz H_AHH`H51`%HQ Hz_HMuHH=~`NHHA`HH_H5Q`q%H H#_A@HHB`H5 `@%Hٛ H^HH^AHH `H5_$H H^HMuHH=_HH_H _H~ HH5_'^H _HQ HH5_d^H _H$ HH5_?^HMuHGH=(DHHuA7AfH DH HH5]_} ^H CH HH5E_X]H CHŲ HH5-_3]H CH HH5_]H CHk HH5^v]H eCH> HH5^Q]H @CH HH5^,]H CH HH5^z]H BH HH5^U\HMuHH=]fHHwH CHL HH5W^ \H .CH HH5C^r\H F^H HH5Q^M\H HH5],[H PEH HH5][H +EH HH5]o[HMuHOH= ?IHz\H5p H= HRI $uL H= IHL\H5 H= HMRI $uL IH \H LHH ID$HH=N 1uHHxRI $uLmH5 HޕIHQYH5 H= HBRI $uL&HMuHH= HH[H5 H=• HZbHMuHHHY[H HHH HEHH= tIHNbHMuHxH5a LHHQH5> H= HuQHMuH1I $uL#HD$(HD$0HD$8HD$@HD$HHD$PHL$8HT$0HHt$(I:H=*[nIHu4Ho H8JH=[>IHmH5[LIMHuLbHDHn H9Et-HEn H5CH8HMH$ 1HHMHR uHHA HuHm H5ZH8=H v&H5BHm H81׬ Hm H5BH81c\H AƅuHVm H5CH80tH9m H5CH8H|$(HHm Il$XL(I9Hu"AdIEHI]1H9~I;lHE1L9~KtH9HE@t~HFHs]@tTHXHtHJ1H9~5H;t+HHHH9HuH;5l Ies H輎HHUte@t\sVA@tLHXHt$HJ1H9L;lHHtHI9usL;-l tjLH+uVzHuhHD$(H|$0Ht HuKHD$0H|$8HH &H W"H=WHL$PHT$HLHt$@}mx@H52 H= 1gHHt0H贈HMu/HA." %"A*"A ."AI$HL$8HT$0Ht$(HH|$@Ht HuT H|$HHt Hu@ H|$PHlTHcT$ YTH OIHUH H5՝ LHV(LI $uL H=y 4L  L H Hː H=L IHLUH4 H5- LHKI $uLX H= L L H " Hc H= wIHTH H5 LHfZKI $uL H= DL = L H H H= IH|THD H5՝ LHJI $uLh H= L ͤ L H 2 H H= IHTH LI$HH H5Н HdzJI $uL H= BL K L H H H=ڂ IHSH LI$HH- H5 HJI $uLT H= L ɣ L H  H H=0 sIH SH H5 LHbII $uL H= @ L Y L H Hߏ H= IHRH LI$HH+ H5L H!II $uLR H= L ע L H  H= H= qIH>RH LI$HH H5 HNHI $uLH=q , L U L~ H H H=D IHQHl LI$HH H5 H /HI $uL>H= L ӡ L H  H H= ]IHJQH" LI$HH H5 H: GI $uLH=]  L Q Lj H ~ H H=~ IHPHX LI$HH H5 H =GI $uL*H=ӊ L Ϡ L H Hō H=F~ IIHVPHΞ LI$HHy H5b H& FI $uLH=I  Hd HHƜ IHOHc HH=} I\$(L fHnHflAD$L H  H vHH9FLI<$uLH H5n HHO UHMuHH=r - L ~ L H H H=| HHOH- H5 HH IUHMuHQH= L  L H  H H= | pHHNH] HHHH H5r HN THMuHH=q ,L L~ H H H=d{ HH4NH HHHH H5i H YTHMuH?H= L L H H H=z ^HHMH H5L HHM SHMuHH=p +L L} H H H=#z HHSMH+ H5 HHSHMuHOH= L , L H  HB H=y nHHLH HHHH H5 HLSHMuHH=o *L L| H HA H=x HHrLHj HHHH H5? HRHMuH=H= L * L H  H H=9x \ HHKHY HHHH H5v H:#RHMuHH=] L Lj H ~ H H=w HHKHX HHHH H5 HQHMuH+H=Ԅ L ( L H H H=v J HHKH HHHH{ H5 H(3QHMuHH=K L LX H l H% H=>v HHJHF HHHH H5 HPHMuHH=ƒ }L & Lς H Ht H=u 8 HHJH HHHHi H5: HCPHMuHH=9 L LF H Z HӃ H=t HHIH4 HHHH H5 HOHMuHH= kL $ L H ю HR H=Ct & HH#IH HHHHW H5 HSOHMuH~H=' L L4 H H HɄ H=s HHHH" HHHH΁ H5 H{NHMuHH= YL " L H H H=r  HH1HH HHHHE H5 HcNHMuHlH= L L" H 6 H7 H=Hr HHGH HHHH H5e HiMHMuHH= GL L H H^ H=q HH?GH HHHH3 H5 HsMHMuHZH= L L H $ HE H=p yHHFH HHHH H5 HWLHMuHH=z 5L  L~ H H| H=Mp HHMFHu HHHH! H5b HLHMuHHH=~ L L} H  H# H=o gHHEHd HHHH~ H5 HE LHMuHH=h~ #L  Lu} H Hz H=n HH[EHے HHHH~ H5 HKHMuH6H=} L L| H H! H=Rn UHHDHR HHHH} H5 H3KHMuHH=V} L  Lc| H w H~ H=m HHiDHɑ HHHH| H5 HJHMuH$H=| L L{ H H7 H=m CHHCH HHHHt| H5 H!+JHMuHH=D| L  LQ{ H e Hv H=Wl HHwCH? HHHH{ H5 HIHMuHH={ vL Lz H ܇ H~ H=k 1HHBH HHHHb{ H5c H;IHMuHH=2{ L  L?z H S H{ H=k HHBH- HHHHz H5 HHHMuHH=z dL Ly H ʆ H;| H=\j HH BH HHHHPz H5 HKHHMuHwH= z L  L-y H A H| H=i HHAH[ HHHHy H5؆ HtGHMuHH=y RL Lx H H| H= i HHAH HHHH>y H5w H[GHMuHeH=y L  Lx H / Hz H=ah HH@H HHHHx H5f HbFHMuHH=x @L Lw H Hy H=g HH(@H HHHH,x H5e HkFHMuHSH=w L  L w H  H^y H=g rHH?H HHHHw H5́ HPEHMuHH=sw .L Lv H Hx H=ff HH6?H HHHHw H5 H{EHMuHAH=v L  Lu H H\x H=e `HH>H HHHHv H5r H>EHMuHH=av L Lnu H Hw H=e HHD>H\ HHHHv H5| HDHMuH/H=u L Lt H Hjx H=kd NHH=Hu H5 HH=$DHMuHH=`u L Lmt H HBw H=c HHc=Hu H5 HHCHMuH?H=t H=t WHt$XH?)D$PїHH=H5n H=s H_hCHMuHHs H;s H9Xu)H&s Ht HH-s .H= AHH= Hr H5r HH<H5/y H'MIHBHMuHMH5y H=s Lj0I $uL!Hr Hsr H9Xu)H^r Ht HL%Or .H=^ IH=M H.r H5/r IM;H5x LoLHH/I $uLH5fx H=Wr HBHMuHiH2r Hq H9Xu)Hq Ht HH-q .H= HH= Hfq H5gq BHH7;H5x HKIHAHMuHH5w H=q L7/I $uLHzq Hp H9Xu)Hp Ht HL%p .H= IH= Hp H5p IM:H5w LJHH.I $uL%H5w H=p H@HMuHHp Hp H9Xu)Hp Ht HH-o .H=6 aHH=% Ho H5o HH9H5w HGJIHQ@HMuHmH5v H=/p L-I $uLAH p HSo H9Xu)H>o Ht HL%/o .H=~~ IH=m~ Ho H5o IM?9H5w LIHHY-I $uLH5v H=wo H~?HMuHHRo Hn H9Xu)Hvn Ht HH-gn .H=} HH=} HFn H5Gn bHH8H5v HHIH?HMuHH5v H=n LW,I $uLHn Hm H9Xu)Hm Ht HL%m .H=} 9IH=| H~m H5m IM7H5?v LHHH,I $uLEH5v H=n H0>HMuHHm Hl H9Xu)Hl Ht HH-l .H=V| HH=E| Hl H5l HHG7H5u HgGIH=HMuHH5u H=Om L>+I $uLaH*m H3l H9Xu)Hl Ht HL%l .H={ IH={ Hk H5k :IM6H5u LFHH*I $uLH5fu H=l H/<HMuHHrl Hkk H9Xu)HVk Ht HH-Gk .H=z HH=z H&k H5'k HH5H5t HEIHg<HMuHH5~t H=k Lw)I $uLHk Hj H9Xu)Hj Ht HL%j .H=.z YIH=z H^j H5_j IMO5H5Gt L?EHHx)I $uLeH5t H='k H;HMuH9Hk Hi H9Xu)Hi Ht HH-i .H=vy HH=ey Hi H5i HH4H5s HDIH;HMuHH5s H=oj L(I $uLHJj Hi H9Xu)Hh Ht HL%h .H=x IH=x Hh H5h ZIM3H5/t LCHH-(I $uLH5t H=i HOF:HMuHHi HKh H9Xu)H6h Ht HH-'h .H=x 1HH=w Hh H5h HHW3H5t HCIH9HMuH=H5s H=h L]'I $uLHh Hg H9Xu)Hng Ht HL%_g .H=Nw yIH==w H>g H5?g IM2H5os L_BHH&I $uLH5Fs H=Gh H8HMuHYH"h Hf H9Xu)Hf Ht HH-f .H=v HH=v Hvf H5wf 2HH2H5r HAIH}8HMuHH5r H=g L'&I $uLHjg He H9Xu)He Ht HL%e .H=u IH=u He H5e zIM_1H5r L@HH%I $uLH5r H=f Ho7HMuHHf H+e H9Xu)He Ht HH-e .H=&u QHH=u Hd H5d HH0H5/r H7@IH/7HMuH]H5r H=f L$I $uL1He Hcd H9Xu)HNd Ht HL%?d .H=nt IH=]t Hd H5d IM0H5q L?HHL$I $uLH5q H=ge H\6HMuHyHBe Hc H9Xu)Hc Ht HH-wc .H=s HH=s HVc H5Wc RHHg/H5q H>IH5HMuHH5Vq H=d LG|#I $uLHd Hb H9Xu)Hb Ht HL%b .H=r )IH=r Hb H5b IM.H5p L>HH#I $uL5H5p H=c H5HMuH Hc H b H9Xu)Ha Ht HH-a .H=Fr qHH=5r Ha H5a HH.H5?p HW=IH4HMuH}H5p H=?c L1"I $uLQHc HCa H9Xu)H.a Ht HL%a .H=q IH=}q H` H5` *IMo-H5gp L<HH!I $uLH5>p H=b H3HMuHHbb H{` H9Xu)Hf` Ht HH-W` .H=p HH=p H6` H57` rHH,H5o H;IHE3HMuH H5o H=a Lg I $uLHa H_ H9Xu)H_ Ht HL%_ .H=p IIH= p Hn_ H5o_ IM,H5?o L/;HHk I $uLUH5o H=a Hr2HMuH)H` H^ H9Xu)H^ Ht HH-^ .H=fo HH=Uo H^ H5^ HHw+H5n Hw:IH1HMuHH5n H=_` LI $uLqH:` H#^ H9Xu)H^ Ht HL%] .H=n IH=n H] H5] JIM*H5gn L9HH I $uLH5>n H=_ H?$1HMuHH_ H[] H9Xu)HF] Ht HH-7] .H=m !HH=m H] H5] HH'*H5m H9IH0HMuH-H5m H=^ LPI $uLH^ H\ H9Xu)H~\ Ht HL%o\ .H=>m iIH=-m HN\ H5O\ IM)H5/m LO8HHI $uLuH5m H=7^ H/HMuHIH^ H[ H9Xu)H[ Ht HH-[ .H=l HH=ul H[ H5[ "HH(H5l H7IH[/HMuHH5fl H=] LI $uLHZ] H[ H9Xu)HZ Ht HL%Z .H=k IH=k HZ H5Z jIM/(H5k L6HHI $uLH5k H=\ H_.HMuHH\ H;Z H9Xu)H&Z Ht HH-Z .H=k AHH=k HY H5Y HH'H5Gk H'6IH .HMuHMH5k H=\ LI $uL!H[ HsY H9Xu)H^Y Ht HL%OY .H=^j IH=Mj H.Y H5/Y IM&H5j Lo5HH?I $uLH5j H=W[ H:-HMuHiH2[ HX H9Xu)HX Ht HH-X .H=i HH=i HfX H5gX BHH7&H5j H4IH,HMuHH5j H=Z L7oI $uLHzZ HW H9Xu)HW Ht HL%W .H=h IH=h HW H5W IM%H5wj L3HHI $uL%H5Nj H=Y H+HMuHHY HW H9Xu)HW Ht HH-V .H=6h aHH=%h HV H5V HH$H5'j HG3IHq+HMuHmH5i H=/Y L$I $uLAH Y HSV H9Xu)H>V Ht HL%/V .H=~g IH=mg HV H5V IM?$H5i L2HHI $uLH5^i H=wX H*HMuHHRX HU H9Xu)HvU Ht HH-gU .H=f HH=f HFU H5GU bHH#H5h H1IH#*HMuHH5h H=W LWI $uLHW HT H9Xu)HT Ht HL%T .H=f 9IH=e H~T H5T IM"H5Gh L1HH^I $uLEH5h H=W HP)HMuHHV HS H9Xu)HS Ht HH-S .H=Ve HH=Ee HS H5S HHG"H5g Hg0IH(HMuHH5~g H=OV LI $uLaH*V H3S H9Xu)HS Ht HL%S .H=d IH=d HR H5R :IM!H5g L/HHI $uLH5g H=U H/(HMuHHrU HkR H9Xu)HVR Ht HH-GR .H=c HH=c H&R H5'R HH H5Gg H.IH'HMuHH5g H=T LwCI $uLHT HQ H9Xu)HQ Ht HL%Q .H=.c YIH=c H^Q H5_Q IMO H5f L?.HHI $uLeH5f H='T H&HMuH9HT HP H9Xu)HP Ht HH-P .H=vb HH=eb HP H5P HHH5Gf H-IH9&HMuHH5f H=oS LI $uLHJS HP H9Xu)HO Ht HL%O .H=a IH=a HO H5O ZIMH5e L,HH}I $uLH5e H=R HOf%HMuHHR HKO H9Xu)H6O Ht HH-'O .H=a 1HH=` HO H5O HHWH5Oe H,IH$HMuH=H5&e H=Q LI $uLL j LQ 1H=*A H ^ Ha ?IHHf LI$HH5a H=Q #JI $uLL i L_Q 1H=@ H o^ HZ IHxH5Y H=)Q HI $uL;L Di LP 1H=@ H ^ Ha iIH&H5` H=P H_I $uLL h LP 1H=? H ] H\` IHH5D` H=eP HWI $uLwL h L9P 1H=? H I] HZ_ IH|H5B_ H=P HI $uL5[IHGHU HID$HU HHU HID$HU HPHU HID$HU HPH#V HID$HV HPH!V HID$HV HP HV HID$HV HP(H%W HID$HW HP0H#W HID$HW HP8HqW HID$HbW HP@HW HID$HpW HPHHW HID$HvW HPPH{W HID$HlW HPXHW HID$HzW HP`HW HID$HW HPhHX HID$HX HPpH+Y HH!Y ID$HPxH)Y HID$HY HH$Y HID$HY HHY HID$HY HHY HID$HY HHY HID$HY HH@Z HID$H1Z HH;Z HID$H,Z HH6Z HID$H'Z HHZ HID$HZ HHZ HID$HZ HH[ HID$HZ HH[ HID$H [ HHe[ HID$HV[ HHh[ HID$HY[ HHc[ HID$HT[ HH^[ HHT[ ID$HHQ[ HID$HB[ HHL[ HID$H=[ HH_[ HID$HP[ HHb[ HID$HS[ HH[ HID$H[ H H[ HID$H[ H(H[ HID$H[ H0H[ HID$H[ H8H[ HID$H[ H@H$\ HID$H\ HHH\ HID$H\ HPH\ HID$H \ HXH\ HID$H\ H`H\ HID$H\ HhH\ HID$H\ HpH\ HH\ ID$HxH!] HID$H] HH] HID$H ] HH/] HID$H ] HHZ] HID$HK] HHeK HH[K ID$H=I HH5AO L I $uL-yIHBHY H57M HG HX H5L L) HzQ H5KK L  HLZ H5M L H[ H57M L HW H5IL Lu HbO H5J Lb HO H5J LuO H[ H5L LW< HPW H5K L9) HbW H5K L HlW H5K L HnY H5'L L HU H5J L HY H5K L HtP H5I L HO H5I Lg HT H5qJ LI HJN H5;I L+~ H,T H5%J L k H6X H5K LX HxX H5)K LE HY H5KK L2 HT H5I L HY H5/K Lw HU H5I LYHQ H5H L;HO H5H LH^Q H5H LHXQ H5H LHU H5I LHX H5MJ LHW H5I LtHK H5YG LiaH R H5SH LKNHS H5H L-;HX H5I L(HN H5aG LHrN H5kG LHDP H5G LHP H5G LHP H5aG LyHBL H5F L[HU H5EH L=HnR H5wG LH5(V H=iD L}I $u^L{TAFAI $uL_M_H=] H= D ^H=f] Hu1H=S] ARAE1AaAIAfAqE1AA]IAAIAA8AA'AAAĀAAҀAAAAA;AAA AAAA%AA3AE}AAAqlE1AZAXAAGE1AA3AA"E1AAAӃAE1A߃AAAE1AAAAE1AAA-AE1A9AzAKAiE1AWAUAiADE1AuA0AAE1AA AAE1AAAÄAE1AτAAAE1AAAAE1A AwAAfE1A)ARA;AAE1AGA-AYAE1AeAAwAE1AAAAE1AAAAE1AAAхAE1A݅AtAAcE1AAOA A>E1AA*A+AE1A7AAIAAVAAbAAnAAzA$AA+A1A2A;A<xA=mA>bA?WA@LAAAAB6AC+AD AEAF AGAHAIAJAKALAMANAOAPAQARAS{ATpAUeAVZAWOAXDAY9AZ.A[#A\A] A^A_A`AaAbAcAdAeAfAgAhAdAIILܹH DDH=H{HHU w藹mH_H H5nH8.DHHEHA4AHHMHA5A HMHA6AHMHA7AиrDH AH=AGjDI $}mL膸pmAE1AE1AE1AE1AE1AE1AE1AE1A&AA6AiADASAPA=A\A'AqAA}AAAAA€AπA݀A;A{AkA[A"KA0E;A>q+AOAi5 AvYA1AAA AAȁAց2A{AkA6[AKA;A*2+A8}AF ATI Ab Ap A~ A A_ A A) AĂv {A҂ kA0 [A KA&;A y+AA& A4ABAPA^JAlAy\AAA{AkA[A΃KA݃;A+AA  AA(A7AFAUAdAsAA{AkA[AKÄ́;A܄+AA A AA'A6AEATAcArA{AkA[AKA;A̅+AۅA AAAA&A5ADASA`Al~Ax$qA+dA2WA9JA4A7A5A$A6AA7ADAAsACAA2A^A!AkA5AyAYAA1AAAAAA AAAˁAAفA2AAwAAfAA6UAADAA3A-A2"A;A}AIAAWAI AeA AsA AA AA AA_ AA AA) xAǂAv gAՂA VAA0 EAA 4AA&#A AyAAA)AA7AAEAASAAaAJAnAA{A\AAyAAhAAWAAFAăA5AЃA$AAAAAAA AAAA*AA<AAHAAZAAfAzAxAiAAXAAGAA6AA%AAA҄AAބAAAAAAAAAA,AA8AAJA{AVAjAhAYAtAHAA7AA&AAAAA…AA΅AAAAAAAA AAAA(A|A:AkAFAZH([]A\A]A^A_H=f 1H=" H" H9tH Ht H=" H5" H)HH?HHHtH HtfD=" u/UH= Ht H= }ha" ]{f.H GPHGXHDATIUHSHHHt HՅu!H1Ht[LH]A\[]A\ff.HGHHGHGPHHGPHG`HHG`H! H@HGhHtHfH H@Hu310Ht!HD HPH HHPHHH H5:+ 18HHt HHH|$ڬHtH|$HHHHG@HtHHG@fDHH|$蚬H|$HG@HtHHG@Hff.@AVAUATIUSHHHGH$ILl$HD$tFHHOH $Hs LH5aH81/H1[]A\A]A^1LLHt4H$H@uH$ LH5H811@u H $HuH[]A\A]A^fDAUIATIUHSHHGHHt]H=豥u:LLHIZMtHL[]A\A]IHt"HE1[L]A\A]fH[]A\A]H H5H8ۦfAWL~AVAUATUSHL$M~qLHG HIHHD$MLfH|$L3H $HLIL4LHHLۨHHLͨL)IuH HH[]A\A]A^A_f.UHHSHHHGL@@tV$u~HHH[]AucHHNH HvHH[]A@HHN1HtHCH5aHH H81襪H H5H8ZH1[]ÐHtHL$HT$ L$Ht$Hp)fDHL$HT$ߢL$Ht$H3HCH5"HHH H81@HL$HT$藢L$Ht$HHCH5HH H812ff.Ht 閤fDHHSHGHHHtHHCH[fHGH8ԨHCHHu[fHwPH1H=kff.HGHHtfD郩UHGHL@t3H LH H81軩AHEu&]fDH HHH5H81誨Hmt 1]H蠢1ff.HHt HSHHHt8Ht/HPHHHP HHH(tHH[H HfDUHSHHt:HHH}H/tH]H1[]@ۡH]H1[]fDH ff.@UHGHHunHH}HtHEH/t?HHtHDžH/tHEH]H@Kf;fuHUHlH9B0tHd]ATfHnfHnUflHLGXLg`GXHohHOhMtI(tLMtI,$t0HtHmt H]A\@HH]A\陠fL舠fDLxfDAUfIATIUHSHH(HGXHT$Ht$HD$HG`GXHD$HGhHGhH|$HD$kH{XHt$Ht H|$蜞HD$HtHHD$HtHHD$Ht HHD$HT$IUI$HD$HEHH8L`HHT$HhHPHT$HPHtH/t=MtI,$tAHtHmtH(1[]A\A]HHfD;fL(fDH|$IEI$HEHtH/t8H|$HtH/tHH|$HtH/t(H([]A\A]軞f諞f蛞fHt3HLGXHwXMtI(t1ÐHLd1HDH5 ff.@USHHH- HHEHkHHEHt H/tAHEHHHHEHtH/t H1[]۝H1[]f˝fAUIATIUSHiHtTH5m HLALH HmItHL[]A\A]ÐHhHL[]A\A]f.HE1[L]A\A]ff.@ATUSHHpHtHCpH/~H{ HtHC H/SH{@HtHC@H/(H{HHtHCHH/H{PHtHCPH/H{XHtHCXH/H{`HtHC`H/|H{hHtHChH/QH{8HC8Ht H/&HHtHǃH/HHtHǃH/HHtHǃH/HHtHǃH/tfLcxMtQ~71H9~ILkxMt3~)E1@K|HtLՅuID91[]A\A]A^H9t+HXHt/HJH~F1 fHH9t7H9tuf.HH9tHu1H;5 f1ff.fHLGH?t%HuhHHu6I@H6HfHtkHH>HHDH9 HIH5̶H811HHytH IH5жH81Ɲ@H IH5qH81襝HysSHGXHHtHHCX[fHGHxHt螜HCXHu[H H[AUAATUSHH9H H9GHIH9F A|$ HUI9T$HEIL$H9@H@t Hu ED$ D8@ !H}HA cIL$0It$HA@HEȃU\DA9uBHHĘ1Au&1@Hy H9ut1AH[]A\A]@I9uuHDLYHHH;> H;- uH9u8HmuHD$ 誕D$ @1AH[]A\A]@HfDHM0H}H@HE@軓L:fL蘓5fDIt$HfDD@Dff.ATUHHAIHtHHL]A\f.KHuHEHuH H8ǔ1虛HHtHH H8蟔HmuH`ff.AWAVIAUATIUHSHhLG0MtHwHhHL[]A\A]A^A_HVLIHL1IHMLLHH\ImItBHhL[]A\A]A^A_L~HvHu;HhLH1[]A\A]A^A_Af.LhfDE1HBHD$HtJ<8LD$HIH$MLD$tWIF I9IGHL1HHAoDADHH9uLHAt ITITH|$LD$ܕLD$HHD$KD1AL|$ HHD$HL$PLHD$XH|$HLl$0IHl$8IHIHD$HLD$(EHD$PHL$HPH#HHD$XHHD$PJDHL$HD$XJILHLLuL|$ LD$(ILl$0Hl$8HHL$LLHAIH\$HHD$ HHtzH\$H~YLd$1 HH9tGIHOHt1Hx H1Hf.H; toUHHHH; H;- u4H;- t+HӆHmtDH]fD1fDf1H*f.GE@HD$ D$ 몸ff.@LGH?t HA`HtH>HHA`@HH IH5eH81虇1HfH?IIHwHуt LWL_8HFLLLfHtMHIfDHH] HH5H811HfHLGH?tHuXHHu'I@1HHtcHu~H>HDH HIH5TH81袆1HHytH IH5H81v@H IH5!H81UHywLVIIM1fHI9tM9Duf.1fDITHBtv@tmL9tIXHt,HqH~S1HH9t?H;TufDLDHH9xHuH; ffDHI9k1HHtGHH9t.HXHtRHJH~q1DHH9t_H;tuHfDH H5H81HHDHH9tHuH;5h tfDH HNH5~HWH81訄1@HG@t~HFHtt@tkH9t.HXHt*HJH~A1DHH9t/H;tufHH9tHu1H;5 f1D~t@HHt HSHHHt8Ht/HPHHHP HHH(tHH[H H}fDAWAVAUIATUHSHHGHHtIMt"HL[]A\A]A^A_蛃IMuH H8~uH LH5H81HHHtHHtH5 HIHteLHIHtjHI/Ht#I.t'Hmt*HpI3DLx|Ln|Hd|Hm:HO|-1fDHGt{HGHPHvHtOHt1@HtHcWHHcHHE@GWHH f.GWHH HUHH@`HtvHHtjHHt`H H9EuH8Hmt)H]HH53DHI"tHuH? H5(H8qH@DAWAVAUIATAUHSH8oH= IHH@XMwhIGhfH$IG`AGXHD$uHHH8H H9GLu MH I9tL; LCrE1H5a EAH/DC D׃HcHD9t1f}0H9})HcHTD9~߉9|D9A9HHHD9pL I$Hc 1LLnHHt6XlH#qI,$uLdoHmHtH8[]A\A]A^A_I,$uLH8[]A\A]A^A_+oHaHHmIH I|$XLI9t"Ht+I@.L*t111LsqH H5b E1H= #r~$IXMW`MGhMwhD$AGXHt H/Mt I*Mt I(gEH5o HDW AD׃V1UfDIGXfH$IG`AGXHD$IGhIGhHD$ELHuIM~$IXHD$Mo`D$IohAGXIGhHt H/Mt ImHt HmH= EHD-x DHcHD;t1fDU9)HcHDA9}׉IGXfH$IG`AGXHD$IGhIGhHD$LDHÒ1H=jIHHqHHHsImILulG1A9A9 HcHLE9p;D;-\ DL$)PIcHHHHHA9HNHHA9HNHtHHHpL$AEpM D- I$%ImuLkH$HtHHD$HHH\$HtHH$HHtH\$H HHH$HH DD9-d A@IcHlHHH? HcD-1 D-& HLA9LjLD$L$jLD$L$RLL$jL$LIGXfAH$IG`AGXHD$IGhIGhHD$H=X H5! HGHH; .1jHHyH|$'lH|$HHHIHC H9t cHiHiTHiaLiDi*H5k HL$HV}mHL$HIHL H@H 8pHH H DpHv L I$HcHL^HHHU IHD$LD$iLD$HD$I8M H/ZhPHD$kLD$HL$5L2 H oHIH1I;|HH91H9ItH9HT$(HL$ LD$H|$HT$(HL$ LD$H|$H@AWH AVAUATIUSHH(HD$Hq HD$HD$HHLfHHIFLHIFHEIFH5" HHIM'hIH;H H5 HhmIFH LMH=c6b&LHLAIfMI.ImeHI$fInfHnLx(fl@I,$tOH[]A\A]A^A_LbHxbHhbHmLHD$KbHD$H[]A\A]A^A_L(bA(I.HEAHHEMt ImH +DDH=MuP1H[]A\A]A^A_fHmuHaI/H ڧ(H=7z1I$HI$H (H=A1rf.L(a;AA(Hm!E1H`fL`L`A(DHHս E1L 8RH HH8H51{fXZ1H[]A\A]A^A_Hy1H5ƦH褸1}DH }(H=#1T@L`LHLeIHfDA(DH (H=x1@L_EeIAA(kLp_bA(H:H% H5}H8_fAVAUATUHHHHHEH5 HHHHHcIHH H5 H-dHEL5 LMH=|]uZLLHAIVbMt!HmImtHL]A\A]A^aHuH H5|H8^Ao'HmuH^ImH SDH=HE1]LA\A]A^L]HL]A\A]A^fH]MAn'{DLLHRcIHQHH E1E1RL HգH8H H518cXZLH]A\A]A^L(]Hy1H5HTHHcHAj'DHmAl'H\fAUH5G ATUHGHHHIM ID$H5c LHH0II$HI$M&HH}H5 HGHH.HH0HEH5 HHHrIHEHM<HEHH=h LP[HHDI,$H5J H*[IH.HmLLZHHI,$ImHEIHPHUHEHt.L]A\A]DH([TL[H[L]A\A]LZBHZX&I,$uLZLE1H H=IEHPLZ-LxZ`IMH &H=_BL]A\A]f`IHH Z&E1H=fK`HA&f.DLE1H H=̀IEH[fA&HmuHYf_I&fDA&LHY,AUH?IATUHSHHwHHt8HF8HDHHH1L[]A\A]DL I9H= H9LXM MQM~1DILI9RH9IHI9ucHH1L[]A\A]Q\LY I9H=! H9LXMpMQM~'1DILI9H9HI9utHF8HDHLL- MH=vV~1LHAI[MHL[]A\A]HDHI9tHkHEHH5 HH,IM/H5L_QGH5LxTIHAoEHC AoMK0IE HC@HC HCHHCPHHH(tdHEH5 HHHHHHH/tSHD HHI,$tH[]A\A]M땐LH$MH$H[]A\A]DMfMfDSIMH (H=uRH1[]A\A]DH- L- HEHHH=_k2L1HLHPHHHm)tnH H=t1@CPH>&)fD9)@RHrD)@Ht$ $YL$t$ wH1LRHHE@)KOHuH H5qjH8L)fAWAVAUIATUSHHGH5 HHHHH H9]mLuM`L}IIHmqHLL4$HD$$I.IWMI/TH= H5^ HGHHHHH=u H$Ll$H5 HWH9LL L9<LXMMQM~#1ILH9L9 HI9uHt$1HB8HHIMI$A&HI$u LLJHt HmlH DE1H=rgfHJLILIHHH9t4HuH P H9t#HHL9tHuI9fDHOALqE1 uLH=gHLLAI=MM^@H9]FL}M9H]IHHmfInfInHHfl)$I/II.teMH+tfLLcHII$MHI$tqImtZHL[]A\A]A^A_úLIfDLHfDHpHfDH`H9LPHfDL@HfHt$1H)$IIfDLHHt$HHH$Lt$IfDHGNHHAj&aDMHnI$A&HI$/I/LA~&PGI$HA&HI$HI$tTImtA&@LFJHH H5BeH8GeA~&LFff.ff.AWAVAUATIUHSH(HGHXpHtHSHtH(H[]A\A]A^A_HPhHsHRHfH I9L$Ml$IMHMulH; H; HH{LHIHHHSI,$H([]A\A]A^A_IcD$IHcHEII)HEH;\ H;_ teHXpL`hHt H{oMIT$HMvLDHE1H9aHEHHJM!HEJ(HH93HDHD@HH; H5V 1HiDIHLd$H5 HD$IUH9 H=< H9LXMMHM~"1ILH9H9HI9uHt$1HB8IDHLIm/LHD$CHD$fDMHEJ(fLEIHHHD@LIIHt"HFI.IL[CFH HEH;% tH;( I'fDI&I AT$AL$HH H;Ԡ IH;ԟ qHELLLHD$BHD$LLNCt111L茡7EHEHPH H5jH81WH1HE11gL3FNH HH2Bt)ID$L`DH H54jLH81G1BLDI_El$AD$II IFHELII$H}HHIT$IbHHIM`@IH MfXL(M9MIEHIT$|A$@mbA@TI$XHt!HJ1L;lCHH9CM$M9$MuL;-: HHH9t8HuH  H9t'HHH9tHuH9Jf.IMA1HY1 uImH=^W?u*LHHD$DHD$H#CHtC1HETFInH- H8MABIT$H H5 ^HD$H:k@HD$M}1M;dHI91I9ItI9LHfDHGt{HGHPHvHtOHt1B@HtHcWHHcHHE@GWHH f.GWHH HUHH@`HHHHHH@H; uHHEHPHveHHtuHADHmt)H]HH5 HHtH@fDHHD$>HD$@HtHcUHHcHHEEUHH fEUHH HkFAH4HXH@`HHHHHHH H9GuXHGHGHPHvWHt}HthH|$@H|$H/HD$=HD$H5مHHuHHtHcWHHcHHE맋GWHH 똋GWHH HH|$HH|$r)@HuH H5H8>=H H5{H8#=ff.HGHH;~ u 1IHtVHD$ IHL$ HT$@ILL HaJF{7HzFH31D$LT$L $3D$LT$L $@LLT$Ht$L $1L $Ht$LT$HLT$LL$H4$1H4$LL$LT$qHvH;I}H6HFDDAWH AVfHnAUATIHHUfHnSHflHho H HD$PL HD$XH HT$0)D$@)T$ HHL4HvHHwLLyLL$ MH LHHEHD$(IMcLML-H Mi1HI9KL;luIHAHD$0I,@HVHFLLyHD$0HFLL$ HD$(MLL$ HD$(HT$0H HI$It$ H=AHEQHj5 5, j5 Pj5  IHEHPMHHEtwHhL[]A\A]A^A_DH,HHH HLLL$ IfHuZHVH HT$0HFHD$(Hp0|LyMKHT$0H f.H tvAHHH HvH5jVSL {H815XPZH Dv H=XE1H 1 HxHFHLyHD$(HT$ HH uHuIHII?IA[fDLL$ #fDE1IE0JtHD$I99fDH I9EH9F#A} ~ IUH;VHFIMH9@H@t HA} D^ D8@ IMH@HHE|$A HN0HHA@HEȃDA9u5L $HtFLT$H0LT$L $IM9JtI9Kd@H i I9uuH9uuLHL$LT$L $T.HHtnH;= L $H;= LT$HL$H9LT$LL$H<$1LT$LL$H<$H/6I0HOUfDHHEH s 8PH=U%1H q HH LHIHtVHD$ IHL$ HT$@ILL ^sHJO+0HzOHx,1D$LT$L $^,D$LT$L $@LLT$Ht$L $*L $Ht$LT$HLT$LL$H4$Z*H4$LL$LT$qHvH;I}HC/HODDAWH/ AVfHnAUATIHHUfHnSHflHhoJ H; HD$PL 7 HD$XH/ HT$0)D$@)T$ HHL4HvHHwLLyLL$ MHy LH覲HEHD$(IMcLML- Mi1HI9KL;luIHAHD$0I,@HVHFLLyHD$0HFLL$ HD$(MLL$ HD$(HT$0H a HI$It$ H=AHEQHj5j 5ܿ j5 Pj5[ E IHEHPMHHEtwHhL[]A\A]A^A_DH,HHH HLLL$ IfHuZHVH HT$0HFHD$(H )|LyMKHT$0H W f.H $oAHH HnoH5OSL OtH81.XQZH n_ H=nQE1薷H HxHFHLyHD$(HT$ HH nHnIHII?IA[fDLL$ #fDE1IE0JtHD$I99fDH I9EH9F#A} ~ IUH;VHFIMH9@H@t HA} D^ D8@ IMH@HHE|$A HN0HHA@HEȃDA9u5L $HtFLT$H)LT$L $IM9JtI9Kd@H  I9uuH9uuLHL$LT$L $'HHtnH; L $H;= LT$HL$H9LT$LL$H<$*LT$LL$H<$H/6I)HfQUfDHHEH 5l QH=Nմ1H ! HHI LH螭IHtVHD$ IHL$ HT$@ILL lHJkQ(HzXQH(%1D$LT$L $%D$LT$L $@LLT$Ht$L $B#L $Ht$LT$HLT$LL$H4$ #H4$LL$LT$qHvH;I}H'H_QDDAWHw AVfHnAUATIHHUfHnSHflHho H~ HD$PL HD$XH HT$0)D$@)T$ HHL4HxHHyLLyLL$ MHö LHXHGHD$(IMeLML- Mk1DHI9KL;luIHAHD$0I,@HVHFLLyHD$0HFLL$ HD$(MLL$ HD$(HT$0H  HI$It$HH=!,AHEQHj5 5 j5Ե Pj5  IHEHPMHHEtwHhL[]A\A]A^A_DH,HHH j HLLL$ IfHuZHVH G HT$0HFHD$(H!|LyMKHT$0H  f.H gAHH~ H'hH5GSL lH81X'X>RZH g H=NJE1FH ~ HxHFHLyHD$(HT$ HH >gH@gIHII?IA[fDLL$ #fDE1IE0JtHD$I99fDHq~ I9EH9F#A} ~ IUH;VHFIMH9@H@t HA} D^ D8@ IMH@HHE|$A HN0HHA@HEȃDA9u5L $HtFLT$H:"LT$L $IM9JtI9Kd@H | I9uuH9uuLHL$LT$L $HHtnH;| L $H;=W| LT$HL$H9LT$LL$H<$V#LT$LL$H<$H/6IY"H%RUfDHHEH d$ uRH=G腭1H { HH LHNIHtVHD$ IHL$ HT$@ILL dHqJ*R!HxRH1D$LT$L $D$LT$L $@LLT$Ht$L $L $Ht$LT$HLT$LL$H4$H4$LL$LT$qHvH;I}H HRDDAWH AVfHnAUATIHHUfHnSHflHho Hw HD$PL HD$XH HT$0)D$@)T$ HHL4HvHHwLLyLL$ MHٮ LHHEHD$(IMcLML-X Mi1HI9KL;luIHAHD$0I,@HVHFLLyHD$0HFLL$ HD$(MLL$ HD$(HT$0H x HI$It$ H=AHEQHj5ʢ 5< j5 Pj5  IHEHPMHHEtwHhL[]A\A]A^A_DH,HHH x HLLL$ IfHuZHVH w HT$0HFHD$(H|LyMKHT$0H w f.H `AHHXw H`H5z@SL eH81 XPZH T` H=.CE1H Aw HxHFHLyHD$(HT$ HH _H_IHII?IA[fDLL$ #fDE1IE0JtHD$I99fDH!w I9EH9F#A} ~ IUH;VHFIMH9@H@t HA} D^ D8@ IMH@HHE|$A HN0HHA@HEȃDA9u5L $HtFLT$HLT$L $IM9JtI9Kd@H yu I9uuH9uuLHL$LT$L $dHHtnH;Mu L $H;=u LT$HL$H9LT$LL$H<$LT$LL$H<$H/6I HPUfDHHEH ]Z PH=j@51H t HH LHIHtVHD$ IHL$ HT$@ILL ]H!JP;HzPH1D$LT$L $nD$LT$L $@LLT$Ht$L $L $Ht$LT$HLT$LL$H4$jH4$LL$LT$qHvH;I}HSHPDDAWH AVAUIATUSHHHHD$(HD$0HD$8HuHL4H}HL&LALd$(MIEH5 LHHHHH*Hq H9ELuMLmIIEHmdfInfInHt$0Lfl)D$0_I.IBIEMuHIEI,$Hq HHH[]A\A]A^A_fHHeq HZH5:SL `ZAH ZH81X'ZH TZH=V=HH1[]A\A]A^A_HuL&Ld$(DLy1H ˧ MHL9H;LuM$Ld$(MMGRLxLhHXLHHt$8HILd$8HD$0ݹIDH(IEH @YH=B<HH1[]A\A]A^A_@(fHHA0E1HD$@JtH9HKp H9AH9Fy S~ HQH;VHFHyH9AHAt HDY DV DD@@8A fHyHA@HE|$A >L^0HHA@IEDA9A<DDE9u;H $HLL$E1H LL$H $A@IM9fDL%n L9uuL9uuHϺLL$H $HHtbH;n H $H;= LMH=("LLHAI MHmI,$0L #f.Hh HX BHIcEHHcHEHf@H=٘ H H5 >HHfAsA$|H= |H@+IM~MMfII$I.fInfInHLfl)$:I/HLeAsA$I,$LA@AmAEHH h@AmAEHH HCfHaA$Hm@HAvx@LLLHj IHfDA$ HuHd H5MH80 HHfD H"AvA$A$Av HjHc H5\%H8OH@`HCHH3LIH"H@H-7d H9uPfImHEHvTHHtaL HImLWLH5O#cIHH@HtIcEHHcHEHAmAEHH Im>AmAEHH HwAtA$]H@`HHHLIHH9huTID$ID$HPHvbHHtkL HI,$LMH5NHbIHuImL1Ht HAl$uHcAl$AD$HH Al$AD$HH HtLlHdOHuHa H5JH8dnAv@@AWH AVAUATUSHHHD$PH@HD$XHHD$`HHD$hHa H|$HD$0HD$8HD$@HD$pHD$HH3LHL$0HT$PILL IHzy^9HuMDHD$HL|$0H\$8Hl$@HD$HȞ L(hE111HALAIH H8H L(hE111HAHAIH H8H> L(hE111HAHAIH H8AEA;D$aHr H {~ H9HHb~ HaHH-N~ HuHEH5 HHHGIMHmH H } H9HH} HeHH-} HYHEL\$HH5O HHL\$IHEHHEMH/Hx H a} H9H HH} H HH4} HHCL\$ HLT$H5c HHLT$L\$ IMH+ H] I9@ LHt$XL\$(LT$ LD$HD$PLd$XLl$`褦LD$LT$ L\$(HLHIHm H\ I9BkLHt$XL\$ LT$HD$PH\$XLt$`;LT$L\$ HMH+ HI/ H'\ I9CLHt$XL\$HD$PHl$XإL\$ILHmJ MH+o L;P\ L;\ u L;(\ b I* HD$HH=MLHp ILL ɍ j5I AVj5w AUj HT$PLT$Hә H@LT$HII* LIH I,$ImMI.vMI/LfDH.HHH CHCAHMEIHHZ HuDH5#SL IH81oX^ZH CH='E1]HĈL[]A\A]A^A_H8HFHD$HD$HHnoH^L>Hl$@)L$0}@HLAHD$00LAH LHLD$轄HD$0HELD$LuIDHD$|fDHL\$L\$KHL\$ LT$nL\$ LT$HPNA;FL޼IH HɼIH HLD$诼LD$HH KH95HD$LLD$ LILT$.LT$HH_LD$ LLT$HI{HHD$ LT$L\$ HHHt$HE1LjH= A5 H Pj5; ASL\$Pj 5l SHT$`LT$X HPLT$L\$HII* H+u I+[ Hm@ LIHD$E11D$_Rf.HL}E1LL\$pL\$MMtImuLTM;@fDID$0LD$1HD$ H\$LMI̐NlM9HW I9FI9EUMA~ A} IVI;UIEIvH@H9@t HEV EM DD@@8urA PI~HA bIu0IHA@IDDAA DDE9u'HHII9HD$@HHHV H ?H5#jL WDAH?H81Y^v^/L;5U uuL;-U urLLIHcH;U L;-U u L;-U ImLD$H\$L@LD$H\$LIf.MID$0LD$1HD$ H\$LMI̐NlM9HkU I9FI9Ee]A~ A} IVI;UIEIvH9@H@t HEV EM DD@@8~A I~HA@HE|$ A Iu0IHA@IDDA{A DDE9u)HH3DII9HD$8HHHS AH5jL AH <H8H'=13_l^AXfDL;5S uuL;-tS urLLrIHcH;WS L;-S u L;-/S ImLD$H\$L@LD$H\$LIf.D$~E1E11D$^E1E1E1E11Mt I+Ht HmMt I*HtH+tJMtI(tWT$t$H J;H=MI,$`E1pHLD$LD$@LfDLLT$ LD$LT$ LD$CHLT$ LD$nLT$ LD$+LLD$KLD$D$E1E11D$^E1E1DHL\$(LT$ LD$L\$(LT$ LD$DD$E1E11D$^E1pHL\$L\$HfHL\$ LT$L\$ LT$L-LLT$LT$D$E11E1D$_I~HA@HE|$ DLh5I~HH= L\$ LT$LT$L\$ HVt^Pj^FfDHLT$LT$LL\$L\$;fHLT$sLT$zfLD$(TD$(LV0LILNIuHD$E1E1E1D$_f.H= Hn H5n H8fDLWHLHL\$kL\$tLL\$SL\$RfD$E1E1D$_DH=р H@cILD$(D$( L;Q;L&G&DH}^)D$E1E1D$_.H=7 H@m L\$H5D$1E1D$'`D$E11D$5`LL\$ LT$LT$L\$ LL\$ LD$LD$L\$ HFLHD$E1E1E1D$D_-D$E1E1E1D$`E1E11D$_11E1D$E1ff.AWHWw AVAUATUSHHHD$PH0HD$XH(HD$`HHD$hH@G H|$HD$0HD$8HD$@HD$pHD$HHLHL$0HT$PILL S/H_yhTHuMDHD$HL|$0H\$8Lt$@HD$Hh H(hE111HALIHH8H$ H(hE111HAHHHH8 H L(hE111HALAIHH8 EA;D$Hj H c H9H<Huc HHL ac M@IALL$LH5Uo HHeLL$IMjI)XHi H c H9HHb HuHHb HHCH5Ar HHHWIHHMHHH.C I9A2LHt$XLL$HD$PLd$XHl$`ڌLL$ILMoH)HB I9GHt$XLHD$PLt$X苌MII. MǺ# AUMI(L;B L;B ^ L;B Q LLT$LT$wI*#H h H ea H9HkHLa HHL=8a M7IGH5Tm LHHIMI/Hg H ` H9HH` HHH ` HHALL$ HHL$H55p HHHL$LL$ IHHM\HHHA I9FsHt$XLLLL$HD$PHl$XLl$`ĊLL$IA4V% MH+H@ I9ALHt$XLL$HD$PL|$XhLL$ILI/AKV% MH+L;@ L;@ u L;@  I* Hf H =_ H9HH$_ H HL _ MIALL$LH5Dk HHLL$IMI) He H=^ H9x H^ HHH^ HHCH5(m HHHIHHMHHH? I9ALHt$XLL$HD$PLd$XLl$`ɈLL$ILMdH)H> I9GHt$XLLHD$PLt$XwII.Iߺ' AVM H+&L;> L;> L;> LLT$LT$I*oHD$HH=MLHp IL h LjAQAUjAQUjHT$PLT$Ha| H@LT$HII*LDH&HHH &H&AHMEIHH= H\'H5SL +H81GX|TZH & H= E15oHĈL[]A\A]A^A_H0HFHD$HD$HLvo&H^L>Lt$@)d$0@HLAHD$0pLAHl LHLD$gHD$0HLD$LuINDHD$|fDLpHLL$[LL$HHGAU# H+H߉T$LL$LL$T$LːMt I/ Ht H+H 5%DH=mI,$A+Ht HmM}ImrLef.HA;E*"Lf.ޅf( LL$ f.L$ D$ HL$ nf.L$ f( f/V f/T$ f.L$F Lt$f(T$ IHHL$HL$HIT$ HD$(f(HL$LL$(HI9D$LL$ HL$LL$ HH&Hd HL5: jH APH=SjPAWjPHT$`LL$hHL$Xx HPHL$LL$HIBH) I)z I/ H+ I,$t0HmDHHILID$0LD$1HD$ H\$LMI̐NlM9H: I9FI9EUMA~ A} IVI;UIEIvH9@H@t HEV EM DD@@8urA I~HA Iu0IHA@IDDAA DDE9u'HHII9HD$@HHHQ8 H \!H5sjL &AH!H81Y^\TL;5A8 uuL;-48 urLL2IHcH;8 L;-7 u L;-7 yImLD$H\$L@LD$H\$LIf.H߉T$4T$9M&ID$0LD$1HD$ H\$LMI̐NlM9H7 I9FI9EUMA~ oA} yIVI;UIEIvH9@H@t HEV EM DD@@8urA eI~HA Iu0IHA@IDDA\AW DDE9u'HHqII9HD$8UH'HH5 AH5jL ($H H8Hv1s_RTAX&fDL;55 uuL;-5 urLLIHcH;5 L;-U5 u L;-o5 !ImLD$H\$L@LD$H\$LIf.LLT$ LD$LT$ LD$E1 ATDLLT$sLT$f AT^H  TE1H=0fL Lp}I~HA@HE|$ 1DLHI~HA@HE|$ LLT$#LT$g% AOVMI*L׉T$T$DH H=UpeLcI_HfMGHII/ fInfHnLǺflHt$PLD$)D$P|H+LD$I:HLT$ LD$LT$ LD$# AU%LH=re HQ H5Q HZTEPT;LD$(|D$(L.L!LLL$3LL$XfLT$T$ IuHUH=d H2Q H53Q IfD# AUHLL$LL$gfH=Qd 4Is@LL$IfD# AULaDLD$(LD$(@L[LwFHcTHLL$LL$ LLT$LT$?# AU1H=qc HO H5O HIuHkDDHLT${LT$LiyHLL$WLL$XHEiL8RH=b HzIIYHIIHHI)fInfHnHϺflHt$PHL$Hl$`)D$PyH+HL$IHHL$HL$# AU1H)pHωT$tT$[ D$L$H ATAL$ L$ H A UL$ D$(yL$ T$(H AU@DDDDH5h H=l 1*IHHA+UgLI/ vLT$\T$aH5h H=l 1~*IH3HAVLI/$ #H5\h H=el 1>*IHHAKUKI/ h% AVH=q` HL H5L IH5g H=k 1)IHHAkUYKI/ hHLL$HLL$% AV1!I;H=_ ILLT$LT$JHH=_ HK LL$H5K LL$HDD AULE1% AV#H AHHTHLT$VLT$1AU I)LωT$HL$$HL$T$fDH=^ LL$蟿LL$HT% AV?=LL$ HL$ImLMFMI^IHI.fInfHnHt$PHflƺLL$ LD$Ll$`)D$P'uLD$LL$ II(TLLL$CLL$=AU! MqMOIYIHI)fInfInHt$PHflź)D$PtI.I7LLT$LT$ AU nH5d H=h 1&IHHA^VnHI/& }LHL$]HL$dH=] HMI H5NI iI)' ApV/LLD$LD$H=\ 螽IQLL$I' ArVo' AuV1H=s\ VH"AwV' fI1' AVIYHAIIHHI)fInfHnHϺflHt$PHL$Ll$`)D$PrH+HL$IHHL$HL$MOMI_IHI/$fInfInHt$PHflźLL$)D$PyrLL$II)LLT$LT$' AVLLD$ LL$nLL$LD$ H5b H=f 1$IHHAV FI/( /* AVcL$ AV A'U AGU AgULLL$LL$LHL$HL$3& AZV( AVLYfAWH7V AVfHnAUATUHSH(HfHnflHxo-` L-&` HD$`H% H|$L5` HD$hHD$@)D$P)T$0HUILH! L.Ll$0}DHH^H\$@LvLt$8f.H(L`f.kD$LDf.lk\D$HF H ? D$H9H H? H HL=o? M IGH5#P LHHp IIHMC IHGD$LD$ LD$ HIH I9@LHt$XLL$XLL$(LD$ HD$PiLD$ LL$(IMI)MI/L;-7 L;- u L;- !DImE Lt$D$MIEIHFD$IHKHI LMLH5 H H=%}jAHP5+X jPASL\$HjP] L\$XIHPHIm I/ I+AALLL$LL$fHKLAMH\$@fH AHH HXH5UL H818X4AZH 1H=>E1&PHxL[]A\A]A^A_@Ha $@HLL\$L\$ fE1LLL$LL$MHHmHLLvHLALt$8HD$0fDLLD$SLD$zfL8HH [H]IHII?IALl$0kfD1IF0ItHD$ I9BH I9FH9F+#A~ (~ VIVH;VHFI~H9AHAt HEn D^ DD@@8A {I~HA@HE|$ A SLn0HHA@IEDA( A DD.E9u8LL$HtHLD$E1HQLD$LL$AHI9ItI9I\@L- M9uuL9uuLLL$LD$оHHtqH; LD$H;=r LL$L9LL$(LD$H|$uLL$(LD$H|$AH/QE8EJrHAE1E11E1E1yAE1MtI/tjMt I)MtI(t8H H=sLMI,$E1Lljt$Dt$fDLt$LL$LD$"t$LL$LD$mLωt$LD$t$LD$VfE1E1E1E1ADLL\$購L\$hfH O@HJ LLLD$iDLD$HIHD$0If.HL$0HT$PILL 0L3: ALLD$ LD$ LLT$LT$QfLLL$L\$λLL$L\$LKBE1fDLAƅAE1Imm Lt$E1E1St$E1E1@D4.赾H E1E1AE1E1QD$D$HE1E1AE1E1DH=qJ Hz6 H5{6 ֭IfDH=IJ L\$'L\$ILD$LD$H ALHZL8L(MLD$D$E1E1E1ɻ:B@H=I H5 H55 IfDLE1E1MLAH|$(Mv HGe LoIEH3IqIڴIILImHHIHHIFH5B LHHIM襵IHHHHH5&D HL#HmHi3 H* H9XH H* H HH-* HL HEH5IE HHH IM HmH5: LL苵I.IGH-F HHH=/ LHLI觳M I/I,$IFH5.8 LHHs HH= I.HL H9]L}MLeII$HmHD Ht$0LL|$0HD$8UI/IMtI,$IFH5~C LHHIIHIMHI9_8MwM+MgII$I/Ht$0LLt$0HD$8,UI.Hh7HI,$&HEHXpHH{LRIHH=g HHI,$IMLHSImHUHJHxHMHuqHH$賭H$_DHH HH5ATL AH H81_X6ZH 5H=P<1HH[]A\A]A^A_HuH>H|$(DMHcGIHcHEI@Ly1H K: MHL9H;LuI|H|$(HMG1L訬L蘬L般Hx6HPH H5H81DHUHBAWl7HEHDH rH=Y;HH1[]A\A]A^A_AWS7HuLfDH;DoGII IfDoGII DH蠫gH萫^L耫rLprHH53<HHۮH5I英I0AV7HA0E1HD$@JtH9H H9AH9Fy ~ AHQH;VHFHyH9AHAt HDY DV DD@@8A &HyHA@HE|$A L^0HHA@IEDAADDE9u;H $H LL$1HjLL$H $fDIM9fDH  H9uuH9uuHϺLL$H $HHtgH; H $H;= LL$H9LL$HL$H<$蠭LL$HL$H<$H/VNy*HD$(蜬H[6fK|fDLبAV7E1I/tlI,$tEHtHmtIMlI.bO7I,$AWF1E1DLhfDHXfDLHfDI/uL2AV7fLLE1AV7+DLHL蚭IHfL踧'xHt$8HIHD$0H`< HD$86NIafDL`AV$71fDH=6 HZ! H5[! VHfDE1AW7CDHL$(HT$0MLL H1,H|$(YLH$輦H$AW!7H=Q6 4H@I:fHt$81L)D$0MMHDLHMAW:7AV.7E11YsH蓩HHT1AV.7#D;ILL$H $¥LL$H $@HLL$Ht$H $H $Ht$LL$eHLL$HL$H4$£H4$HL$LL$-HvH HyHDD DD6H@`HtzHHtnHHtdHj H9EeHxeHmIH貤v6;Ht AV.7H5H8BHH H5qH8@AWH7 AVAUATIUSH(H LvHD$HD$H\$HHMmIHFHHD$zHLl$I9H5- HFIELHHDHHH/IELID$LHHHH/AHmHHMImMH^H NHILH?H?L MLIL@HH AVHH5-H81èXV%ZH H=A1H(D[]A\A]A^A_@H IHH5M5 HHVAHHD$IEpf.H% H b H9HHI HHL-5 MHI9EfHt$1L)D$MHHH I.NLʡADLnLl$f諡fDKHd$ H  H9HSH HHL=p MWH8I9GnfHt$1L)D$MGHHI.HEH5j. HHHIMHI9F`M~MSI^IHI.fInfInHt$Hfl)D$XGI/IMRH+ImLe[fD1HL$HT$ML ?HIH%TLH)H5AH8贠H 2%H=<.$fHȟKL踟H=# H H5 ƒIqfDH %AH=e.軥HH=1# I@M}M(MuIIImHt$LL|$HD$EI/HLߞf.%AI.kDH AH=-L舞Ht$LLLl$HD$EIDۡHC%lH=" H* H5+ FIfDH F%AH=J,2I_HMwHII/Ht$LH\$HD$]DH+H`H舝SH=a! I@%ALH蛣I9H+uH"H e%AH=f,@LL!fDH %AH=+UH "&AH=+/f.AWAVAUATUHHH"IHHEH H  H9HiH HHL5 M5IFH5- LHHoIIHIM7HL^IHbIELMH=šLHLAIfM}ImtfI.tPHmt9I,$t"HL]A\A]A^A_ÐLHmL8fDH(fDLfDLI.u苞HI.A vImD0E1H H=)D@LHLbIHtImzf.HBIILl$H$HD$t'LߟIHE11LLLtH$H@uH$H"H5E1H81ٟ@AvDH=9+ H: H5; IfDAvHL葙@IH=* IF@A vsDLHlHH5H86L 9ff.AWAVAUATUHHH貝IHHEH H C H9HiH* HHL5 M5IFH5r* LHHoIIHIM7HLIHbIELMH=RLHLAIM}ImtfI.tPHmt9I,$t"HL]A\A]A^A_ÐLؗmLȗfDH踗fDL託fDL蘗I.uHI.AuImD)E1H H=xI&D@LHLIHtImzf.HBIILl$H$HD$t'LoIHE11LLL萛tH$H@uHHH5vE1H81i@AuDH=' H H5 vIfDAuHL!@sIH=q' 蜆IF@AusDLؕlHH5-H8}6L谕9ff.ATUHH=b H H9GH HHH- HHEH5J HHHHMHQHUHt4HtH]A\HHD$ HD$H]A\fDtHH -H=#H1]A\L%)& IT$L|HHr H- H@H Ht[HE$tfH=% H@賚fDHt$E9u*HHċfDIL9LL$H\$Ld$DL9uuM9uuLHVIHtH;?L;=M9LI/oLL\$L$MH\$Ld$y;L$ L$Ht8RfDLL\$L$MH\$Ld$O4&fDHL$0HT$@ILL xHix%LL$0<HQ?RTfDLD$,ԇD$,$H LxIwHH{HLT$L$L$LT$HRDD>DD>ff.AWH AVAUATIUSHHhL  L-_HD$@HHD$HHD$PLL$0Ll$8HHLHH0HOHHIH H|HIHHI?SIH5άH8L A1YX0ZH H=E1GHhL[]A\A]A^A_DHHLH\MHI$It$HH=AHEHv HH AUjRPjRLPj5 I! IHEHPM%HHEXH藅KfoLvLY)D$0MnLL$0\@LYMHH  LHL\$L$0 L$L\$HIIHD$0M~~HML5z H1DHH9L;tuM4MLt$8IRLvLt$8LLL$0f.LLYLL$0MLt$8DMHHEH ?:0H=]IF0L\$E1HD$ L$Ld$IHH\$LMDO|L9HH9CI9G{ KA XHSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A Iw0IHA@IDDAADD>E9u*HH蔅fDIL9LL$H\$Ld$DL9uuM9uuLH&IHtH;L;=M9LI/oLL\$L$MH\$Ld$y;L$مL$Ht8/fDLL\$L$MH\$Ld$O4&fDHL$0HT$@ILL H9x%LL$0<HЁQ?