ELF>@ @8@m m p{ p{)p{)=W { {){)$$Ptd   QtdRtdp{ p{)p{)GNU1׆}UtV2C?K 38C p"  #@012a@AT% $@ RmHJݼ(y a aף{6}tkm@tOjt#H)M ,,qғ75kΑ(B9 S fQ{`ܪd ^? p^' ] Le =  @y PZ xWg(+ @Z  L%__gmon_start___init_fini_ITM_deregisterTMCloneTable_ITM_registerTMCloneTable__cxa_finalize_Py_NoneStructPyBaseObject_TypePyExc_DeprecationWarningPyErr_WarnFormatPyExc_TypeErrorPyErr_FormatPyThreadState_Get_Py_CheckRecursionLimitPyObject_Call_Py_CheckRecursiveCallPyErr_OccurredPyExc_SystemErrorPyErr_SetString_PyThreadState_UncheckedGetPyFrame_NewPyEval_EvalFrameExPyObject_GetAttrPyLong_FromSsize_tPyTuple_PackPySlice_New_Py_EllipsisObjectPyTuple_NewPyCode_NewPyObject_GC_UnTrackPyObject_CallFinalizerFromDeallocPyErr_FetchPyThread_free_lockPyErr_RestorePyBuffer_ReleasememcpyPyErr_NormalizeExceptionPyException_SetTracebackrandom_interval_PyType_LookupPyExc_AttributeErrorPyErr_SetObjectrandom_bounded_uint64PyCapsule_NewPyDict_SetItemPyErr_ClearPyObject_RichCompareBoolPyObject_GetAttrStringPyExc_ValueErrorPyOS_snprintfPyErr_WarnExPyDict_SetItemStringPyErr_ExceptionMatchesPyDict_GetItemStringPyModule_GetNamePyExc_ImportErrorPyCapsule_IsValidPyCapsule_GetNamePyCapsule_GetPointerPyModule_GetDictPyDict_NewPyImport_ImportModuleLevelObjectPyList_NewPyInterpreterState_GetIDPyModule_NewObjectPyLong_Type_Py_TrueStruct_Py_FalseStructPyFloat_TypePyObject_RichComparePyUnicode_TypememcmpPyObject_IsTrue_PyUnicode_ReadyPyList_TypePyTuple_TypePyObject_GetItemPyExc_OverflowErrorPyDict_NextPyUnicode_ComparevsnprintfPy_FatalErrorPyEval_EvalCodeExPyExc_NameErrorPyErr_GivenExceptionMatchesPyObject_GC_TrackPyLong_AsLongPyObject_FreefreePyNumber_IndexPyLong_AsSsize_tPyExc_IndexErrorPyLong_AsUnsignedLongPyFunction_TypePyCFunction_TypePyTraceBack_Here_PyObject_GetDictPtrPyObject_NotPyObject_SetAttrPyUnicode_FromStringPyUnicode_FromFormatPyMem_Realloc_PyDict_GetItem_KnownHashPyMem_MallocPyNumber_InPlaceMultiplyPyNumber_MultiplyPyLong_FromLongPyList_AsTuplePyList_AppendPyUnicode_FormatPyDict_SizePyObject_GenericGetAttrPyObject_SetItemPyExc_NotImplementedErrorrandom_logseriesrandom_geometricrandom_zipfrandom_poissonrandom_negative_binomialrandom_waldrandom_rayleighrandom_lognormalrandom_logisticrandom_gumbelrandom_laplacerandom_powerrandom_weibullrandom_paretorandom_vonmisesrandom_standard_trandom_standard_cauchyrandom_noncentral_chisquarerandom_chisquarerandom_noncentral_frandom_frandom_gammarandom_normalrandom_exponentialrandom_betaPyDict_TypePyDict_GetItemWithErrorPyExc_KeyErrorPyNumber_AddPyNumber_InPlaceAddPyExc_BufferErrorPyMem_FreePyErr_NoMemoryPyGILState_EnsurePyGILState_ReleasePyMethod_TypePyExc_StopIterationPyThread_allocate_lockPyObject_GetBuffer_PyObject_GenericGetAttrWithDictPyDict_DelItemPyType_ModifiedPyExc_RuntimeErrorPy_GetVersionPyFrame_TypePyBytes_FromStringAndSizePyUnicode_FromStringAndSizePyImport_AddModulePyObject_SetAttrStringPyUnicode_InternFromStringPyUnicode_DecodePyObject_HashPyFloat_FromDoublePyLong_FromString__pyx_module_is_main_numpy__random___generatorPyImport_GetModuleDictPyType_ReadyPyImport_ImportModulePyCapsule_TypePyExc_ExceptionPyCFunction_NewEx_PyDict_NewPresizedPyBool_TypePyNumber_RemainderPyUnicode_DecodeASCIIPyBytes_FromStringPyBytes_TypePySequence_TuplePySlice_TypePyObject_Size_PyList_ExtendPyObject_GetIterPyErr_WriteUnraisablePyErr_PrintExPyObject_MallocmallocPyObject_IsInstancePySequence_ListPyNumber_MatrixMultiplyPyNumber_NegativePyExc_UnboundLocalError_PyLong_CopyPyNumber_AbsolutePyNumber_OrPyLong_FromSize_tPyNumber_FloorDividePyEval_SaveThreadrandom_multivariate_hypergeometric_countPyEval_RestoreThreadrandom_multivariate_hypergeometric_marginalsstrlenmemsetPyExc_ZeroDivisionErrorPy_OptimizeFlagPyExc_AssertionErrorPyErr_SetNonerandom_hypergeometricrandom_triangularPyFloat_AsDoublerandom_standard_normal_fillrandom_standard_normal_fill_frandom_standard_uniform_fillrandom_standard_uniform_fill_frandom_standard_gammarandom_multinomialrandom_binomialrandom_standard_gamma_frandom_uniformrandom_standard_exponential_fill_frandom_