/TfDLD$,褁D$,$HLxIwHH{HLT$L$фL$LT$H/DD>DD>ff.AWH AVAUIATUSHHHD$@H$HDŽ$H$ HL4H, HL&HILd$@HHD$HLHD$PHD$XHD$`HD$hHD$pHD$xvHD$HHh I|$H9t:HXH>HqHa1DHH9KH;TuHH5 LHHH|$HHHGH5t HHHD$PH|$HHH/eH|$PH-HD$HHH9H9%H;=裃AƅvH|$PH/HD$PEI|$HC H9ZHH5 LHHH|$PHH5 H9{HGH;JP HGHE1H-H/HD$PE{H5 L/HD$PHH$H9H9H;=蘂Å$H|$PH/HD$P H_HD$I9JIT$ID$H5( LHD$HHHH|$PH111rHD$HHH|$PH/=HD$PH|$H2>HHH|$HH/ID$H5 LHD$HHHHH|$HHHGH5 HHHD$PH|$HHH/HD$HH|$P=IHnH|$PH/HH$HD$PH9XxHHHH=H|$PH9HGH5 HHHD$HLT$PHI*NHD$PL~HD$PH 5HD$XIHHD$PHD$PIG譀HD$PHHH1H9X#HH5HHHHCH5 HHHIL|$`MH+HT$`H5 H|$P€H|$`H/ HT$PHt$XHD$`H|$H^HD$`IH H|$HH/LHD$HH|$XH/HHD$XH|$PH/&HD$PL|$`HD$`L;|$ IH5V MwHHZ IH5 HD$PHH HD$XH5H9G)!HGHD$XHp!HWHHH|$PHT$PH/*HD$XH|$PH$H$HDŽ$ H|$XHD$`HtH/HD$`HD$XH H|$PH/:HD$PH|$`H/HD$`txHL$xHT$pHHt$hHL$HIZH|IE L|$(IHD$LLl$H\$ HH|$H`HL$LLHL<L@|LLL2|LLL$|I)HuH\$ L|$(HD$H8HD$`HHD$`H|$hHt H/aHD$hH|$pHt H/0HD$pH|$xHt H/H5 1HHD$xHD$xH+H\$xH!H+HD$xHD$H: HHHVH5?SL AH <H81}X%lZH H=1HĘ[]A\A]A^A_DHuL&Ld$@DLyL%= 1M fHL9 L;duM$Ld$@M IODwH/HD$PH4I|$H9FH5O LHD$XIHNH5 H9H@H;CF1II/HD$XH5 L9HD$XIH H5 HDÅ H|$XH/HD$X%H5 LHD$XIH%H5 HHD$PHo%H|$XH/HH HD$XH9X%HH&HL=L|$XM&H5 LEHD$`IH%H|$XH/H\$PHl$`HD$XH+HD$PH|$`H/HD$`HHPH9H9L#H7H%HL=#L|$`M"H5f LHD$PH9%H|$`H/?HD$`cyHD$`IH`%H H5 HyDHT$`H5 H|$PHD$XIHy$H|$PH/T"HD$PH|$`H/#HD$`H|$XH/l#HD$XHH'H9X#HH"HL=L|$`M#H5ELuHD$PH"H|$`H/JH5Q LHD$`,HD$`IH!H|$PH-C1HD$HAH9o fHnfInL$flI4L)$H|$HHD$XHt H/zHD$HH|$`H/ HD$`L|$XM!H|$PH/IH5lHD$PL|$XHD$XHHm#IH5HD$PH"HD$`H9h!HPHT$`H!H@HLAHH|$PHD$PH/VHD$`H|$PLHHDŽ$H$H|$`HD$XHtH/HD$XHD$`H"H|$PH/HD$PH|$XH/HD$XpHL$hHT$pHHt$xHD$Hl$HHI HLXIH911HLHD$XHHH5kLSoH|$XH/11HLHD$XHD$XHH7LL( H|$XH/wLHLHD$X H/H|$xHt H/!HD$xH|$pHt H/!HD$pH|$hHt H/!H5 1HHD$h`HD$hH+!H\$hH!H+1!HD$hHAHD$HDHH9HuH;<fDHXHt,HqH~K1HH9t7H;TuHH9rHuH;`fDHHZH9XmHAHHL=-L|$`MfLLoÃwH|$`H/ HD$`DHH H9HHHHL=L|$`MH5 LAHD$PHbH|$`H/HD$`qqHD$`IHH HH HD$`HPI|$H5 HD$XIHH@H;^IHD$XHD$HHH|$XH/^HD$HHL$`HD$XHpP H@1 =HA HHD$HHHD$`HHHeHC(TtIHw D$HHAA  IQ0IIH@HDHT$E1Ld$0E1ILl$8MMΉ\$ AHr0HzH@HD;D$I1LLoIIIJTz  HZHtLH)I9 J  uHrHkl=fDAhlA 1E1H|$HHH/LT$PMt I*H|$XHt H/Ht H+H|$`Ht H/|H /DDH=1MI/LHD$kHD$kLT$P_Lk`HxkskkzfD[kGfDID$0E1Lt$MHD$ ILLl$MDMdM9HI9EI9D$ A} q A|$ I IUI;T$ID$IuH9@H@t HEU EL$ DD@@8u}A  I}HA@HE|$ A  It$0IHA@IDDA_ A" DDE9u'HHlHL9fDL AM9uuM9uuLLLL$2jIHt\H;L;%LL$M9LmI,$pmILl$LMLt$y9HD$@lH?ltDILl$LMLt$O$@H=q H5 1CHD$PHHuH|$PH/tGHD$PH|$HAlAE11Hy%hrhH=-(YI1AmA*E1ADShfDChfD3h=fD#h\fDD$ LHډHHH|$HD$XIHH|$PH/UHD$PH|$`H/3HD$`H|$HH/HD$HH|$XH/=HD$XIH57IH IH5HD$HHHHD$`HH9GHGHD$`H|HWHHH|$HHT$HH/HD$`H|$HH$H$HDŽ$ H|$`HD$XHtH/ HD$XHD$`HH|$HH/ HD$HH|$XH/HD$XI UdHL$xHT$pHHt$hHHl$HHHLL11LHIHD$XH211HLHD$HH HT$XHL= H|$XH/HT$HLLHD$X  H|$HH/HD$HHNH|$hHt H/HD$hH|$pHt H/HD$pH|$xHt H/H51LHD$x[HD$xI/L|$xMuI/r HEH5?jH LT$PAxlAE11HcGL=M IT$I9HXHg HqH~1fDL;|qHH9uIOHq1E1HRH5AlAH81ifDH|$HAlABeiH2XiQAzlALT$PAlAbA}lA1E1]LT$PAlAhHPfHAlA1E1DAlA1E1H/?HD$P-bAlAOrhLD$bD${HHT$(K`HT$(?MI)@HD$HLT$PA|oAGaaaL_!L_dHAlA1E1H|$HAlAaH=HH5?THHHH5Lt$H8aLL$`It$H|I}HXH=vYQH+AlAf;`HD$HHIfHA HhHD$HHHD$`HHHIHA(8gIH[HD$HH|$HAlA4H"IHD$H\$LLL|$I IIMIf.LL%GHLH0H3I7H0HII)IuH\$L|$_[u_k_a_IAHHD$@H|$HAlApeHH1AlA_HDI|$HD$P^5^DDH|$HAlA^E1AlAUH=UHH5QHH=3OHi^H;M^IHtAH9AH9DKL;=>LzbI/A7ESAlA1E1IdIaAlAL]*H=HH5PI1AKoAD?] ]AMoAD1E1H=MI;K]A]7]H*]H]DD]\HD$PpAoAL1E1oH|$`Ht H/|HD$`H|$HHt H/iHD$HH|$PHt H/HD$PH|$XHt H/H QHD$XH=EHL$`HT$XHHt$H辻Ht$HHL$`1HT$XcHD$PHH1L0I/HH|$PH/HD$PHH;-#AH;-DH;-H_HmAEy9A$pHHL$x1E1HT$pHt$hAN0ApHl$HE Ht Hm~HD$HH|$XHt H/nHD$XH|$`Ht H/[HHL$xHD$`HT$pHt$hHHHmEeHZZLA|oAG}ZLT$PHD$Ho1AlAop>Zz4ZAmA!H5OL[A mA 1@YaYt1AYoAFzTH=XHH5LIoH|$HAlAH|$HAlAAmH+A!1EYAlA1E1A.mH$11H|$HA[oAF3XLXI/E% LXH$11AfoAGXlXXX1AoAN1AnoAG TXHA HLHD$HHHD$`HHH(HC(_IH:H?D$HLT$PApoAGE1Sf.GADEAoI/AN1E1EH;LPXHA!1lWH;PZLWHHtFH;H;dH;-zHt[HmlAmA*1E11AoAFAoHH51AlAH8gWSH$1H$11LzVAmA!LT$PAoALIV?V>5VA(nA71E1HHI9HuL;=UUU1AmA.pLAN1E1UXHUuUUU|wU mUcU.HVU AmA.1E1Ap+UKA pHmzHU1ffA.GEAT=1AoAFxTTAnA@H|$`Ht H/HD$`H|$HHt H/HD$HH|$PHt H/HD$PH|$XHt H/H DDHD$XH=H|$HL$`HT$PHt$X葳Ht$XHL$`1HT$PZHD$HHH 1HH+I^H|$HH/EHD$HML;%L;%L;%ʰLWI,$y;AoHD$HL$h1A:HT$pHt$xHAnLd$XMt I,$HD$XH|$PHt H/HD$PH|$`Ht H/HD$HL$hHD$`HT$pHt$xHxHHD$nI,$aLqRgR&]RHPRHGHD$HHHWHHH|$PHT$PH/GHD$HL|$`1AH|$P1AnA/QQCQVQi1AanA9`1ALnA9MAnA@AnA?AnA?Ld$XLl$`THLHLA,pCHD$HHD$XHD$`H=HMH5N9DIH=AIH=HH5DIH|$HAInA91AGnA9SPPhH$E1aH|$HAnA.1AnA.ArnH+^A:1H=HrH5sNCItP.PALOH;ެaH5LN1A1nA/rAniAnAAAnK1AnA.8H|$HAnA/H= @IrXONOLAO07O1A&nA7ApnA:1AnA.H=?IEHA:1NNHNN=H|$P NFNYL|$X1AFpAN'A*oA:H]NGLt$`Ll$PyQLAoHLL2HD$XHD$PHD$`AoAWH?AVAUATIUSHHHL-FHD$0HD$8Ll$(HRIL$t$9g+1kH=H:H5;&1I2fDDu@LD$D$L $=LD$D$L $@H=D.I@LLD$Ht$L $;L $Ht$LD$i+11E1E1I.LLL$T$4$0=4$T$LL$HLD$LL$H4$b;H4$LL$LD$dOHvHIyHL$<$D=I[7+=L $ImE17+6DU=L $IrM1E1+7DB!=L $IE17+LH;HEHHEI/lI}H5HGHH-IMH0I9GI_HIoHHEI/BfInfHnHt$0Hfl)D$0H+I0MHmI.IELPHHH(I$HHL[]A\A]A^A_HHHHH HIHHH?L HLIL@HHeSHÃH5_H81?X)ZH hH=jn HH1[]A\A]A^A_@HIHML6Lt$(L1L;DHH9uI@0E1HD$JtI9"HKI9@H9FMEAx ~ IPH;VHFIxH9AHAt HEX DV DD@@8wA IxHA@HE|$A L^0HHA@IEDAADDE9L$H(HL$H:HL$L$:HnHL$(HT$0ILL ɁH苮Lt$(+)@L7H7mL6!H6L66fDH6Ht$8LLLt$8HD$0MIDA*AHmuHe6DDH |H=kLHH1[]A\A]A^A_H=HRH5SN)IOfDH=&I5@HHEI/A*AoL5bH-L%"HEHHtOH=S4uT1HLH`9HbHA&*ASHmH1L;HHuA"*ADM9uL9uLǺHL$L$h5HHH;ML$H;=HL$L9HL$LD$H<$ 9HL$LD$H<$H/II9fDO4MLt$(HIDA8*A:IM4fAL*AAY*AKHL$D$L$3HL$D$L$)@LHL$Ht$L$2L$Ht$HL$*HHL$LD$H4$1H4$LD$HL$HvHrIxHHDDsDDf6HH؏H5aQH83)H2DAWHAVAUATUHSHHH HD$0HD$8H\$(HILtf@tAk.H.HlH=I@H=ѹI@HuLC(H ntE1H=r#fDL}M#LuIIHmfInfInHt$0Lfl)D$0I/IL'DLANtDLAPtgD-IoH=ٸHZH5[IfDLStADD$HL$"'D$HL$Q@HHt$HL$^%HL$Ht$i_H=ItI@#-H2I$LUtAHI$mAUtjHHL$Ht$$Ht$HL$HvHYHyH/)DDIDD\ZH ek&H=ZE1HhL[]A\A]A^A_DH&H|H\LM$HIt$ E1H=;IL:H jAPQjAPAQLj 5PHIHPHvHIt5H IH1Hm?H>$2fL($fDoHVLI)D$0MHD$0,@LIMHLHLL$īLL$L H0IHD$0MHML5 H1DHH9L;tuITHHT$8IQHVHT$8HHD$0zf.HLIHD$0M~HT$8QH ciw\H=XfDHIH .iru\E1H=PX˱fDLaIF0LL$E1HD$ Ld$IHH\$LMfO|L9H+H9CI9G%{ [A pHSI;WIGHsH9@H@t H{ E_ D@@8@ ?HsH@HHE|$ A Iw0IHA@IDDA9A<DD>E9u2HH#L~IL9LH\$Ld$fL9uuM9uuLH~!IHtH;g~LP~L;=~M9L2%L#~I/aLLL$H\$MLd$y-+$Ht6(\j@LLL$H\$MLd$KTfHL$0HT$@ILL jH著xHD$0]-\fDL LD$,D$,LI}+@H@L1}@L L}ya@IwHH{HLL$#LL$L|H!\DDD>DD>ff.AWHWAVIAUATIUSHLs|HD$PHHD$XHPHD$`HHD$0HD$8HD$@HD$hHD$pLT$HH=LHCHkHJcH@HFHD$HHFHD$@HFLEHD$8HHD$0II5MMIHXM1HL9KH;\uIHD$8HHUIL-cH1fDHCH9HL;luIHD$@HIMHLHL\$LD$莥L\$LzHLD$HD$HIM~VHL$0HT$PMLL gH觔uLL$0HL$8HD$@HT$HLz&fIuMgLL$0HL$8HD$@HT$HIHIvHH=!AHEARj5IPj5QHj5NIHEHPMBHHEHĈL[]A\A]A^A_H&HIH bHbAHMEIHHgyHfH5BATL gH81"XIZH bb6H=QE1[HGHVHT$HHFoHNLHD$@)D$0HLAHD$0@LAHLHLD$L\$xHD$0HLD$L}L\$LxI @LqIE0LD$1L\$HD$ HLLd$IIHM|L9HxH9EI9G} KA `HUI;WIGHuH9@H@t H} E_ D@@8@ OH}HA Iw0IHA@IDDAADD>E9uFHt HLuwu)fLD$L\$HLLd$IfHCI9HL9uuM9uuLH&IHt^H;wLvL;=vu M9FI/tLD$L\$HLLd$_f.HD$@HiHHNvH Y_H5p?jL dAHcH81Y^I@LHC0LD$1HD$ L\$Ld$ILIHf.MlM9H[vI9GI9E]UA FA} WIWI;UIEIwH9@H@t HA EM D@@8u{@  IHA )Iu0IHA@IDDA!A1DD.E9u0H.H2LuDHH9HD$8 HHHtAH5=jL bH ~]H8Ha1+_IAXfDM9uuM9uzLLrIHkH;WtL@tL;-tu M9Imx LLD$L\$HLd$fLLD$L\$HLd$IRfHHEtH \IH=KI@LLs@HuH@HHE|$ LhLYs1IwH@HHE|$ HtIII@LD$,LrD$,#@HLr,@LLr @IwHLD$,<D$,LrlLLurLhLYriL\$zL\$HfIIuHDD>DD.DD>gDD. H ZAH(}IQAWHwAVAUIATUSHHHHD$(HD$0HD$8HuIL4HHH.LAHl$(MHEH Gu [HqH9H;pL`pMI|$1IHHHAT$I.IrML5bM9iHqI9D$I9FA|$ A~ IT$I;VID$INH9@H@t HA|$ Av 8m@ @IL$0MD$HIDH@ IN0IH@IDHȃ.WDA9HmE1HAI,$EOHHEoH\H5g8SL @XAH dXH81XK,ZH 4X;H=GE1֠HHL[]A\A]A^A_@HuH.Hl$(DH5HaH5H=ݬ1jIHHRI,$,q;DH WH=GE12WDLyH--1MHL9I;luI,Hl$(HMGH5Hb8HEE1HEH50HHHEIMGID$L5+HH H=. 1LLI-MTI,$H"mI9FLpf. I.AEXHEHH5tHHlIMnIFL%{HH H=- 1LLIuMI.ID$ID$HPH}H H LHcЉH9P g I,$/I}A]PHH5HGHHyЅ H@lHIHmMI/LfDH@hHk H@H^ 1HI H kI9uuI9I,$uLPH5H=1{gIHHI,$-xLT$t$t$T$DH LLHL$;IHH; kAH;jDAHL$I93LI/A,E%,AwE11I,$QE1E1H}HEL I$@AFDH}LeI$I,$IHHEH9H;iLppMt9I~t21IHHHAVI/Iu0L &fH@hH H@H 1HIM[H5<LLI.IHn HEH9H;iLxpMtKItDH HD$HHAWLL$II)u;LHD$ LD$'H@hHcH@HVHIMW"-A{DI,$AuDfDLLD$K MLD$u'+I,$ 1E1E1Az-I.tLMtI(t1H LQDE1H=@HE1DL fDLLD$ LD$@H}%HEL0IBDL TLx :LD$b D$OHE0E1HD$@KtH9HgH9EH9F } ~ HEH;FHVHMH9@H@t HD] DV DD8A H}HA@HE|$A HN0HHA@HEʃ  DA9u1HLL$H LL$IM9fDH YfH9uuH9uuHHL$LL$H HHt^H;1fLL$H;=e?HL$H91LL$H|$ LL$H|$H/Vy"HD$( H+;,`K,OH}HEL@IH5LLLD$ LD$=I(;HEH9H;dHXpH4H{) IHHHSI/IMH5}LLLD$ LD$I(&H5wLL <I.JHj HHLE1`H}LuILqLHt+AT$HEfDL1fH@hH#H@HHI@I/ELJDfIHHHI/ILHD$LD$rE11-Az~HHcH5+H8N HH Kx,H=k;莔fH5H=ʠ1^IHHGI,$,u+D1aIHHHDH}HEL@I`@HL$(HT$0ILL yOL!|Hl$(L1Lk IH[p-A#H Jw,E1H==:`H} LE IL8-H iJH=9E1*LLL1L IHCI.uLH I-H=z9蝒AD$AT$HH HHcЉH9;fAD$AT$HH HcЉH9c@H}LE(I1ҾHxIHH5юHL.I/1ҾHxIHH5 HLI/eI$ML1IHxHHbDIn-f,pD$D$HI.uLH 5Hs-E1H=7ҐD$-A{HE11 -A{6D$LL$D$LL$@I~-3fHLL$Ht$Ht$LL$)HLL$Ht$nHt$LL${p-AHrH]H51H8W@HvH?H}H(-A{fDIHSHHOf.LL*-AzE1DH"H\H5DH8I|$HfDIvHLHD$LD$D6&-A{ L6-E1|A,pDDL8WL+DE1"-A{E1&-A{@,LA-}LC-MA}E1ZLM-~LO-MA~E1/@H@`HHHLIHtxH[I9Fu*fL舄I. L4LH5F[IHuP-A1,AwyHHZH5CH8PHH Cq,H=r3蕌-AzE11E1 H Cu,H=83[LNH Cz-H= 3,QL-f.AWHoAVAUATUHSHHhL-FZHD$PHD$XLl$0HILI/IL9d$ HEIHPHUHHeHH>>H .>HIHH>H?L pCHLIL@HHTUHoBH5H81XI1ZH =qH=-E1蒆HhL[]A\A]A^A_HIHHD$8HD$@HD$HLL=Mt$LLHH H@HH* HLLHHzLL=Mt$LLUHH H@HH LLHD$@HHH@HD$HH;xS HGHD$HH HWHHH|$@HT$@H/ HD$HH|$@Ht$PHD$PHD$XH|$HHD$8HtH/`HD$8HD$HHH H|$@H/: HD$@H|$8H/ HD$8IH HL`M9t MCHPHuL0LHMtIMtI$MtIH{ LD$LL$|HLL$LD$HHD$8I IfInfInHD$8flH8HXLhLHHt H/Ht H+Mt ImH51HNHmIM I.'L@L6Lt$0L1M;DEHH9uI@0E1HD$KtI9bH RI9@H9F}Ax b~ IPH;VHFIxH9AHAt HEX DV DD@@8A uIxHA@HE|$A ML^0HHA@IEDA~ A DDE9^LD$HgHL$HHL$LD$&fD0H HL$0HT$PILL =LCjLt$0C;1/@LLIfDIfDLpL`SOfDL@AE2AE11H|$8Ht H/H|$@Ht H/H|$HHt H/Mt I.MtI/tJH 8DDE1H='讀HHEHHEHHLxfDkZfD[^fDKbfDL8^1LLI.Iuf.LME1A2AfDHHl$8EfD%fDfDHANLE1Ax2AH8Zf.H=!HiH5iHfDE1E1A42AHMLAz2H8HI.AE1HLLHHfDAG2E1E11AfDM9uL9uLǺHL$LD$HHSH;LLD$H;=LHL$L9HL$ LD$H|$HL$ LD$H|$H/'II9fDO4MLt$0H)D+IH|$81E1AA62H@H=I~,H8@ HuHGKH5 H8 I.A2AE1E1M4CMA2f.Ht$(M11E1E1A92AHt$(M11E1AA>2f.E11A@2Azf.H=A}HfH5fIfDL0LHMI1AB2A DH=|I@LAE1f.HEfDHH|$@*ffDfD6fDLA2E1E1|AQHHH0IH5 E1E1AG2AH8fH((L( fDH'HL$ D$LD$HL$ D$LD$f.LHL$ Ht$LD$ LD$Ht$HL$ pmDHHL$ LD$Ht$Ht$LD$HL$ B5DHvHIxHHHLE1E1A1AH8<HHLA1H8HD$@HmAE1E11A1Ht$X11_H|$@Ht$X1NH|$@Ht H/HD$@H|$HHt H/H 01LL$H=T LD$HD$H9yLD$HL$HHT$@Ht$8LILD$LL$UHL$HHT$@1Ht$8LD$LL$LL$LD$H1HHLD$LL$HD$BHmL\$LL$LD$II+ML;FL;F9M90LLD$LL$LT$LT$LL$LD$I* y~A1ILLLE11AdDDHAE1E11DD]E11A2AA1Ht$8Ht H.*HD$8H|$@Ht H/.HD$@H|$HHt H//ILLLHD$HMdI*wLLD$LL$LD$LL$61MLD$LL$`LD$LL$LD$LL$_LLT$LD$LL$LT$LD$LL$ HL\$LD$LL$HD$ LT$ L\$LD$LL$HLD$LL$VLD$LL$LD$LL$8LD$LL$LD$LL$LD$LL$A1A1Ll$@H\$HLD$LL$Ht$Ht$LHHA1ELL$LD$HD$8HD$@HD$HnA1E1E1A1ARAz2OE1Ax2A3AWHwfAVfHnAUATIUHHH0SHH$H A)D$pfHnflH|$HDŽ$H$)$H L4H H HHLEHD$pML-vM 1HL9k L;luIHD$xH MhM~]HxLH7mHH$ME+HV HFLEH$HFHD$xHHD$pMH|$pHGLIGH I I [IIH\$xH HD$XHD$`H$L(H\$HD$hhE1ɹ1HAHAHD$XIHHD$`HH|$XH/ HD$XLd$`HD$`At$HCI|$ Mt$H5$uLHD$HúLt$ qpHtHsLNf/F L-@L9 {HD$(HDHLpMt M93 HPHuHLhHL$Ht HD$HMtIMtIEHeH `H9HH`HHH=`H|$`HDHGH5oHHHD$XLL$`HI)H|$X1HD$`H v?H$H9OfHnfHnfl)$)H|$`HD$hHtH/HD$hHD$`H|$XggHH/-HD$XH|$HD$XH HD$`HHD~D$hHL$HD$hHD$`D$XHD$X@HtHHD$(HHMt I.qMt ImQHJdH _H9HH^H3HL-^Ll$hMIEH5wLHHHD$XHH|$hH/C HD$hHD$hIHLHHD$hHXHD$`IHHcH .^H9HnH^HHL^M'ICL\$LH5enHHL\$IMsI+H5jH|$`LLL$LL$I)Hl$XHT$`Lt$hHELM H=.HT$HT$LHAIMH|$XH/fHD$XH|$hH/>HD$hH|$`H/IEAuHD$`I} IEHD$8H{fH= nI*HzHD$HL$Ht HHHHD$HD$H-iLHIPLD$0HHT$(GIHNH@HT$(LD$0HHLLIH.HD$H--iLHIPLD$0HHT$(HHEH@HT$(LD$0HHLHD$`HH'H@HD$hH;:HGHD$hHHWHHH|$`HT$`H/qHD$hH|$`H$H$HDŽ$qH|$hHHt H/HD$hHH|$`H/2HD$`Hm(H|$HD$(HD$Ld$H1Ld$ Lt$0LH`HH HD$H\$@IHHD$HD$8Ll$8MHLD$HMLLLHH\$H;l$uLt$0Ll$8H\$@Ld$HH|$(H5rs1L5I.HHHmIEI,$ImM H+nMImLJDHtHHFH$o&H>)d$pLEHjLHLD$cLD$HHD$pL}I6DH "AHH8Hd&H5ATL .'H81XeZH !H=E1ujHĸL[]A\A]A^A_HH !H!AHMEIvfMyHcGIHcHEIT@fDHDGII 'DDGII If.HFLmHD$xHHD$p8fDH|$HD$`IH*HD$hHHvHD$hHD$`HD$`HCf.HLhHL$HfDIE0LD$1HD$(Ld$ILMfMlM9 H7I9GI9EUMA  A}  IWI;UIEIwH9@H@t HEW EM DD@@8urA u IHA  Iu0IHA@IDDAADDE9u'H%HqHH9HD$xUHP HH5H H5jL '$AH@#H81sY^eL 5M9uuM9usLLLL$ IH_H;5L;-Q5uLL$ M9, Im LLD$Ld$LLD$Ld$IRffDHD$HZH@H9t?HXH+HqHF1f.HH9+H;TuH%ZH TH9HHTH HHTH\$XHHCH5cHHHILl$hH|$XMH/ H|$H5?aHD$XHGHHHH\$XH{H|$hH-s3H9oHYfHnfH$flźH$)$}HD$`H|$XH/q HD$XLt$`MH|$hH/ H|$`L5u3HD$hL9H;='3 H;==3 :ADžH|$`H/HD$`EvH5`H|$]2HD$`IHH5u2HHD$hIHGH|$`H/HD$`H|$hL9H;=m2 H;=2 ADž?H|$hH/HD$hEH5=iH|$1HD$`IHoHD$XH9h8H@HD$XH&IUHAH$HH|$`HT$`H/HD$XH|$`LHHDŽ$H${H|$XHD$hHHtH/wH\$hHD$XH|$`HH/H5-jH|$h1HD$`hHD$`IHH|$hH/=HD$hH|$`L9H;=0H;=1sÅ}H|$`H/HD$`tBHgHL5g@fDHH9HuH;0fDHgHL5~gH=nH$HDŽ$L$WHD$`IHz HMH|$`H/LHD$`E111D$2H|$XE1E1AgH|$XPgH4H/LL$` HD$XMt I)HD$`H|$hHt H/H 7HD$hH=k.aH|$(HL$hHT$XHt$`1wH.H9EP HEH|$]IH HD$0L\$0HI LXHHHD$0=LL$0HHt HmI)H|$`Ht H/urHD$`H|$XHt H/uTHD$XH|$hHt H/u6HD$(Ht$LLHD$hHL@LIyHuH-H5H8fDSH IfDD$H|$`1A fE1E1E1E11DHt H/H|$hHt H/Ht HmMt I+Mt I)T$H nDH=R_MtI,$E1MtImt.MtI.tHrwDLE1LfDLLL$LL$cLdLL$L\$LL$L\$fLL$L\$aLL$L\$fHLL$L\$>LL$L\$1LIfDH5fH=Zj1;(HD$`IH HIH|$`H/?HD$`A'fD$H|$XE1E111E1E1HtH/t H|$`fLL$L\$H|$`LL$L\$DLLL$[LL$D$"AKfD3fDL HL$pMLHH$L FEeT@L8IHA@HE|$(|L,HLL$0HH=1]HCef[fDKfD;fDL(LHcLD$ D$ IFL0L D$:E1E11H|$XE1AgDH=1\HIH5I薿IIuH+HH|$`BH5H=[躼ID$:H|$`AggH=3\HIH5IHLE1LHIHD$:E1E11H|$XE1AgD$:E1E11H|$XE1AgkaLL$`LOEHeH=a[輻HRge1D$:E1E11H|$XE1Ag6D 0HGHD$`H4HWHHH|$XHT$XH/HD$`H|$XH$D$:AgH|$XE11E1E1D$:E11E1H|$XE1AgH=YHGH5GLIDDLWLL$`kgeH=YIND$:1E1E1H|$XAgL\$IRH|$XmgD$8E1E11AgHD$(Ht$LLLL$ 1E1E1HL\$2EH|$XL\$LL$ 9YD$?AhH|$XE1E11E1jDDD$:AgE1yH'H A OIHH%H5 1AgH8UH|$XE1E1D$:H&HH8D$CE1E11H|$XA=hD$4E1E11H|$X1E1A$g!H%HA?hH84HD$`I.D$CD$4E1E11H|$XE1E1A&gAShH$HHD$9E1E1AgH$11H|$`H$1D$9E1AgD+D$9AgD$9E1AgL ceaH@`HHHHHH@L-J$L9uSD,L}IGHvTIItaHIHmHgHH5*3#HHH@MtHcEIHcHEID}EII Hm]D}EII IyH|$XE1E1E1D$CAh+D$E1E11H|$X1E1A#fH=^UHGCH5HCøHbD$(H|$`AtfD$2E1E11H|$X1AgD$(E1E111E1AvfI,D$(H|$`AyfygHRLwMiHGIHH|$hHD$hH/AH|$hfIn~L$XH$H#HflH$)$UkHD$`I.8L~+D$(E1E11E11AfTJ?D$(Afv% D$)E1E11H|$X1E1AfTD$)E1E11H|$X1E1Af/D$)Afe&H$E1H@`HHHHIHL9huTID$6M|$IGHvbIItpLII,$LuH5 HIHuHmHMtIcD$IHcHEIE|$AD$II E|$AD$II IoD$*AfD$*E1E11H|$X1E1AfD$*E1E11H|$X1E1AfrD$*AfJA?hH@`HtaHHtULIHtHL9hu"LEI.ILH5| HIHuIHuHH5H81HHzH5cH8 nfDAWHTAVAUATIUSHHHL-vHD$0HD$8Ll$(Ht~HL4HHu{HHIHT$(HI$It$ H=iMHEHN\IHEMHHEHHHHHH HIHHH?L HLIL@HHUSH H5vH81 X-ZH XH=jE1NHHL[]A\A]A^A_LyMLHHT$(LS1L;DHL9uI@0E1HD$JtI9ZH+I9@H9FAx ~ IPH;VHFIxH9AHAt HEX DV DD@@8A IxHA@HE|$A mL^0HHA@IEDAhAkDDE9VL$H`LL$HLL$L$ 0HHT$0HL$(ILL Hk5HT$(3-@HHEt&H 9).H=FL@HȽfDM9uL9uuzLǺLL$L$HH@H;L$H;=LL$uwL9trLL$LD$H<$LL$LD$H<$H/tSIM9fDKHHT$(IODIfD$LL$L$ּD$LL$L$LLL$Ht$L$ L$Ht$LL$RGHLL$LD$H4$ҺH4$LD$LL$$HvHIxHpDDDD-AWH?DAVAUATUSHHL-FH$H0 H|$0HDŽ$H$HDŽ$L$HILH HH HHH AHOL EHIHLOHHeH SH5H81X5iZH hJH= JHD$HD$H[]A\A]A^A_DH HNHnH$L>L$(oHnLq)$ML$HD$xLHDŽ$HDŽ$蚾IH HV H(hE1ɹAHƺLHD$HD$xHiH$HH|$xH/ H$H7HD$xHDŽ$HD$H<H9XWH7HHH7HD$HD$xHMHH@H5AHHIL$H|$xMH/ H<H527HD$xH9pH7HCHH6HD$(HHHGH5hFHHIMH\$(HHD$HHP HI9FHD$LMH$HDŽ$H$HNH$/_HD$HD$xH|$+I/ H$H H9G HD$xH$HDŽ$H$^H$H|$xH/ H$HD$xHD$H@H$H/ H$H;=HDŽ$H;=s L9j ÅH$H/9 HDŽ$UHD$HXL9k .HD$ HHLxM9t M HPHuL0H@HD$MtIMtIHD$HtHH9H54H9pH4H+HH=4H$HhHGH5BHHHD$xL$HI)H|$x1HDŽ$H &H$H9OfHnfHnfl)$\H$H$HtH/zH$H|$xBjHDŽ$HH/HD$xL(HD$xHHD$H$H~$HDŽ$HDŽ$D$xHD$x@Mt I.Mt I/Ht$HtHHD$ HHH7H52H9pH2HgHL=2L|$xMIGH5JLHHDIL$H|$xM6H/h H_7HH2HD$xH9PgH&2H<HL=2L|$xM`IGH5Y?LHHIH|$xMH/H$1HD$xH H$H9OfHnL$D$)$ZH|$xH$Ht H/vHD$xI.eL$M1H$H/H$HDŽ$HL$HD$8IFAvHDŽ$I~ HD$ HnOH-=HD$@HD$0HLMoLiHD$(HH@HHaH|$(LLHD$(HHD$0L=P=HLLmL HHH@HHiLHH$HHH@H;3gHoHZHGHEHH$H$H/H$H$H$HDŽ$XH$Hmr H$HH$H/H$HDŽ$H/rHDŽ$ֶH|$@HD$`JLt$hL|$ E1HD$HLHHHD$PLHHD$XMHD$0fE1HhHBHT$T$CIXM9u sX^Il HD$PJ 8LDfHfY@H9uHD$XI9tHL$ LHYML|$HL9l$@BLt$hH|$`胯H\$(H5G1H IHHD$HHPM<I/ HD$8HT$HHHD$HHnI.H\$8H\$HT$HHD$HHNHT$8HHHD$HHH賯fDHiL=71H]HH9cM;|uMLqL$MIL$H-DAH1HH9I;luI,HH$IfDH舭{&fDHh[bfDLH;sfD+fDIG0LT$E1HD$H\$LILMfDNtL9H H9EI9F } ; A~ E HUI;VIFHuH9@H@t HDU EN DD@@8u{A  H}HA@HE|$A  Iv0IHA@IDDA ADDE9u&HE1HmAIM9H\$fL9uuM9uuHLHHtH;H;=L9H|$ ǯH|$ AH/ ElLLT$IMH\$EyOHDŽ$賮H NAL @Hi@LLT$IH\$MOH趡HD$(HE11A{j E1A}jH|$xE1HD$(AkAjHD$(Aj豛#觛H|$xE1E1HD$(AiHD$8HD$lHAkfAWH.AVAUIATUHSHHHHD$0HD$8H\$(HIL4H2HHHIHT$(HHIuHIE1IH=HEHSjSSjSSjS56IHEHPMHHEH胚fDHHHHH HIHHwH?L HLIL@HHEUHH5fH81X0ZH HEH=E1(HHL[]A\A]A^A_LyMHHHT$(L -1M;DHL9uI@0E1HD$KtI9ZHI9@H9FAx ~ IPH;VHFIxH9AHAt HEX DV DD@@8A IxHA@HE|$A mL^0HHA@IEDAhAkDDE9VL$H`LL$HךLL$L$ 0ÛHHT$0HL$(ILL L[HT$(0@HHEt&H )k0H=&@H踗fDI9uH9uuzLǺLL$L$HH@H;L$H;=LL$uwH9trLL$LD$H<$蟛LL$LD$H<$H/tSIM9fDKHHT$(IODIfD$LL$L$ƖD$LL$L$LLL$Ht$L$L$Ht$LL$RGHLL$LD$H4$”H4$LD$LL$$HvHIxHpDDDD0AWH_)AVAUIATUHSHHH6HD$0HD$8H\$(HIL4H2HHHIHT$(HHIuHIE1IH='HEHSjSSjSSjS0IHEHPMHHEH3fDHHHHFH 6HIHH'H?L xHLIL@HHUHH5H81謚XEZH H=E1#HHL[]A\A]A^A_LyMHHHT$(L'1M;DHL9uI@0E1HD$KtI9ZHI9@H9FAx ~ IPH;VHFIxH9AHAt HEX DV DD@@8A IxHA@HE|$A mL^0HHA@IEDAhAkDDE9VL$H`LL$H臕LL$L$ 0sHHT$0HL$(ILL L HT$(E@HHEt&H .FH=~y!@HhfDI9uH9uuzLǺLL$L$褒HH@H;L$H;=CLL$uwH9trLL$LD$H<$OLL$LD$H<$H/tSIM9fDKHHT$(IODIfD$LL$L$vD$LL$L$LLL$Ht$L$誏L$Ht$LL$RGHLL$LD$H4$rH4$LD$LL$$HvHIxHpDDDDEAWH$AVAUATIUSHHHL-HD$0HD$8Ll$(HHL4HBHHHIHT$(HI$HIt$HE1H=8HEHHL &AUjPAQjPAQjP+IHEHPMHHEHӏfDHHHHH HIHHH?L HLIL@HHSHiH5H81LXKZH 2H=rE1:HHL[]A\A]A^A_LyMLHHT$(LY"1L;DHL9uI@0E1HD$JtI9ZHkI9@H9FAx ~ IPH;VHFIxH9AHAt HEX DV DD@@8A IxHA@HE|$A mL^0HHA@IEDAhAkDDE9VL$H`LL$H'LL$L$ 0HHT$0HL$(ILL _HHT$( K@HHEt&H yzKH=N@HfDM9uL9uuzLǺLL$L$DHH@H;)L$H;=LL$uwL9trLL$LD$H<$LL$LD$H<$H/tSIM9fDKHHT$(IODIfD$LL$L$D$LL$L$LLL$Ht$L$JL$Ht$LL$RGHLL$LD$H4$H4$LD$LL$$HvHIxHpDDDDKAWH'AVfHnAUATUHH(HSfHnHflHxL5kH<$HD$0HD$`HD$hLt$8Lt$@)D$PHPLELz LzrADI)P=HLL$zLL$%fLMIrzDL`z$HPzL@zAEDL zHL $ zL $LL $yL $}AqE+I6AsEI/Ly}^fDLD$,yD$,5qADHFLyHD$@HFHD$8HHD$0MLL$0HD$8HT$@H IHIvHH={AHEQHj55& j5Pj5EIHEHPMcHHE5HxL[]A\A]A^A_HHHVH HT$@oHFL)T$0FLHLT$LyHeLT$HHD$0HUIUDH AHHXHMH5zATL H81xXSZH Sv H=E1 HH HAHMEI@H H fHqLML-M1HI9L;luIHHD$@L{HFHHYHD$8HT$0{fDE1IE0JtHD$ I94fDHI9EH9F#A} P~ ~IUH;VHFIMH9@H@t HA} D^ D8@ vI}HA HHN0HHA@HEȃ5DA9u9HtQLL$HLT$rLT$LL$DIAL9IJtI9K fH !I9uuH9uuLHL$LL$LT$ pHHtMH;LT$H;=LL$uHL$H9;H/NdHD$8rH_HHDH OH5fjL AHH81tY^Sf.IE0E1HD$(@N|M9"H{I9EI9G A} A IUI;WIGIMH9@H@t HA} Aw 8@ IMH@HHE|$(@ IO0IH@IDHȃD>A9ufLL$Ht2L\$HLT$pLT$HtDSf.HHEH Ų SH=e{HL$0HT$PMLL eHSLL$LT$H|$tpLL$LT$H|$IMH@HHE|$ yHk\fDSfDLL$D$LT$kLL$D$LT$Df.LLL$Ht$LT$iLT$Ht$LL$DHLL$LT$Ht$iHt$LT$LL$TDHvHDLD$ L\$LL$LT$jD$ L\$LL$LT$ LL\$LL$LT$!iLT$LL$L\$LL\$LL$LT$hLT$LL$L\$IwHmI}HHDmHu*H uAD>JD>>SfDAWHfAVfHnIHAUHATIUSHxHD$`H)D$0fHnflHD$@HD$h)D$PHiHLHHH HLyHD$0LL-IH1@HCH9HL;luIHD$8HI_HLL$0HT$@H iNH>HFLyHD$@HFHD$8HHD$0MLL$0HD$8HT$@H IHIv H=AHEQHj5$5j5nPj5IHEHPMcHHE5HxL[]A\A]A^A_HHHVH yHT$@oHFL)T$0FLHLT$LyHjLT$HHD$0HUIUDH AHHH²H5ATL H81wmX2MZH íH=%E1e HH HAHMEI@H H fHgLML-mM1HI9L;luIHHD$@L{HFHHYHD$8HT$0{fDE1IE0JtHD$ I94fDH9I9EH9F#A} P~ ~IUH;VHFIMH9@H@t HA} D^ D8@ vI}HA HHN0HHA@HEȃ5DA9u9HtQLL$HLT$ hLT$LL$DIAL9IJtI9K fH I9uuH9uuLHL$LL$LT${eHHtMH;dLT$H;=LL$uHL$H9;H/NdHD$88hH_HHH H5֊jL AHH81VjY^Mf.IE0E1HD$(@N|M9"HI9EI9G A} A IUI;WIGIMH9@H@t HA} Aw 8@ IMH@HHE|$(@ IO0IH@IDHȃD>A9ufLL$Ht2L\$HLT$eLT$L\$LL$u2fK/H QI9uu I9utIM9DLLLT$HL$ L\$LL$cLT$HItwH;LL$L;=L\$1HL$ I9#LL\$LL$LT$fL\$LL$LT$I/LDLT$eLT$HtDMf.HHEH 5D iMH={HL$0HT$PMLL ڬH MLL$LT$H|$eLL$LT$H|$IMH@HHE|$ yHSa\fDMfDLL$D$LT$aLL$D$LT$Df.LLL$Ht$LT$I_LT$Ht$LL$DHLL$LT$Ht$_Ht$LT$LL$TDHvHDLD$ L\$LL$LT$``D$ L\$LL$LT$ LL\$LL$LT$^LT$LL$L\$LL\$LL$LT$^^LT$LL$L\$IwHmI}HHDOcHu*H AD>JD>> MfDAWHfAVfHnIHAUHATIUSHxHD$`HC)D$0fHnflHD$@HD$h)D$PHiHLHHH HLyHD$0LL-9H1@HCH9HL;luIHD$8HI_HLL$0HT$@H ٻNH>HFLyHD$@HFHD$8HHD$0MLL$0HD$8HT$@H IHIvHH=_AHEQHj55j5^Pj5oIHEHPMcHHE5HxL[]A\A]A^A_HHHVH HT$@oHFL)T$0FLHLT$LyHZELT$HHD$0HUIUDH dAHH8H;H5ZATL H81bX KZH 3H=ŔE1 HH HAHMEI@H H fH\LML-M1HI9L;luIHHD$@L{HFHHYHD$8HT$0{fDE1IE0JtHD$ I94fDHI9EH9F#A} P~ ~IUH;VHFIMH9@H@t HA} D^ D8@ vI}HA HHN0HHA@HEȃ5DA9u9HtQLL$HLT$}]LT$LL$DIAL9IJtI9K fH I9uuH9uuLHL$LL$LT$ZHHtMH;ԷLT$H;=LL$uHL$H9;H/NdHD$8]H_HH$H /H5FjL zAH H81_Y^Kf.IE0E1HD$(@N|M9"H[I9EI9G A} A IUI;WIGIMH9@H@t HA} Aw 8@ IMH@HHE|$(@ IO0IH@IDHȃD>A9ufLL$Ht2L\$HLT$[LT$L\$LL$u2fK/H I9uu I9utIM9DLLLT$HL$ L\$LL$XLT$HItwH;pLL$L;=)L\$1HL$ I9#LL\$LL$LT$$\L\$LL$LT$I/LDLT$[LT$HtD Kf.HHEH -WKH=2E{HL$0HT$PMLL SHqKLL$LT$H|$T[LL$LT$H|$IMH@HHE|$ yHV\fDKfDLL$D$LT$VLL$D$LT$Df.LLL$Ht$LT$TLT$Ht$LL$DHLL$LT$Ht$THt$LT$LL$TDHvHDLD$ L\$LL$LT$UD$ L\$LL$LT$ LL\$LL$LT$TLT$LL$L\$LL\$LL$LT$SLT$LL$L\$IwHmI}HHDXHu*H UAD>JD>>JfDAWHfAVfHnIHxAUHX ATIUSHxHD$`H)D$0fHnflHD$@HD$h)D$PHlHLHHH HLyHD$0LL-\H1HCH9HL;luIHD$8HI_HLL$0HT$@H INH>HFLyHD$@HFHD$8HHD$0MLL$0HD$8HT$@H IHIvHH=bAHEQHj55vj5~Pj5IHEHPMcHHE5HxL[]A\A]A^A_HHHVH YHT$@oHFL)T$0FLHLT$LyHLT$HHD$0HUIRDH ԘAHHHH5xATL H81WXXHZH H=mE1E HH gHiAHMEI@H iH fHQLML-MM1HI9L;luIHHD$@L{HFHHYHD$8HT$0{fDE1IE0JtHD$ I94fDHI9EH9F#A} P~ ~IUH;VHFIMH9@H@t HA} D^ D8@ vI}HA HHN0HHA@HEȃ5DA9u9HtQLL$HLT$RLT$LL$DIAL9IJtI9K fH qI9uuH9uuLHL$LL$LT$[PHHtMH;DLT$H;=LL$uHL$H9;H/NdHD$8SH_HHH H5ujL AHH816UY^Hf.IE0E1HD$(@N|M9"HˬI9EI9G A} A IUI;WIGIMH9@H@t HA} Aw 8@ IMH@HHE|$(@ IO0IH@IDHȃD>A9ufLL$Ht2L\$HLT$PLT$L\$LL$u2fK/H 1I9uu I9utIM9DLLLT$HL$ L\$LL$MLT$HItwH;LL$L;=L\$1HL$ I9#LL\$LL$LT$QL\$LL$LT$I/LDLT$PLT$HtDHf.HHEH 1IH=ڄ{HL$0HT$PMLL 7HHLL$LT$H|$PLL$LT$H|$IMH@HHE|$ yH3L\fDHfDLL$D$LT$KLL$D$LT$Df.LLL$Ht$LT$)JLT$Ht$LL$DHLL$LT$Ht$IHt$LT$LL$TDHvHDLD$ L\$LL$LT$@KD$ L\$LL$LT$ LL\$LL$LT$qILT$LL$L\$LL\$LL$LT$>ILT$LL$L\$IwHmI}HHD/NHu*H ŐAD>JD>>HfDAWHfAVfHnIHAUH ATIUSHxHD$`H#)D$0fHnflHD$@HD$h)D$PHiHLHHH HLyHD$0LL-iH1@HCH9HL;luIHD$8HI_HLL$0HT$@H NH>HFLyHD$@HFHD$8HHD$0MLL$0HD$8HT$@H iIHIvHH=UAHEQHj5t5j5Pj5OIHEHPMcHHE5HxL[]A\A]A^A_HHHVH ɥHT$@oHFL)T$0FLHLT$LyH%LT$HHD$0HUIUDH DAHHH0H5:nATL nH81MXR/ZH H=E1 HH ׍HٍAHMEI@H ٤H fHhGLML-M1HI9L;luIHHD$@L{HFHHYHD$8HT$0{fDE1IE0JtHD$ I94fDHI9EH9F#A} P~ ~IUH;VHFIMH9@H@t HA} D^ D8@ vI}HA HHN0HHA@HEȃ5DA9u9HtQLL$HLT$]HLT$LL$DIAL9IJtI9K fH I9uuH9uuLHL$LL$LT$EHHtMH;LT$H;=mLL$uHL$H9;H/NdHD$8HH_HHH H5&kjL ZAHH81JY^4/f.IE0E1HD$(@N|M9"H;I9EI9G A} A IUI;WIGIMH9@H@t HA} Aw 8@ IMH@HHE|$(@ IO0IH@IDHȃD>A9ufLL$Ht2L\$HLT$ELT$L\$LL$u2fK/H I9uu I9utIM9DLLLT$HL$ L\$LL$lCLT$HItwH;PLL$L;= L\$1HL$ I9#LL\$LL$LT$GL\$LL$LT$I/LDLT$ELT$HtD;/f.HHEH /H=rz%{HL$0HT$PMLL HHQ@/LL$LT$H|$4FLL$LT$H|$IMH@HHE|$ yHA\fD2/fDLL$D$LT$mALL$D$LT$Df.LLL$Ht$LT$?LT$Ht$LL$DHLL$LT$Ht$a?Ht$LT$LL$TDHvHDLD$ L\$LL$LT$@D$ L\$LL$LT$ LL\$LL$LT$>LT$LL$L\$LL\$LL$LT$>LT$LL$L\$IwHmI}HHDCHu*H 5AD>JD>>*/fDAWHfAVfHnAUIATUHHpHSHHhHD$PH)D$ fHnflHD$0HD$X)D$@HfL4HvHH HHMHD$ IL%M1 HL9L;duIHD$(HLaM%HL$ HT$0AHNHFHMHD$0HFHD$(HHD$ H'HL$ HD$(HT$0MHIuHE1H=IAI$j55vj5Pj5-QLHI$HPHgHI$IH1IHHm:HhL[]A\A]A^A_HHHVHT$0oHFH)T$ .fDHMHLHH $H $HHD$  L}H[H AHHHH5cSL H81@CX[ZH  H=uE1.fHH GHIAHMEI@HI @L<H<LEL=%M1 HI9L;|uIHHD$0IL$HFHLeHD$(HT$ pfDID$0E1HD$JtI92HI9D$H9F|tA|$ @~ nIT$H;VHFI|$H9AHAt HE\$ DV DD@@8A I|$HA@HE|$A 'L^0HHA@IEDAJADDE9uHHH H H5+ajL _AH H81@Y^a[fLM9uuL9usLLT$LL$H $:HHYH;ŗH $H;=LL$uLT$L9H/0 KH C$[H=XrfDHI$zH [E1H=rDIG0E1HD$@JtI9H;I9GH9FA @~ gIWH;VHFIH9AHAt HE_ DV DD@@8A >IHA@HE|$A L^0HHA@IEDA:A=DDE9uSL$Ht LL$H:LL$L$u+fKLM9uu L9utIM9DLLT$LL$L$8HHtmH;nL$H;=(LL$LT$L9LL$LD$H<$'E9uTHtH1u>@MLL$Ll$H\$Ld$OfDL9uu M9u*t&IL9LL$H\$Ld$~fDLH/IHtH;L;=M9L2I/tMLL$Ll$H\$Ld$GfDL$1L$HtLJ@HL$0HT$@ILL &yH9x-LL$0H-Y]QJ5fDLD$ -D$ L+8 L++IuHI|$HzfDLD$,4-D$,H+EZLh+8BIwHH{HDD>DD>DD$DD>ff.AWH_AVAUATIUSHHhL5։HD$@H HD$0HD$HHD$PLt$8HHLHHHZHH[rH KrAHOL ;rEHwIHLOHH H8wSH5*RH811XMZH rI H=dE1论HhL[]A\A]A^A_@HHZL^L\$8LLL$0 oL^Li)D$0MLL$0HI$It$HH=I*AHEHHH%AVjRPjRLPj5زIHEHPM&HHE2H*%LyL-1MfHL9L;luM LL$0MqMoM2L\$8&MLLiLL$0M~HML=H1HH9L;|uMML\$8IIE0LT$E1HD$Ld$MH$HLDLlM9HI9D$I9E A|$ A} IT$I;UIEIt$H9@H@t HE\$ EM DD@@8u}A I|$HA@HE|$A bIu0IHA@IDDAADDE9u'HHI+HL9H$fM9uuM9uuLL(IHtH;ׅL;-2M9)L,ImFtILT$HLd$H$yNHD$0+H ?nAL 1nHM$DILT$HLd$H$O 3HHEH m NH=z`蝶IG0L$E1HD$ Ll$MLd$IHH\$LDO|L9H H9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A UIw0IHA@IDDA]AmDD>E9uTHtH(u>@MLL$Ll$H\$Ld$OfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHP&IHtH;9L;=M9L *I/tMLL$Ll$H\$Ld$GfDL$(L$HtM@HL$0HT$@ILL pH艜x-LL$0H%Y]M5fDLD$ $D$ L8#8 L #+IuHI|$HzfDLD$,$D$,H"EZL"8BIwHH{HDD>DD>DD$DD>ff.AWHAVAUATIUSHHhL5&HD$@H HD$0HD$HHD$PLt$8HHLHHHZHHiH iAHOL iEHnIHLOHHYHnSH5zIH81)XNZH \i H=\E1HhL[]A\A]A^A_@HHZL^L\$8LLL$0 oL^Li)D$0MLL$0HI$It$HH=!AHEHHHuAVjRPjRLPj5(ڽIHEHPM&HHE2H("%LyL-ݩ1MfHL9L;luM LL$0MqMoM2L\$8&MLLiLL$0M~HML=H1HH9L;|uMML\$8IIE0LT$E1HD$Ld$MH$HLDLlM9H~I9D$I9E A|$ A} IT$I;UIEIt$H9@H@t HE\$ EM DD@@8u}A I|$HA@HE|$A bIu0IHA@IDDAADDE9u'HH"HL9H$fM9uuM9uuLL> IHtH;'}L;-|2M9)L#ImFtILT$HLd$H$yNHD$0"H eAL eHvN$DILT$HLd$H$O 3HHEH Me NH=WIG0L$E1HD$ Ll$MLd$IHH\$LDO|L9H[|H9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A UIw0IHA@IDDA]AmDD>E9uTHtH( u>@MLL$Ll$H\$Ld$OfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHIHtH;zL;=GzM9L[!I/tMLL$Ll$H\$Ld$GfDL$O L$Ht}N@HL$0HT$@ILL gHٓx-LL$0HhY]N5fDLD$ <D$ L8 Lp+IuHI|$HzfDLD$,D$,H EZL8BIwHH{HDD>DD>DD$DD>ff.AWHWAVAUATIUSHHhL5vxHD$@HHHD$0HD$HHD$PLt$8HHLHHH\HH`H `AHOL `EH.fIHLOHHwHeSH5@H81b XfGZH `2H=SE1PHhL[]A\A]A^A_fDHHXL^L\$8LLL$0 oL^Li)D$0MLL$0HI$It$HH=AHEHWHHůAVjRPjRLPj5Ь*IHEHPM&HHE0Hx#LyL-1MfHL9L;luM LL$0MqMoM2L\$8&MLLiLL$0M~HML=LH1HH9L;|uMML\$8IIE0LT$E1HD$Ld$MH$HLDLlM9H#vI9D$I9E A|$ A} IT$I;UIEIt$H9@H@t HE\$ EM DD@@8u}A I|$HA@HE|$A bIu0IHA@IDDAADDE9u'HHHL9H$fM9uuM9uuLLIHtH;wtL;-5t2M9)LIImFtILT$HLd$H$yNHD$0DH \AL \HJG"DILT$HLd$H$O 3HHEH \~GH=jO=IG0L$E1HD$ Ll$MLd$IHH\$LDO|L9HsH9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A UIw0IHA@IDDA]AmDD>E9uTHtHxu>@MLL$Ll$H\$Ld$OfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHIHtH;qL;=qM9LI/tMLL$Ll$H\$Ld$GfDL$L$HtQG@HL$0HT$@ILL F_H)x-LL$0HY]VG3fDLD$ D$ L8 L+IuHI|$HzfDLD$,$D$,HpEZLX8BIwHH{HDD>DD>DD$DD>ff.AWHAVAUATIUSHHhL5oHD$@Hh HD$0HD$HHD$PLt$8HHLHHHZHHKXH ;XAHOL +XEH|]IHLOHHnHF]SH58H81XsLZH W}H=JE1螠HhL[]A\A]A^A_@HHZL^L\$8LLL$0 oL^Li)D$0MLL$0HI$It$HH= AHEHHH5AVjRPjRLPj5zIHEHPM&HHE2H%LyL-1MfHL9L;luM LL$0MqMoM2L\$8&MLLiLL$0M~HML=H1HH9L;|uMML\$8IIE0LT$E1HD$Ld$MH$HLDLlM9HsmI9D$I9E A|$ A} IT$I;UIEIt$H9@H@t HE\$ EM DD@@8u}A I|$HA@HE|$A bIu0IHA@IDDAADDE9u'HH9HL9H$fM9uuM9uuLLIHtH;kL;-k2M9)LImFtILT$HLd$H$yNHD$0H /TAL !THWL$DILT$HLd$H$O 3HHEH SLH=F荜IG0L$E1HD$ Ll$MLd$IHH\$LDO|L9HjH9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A UIw0IHA@IDDA]AmDD>E9uTHtHu>@MLL$Ll$H\$Ld$OfDL9uu M9u*t&IL9LL$H\$Ld$~fDLH@ IHtH;)iL;=hM9LI/tMLL$Ll$H\$Ld$GfDL$L$Ht^L@HL$0HT$@ILL VHyx-LL$0H Y]cL5fDLD$ D$ L( 8 L +IuHI|$HzfDLD$,t D$,HEZL8BIwHH{HDD>DD>DD$DD>ff.AWHAVAUATIUSHHhL5gHD$@H HD$0HD$HHD$PLt$8HHLHHHZHHOH OAHOL {OEHTIHLOHHIfHTSH5j/H81XBOZH LO H=~BE1HhL[]A\A]A^A_@HHZL^L\$8LLL$0 oL^Li)D$0MLL$0HI$It$HH=IAHEHHHeAVjRPjRLPj5ʣIHEHPM&HHE2H%LyL-͏1MfHL9L;luM LL$0MqMoM2L\$8&MLLiLL$0M~HML=H1HH9L;|uMML\$8IIE0LT$E1HD$Ld$MH$HLDLlM9HdI9D$I9E A|$ A} IT$I;UIEIt$H9@H@t HE\$ EM DD@@8u}A I|$HA@HE|$A bIu0IHA@IDDAADDE9u'HHHL9H$fM9uuM9uuLL.IHtH;cL;-b2M9)L ImFtILT$HLd$H$yNHD$0H KAL qKH&O$DILT$HLd$H$O 3HHEH =K yOH=j>ݓIG0L$E1HD$ Ll$MLd$IHH\$LDO|L9HKbH9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A UIw0IHA@IDDA]AmDD>E9uTHtHu>@MLL$Ll$H\$Ld$OfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHIHtH;y`L;=7`M9LKI/tMLL$Ll$H\$Ld$GfDL$?L$Ht-O@HL$0HT$@ILL NHyx-LL$0HXY]2O5fDLD$ ,D$ Lx8 L`+IuHI|$HzfDLD$,D$,HEZL8BIwHH{HDD>DD>DD$DD>ff.AWHAVAUATIUSHHhL-f^HD$@HHD$0HD$HHD$PLl$8HHLH HHzHHFH FAHOL FEHLIHLOHH]HTGSH5&H81PX]ZH FH=9E1>HhL[]A\A]A^A_@H>HZHVHT$8HHD$0 oHVLq)D$0MVHD$0M$HIt$ E1H=FILKH jAPQjAPAQLj5PHIHPHGHIHIHHmHKfDLyL51MfHL9L;tuIHD$0HMwMBHT$8LLQHLqHD$0M~HML= H1HH9L;|uIHHT$8IrH D]H=7CfDHIH nD]E1H=7 fDIF0LT$E1HD$Ld$MH$HLDLtM9H[I9D$I9F A|$ A~ IT$I;VIFIt$H9@H@t HE\$ EN DD@@8u}A [I|$HA@HE|$A 2Iv0IHA@IDDAADDE9u'HHIHL9H$fM9uuM9uuLLIHtH;YL;5YM9LI.tILT$HLd$H$yOHD$0H @BAL 2BHt]fDILT$HLd$H$KIG0L$E1HD$ Lt$MLd$IHH\$LDO|L9HCYH9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A ]Iw0IHA@IDDAeAuDD>E9uTHtHu>@MLL$Lt$H\$Ld$KfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHIHtH;qWL;=/WM9LCI/tMLL$Lt$H\$Ld$GfDL$7L$Ht];@HL$0HT$@ILL [@Hpx%HD$0re]fDL0LD$ D$ LhhPLP[8IvHI|$HfDLD$,D$,H=RL0:IwHH{HDDnDD>DDTDD>ff.AWHoAVAUATIUSHHhL-VUHD$@HHD$0HD$HHD$PLl$8HHLH HHzHH=H =AHOL =EH CIHLOHHTHBSH5H81@X`ZH =H=1E1.HhL[]A\A]A^A_@H>HZHVHT$8HHD$0 oHVLq)D$0MVHD$0M$HIt$ E1H=VIL;~H jAPQjAPAQLj5PHIHPHGHIH IHHmH;fDLyL5}1MfHL9L;tuIHD$0HMwMBHT$8LLQHLqHD$0M~HML=H1HH9L;|uIHHT$8IrH ;`H=/3fDHIH ^;`E1H=.fDIF0LT$E1HD$Ld$MH$HLDLtM9HsRI9D$I9F A|$ A~ IT$I;VIFIt$H9@H@t HE\$ EN DD@@8u}A [I|$HA@HE|$A 2Iv0IHA@IDDAADDE9u'HH9HL9H$fM9uuM9uuLLIHtH;PL;5PM9LI.tILT$HLd$H$yOHD$0H 09AL "9Ht`fDILT$HLd$H$KIG0L$E1HD$ Lt$MLd$IHH\$LDO|L9H3PH9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A ]Iw0IHA@IDDAeAuDD>E9uTHtHu>@MLL$Lt$H\$Ld$KfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHxIHtH;aNL;=NM9L3I/tMLL$Lt$H\$Ld$GfDL$'L$Ht`;@HL$0HT$@ILL ;Hgx%HD$0re`fDL LD$ D$ LXhPL@[8IvHI|$HfDLD$,D$,H=RL0:IwHH{HDDnDD>DDTDD>ff.AWHvAVAUATIUSHHhL-FLHD$@H HD$0HD$HHD$PLl$8HHLH HHzHH4H 4AHOL 4EH9IHLOHHyKH9SH5H810X\ZH |4yH=6(E1}HhL[]A\A]A^A_@H>HZHVHT$8HHD$0 oHVLq)D$0MVHD$0M$HIt$ E1H=IL+uH jAPQjAPAQLj5IuPHIHPHGHIHIHHmH+fDLyL5t1MfHL9L;tuIHD$0HMwMBHT$8LLQHLqHD$0M~HML=H1HH9L;|uIHHT$8IrH 2A]H=8&#{fDHIH N23]E1H=&zfDIF0LT$E1HD$Ld$MH$HLDLtM9HcII9D$I9F A|$ A~ IT$I;VIFIt$H9@H@t HE\$ EN DD@@8u}A [I|$HA@HE|$A 2Iv0IHA@IDDAADDE9u'HH)HL9H$fM9uuM9uuLLIHtH;GL;5uGM9LI.tILT$HLd$H$yOHD$0H 0AL 0Ht\fDILT$HLd$H$KIG0L$E1HD$ Lt$MLd$IHH\$LDO|L9H#GH9CI9G { A HSI;WIGHsH9@H@t H{ E_ D@@8@ HsH@HHE|$ A ]Iw0IHA@IDDAeAuDD>E9uTHtHu>@MLL$Lt$H\$Ld$KfDL9uu M9u*t&IL9LL$H\$Ld$~fDLHhIHtH;QEL;=EM9L#I/tMLL$Lt$H\$Ld$GfDL$L$Ht\;@HL$0HT$@ILL 2H^x%HD$0re\fDLLD$ D$ LHhPL0[8IvHI|$HfDLD$,D$,H=RL0:IwHH{HDDnDD>DDTDD>ff.AWHxAVfHnAUATIHxhHUfHnSflHHxL=CHD$`H{HD$0HD$hHD$8L|$@)D$PHHLHHXHLLiLL$0MHD$8HT$@HI$It$HH=AHEHAWj5l5{j5wPj5(xIHEHPMHHEHxL[]A\A]A^A_fDHHFLLiHD$@HFLL$0HD$8MLL$04DHfHHHH*H *AHOL s*EH/LOODx@H CLiL5-w1MHL9L;tuM LL$0MIM|kH )AL A/HH@H/H5 SH81X$HZH )H=E1nrfHvLHL$:kL$HHD$8IMHML5vH1HH9L;tuIHHD$@IILLLL$0HVHT$@HFHD$8@HFHLiHD$8HT$0jfDLL$0HT$@BfIF0L$E1HD$ Ll$MLd$IHH\$LDOtL9H?H9CI9F { A~ HSI;VIFHsH9@H@t H{ E^ D@@8@ HsH@HHE|$ A Iv0IHA@IDDA9AIDD6E9uTHtHu>@MLL$Ll$H\$Ld$K&fDL9uu M9u*t&IL9LL$H\$Ld$~fDLH@IHtH;)>L;5=DM9;LI.QtMLL$Ll$H\$Ld$GfDL$L$H HHHEH u&[HH=Ro8LAIF0E1L$MHD$ILLd$MMtM9 H=I9D$I9FA|$ A~ IT$I;VIFIt$H9@H@t HE\$ EN DD@@8A I|$HA@HE|$A Iv0IHA@IDDAADDE9u3HLT$H@LT$DHL9fDM9uuM9uuLLLT$IHtdH;;LT$L;5{;M9LLT$LT$I.CkILd$LML$yOHD$0H $AL $HG?ILd$LML$O DHL$0HT$PILL (HTH3HHLD$,4D$,HLhIvHrH{HHLD$ LT$D$ LT$fL$7L$HoHLLT$LT$tLLT$LT$TIvHII|$HDD6DD4I(S[HYH PH9HsHPH7HLPM?I@LD$LH5bHHLD$HIHHIH:HLHHH+1H;-3H;-?3H;-U3HOÅHmHD$HE1LLHLH HEUHT$wq[IXMA4 7HuLH~PL}L%_M1DHI9L;duIHHD$HHHLd$0HmHDHtE0HH HD$LnL&Ll$8Ld$0H.LvLt$HHFHD$HD$@DHEL=MkHH H=u=1LHI;MHuH1H5H8f3[ fDHq1I$H e1HHD$HHHjHHjI,$KMIHD$HE1LLHLH HEUHT$}oY^IHHm HD$H;0ILI/I(ZMt I,$ImHĈH[]A\A]A^A_L&HILd$0HIHaLHHL$ZHD$0IH:HL$HfDL&HD$ILd$0HÅAP4E1AH AL HH/HH5SH81@Xr3ZH H= 1/af.H]LHHL$YHL$HIHD$8HHL}L%CeMy1fDHI9{L;duIHNHD$@HfLMIDHpsH`A3 E1HmH DLD$H= $`I/LD$/MI(LL;=A.uL;-0.uLL.IHH;.L;--u L;--Im HL$H\$MHD HL$0HT$PILL HG\3L 3H IH=X ^I/E1LLD$LD$f.LsHLD$T$T$LD$RL{HxHhL;5,t L;5,HRH HH9H HHHRHHHH HCLD$HH5sWHH LD$IHHMHH H+I9AKLHt$XLD$XLD$LL$HD$PJuLL$LD$HMHkI*HELD$HH5taHHLD$IHEHMvHEHH5PL׺LD$LT$nLT$LD$HHEI*N H;-?+H;-*u H;-+Hm;1IHx*I9FHt$XLLD$XLD$HD$P)tLD$MHú2A66HOI* I/PID$0HL$E1HD$ H\$LMMNlM9"H*I9GI9EA bA} oIWI;UIEIwH9@H@t HEW EM DD@@8uzA HIHA@HE|$ A  Iu0IHA@IDDAT AV DDE9u%Ht7HDIL9H\$HL$H\$MKHHH AHOL EH,LOOD@HHx.HhHD$@Ld$0Lt$HHD$RID$0HL$E1HD$ H\$LMMNlM9H(I9GI9E A bA} oIWI;UIEIwH9@H@t HEW EM DD@@8~A *IHA@HE|$ A  Iu0IHA@IDDA A DDE9u)HHDIL9HL$H\$@L;=1'uuL;-$'uuLL&IHtH;'L;-&zL;-&mLImjHL$H\$My6HL$HL$HP3HL$H\$MKjf.HLD$LD$_fLLD$LD$MIfD3H  H=1W\@HLL$LL$HLD$LD$fHD$HE1LLHLH HSHT$&dIXZM H+HLD$LD$LLD$[LD$LúE1A{4H+HLD$T$T$LD$sfLHLD$LD$A61*@HLD$cLD$f1HHHf. A4E1BDW31A6I*LLD$T$T$LD$fAI4H=qVH@H5@ֹIDfDLLD$LD$L1H3IH2fLD$H*fDLAK4E1fH=UĶI@HAN4HHD$LD$ffHt$X1H)D$PHulIDHLL$LD$LL$LD$LxLLD$cLD$fHLT$LD$>LT$LD$ 3fH=ZH?H5?&HfDE1A3 DH=iZdH@LD$,D$,$HLLD$LD$fLpLXSIuHIHH S  4H=XRfDCHLD$,D$,QKA4 HHx H5T$H8MT$_@LxL`IuHIHA 4VHEH =H9HH<HHH<HUH5NOHHHH+HL0IHHmL; L;L;LLD$LD$dI([HD$HE1LLHLH HSHT$]AZA[IH"A4HL$HL$HI3DDt4H=sQH;H5;شH,A]4FLLD$LD$9H=*Q HAv40H1A5H=PH:LD$H5:ILD$HLAy4H=LD$IH=PLD$_LD$H1A5-H AL H B3-DD;IYHMQHII)TfInfHnL׺flHt$PLD$LT$)D$PfH+LT$LD$HHLT$LD$LT$LD$eH׿DD LD$IyA61Y1A6'DDPHGBH @9H9HH'9HHH-9HH53THHHHmHL|HHH+H;-WH;- H;-+H%ÅuHmHD$HE1LLHLH HEUHT$Z_AYIHA 5$ fDA4H=MH8H58VIA4InHsMVHEII.fInfHnL׺flHt$PLD$LT$)D$P{dHmLT$LD$HEHLT$LD$葽LT$LD$$LA4E1"H=MILD$HH>iA4HHD$LD$iA4E1LLLT$LD$LD$LT$H?H {6H9H]Hb6H9HHN6HH5vQH>HHH+HLȼHHHmH;@H;_@FH;t9HnŅH++HD$E1LLHLH HRSHT$WY^IH&A45D!4H=BKH[5H5\5觮HqH= KH]LLT$LD$ILD$LT$PH2A4!!A4H Lú!E1A4H-@H̺_H=H i4H9HHP4H`HH-<4H9H5lOH<HH HmHLźHHH+QH;-H;-^H;-t HnÅHm%*HD$E1LLHLH HEASUHT$U[IXM#A_5(XE1A4#EH=@I#H0H=,IH53H563葬HE1#A4A4#E1H+#A4HE1%A'5HHٸoH̸H;H Y2H9HH@2HHH,2HH5MH<HHH+pHLƸHHAHm-H;@H;]@H;rHlŅH+sHD$LLHLH HAPE1SHT$SAYAZIH*A5{E1A%5%C@녺%A"5R% 5bH=GHH=GH0H50jHH9H 0H9HsH0HOHH-0H(H5?HHHHmHL0HHH+H;- H;-{H;-nHٺÅIHm1t[HD$E1LLHLH HEQUHT$ R^_IHA5,H H=v=H;=OtuIGt H;u_LwHHH=QHt$XH\$XHD$P:HHtRH+tBH0Hmt.5zLظHHk.5VHT봺.A5.5-H+A5+E1HT+A5HE1A5+E1+A5|H=lDOHH=XDH!.H5".轧HH}E1)A}5E1AM5'E1'AK5H=CӤHH=CH-H5-AHuE1A{5)H)Ax5)v5H={C^H"H=gCH@-H5A-̦H'AP5_AR5'E1'AWH/IAVAUIATUSHXH<$HD$(HD$0HD$8HHL4HH HLAH\$(MHCuDHA6H9t8HXH<HqHW1HH9CH;TuH}5H 1H9HH}1HLHL-i1MIEH5:LHHzHIEHHtIEHWMIHHHXԶIHH4H 0H9H%H0HHL=0MIGL\$LH5!DHHZL\$IMYI/,HU4H N0H9HH50HHL=!0MIGLD$LL\$H50?HHL\$LD$IMWI/ H HD$I9@LHt$8L\$LD$HD$0H\$8Ll$@WLD$L\$IMIm M,I/H5:LLL\$L\$vI.=HEHHH=L\$`L\$LLHIML\$LHm'I,$I+H<$H5CHGHHIMHD$I9BIjH}MZHEII*fInfHnLߺflHt$0L$)D$0)VHmL$I^ qM~I+I,$IEHPIUIEHueL[HH% HH5GAUL AH CH81ǴX qZH \H=EE1=HXL[]A\A]A^A_HH9HuH; fDHA1H *-H9H H-H(HL-,M IEH56LHHIMImH HD$I9FHt$8LMHD$0H\$8TIMI,$IEH5<LHHIMH5C1LLT$LT$HIgI*L= M9L;%t yL;% lL脱Aƅ]qE1E11I,$!L$$MIE1fDE1Mt I+v Mt I. Mt I/ Mt I( H H=L$;ML$tIm E1MH>HH\$(QIIĺ7rMI+M9L;%~L;%L莮ÅtI,$ZJH,H(H9XH(H&HLv(MH5F2LLT$LT$HII*\HD$I9FHt$8LLHD$0Ll$84PHHHmImgH<$H5w<HGHHIMcHD$I9B.IjH!MrHEII*tfHnfHnHt$0Lflź)D$0OHmIMI.I,$HIHPHlL耨#I,$uLiH2+H&H9XH&HHL&M~IBLT$LH5z0HHLT$IIHMIHDIEH5:LHHRIM2111LjHH5I,$ 詪IH Hh4HH.H1*H%H9XH%HHH%HHCH53HHHIMH+# H5_1LHT I/n HLLIHI. I,$} Hm[ H59H<$LT$CLT$HH8HD$H9E;LeM.L]I$IHmSfInfInLߺflHt$0LT$L$)D$0LI,$L$LT$I`MԺrMI+ I.LLL$0_L$HHImIfI*L莥fIG0E1HD$@JtI9HI9GH9F A  ~ )IWH;VHFIH9AHAt HE_ DV DD@@8A IHA@HE|$A oL^0HHA@IEDA A. DDE9u2HLL$1HݦLL$DIM9fDHI9uuH9uuLLL$uHHt[H;^LL$H;=H9LL$H|$$LL$H|$H/K^y"HD$($H{pKoLhLLD$L\$NLD$L\$LLD$L\$&LD$L\$LL\$L\$SHL$L$@ H=2L\$gL\$IL貢vL襢L蘢dL耢 E1A|$A'E1rWLLD$LT$$;LD$LT$$cfLLD$LT$$ LD$LT$$BfLLD$LT$$ۡLD$LT$$!fLLT$$谡$LT$ 1LL$I菡L$fDLL\$sL\$fHL$(HT$0MLL H H\$(*L L\$fDLLL\$L\$fHL\$ӠL\$fE1ҺUqfDH=Y0HH5输IfDH=10H*H5+薓IfDE1ҺqfDH=/ԐI@H=/輐I@kI1qME1IE1E1;H~ME1E1E1WqDL訟tL蘟YE1E1E1E1E1ۺZqHmHLD$L\$$P$L\$LD$ffDInHLvLiQL\DDLBHLT$0LT$LLT$LT$lLHD$LT$JH,LH֘tH=(HKH5LIyr rH=B(H+H5,觋ItH= (I`H= (I袞LT$HpLL$)L$GE11MԻr dLT$I"rLHt$8L$HD$0H\$8>L$IMLL$詗L$L׻r蓗E1ҺMrϝILuML]IIHm6fInfInLߺflHt$0L\$H\$@)D$0=I.L\$ILL\$L\$sHLT$L$ږL$LT$HmD{H赖l;rE1E11LLT$L$舖LT$L$qL躞IrgLL\$LT$ILT$L\$E1r(Ht$8HL$LT$8HD$0uH{aLD$(TqD$(AbHD$ 1E11HD$E1E1HD$HD$2LLL$HD$pLD$LL$LLD$pLD$LoLoHnULn=H5H= 1HHHAbHmHAbpfIt$HfDI~HLoHIt$HfDH{HsuIHD$ 1E11E1AbHD$HD$HD$frHStaAbLl$0I1E1HD$ E1E11HD$E1AazHD$HD$HD$(DDDDLHD$nLL$HLT$nLT$7LHD$vnLD$HD$ 1E11E1AbHD$HD$HD$itI`LLL$HD$nLT$LL$~LmHH HRH5A"cH81sVHD$ 1E11E1AbHD$HD$HD$LLT$HD$wmLD$LT$ LLD$[mLD$HrHD$IHu;I1E1E1E11A$cHD$ HD$HD$,HmHLlHD$ 1E11E1AbHD$HD$HD$sLT$I;DDDDqHLL$olLL$L]lloH ~AHbcaLLL$#lLL$&HD$ 1E11E1AbHD$HD$HD$HLT$kLT${LHD$kLT$HD$ 1E11E1AbHD$HD$HD$HD$ 1E1E1HD$ApbHD$HD$HD$(eLHLD$mLD$WHD$ 1A6cHD$E1E1E1HD$HD$LLT$LL$jLT$LL$Ac_LLT$jLT$HLD$xjLD$H5H=1HHHAc*HmHAcjLjZLl$H$HDŽ$H$LMILLD$iLD$HD$ 1E11E1AbHD$HD$HD$oITLLD$biLD$xHD$ 1E1E1HD$A1cHD$HD$WHD$ 1E11HD$E1E1AbHD$HD$HD$ 1E1E1E1A3cHD$HD$HD$HLD$ LL$hLD$ LL$LLT$nhLT$HHLL$WhLL$LLL$@hLL$HoL׺H$LT$H$HD$HDŽ$H$LT$HD$MHD$ 1E11E1AbHD$HD$HD$LLD$gLD$1E1E1A@cHD$ HD$HD$HD$ 1E11E1AbHD$HD$HD$LT$H5:SLT$Ix1E1AcHD$ HD$HD$LLL$0HD$8_LT$8LL$0H=8LT$PLT$I 1E1AcHD$ HD$xHD$ 1E1E1HD$A"cHD$HD$A1E1ME1HD$ AcHD$HD$LLT$8LL$0HD$@^HL$@LT$8LL$0.Ml$MMMT$IEII,$fInfInL׺flH$LL$LT$)$ ImLT$LL$I LLD$ LT$LL$^LD$ LT$LL$LLD$@LT$8LL$0]LD$@LT$8LL$0LL$01E1E1HD$ AcHD$HD$LLT$8LL$0HD$@]LD$@LT$8LL$0LLT$8LL$0HD$@a]L\$@LT$8LL$0LL$0Ad1E1ۻE1kHxH4HHH <LD$8HEH H5ܗLL$H81bLL$LD$81E1HD$ AcE1E1LL$0HD$HD$H;7LLL$LD$aLD$LL$HH& I(HALL$HHL$LAHL$LL$HI LL$ HHD$0AHL$LL$ HHD$LT$0l HAHL$LL$ HHD$PLT$0 LT$HHLL$0HL$ AHL$ LL$0HHD$LT$HL\$P HL$PL\$HLT$0LL$ LL$ LT$0L\$HHL$PZ H)L\$ LT$HLL$0S[LL$0LL\$8LL$07[L\$8LL$0!LL\$8LL$0[L\$8LL$0LLL$0ZLL$0NHL\$8LL$0ZL\$8LL$0H=HLL$0H5MLL$0IkHD$ 1E1E1HD$A cHD$HD$LL$01E1AdH=LL$0JLL$0ILL$0AdPLL$01Ad9H=LT$8LL$0JLL$0LT$8IH=HLT$8H5LL$0LLT$8LL$0ILL$0A%eE1LL$01E1AdHLL$0fYLL$0#IPHI@HHL$HHH@HL$HD$ gLL\$PHD$HYL\$PLT$HLL$0LLL$0HD$HXL\$HLL$0pLL$0A d1E1ۻMLL$01A eLL$01MAdE1ۻLL$01AdLL$01ME1AdLLT$LL$MXLL$LT$L\$1E1ۻLL$0AeZHYHLiHIEH)fHnúLLT$8D$H$LL$0)$H+LL$0LT$8IHLD$@LT$8LL$0WLD$@LT$8LL$0tIZHMjHIEI*fHnúLLL$8D$H$LD$0L$)$H+LD$0LL$8IHLD$@L\$8LL$0VLD$@L\$8LL$0LL$0Ac1E1HD$ E1E1HD$HD$HLD$8H5NLL$H81w\HLD$PL\$HLL$0kVLD$PL\$HLL$0DLHD$XJVHL$XL\$PLT$HLL$0LLD$PL\$HLL$0VLD$PL\$HLL$0LL$01ME1E1A1LL$0HItuHI/LL$0tXLL$0A#d1E1E1LL$0Ad1E1ۻLLD$8LL$0nQLL$0LD$8LWQLL$01E1AdvLL$01AgdOHl$LL$0E1LL$E1HD$ AcHD$HHHD$ HHH٭HL$HH5pL\$@AcE1H81LT$8LL$VLL$LT$8E1HD$ L\$@LL$0HL$HHD$LL$HD$pALL$0E1E1E1HD$ AcHD$HD$7LL$0Ad1E1ۻ-LL$0Avd1E1ۻLL$0E1AdH=YLL$0O@LL$0HH=;HLL$0H5BLL$0HjH5H=1LL$0蜨LL$0HHLL$8HD$0,LD$0LL$8I(LL$0JAd1E1ILLL$0HD$8NLT$8LL$0^LL$01E1AdLL$0Ac1E1LL$E1E1HD$ HD$HD$HLT$ LL$NLT$ LL$LHL$PL\$HLL$0[NLL$0L\$HHL$PHHL$@L\$8LT$0LL$ +NHL$@L\$8LT$0LL$ BHLL$0HD$8NLD$8LL$0'LL$0E1AdLAdM1E1LL$01E1AdfAWHfAVfHnAUIATUSHpHHH$H`)D$pfHnflH|$HDŽ$H$)$HHL4HH HIHLyHD$pLL%@H1HH9SL;duM$Ld$xMIMLl$pL$LfDHHFLyH$HFHD$xHHD$pMLl$pLd$xL$H HD$@HD$HHD$PH(HD$XHD$`HD$hhE111HALHHH8HD}H(hE111HALHHH8Eu sH5H H5# HHHD$I9tHH=H9xoHHHL=MmIGH5`LHHILl$@MiI/HyH=BH9xH)HHL=MLIGH5aLHH|ILl$HMI/yHH|$@HD$ H9GHD$HH$HDŽ$L$H$IH|$HH/H|$@E1HD$HA AEXMH/HL|$@I?'HD$@E1HAwI HHLLHD$(1AIH3HD$HH;D$Mt I,$IEH5?LHD$HHHHH|$HHLwHD$@HH|$HH/4HD$HH|$@H/HD$L%^HD$@LLMpLD$LLHD$HH@LD$HHSH|$LLHD$HM#HD$L%LLMpLD$LLHH)H@LD$HHLLHD$HHH"H@HD$PH;D$ 8HGHD$PHrHWHHH|$HHT$HH/HD$PH|$HH$H$HDŽ$!H|$PHD$@HtH/HD$@HD$PAXHhH|$HH/QHD$HH|$@H//HD$@E1CLHD$HD$Hx`H H|$(Hl$Hl$(IL|$(H\$ LII0LLL0I@H0I8H0H6DWIAMIE DI01)fDH(H0HH@(HA;}}LHH@HHtˀ8H(HI8A!HI(H0HA;}|HH9Hl$H\$ L|$(H|$DLt$H5_1LݞHD$hIHD$HIH|$hHH/dHD$hI?MImHm? H+H&EHHLvL$oLfL.)T$pHLHLyHD$pHH]IH AHHHޏH5jAUL H81gJXaWZH 0 H=E1UHĸL[]A\A]A^A_HH gHiAHMEIvfL5a@u{HH0H;0HHH0HH0HcI E1LMCMtImuLCHt HmHM+y2EHF(HJ(H)0HLcJ4LN(L;(}INL(HJ(H0fDH@0HH@(HH(H+0H0HBHB;LGf.0D$/LLIHH=D$'fH=3 I*HHD$I9/HH=H9xH~HMHL=jL|$HMIGH5iLHH>IL|$@LT$HMI*H}H=HD$HH9xHH|HL=нL|$HMIGH5WLHHIMXH|$HH/HH|$@HD$HH$HD$ H1H9O9fHnfInL$fl)$H|$HHD$PHt H/ HD$HI/L|$PH|$@M/H/` HD$PHD$@HH|$PH/M L|$@HHD$PHD$@AwI L5HD$IGLHD$8HD$LIRLT$0HHT$(zDIHH@HT$(LT$0HH LLIHHD$H5`HHD$PHEH|$ H9xLpMH@IHH|$PHD$PH/mH|$PH$L$HDŽ$HD$@I.`HD$@AZHH|$PH/HD$PH|$@H/HD$@E14DHD$ HD$Hx`H H|$~`Hl$(Hl$8Ll$IH\$0H\$L|$MID$LLL\OJDIL9uLl$Hl$(H\$0L|$H|$ p=H51L菗HD$hImLl$hMRImYHD$hI?L >Ll$H5I.IH)IH5tHD$HHH>HD$PHH9GHGHD$PHHWHHH|$HHT$HH/HD$PH|$HH$H$HDŽ$H|$PHD$@HtH/HD$@HD$PH_H|$HH/HD$HH|$@H/HD$@;HL$XHT$`HHt$hHD$ -HD$D$LHP`Hx rMH ?HD$@IH/HD$ HL$XHD$@HT$`Ht$hH@1LLl$XH5芕HD$`I/4L|$`MI/HD$`L|$XHD$XHmH;eLEL%MM1HI9L;duIH[H$I9fDLfHLyLd$xHD$pfDID$0E1Lt$IHD$(LLl$ILDMdL9HH9EI9D$\T} A|$ HUI;T$ID$HuH9@H@t HDU EL$ DD@@8uvA ZH}HA It$0IHA@IDDA A DDE9u*H(H<fDHL9HD$x=H`HH)AH5L`jL H H8H/1?_CWAX^fDL L9uuM9usLHLL$ :IH_H;L;%uLL$ M9 I,$LLl$Lt$Iم @LLl$Lt$IO$WH=H т.M後 A AXH|$@HD$HHtH/LT$HtuMtI*tZH|$PHtH/t:H ~DDH=sMI/E1fDk8fLX8fDK8LT$H|ID$0L|$1MHD$(Ll$ILŐMdM9H˕I9GI9D$ A WA|$ IGI;D$IT$IwH9@H@t HEW EL$ DD@@8u}A IHA@HE|$(A nIt$0IHA@IDDA A? DDE9u'HH9HH9LLl$fDL 1M9uuM9uuLLLL$ "7IHtH; L;%ɓuLL$ M9gL:I,$geLL|$Ll$y)9HJW@LL|$Ll$IjfH|$@1E1E1A AWH'fL53H|$@E1E1A AW5L5zH HH1IHHD$@H;D$ HFH=HD$@H9x HݱH HL=ɱMb IGH5LHH HD$HH I/H5L豑IH HH=WH9xX H>H4 HL=*M H5LZHD$PH I/nH[H|$HHD$ H9G! HD$PH$HDŽ$L$H$HD$@I,$H|$PH/L|$@LT$HHD$PA AXM(I*HD$@HD$HHH|$@H/TL|$HMHD$@HD$HHL$pMLHH$L ~ժ!OWLk3L|$@\3L7H}HA@HE|$(23(33IE;326H |+Z E1H=~A AaXLT$HHD$@AW:f.22H|$@E1E1A A XpLD$ d2D$ LS2$H0,L0#H|$@E1E1A AXHD$HL19It$H^HH|$H111&L11qH=QH:H5;$IH|$@E1A A(XeH=!I]A A*XDI/tH|$@E1-DLE11H|$@n7I>LD$ 0D$ DD}HLAXH80HD$HH|$HHD$HHH|$@A H|$0L.3HH|$@E1E1A AeYEDL.[8H|$@E1A A-XH=HH5%#IpIt$HIHu56I|H=~a I0H=IH|$@E1A AsXL)LA E1)H|$@H|$@A AZI,$tA AwXqLAwXA )WMAuX&H=:IH=&HߥH5ILK)#LwMHGIHH|$HHD$HH/HD$PfInH|$HfInflH$H$)$HD$@I.L((+H|$@E1A AYH$16H$11%AYI/H|$@E1E1A ZH|$HHt H/HD$HH|$PHt H/H ln YHD$PH=PcH|$ HL$PHT$HHt$@zHL$PHT$H1Ht$@.Hv1HLHD$(I/L\$(I6I+MWL;L;L;T$LLT$+LT$AI*EHt$@EcHt H.HD$@H|$HHt H/HD$HH|$PHt H/HD$ HL$XHD$PHT$`Ht$hH蘢~AY.LE1E1A &H|$@9WH|$@E1A A!ZgHHH9HuH;ЃDH H5lMAXA H8&H|$@ &H|$@E1E1A AcXAYHD$ HL$XE1A HT$`Ht$hH蔡H|$@%Z%mI*DPL%:LLT$(n%LT$(LHD$0W%LT$0L\$(AY`H5%AYH %%0H|$@E1A AX0H|$@A AXH|$@A AZLt$PLd$HHt$'Ht$LHLAY蚃HD$@HD$HHD$PAYff.@AWHAVAUATIUSHHH$H H$H-H$H-H$HmH|$ HDŽ$HDŽ$H$H$H$HL4H1'H>qHJcHfDHFH$HFH$HFLL}H$L$Iz IMt I+M- H$H$HT$H$HT$H$HDŽ$HDŽ$HDŽ$HDŽ$IH$HHD$HHHLH9Pj&H3H+HL%L$Mn&ID$L\$LH5HHb+L\$IL$M_+H$H/% H$1HDŽ$H ~H$H9O+fInfHnL\$fl)$H$L\$H$IHtH/L$HDŽ$M+H$H/ HDŽ$H$I+ HEH5jHHDŽ$HH.IL$M=.H5NL9ID$L-~L9X4E1I|$AI,$HDŽ$ErHEH5دHHH6IL$M5H5жL9/ID$L90:IT$HH?HH1H)Hu H=/I,$uLV HDŽ$H5+H=1uyH$IH2H H$H/YD$89E1E11HD$HE1E1L$HD$(E1E1AHDŽ$HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0#fIuOM~SLEL-MS1DHI9cL;luIH)H$IM"L$DH+DHBHX"H\$HD$HFLH$L$f.HHFHD$H$HFHD$H$fI,$U3HDŽ$THD$H fDHLxI9t MKHPHuL0LhMtIMtIMtIEHؠHH9Pk6HhH6HL TL$M3IAH5LHH6IL$M6H$H/u*HEH5:HHDŽ$HH6IM6H&zI9A1MQM1MaII$I)@0H$LL$LT$(HDŽ$LT$(H$I*1H$H8I,$g*H$HyH9G?8H$H$HDŽ$H$0H$H$H/)L$H$HDŽ$MX8H/g/L$HDŽ$HDŽ$Mt I.0Mt I/0Mt Im0H5L5H$IHJHxH; yI9HT$hu I9)DI//HDŽ$EH.HǖH9POHHPHL=L$MOH5BLwH$IHPH$H/,H$1HDŽ$H wH$H9OSH$H$H$[H$H$IHtH/IL$HDŽ$L$MSI),H5kH$HDŽ$H9HGH;wWE1HAH/IHDŽ$EzH5ðH=<1sH$IH~H詔H$H/3eD$8Y9E1E11HD$HE1E1L$HDŽ$AE1HD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0DL\$L\$@LLyL$@LyHLH躠H$IHHIgDLHD$HH\$L$pH ^AL dHHuHjdH5>ATH81?X 8ZH ^YH=SE1-HL[]A\A]A^A_HqLHH~3H$IM!LEL-%Mw$1HI9k"L;luIHJ$H$I$fDH9u I9u LHIH H;tL;-qtu I9)Im#, MLL|$Ld$Lt$ Hu)L bMLHH$H$7qfL6L\$L\$H.Q_HDŽ$fDHD$H9 HIHFHHH9PHHvHtIHL5bL$MHH5LrH$IHIIH$H/'HHHDŽ$H9PJHHKHL5͐MJH5}L=rIHKI.y'H0HH9PLHpH'HL5\MLH5<LLT$qLT$HIPI.1HqI9B*QL׺H$LT$HDŽ$L$}LT$H$MI/U1L$MRQI.DH5)H$4qIHRH$H/'DH$1HDŽ$H qH$H9O9SfHnfInfl)$ǺH$H$Ht H/FHDŽ$I/CL$M&TH$H/CHD$H5ۖHDŽ$L$HDŽ$Hxz]H;HH9PdHkHfHHWHD$0H$HcH5ҟHoIH#hH$H/?PH5eH|$LT$HDŽ$oLT$HHD$0H$kH=oA1HDŽ$I9z/mfHnHLLT$(D$0H$H$H4LHD$)$H$LT$(H$Ht H/\H$HDŽ$H/QH$HDŽ$HD$0H nI*~SHoH$H;=7oH9HT$hQFH9HF H$AƅpH/ZHDŽ$EdHBH{H9PHbHHLNMH5LLT$(mLT$(HHD$0H$^I*oH͓H=H9xUH݋HHHɋHD$0H$H:H5HlmIHÊH$H/pH5H|$LD$(HDŽ$,mLD$(HHD$0H$ԍH*mI9@HHD$0LǺH$LD$(HDŽ$H$ͶLD$(IMH$H/oHDŽ$MLI. yH5rLLT$(}lLT$(HHD$0˖I*xH$HqlH9G-HD$0H$HDŽ$H$H$HT$0HHD$HHxH$HD$0HՈH$H/xHDŽ$LIH$L$UIHچH;:lAH;D$hD7mI9.mLLT$LT$A/I*EwH$HL$H$H/kH$HDŽ$H/CL$I>I/HDŽ$MHV L(hE1H|$1HA1AH$IHZHH$H/GCHT$HDŽ$HHD$HH CI9\IAH5LLL$HD$.jLL$HIh[H5H9NH@H;jbIPHH?HH1H)Hu HNI(uLLL$ LL$H5H=S1LL$'fLL$HHD$0'HHLL$贇HLL$HD$HHLL$ND$8;E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0fL\$ L$L\$R@L L0LhMIE0E1L|$HD$(MLt$ILLd$MĐOlL92H[iH9EI9E}  A}  HUI;UIEHuH9@H@t HD] EU DD@@8u{A #H}HA@HE|$(A #Iu0IHA@IDDA'A:DDE9u&HtHH, fDIM9LLd$Lt${fMLL|$Ld$Lt$K"fHygE1HD$hH$H9HT$hE1HDŽ$HH;ygAHH$HĄHD$`HH9PUKHHSHL-L$MJH5ӚLcfH$IHRH$H/E:HDŽ$ H$IHSHT$`HH$HP IHATHH߃H9PGVHƃHVHLMUH5RLLD$(LT$eLT$LD$(HH$I!VI*=@H$H5LLD$LD$=TH$H/KBH$H$LLD$HDŽ$aLD$HH$I\H$H/LDH$HDŽ$H/DHDŽ$I(CH$H$H$H/CHD$hHDŽ$HD$0HL$H; eH;L$hH9H  Cb$HD$H9A8H5PHdIH{VHDŽ$HdI9EVIEH$HVMEHIImAH$H$LLD$H$HDŽ$荭H$LD$H$IHtH/ EL$HDŽ$MgVI(*@L$1ɺHHDŽ$LL\$|L\$HHD$H$~VLHL\$EL\$HI9ZH$H/fCHDŽ$I+CCH5H|$ bHD$H$Hj_HDŽ$HHbH9Cz\HCH$HfHSHHH$H$H/RNH$H$H$H\$HfHnD$`)$H$HD$Ht H/OH|$HDŽ$_H$H/MH5LHDŽ$aIHgHD$ LD$ HHD$H$fH\$LD$ HH$HX8 LD$ HHD$H$ cHH5HLD$ LD$ UH$H$LLD$ X]LD$ HHD$H$tI(`UH$H/DUH$HDŽ$H/UH$Hd~HDŽ$HDŽ$HD$H`H9XvH'~HovHH~HD$@H$HwH5֋H`HD$@H$H?wH$H/VH$H_1HDŽ$H9_pfHnH D$)$莩H$HD$@H$HtH/bbH$HD$@H|$@HDŽ$DqH$H/[H5H$HDŽ$_HD$@H$H$mH$H/ZHH|HDŽ$H9XH|HFHH|HD$@H$HڂH5ǏH^HD$@H$HqH$H/`HDŽ$HD$@HH$H)H$HDŽ$HCHD$@H$HHnH5HqH$H$H$BZHD$@HH$H/nH$HDŽ$H/nH$HDŽ$H/nH\$HDŽ$HHD$ HHOnHD$HHD$@HD$(HD$HD$ HD$PHD$XHD$@E3H\$H5cH{2gH5ӌH\H$HHąH\H9CljLCM_jHCIHH$H$H/eHԕH$H$L$LD$8H$=LD$8H$I(yeH$HׅH$H/ehHT$H$HDŽ$HHD$8HH*hHDŽ$H\$H5H[H$HH=uH51H{ÅtH$H/QHDŽ$QE}PH5H|$F[H$HHL{H51HQ{ÅsH$H/C{HDŽ$PHHcxH9X6zHJxHzHH6xH$HyH5fHZH$HHyH$H/8tHDŽ$gH$HHtH5HDZIHH5H$HHD$8LT$8sI*H$H5nH$iVHHbH$H/vH$HDŽ$H/NHt$HHDŽ$DIH6H5HHHD$8LT$8I*gyHIfDMHBH BAHOL BEHGLOOD@H=HwL\$H5wYL\$I@HD$HLE11HD$(E1E1L$HD$ E1E1AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8d8@Mt I)H$Ht H/H$Ht H/Mt I.Mt I*Mt I(.t$8DE1H _AH=6 Mt I,$9Mt I/:HL$0HtHHD$8HH+HL$`HtHHD$0HHMt ImHL$HtHHD$0HH Ht$HtHHD$HHH$Ht H/HT$@HtHHD$HHHL$XHtHHD$HHHt$PHtHHD$HHHT$ HtHHD$HHHL$(HtHHD$HHHt$HHtHHD$HHHtH+tbHmtKH$HtH/tXH\$H-HHD$HHH= H(fDHfD fH HH'H6LLHHLxHhHXKfDH8 H(LD\$pLD$hD\$pLD$hLD\$hD\$hLD\$|LT$pLD$hD\$|LT$pLD$h4DD\$|LT$pLD$hD\$|LT$pLD$hD\$|LT$pLD$hlD\$|LT$pLD$hLD\$|LT$pLD$h9D\$|LT$pLD$hDH=цL\$L\$Ikf[L\$IfDLD$8f8E11HD$HE1E1L$HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0H$L$HT$H$HT$HGH$HHWHHH$H$H/H$H$H$LD$8{8E11HD$HE1E1E1HD$(E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0IE0E1L|$IHD$(LLt$MLd$MMlM9HRI9D$I9EA|$ AA} KIT$I;UIEIt$H9@H@t HE\$ EU DD@@8u}A I|$HA@HE|$(A Iu0IHA@IDDAZ$A(DDE9u'HHHI9LLt$Ld$L|$@I9uuI9uuLL&IHtH;QL;-PI9LIm nILt$LLd$L|$y2H7DILt$LLd$L|$Krf. fDH\$afDHD$HE1E11HD$(E1E1E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$88fI%[fDLHHOAHDŽ$HHD$0HtHHtHD$`H0HH$HH$H/&L(L$AŅ1D$8&9E1E11HD$HE1E1L$HD$(AE1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0A|$@I,$̀HDŽ$HEH5HHH2 IL$M L111fH$IH H$H/L$L;%HDŽ$HDŽ$H5sHkH9P'HkH)HL=kL$M'H5ILLH$IH-)H$H/r H$1HDŽ$H LH$H9O)H$H$H$bH$H$IHtH/hL$HDŽ$M)H$H/4 H5rL$HDŽ$L9J"IAL9.E1IyAI)HDŽ$EH5ޅH=G1 HH$IHGFHiH$H//D$89E1E11HD$HE1E1AHDŽ$E1HD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0Lx5fwLHD$LT$pL\$L\$H$H$H$LOH=J}LT$(LT$IG7)HD$HE1E11HD$(E1L$E1HD$ E1AHD$PHD$XHD$@HD$HD$HD$`HD$0D$89[EfDLLT$(LT$(fDH;IJLIHr6H;IAH;IDI9LI,$AL$ESD$88E1E11HD$HE1E1AHD$(E1E1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0+LL!fL1LL$(H$)$-LL$(H$MLHKL8ZL(:LHDŽ$LD$0D$0HChKL.^6HD$HE1E11HD$(E1L$E1HD$ E1AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8u9IvD$88H$Ht H/H$HDŽ$Ht H/H$HDŽ$Ht H/qHDŽ$Mt I)jHD$L%HxXI9HID$HHW@A$@HXH]HJHx1@HH9cL;dut$8H C/H=e$wH|$H$H$H$VHH5H=1AIHr2HcI,$AD$8 9@HD$LLLD\$hE11E1HE1E1+dD\$hE1E1HD$HL$HD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0IfHI9HuL;%H;CAH;QCDI9LdI,$AEx1L$I,$>HDŽ$Ez@D$8w9E1E11HD$HE1E1L$HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0HD$HE1E11HD$(E1L$E1HD$ E1AHD$PHD$XHD$@HD$HD$HD$`HD$0D$89! ID$88MnIuHAI|$H|H7LgMHGI$HH$H$H/ H$H$H$L$H$蹊H$I,$~LqD$88HD$HE1E11HD$(E1E1E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$89LFLLT$4LT$.DDZH$LnH$IH5H;K@L;l$hI9LL$…9Im#HDŽ$x=H5xH$1H$IHPFH;?L;l$hI9LL$…EIm,HDŽ$KH9\$>HdH\H9P5KHf\HUHL-R\L$MFH5kLj>H$IHOH$H/2H5wH|$HDŽ$H$IHVH$H>H9GUH$HDŽ$L$χH$H$H/6L$HDŽ$MLH$H/V;H$1H$HDŽ$H$IHTH$H/;H$H;==HDŽ$H;|$h+H9+H$AŅWH/DHDŽ$E5iH0vH|$HH vH5iH$40*ϻLFH 4AL &Hx7隷DDDDD$89E1E11HD$HE1E1L$HD$(AE1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$03yH5`H|$ +IH/HD$LD$HH$I7HeLD$HHdH$HPI$H$L` hLD$HH$I3HT$`H5aHLD$%"H$H$LLD$'LD$HH$I;I(!H$H/!H$HDŽ$H/!H$HDŽ$HDŽ$HD$HD$HE1HD$(HD$ HD$PHD$XHD$@HD$RH=C]HHH5HIVHD$HE1E1L$HD$(1E1E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8:]H.sHDŽ$XLD\$D\$9H=b\EI郶HD$HE1E1L$HD$(1E1E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8:DDHD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8-9H=;[HFH5F蠾I6DǹI\$H1DHH9-I;|u+H=ZHYFH5ZFEIxHD$HE1E1L$HD$(1E1E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8:HLL$}LL$߼HD$iLL$饼I,$E_LHMH=Y׺IHD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8/93H=eYHI HD$HE11L$HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8 :HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$89 H=NXHCH5C賻IBLLLD$fLD$鬿H= XHvCLT$H5rCmLT$IHD$HE11L$HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8 : HGH$HHWHHH$H$H/QH$H$H$雬HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8D9HHD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$89LMɿH=UܶI/#LD$ LD$顽E1mfA.D$ADE LLD$LD$H$H$ؾHGH$HHWHHH$H$H/H$H$H$HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$89mHD$HE1L$1HD$(E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8:D$8f<E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$HD$p+LLD$LD$һLD$LD$頻MBMɮMrIII*fInfInߺLflH$LD$)$.jLD$H$I(LO靮EMD$8$:E11HD$HE1L$E1HD$(AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0-HD$LT$魯L鰼L\$L\$醼H;qgIHXDH;AH;D$hDI9LI/AEC[D$8H9E1E11HD$HE1E1L$HD$(AE1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0 LD$L$LD$HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8(:mLT$(LT$(خHGH$HHWHHH$H$H/!H$H$H$jH;uLIHM7H;AH;DI9|LI/AuL$ED$89E1E11HD$HE1E1AHD$(E1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0 HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8>:LT$T$iOHD$hE1E1E1HD$H1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$D$8#<HD$0H=:NHC9H5D9蟱I餴L_uAx6I(aH5QLLL$LL$HILHLL$(HD$vLD$LL$(HHD$0H$I(vHH$H;=.H9HT$h8H9/LL$H$LL$AH/,HDŽ$ETLL$(H|$LXH@LL$(H=08D$H9xH8H_ HL7MH5ILLL$(LD$LD$LL$(HHD$0H$BI(D$LL$LL$HIB$H$HyH9G#H$LL$HDŽ$L$$cLL$H$I.H$HD$0H$H$H/SH$H;=~HDŽ$AH;|$hDH9}LL$BH$LL$AI"H/ HDŽ$E(Ht>H}6H9P'Hd6H'HHP6HD$0H$H'H5IHLL$LL$HIG*H$H/H5dMLLL$(LD$HDŽ$LD$LL$(HH$Iw*I([H5P1LLL$8LL$HHD$0(H$HH9GB LoM5 HGIEHH$H$H/fInźLL$H$D$0H$)$`LL$H$Im HT$0HHD$HHhH$HD$0H7H$H/H$H;=6HDŽ$AH;|$hD H9LL$H$LL$A;H/HDŽ$E<D$\`LL$fT`胻LL$HHD$0HH$<LLL$胹LL$HHD$0H$q:H$H/!H$H;=?HDŽ$AH;|$hDH9LL$LL$ABH$H/T#HDŽ$LL$E&H5xOH=S1LL$LL$HH$IHHLL$3H$LL$H/[LL$AE11D$8;E1E1L$HDŽ$E1HD$HHD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0@DELLL$YLL$sLL$ELL$6IH)HDŽ$騠餱1L|$MILl$ HIHItH9,)LLl$ ML|$˶8H=wFZIHD$hE1E1E1HD$H1AHD$(HD$ HD$PHD$XHD$@HD$HD$D$8%<HD$0LLL$5LL$bD$89AHD$HE1E11HD$(E1L$E1HD$ E1AHD$PHD$XHD$@HD$HD$HD$`HD$0D$88荵v胵LT$(JLT$mT$HHD$hE1E1E1HD$H1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$D$8(<HD$0\HD$hE1E11HD$HE1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$D$8-<HD$0oÚD3LLL$TLL$HD$hAE11D$84<E1E1L$HD$HHD$(HD$ HD$PHD$XHD$@HD$HD$HD$0BLL$ȳLL$HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8:·HD$hE11E1HD$HAL$HD$(HD$ HD$PHD$XHD$@HD$HD$D$8/<HD$0>H=BHx-LD$H5t-ߥLD$I騩EHD$hE11AHD$HL$HD$(HD$ HD$PHD$XHD$@HD$HD$D$81<HD$0闶H=ALD$趢LD$ILL$LL$5HD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8;H54LLL$1LL$ɣD$8:E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0AHD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8z<ӴMH$11aI/EbIL9 MH$14HD$HE1L$E1HD$(1AHD$ HD$PHD$XHD$@HD$HD$D$8<'MD$8<E11HD$HE1L$E1HD$(AHD$ HD$PHD$XHD$@HD$鿳HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`D$8V:EЮGH=|>H)H5)HD$0m袮ڪ蘮鲪L苮铪D$8<E1E11HD$HAL$HD$(HD$ HD$PHD$XHD$@HD$锲Ht$H1{gIHD$8@E1E1A/L$NfDӭCHD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8T=бHD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8%;PHD$1HIıȬI/EnBL譬~LL$HD$(虬LD$(LL$_腬bH=1<HD$00eb[:LND$87=E1E11HD$HAE1L$HD$(HD$ HD$PHD$XHD$@HD$HD$KMD$8<E11HD$HE1L$E1HD$(AHD$ HD$PHD$XHD$@HD$HD$گHD$hE1E11HD$HAL$HD$(HD$ HD$PHD$XHD$@HD$HD$D$86<HD$0eHD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8X:ۮH;TLǺLL$(LD$跪LD$LL$(HHD$0eGH|$0H;=AH;=HD@H97LD$8LL$(TH|$0LL$(ALD$8HHD$HHExI(BEp鯜LL$AE11D$8;E1E1L$HD$HHD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0雭H$1HϣE1ff.GADED$8U=E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$HD$HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$89鉬EvD$8);IE11HD$HE1E1AHD$(E1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0LLL$vLL$g7]HD$HE1E11HD$(E1L$E1HD$ E1AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8w9BHD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`D$8[:˪HD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$D$8<fLLD$LD$AHD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$D$8<ukIBH$HMrHII* H$MH$AHT$01遒H5/<H=h@1AH$IH>AHH$H/=D$8d=E1E11HD$HE1E1L$HD$(AHDŽ$HD$ HD$PHD$XHD$@HD$HD$ƨHD$HE1E11HD$(E1AHD$ HD$PHD$XHD$@HD$HD$HD$`LL$D$8';aHD$HE1E11HD$(E1AHD$ HD$PHD$XHD$@HD$HD$HD$`D$8p:H5 4H|$ "H$IH"H'I9EY"I]HL"IEHHH$H$H/"fInH$fHnH$flƺ)$IH$H+!L$MD;H$H/!H$E1E11H$1HDŽ$H$IH:H$H/E!H$H5e5HDŽ$HDŽ$HT$`HGH|$HHEЅ,D$8?E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$ HLL$萡LL$ELulE1ffA.AADE=HD$HE1L$E1HD$(1AHD$ HD$PHD$XHD$@D$8<hOD$8t:IE11HD$HE1E1AHD$(E1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0HD$(fLD$(ΏLT$(RLT$(XC锝H=/HLL$H5OLL$IH=k;H571LL$8LL$HHD$0HH$<H$LL$H/LL$AHDŽ$D$88;HD$HE1E11HD$(E1E1HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0鱣HD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$D$8+=4连 H$H`LL$蛞LL$LL$E1E11HD$HE1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$HD$`D$8U;酢LL$ LL$$HD$HE1L$E1HD$(1AHD$ HD$PHD$XHD$@D$8<HD$HE1L$E1HD$(1AHD$ HD$PHD$XHD$@HD$D$8<鶡LL$E1E11HD$HE1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8S;1H=r,LL$PLL$ILL$虜LL$'D$8w=E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$HD$鎠HD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8v=HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8U9铟HD$HE1E11HD$(E1AHD$ HD$PHD$XHD$@HD$HD$D$8=AHD$8E1E1E1HD$H1L$A HD$(HD$ HD$PHD$XHD$@HD$HD$HD$D$8>ɊT7HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`D$8:\HD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8:͉HD$8E1E11HD$HE1L$A HD$(HD$ HD$PHD$XHD$@HD$HD$HD$D$8>UH=yuI$H=H+H5,wIHD$8E1E1E1HD$H1A HD$(HD$ HD$PHD$XHD$@HD$HD$HD$D$8>龈E1E1E1A+D$8^@L$鋈E1E1E1A+D$8\@L$gH=tHH=H H5vHÅLD$跃LD$HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`D$8;飇L+錆E1E1E1A%D$8M@L$r鳄MHMrMpIII(h&H$Ht$LL$LL$(H$f)LL$(II)rLLT$(臂LT$({rHD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`D$8:vHD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`LL$D$8;MD$8:E11HD$HE1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0 VH&H$LHDŽ$L${'H$HD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8?鰄HD$HE1E11HD$(E1L$AHD$ HD$PHD$XD$8=]HD$HE1E11HD$(E1L$AHD$ HD$PHD$XD$8= HD$HE1E11HD$(E1L$AHD$ HD$PHD$XD$8=鷃HD$HE1E11HD$(E1AHD$ HD$PHD$XD$8 ={HD$HE1E11HD$(E1AHD$ HD$PHD$XD$8=0H=bEoHD$@|LL$~LL$HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`D$8;}HD$HE1E11HD$(E1L$E1HD$ AHD$PHD$XHD$@HD$HD$HD$`HD$0D$8H9LL$AI1D$8;E1E1E1HD$HE1HD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0wHD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`LL$D$8;H=H51LL$LL$HHD$0HH$`CH$LL$H/5LL$AHDŽ$D$8;yH= HH5.oHD$@ zHD$8A E11D$8>E1E1L$HD$HD$HHD$(HD$ HD$PHD$XHD$@HD$HD$HD$8E11E1HD$HA L$HD$(HD$ HD$PHD$XHD$@HD$HD$HD$D$8>wLzKLD$zD$ҟzzzE1酞z{z驠HD$8E1E1E1HD$H1L$A HD$(HD$ HD$PHD$XHD$@HD$HD$HD$D$8>~E1E1E1AD$8?L$~HD$HE1E11HD$(E1A HD$ HD$PHD$XHD$@HD$HD$D$8>(~yƝy|y|Ly9|E1E1E1AD$8 @L$}LwMiHGIHH$H$H/ Ht$fInƺH$D$0)$H$I.iLxiHD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`D$8:|D$8:E1E11HD$HAE1L$HD$(HD$ HD$PHD$XHD$@HD$HD$HD$`HD$0X|H|$XE1A1H$19IHH H|$8Hu0I,$hLw[LwEE1nw鯝dw{ZwH$HD$HE1E11HD$(E1L$A HD$ HD$PHD$XHD$@HD$HD$D$8=J{H5H=,1H$IHHH$H/D$8=E1E11HD$HE1E1L$HD$(AHDŽ$HD$ HD$PHD$XHD$@HD$HD$zvLL$v6D$8;E1E11HD$HE1E1L$HD$(AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0yHD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`LL$D$8|;yuLL$HD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`LL$D$8;yHD$8L|$ ME1Ld$L|$(E11Hl$(AHl$pHD$L$D$8>HD$HD$xIHD$8L|$ E1Ld$L|$(E11Hl$(AHl$pHD$L$HD$HD$D$8>;xs*HD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8=wHML|$ Ld$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HD$L$HD$D$8?dwHML|$ Ld$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HD$L$HD$D$8?wrՠHML|$ Ld$HD$HHD$8E1E1L|$(Hl$p1LT$(HD$AE1L$D$8?HD$HD$vLrGHML|$ Ld$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HD$L$HD$D$8?)vLHPILd$HHH L|$ HDHLd$1H5L|$(Hl$pH81WwLT$E1HD$8Lt$HL$E1ALT$(E1HD$D$8?HD$HD$uqƞLL|$ Ld$IHD$PHD$8E1E1L|$(E1Hl$(1HD$Hl$pA HD$HD$D$8v> uLp1LLL|$ Ld$E1HD$PHD$8E11L|$(LD$A Hl$(L$HD$Hl$pHD$D$8a>tL8p&HD$HE1E1L$HD$(1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`HD$0D$8;#tD$8w@A,E1E1L$tD$8u@E1E1A,L$sHD$8L|$ ME1Ld$L|$(E11Hl$(AHl$pHD$L$D$8>HD$HD$sIHD$8L|$ E1Ld$L|$(E11Hl$(AHl$pHD$L$HD$HD$D$8>+sE1E1E1A+D$8g@L$snfHD$HD$8E1E1L|$ Ld$1AL|$(HD$Hl$(L$HD$Hl$pD$8>rHD$8LD$E1E1L|$ Ld$E11L|$(HD$AHl$(Hl$pHD$D$8>irH=~^I۔H=H H5!`I鹔HD$HD$8ME1L|$ Ld$1AL|$(HD$Hl$(L$HD$Hl$pD$8>qHD$8L|$ ME1Ld$L|$(E11Hl$(AHl$pHD$L$D$8>HD$HD$uqHD$HD$8ME1L|$ Ld$E11L|$(HD$AHl$(Hl$pHD$D$8>5qIHD$8L|$ E1Ld$L|$(E11Hl$(AHl$pHD$L$HD$HD$D$8>pHD$HD$8E1E1L|$ Ld$1AL|$(HD$Hl$(L$HD$Hl$pD$8>wpLkZLk;HD$HD$8MML|$ Ld$E11L|$(HD$AHl$(L$HD$Hl$pD$8>pIHD$8L|$ MLd$L|$(E1E1Hl$(L$1AHD$Hl$pHD$HD$D$8>oL0k霔&k HD$8L|$ E1E1Ld$L|$(Lt$1Hl$(AHl$pHD$HD$D$8>ToHGH$H0HWHHH$H$H/tH$j͑}jH$ݑfj#LYjYL$LIL$&{HD$8E1E1E1HD$H1L$E1HD$(A HD$ HD$PHD$XHD$HD$HD$D$8!>8nHD$8E1E1E1HD$H1A HD$(HD$ HD$PHD$XHD$HD$HD$D$80>mHD$8E1E1E1HD$H1E1A HD$(HD$ HD$PHD$HD$HD$D$8D>pmHD$8L|$ ME1Lt$PLd$1A L|$(HD$E1Hl$(Hl$pD$8T>L$HD$HD$mLLD$Ld$E1HD$PHD$8E1E1L|$ 1L|$(A Hl$(Hl$pHD$HD$D$8_>lHD$@LL|$ Ld$L|$(H\$PHDt$|HD$HD$8Hl$(Hl$pHD$HD$ggЖg阖MFHD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8?kHD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8?ukLLL$fLL$0HD$hHD$0tH$LHML|$ Ld$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HD$L$HD$D$8?jHML|$ Ld$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HD$L$HD$D$8?jL fZf2HMLLd$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HT$ L$HD$HD$D$8s? jLe頒eWH$LǺHDŽ$H$ H$`HMLLd$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1H|$ L$HD$HD$D$8_?\idBdHMLLd$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1Ht$ L$HD$HD$D$8R?hLnd鈐ddAfH$1L)$ H$SHMLLd$HD$HHD$8E11L|$(LD$ALT$(Hl$pE1HT$ L$HD$HD$D$8>?BhD$8c@E1E11L$A+hHD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8`=gHD$HE1E11HD$(E1E1AHD$ HD$PHD$XHD$@HD$HD$HD$`LL$D$84;IgLL$E1E11HD$HE1L$AHD$(HD$ HD$PHD$XHD$@HD$HD$HD$`D$8;fLd$L|$(Lt$Hl$(Hl$pHHMHHH AHEHE11H5H81gHD$E1E1HD$L$AHD$ HD$8HD$HD$D$8 ?fH;gfH$IHH$H/H$HDŽ$HGLAIHH$AH$H'H$AIHjHD$LD$H$H/HDŽ$ HD$AE11HD$L$E1E1HD$ HD$8HD$HD$D$8 ?dHE11H5)H81WfHD$E1E1HD$L$AHD$ HD$8HD$HD$D$8 ?d`BH`x`LLD$L|$(MHD$ HD$8E1E1Hl$(Ld$1AHD$Hl$pHD$L$D$8'?dHD$HML|$(Ld$Hl$(HD$HHl$pLt$I LT$hE11HiH5 H818eHD$LT$hHD$HD$L$AE1HD$ HD$8D$8'?HD$ocL|$(Hl$(MHl$pLd$Lt$AH$H/LT$HDŽ$LT$t`HD$AE11HD$L$E1HD$ HD$8HD$HD$D$8/?bLLT$V^LT$IHΩLLT$hH cH5HEH0E11H81dHD$LT$hHD$HD$L$AE1HD$ HD$8D$8/?HD$8bLT$]LT$Ld$L|$(Lt$IHl$(E1Hl$p]LLD$Ld$E1HD$ HD$8E1E1L|$(1Hl$(AHD$Hl$pHD$D$8?aHWHHGL H@H$HD$HE1E11HD$(E1L$AHD$ HD$PHD$XHD$@HD$HD$D$8=aL\vLLL$\LL$7w\LL$]LLL$(`\LL$(H|$_:ILL|$ Ld$HD$PHD$8E1E1L|$(1Hl$(A HD$Hl$pHD$L$HD$D$8S>T`D$8O@E1E11L$A%E1.`L[郆[b[:LL|$(MHl$(HD$ HD$8E11Ld$LD$AHD$Hl$pHD$L$D$8>_HD$8Lt$ ME1Ld$L|$(E11Hl$(AHl$pHD$L$D$8>HD$HD$[_ILLd$L|$(HD$ HD$8E1E1Hl$(1Hl$pAHD$L$HD$HD$D$8>^LLd$L|$(E1HD$ HD$8E11Hl$(L$ALD$Hl$pHD$HD$D$8>^1Z"LLD$Ld$E1HD$ HD$8E1E1L|$(1Hl$(AHD$Hl$pHD$D$8>Y^LLd$L|$(E1HD$ HD$8E11Hl$(L$ALD$Hl$pHD$HD$D$8>]H=4JIH= HH5LILEY|;Y[1YzH$Ygf.DUf(SH(-Hf/\f(f\H,H*XZ H\$^T$YYX HY\ HYX HY\ HYX HY\ ~HYX zHY\ vHYf(L$YL$X ]HT$f(5G\$^f(\EHf/YX -HX\vIH~DHD\`L$Hf(T$OYL$H9T$\uH(f([]f1ff.fHHH?Pf*YIHff.HGH?H~CAUIATIUHS1HfDH}UADHI9uH[]A\A]fDH~SAUIATIUHS1HfDI}AUf*YmHAHH9uH[]A\A]AUIATL%R UH-JSHBH(I}AUfHH HH*YLH9IEI}AL$HcAT$\$L$D$f(fW%Gf(bUT$L$f($YD$Xf/KH(f([]A\A]f.fWvGQW EH([]\A\A]f(fDHWAWIAVE1AUL-$ATL%UHSHH(f.HL$HcLT$\$L$D$f(fW%Ff(^TT$L$f($YD$Xf/BLIL9tkI?AWfHH HH*AY I;DrIGM8LfW;FV D\BLIL9uH([]A\A]A^A_Lf(bff.@AVAUIATL%UH-SHHI}AUfɉ *YL9IEI}AL$HcAT$ \D$L$A(W-SEA(gVT$ L$(fA*YDYD$X/HH([]A\A]A^@f*YDWDQ DH[]\A\A]A^(AWAVAUATUSH(Ht$H,HHL%wE1L-mAH UL$McLBT$\D$L$A(W-9DA(MUT$L$(fA*YCYD$X/BLIL9t$tpH;Sfɉ *@DAY A;DrHCL @.Lf*Y^CWCO NC\@H([]A\A]A^A_fHh(VH~[AUIATIUHS1HfDI}AUfWCRfWCAHH9uH[]A\A]fff.@H~sAUIATE1UHSHHDI}AUf*YmBWBZeRZWBBDIL9uH[]A\A]AWAVIAUL-00ATL%'8UHSH(H(IFI>L$HcT$\$L$%L@D$Yf(Y]OT$L$f($YD$Xf/wAI>AVfII LH!H*AYLtfW aAI94?H(f([]A\A]A^A_@IFI>fW/A Q ?I>Y $AVfW AP $fW@f(XYf/vX N?AzfW @m@SHH0=XD$f/H;SHD$ ,t$T$\f/r> D$f(^O\$f(f/rH0f([fD$L$ \^D$O|$L$ D$f(Y\f( ^zOT$\$f(\f/H0f([ff(\%$>$>Yd$(ff.Qf(^d$fHff(D$YXf/sf(L$H;YYD$SL$=f(YYY\f/wbL$ ND$D$N%~\d$f(L$ f(X=YD$(YYXf/D$(L$(YL$H0[f( S%H~CAUIATIUHS1HfDLADHH9uH[]A\A]DAWAVIAUATL%UH-SHHDIFI> HcL$ \D$%/<fAnfZAYAf(Y2KL$ f(fA*YO=YD$XZf/wDI>AVfAA A*AYfA~t W?=fA~D9l/HfAn[]A\A]A^A_IFI>f*Y<W<0I <I>YL$AVf*Y<W<HL$W<(XY/vX g<AfA~RW z<fA~AH~CAUIATIUHS1HfDLADHH9uH[]A\A]Df.{&ff.{ 7uf(fuifUHH .;D$ud$f.R=;/|$%e;d$H}UfH*YT$T$ z :;t$T$ \/r- ;D$ (^cM\$ /rH ]fD$:L$\^D$M|$L$D$ (Y\( :^LT$ \$\/#H ]Dt$\5v:fn:Yt$.;Q=G:d$^|$H f(D$YX:/s(L$H}YYD$ UL$9f(*YYD$YY9\/w_L$LD$D$ yL=9\|$ (L$(X{9YD$YYX/D$D$YD$ H ]@H f]H ]P K%8ff.HHH?PHHff.HHH?PHff.HHH?PHHff.HGH?f.@E„uf.7D„uffff.H$L$YD$X$Hff.UHH0H?D$UH=fHH HH*Y H= H;sD$H0]Y@HEH}HL$ HcT$(\D$L$ D$f(fW%~7f(ET$(L$ f(D$YD$Xf/_Hf(D$H0]YfW7F q5\D$H0]Yff.fHHH?$L$PYD$X$HHf.\{Bff.{ L$L$HYfDufHYfuL$L$HY@HL$ qYD$ HfDUHH0D$ f/L$(f/f(4f/vf/n@H}UH}D$U l^L$ D$D$ME M^L$(D$D$.EXD$(f/rL$XL$ff/vf/JL$H0]^f(fD\$ f.|d$ ff.D$ HJf(\$(f.{ht$(ff.D$(HL$L$X^H0f(]fD~H2\$(f.<f(zuHL$L$H?U\$ H f(D$(XYfHnf/wf}DPfaff D$Cf(D$^L$ L$CL$f(^T$(T$ f(_\\$f(L$SA\$T$ D$\f(4AXD$9CL$H0]\f(AHY1f.{*ff.{HXfufHDuHXUHHh1$Yf.f(ff.{sHL$L$f(XYY 1f. c}ff.{;f(HT$T$XY$H]^f(@ufufeL$L$f(Xm@}HT$T$XUHHHD$L$H]^f(ÐUHH0H?D$UH=fHH HH*Y H=H;s^L$H0f(]BHEH}HVL$ HcT$(\D$L$ D$f(fW-&1f(>T$(L$ f(D$YD$Xf/WH{f(FffW0@ )/\$UfHH0f.zu H0f(]f @H}^L$UH=pfL$HH HH*YH=DH;HEH}H L$(HcT$\$ \D$T$D$f(fW-/f(=\$ T$f(D$L$(YD$Xf/wHL$:L$f(H0f(]2?ffW/a?-L$\ff.fHD$~ I/fW@~ 8/fW ^L$H>ff.fSHH$L$ ff/wFH;Sf/8-r n-\\f(u>L$Y$H[\@XO>YD$X$H[ff.AVfI~SHHL$H;Sf(\f/v=fW7.=L$H[YfInA^\SHH$L$DH;Sff/v \^=YD$X$H[fH$L$\YD$X$H8;UHH0H?D$UH=efHH HH*Y H=BH;s,Xff.QD$H0]YfDHEH}HL$ HcT$(\D$L$ D$f(fW%,f(]:T$(L$ f(D$YD$Xf/GHf(6ffWv,Q< *\f(@f(ff.UHH D$L$Y w*f. f({qff.{_f(Hl$\$L$)L$\$l$f(f.wtQYf.wKQf(H ^]Duf(uHL$D$L$\$ff(f(L$?L$f(f(l$d$\$?l$d$\$f(ZAWf(AVAUATUSHHHf/)D$f.(zuE1HHL[]A\A]A^A_fDD$fW*E1Z8 ZD$f.IL$H;SL$Yf/L$w@ff(f.>Qf(f(L$19L$Y (X (D$0(Yf(L$ \ (\-(f((\=(^ (f(|$8XfI~X(fI~(^\fI~fDH;SH;f(\'T$ST$ 'D$f(fTB\fIn^L$XD$ YXD$X?(=L$f/ 4(L,\$rfInf/0M_=(f/v f/Gf(L$(8D$fIn8L$(t$8D$D$ Y^X_8T$XT$ID$fI*YL$0\f\L$H*f.Eńu3f.&D„u!L$T$WL$T$\f/}@D$ff.{,HT$T$YHH]ufuT$#T$ff.AWfAVH*AUIATIUSHH$t H9r Me$AEf(\A}|$Pf/t$H# L$PT$Yf(fD$HAEX\$AM(f(L$:\$Y\$HH,fL$T$f.Im0f(\$h Qf(Y%\$HY%%6-~%\f(fTf.05$d$PXf(t$(fD(D$AE8ffA(H*XX)%Xf(Am@$A\f(\$xA]P$Y|$8^f(A}H\X$fD(\$@A]Xf(\^f(YXYf(Yl$H\\$`^A]`Yf(XYfA(AXXL$pAMhAYf(D$AEpfA(^D^Xf(D$XAExfA(XD$ AMl$LH)H$fH;Sd$ H;Yd$Sd$f/d$f(Vf/d$"t$l$@fH*\YT$0f(^XXD$8\X\$(fT^\f/L$[7L$T$0L,MI)LH?HL1H)H~#D$hYD$(fH*\f/fd$P^d$HI*YI94f(jf/$M)f/L$(MGH[]LA\A]A^A_Df.X!ADEf/d$Xwcd$0D$2^D$`XD$87L,MSEJd$0\d$L$YYL$`d$0D$2\$x^D$p\f(6L,M9EL$d$0\d$X=YYL$pqfDHEI9IVf(ff(H*H^\YH9uIFH9HUff(H*H^\^H9uWf(^4!t$hHX+!HHf(XYX!