standard_exponential_inv_fill_frandom_standard_exponential_inv_fillrandom_standard_exponential_fillPyNumber_InPlaceTrueDividePyNumber_SubtractPyLong_FromUnsignedLongPyInit__generatorPyModuleDef_Initrandom_standard_uniform_frandom_standard_uniformrandom_standard_exponentialexplograndom_standard_exponential_fexpflogfrandom_standard_normalrandom_standard_normal_fpowsqrtpowfsqrtfrandom_positive_int64random_positive_int32random_positive_intrandom_uintrandom_loggamrandom_gamma_ffloorrandom_binomial_btperandom_binomial_inversion__isnanacosfmodrandom_geometric_searchrandom_geometric_inversionceilrandom_buffered_bounded_uint32random_buffered_bounded_uint16random_buffered_bounded_uint8random_buffered_bounded_boolrandom_bounded_uint64_fillrandom_bounded_uint32_fillrandom_bounded_uint16_fillrandom_bounded_uint8_fillrandom_bounded_bool_filllogfactoriallibm.so.6libpthread.so.0libc.so.6_edata__bss_start_endGLIBC_2.2.5< ui xVui xp{) x{){){)_+x+_+P+_+H+_+X+_++_+x+`++`+x+ `+x+(`+P+0`++8`+H+``++h`+P+p`+x+x`+P+`++`+X+`+x+`+ +`+@+`++`++`++`+P+`+x+a+x+a+P+a+H+ a+H+(a++0a+x+@a++Ha+x+Pa+P+Xa+H+a++a++a+x+a++a++a+x+a++a++a++a+x+a++a+x+ b++(b++0b+x+@b+x+Pb++Xb+x+b++b++b+x+b+X+b+x+b+X+b+x+b+X+b+x+c+H+c++c+x+ c+H+(c++0c+x+@c+H+Hc++Pc+x+`c++hc++pc+x+c++c+x+c++c++c+x+c+p+c++c++c+x+d+`+d+@+d+x+ d+`+(d+@+0d+x+@d++Hd+x+`d+X+hd+x+d+@+d+x+d++d+@+d++d+x+d+@+d+x+e++e++e+x+e+H+ e++(e++@e+`+He++Pe+x+`e++he++pe+x+xe++e++e+x+e++e++e++e++e++f++f++f++f++ f++0f+H+@f+X+Hf++Pf+H+`f++f++f++f++f++f+f++f+@pf++f+g++ g+@g++Hg+hg++pg+@g++g+g++g+g++g+8h++h+0h++8h+`Xh++`h+h++h+`h++h+h++h+h++i+  i++(i+Hi++Pi+pi++xi+i+x+i+i+p+i+ i+h+i+ j+`+j+`8j+X+@j+@`j+P+hj+@j+H+j+j+@+j+`j+8+j+pk+0+k+(k+(+0k+Pk+ +Xk+`xk++k+ k++k+k++k+k++k+l++ l+o@l++Hl+ hl++pl+l++l+l++l+ l++l+m++m+0m++8m+`Xm++`m+ m++m+m++m+m++m+m++n+ n++(n+Hn++Pn+`pn++xn+n+x+n+@n+p+n+n+h+n+o+`+o+8o+X+@o+``o+P+ho+o+H+o+o+@+o+o+8+o+p+0+p+п(p+(+0p+ Pp+ +Xp+@xp++p+p++p+p++p+@p++p+Pq++ q+@q++Hq+hq++pq+q++q+q++q+ؾq++q+r++r+0r++8r+`Xr++`r+ r++r+`r++r+r++r+r++s+ s++(s+Hs++Ps+ps++xs+os+x+s+s+p+s+s+h+s+t+`+t+8t+X+@t+`t+P+ht+t+H+t+t+@+t+@t+8+t+0u+0+u+(u+(+0u+Pu+ +Xu+xu++u+(u++u+u++u+u++u+v++ v+@v++Hv+hv++pv+zv++v+v++v+v++v+w++w+0w++8w+fXw++`w+aw++w+w++w+w++w+\w++x+`a x++(x+@Hx++Px+@px++xx+Vx+x+x+x+p+x+_x+h+x+y+`+y+8y+X+@y+`y+P+hy+y+H+y+y+@+y+y+8+y+Wz+0+z+@K(z+(+0z+Pz+ +Xz+xz++z+pz++z+z++z+z++z+ {++ {+@{++H{+h{++p{+{++{+Q{++{+j{++{+0|++|+0|++8|+X|++`|+|++|+|++|+|++|+|++}+ }++(}+dH}++P}+dp}++x}+^}+x+}+^}+p+}+H}+h+}+ ~+`+~+8~+X+@~+`~+P+h~+X~+H+~+~+@+~+A~+8+~+A+0++R(+(+0+pP+ +X+yx++++++H++++++++ +;@++H+5h++p+8+++ >+++/+++)+++0++8+X++`++++r+++Ё++؁++++1 ++(+3H++P+p++x+ +x++,+p+Ȃ++h++`+`++8+X+@+``+P+h+,+H+++@++؃+8++k+0++(+(+0+`P+ +X+#x++++++Ȅ++Є++++'++ +@++H+h++p++++"++++++ +++0++8+X++`+`++++++І++؆++++ ++(+H++P+p++x++x++`+p+ȇ++h+++`++P8+X+@+`+P+h+@+H+++@++؈+8+++0++0(+(+0+`P+ +X+`x+++`+++ ȉ++Љ++++++ +@++H+Ph++p+++++++++++++ 0++8+X++`++++E+++Ћ++؋++++ ++(+ H++P+p++x++x++ +p+Ȍ++h+++`++8+X+@+`+P+h++H++@+@++؍+8+++0++(+(+0+P+ +X+`x++++++Ȏ++Ў++++`++ + @++H+h++p+++++++{+++l+++0++8+X++`+++++++А++ؐ++++` ++(+@H++P+p++x+`+x++l+p+ȑ++h+++`++8+X+@+`+P+h++H+++@++ؒ+8+++0++(+(+0+P+ +X+]x++++++`Xȓ++Г+8+++{++ +@++H+ph++p+K++++++@++++++0++8+X++`++++P+++Е++ؕ++++ ++(+H++P+pp++x++x+++p+Ȗ++h++++`++@D8+X+@+`+P+h++H++ ;+@++$ؗ+8++`+0++(+(+0+P+ +X+xx+++p+++Ș++И+P+++++ +@++H+`h++p++++++++++p+++0++8+@X++`+0++++++К++ؚ+X+++7 ++(+H++P+p++x++x+++p+ț++h+++`++8+X+@+`+P+h+ {+H++ +@++@.؜+8++`!+0++(+(+0+P+ +X+x++++++xȝ++Н+ +++z++ +@++H+Ph++p+`++++++0+++@+++0++8+sX++`+s+++ +++П++؟++++H ++(+oH++P+p++x++x+++p+Ƞ++h+++`++8+X+@+`+P+h++H++ +@++ء+8+++0++(+(+0+ P+ +X+px++++++Ȣ++Т++++ ++ +@++H+h++p+k+++@+ ++(+`+h+Ax++Ф+*P+X+'`+(+++@+إ+'@+YH++^+V+pȦ+0U+@0+ X+ h++p+++0m++++0+++ +8+++P+ ++Pب+% +(+ nH+YP+p+!x+`]+++ȩ+P[++[++Z8+@+pY`+h+W+Ȫ+++0++T +(+M@+^H+i`+ph+ h+Ы+ +@X+`++@+Ȭ+oج+@+^H+ `+ph+>+Э+>+`++@+0+0u@+0++ +++خ+`0+@@+0H+vP+u`+0x+` +ȯ+q +(+ u@+^H+r`+ph+`q+hа+P +@(+P+X+`++++@+ȱ+`ر+@@+H+pX+h+'p+x+ +6Ȳ++C++P+ +g(+ 8+X+@+ H+`X+@S+`+ h+x+L+++6+F+++A+9++ȳ+س+ 8++R+G++++++!