^^XT$(YfH*^f(T$0\$D1\$T$0f(f(\f/Xd$0f/IFfEfEL*HEfEL*LfL*$L)HH*fE(D$EYfA(fA(D$A^fE(D$AYEYfD(l$DYD$$D$D$N0D$$D$fA(^D$!0T$HD$$t$Yt$PAYf(^/=5L$-~H)f(f(D$D$^d$0D$DY$D$D$D$D$\f(^\ fD(^D\D^ f(A\fEL*DX\$(DY$A^DEXfEM*AYDXf(A^A^AXfD(D\f(E^A\fD(A^D\f(E^A\fD(A^A^Xf(A^A^D\f(E^\A^A\fD(A^\A^D\f(E^\A^A\A^\^L$A^A^XXf/f($YT$\X1L,~fDH,ffUH*f(fT\fVf(f(t$P\l$Hf( f.Bhf/B]J8rz Hj0L$J@Yt$P5$JH|$HL$8JPYt$(L$xJXL$@J`L$`Jh\$hL$pJpL$JxL$XL$ D$hT$L$k1T$L$Ff.AUIATIUSHH8D$t H9rWd$-1fMeI*AE\AeAm f($l$,$Y)d$$L$f(AEfYYAeXX f.Q-f(YXf/aH,Im0$H;S$1f/f(vp@HH9})$H;S$f/1f(HH9|Lf\H)HH*YT$YfH*YT$^f/wH8[]A\A]Ðf.Bf/Bz ZHj0|$9H81[]A\A]f(d$(\$ T$ $s/d$(T$-w $\$ YXf/f(l$ d$$(/$d$l$ f(YXH,ff.Hf.<EurfUHH*f/rYf/rM]O\|Yf/f(r&!IH]L)fD1D]Bf;IH]L)UHH$f(L$'L$ff.$f/\Yf.ff.H $d $XD$H $ $ff(f.%QXD$YXH]D HYfHH*X$YHDfHnf.ff.{RHHX]Ð9HD$G@$Yf.{kff.zufKfDH $ $XD$f]HrXft$f(N,$f(UHH0$f(L$T$$vT$ff. <$f/\%f(Yf.ff.H\$ d$T$UXT$d$\$ D$(H\$ d$T$T$fd$\$ f(f.QXYXT$(@%H\$Yd$f(\fd$\$HH*X$Yf.Yff.H\$d$od$\$f(X!f%Xl$YYf.{Dff.{2f(HT$T$XY$H0]^f(ufuHT$ T$X@KHD$(s@%$Yf.{sff.f.fDH\$ d$T$Hzd$\$f(Xu\$d$D$f(\$ d$5)\$ d$L$f(f.f(f(SHXf(H ^L$\$l$\$L$Y f(YYYf(YXff.w\Q\YT$\$H;XT$S\$T$f(f(X^f/s Y^f(H f([f(\$T$2(\$T$f(UHH@D$8f(L$  f/D$f/D$s|$f/ -ft$Yl$YXf.Qf(XL$f(Xf.JQ\T$f(f(X^f(YXXL$^L$0&f(^3"XD$L$ \f/D$sH}UYw"\$0H}f(YXXL$f(L$^\d$(YL$ UL$ f(\Y\f/D$\H}UD$D$(T"f( f/D$vfWXT$8 jf(fTnT$X0{!t$T$\f/vfWiH@]H}UX\YH@]f^D$f.QD$HYD$ XD$8f/vXf/q\dD=hf(|$|$^Xf|$D$0(%D$qf(L$ %L$ f(f($l$f(hff.fSHH0D$ fW* D$(H;Sf/D$ D$H;SYD$(f!T$f(fWYf/~f(T$\$q\$D$f(\L$^BXa#L,MaT$ff.E„EH0L[f/ArA@HHH?D$\$PL$$f(f/vf(fDYHXf/wHÐHD$a~T$$fWf(~ $f(fW f(^Tf(fTf.v3H,ff(% fUH*fTXfVf(f/HsH,HDHf/ Ir_H?D$\$APL$$f(f/f(vYHXf/wHfD$5~ L$$fWf(e~} $f(fW f(,^f(fTf.w*f/ HtH,HH,ff(%fUH*fTXfVf(ff.AVSHH(\vf(D$T WD$H;SH;D$S%>\d$fI~ #^L$f(= f(0 fTf.v;H,f=H*f(fT\ fUf(fVf/ ]5f/Kf(L$T$^X~T$l$f(fInYf(\a^Yf(\M^f/H(H,[A^f.f(f(H8H\\T$H?L$D$(f(d$^l$ 4$P4$d$L$T$f/r;l$ \$(Yf(Yff.wPQXH8f(f(\ x\fYYf.w*Q\H8f(f($+$f(f($$f(1HATIIUHI SHLHI LHL IIL III LH I ĸH9wfDH;SD!H9r[]A\DH;SL!H9sH;SL!H9rf.ff.@AWAVAUATIUSHHttHHH?IIH9wqHEAEDjAME9v,D1AAA9sH}UAME9wI IHL[]A\A]A^A_@HEHEuQLbIHHI9v+L1HIIH9sfDH}UIHI9wHL$DHEH}L!I9rL$oDHEH}D!D9wL$ODЉI>@I0fDAWAVAUATAUSHteHGIH?AEu`JL$ Dl$ AME9v%1AAA9sI>AVAME9wI AHD[]A\A]A^A_@IFI>D!9rD$f.AfAWAVAUATUSHt$H\$PfA։IAMfAAEEF-DD$ H?AUDD$ A$3EAfA9v^AAƙAAf9r&ID3A$xA<$3AfA9v%uI}AUA$3AfA9wD$H[]A\A]A^A_+A$PA$D!f9s!uI}AUA$D!f9rfD$H[]A\A]A^A_fuDH?AUA$D$fH[]A\A]A^A_@+AxA9+A)ff.@AWAVAUATUSHt$ H\$PA։IAMAAEE~H?AUA$3D@A8v^AEAƙAA8r#Ff3A$xDA<$3@A8v%uI}AUDA$3@A8wD$ fH[]A\A]A^A_f+A$PA$D!@8s!uI}AUA$D!@8rD$ H[]A\A]A^A_uDH?AUA$D$ H[]A\A]A^A_D+AxA9+A)ff.@USHHl$ tAHL˅tmA)EH[]fDH?QEAWAVAUATIULSH(HL$HujH~OHHHfInLHHLHH9uHt$HHtLdH([]A\A]A^A_IHH9EH|$~DrE1Dt$D$fI}AUt$I9s$D$19s@I}AUI9wH LJLIL9|$uLf.HEE1HLrH|$H\$fI}AUIHHL9s-HD$1IHH9sI}AUIHH9wHLJDIL9|$ufH|$1fI}AULHDHH9\$ufDH|$n1f.I}AULHDHH9\$uBIII LHI LHI LHI LHI LH I H|$E1fDI}AUL!H9rLJDIL9|$uIII LHI LHI LHI LHI H|$E1I}AUD!9wLJDIL9|$u[1OfAWAVAUATMUHSHt$H~pHAHHfnLHfpHLHH9uHH@t,t$HHHA4H9~HAtH9~AtH[]A\A]A^A_@IӃEDrE1Dt$\$ H~f.I}AUt$I9v$D$ 19s@I}AUI9wH L$C IL9uH[]A\A]A^A_fIII LHI LHI LHI LHA HE1I}AUD!9rD$CIL9uH[]A\A]A^A_ÐH1DI}AUD$AHH9uH[]A\A]A^A_1zAWAVAUATUSLHfHHAHCHfnLHfaHfpLʐHH9uHHt\Hxf4CHH9~KHxftH9~=HxftH9~/HxftH9~!HxftH9~Hft H9}ft H[]A\A]A^A_DIAfHEH~DrII1H$A1D$ ufI?AWfD9s@D$ Af9s1t1fA9vI?AWȉfA9wۿHDfKH;$2t1뒐AHHH HHH HHH HH HM4I11t@1!fA9sI?AW!fA9rDHfSL9uH[]A\A]A^A_@HI,IHDfCH9mI?AWATfHSH9uP1DAWAVIAUATUSH@t$ uVH~ff.{,HT$FT$YH;H[ufuHT$H8PHfHn\f(T$fWfDUHHWt9OGHGHL$L$H]^f(Ef(ȅtEEHEH^]f(fDAVUHSH D$f/L$f/DHEH8PHEfI~H8P ^L$fH~fIn ^L$$fHneX$=`f/rf/d $H []^A^f(fDT$f."T$ff.D$Hbf(\$f.T$ff.D$H $# $XH []A^^f(fvHEH8Pf(|\{\$f.ef(fW qkHE $H8Pf(2\1 $fWtaH9fL$HHNfMHMfI)H*I9HD$8LI*LOfIH*fI*L)|$f(^f(YXuD$fH*ID$Y%\YfH*Y^X4f.!f(Qf(IGfIYT$0M)XL$(\$XD$ fH*HD$8HH*YfI*^2L,fID$H*誥HD$8D$fL)HH*芥L\$L)HXfHH*\$_XD$KD&Lt$@D$fH*;d$L9|$8L$(T$0Xd$HY D$n%X6Xf(fTf.]T$0fDH;SH;D$SL$|$\YD$ ^XD$f/wf/D$0sML,fID$M)H*aHD$8D$(fL)HH*Ad$(IFXfH*d$(XD$(HD$@JD D$(fH*XD$(T$HL$\\Y\f/s=f(T$\Yf/f(T$Xf/L;l$PMOM)L9MOffH*fI~DH,f5fUH*f(fT\fVf(H\$Xf(T$ \$0T$ \$f(ff.[ff.ff.f/Ir fDHH?D$APfW/ L$$\$^f(HH,f[ff.SHH@D$8f(L$ 3Kw[f/D$ o?|$ f/'fl$D$^XD$0,f(^XD$L$\f/D$s}H;SY@\$0H;f(YXXL$f(L$ ^\d$(YL$SL$f(f\Y\f/D$^H;SD$D$(f(f/D$vfWXT$8 5f(fT9T$XF|$T$\f/w,H@[fDH;SX\ƎYH@[fWH@[f -ft$Yl$YXf.w[Qf(XL$f(Xf.wTQ\T$ f(f(X^f(YXXL$^L$0@f(|$f(f(L$L$f(SHH05؍D$ \f(D$(H;Sf/D$ D$H;SYD$(T$\f(Yf/r~f(T$\$i\$D$f(TL$^:XYL,MaT$ff.E„EH0L[f/ArAHH%.200s() keywords must be strings%s() got an unexpected keyword argument '%U' while calling a Python objectNULL result without error in PyObject_Call%.200s() takes no arguments (%zd given)%.200s() takes exactly one argument (%zd given)%.200s() takes no keyword arguments__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)%.200s does not export expected C variable %.200sC variable %.200s.%.200s has wrong signature (expected %.500s, got %.500s)%.200s does not export expected C function %.200sC function %.200s.%.200s has wrong signature (expected %.500s, got %.500s)Interpreter change detected - this module can only be loaded into one interpreter per process.unbound method %.200S() needs an argument__annotations__ must be set to a dict object__kwdefaults__ must be set to a dict objectchanges to cyfunction.__kwdefaults__ will not currently affect the values used in function calls__defaults__ must be set to a tuple objectchanges to cyfunction.__defaults__ will not currently affect the values used in function callsfunction's dictionary may not be deletedsetting function's dictionary to a non-dict__qualname__ must be set to a string object__name__ must be set to a string object%.200s.%.200s is not a type object%.200s.%.200s size changed, may indicate binary incompatibility. Expected %zd from C header, got %zd from PyObject%s.%s size changed, may indicate binary incompatibility. Expected %zd from C header, got %zd from PyObject'%.200s' object is unsliceable%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 BaseExceptioninvalid vtable found for imported typeCannot convert %.200s to %.200svalue too large to convert to intnumpy/random/mtrand.cpython-39-x86_64-linux-gnu.so.p/numpy/random/mtrand.pyx.c%.200s() takes %.8s %zd positional argument%.1s (%zd given)numpy.random.mtrand.RandomState.randomnumpy.random.mtrand.RandomState.__reduce__numpy.random.mtrand.RandomState.__getstate__numpy.random.mtrand.RandomState.__str__numpy.random.mtrand.RandomState.randnumpy.random.mtrand.RandomState.randn'%.200s' object has no attribute '%U'numpy.random.mtrand.RandomState._initialize_bit_generatornumpy.random.mtrand.RandomState.__repr__cannot fit '%.200s' into an index-sized integer'%.200s' object is not subscriptablenumpy.random.mtrand.RandomState.normalnumpy.random.mtrand.RandomState.laplacenumpy.random.mtrand.RandomState.logisticnumpy.random.mtrand.RandomState.lognormalnumpy.random.mtrand.RandomState.gumbelnumpy.random.mtrand.RandomState.__setstate__numpy.random.mtrand.int64_to_longnumpy.random.mtrand.RandomState.hypergeometricnumpy.random.mtrand.RandomState.triangularnumpy.random.mtrand.RandomState.uniformnumpy.random.mtrand.RandomState.bytesnumpy.random.mtrand.RandomState.__init__hasattr(): attribute name must be stringModule 'mtrand' has already been imported. Re-initialisation is not supported.compile time Python version %d.%d of module '%.100s' %s runtime version %d.%dShared Cython type %.200s is not a type objectShared Cython type %.200s has the wrong size, try recompilingbase class '%.200s' is not a heap typeextension type '%.200s' has no __dict__ slot, but base type '%.200s' has: either add 'cdef dict __dict__' to the extension type or add '__slots__ = [...]' to the base typemultiple bases have vtable conflict: '%.200s' and '%.200s'numpy.random._bounded_integersPyObject *(PyObject *, PyObject *, PyObject *, int, int, bitgen_t *, PyObject *)int (double, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type)int (PyArrayObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type)PyObject *(void *, bitgen_t *, PyObject *, PyObject *, PyObject *)PyObject *(PyObject *, PyArrayObject *)PyObject *(void *, void *, PyObject *, PyObject *, int, PyObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyObject *)PyObject *(void *, void *, PyObject *, PyObject *, int, int, PyObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type)PyObject *(void *, void *, PyObject *, PyObject *, PyArrayObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyArrayObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type, PyArrayObject *, PyObject *, __pyx_t_5numpy_6random_7_common_constraint_type)_ARRAY_API is not PyCapsule objectmodule compiled against ABI version 0x%x but this version of numpy is 0x%xmodule was compiled against NumPy C-API version 0x%x (NumPy 1.16) but the running NumPy has C-API version 0x%x. Check the section C-API incompatibility at the Troubleshooting ImportError section at https://numpy.org/devdocs/user/troubleshooting-importerror.html#c-api-incompatibility for indications on how to solve this problem.FATAL: module compiled as unknown endianFATAL: module compiled as little endian, but detected different endianness at runtimenumpy.random.mtrand.get_bit_generatornumpy.random.mtrand.set_bit_generatornumpy.random.mtrand.RandomState.rayleighnumpy.random.mtrand.RandomState.exponentialnumpy.random.mtrand.RandomState.shufflejoin() result is too long for a Python stringnumpy.random.mtrand.RandomState.get_statenumpy.random.mtrand.RandomState.seednumpy.random.mtrand.RandomState.poissonnumpy.random.mtrand.RandomState.noncentral_fnumpy.random.mtrand.RandomState.set_statenumpy.random.mtrand.RandomState.tomaxintnumpy.random.mtrand.RandomState.multinomialnumpy.random.mtrand.RandomState.random_samplenumpy.random.mtrand.RandomState.dirichletnumpy.random.mtrand.RandomState.standard_exponentialnumpy.random.mtrand.RandomState.standard_normalnumpy.random.mtrand.RandomState.standard_cauchynumpy.random.mtrand.RandomState.random_integersnumpy.random.mtrand.RandomState.waldnumpy.random.mtrand.RandomState.vonmisesnumpy.random.mtrand.RandomState.noncentral_chisquarenumpy.random.mtrand.RandomState.fnumpy.random.mtrand.RandomState.betanumpy.random.mtrand.RandomState.negative_binomialnumpy.random.mtrand.RandomState.chisquarenumpy.random.mtrand.RandomState.paretonumpy.random.mtrand.RandomState.weibullnumpy.random.mtrand.RandomState.standard_gammanumpy.random.mtrand.RandomState.standard_tnumpy.random.mtrand.RandomState.powernumpy.random.mtrand.RandomState.geometricnumpy.random.mtrand.RandomState.logseriesnumpy.random.mtrand.RandomState.zipfnumpy.random.mtrand.RandomState.gammanumpy.random.mtrand.RandomState.randintnumpy.random.mtrand.RandomState.permutationtoo many values to unpack (expected %zd)need more than %zd value%.1s to unpacknumpy.random.mtrand.RandomState.multivariate_normalnumpy.random.mtrand.RandomState.binomialnumpy.random.mtrand.RandomState.choice'%.200s' object does not support slice %.10s_cython_3_0_11.cython_function_or_methodnumpy.random.mtrand.RandomState RandomState(seed=None) Container for the slow Mersenne Twister pseudo-random number generator. Consider using a different BitGenerator with the Generator container instead. `RandomState` and `Generator` expose 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 bit generator using 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. `RandomState` is effectively frozen and will only receive updates that are required by changes in the internals of Numpy. More substantial changes, including algorithmic improvements, are reserved for `Generator`. Parameters ---------- seed : {None, int, array_like, BitGenerator}, optional Random seed used to initialize the pseudo-random number generator or an instantized BitGenerator. If an integer or array, used as a seed for the MT19937 BitGenerator. Values 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 the `MT19937` BitGenerator is initialized by reading 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. See Also -------- Generator MT19937 numpy.random.BitGenerator Bad call flags for CyFunction__pyx_capi____loader__loader__file__origin__package__parent__path__submodule_search_locations%.200s() needs an argumentkeywords must be stringsMissing type objectcannot import name %San integer is required%s (%s:%d)at leastat mostrandomnumpy/random/mtrand.pyxexactly__reduce____getstate__randnBitGeneratorlaplacelogisticlognormalgumbel__setstate__name '%U' is not definedhypergeometrictriangularuniformbytes__init__builtinscython_runtime__builtins__does not match_cython_3_0_114294967296complexnumpydtypeflatiterbroadcastndarraygenericnumberunsignedintegerinexactcomplexfloatingflexiblecharacterufuncnumpy.random.bit_generatorSeedSequenceSeedlessSequencenumpy.random._commondoubleLEGACY_POISSON_LAM_MAXuint64_tMAXSIZE_rand_uint64_rand_uint32_rand_uint16_rand_uint8_rand_bool_rand_int64_rand_int32_rand_int16_rand_int8check_constraintcheck_array_constraintdouble (double *, npy_intp)kahan_sumdouble_fillvalidate_output_shapecontdisccont_broadcast_3discrete_broadcast_iiinumpy._core._multiarray_umathnumpy.core._multiarray_umath_ARRAY_API_ARRAY_API is NULL pointernumpy/__init__.cython-30.pxdnumpy.import_arrayinit numpy.random.mtrandnumpy.random.mtrand.ranfnumpy.random.mtrand.sampleset_bit_generatorrayleighshuffleget_statenumpy.random.mtrand.seedpoissonnoncentral_fset_statetomaxintmultinomialrandom_sampledirichletstandard_exponentialstandard_normalstandard_cauchyrandom_integerswaldvonmisesnoncentral_chisquarebetanegative_binomialparetoweibullstandard_gammastandard_tpowerlogserieszipfrandintpermutationmultivariate_normalnumpy.PyArray_MultiIterNew2numpy.PyArray_MultiIterNew3assignmentdeletionchoice__module__func_doc__doc__func_name__name____qualname__func_dict__dict__func_globals__globals__func_closure__closure__func_code__code__func_defaults__defaults____kwdefaults____annotations___is_coroutine__repr__get_bit_generator ^UL :1(f]T& l& Z |Ԏ This is an alias of `random_sample`. See `random_sample` for the complete documentation. This is an alias of `random_sample`. See `random_sample` for the complete documentation. Sets the singleton RandomState's bit generator Parameters ---------- bitgen A bit generator instance Notes ----- The singleton RandomState provides the random variate generators in the ``numpy.random``namespace. This function, and its counterpart get method, provides a path to hot-swap the default MT19937 bit generator with a user provided alternative. These function are intended to provide a continuous path where a single underlying bit generator can be used both with an instance of ``Generator`` and with the singleton instance of RandomState. See Also -------- get_bit_generator numpy.random.Generator Returns the singleton RandomState's bit generator Returns ------- BitGenerator The bit generator that underlies the singleton RandomState instance Notes ----- The singleton RandomState provides the random variate generators in the ``numpy.random`` namespace. This function, and its counterpart set method, provides a path to hot-swap the default MT19937 bit generator with a user provided alternative. These function are intended to provide a continuous path where a single underlying bit generator can be used both with an instance of ``Generator`` and with the singleton instance of RandomState. See Also -------- set_bit_generator numpy.random.Generator seed(seed=None) Reseed the singleton RandomState instance. Notes ----- This is a convenience, legacy function that exists to support older code that uses the singleton RandomState. Best practice is to use a dedicated ``Generator`` instance rather than the random variate generation methods exposed directly in the random module. See Also -------- numpy.random.Generator 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. .. note:: New code should use the `~numpy.random.Generator.permutation` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.permutation: which should be used for new code. Examples -------- >>> np.random.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) # random >>> np.random.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) # random >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.permutation(arr) array([[6, 7, 8], # random [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. .. note:: New code should use the `~numpy.random.Generator.shuffle` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- x : ndarray or MutableSequence The array, list or mutable sequence to be shuffled. Returns ------- None See Also -------- random.Generator.shuffle: which should be used for new code. Examples -------- >>> arr = np.arange(10) >>> np.random.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] # random 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], # random [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. The Dirichlet distribution is a conjugate prior of a multinomial distribution in Bayesian inference. .. note:: New code should use the `~numpy.random.Generator.dirichlet` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- alpha : sequence of floats, length k Parameter of the distribution (length ``k`` for sample of length ``k``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n)``, then ``m * n * k`` samples are drawn. Default is None, in which case a vector of length ``k`` is returned. Returns ------- samples : ndarray, The drawn samples, of shape ``(size, k)``. Raises ------ ValueError If any value in ``alpha`` is less than or equal to zero See Also -------- random.Generator.dirichlet: which should be used for new code. Notes ----- The Dirichlet distribution is a distribution over vectors :math:`x` that fulfil the conditions :math:`x_i>0` and :math:`\sum_{i=1}^k x_i = 1`. The probability density function :math:`p` of a Dirichlet-distributed random vector :math:`X` is proportional to .. math:: p(x) \propto \prod_{i=1}^{k}{x^{\alpha_i-1}_i}, where :math:`\alpha` is a vector containing the positive concentration parameters. The method uses the following property for computation: let :math:`Y` be a random vector which has components that follow a standard gamma distribution, then :math:`X = \frac{1}{\sum_{i=1}^k{Y_i}} Y` is Dirichlet-distributed References ---------- .. [1] David McKay, "Information Theory, Inference and Learning Algorithms," chapter 23, https://www.inference.org.uk/mackay/itila/ .. [2] Wikipedia, "Dirichlet distribution", https://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() >>> import matplotlib.pyplot as plt >>> 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 generalization 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``. .. note:: New code should use the `~numpy.random.Generator.multinomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. .. warning:: This function defaults to the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. Parameters ---------- n : int Number of experiments. pvals : sequence of floats, length p Probabilities of each of the ``p`` different outcomes. These must 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. See Also -------- random.Generator.multinomial: which should be used for new code. Examples -------- Throw a dice 20 times: >>> np.random.multinomial(20, [1/6.]*6, size=1) array([[4, 1, 7, 5, 2, 1]]) # random 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], # random [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]) # random 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]) # random not like: >>> np.random.multinomial(100, [1.0, 2.0]) # WRONG Traceback (most recent call last): ValueError: pvals < 0, pvals > 1 or pvals contains NaNs multivariate_normal(mean, cov, size=None, check_valid='warn', tol=1e-8) 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. .. note:: New code should use the `~numpy.random.Generator.multivariate_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. cov is cast to double before the check. 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. See Also -------- random.Generator.multivariate_normal: which should be used for new code. 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) Here we generate 800 samples from the bivariate normal distribution with mean [0, 0] and covariance matrix [[6, -3], [-3, 3.5]]. The expected variances of the first and second components of the sample are 6 and 3.5, respectively, and the expected correlation coefficient is -3/sqrt(6*3.5) ≈ -0.65465. >>> cov = np.array([[6, -3], [-3, 3.5]]) >>> pts = np.random.multivariate_normal([0, 0], cov, size=800) Check that the mean, covariance, and correlation coefficient of the sample are close to the expected values: >>> pts.mean(axis=0) array([ 0.0326911 , -0.01280782]) # may vary >>> np.cov(pts.T) array([[ 5.96202397, -2.85602287], [-2.85602287, 3.47613949]]) # may vary >>> np.corrcoef(pts.T)[0, 1] -0.6273591314603949 # may vary We can visualize this data with a scatter plot. The orientation of the point cloud illustrates the negative correlation of the components of this sample. >>> import matplotlib.pyplot as plt >>> plt.plot(pts[:, 0], pts[:, 1], '.', alpha=0.5) >>> plt.axis('equal') >>> plt.grid() >>> plt.show() 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. .. note:: New code should use the `~numpy.random.Generator.logseries` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. random.Generator.logseries: which should be used for new code. 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", https://en.wikipedia.org/wiki/Logarithmic_distribution Examples -------- Draw samples from the distribution: >>> a = .6 >>> s = np.random.logseries(a, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s) # plot against distribution >>> def logseries(k, p): ... return -p**k/(k*np.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``). .. note:: New code should use the `~numpy.random.Generator.hypergeometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Each sample is the number of good items within a randomly selected subset of size `nsample` taken from a set of `ngood` good items and `nbad` bad items. See Also -------- scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. random.Generator.hypergeometric: which should be used for new code. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{g}{x}\binom{b}{n-x}}{\binom{g+b}{n}}, where :math:`0 \le x \le n` and :math:`n-b \le x \le g` for P(x) the probability of ``x`` good results in the drawn sample, g = `ngood`, b = `nbad`, and n = `nsample`. Consider an urn with black and white marbles in it, `ngood` of them are 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. https://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", https://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) >>> from matplotlib.pyplot import hist >>> 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. .. note:: New code should use the `~numpy.random.Generator.geometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.geometric: which should be used for new code. 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 discrete probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. .. note:: New code should use the `~numpy.random.Generator.zipf` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Distribution parameter. Must 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. random.Generator.zipf: which should be used for new code. Notes ----- The probability density for the Zipf distribution is .. math:: p(k) = \frac{k^{-a}}{\zeta(a)}, for integers :math:`k \geq 1`, 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 = 4.0 >>> n = 20000 >>> s = np.random.zipf(a, n) Display the histogram of the samples, along with the expected histogram based on the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import zeta # doctest: +SKIP `bincount` provides a fast histogram for small integers. >>> count = np.bincount(s) >>> k = np.arange(1, s.max() + 1) >>> plt.bar(k, count[1:], alpha=0.5, label='sample count') >>> plt.plot(k, n*(k**-a)/zeta(a), 'k.-', alpha=0.5, ... label='expected count') # doctest: +SKIP >>> plt.semilogy() >>> plt.grid(alpha=0.4) >>> plt.legend() >>> plt.title(f'Zipf sample, a={a}, size={n}') >>> 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. .. note:: New code should use the `~numpy.random.Generator.poisson` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- lam : float or array_like of floats Expected number of events occurring in a fixed-time interval, must be >= 0. A sequence 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. See Also -------- random.Generator.poisson: which should be used for new code. 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 int64 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. https://mathworld.wolfram.com/PoissonDistribution.html .. [2] Wikipedia, "Poisson distribution", https://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, density=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` successes and `p` probability of success where `n` is > 0 and `p` is in the interval [0, 1]. .. note:: New code should use the `~numpy.random.Generator.negative_binomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : float or array_like of floats Parameter of the distribution, > 0. 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 failures that occurred before a total of n successes was reached. .. warning:: This function returns the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. See Also -------- random.Generator.negative_binomial: which should be used for new code. Notes ----- The probability mass function of the negative binomial distribution is .. math:: P(N;n,p) = \frac{\Gamma(N+n)}{N!\Gamma(n)}p^{n}(1-p)^{N}, where :math:`n` is the number of successes, :math:`p` is the probability of success, :math:`N+n` is the number of trials, and :math:`\Gamma` is the gamma function. When :math:`n` is an integer, :math:`\frac{\Gamma(N+n)}{N!\Gamma(n)} = \binom{N+n-1}{N}`, which is the more common form of this term in the pmf. The negative binomial distribution gives the probability of N failures given n successes, with a success on the last 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. https://mathworld.wolfram.com/NegativeBinomialDistribution.html .. [2] Wikipedia, "Negative binomial distribution", https://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): # doctest: +SKIP ... 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. .. note:: New code should use the `~numpy.random.Generator.triangular` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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 must fulfill the condition ``left <= mode <= right``. right : float or array_like of floats Upper limit, must 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. See Also -------- random.Generator.triangular: which should be used for new code. 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" https://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, ... density=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. .. note:: New code should use the `~numpy.random.Generator.wald` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : float or array_like of floats Distribution mean, must be > 0. scale : float or array_like of floats Scale parameter, must 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. See Also -------- random.Generator.wald: which should be used for new code. 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, https://web.archive.org/web/20090423014010/http://www.brighton-webs.co.uk:80/distributions/wald.asp .. [2] Chhikara, Raj S., and Folks, J. Leroy, "The Inverse Gaussian Distribution: Theory : Methodology, and Applications", CRC Press, 1988. .. [3] Wikipedia, "Inverse Gaussian distribution" https://en.wikipedia.org/wiki/Inverse_Gaussian_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, density=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. .. note:: New code should use the `~numpy.random.Generator.rayleigh` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. Must be non-negative. 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. See Also -------- random.Generator.rayleigh: which should be used for new code. 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," https://web.archive.org/web/20090514091424/http://brighton-webs.co.uk:80/distributions/rayleigh.asp .. [2] Wikipedia, "Rayleigh distribution" https://en.wikipedia.org/wiki/Rayleigh_distribution Examples -------- Draw values from the distribution and plot the histogram >>> from matplotlib.pyplot import hist >>> values = hist(np.random.rayleigh(3, 100000), bins=200, density=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 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. .. note:: New code should use the `~numpy.random.Generator.lognormal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must be non-negative. 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. random.Generator.lognormal: which should be used for new code. 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. https://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, density=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.standard_normal(100) ... b.append(np.prod(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, density=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). .. note:: New code should use the `~numpy.random.Generator.logistic` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must be non-negative. 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. random.Generator.logistic: which should be used for new code. 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. https://mathworld.wolfram.com/LogisticDistribution.html .. [3] Wikipedia, "Logistic-distribution", https://en.wikipedia.org/wiki/Logistic_distribution Examples -------- Draw samples from the distribution: >>> loc, scale = 10, 1 >>> s = np.random.logistic(loc, scale, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=50) # plot against distribution >>> def logist(x, loc, scale): ... return np.exp((loc-x)/scale)/(scale*(1+np.exp((loc-x)/scale))**2) >>> lgst_val = logist(bins, loc, scale) >>> plt.plot(bins, lgst_val * count.max() / lgst_val.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. .. note:: New code should use the `~numpy.random.Generator.gumbel` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must 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 ``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 random.Generator.gumbel: which should be used for new code. 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, density=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, density=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. .. note:: New code should use the `~numpy.random.Generator.laplace` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must 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 ``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. See Also -------- random.Generator.laplace: which should be used for new code. 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. https://mathworld.wolfram.com/LaplaceDistribution.html .. [4] Wikipedia, "Laplace distribution", https://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, density=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. .. note:: New code should use the `~numpy.random.Generator.power` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Parameter of the distribution. Must 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 ``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 <= 0. See Also -------- random.Generator.power: which should be used for new code. 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. https://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 # doctest: +SKIP >>> rvs = np.random.power(5, 1000000) >>> rvsp = np.random.pareto(5, 1000000) >>> xx = np.linspace(0,1,100) >>> powpdf = stats.powerlaw.pdf(xx,5) # doctest: +SKIP >>> plt.figure() >>> plt.hist(rvs, bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('np.random.power(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('inverse of 1 + np.random.pareto(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> 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}`. .. note:: New code should use the `~numpy.random.Generator.weibull` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape parameter of the distribution. Must be nonnegative. 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 random.Generator.weibull: which should be used for new code. 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", https://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. .. note:: New code should use the `~numpy.random.Generator.pareto` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape of the distribution. Must be positive. 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. random.Generator.pareto: which should be used for new code. 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", https://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, density=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. .. note:: New code should use the `~numpy.random.Generator.vonmises` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. random.Generator.vonmises: which should be used for new code. 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 # doctest: +SKIP >>> plt.hist(s, 50, density=True) >>> x = np.linspace(-np.pi, np.pi, num=51) >>> y = np.exp(kappa*np.cos(x-mu))/(2*np.pi*i0(kappa)) # doctest: +SKIP >>> plt.plot(x, y, linewidth=2, color='r') # doctest: +SKIP >>> 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`). .. note:: New code should use the `~numpy.random.Generator.standard_t` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must 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. See Also -------- random.Generator.standard_t: which should be used for new code. 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" https://en.wikipedia.org/wiki/Student's_t-distribution Examples -------- From Dalgaard page 83 [1]_, suppose the daily energy intake for 11 women in kilojoules (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? Our null hypothesis will be the absence of deviation, and the alternate hypothesis will be the presence of an effect that could be either positive or negative, hence making our test 2-tailed. Because we are estimating the mean and we have N=11 values in our sample, we have N-1=10 degrees of freedom. We set our significance level to 95% and compute the t statistic using the empirical mean and empirical standard deviation of our intake. We use a ddof of 1 to base the computation of our empirical standard deviation on an unbiased estimate of the variance (note: the final estimate is not unbiased due to the concave nature of the square root). >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> t -2.8207540608310198 We draw 1000000 samples from Student's t distribution with the adequate degrees of freedom. >>> import matplotlib.pyplot as plt >>> s = np.random.standard_t(10, size=1000000) >>> h = plt.hist(s, bins=100, density=True) Does our t statistic land in one of the two critical regions found at both tails of the distribution? >>> np.sum(np.abs(t) < np.abs(s)) / float(len(s)) 0.018318 #random < 0.05, statistic is in critical region The probability value for this 2-tailed test is about 1.83%, which is lower than the 5% pre-determined significance threshold. Therefore, the probability of observing values as extreme as our intake conditionally on the null hypothesis being true is too low, and we reject the null hypothesis of no deviation. standard_cauchy(size=None) Draw samples from a standard Cauchy distribution with mode = 0. Also known as the Lorentz distribution. .. note:: New code should use the `~numpy.random.Generator.standard_cauchy` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.standard_cauchy: which should be used for new code. 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", https://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm .. [2] Weisstein, Eric W. "Cauchy Distribution." From MathWorld--A Wolfram Web Resource. https://mathworld.wolfram.com/CauchyDistribution.html .. [3] Wikipedia, "Cauchy distribution" https://en.wikipedia.org/wiki/Cauchy_distribution Examples -------- Draw samples and plot the distribution: >>> import matplotlib.pyplot as plt >>> 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 generalization of the :math:`\chi^2` distribution. .. note:: New code should use the `~numpy.random.Generator.noncentral_chisquare` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must be > 0. .. versionchanged:: 1.10.0 Earlier NumPy versions required dfnum > 1. nonc : float or array_like of floats Non-centrality, must 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. See Also -------- random.Generator.noncentral_chisquare: which should be used for new code. 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. References ---------- .. [1] Wikipedia, "Noncentral chi-squared distribution" https://en.wikipedia.org/wiki/Noncentral_chi-squared_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, density=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), density=True) >>> values2 = plt.hist(np.random.chisquare(3, 100000), ... bins=np.arange(0., 25, .1), density=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, density=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. .. note:: New code should use the `~numpy.random.Generator.chisquare` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Number of degrees of freedom, must 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 chi-square distribution. Raises ------ ValueError When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``) is given. See Also -------- random.Generator.chisquare: which should be used for new code. 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" https://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]) # random 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. .. note:: New code should use the `~numpy.random.Generator.noncentral_f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Numerator degrees of freedom, must be > 0. .. versionchanged:: 1.14.0 Earlier NumPy versions required dfnum > 1. dfden : float or array_like of floats Denominator degrees of freedom, must be > 0. nonc : float or array_like of floats Non-centrality parameter, the sum of the squares of the numerator means, must 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. See Also -------- random.Generator.noncentral_f: which should be used for new code. 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. https://mathworld.wolfram.com/NoncentralF-Distribution.html .. [2] Wikipedia, "Noncentral F-distribution", https://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, density=True) >>> c_vals = np.random.f(dfnum, dfden, 1000000) >>> F = np.histogram(c_vals, bins=50, density=True) >>> import matplotlib.pyplot as plt >>> 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 must 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. .. note:: New code should use the `~numpy.random.Generator.f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Degrees of freedom in numerator, must be > 0. dfden : float or array_like of float Degrees of freedom in denominator, must 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`` 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. random.Generator.f: which should be used for new code. 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", https://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 : >>> np.sort(s)[-10] 7.61988120985 # random 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. .. note:: New code should use the `~numpy.random.Generator.gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. Must be non-negative. scale : float or array_like of floats, optional The scale of the gamma distribution. Must be non-negative. 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. random.Generator.gamma: which should be used for new code. 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. https://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", https://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 # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1)*(np.exp(-bins/scale) / # doctest: +SKIP ... (sps.gamma(shape)*scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> 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. .. note:: New code should use the `~numpy.random.Generator.standard_gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats Parameter, must 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 ``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. random.Generator.standard_gamma: which should be used for new code. 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. https://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", https://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 # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ # doctest: +SKIP ... (sps.gamma(shape) * scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() 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]_. .. note:: New code should use the `~numpy.random.Generator.normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must 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 ``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. random.Generator.normal: which should be used for new code. 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 normal is more likely to return samples lying close to the mean, rather than those far away. References ---------- .. [1] Wikipedia, "Normal distribution", https://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)) 0.0 # may vary >>> abs(sigma - np.std(s, ddof=1)) 0.1 # may vary Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=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() Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> np.random.normal(3, 2.5, size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random standard_normal(size=None) Draw samples from a standard Normal distribution (mean=0, stdev=1). .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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 A floating-point array of shape ``size`` of drawn samples, or a single sample if ``size`` was not specified. See Also -------- normal : Equivalent function with additional ``loc`` and ``scale`` arguments for setting the mean and standard deviation. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use one of:: mu + sigma * np.random.standard_normal(size=...) np.random.normal(mu, sigma, size=...) Examples -------- >>> np.random.standard_normal() 2.1923875335537315 #random >>> 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) Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.standard_normal(size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random random_integers(low, high=None, size=None) Random integers of type `numpy.int_` between `low` and `high`, inclusive. Return random integers of type `numpy.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 `numpy.int_` type translates to the C long integer type 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 -------- 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 # random >>> type(np.random.random_integers(5)) >>> np.random.random_integers(5, size=(3,2)) array([[5, 4], # random [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 ]) # random 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, density=True) >>> plt.show() randn(d0, d1, ..., dn) Return a sample (or samples) from the "standard normal" distribution. .. note:: This is a convenience function for users porting code from Matlab, and wraps `standard_normal`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. If positive int_like 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. A single float randomly sampled from the distribution is returned if no argument is provided. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, must be non-negative. 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 -------- standard_normal : Similar, but takes a tuple as its argument. normal : Also accepts mu and sigma arguments. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use:: sigma * np.random.randn(...) + mu Examples -------- >>> np.random.randn() 2.1923875335537315 # random Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.randn(2, 4) 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. .. note:: This is a convenience function for users porting code from Matlab, and wraps `random_sample`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. 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, must be non-negative. If no argument is given a single Python float is returned. Returns ------- out : ndarray, shape ``(d0, d1, ..., dn)`` Random values. See Also -------- random 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`. .. note:: New code should use the `~numpy.random.Generator.uniform` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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 or equal to high. The high limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. 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)``. random.Generator.uniform: which should be used for new code. 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. The ``high`` limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. For example: >>> x = np.float32(5*0.99999999) >>> x 5.0 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, density=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 .. note:: New code should use the `~numpy.random.Generator.choice` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. .. warning:: This function uses the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. 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 it 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. Default is True, meaning that a value of ``a`` can be selected multiple times. 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 random.Generator.choice: which should be used in new code Notes ----- Setting user-specified probabilities through ``p`` uses a more general but less efficient sampler than the default. The general sampler produces a different sample than the optimized sampler even if each element of ``p`` is 1 / len(a). Sampling random rows from a 2-D array is not possible with this function, but is possible with `Generator.choice` through its ``axis`` keyword. Examples -------- Generate a uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3) array([0, 3, 4]) # random >>> #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]) # random 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]) # random >>> #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]) # random 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'], # random dtype='>> np.random.bytes(10) b' eh\x85\x022SZ\xbf\xa4' #random randint(low, high=None, size=None, dtype=int) 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`). .. note:: New code should use the `~numpy.random.Generator.integers` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- low : int or array-like of ints Lowest (signed) integers to be drawn from the distribution (unless ``high=None``, in which case this parameter is one above the *highest* such integer). high : int or array-like of ints, optional If provided, one above the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). If array-like, must contain integer values 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. Byteorder must be native. The default value is long. .. versionadded:: 1.11.0 .. warning:: This function defaults to the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. Which corresponds to `np.intp`. (`dtype=int` is not the same as in most NumPy functions.) 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_integers : similar to `randint`, only for the closed interval [`low`, `high`], and 1 is the lowest value if `high` is omitted. random.Generator.integers: which should be used for new code. Examples -------- >>> np.random.randint(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) # random >>> 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], # random [3, 2, 2, 0]]) Generate a 1 x 3 array with 3 different upper bounds >>> np.random.randint(1, [3, 5, 10]) array([2, 2, 9]) # random Generate a 1 by 3 array with 3 different lower bounds >>> np.random.randint([1, 5, 7], 10) array([9, 8, 7]) # random Generate a 2 by 4 array using broadcasting with dtype of uint8 >>> np.random.randint([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7], # random [ 1, 16, 9, 12]], dtype=uint8) tomaxint(size=None) Return a sample of uniformly distributed random integers in the interval [0, ``np.iinfo("long").max``]. .. warning:: This function uses the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. 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.RandomState() # need a RandomState object >>> rs.tomaxint((2,2,2)) array([[[1170048599, 1600360186], # random [ 739731006, 1947757578]], [[1871712945, 752307660], [1601631370, 1479324245]]]) >>> rs.tomaxint((2,2,2)) < np.iinfo(np.int_).max array([[[ True, True], [ True, True]], [[ True, True], [ True, True]]]) 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. .. note:: New code should use the `~numpy.random.Generator.standard_exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.standard_exponential: which should be used for new code. 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]_. .. note:: New code should use the `~numpy.random.Generator.exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats The scale parameter, :math:`\beta = 1/\lambda`. Must 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 ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized exponential distribution. Examples -------- A real world example: Assume a company has 10000 customer support agents and the average time between customer calls is 4 minutes. >>> n = 10000 >>> time_between_calls = np.random.default_rng().exponential(scale=4, size=n) What is the probability that a customer will call in the next 4 to 5 minutes? >>> x = ((time_between_calls < 5).sum())/n >>> y = ((time_between_calls < 4).sum())/n >>> x-y 0.08 # may vary See Also -------- random.Generator.exponential: which should be used for new code. References ---------- .. [1] Peyton Z. Peebles Jr., "Probability, Random Variables and Random Signal Principles", 4th ed, 2001, p. 57. .. [2] Wikipedia, "Poisson process", https://en.wikipedia.org/wiki/Poisson_process .. [3] Wikipedia, "Exponential distribution", https://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 normalization, 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. .. note:: New code should use the `~numpy.random.Generator.beta` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Alpha, positive (>0). b : float or array_like of floats Beta, positive (>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 ``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. See Also -------- random.Generator.beta: which should be used for new code. random(size=None) Return random floats in the half-open interval [0.0, 1.0). Alias for `random_sample` to ease forward-porting to the new random API. 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 .. note:: New code should use the `~numpy.random.Generator.random` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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). See Also -------- random.Generator.random: which should be used for new code. Examples -------- >>> np.random.random_sample() 0.47108547995356098 # random >>> type(np.random.random_sample()) >>> np.random.random_sample((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) # random Three-by-two array of random numbers from [-5, 0): >>> 5 * np.random.random_sample((3, 2)) - 5 array([[-3.99149989, -0.52338984], # random [-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 bit generator used by the RandomState instance. By default, RandomState uses the "Mersenne Twister"[1]_ pseudo-random number generating algorithm. Parameters ---------- state : {tuple(str, ndarray of 624 uints, int, int, float), dict} 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``. If state is a dictionary, it is directly set using the BitGenerators `state` property. 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(legacy=True) Return a tuple representing the internal state of the generator. For more details, see `set_state`. Parameters ---------- legacy : bool, optional Flag indicating to return a legacy tuple state when the BitGenerator is MT19937, instead of a dict. Raises ValueError if the underlying bit generator is not an instance of MT19937. Returns ------- out : {tuple(str, ndarray of 624 uints, int, int, float), dict} If legacy is True, 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``. If `legacy` is False, or the BitGenerator is not MT19937, then state is returned as a dictionary. 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) Reseed a legacy MT19937 BitGenerator Notes ----- This is a convenience, legacy function. The best practice is to **not** reseed a BitGenerator, rather to recreate a new one. This method is here for legacy reasons. This example demonstrates best practice. >>> from numpy.random import MT19937 >>> from numpy.random import RandomState, SeedSequence >>> rs = RandomState(MT19937(SeedSequence(123456789))) # Later, you want to restart the stream >>> rs = RandomState(MT19937(SeedSequence(987654321))) This function is deprecated. Please call randint({low}, {high} + 1) insteadRandomState.standard_exponential (line 581)RandomState.noncentral_chisquare (line 2009)RandomState.multivariate_normal (line 4083)x must be an integer or at least 1-dimensionalprobabilities are not non-negativenumpy._core.umath failed to importnumpy._core.multiarray failed to importmean and cov must have same lengthlegacy can only be True when the underlyign bitgenerator is an instance of MT19937.get_state and legacy can only be used with the MT19937 BitGenerator. To silence this warning, set `legacy` to False.covariance is not symmetric positive-semidefinite.cov must be 2 dimensional and squarecheck_valid must equal 'warn', 'raise', or 'ignore'can only re-seed a MT19937 BitGeneratora must be 1-dimensional or an integerThis function is deprecated. Please call randint(1, {low} + 1) insteadShuffling a one dimensional array subclass containing objects gives incorrect results for most array subclasses. Please use the new random number API instead: https://numpy.org/doc/stable/reference/random/index.html The new API fixes this issue. This version will not be fixed due to stability guarantees of the API.RandomState.triangular (line 3267)RandomState.standard_t (line 2173)RandomState.standard_normal (line 1408)RandomState.standard_exponentialRandomState.standard_cauchy (line 2098)RandomState.random_sample (line 389)RandomState.random_integers (line 1312)RandomState.permutation (line 4700)RandomState.noncentral_f (line 1846)RandomState.noncentral_chisquareRandomState.negative_binomial (line 3528)RandomState.multinomial (line 4282)RandomState.exponential (line 504)Providing a dtype with a non-native byteorder is not supported. If you require platform-independent byteorder, call byteswap when required. In future version, providing byteorder will raise a ValueErrorNegative dimensions are not allowedInvalid bit generator. The bit generator must be instantized.Fewer non-zero entries in p than sizeCannot take a larger sample than population when 'replace=False' 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 discrete probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. .. note:: New code should use the `~numpy.random.Generator.zipf` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Distribution parameter. Must 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. random.Generator.zipf: which should be used for new code. Notes ----- The probability density for the Zipf distribution is .. math:: p(k) = \frac{k^{-a}}{\zeta(a)}, for integers :math:`k \geq 1`, 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 = 4.0 >>> n = 20000 >>> s = np.random.zipf(a, n) Display the histogram of the samples, along with the expected histogram based on the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import zeta # doctest: +SKIP `bincount` provides a fast histogram for small integers. >>> count = np.bincount(s) >>> k = np.arange(1, s.max() + 1) >>> plt.bar(k, count[1:], alpha=0.5, label='sample count') >>> plt.plot(k, n*(k**-a)/zeta(a), 'k.-', alpha=0.5, ... label='expected count') # doctest: +SKIP >>> plt.semilogy() >>> plt.grid(alpha=0.4) >>> plt.legend() >>> plt.title(f'Zipf sample, a={a}, size={n}') >>> 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}`. .. note:: New code should use the `~numpy.random.Generator.weibull` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape parameter of the distribution. Must be nonnegative. 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 random.Generator.weibull: which should be used for new code. 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", https://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. .. note:: New code should use the `~numpy.random.Generator.vonmises` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. random.Generator.vonmises: which should be used for new code. 