+ +W(+@8++@+`H+0X+@ +`+h+x+`*+++@*+++ *+ȴ+ش+*+++*++P+* +(+8+*@+H+X+ *`+h+x+ *++С+ *+++ *+ȵ+ص+@|*+++o*+++@]* +(+8+Q*@+H+X+@*`+h+`x+7*+~++@-*+L++#*+uȶ+pض+*+l++` *+d++* +_(+@8+)@+H+|X+@)`+h+x+)+F+y+)+z+`+)+ȷ+ط+)++P+ )+n++) +x(+ T8+)@+H+X+)+%+/+g+0+)+.h})p})x}) })}) })}) })})})})})!})})%})})})})})~)~)4~)9~): ~);(~)=0~)>8~)@~)H~)P~)X~)`~)h~)p~)`x~)b~)~)d~)e~)~)h~)j~)~)~)q~)v~)~) ~)|~)}~)~))))) ) ()0)8)@)H)P)X)`)h)p)x)))))))))))) )))))) )()0)8)@)H)P) X)`) h)p)x))))))))))Ȁ)Ѐ)؀)))) )")#)$)&)' )()0)(8))@)*H)+P),X)-`).h) p)/x)0))1)2)3)5)6)7)8)ȁ)<Ё)?؁)@)A)B)C)D)E)F)G) )H()I0)8)J@)KH)LP)MX)N`)Oh)Pp)Qx)R)S)T)U)V)W)X)Y)Z)Ȃ)[Ђ)\؂)])^)_)a)c))f)g)h )i()k0)8)l@)mH)P)nX)o`)ph)p)rx)s)t)u)w)x)y)z){))~ȃ)Ѓ)؃))))))))) ) ()0)8)@)H)P)X)`)h)p)x))))))))))Ȅ)Є)؄))))))))) )()0)8)@)H)P)X)`)h)p)x))))))))H GH5(%(@%(h%(h%(h%(h%(h%(h%(h%z(hp%r(h`%j(h P%b(h @%Z(h 0%R(h %J(h %B(h%:(h%2(h%*(h%"(h%(h%(h% (h%(h%(hp%(h`%(hP%(h@%(h0%(h %(h%(h%(h%(h %(h!%(h"%(h#%(h$%(h%%(h&%z(h'p%r(h(`%j(h)P%b(h*@%Z(h+0%R(h, %J(h-%B(h.%:(h/%2(h0%*(h1%"(h2%(h3%(h4% (h5%(h6%(h7p%(h8`%(h9P%(h:@%(h;0%(h< %(h=%(h>%(h?%(h@%(hA%(hB%(hC%(hD%(hE%(hF%z(hGp%r(hH`%j(hIP%b(hJ@%Z(hK0%R(hL %J(hM%B(hN%:(hO%2(hP%*(hQ%"(hR%(hS%(hT% (hU%(hV%(hWp%(hX`%(hYP%(hZ@%(h[0%(h\ %(h]%(h^%(h_%(h`%(ha%(hb%(hc%(hd%(he%(hf%z(hgp%r(hh`%j(hiP%b(hj@%Z(hk0%R(hl %J(hm%B(hn%:(ho%2(hp%*(hq%"(hr%(hs%(ht% (hu%(hv%(hwp%(hx`%(hyP%(hz@%(h{0%(h| %(h}%(h~%(h%(h%(h%(h%(h%(h%(h%(h%z(hp%r(h`%j(hP%b(h@%Z(h0%R(h %J(h%B(h%:(h%2(h%*(h%"(h%(h%(h% (h%(h%(hp%(h`%(hP%(h@%ڿ(h0%ҿ(h %ʿ(h%¿(h%(h%(h%(h%(h%(h%(h%(h%(h%z(hp%r(h`%j(hP%b(h@%Z(h0%R(h %J(h%B(h%:(h%2(h%*(h%"(h%(hSH5+1|H*HuO H5+1UHf*HtH*H5*1/H8*HtH5$+1H *HtH5+1H*HqH5+1H*HNH5+1H*H+H5+1Ha*HH5+1eH6*HH5*1BH *HH5*1H*HH5+1H*H|H5+1H*HYH5R+1H_*H6HO(1HHHH+*H H5{ +1gH*HH5+1DH*HH5e+1!H*HH5j+1H*H~H5+1HT*H[H5*1H)*H8H5+1H*HH5+1rH*HH5+1OH*HH5*1,H}*HHHHH*HH5+1H4*HkH+H5U+1H*HAH52+1H*HH5*1{H*HH5*1XH*HH5*15HV*HH5*1H+*HH5C+1H*HoH58*1H*HLH5+1H*H)H"(H5*1Hx*HH5@+1\HM*HH5*19H"*HH5b*1HH5+1H*HzH5S*1H*HWH5*1H*H4H5*1HZ*HH5B+1nH/*HH5*1KH*HH5*1(H*HH5+1H*HH5~+1H*HbH5+1HX*H?H5x+1H-*HH *HH*HH5$*HP1QH*HH5*1.H*HH5?