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 # doctest: +SKIP >>> plt.hist(s, 50, density=True) >>> x = np.linspace(-np.pi, np.pi, num=51) >>> y = np.exp(kappa*np.cos(x-mu))/(2*np.pi*i0(kappa)) # doctest: +SKIP >>> plt.plot(x, y, linewidth=2, color='r') # doctest: +SKIP >>> 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`. .. note:: New code should use the `~numpy.random.Generator.uniform` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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 or equal to high. The high limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. 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)``. random.Generator.uniform: which should be used for new code. 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. The ``high`` limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. For example: >>> x = np.float32(5*0.99999999) >>> x 5.0 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, density=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. .. note:: New code should use the `~numpy.random.Generator.triangular` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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 must fulfill the condition ``left <= mode <= right``. right : float or array_like of floats Upper limit, must 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. See Also -------- random.Generator.triangular: which should be used for new code. 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" https://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, ... density=True) >>> plt.show() tomaxint(size=None) Return a sample of uniformly distributed random integers in the interval [0, ``np.iinfo("long").max``]. .. warning:: This function uses the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. 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.RandomState() # need a RandomState object >>> rs.tomaxint((2,2,2)) array([[[1170048599, 1600360186], # random [ 739731006, 1947757578]], [[1871712945, 752307660], [1601631370, 1479324245]]]) >>> rs.tomaxint((2,2,2)) < np.iinfo(np.int_).max array([[[ True, True], [ True, True]], [[ True, True], [ True, True]]]) sum(pvals[:-1].astype(np.float64)) > 1.0. The pvals array is cast to 64-bit floating point prior to checking the sum. Precision changes when casting may cause problems even if the sum of the original pvals is valid.state must be a dict or a tuple. 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`). .. note:: New code should use the `~numpy.random.Generator.standard_t` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must 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. See Also -------- random.Generator.standard_t: which should be used for new code. 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" https://en.wikipedia.org/wiki/Student's_t-distribution Examples -------- From Dalgaard page 83 [1]_, suppose the daily energy intake for 11 women in kilojoules (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? Our null hypothesis will be the absence of deviation, and the alternate hypothesis will be the presence of an effect that could be either positive or negative, hence making our test 2-tailed. Because we are estimating the mean and we have N=11 values in our sample, we have N-1=10 degrees of freedom. We set our significance level to 95% and compute the t statistic using the empirical mean and empirical standard deviation of our intake. We use a ddof of 1 to base the computation of our empirical standard deviation on an unbiased estimate of the variance (note: the final estimate is not unbiased due to the concave nature of the square root). >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> t -2.8207540608310198 We draw 1000000 samples from Student's t distribution with the adequate degrees of freedom. >>> import matplotlib.pyplot as plt >>> s = np.random.standard_t(10, size=1000000) >>> h = plt.hist(s, bins=100, density=True) Does our t statistic land in one of the two critical regions found at both tails of the distribution? >>> np.sum(np.abs(t) < np.abs(s)) / float(len(s)) 0.018318 #random < 0.05, statistic is in critical region The probability value for this 2-tailed test is about 1.83%, which is lower than the 5% pre-determined significance threshold. Therefore, the probability of observing values as extreme as our intake conditionally on the null hypothesis being true is too low, and we reject the null hypothesis of no deviation. set_state can only be used with legacy MT19937 state instances. rayleigh(scale=1.0, size=None) Draw samples from a Rayleigh distribution. The :math:`\chi` and Weibull distributions are generalizations of the Rayleigh. .. note:: New code should use the `~numpy.random.Generator.rayleigh` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. Must be non-negative. 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. See Also -------- random.Generator.rayleigh: which should be used for new code. 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," https://web.archive.org/web/20090514091424/http://brighton-webs.co.uk:80/distributions/rayleigh.asp .. [2] Wikipedia, "Rayleigh distribution" https://en.wikipedia.org/wiki/Rayleigh_distribution Examples -------- Draw values from the distribution and plot the histogram >>> from matplotlib.pyplot import hist >>> values = hist(np.random.rayleigh(3, 100000), bins=200, density=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 random_integers(low, high=None, size=None) Random integers of type `numpy.int_` between `low` and `high`, inclusive. Return random integers of type `numpy.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 `numpy.int_` type translates to the C long integer type 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 -------- 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 # random >>> type(np.random.random_integers(5)) >>> np.random.random_integers(5, size=(3,2)) array([[5, 4], # random [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 ]) # random 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, density=True) >>> plt.show() randint(low, high=None, size=None, dtype=int) 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`). .. note:: New code should use the `~numpy.random.Generator.integers` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- low : int or array-like of ints Lowest (signed) integers to be drawn from the distribution (unless ``high=None``, in which case this parameter is one above the *highest* such integer). high : int or array-like of ints, optional If provided, one above the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). If array-like, must contain integer values 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. Byteorder must be native. The default value is long. .. versionadded:: 1.11.0 .. warning:: This function defaults to the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. Which corresponds to `np.intp`. (`dtype=int` is not the same as in most NumPy functions.) 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_integers : similar to `randint`, only for the closed interval [`low`, `high`], and 1 is the lowest value if `high` is omitted. random.Generator.integers: which should be used for new code. Examples -------- >>> np.random.randint(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) # random >>> 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], # random [3, 2, 2, 0]]) Generate a 1 x 3 array with 3 different upper bounds >>> np.random.randint(1, [3, 5, 10]) array([2, 2, 9]) # random Generate a 1 by 3 array with 3 different lower bounds >>> np.random.randint([1, 5, 7], 10) array([9, 8, 7]) # random Generate a 2 by 4 array using broadcasting with dtype of uint8 >>> np.random.randint([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7], # random [ 1, 16, 9, 12]], dtype=uint8) 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. .. note:: New code should use the `~numpy.random.Generator.power` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Parameter of the distribution. Must 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 ``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 <= 0. See Also -------- random.Generator.power: which should be used for new code. 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. https://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 # doctest: +SKIP >>> rvs = np.random.power(5, 1000000) >>> rvsp = np.random.pareto(5, 1000000) >>> xx = np.linspace(0,1,100) >>> powpdf = stats.powerlaw.pdf(xx,5) # doctest: +SKIP >>> plt.figure() >>> plt.hist(rvs, bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('np.random.power(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('inverse of 1 + np.random.pareto(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> 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. .. note:: New code should use the `~numpy.random.Generator.pareto` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape of the distribution. Must be positive. 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. random.Generator.pareto: which should be used for new code. 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", https://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, density=True) >>> fit = a*m**a / bins**(a+1) >>> plt.plot(bins, max(count)*fit/max(fit), linewidth=2, color='r') >>> plt.show() 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]_. .. note:: New code should use the `~numpy.random.Generator.normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must 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 ``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. random.Generator.normal: which should be used for new code. 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 normal is more likely to return samples lying close to the mean, rather than those far away. References ---------- .. [1] Wikipedia, "Normal distribution", https://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)) 0.0 # may vary >>> abs(sigma - np.std(s, ddof=1)) 0.1 # may vary Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=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() Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> np.random.normal(3, 2.5, size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random 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. .. note:: New code should use the `~numpy.random.Generator.noncentral_f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Numerator degrees of freedom, must be > 0. .. versionchanged:: 1.14.0 Earlier NumPy versions required dfnum > 1. dfden : float or array_like of floats Denominator degrees of freedom, must be > 0. nonc : float or array_like of floats Non-centrality parameter, the sum of the squares of the numerator means, must 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. See Also -------- random.Generator.noncentral_f: which should be used for new code. 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. https://mathworld.wolfram.com/NoncentralF-Distribution.html .. [2] Wikipedia, "Noncentral F-distribution", https://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, density=True) >>> c_vals = np.random.f(dfnum, dfden, 1000000) >>> F = np.histogram(c_vals, bins=50, density=True) >>> import matplotlib.pyplot as plt >>> 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 generalization of the :math:`\chi^2` distribution. .. note:: New code should use the `~numpy.random.Generator.noncentral_chisquare` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must be > 0. .. versionchanged:: 1.10.0 Earlier NumPy versions required dfnum > 1. nonc : float or array_like of floats Non-centrality, must 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. See Also -------- random.Generator.noncentral_chisquare: which should be used for new code. 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. References ---------- .. [1] Wikipedia, "Noncentral chi-squared distribution" https://en.wikipedia.org/wiki/Noncentral_chi-squared_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, density=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), density=True) >>> values2 = plt.hist(np.random.chisquare(3, 100000), ... bins=np.arange(0., 25, .1), density=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, density=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` successes and `p` probability of success where `n` is > 0 and `p` is in the interval [0, 1]. .. note:: New code should use the `~numpy.random.Generator.negative_binomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : float or array_like of floats Parameter of the distribution, > 0. 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 failures that occurred before a total of n successes was reached. .. warning:: This function returns the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. See Also -------- random.Generator.negative_binomial: which should be used for new code. Notes ----- The probability mass function of the negative binomial distribution is .. math:: P(N;n,p) = \frac{\Gamma(N+n)}{N!\Gamma(n)}p^{n}(1-p)^{N}, where :math:`n` is the number of successes, :math:`p` is the probability of success, :math:`N+n` is the number of trials, and :math:`\Gamma` is the gamma function. When :math:`n` is an integer, :math:`\frac{\Gamma(N+n)}{N!\Gamma(n)} = \binom{N+n-1}{N}`, which is the more common form of this term in the pmf. The negative binomial distribution gives the probability of N failures given n successes, with a success on the last 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. https://mathworld.wolfram.com/NegativeBinomialDistribution.html .. [2] Wikipedia, "Negative binomial distribution", https://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): # doctest: +SKIP ... 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) Here we generate 800 samples from the bivariate normal distribution with mean [0, 0] and covariance matrix [[6, -3], [-3, 3.5]]. The expected variances of the first and second components of the sample are 6 and 3.5, respectively, and the expected correlation coefficient is -3/sqrt(6*3.5) ≈ -0.65465. >>> cov = np.array([[6, -3], [-3, 3.5]]) >>> pts = np.random.multivariate_normal([0, 0], cov, size=800) Check that the mean, covariance, and correlation coefficient of the sample are close to the expected values: >>> pts.mean(axis=0) array([ 0.0326911 , -0.01280782]) # may vary >>> np.cov(pts.T) array([[ 5.96202397, -2.85602287], [-2.85602287, 3.47613949]]) # may vary >>> np.corrcoef(pts.T)[0, 1] -0.6273591314603949 # may vary We can visualize this data with a scatter plot. The orientation of the point cloud illustrates the negative correlation of the components of this sample. >>> import matplotlib.pyplot as plt >>> plt.plot(pts[:, 0], pts[:, 1], '.', alpha=0.5) >>> plt.axis('equal') >>> plt.grid() >>> plt.show() multinomial(n, pvals, size=None) Draw samples from a multinomial distribution. The multinomial distribution is a multivariate generalization 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``. .. note:: New code should use the `~numpy.random.Generator.multinomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. .. warning:: This function defaults to the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. Parameters ---------- n : int Number of experiments. pvals : sequence of floats, length p Probabilities of each of the ``p`` different outcomes. These must 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. See Also -------- random.Generator.multinomial: which should be used for new code. Examples -------- Throw a dice 20 times: >>> np.random.multinomial(20, [1/6.]*6, size=1) array([[4, 1, 7, 5, 2, 1]]) # random 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], # random [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]) # random 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]) # random not like: >>> np.random.multinomial(100, [1.0, 2.0]) # WRONG Traceback (most recent call last): ValueError: pvals < 0, pvals > 1 or pvals contains NaNs 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. .. note:: New code should use the `~numpy.random.Generator.logseries` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. random.Generator.logseries: which should be used for new code. 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", https://en.wikipedia.org/wiki/Logarithmic_distribution Examples -------- Draw samples from the distribution: >>> a = .6 >>> s = np.random.logseries(a, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s) # plot against distribution >>> def logseries(k, p): ... return -p**k/(k*np.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. .. note:: New code should use the `~numpy.random.Generator.lognormal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must be non-negative. 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. random.Generator.lognormal: which should be used for new code. 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. https://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, density=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.standard_normal(100) ... b.append(np.prod(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, density=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). .. note:: New code should use the `~numpy.random.Generator.logistic` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must be non-negative. 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. random.Generator.logistic: which should be used for new code. 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. https://mathworld.wolfram.com/LogisticDistribution.html .. [3] Wikipedia, "Logistic-distribution", https://en.wikipedia.org/wiki/Logistic_distribution Examples -------- Draw samples from the distribution: >>> loc, scale = 10, 1 >>> s = np.random.logistic(loc, scale, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=50) # plot against distribution >>> def logist(x, loc, scale): ... return np.exp((loc-x)/scale)/(scale*(1+np.exp((loc-x)/scale))**2) >>> lgst_val = logist(bins, loc, scale) >>> plt.plot(bins, lgst_val * count.max() / lgst_val.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``). .. note:: New code should use the `~numpy.random.Generator.hypergeometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Each sample is the number of good items within a randomly selected subset of size `nsample` taken from a set of `ngood` good items and `nbad` bad items. See Also -------- scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. random.Generator.hypergeometric: which should be used for new code. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{g}{x}\binom{b}{n-x}}{\binom{g+b}{n}}, where :math:`0 \le x \le n` and :math:`n-b \le x \le g` for P(x) the probability of ``x`` good results in the drawn sample, g = `ngood`, b = `nbad`, and n = `nsample`. Consider an urn with black and white marbles in it, `ngood` of them are 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. https://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", https://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) >>> from matplotlib.pyplot import hist >>> 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. .. note:: New code should use the `~numpy.random.Generator.gumbel` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must 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 ``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 random.Generator.gumbel: which should be used for new code. 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, density=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, density=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. .. note:: New code should use the `~numpy.random.Generator.geometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.geometric: which should be used for new code. 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. .. note:: New code should use the `~numpy.random.Generator.gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. Must be non-negative. scale : float or array_like of floats, optional The scale of the gamma distribution. Must be non-negative. 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. random.Generator.gamma: which should be used for new code. 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. https://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", https://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 # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1)*(np.exp(-bins/scale) / # doctest: +SKIP ... (sps.gamma(shape)*scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> 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 must 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. .. note:: New code should use the `~numpy.random.Generator.f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Degrees of freedom in numerator, must be > 0. dfden : float or array_like of float Degrees of freedom in denominator, must 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`` 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. random.Generator.f: which should be used for new code. 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", https://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 : >>> np.sort(s)[-10] 7.61988120985 # random 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. 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]_. .. note:: New code should use the `~numpy.random.Generator.exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats The scale parameter, :math:`\beta = 1/\lambda`. Must 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 ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized exponential distribution. Examples -------- A real world example: Assume a company has 10000 customer support agents and the average time between customer calls is 4 minutes. >>> n = 10000 >>> time_between_calls = np.random.default_rng().exponential(scale=4, size=n) What is the probability that a customer will call in the next 4 to 5 minutes? >>> x = ((time_between_calls < 5).sum())/n >>> y = ((time_between_calls < 4).sum())/n >>> x-y 0.08 # may vary See Also -------- random.Generator.exponential: which should be used for new code. References ---------- .. [1] Peyton Z. Peebles Jr., "Probability, Random Variables and Random Signal Principles", 4th ed, 2001, p. 57. .. [2] Wikipedia, "Poisson process", https://en.wikipedia.org/wiki/Poisson_process .. [3] Wikipedia, "Exponential distribution", https://en.wikipedia.org/wiki/Exponential_distribution choice(a, size=None, replace=True, p=None) Generates a random sample from a given 1-D array .. versionadded:: 1.7.0 .. note:: New code should use the `~numpy.random.Generator.choice` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. .. warning:: This function uses the C-long dtype, which is 32bit on windows and otherwise 64bit on 64bit platforms (and 32bit on 32bit ones). Since NumPy 2.0, NumPy's default integer is 32bit on 32bit platforms and 64bit on 64bit platforms. 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 it 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. Default is True, meaning that a value of ``a`` can be selected multiple times. 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 random.Generator.choice: which should be used in new code Notes ----- Setting user-specified probabilities through ``p`` uses a more general but less efficient sampler than the default. The general sampler produces a different sample than the optimized sampler even if each element of ``p`` is 1 / len(a). Sampling random rows from a 2-D array is not possible with this function, but is possible with `Generator.choice` through its ``axis`` keyword. Examples -------- Generate a uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3) array([0, 3, 4]) # random >>> #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]) # random 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]) # random >>> #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]) # random 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'], # random dtype=' 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 chi-square distribution. Raises ------ ValueError When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``) is given. See Also -------- random.Generator.chisquare: which should be used for new code. 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" https://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]) # random bytes(length) Return random bytes. .. note:: New code should use the `~numpy.random.Generator.bytes` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- length : int Number of random bytes. Returns ------- out : bytes String of length `length`. See Also -------- random.Generator.bytes: which should be used for new code. Examples -------- >>> np.random.bytes(10) b' 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) .. note:: New code should use the `~numpy.random.Generator.binomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. random.Generator.binomial: which should be used for new code. 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. https://mathworld.wolfram.com/BinomialDistribution.html .. [5] Wikipedia, "Binomial distribution", https://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%. a must be greater than 0 unless no samples are taken'a' cannot be empty unless no samples are takenUnsupported dtype %r for randintRandomState.standard_gamma (line 1586)RandomState.multivariate_normalRandomState.logseries (line 3994)RandomState.lognormal (line 2997)RandomState.hypergeometric (line 3863)RandomState.geometric (line 3801)RandomState.dirichlet (line 4426)RandomState.chisquare (line 1933) 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. .. note:: New code should use the `~numpy.random.Generator.wald` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : float or array_like of floats Distribution mean, must be > 0. scale : float or array_like of floats Scale parameter, must 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. See Also -------- random.Generator.wald: which should be used for new code. 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, https://web.archive.org/web/20090423014010/http://www.brighton-webs.co.uk:80/distributions/wald.asp .. [2] Chhikara, Raj S., and Folks, J. Leroy, "The Inverse Gaussian Distribution: Theory : Methodology, and Applications", CRC Press, 1988. .. [3] Wikipedia, "Inverse Gaussian distribution" https://en.wikipedia.org/wiki/Inverse_Gaussian_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, density=True) >>> plt.show() standard_normal(size=None) Draw samples from a standard Normal distribution (mean=0, stdev=1). .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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 A floating-point array of shape ``size`` of drawn samples, or a single sample if ``size`` was not specified. See Also -------- normal : Equivalent function with additional ``loc`` and ``scale`` arguments for setting the mean and standard deviation. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use one of:: mu + sigma * np.random.standard_normal(size=...) np.random.normal(mu, sigma, size=...) Examples -------- >>> np.random.standard_normal() 2.1923875335537315 #random >>> 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) Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.standard_normal(size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random 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. .. note:: New code should use the `~numpy.random.Generator.standard_gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats Parameter, must 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 ``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. random.Generator.standard_gamma: which should be used for new code. 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. https://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", https://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 # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ # doctest: +SKIP ... (sps.gamma(shape) * scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> 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. .. note:: New code should use the `~numpy.random.Generator.standard_exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.standard_exponential: which should be used for new code. 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. .. note:: New code should use the `~numpy.random.Generator.standard_cauchy` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.standard_cauchy: which should be used for new code. 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", https://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm .. [2] Weisstein, Eric W. "Cauchy Distribution." From MathWorld--A Wolfram Web Resource. https://mathworld.wolfram.com/CauchyDistribution.html .. [3] Wikipedia, "Cauchy distribution" https://en.wikipedia.org/wiki/Cauchy_distribution Examples -------- Draw samples and plot the distribution: >>> import matplotlib.pyplot as plt >>> 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. .. note:: New code should use the `~numpy.random.Generator.shuffle` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- x : ndarray or MutableSequence The array, list or mutable sequence to be shuffled. Returns ------- None See Also -------- random.Generator.shuffle: which should be used for new code. Examples -------- >>> arr = np.arange(10) >>> np.random.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] # random 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], # random [6, 7, 8], [0, 1, 2]]) seed(seed=None) Reseed a legacy MT19937 BitGenerator Notes ----- This is a convenience, legacy function. The best practice is to **not** reseed a BitGenerator, rather to recreate a new one. This method is here for legacy reasons. This example demonstrates best practice. >>> from numpy.random import MT19937 >>> from numpy.random import RandomState, SeedSequence >>> rs = RandomState(MT19937(SeedSequence(123456789))) # Later, you want to restart the stream >>> rs = RandomState(MT19937(SeedSequence(987654321))) 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 .. note:: New code should use the `~numpy.random.Generator.random` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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). See Also -------- random.Generator.random: which should be used for new code. Examples -------- >>> np.random.random_sample() 0.47108547995356098 # random >>> type(np.random.random_sample()) >>> np.random.random_sample((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) # random Three-by-two array of random numbers from [-5, 0): >>> 5 * np.random.random_sample((3, 2)) - 5 array([[-3.99149989, -0.52338984], # random [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) randn(d0, d1, ..., dn) Return a sample (or samples) from the "standard normal" distribution. .. note:: This is a convenience function for users porting code from Matlab, and wraps `standard_normal`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. If positive int_like 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. A single float randomly sampled from the distribution is returned if no argument is provided. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, must be non-negative. 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 -------- standard_normal : Similar, but takes a tuple as its argument. normal : Also accepts mu and sigma arguments. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use:: sigma * np.random.randn(...) + mu Examples -------- >>> np.random.randn() 2.1923875335537315 # random Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.randn(2, 4) 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. .. note:: This is a convenience function for users porting code from Matlab, and wraps `random_sample`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. 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, must be non-negative. If no argument is given a single Python float is returned. Returns ------- out : ndarray, shape ``(d0, d1, ..., dn)`` Random values. See Also -------- random 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. .. note:: New code should use the `~numpy.random.Generator.poisson` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- lam : float or array_like of floats Expected number of events occurring in a fixed-time interval, must be >= 0. A sequence 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. See Also -------- random.Generator.poisson: which should be used for new code. 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 int64 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. https://mathworld.wolfram.com/PoissonDistribution.html .. [2] Wikipedia, "Poisson distribution", https://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, density=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. .. note:: New code should use the `~numpy.random.Generator.permutation` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. See Also -------- random.Generator.permutation: which should be used for new code. Examples -------- >>> np.random.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) # random >>> np.random.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) # random >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.permutation(arr) array([[6, 7, 8], # random [0, 1, 2], [3, 4, 5]]) ' object which is not a subclass of 'Sequence'; `shuffle` is not guaranteed to behave correctly. E.g., non-numpy array/tensor objects with view semantics may contain duplicates after shuffling. 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. .. note:: New code should use the `~numpy.random.Generator.laplace` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. 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. Must 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 ``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. See Also -------- random.Generator.laplace: which should be used for new code. 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. https://mathworld.wolfram.com/LaplaceDistribution.html .. [4] Wikipedia, "Laplace distribution", https://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, density=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. The Dirichlet distribution is a conjugate prior of a multinomial distribution in Bayesian inference. .. note:: New code should use the `~numpy.random.Generator.dirichlet` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- alpha : sequence of floats, length k Parameter of the distribution (length ``k`` for sample of length ``k``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n)``, then ``m * n * k`` samples are drawn. Default is None, in which case a vector of length ``k`` is returned. Returns ------- samples : ndarray, The drawn samples, of shape ``(size, k)``. Raises ------ ValueError If any value in ``alpha`` is less than or equal to zero See Also -------- random.Generator.dirichlet: which should be used for new code. Notes ----- The Dirichlet distribution is a distribution over vectors :math:`x` that fulfil the conditions :math:`x_i>0` and :math:`\sum_{i=1}^k x_i = 1`. The probability density function :math:`p` of a Dirichlet-distributed random vector :math:`X` is proportional to .. math:: p(x) \propto \prod_{i=1}^{k}{x^{\alpha_i-1}_i}, where :math:`\alpha` is a vector containing the positive concentration parameters. The method uses the following property for computation: let :math:`Y` be a random vector which has components that follow a standard gamma distribution, then :math:`X = \frac{1}{\sum_{i=1}^k{Y_i}} Y` is Dirichlet-distributed References ---------- .. [1] David McKay, "Information Theory, Inference and Learning Algorithms," chapter 23, https://www.inference.org.uk/mackay/itila/ .. [2] Wikipedia, "Dirichlet distribution", https://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() >>> import matplotlib.pyplot as plt >>> 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 2288)RandomState.rayleigh (line 3113)RandomState.logistic (line 2911)RandomState.binomial (line 3376)state dictionary is not valid.probabilities do not sum to 1RandomState.weibull (line 2480)RandomState.uniform (line 1073)RandomState.tomaxint (line 625)RandomState.shuffle (line 4575)RandomState.poisson (line 3622)RandomState.negative_binomialRandomState.laplace (line 2693)_RandomState__randomstate_ctorpvals must be a 1-d sequenceRandomState.randint (line 688)RandomState.pareto (line 2377)RandomState.normal (line 1477)RandomState.gumbel (line 2787)'a' and 'p' must have same sizeRandomState.standard_normalRandomState.standard_cauchyRandomState.random_integersRandomState.randn (line 1244)RandomState.power (line 2584)RandomState.gamma (line 1668)RandomState.choice (line 857)mean must be 1 dimensionalRange exceeds valid boundsRandomState.zipf (line 3705)RandomState.wald (line 3190)RandomState.standard_gammaRandomState.rand (line 1200)RandomState.hypergeometricRandomState.bytes (line 821)probabilities contain NaNRandomState.seed (line 232)RandomState.random_sampleRandomState.noncentral_f'p' must be 1-dimensionalnumpy/random/mtrand.pyxa must be 1-dimensionalRandomState.permutationRandomState.multinomialRandomState.f (line 1752)RandomState.exponentialRandomState.triangularRandomState.standard_tRandomState.__setstate__RandomState.__getstate__RandomState.set_stateRandomState.logseriesRandomState.lognormalRandomState.get_stateRandomState.geometricRandomState.dirichletRandomState.chisquarestandard_exponentialnoncentral_chisquareRandomState.vonmisesRandomState.tomaxintRandomState.rayleighRandomState.logisticRandomState.binomialRandomState.__reduce__you are shuffling a 'numpy.random.mtrandmultivariate_normalRandomState.weibullRandomState.uniformRandomState.shuffleRandomState.randintRandomState.poissonRandomState.laplacengood + nbad < nsamplecline_in_tracebackasyncio.coroutinesarray is read-onlyRandomState.randomRandomState.paretoRandomState.normalRandomState.gumbelRandomState.choiceDeprecationWarningset_bit_generatornegative_binomialget_bit_generatorRandomState.randnRandomState.powerRandomState.gammaRandomState.bytes__randomstate_ctormay_share_memoryRandomState.zipfRandomState.waldRandomState.seedRandomState.randRandomState.betauniform_samplessum(pvals[:-1]) > 1.0standard_normalstandard_cauchyrandom_integers_poisson_lam_maxcollections.abc_bit_generatorunique_indicesstandard_gamma_legacy_seedinghypergeometricRuntimeWarningrandom_samplecount_nonzero__class_getitem__bit_generatorRandomState.fOverflowErrorsearchsortedreturn_indexrandoms_datanumpy.linalgnoncentral_fnewbyteorder_is_coroutine_initializingresult_typepermutationmultinomialfinal_shapeexponentialcheck_validUserWarningRandomStateImportErrortriangularstandard_tstacklevel__pyx_vtable__mode > rightlogical_orless_equalleft == rightissubdtypeflat_foundempty_likealpha_dataValueErrorIndexErrorwriteablesingletonset_statelogserieslognormalleft > modeisenabledis_scalarhas_gaussget_stategeometricdirichletchisquarealpha_arrTypeErrorMT19937warningsvonmisesval_datatomaxintsubtract__setstate__reversed__reduce__rayleighpop_sizeoperatoronsamplen_uint32logisticlnsampleitemsizeisscalarisnativeisfinite__getstate___endpointbinomialallcloseSequenceweibullval_arruniformtotsizetobytesstridesshufflereshapereplacerandomsrandintpoissonnsample_mt19937laplacegreaterfloat64_dtypedisablecastingcapsulebuf_ptrbg_type at 0x{:X}asarrayalpha <= 0_MT19937unsafeuniqueuint64uint32uint16stridesamplereducerandom_rand_pickleparetoorightongoodoffsetobject_normaln_uniqmultin_maskedlngoodlengthlegacykwargsinvacc__import__ignore_highgumbelfrightformatenabledoublecumsumchoicebitgenastypearangezerosx_ptrvalueuint8statesigmashapescalerightravelrangerandnraisepvalspowerp_sump_arronbadomodeoleftnumpyniterngoodn_arrmnarr_lowlnbadkappaisnanint64int32int16indexgaussgammafoundfmodefleftflagsfinfoequal__enter__emptydtypediricdfnumdfden__class__bytesarrayalpha__all__ahighzipfwarnwaldtype__test__temptakesqrt__spec__sortsizesideselfseedrtolranfrandprodparrnoncndimnbad__name__modemnixmean__main__longlocklessleftitemintpint8high__exit__copyboolbetaatolargsalow?tolsvdsum__str__retrespsdpospixoutnewmsglowloclamkeyidxgetepsdotcovcntcdfbufarranyalladdacc_*qh???/*p?3f?(_?xY?յS?N?J?F?DB?Qt??u+ ?A?Į?"?ʝ?G??i>l>7>>*J>>>^>>F>>7P>>K{>>>u>;->>>|>eO>4(>8>L>N>ȿ>>>>·>ε>߳>>>0>U>~>|>ަ>Y>IP>w>ҟ>B>e>)>~>_>>C>{>>J>﷍>X(>'>N>Í>x >b>x>>!>^}>;z>Хw>@t>wr>byl>i>g>Sd>3a>^>]\>&Y>z)W>T>P R>O>L>5~J>3H>E>nC>@>VK>>;>9>07>4>2>>0>p->+>s)>7'>%>"> >s>L>*>= >T>T>4>>y >ϣ > >>L>>l>=+==0 =C==8==hp==0== =n==|===;=ں=Z=o,=ް=ߗ=.W===%==r=W= C=4=M,=4*=D.=y8=H=~=x=3E3lM3FT3/[3b3i34p3fw3&~3[3B3ψ3g37!3>3T3d3n3r3Fq3j3_31P3r<3$3k 33ȸ3q3|{3P3#3C333dY3"3+3®3r35333x37333p3-33731b33l44(44h4C44 4` 4M 47 44?4nB44L4 i4a4T044542`44p. 4!4"4i$4%4@'4t(4>*4+4,4j.4/4'P1424):4454&)7484c:4;4$=4+>4@4A4KC4vD4B(F4G4:I4J4rTL4M4GuO4Q4R44T4U4EiW4Y4 Z4G\4]4_4:a4b4d4Bf4\g4ji4bk4m4n4p4\r4}"t4Yu4Hw4[y4X{46.}4 4q4a4]S4F4N<434,4+(4{%4$4o&4,*4'04m84 C4P4_4q474{4w4>ԕ44s4<4d444$4 (4a44lߢ4$4l44x 4_444{4 4EP4±4{:4귴4);4nķ4S444<*4տ44A4.44ע44f4RW4R4*Y4Fk44δ444444g44k4<444y44u4_45555@5ó 5 5]5^555q5v 5!5%5V*5s/5;S55:<5D5NO5^5Nv5QHqoMֻanjDotTrotou$w'xx,jyy7\z׻z{W{S{{.|3|]|ȃ|||I||}C0}F}Z}m}}S}(}}-}}"}}|}M}~i ~~~B(~o0~C8~?~F~M~T~Z~a~f~l~r~]w~v|~`~ ~~$~m~~~w~:~ަ~f~ѭ~#~Z~y~~q~K~~~^~~a~~~`~~~~~~~~f~*~~~-~~J~~=~~~\~~~$~U~}~~~~~pH`  i   6  H  A!B+m 5XttW3 `wK\ L   s   G {V~~~d~~x~~K~~~~~~~~~)~~~a~~~{~;~~A~~~m~~z~~~"~k~]~~~ԃ~|~s~j~Ua~W~K~?~2~$~~~ }}} }}i}A}}|Q|D|{3N{zeyww7ms?7E?P?'{{?*!?bv?mU?9U1T?/v?x]?&1$-?~ n?cK[!?I?\Omg?f?uLi=?sڂl?x?Qf?ij?%ᨯC?+?Dܻ?z?cE#;?^E#?$O?2m?P"K?>?{s?%;?omo?3;?J9?++?*T[?};1s?HeC?$`?vE!=?ſ-r?MBц?K=?Q}6Ei?7u? !?z}k? ~?@?`x?*?8? Qi?oTC?_(4?ָ?@je?!u v?7Zi?{ ?I?]T?9]??}?8aD:?Yζi?Ɲҷ?r^sSw?ꍰ07?d>[?%۹? Ə{?'HB>?vX#?l1&?:l?磽!O?ލ?&?ڋ?タ+ j?A1?N0Z?0H?}G?(V?5$1!?pB9 ?b"FS?)vEW(?vG}rO?~ /? {^?Z? ?ބS?i"?lR?3Sn?>N?Ґ]b?,|y2?jG>?TLҫ?~>\O??@YH?/֎@?9O"H?>?1 7?8?Ox?]4?5D9g?r|?>ܸ$8? [B/?I䠟?O?y%d?bPޱ?c?PR?j?F}?9(Q1?c?(ڦ^w?0U^Q?1j?T ξ?x.BTv?Imb.??6YJ?)ِ?\C}?%d?w?SN?эv?pa?,Q&?@oű?SuFe?PV?;?I?viׯ?4D?.g?X1Iα?Jy?!dJ?پz?j»?8G;?L|{ʎ?mwn?k9:9??Ry?A&E?U?Ŗ<?k&_?G??~#? V#?_?S?Q| z? Y&?$?htQz? 3Tݜ?pXP?N梚?H*g?gS(u?1c?w@rT?Q=I?QA?]1%? RD?lj?W'n?-BU؊?h?t4? n?boQ?qvi?_)N?]tQW}?6H#z? 67w?"Ηs?C@Wi=q?ḰXl?f?$ka?%> T+Y? O?K 2=?]d<A]X`<+M[Ij<[5q&<.8eG< h#ឪV <;LC%K<ꆭh NVeΙVn6nvK zicp%E tQ)2U1WQ9Lin?23F:L"3\LQ V f[_rWDdx h+*k2=Ko:qr Mu\x?A{FS~8;b=ZV`bBtu9=JE>XدGwdO 8cx AFẙi&zqVYםΡag6 X83:뇡koɣj_ۤ| Mg^ݧt|Ψ_ΓXp2X^ttH蟿W;ޭl~$\z[߁İPp:J+N!X ɦ֬ ᆴX7(. Ɉ?5}h.G{tr&oya=cA/˺DH0⤮<)9O@ᣩTrVj֋@?˷dsI^i@(0߾ta&⊂l1EA1T[n&mi#d)B}QJwt}B < EOvpc/F<Ң"Ae އ0~ Rfq(*QtH3D@M`P}hwx%ƿ8*JG+[EliPIw+ E>ҙ02yΩ4A (Nt.Ȱ--̕^&܌z#;ޖu~g6X .pmF 3n bH޵LaEZvpR(-x_b˿ӰdyQӶVg<7܆ut7$MH𯋉ld"rqտH)݄ /0 wپ}2}K D5z&R cM,}uc?Ѡp5.bJ3ʸT[vv+\[U@ضBi"7oLeiFγ>SR(D2Z> B0$y1gWr-ެ @樫(afoeW-|&aY +M?V#z?u?q?}n?k?Lh?e?Rc?`?Zw^?*+\?Y?RW?U?_S?XQ?߱O?M?3K?J?GH?F?jD?`C?(`A?j??>?x,>N>>q>>>j>>k>>Π>>F>>>'>\>#>u>J>*>_F>d>+>$>w>>>JK>y>|>iݿ>>I>;>ʾ>t>5<> ~>>>O>>>~3>T>ե>(>g~>ՠ>G/>>>F>J> >:n>bԓ>Q<>>x>~>>>^>Ј>D>l>1>>%>\D>@|>?y>Bv>Hs>Qp>#^m>mj>|g>md>a>^>$[> Y>=3V>[S>P>M>J>~H>UE>B>?>=>S:>7>"4>=22>T/>d,>m+*>m'>c$>N?">,>>m>t>F>>1*> > >Y>>ʗ>>I=_={==^==&=_=g=='0===P6=˙=\= s==d= =yo=/=6=.=fЍ=x=i'=܀=a1y=p=xIh=_==W=TO=G=>=N6=.=&===-H==<א<̀<<<.4V?4=3@4A4A4qB4C4D4udE4-CF4K"G4H4H41I4J4vK4\fL4HM4+N4aO4O4bP4ٽQ4R4ԊS4crT4ZU4CV4-W4ZX4Y4UY4Z4[4(\4_]4^4_4C`4va4alb40cc47[d4~Te4Of4Jg42Hh4Fi4Fj4Hk4Kl4MPm4Vn4^o48hp4sq4r4s4 t4u4v4Cw4x4 z42{40S|4u}4~44v4@ 4L4>4ق4v444lV44R4F44p4 I44"4_44Ќ4l4L4`4ԏ4坐4y4ݖ4%44r&4k44(4444.4Q4N4t44\۶4H94̻4p44~X4w4p_4~444wE`mru\zw8xky5zz/ {ԃ{{7|3}|&|H|}C}g}ۇ}}a}g}]}~~4%~5~C~Q~g^~ij~u~>~2~~r~դ~Ƭ~N~u~C~~~~k~~~~~~t~~~6 < :#%](*.-z/13579;=?EABD:FGNIJ8LMNLPQR T=UdVWXYZ[\]^~__`;abbcod.eefLggh~~7~~/~7~~ ~ ~w~G]~>~Y~,}6}b}|O|06{x?yjD?l[T?w'??o?Wp?xI?-3?x^j??N?R:e?4:>?l?*?%z?PՋt?4?e;?$"?zaWF}?Gz‘B?Oq1? OU?ߺH?7a?nV,? K?Xhw?հ<?Vp\?m?)?zP?ZcX?*;Q^?#*'g? U7?e&$ ?jJo?\Ȭ)?L&?FS?leZ&?g ?NIO??xRr!?P_hy?y6IJO?_5%?[X~?1>?bU?+À?PX?5:pɗ0?8d?;U?J?͓?)m?ېZ]G?/|!? ?iT??Wq?PF9 ?ߓ^??ۮY?3???i?Z8o? O5?ٸ?P?R9?igP?La;?L?!ވ?%o?{7=8?Ҁt?DvC?6?=p\?;So&?mj?W?j?$O?z5Ѽ?Ҏ?C|P?yh|?%H?/ZM?f!w;??>ǭ?MAz?G?y?.?P9կ?TT}?g4K?#$O? Y?BM?6C;?B"_U?~t$?œ߉?52?Ҙl'?DɤT?<(i?qE8 ? Uī?OQM?o^?Sq͒?Gط5?zx?1zd}?:R!?Wg?~& ~k?=~-2?ZҿҶ?'|j_]?it?[?8R?uqb?#h?z|J?G~`?\!>?GF?vJ?l󈬚?5hȩmE?㭍?-l ?uG?1i%?調?M?e*|?zè?^V?4<%F?B}u?c-@c?n? R=?Kr?*}T#?,"k>?R) ?K{o?vaӽ?命8? t;I_? h?3xk?3Ӻ?b3?vZ9S?LJisk?M$a.?ftW?+ ?"@|?&#?p>_?1fҲ? DE?} ?/?%,?0?5nl+,&?QG?b. ?,*(>?p_8?cU)?h*?'wާ?dИۦ?ԭ<ڥ?]']ۤ?ݣ?=|?j?.?ĥׁ?u? ̓0?"NR? y? ڥ?d֔?^8 ?0`4I?IrO*?O'?x A?B?/)?7h`|?] ٨v?p?gC_e?T?yx;I< <[,L< Ŀk<4xV<=A[<'?}y<NG<~;[xo6xu{fUY>9>{ppCBwS(:5^dܓAN}8) YfHqն&|s f2,2Ztզޗ .n ZR'ӯB)[l@u Pҍ'TȈt(5wI'L/$;nXMØT`OArW,+jtȳRfARnqӊ<KZW$eKs) 4<=>)G'QA@Y.(5bX jz>lq{2Xx{~JH҄Cc`Qz%~ )Q\HsrUb'Bkq-hnק Ψ;3Kd)P^٨Tv$Hx"$ 5..&$ŗ: ٬ @r鷯?Q?Q?9v?(\@ffffff@0C.@4@x&?@?UUUUUU?a@X@`@|@@MA>@@-DT! @h㈵>.A-DT! -DT!@C3N@Si@?>Aޓ=?3?r?q?0@; ` ] Zd  8DU<ܾڿh#cLDpdd,dddxee e@eee0f gl g h jX @jl j j j Pk k 0l0 lT pm Po o p, p `s s t`u,@vPvxxy}p}0~~`$@PD 0TЈdPЊ@(Hp\0`В$ TЕИT@t0зT`(0@Pd`0HPdP 8 # !C!\"i#`qh/t/@x/y4000112px23h33p45'`5L5Q,6r6w70}7,8@8У(9`9`::;4<@<L==Pd>> |?@@' A1ApZB0z C@|CCд`D`DDDDP0Ep|EE,FxFFpF@GPxGG`Ht-BBE H(H0K@ 0D(A BBBE m0A(A FBBHX>tpBBB B(A0A8D`> 8A0A(B BBBJ 8>vaGED D(F0v(A ABBJ<>4vyGED D(G0O(A ABBHH ?tvBBE I(H0K8K` 8E0A(B BBBE 4l?wPAG@y EJ x EC ( AE 8?yIGED D(F0b(A ABBFH?yBBE B(H0H8KP 8F0A(B BBBD 8,@X{IGED D(F0b(A ABBFh@l{7H|@{AG0 AC q AF B AE J EA J AE @ ~GJ@~GJ@~GJA$~ $A ~58AL~$D _0PAd~3AG@F AI  AL eAAp(D c AlD i K J N WAD U4AAG@ AO  EG 1A BdPDe G J F K 4BAG  AM XB|/AG ]A xB AG@F EM ,BSAK@L EC  EG BCD z(BAG V AI XA$CTiBFG0IAK8CNAG DAXC̆(D _ pCBAG@\ AK CAG0 EF HC BFB B(A0A8Gg 8D0A(B BBBG DuAG @ DL L(D< BFG E(D0A8J 8A0A(E BBBF LxDRBED A(G`e (A ABBB t (C ABBA 0Dka L nJHA G L,D8AG  AF Y EB ,E&AG@ AL PEIO0 EA ,tECAGP# AD _ AH EAG@ DD Eܢ_D ZE$D  Ẹ5D b J  I (FVBAG@CFB HF L@ M r E 0lFMGG T ABF hHFZBBB B(D0A8D@} 8D0A(B BBBE HFBBB B(D0A8DPm 8D0A(B BBBE x8G BDB B(A0A8DP 8A0A(B BBBD C 8A0A(B BBBC a 8A0A(B BBBE xG@BDB B(A0A8DP 8A0A(B BBBC B 8A0A(B BBBD ` 8A0A(B BBBF (0HdIACD g AAG H\HBBB B(D0D8D`b 8A0A(B BBBH H\BBB B(D0D8DP 8A0A(B BBBE  8A0A(B BBBC d 8A0A(B BBBB m 8A0A(B BBBA d@I{BBB B(A0A8GP 8A0A(B BBBF I 8A0A(B BBBE xI̱`BBE B(A0A8DPR 8K0A(B BBBI _ 8A0A(B BBBG L 8A0A(B BBBJ ,$JwKHE }ABHdTJBIB B(A0D8D`' 8A0A(B BBBG B8A0A(B BBB J8AG0 AA 4JAG@ EA x EC > AI K%,K/DjDKиavj \K(D i K J N xKHD CK̹g\ ~K$YD H KlqDe G J F lKȺHD | A 4LAO  EC  AT z EI @LԽAK@ AQ dL8AW0 EG L`t _L]D g U L(AK0 EH LAG C AL $MzAG s AL hEP,MhBAD D@  AAFK   AABR ; AAF MJAG  AM M;D v(MMa J n J L D AHM|BBB E(A0D8G[ 8D0A(B BBBA 4N0HN,\N(n\ LtN4N|AGP AG ^ AA L AC NAG@ DD P   ` |  o`   @K!X) o!oo o 6`F`V`f`v`````````aa&a6aFaVafavaaaaaaaaabb&b6bFbVbfbvbbbbbbbbbcc&c6cFcVcfcvcccccccccdd&d6dFdVdfdvdddddddddee&e6eFeVefeveeeeeeeeeff&f6fFfVfffvfffffffffgg&g6gFgVgfgvggggggggghh&h6hFhVhfhvhhhhhhhhhii&i6iFiVifivii0L( @ P@z *p3p;pEpN [en0{0@@PP``ήݮй@ l ```D0 ppDǨLzp& x MϬР@SO` e0@=@ x s`j@_V'FuF;~|/ !BE, : `@O1fp5m >0X@/0E`6{=jg_,p`TwI]p:T@"+ <!s`]i`ì{ݬp@@5N@Up@/@p@ݬì{`i]`s <!T@"+]p:wI,p`Tg_=j6{0E`@/0Xm >fp51O `@:E, B !~|/uF;'FV@_`j s x=@0@e `OSϬР@Mx & zpǨLGCC: (GNU) 10.2.1 20210130 (Red Hat 10.2.1-11)8`  !!@K ` ` i | 0T \          H E}     0  @  P1 pXa@  X й@ A ` P P l P p7  о  z QS `   A w P  " ); iV jr k~ l` H  ] c  pM T5 J h @}  `  h h hm: ] p~ ^ o  p  * P|T v `  pj A  `&G [  n > H    $  !  5  T  ` q  `       @   !  -  P  ` n  } "      @   } "  ;  #d  z          `} "  <  W  |    } '     &  ` !M  k  | "    | "    $" | J ,t  @* ! @- 8 %a ` |  @  `   $* ` F l `  `  `  `  @  @ ' @ A @ e   @(  % 5 !\ @ u      (D !m , @  @| '  @(0 @ O #x  `   #  @ E b !       C f } P  >  G L  ' | !O c  v 8    d  b  6  `  4    { 0 > &g { 5 0  ,    (    @      8     0 $ < g K  X d  r  0  `       ~       0  `l D Z 0 r Y          i  (     )  5 I `4r   ` 6  R  Q     "  . F  S i  u % 3 K  |  ]      ~   ' ` b N  ^ j D y Y  x    r    =    h    ]     , F T z ^ 8   W  Q    K    E  6  ?  / ! 9 / ,  W 3 e V p   % {    x  `u      ,  ( ;   !c  h u     "   p   ` o  x   S      !  ! O ! - (! h =! ' K! ! Y!  g!  t! z !  ! X ! X ! H !  !  !  !  "  " L (" u 5"  F" v P" t Z"  h"  t"  " r "  "  " " p " 8 " x #  # T:# ` Q#  `# k m# h #  #  #  #  # f # X #  # q $ ( %$ M$  _$ $ a $  $  $ X $  $ $ S $ #% <%  J% N W% I d% H w%  %  % I %  % % { % & -& p 7&  E&  V&  e& @ r& 5 & 0 & & &  & H &  & ' F #'  1' + >' ['  ' 8 ' q '  ' @{ ' C (  (   ( (I( @#r( ( ( ( ` (  ( (  )  )  $)  2)  A)  R)  c)  r)  ~) n )  ) @ )  )  ) @k ) & * * C*  R*  ^*  n* * * p *  *  * @] + #)+ K+ q+ ! ~+  +  + + 8 +  +  +  + q!,  1, P  Y,  g, ( ,  , , E  , , -  *-  ?- X-  e-  s-  - ` - : -  - P -  -  -  . x .  &.  ;. @ L.  Z.  g. l q. { .  . .  . w.  . .  / : @8/ 0 I/  W/  g/ /  /  /  /  /  /  /  / @ / ( 0 %0 B L0 i0 @ 60 P 0 0 0 z 1  1 ) H1  V1 |1 ( !1  1 t 1  1  1  1 2 ' ,2  82 = C2  P2  ]2  j2  z2  2  2 ! 2 x 2  2  3  3 j !3 9 ,3 m ;3 f J3 _ Y3  g3 p w3 3 3 X 3 @ 3 Q 3 h 3 h 4  4 z %4  64  ^4  k4 ~ k 4  4  4 ` 4  4  4 f 5 /-5 t ;5 W5 n e5  r5  5 c5 5 c5  6 %6 L6 [6 `6 z6 :6 .6 L6 P7 7 >7 O7 [n7 7 7 70 7 7 pg38 ~o8 `m8 #8 9 99 $T9 9p 9 89x : `9 pI: :`: "x: ": %!: : 'H ; 1; @/Hi; 6H; =H; 0EH< L-O< Rd< @Ut< U< < < ]U=X *= A= ]= t= = = = = > wZ>` q>X >P >H >@ >8 >0  ?( &? =? X? n? ? ? y? @ #@ 9@ T@ j@ @ @ @P @ 0 A@ A8 9A qA0 A( A A A ɣA Bh B` 6BX WBH qB@ B8 B B C 3C kC C C D i F:j Ouj 0Xj ` j i k s Tk | k (k0 k( k  l $l :l Ul kl l l l l l l m /m Jm `m {m m m m m n 1n Mn dn n n n n n n o 0o Lo co o o o `Fo px *pp Aph ]p` tpX p p p p pP  qH )q@ @q8 \q0 sq( q p1)q q q r .r Erx arp xrh r r r r r =A1s Gs bs xs s sx s` sX sP  tH &t@ LctɒU!4GTfvƓؓ( 7IUm̔ܔ3F`vƕԕ"2Abp ӷ ͖ 0AM\q}З#7G\o~˘ۘ "3BSev̙ݙ" `,FTdxmtrand.pyx.c__pyx_f_5numpy_6random_6mtrand_11RandomState__reset_gauss__pyx_tp_traverse_5numpy_6random_6mtrand_RandomState__pyx_getprop_5numpy_6random_6mtrand_11RandomState__bit_generator__Pyx_CyFunction_get_qualname__Pyx_CyFunction_get_globals__Pyx_CyFunction_get_closure__Pyx_CyFunction_get_code__pyx_tp_new_5numpy_6random_6mtrand_RandomState__pyx_vtabptr_5numpy_6random_6mtrand_RandomState__pyx_mstate_global_static__Pyx_CyFunction_get_annotations__Pyx_CyFunction_get_dict__Pyx_CheckKeywordStrings__Pyx_PyObject_Call__pyx_f_5numpy_6random_6mtrand_11RandomState__shuffle_raw__Pyx_CyFunction_CallMethod__Pyx_PyMethod_New__Pyx_CyFunction_get_name__Pyx_CyFunction_repr__Pyx_PyObject_GetAttrStr__Pyx_PyNumber_IntOrLongWrongResultType__Pyx_CyFunction_get_kwdefaults__pyx_setprop_5numpy_6random_6mtrand_11RandomState__bit_generator__pyx_tp_dealloc_5numpy_6random_6mtrand_RandomState__Pyx_ErrRestoreInState__Pyx__GetException__Pyx_CyFunction_set_doc__pyx_tp_clear_5numpy_6random_6mtrand_RandomState__Pyx_Import__Pyx_CyFunction_clear__Pyx_CyFunction_dealloc__Pyx_ImportVoidPtr_3_0_11__Pyx_ImportFunction_3_0_11__Pyx_copy_spec_to_module__pyx_pymod_createmain_interpreter_id.0__pyx_m__Pyx_CyFunction_traverse__Pyx_IsSubtype__Pyx_CyFunction_Vectorcall_O__Pyx_CyFunction_get_doc__Pyx_PyUnicode_Equals__Pyx_PyDict_GetItem__Pyx_CyFunction_CallAsMethod__Pyx_CyFunction_set_annotations__Pyx_CyFunction_set_kwdefaults__Pyx_CyFunction_set_defaults__Pyx_CyFunction_set_dict__Pyx_CyFunction_set_qualname__Pyx_CyFunction_set_name__Pyx_CyFunction_New.constprop.0__Pyx_CyFunction_Vectorcall_NOARGS__Pyx_CyFunction_Vectorcall_FASTCALL_KEYWORDS_METHOD__Pyx_CyFunction_Vectorcall_FASTCALL_KEYWORDS__Pyx_ImportType_3_0_11.constprop.0__Pyx_SetItemInt_Fast.constprop.0__Pyx_PyObject_GetSlice.constprop.0__Pyx__PyObject_LookupSpecial.constprop.0__Pyx_GetItemInt_Fast.constprop.0__Pyx_ParseOptionalKeywords.constprop.0__Pyx_Raise.constprop.0__Pyx_CreateStringTabAndInitStrings__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_IndexError__pyx_k_Invalid_bit_generator_The_bit_ge__pyx_k_MT19937__pyx_k_MT19937_2__pyx_k_Negative_dimensions_are_not_allo__pyx_k_OverflowError__pyx_k_Providing_a_dtype_with_a_non_nat__pyx_k_RandomState__pyx_k_RandomState___getstate__pyx_k_RandomState___reduce__pyx_k_RandomState___setstate__pyx_k_RandomState__randomstate_ctor__pyx_k_RandomState_beta__pyx_k_RandomState_binomial__pyx_k_RandomState_binomial_line_3376__pyx_k_RandomState_bytes__pyx_k_RandomState_bytes_line_821__pyx_k_RandomState_chisquare__pyx_k_RandomState_chisquare_line_1933__pyx_k_RandomState_choice__pyx_k_RandomState_choice_line_857__pyx_k_RandomState_dirichlet__pyx_k_RandomState_dirichlet_line_4426__pyx_k_RandomState_exponential__pyx_k_RandomState_exponential_line_504__pyx_k_RandomState_f__pyx_k_RandomState_f_line_1752__pyx_k_RandomState_gamma__pyx_k_RandomState_gamma_line_1668__pyx_k_RandomState_geometric__pyx_k_RandomState_geometric_line_3801__pyx_k_RandomState_get_state__pyx_k_RandomState_gumbel__pyx_k_RandomState_gumbel_line_2787__pyx_k_RandomState_hypergeometric__pyx_k_RandomState_hypergeometric_line__pyx_k_RandomState_laplace__pyx_k_RandomState_laplace_line_2693__pyx_k_RandomState_logistic__pyx_k_RandomState_logistic_line_2911__pyx_k_RandomState_lognormal__pyx_k_RandomState_lognormal_line_2997__pyx_k_RandomState_logseries__pyx_k_RandomState_logseries_line_3994__pyx_k_RandomState_multinomial__pyx_k_RandomState_multinomial_line_428__pyx_k_RandomState_multivariate_normal__pyx_k_RandomState_multivariate_normal_2__pyx_k_RandomState_negative_binomial__pyx_k_RandomState_negative_binomial_li__pyx_k_RandomState_noncentral_chisquare__pyx_k_RandomState_noncentral_chisquare_2__pyx_k_RandomState_noncentral_f__pyx_k_RandomState_noncentral_f_line_18__pyx_k_RandomState_normal__pyx_k_RandomState_normal_line_1477__pyx_k_RandomState_pareto__pyx_k_RandomState_pareto_line_2377__pyx_k_RandomState_permutation__pyx_k_RandomState_permutation_line_470__pyx_k_RandomState_poisson__pyx_k_RandomState_poisson_line_3622__pyx_k_RandomState_power__pyx_k_RandomState_power_line_2584__pyx_k_RandomState_rand__pyx_k_RandomState_rand_line_1200__pyx_k_RandomState_randint__pyx_k_RandomState_randint_line_688__pyx_k_RandomState_randn__pyx_k_RandomState_randn_line_1244__pyx_k_RandomState_random__pyx_k_RandomState_random_integers__pyx_k_RandomState_random_integers_line__pyx_k_RandomState_random_sample__pyx_k_RandomState_random_sample_line_3__pyx_k_RandomState_rayleigh__pyx_k_RandomState_rayleigh_line_3113__pyx_k_RandomState_seed__pyx_k_RandomState_seed_line_232__pyx_k_RandomState_set_state__pyx_k_RandomState_shuffle__pyx_k_RandomState_shuffle_line_4575__pyx_k_RandomState_standard_cauchy__pyx_k_RandomState_standard_cauchy_line__pyx_k_RandomState_standard_exponential__pyx_k_RandomState_standard_exponential_2__pyx_k_RandomState_standard_gamma__pyx_k_RandomState_standard_gamma_line__pyx_k_RandomState_standard_normal__pyx_k_RandomState_standard_normal_line__pyx_k_RandomState_standard_t__pyx_k_RandomState_standard_t_line_2173__pyx_k_RandomState_tomaxint__pyx_k_RandomState_tomaxint_line_625__pyx_k_RandomState_triangular__pyx_k_RandomState_triangular_line_3267__pyx_k_RandomState_uniform__pyx_k_RandomState_uniform_line_1073__pyx_k_RandomState_vonmises__pyx_k_RandomState_vonmises_line_2288__pyx_k_RandomState_wald__pyx_k_RandomState_wald_line_3190__pyx_k_RandomState_weibull__pyx_k_RandomState_weibull_line_2480__pyx_k_RandomState_zipf__pyx_k_RandomState_zipf_line_3705__pyx_k_Range_exceeds_valid_bounds__pyx_k_RuntimeWarning__pyx_k_Sequence__pyx_k_Shuffling_a_one_dimensional_arra__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_r_for_randint__pyx_k_UserWarning__pyx_k_ValueError__pyx_k__16__pyx_k__163__pyx_k__4__pyx_k__5__pyx_k__54__pyx_k__6__pyx_k__85__pyx_k_a__pyx_k_a_and_p_must_have_same_size__pyx_k_a_cannot_be_empty_unless_no_sam__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_unless__pyx_k_acc__pyx_k_add__pyx_k_ahigh__pyx_k_all__pyx_k_all_2__pyx_k_allclose__pyx_k_alow__pyx_k_alpha__pyx_k_alpha_0__pyx_k_alpha_arr__pyx_k_alpha_data__pyx_k_any__pyx_k_arange__pyx_k_args__pyx_k_arr__pyx_k_array__pyx_k_array_is_read_only__pyx_k_asarray__pyx_k_astype__pyx_k_asyncio_coroutines__pyx_k_at_0x_X__pyx_k_atol__pyx_k_b__pyx_k_beta__pyx_k_bg_type__pyx_k_binomial__pyx_k_binomial_n_p_size_None_Draw_sam__pyx_k_bit_generator__pyx_k_bit_generator_2__pyx_k_bitgen__pyx_k_bool__pyx_k_buf__pyx_k_buf_ptr__pyx_k_bytes__pyx_k_bytes_length_Return_random_byte__pyx_k_can_only_re_seed_a_MT19937_BitGe__pyx_k_capsule__pyx_k_casting__pyx_k_cdf__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_class__pyx_k_class_getitem__pyx_k_cline_in_traceback__pyx_k_cnt__pyx_k_collections_abc__pyx_k_copy__pyx_k_count_nonzero__pyx_k_cov__pyx_k_cov_must_be_2_dimensional_and_sq__pyx_k_covariance_is_not_symmetric_posi__pyx_k_cumsum__pyx_k_d__pyx_k_df__pyx_k_dfden__pyx_k_dfnum__pyx_k_diric__pyx_k_dirichlet__pyx_k_dirichlet_alpha_size_None_Draw__pyx_k_disable__pyx_k_dot__pyx_k_double__pyx_k_dp__pyx_k_dtype__pyx_k_dtype_2__pyx_k_empty__pyx_k_empty_like__pyx_k_enable__pyx_k_endpoint__pyx_k_enter__pyx_k_eps__pyx_k_equal__pyx_k_exit__pyx_k_exponential__pyx_k_exponential_scale_1_0_size_None__pyx_k_f__pyx_k_f_dfnum_dfden_size_None_Draw_sa__pyx_k_final_shape__pyx_k_finfo__pyx_k_flags__pyx_k_flat_found__pyx_k_fleft__pyx_k_float64__pyx_k_fmode__pyx_k_format__pyx_k_found__pyx_k_fright__pyx_k_gamma__pyx_k_gamma_shape_scale_1_0_size_None__pyx_k_gauss__pyx_k_gc__pyx_k_geometric__pyx_k_geometric_p_size_None_Draw_samp__pyx_k_get__pyx_k_get_bit_generator__pyx_k_get_state__pyx_k_get_state_and_legacy_can_only_be__pyx_k_getstate__pyx_k_greater__pyx_k_gumbel__pyx_k_gumbel_loc_0_0_scale_1_0_size_N__pyx_k_has_gauss__pyx_k_high__pyx_k_high_2__pyx_k_hypergeometric__pyx_k_hypergeometric_ngood_nbad_nsamp__pyx_k_i__pyx_k_id__pyx_k_idx__pyx_k_ignore__pyx_k_import__pyx_k_in__pyx_k_index__pyx_k_initializing__pyx_k_int16__pyx_k_int32__pyx_k_int64__pyx_k_int8__pyx_k_intp__pyx_k_invacc__pyx_k_is_coroutine__pyx_k_is_scalar__pyx_k_isenabled__pyx_k_isfinite__pyx_k_isnan__pyx_k_isnative__pyx_k_isscalar__pyx_k_issubdtype__pyx_k_it__pyx_k_item__pyx_k_itemsize__pyx_k_j__pyx_k_k__pyx_k_kappa__pyx_k_key__pyx_k_kwargs__pyx_k_l__pyx_k_lam__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_legacy__pyx_k_legacy_can_only_be_True_when_the__pyx_k_legacy_seeding__pyx_k_length__pyx_k_less__pyx_k_less_equal__pyx_k_lnbad__pyx_k_lngood__pyx_k_lnsample__pyx_k_loc__pyx_k_lock__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_2__pyx_k_main__pyx_k_masked__pyx_k_may_share_memory__pyx_k_mean__pyx_k_mean_and_cov_must_have_same_leng__pyx_k_mean_must_be_1_dimensional__pyx_k_mnarr__pyx_k_mnix__pyx_k_mode__pyx_k_mode_right__pyx_k_msg__pyx_k_mt19937__pyx_k_mu__pyx_k_multin__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_arr__pyx_k_n_uint32__pyx_k_n_uniq__pyx_k_name__pyx_k_nbad__pyx_k_ndim__pyx_k_negative_binomial__pyx_k_negative_binomial_n_p_size_None__pyx_k_new__pyx_k_newbyteorder__pyx_k_ngood__pyx_k_ngood_nbad_nsample__pyx_k_ni__pyx_k_niter__pyx_k_nonc__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_numpy__pyx_k_numpy__core_multiarray_failed_to__pyx_k_numpy__core_umath_failed_to_impo__pyx_k_numpy_linalg__pyx_k_numpy_random_mtrand__pyx_k_numpy_random_mtrand_pyx__pyx_k_object__pyx_k_object_which_is_not_a_subclass__pyx_k_offset__pyx_k_oleft__pyx_k_omode__pyx_k_onbad__pyx_k_ongood__pyx_k_onsample__pyx_k_operator__pyx_k_oright__pyx_k_out__pyx_k_p__pyx_k_p_arr__pyx_k_p_must_be_1_dimensional__pyx_k_p_sum__pyx_k_pareto__pyx_k_pareto_a_size_None_Draw_samples__pyx_k_parr__pyx_k_permutation__pyx_k_permutation_x_Randomly_permute__pyx_k_pickle__pyx_k_pix__pyx_k_poisson__pyx_k_poisson_lam_1_0_size_None_Draw__pyx_k_poisson_lam_max__pyx_k_pop_size__pyx_k_pos__pyx_k_power__pyx_k_power_a_size_None_Draws_samples__pyx_k_probabilities_are_not_non_negati__pyx_k_probabilities_contain_NaN__pyx_k_probabilities_do_not_sum_to_1__pyx_k_prod__pyx_k_psd__pyx_k_pvals__pyx_k_pvals_must_be_a_1_d_sequence__pyx_k_pyx_vtable__pyx_k_raise__pyx_k_rand__pyx_k_rand_2__pyx_k_rand_d0_d1_dn_Random_values_in__pyx_k_randint__pyx_k_randint_low_high_None_size_None__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_randoms__pyx_k_randoms_data__pyx_k_randomstate_ctor__pyx_k_ranf__pyx_k_range__pyx_k_ravel__pyx_k_rayleigh__pyx_k_rayleigh_scale_1_0_size_None_Dr__pyx_k_reduce__pyx_k_reduce_2__pyx_k_replace__pyx_k_res__pyx_k_reshape__pyx_k_result_type__pyx_k_ret__pyx_k_return_index__pyx_k_reversed__pyx_k_right__pyx_k_rtol__pyx_k_s__pyx_k_sample__pyx_k_scale__pyx_k_searchsorted__pyx_k_seed__pyx_k_seed_seed_None_Reseed_a_legacy__pyx_k_self__pyx_k_set_bit_generator__pyx_k_set_state__pyx_k_set_state_can_only_be_used_with__pyx_k_setstate__pyx_k_shape__pyx_k_shuffle__pyx_k_shuffle_x_Modify_a_sequence_in__pyx_k_side__pyx_k_sigma__pyx_k_singleton__pyx_k_size__pyx_k_sort__pyx_k_spec__pyx_k_sqrt__pyx_k_st__pyx_k_stacklevel__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_dictionary_is_not_valid__pyx_k_state_must_be_a_dict_or_a_tuple__pyx_k_str__pyx_k_stride__pyx_k_strides__pyx_k_subtract__pyx_k_sum__pyx_k_sum_pvals_1_1_0__pyx_k_sum_pvals_1_astype_np_float64_1__pyx_k_svd__pyx_k_sz__pyx_k_take__pyx_k_temp__pyx_k_test__pyx_k_tobytes__pyx_k_tol__pyx_k_tomaxint__pyx_k_tomaxint_size_None_Return_a_sam__pyx_k_totsize__pyx_k_triangular__pyx_k_triangular_left_mode_right_size__pyx_k_type__pyx_k_u__pyx_k_u4__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_uniform_samples__pyx_k_unique__pyx_k_unique_indices__pyx_k_unsafe__pyx_k_v__pyx_k_val_arr__pyx_k_val_data__pyx_k_value__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_writeable__pyx_k_x__pyx_k_x_must_be_an_integer_or_at_least__pyx_k_x_ptr__pyx_k_you_are_shuffling_a__pyx_k_zeros__pyx_k_zipf__pyx_k_zipf_a_size_None_Draw_samples_f__Pyx_GetVtable.isra.0__Pyx__ExceptionReset.isra.0__Pyx__ExceptionSave.isra.0__Pyx_CyFunction_reduce__Pyx_PyInt_BoolEqObjC.constprop.0__Pyx_PyErr_GivenExceptionMatchesTuple__Pyx_TypeTest__Pyx_PyErr_GivenExceptionMatches.part.0__Pyx_CyFunction_get_defaults__Pyx_ImportFrom__Pyx_PyInt_As_long__Pyx_PyObject_FastCallDict.constprop.0__Pyx_PyInt_As_int__Pyx_GetKwValue_FASTCALL__Pyx_CyFunction_get_is_coroutine__Pyx_IterFinish__Pyx_PyInt_As_npy_intp.part.0__Pyx_AddTraceback__pyx_dict_version.2__pyx_dict_cached_value.1__pyx_code_cache__pyx_pw_5numpy_6random_6mtrand_11RandomState_21random__pyx_pw_5numpy_6random_6mtrand_11RandomState_11__reduce____pyx_pw_5numpy_6random_6mtrand_11RandomState_7__getstate____pyx_pw_5numpy_6random_6mtrand_11RandomState_5__str____Pyx_PyObject_FastCallDict.constprop.1__pyx_pw_5numpy_6random_6mtrand_11RandomState_39rand__pyx_pw_5numpy_6random_6mtrand_11RandomState_41randn__Pyx_PyObject_CallMethod0__pyx_f_5numpy_6random_6mtrand_11RandomState__initialize_bit_generator__pyx_builtin_ValueError__pyx_pf_5numpy_6random_6mtrand_11RandomState_2__repr____pyx_builtin_id__pyx_specialmethod___pyx_pw_5numpy_6random_6mtrand_11RandomState_3__repr____Pyx_PyObject_GetItem__Pyx_PyInt_As_npy_intp__Pyx_PyInt_As_int64_t__Pyx_PyObject_GetAttrStrNoError__Pyx_ImportDottedModule.constprop.0__pyx_pw_5numpy_6random_6mtrand_11RandomState_47normal__pyx_f_5numpy_6random_7_common_cont__pyx_pw_5numpy_6random_6mtrand_11RandomState_73laplace__pyx_pw_5numpy_6random_6mtrand_11RandomState_77logistic__pyx_pw_5numpy_6random_6mtrand_11RandomState_79lognormal__pyx_pw_5numpy_6random_6mtrand_11RandomState_75gumbel__pyx_pw_5numpy_6random_6mtrand_11RandomState_9__setstate____Pyx_GetBuiltinName__Pyx__GetModuleGlobalName__pyx_f_5numpy_6random_6mtrand_int64_to_long__pyx_dict_version.128__pyx_dict_cached_value.127__pyx_pw_5numpy_6random_6mtrand_11RandomState_97hypergeometricPyArray_API__pyx_dict_version.126__pyx_dict_cached_value.125__pyx_dict_version.124__pyx_dict_cached_value.123__pyx_dict_version.122__pyx_dict_cached_value.121__pyx_f_5numpy_6random_7_common_discrete_broadcast_iii__pyx_f_5numpy_6random_7_common_disc__pyx_pw_5numpy_6random_6mtrand_11RandomState_85triangular__pyx_dict_version.108__pyx_dict_cached_value.107__pyx_dict_version.106__pyx_dict_cached_value.105__pyx_dict_version.104__pyx_dict_cached_value.103__pyx_dict_version.102__pyx_dict_cached_value.101__pyx_dict_version.100__pyx_dict_cached_value.99__pyx_dict_version.98__pyx_dict_cached_value.97__pyx_f_5numpy_6random_7_common_cont_broadcast_3__pyx_pw_5numpy_6random_6mtrand_11RandomState_37uniform__pyx_dict_version.90__pyx_dict_cached_value.89__pyx_dict_version.88__pyx_dict_cached_value.87__pyx_dict_version.86__pyx_dict_cached_value.85__pyx_dict_version.92__pyx_dict_cached_value.91__pyx_builtin_OverflowError__pyx_pw_5numpy_6random_6mtrand_11RandomState_33bytes__pyx_dict_version.40__pyx_dict_cached_value.39__pyx_pw_5numpy_6random_6mtrand_11RandomState_1__init____pyx_dict_version.6__pyx_dict_cached_value.5__pyx_dict_version.4__pyx_dict_cached_value.3__pyx_pymod_exec_mtrand__pyx_CyFunctionType_type__pyx_builtin_TypeError__pyx_builtin_RuntimeWarning__pyx_builtin_DeprecationWarning__pyx_builtin_UserWarning__pyx_builtin_IndexError__pyx_builtin_ImportError__pyx_vtable_5numpy_6random_6mtrand_RandomState__pyx_type_5numpy_6random_6mtrand_RandomState__pyx_vp_5numpy_6random_7_common_POISSON_LAM_MAX__pyx_vp_5numpy_6random_7_common_LEGACY_POISSON_LAM_MAX__pyx_vp_5numpy_6random_7_common_MAXSIZE__pyx_f_5numpy_6random_17_bounded_integers__rand_uint64__pyx_f_5numpy_6random_17_bounded_integers__rand_uint32__pyx_f_5numpy_6random_17_bounded_integers__rand_uint16__pyx_f_5numpy_6random_17_bounded_integers__rand_uint8__pyx_f_5numpy_6random_17_bounded_integers__rand_bool__pyx_f_5numpy_6random_17_bounded_integers__rand_int64__pyx_f_5numpy_6random_17_bounded_integers__rand_int32__pyx_f_5numpy_6random_17_bounded_integers__rand_int16__pyx_f_5numpy_6random_17_bounded_integers__rand_int8__pyx_f_5numpy_6random_7_common_check_constraint__pyx_f_5numpy_6random_7_common_check_array_constraint__pyx_f_5numpy_6random_7_common_kahan_sum__pyx_f_5numpy_6random_7_common_double_fill__pyx_f_5numpy_6random_7_common_validate_output_shapePyArray_RUNTIME_VERSION__pyx_mdef_5numpy_6random_6mtrand_11RandomState_7__getstate____pyx_mdef_5numpy_6random_6mtrand_11RandomState_9__setstate____pyx_mdef_5numpy_6random_6mtrand_11RandomState_11__reduce____pyx_mdef_5numpy_6random_6mtrand_11RandomState_13seed__pyx_mdef_5numpy_6random_6mtrand_11RandomState_15get_state__pyx_mdef_5numpy_6random_6mtrand_11RandomState_17set_state__pyx_mdef_5numpy_6random_6mtrand_11RandomState_19random_sample__pyx_mdef_5numpy_6random_6mtrand_11RandomState_21random__pyx_mdef_5numpy_6random_6mtrand_11RandomState_23beta__pyx_mdef_5numpy_6random_6mtrand_11RandomState_25exponential__pyx_mdef_5numpy_6random_6mtrand_11RandomState_27standard_exponential__pyx_mdef_5numpy_6random_6mtrand_11RandomState_29tomaxint__pyx_mdef_5numpy_6random_6mtrand_11RandomState_31randint__pyx_mdef_5numpy_6random_6mtrand_11RandomState_33bytes__pyx_mdef_5numpy_6random_6mtrand_11RandomState_35choice__pyx_mdef_5numpy_6random_6mtrand_11RandomState_37uniform__pyx_mdef_5numpy_6random_6mtrand_11RandomState_39rand__pyx_mdef_5numpy_6random_6mtrand_11RandomState_41randn__pyx_mdef_5numpy_6random_6mtrand_11RandomState_43random_integers__pyx_mdef_5numpy_6random_6mtrand_11RandomState_45standard_normal__pyx_mdef_5numpy_6random_6mtrand_11RandomState_47normal__pyx_mdef_5numpy_6random_6mtrand_11RandomState_49standard_gamma__pyx_mdef_5numpy_6random_6mtrand_11RandomState_51gamma__pyx_mdef_5numpy_6random_6mtrand_11RandomState_53f__pyx_mdef_5numpy_6random_6mtrand_11RandomState_55noncentral_f__pyx_mdef_5numpy_6random_6mtrand_11RandomState_57chisquare__pyx_mdef_5numpy_6random_6mtrand_11RandomState_59noncentral_chisquare__pyx_mdef_5numpy_6random_6mtrand_11RandomState_61standard_cauchy__pyx_mdef_5numpy_6random_6mtrand_11RandomState_63standard_t__pyx_mdef_5numpy_6random_6mtrand_11RandomState_65vonmises__pyx_mdef_5numpy_6random_6mtrand_11RandomState_67pareto__pyx_mdef_5numpy_6random_6mtrand_11RandomState_69weibull__pyx_mdef_5numpy_6random_6mtrand_11RandomState_71power__pyx_mdef_5numpy_6random_6mtrand_11RandomState_73laplace__pyx_mdef_5numpy_6random_6mtrand_11RandomState_75gumbel__pyx_mdef_5numpy_6random_6mtrand_11RandomState_77logistic__pyx_mdef_5numpy_6random_6mtrand_11RandomState_79lognormal__pyx_mdef_5numpy_6random_6mtrand_11RandomState_81rayleigh__pyx_mdef_5numpy_6random_6mtrand_11RandomState_83wald__pyx_mdef_5numpy_6random_6mtrand_11RandomState_85triangular__pyx_mdef_5numpy_6random_6mtrand_11RandomState_87binomial__pyx_mdef_5numpy_6random_6mtrand_11RandomState_89negative_binomial__pyx_mdef_5numpy_6random_6mtrand_11RandomState_91poisson__pyx_mdef_5numpy_6random_6mtrand_11RandomState_93zipf__pyx_mdef_5numpy_6random_6mtrand_11RandomState_95geometric__pyx_mdef_5numpy_6random_6mtrand_11RandomState_97hypergeometric__pyx_mdef_5numpy_6random_6mtrand_11RandomState_99logseries__pyx_mdef_5numpy_6random_6mtrand_11RandomState_101multivariate_normal__pyx_mdef_5numpy_6random_6mtrand_11RandomState_103multinomial__pyx_mdef_5numpy_6random_6mtrand_11RandomState_105dirichlet__pyx_mdef_5numpy_6random_6mtrand_11RandomState_107shuffle__pyx_mdef_5numpy_6random_6mtrand_11RandomState_109permutation__pyx_dict_version.304__pyx_dict_cached_value.303__pyx_dict_version.302__pyx_dict_cached_value.301__pyx_dict_version.300__pyx_dict_cached_value.299__pyx_dict_version.298__pyx_dict_cached_value.297__pyx_dict_version.296__pyx_dict_cached_value.295__pyx_dict_version.294__pyx_dict_cached_value.293__pyx_dict_version.292__pyx_dict_cached_value.291__pyx_dict_version.290__pyx_dict_cached_value.289__pyx_dict_version.288__pyx_dict_cached_value.287__pyx_dict_version.286__pyx_dict_cached_value.285__pyx_dict_version.284__pyx_dict_cached_value.283__pyx_dict_version.282__pyx_dict_cached_value.281__pyx_dict_version.280__pyx_dict_cached_value.279__pyx_dict_version.278__pyx_dict_cached_value.277__pyx_dict_version.276__pyx_dict_cached_value.275__pyx_dict_version.274__pyx_dict_cached_value.273__pyx_dict_version.272__pyx_dict_cached_value.271__pyx_dict_version.270__pyx_dict_cached_value.269__pyx_dict_version.268__pyx_dict_cached_value.267__pyx_dict_version.266__pyx_dict_cached_value.265__pyx_dict_version.264__pyx_dict_cached_value.263__pyx_dict_version.262__pyx_dict_cached_value.261__pyx_dict_version.260__pyx_dict_cached_value.259__pyx_dict_version.258__pyx_dict_cached_value.257__pyx_dict_version.256__pyx_dict_cached_value.255__pyx_dict_version.254__pyx_dict_cached_value.253__pyx_dict_version.252__pyx_dict_cached_value.251__pyx_dict_version.250__pyx_dict_cached_value.249__pyx_dict_version.248__pyx_dict_cached_value.247__pyx_dict_version.246__pyx_dict_cached_value.245__pyx_dict_version.244__pyx_dict_cached_value.243__pyx_dict_version.242__pyx_dict_cached_value.241__pyx_dict_version.240__pyx_dict_cached_value.239__pyx_dict_version.238__pyx_dict_cached_value.237__pyx_dict_version.236__pyx_dict_cached_value.235__pyx_dict_version.234__pyx_dict_cached_value.233__pyx_dict_version.232__pyx_dict_cached_value.231__pyx_dict_version.230__pyx_dict_cached_value.229__pyx_dict_version.228__pyx_dict_cached_value.227__pyx_dict_version.226__pyx_dict_cached_value.225__pyx_dict_version.224__pyx_dict_cached_value.223__pyx_dict_version.222__pyx_dict_cached_value.221__pyx_dict_version.220__pyx_dict_cached_value.219__pyx_dict_version.218__pyx_dict_cached_value.217__pyx_dict_version.216__pyx_dict_cached_value.215__pyx_dict_version.214__pyx_dict_cached_value.213__pyx_dict_version.212__pyx_dict_cached_value.211__pyx_mdef_5numpy_6random_6mtrand_1seed__pyx_mdef_5numpy_6random_6mtrand_3get_bit_generator__pyx_mdef_5numpy_6random_6mtrand_5set_bit_generator__pyx_mdef_5numpy_6random_6mtrand_7sample__pyx_mdef_5numpy_6random_6mtrand_9ranf__pyx_pw_5numpy_6random_6mtrand_9ranf__pyx_dict_version.210__pyx_dict_cached_value.209__pyx_pw_5numpy_6random_6mtrand_7sample__pyx_dict_version.208__pyx_dict_cached_value.207__pyx_pw_5numpy_6random_6mtrand_3get_bit_generator__pyx_dict_version.204__pyx_dict_cached_value.203__pyx_pw_5numpy_6random_6mtrand_5set_bit_generator__pyx_dict_version.206__pyx_dict_cached_value.205__pyx_pw_5numpy_6random_6mtrand_11RandomState_81rayleigh__pyx_pw_5numpy_6random_6mtrand_11RandomState_25exponential__pyx_pw_5numpy_6random_6mtrand_11RandomState_107shuffle__pyx_dict_version.176__pyx_dict_cached_value.175__pyx_dict_version.174__pyx_dict_cached_value.173__pyx_dict_version.172__pyx_dict_cached_value.171__pyx_dict_version.170__pyx_dict_cached_value.169__pyx_dict_version.168__pyx_dict_cached_value.167__pyx_dict_version.166__pyx_dict_cached_value.165__pyx_dict_version.164__pyx_dict_cached_value.163__pyx_pw_5numpy_6random_6mtrand_11RandomState_15get_state__pyx_dict_version.12__pyx_dict_cached_value.11__pyx_dict_version.10__pyx_dict_cached_value.9__pyx_pw_5numpy_6random_6mtrand_11RandomState_13seed__pyx_dict_version.8__pyx_dict_cached_value.7__pyx_pw_5numpy_6random_6mtrand_1seed__pyx_dict_version.202__pyx_dict_cached_value.201__pyx_dict_version.200__pyx_dict_cached_value.199__pyx_dict_version.198__pyx_dict_cached_value.197__pyx_dict_version.196__pyx_dict_cached_value.195__pyx_dict_version.194__pyx_dict_cached_value.193__pyx_pw_5numpy_6random_6mtrand_11RandomState_91poisson__pyx_pw_5numpy_6random_6mtrand_11RandomState_55noncentral_f__pyx_pw_5numpy_6random_6mtrand_11RandomState_17set_state__pyx_pw_5numpy_6random_6mtrand_11RandomState_29tomaxint__pyx_dict_version.16__pyx_dict_cached_value.15__pyx_dict_version.14__pyx_dict_cached_value.13__pyx_pw_5numpy_6random_6mtrand_11RandomState_103multinomial__pyx_dict_version.150__pyx_dict_cached_value.149__pyx_dict_version.148__pyx_dict_cached_value.147__pyx_dict_version.146__pyx_dict_cached_value.145__pyx_dict_version.152__pyx_dict_cached_value.151__pyx_pw_5numpy_6random_6mtrand_11RandomState_19random_sample__pyx_pw_5numpy_6random_6mtrand_11RandomState_105dirichlet__pyx_dict_version.162__pyx_dict_cached_value.161__pyx_dict_version.160__pyx_dict_cached_value.159__pyx_dict_version.158__pyx_dict_cached_value.157__pyx_dict_version.156__pyx_dict_cached_value.155__pyx_dict_version.154__pyx_dict_cached_value.153__pyx_pw_5numpy_6random_6mtrand_11RandomState_27standard_exponential__pyx_pw_5numpy_6random_6mtrand_11RandomState_45standard_normal__pyx_pw_5numpy_6random_6mtrand_11RandomState_61standard_cauchy__pyx_pw_5numpy_6random_6mtrand_11RandomState_43random_integers__pyx_dict_version.94__pyx_dict_cached_value.93__pyx_dict_version.96__pyx_dict_cached_value.95__pyx_pw_5numpy_6random_6mtrand_11RandomState_83wald__pyx_pw_5numpy_6random_6mtrand_11RandomState_65vonmises__pyx_pw_5numpy_6random_6mtrand_11RandomState_59noncentral_chisquare__pyx_pw_5numpy_6random_6mtrand_11RandomState_53f__pyx_pw_5numpy_6random_6mtrand_11RandomState_23beta__pyx_pw_5numpy_6random_6mtrand_11RandomState_89negative_binomial__pyx_pw_5numpy_6random_6mtrand_11RandomState_57chisquare__pyx_pw_5numpy_6random_6mtrand_11RandomState_67pareto__pyx_pw_5numpy_6random_6mtrand_11RandomState_69weibull__pyx_pw_5numpy_6random_6mtrand_11RandomState_49standard_gamma__pyx_pw_5numpy_6random_6mtrand_11RandomState_63standard_t__pyx_pw_5numpy_6random_6mtrand_11RandomState_71power__pyx_pw_5numpy_6random_6mtrand_11RandomState_95geometric__pyx_pw_5numpy_6random_6mtrand_11RandomState_99logseries__pyx_pw_5numpy_6random_6mtrand_11RandomState_93zipf__pyx_pw_5numpy_6random_6mtrand_11RandomState_51gamma__pyx_pw_5numpy_6random_6mtrand_11RandomState_31randint__pyx_dict_version.38__pyx_dict_cached_value.37__pyx_dict_version.36__pyx_dict_cached_value.35__pyx_dict_version.34__pyx_dict_cached_value.33__pyx_dict_version.32__pyx_dict_cached_value.31__pyx_dict_version.18__pyx_dict_cached_value.17__pyx_dict_version.30__pyx_dict_cached_value.29__pyx_dict_version.28__pyx_dict_cached_value.27__pyx_dict_version.26__pyx_dict_cached_value.25__pyx_dict_version.24__pyx_dict_cached_value.23__pyx_dict_version.22__pyx_dict_cached_value.21__pyx_dict_version.20__pyx_dict_cached_value.19__pyx_pw_5numpy_6random_6mtrand_11RandomState_109permutation__pyx_dict_version.192__pyx_dict_cached_value.191__pyx_dict_version.190__pyx_dict_cached_value.189__pyx_dict_version.188__pyx_dict_cached_value.187__pyx_dict_version.186__pyx_dict_cached_value.185__pyx_dict_version.184__pyx_dict_cached_value.183__pyx_dict_version.182__pyx_dict_cached_value.181__pyx_dict_version.180__pyx_dict_cached_value.179__pyx_dict_version.178__pyx_dict_cached_value.177__pyx_pw_5numpy_6random_6mtrand_11RandomState_101multivariate_normal__pyx_dict_version.144__pyx_dict_cached_value.143__pyx_dict_version.142__pyx_dict_cached_value.141__pyx_dict_version.140__pyx_dict_cached_value.139__pyx_dict_version.132__pyx_dict_cached_value.131__pyx_dict_version.130__pyx_dict_cached_value.129__pyx_dict_version.138__pyx_dict_cached_value.137__pyx_dict_version.136__pyx_dict_cached_value.135__pyx_dict_version.134__pyx_dict_cached_value.133__pyx_pw_5numpy_6random_6mtrand_11RandomState_87binomial__pyx_dict_version.120__pyx_dict_cached_value.119__pyx_dict_version.118__pyx_dict_cached_value.117__pyx_dict_version.112__pyx_dict_cached_value.111__pyx_dict_version.110__pyx_dict_cached_value.109__pyx_dict_version.116__pyx_dict_cached_value.115__pyx_dict_version.114__pyx_dict_cached_value.113__pyx_pw_5numpy_6random_6mtrand_11RandomState_35choice__pyx_dict_version.84__pyx_dict_cached_value.83__pyx_dict_version.82__pyx_dict_cached_value.81__pyx_dict_version.80__pyx_dict_cached_value.79__pyx_dict_version.76__pyx_dict_cached_value.75__pyx_dict_version.74__pyx_dict_cached_value.73__pyx_dict_version.72__pyx_dict_cached_value.71__pyx_dict_version.70__pyx_dict_cached_value.69__pyx_dict_version.68__pyx_dict_cached_value.67__pyx_dict_version.66__pyx_dict_cached_value.65__pyx_dict_version.60__pyx_dict_cached_value.59__pyx_dict_version.58__pyx_dict_cached_value.57__pyx_dict_version.56__pyx_dict_cached_value.55__pyx_dict_version.54__pyx_dict_cached_value.53__pyx_dict_version.42__pyx_dict_cached_value.41__pyx_dict_version.78__pyx_dict_cached_value.77__pyx_dict_version.52__pyx_dict_cached_value.51__pyx_dict_version.50__pyx_dict_cached_value.49__pyx_dict_version.48__pyx_dict_cached_value.47__pyx_dict_version.46__pyx_dict_cached_value.45__pyx_dict_version.44__pyx_dict_cached_value.43__pyx_dict_version.64__pyx_dict_cached_value.63__pyx_dict_version.62__pyx_dict_cached_value.61__pyx_moduledef__pyx_CyFunction_methods__pyx_CyFunction_members__pyx_CyFunction_getsets__pyx_methods__pyx_moduledef_slots__pyx_methods_5numpy_6random_6mtrand_RandomState__pyx_getsets_5numpy_6random_6mtrand_RandomState__pyx_doc_5numpy_6random_6mtrand_11RandomState_12seed__pyx_doc_5numpy_6random_6mtrand_11RandomState_14get_state__pyx_doc_5numpy_6random_6mtrand_11RandomState_16set_state__pyx_doc_5numpy_6random_6mtrand_11RandomState_18random_sample__pyx_doc_5numpy_6random_6mtrand_11RandomState_20random__pyx_doc_5numpy_6random_6mtrand_11RandomState_22beta__pyx_doc_5numpy_6random_6mtrand_11RandomState_24exponential__pyx_doc_5numpy_6random_6mtrand_11RandomState_26standard_exponential__pyx_doc_5numpy_6random_6mtrand_11RandomState_28tomaxint__pyx_doc_5numpy_6random_6mtrand_11RandomState_30randint__pyx_doc_5numpy_6random_6mtrand_11RandomState_32bytes__pyx_doc_5numpy_6random_6mtrand_11RandomState_34choice__pyx_doc_5numpy_6random_6mtrand_11RandomState_36uniform__pyx_doc_5numpy_6random_6mtrand_11RandomState_38rand__pyx_doc_5numpy_6random_6mtrand_11RandomState_40randn__pyx_doc_5numpy_6random_6mtrand_11RandomState_42random_integers__pyx_doc_5numpy_6random_6mtrand_11RandomState_44standard_normal__pyx_doc_5numpy_6random_6mtrand_11RandomState_46normal__pyx_doc_5numpy_6random_6mtrand_11RandomState_48standard_gamma__pyx_doc_5numpy_6random_6mtrand_11RandomState_50gamma__pyx_doc_5numpy_6random_6mtrand_11RandomState_52f__pyx_doc_5numpy_6random_6mtrand_11RandomState_54noncentral_f__pyx_doc_5numpy_6random_6mtrand_11RandomState_56chisquare__pyx_doc_5numpy_6random_6mtrand_11RandomState_58noncentral_chisquare__pyx_doc_5numpy_6random_6mtrand_11RandomState_60standard_cauchy__pyx_doc_5numpy_6random_6mtrand_11RandomState_62standard_t__pyx_doc_5numpy_6random_6mtrand_11RandomState_64vonmises__pyx_doc_5numpy_6random_6mtrand_11RandomState_66pareto__pyx_doc_5numpy_6random_6mtrand_11RandomState_68weibull__pyx_doc_5numpy_6random_6mtrand_11RandomState_70power__pyx_doc_5numpy_6random_6mtrand_11RandomState_72laplace__pyx_doc_5numpy_6random_6mtrand_11RandomState_74gumbel__pyx_doc_5numpy_6random_6mtrand_11RandomState_76logistic__pyx_doc_5numpy_6random_6mtrand_11RandomState_78lognormal__pyx_doc_5numpy_6random_6mtrand_11RandomState_80rayleigh__pyx_doc_5numpy_6random_6mtrand_11RandomState_82wald__pyx_doc_5numpy_6random_6mtrand_11RandomState_84triangular__pyx_doc_5numpy_6random_6mtrand_11RandomState_86binomial__pyx_doc_5numpy_6random_6mtrand_11RandomState_88negative_binomial__pyx_doc_5numpy_6random_6mtrand_11RandomState_90poisson__pyx_doc_5numpy_6random_6mtrand_11RandomState_92zipf__pyx_doc_5numpy_6random_6mtrand_11RandomState_94geometric__pyx_doc_5numpy_6random_6mtrand_11RandomState_96hypergeometric__pyx_doc_5numpy_6random_6mtrand_11RandomState_98logseries__pyx_doc_5numpy_6random_6mtrand_11RandomState_100multivariate_normal__pyx_doc_5numpy_6random_6mtrand_11RandomState_102multinomial__pyx_doc_5numpy_6random_6mtrand_11RandomState_104dirichlet__pyx_doc_5numpy_6random_6mtrand_11RandomState_106shuffle__pyx_doc_5numpy_6random_6mtrand_11RandomState_108permutation__pyx_doc_5numpy_6random_6mtrand_8ranf__pyx_doc_5numpy_6random_6mtrand_6sample__pyx_doc_5numpy_6random_6mtrand_4set_bit_generator__pyx_doc_5numpy_6random_6mtrand_2get_bit_generator__pyx_doc_5numpy_6random_6mtrand_seedcrtstuff.cderegister_tm_clones__do_global_dtors_auxcompleted.0__do_global_dtors_aux_fini_array_entryframe_dummy__frame_dummy_init_array_entryrandom_loggam.part.0fe_doublewe_doubleke_doublefe_floatwe_floatke_floatwi_doubleki_doublefi_doublerandom_standard_gamma.part.0wi_floatki_floatfi_floatlegacy-distributions.clegacy_gauss.part.0legacy_standard_gamma.part.0__FRAME_END__random_laplacerandom_buffered_bounded_boollegacy_random_zipfrandom_geometric_inversionlegacy_frandom_weibullrandom_flegacy_paretorandom_negative_binomialrandom_standard_cauchy__pyx_module_is_main_numpy__random__mtrandlegacy_chisquarerandom_standard_exponential_fill_flegacy_gaussrandom_standard_gammarandom_binomial_btperandom_logserieslegacy_normalrandom_rayleighrandom_standard_exponentialrandom_uniformlegacy_random_binomialrandom_bounded_uint64_filllegacy_random_multinomialrandom_bounded_uint16_filllegacy_standard_exponentialrandom_logisticlegacy_negative_binomialrandom_standard_uniform_fill_frandom_bounded_uint64random_positive_intrandom_standard_gamma_frandom_triangularrandom_buffered_bounded_uint32legacy_rayleighrandom_powerrandom_bounded_uint8_fillrandom_noncentral_frandom_standard_exponential_inv_fill_flegacy_waldrandom_buffered_bounded_uint8random_betarandom_exponential__dso_handlerandom_gammalegacy_random_poissonrandom_standard_uniform_frandom_loggamrandom_gamma_flegacy_weibullrandom_standard_exponential_frandom_paretorandom_positive_int64legacy_standard_gammarandom_geometric_searchrandom_standard_trandom_vonmisesrandom_bounded_uint32_fillrandom_standard_normal_frandom_positive_int32random_standard_uniformlegacy_powerrandom_normallegacy_exponentialrandom_chisquarelegacy_standard_cauchylegacy_gammarandom_standard_exponential_fillrandom_intervalrandom_waldrandom_noncentral_chisquare_DYNAMICrandom_standard_normallegacy_betalegacy_noncentral_frandom_standard_exponential_inv_fillrandom_lognormalrandom_buffered_bounded_uint16legacy_random_hypergeometricrandom_uintrandom_gumbelrandom_standard_uniform_filllegacy_standard_trandom_standard_normal_fill_flegacy_logserieslegacy_random_geometricrandom_bounded_bool_fill__GNU_EH_FRAME_HDR__TMC_END___GLOBAL_OFFSET_TABLE_legacy_vonmisesrandom_binomial_inversionlegacy_noncentral_chisquarerandom_standard_normal_filllegacy_lognormalPyUnicode_FromFormatlog1pf@@GLIBC_2.2.5PyObject_SetItemPyList_New_PyUnicode_ReadyPyExc_SystemErrorPyDict_SetItemStringfree@@GLIBC_2.2.5PyDict_SizePyException_SetTracebackPyMethod_Type_ITM_deregisterTMCloneTablePyFloat_TypePyTuple_TypePyObject_FormatPyErr_RestorePyList_AsTuplePyObject_ClearWeakRefs_PyThreadState_UncheckedGetPyModuleDef_InitPyEval_RestoreThreadPy_EnterRecursiveCallPyFrame_NewPyMem_Free__isnan@@GLIBC_2.2.5PyCapsule_GetNamePyNumber_InPlaceAddexp@@GLIBC_2.2.5PyNumber_AddPyObject_GetAttrStringPyImport_AddModulePyBytes_FromStringAndSize_PyObject_GenericGetAttrWithDictPyObject_SetAttrStringPyErr_WarnEx_Py_DeallocPyModule_NewObjectPyErr_NoMemoryPyErr_SetObjectPyObject_GC_DelPyErr_NormalizeExceptionPyNumber_MultiplyPyObject_RichCompare_finiPyImport_GetModuleDictPyObject_GC_TrackPyExc_RuntimeErrorPyNumber_LongPyErr_GivenExceptionMatchesPyErr_SetStringPyObject_IsInstance_PyObject_GC_NewPyExc_ExceptionPyExc_ValueErrorstrrchr@@GLIBC_2.2.5PyExc_DeprecationWarningPyExc_TypeErrorPyInterpreterState_GetIDPySequence_ContainsPyTuple_GetItemmemset@@GLIBC_2.2.5PyMem_ReallocPyErr_ExceptionMatchespow@@GLIBC_2.2.5log@@GLIBC_2.2.5PyOS_snprintfPyTraceBack_Herelog1p@@GLIBC_2.2.5fmod@@GLIBC_2.2.5PyObject_CallFinalizerFromDeallocPyObject_NotPyObject_FreePyNumber_InPlaceTrueDividePyLong_FromSsize_tPyFloat_FromDoublePyType_Readyacos@@GLIBC_2.2.5PyLong_FromLongmemcmp@@GLIBC_2.2.5PyLong_AsSsize_tPyModule_GetNamePyErr_ClearPyList_AppendPyCapsule_IsValidPyExc_KeyErrorPyImport_GetModule_PyUnicode_FastCopyCharacters_Py_FalseStruct__gmon_start__expf@@GLIBC_2.2.5PyTuple_NewPyObject_GenericGetAttrPyThreadState_GetPyExc_OverflowErrormemcpy@@GLIBC_2.14expm1@@GLIBC_2.2.5PyNumber_RemainderPyType_ModifiedPyObject_SetAttrPyErr_Occurred_Py_EllipsisObjectPyLong_AsLongPyImport_ImportModulesqrtf@@GLIBC_2.2.5_PyDict_GetItem_KnownHashPy_LeaveRecursiveCallPyObject_VectorcallDictPyTuple_GetSlicePyDict_GetItemStringpowf@@GLIBC_2.2.5PyObject_Sizemalloc@@GLIBC_2.2.5_Py_NoneStructPyExc_ModuleNotFoundErrorPyFloat_AsDoublePyObject_IsTrue_PyType_LookupPyImport_ImportModuleLevelObjectPyObject_HashPyUnicode_ComparePyInit_mtrand_Py_TrueStructlogf@@GLIBC_2.2.5PyDict_NewPyExc_IndexErrorPyObject_GC_IsFinalizedPyBool_TypePyDict_TypePyDict_NextPyBaseObject_Typememmove@@GLIBC_2.2.5PyUnicode_AsUTF8PyLong_TypePyCapsule_Type_PyObject_GetDictPtrPyErr_FetchPyUnicode_FromStringPyObject_GetIterPyEval_SaveThreadPyUnicode_InternFromStringPyExc_ImportErrorPyDict_SetItemPySequence_TuplePyExc_AttributeErrorPyDict_CopyPyExc_StopIterationPySequence_ListPyExc_RuntimeWarningfloor@@GLIBC_2.2.5PyUnicode_TypePyCapsule_NewPyUnicode_DecodePyErr_FormatPyCapsule_GetPointerPySlice_NewPyExc_NameErrorPyUnicode_FromStringAndSizePyModule_GetDict_ITM_registerTMCloneTablePyUnicode_ConcatPyNumber_IndexPyObject_GetAttrsqrt@@GLIBC_2.2.5PyCFunction_Type_PyDict_NewPresizedceil@@GLIBC_2.2.5PyUnicode_FormatPyLong_FromStringPyMem_MallocPyErr_WarnFormat__cxa_finalize@@GLIBC_2.2.5_initPyNumber_SubtractPyUnicode_NewPyTuple_PackPyCode_NewWithPosOnlyArgsPy_GetVersionPyCode_NewEmptyPyObject_GC_UnTrackPyDict_GetItemWithErrorPyList_TypePyImport_Import.symtab.strtab.shstrtab.note.gnu.build-id.gnu.hash.dynsym.dynstr.gnu.version.gnu.version_r.rela.dyn.rela.plt.init.text.fini.rodata.eh_frame_hdr.eh_frame.init_array.fini_array.data.rel.ro.dynamic.got.got.plt.data.bss.comment88$.o``48 @   Ho  vUo!!Pd!!X)nB@K@Kx``s ` `p ~ii$|| 0 0T 0T \ \ N           # 0 /  T