+1 H|*HH5*1HQ*HhH5*1H&*HEH5F*1H)H+L +VA11AQh5*5*RRPRRIHPHH5*19H*HH5*1Hg*HH5*1H<*HsH5*1H*HPH5q*1H*H-L *L*1H *H*H5*nHHf+L W+Q11AAQj5*5*RRPRRHPH[UH1HS1QHtGH5A*HHH+HxHʽHu!HCHP0HHu HCHP0Z[]UHSQH5*;HHu11Ht'HHŅxH u HCHP0Z[]AVEAUIHATIUHSHHHH@u H(HLH5H81mLK M9s#H٬(MHLH5H81AAuMM9vHIPLIAQHPML11L1ZYyH u HCHP01HH[]A\A]A^AVEAUIATIHUS7Ht5H;(Hu1AtHLLH u)HCHP0Hv(H8+t1[]A\A]A^AWIAVIAUIH5ATIUSAPHLHHHHu(L:LH5HH(H81*1KH*HtH=~*KHz*HmH=:*JHXH=5*JHA*HAHHaH59lHH u HCHP0H=<HHA`HHbH5HA@HHH5HHH5*HHt~1HIHu HuHߙ(H5H8HHMu HEHP0Mt7AHHH5YeHtH u2HCHP0&H u HCHP0L%ALH=JjHH&H GH*HH5D\H "H*HH57H HV*HH5H u HCHP0H=HHH Hb*HH5H H5*HH5H H*HH5H H*HH5Z`H mH*HH5n5;H HH*HH5TH #HT*HH5;H H'*HH5"H H*HH5 H u HCHP0H=@`HH3H H*HH5U[H Ha*HH506H H4*HH5 H H*HH5H H*HH5H H*HH5tH H*HH5ew}H HS*HH5ERXH H&*HH5%-3H xH*HH5H SH*HH5H u HCHP0H=*11~-IHlH5*H=*HI $u ID$LP0H=D*113-IH8H5(*H=i*HI $u ID$LP0H=y*11,IHH5*H=*H6RI $u ID$LP03IHHX*LHHK*ID$HH=*1m,HHI $u ID$LP0H5*H?IHH5*H=w*HI $u ID$LP0H u HCHP0}HH7H*HHH*HCHH=*+IHdH u HCHP0H5*L>HH@H5p*H=*H2H u HCHP0I $u ID$LP0HD$HD$ HD$(HD$0HD$8HD$@6HL$(HT$ HHt$IH=EHHH5JHHMHu HEHP0HuH6(H5'H8CH(H9Ct.H(H5/H8H HCHP01H=H H*u HCHP0H*HuH,(H5H8= Hb*t& H5H(H81 H**w' H5H(H81RAŅuH(H5H81\+tHv(H5H81= H|$Hu)3HX(I|$XH0Ruf\rHuHGP0HD$H|$ Ht HuHGP0HD$ H|$(HHHGP0H r\H=~taHL$@HT$8LHt$0M%x@H5*H=*1HHt/H2H u.HCH߽t ]P0"\s]t ]tI$HL$(HT$ Ht$ *H|$0Ht HuHGP0H|$8Ht HuHGP0H|$@HW HN HGP0B HE*IHH*H5*LHl I $u ID$LP0H=*H*H *H9Hu)H*Ht HL%*.H=*:IH=*H*H5*IMH5*L<IH I $u ID$LP0H(*L5I*H b*H9Hu)HM*Ht HH>*.H=e*p9HH=T*H*H5*AHHH5.*HHH, H u HCHP0H*H-*H *H9Hu)H*Ht HL%*.H=*8IH=*Hy*H5z*IMH5*LIH I $u ID$LP0H*L5*H *H9Hu)H*Ht HH*.H=;*F8HH=**H*H5*HHH5*H|HH H u HCHP0Hj*H-s*H t*H9Hu)H_*Ht HL%P*.H=*7IH=*H/*H50*IMH5p*LIH I $u ID$LP0H*1H=*L5*HHZH5*H=*HE H u HCHP0(HH.Hy*H5*Hz HS*H5*H\ H*H56*H> H/*H5`*H  H*H52*H H*H5*H H*H5*H H*H5H*H Hy*H5*H H*H5*Hl H*H5n*HN H*H5*H0 H*H5J*H} Hs*H5*Hp H*H56*Hc H*H5*HV Ha*H5*HI H;*H5*H|< H*H5v*H^/ H*H5*H@" H9*H5*H" H*H5 *H H*H5*HH*H5*HH*H5*HHK*H5*HH*H5*HnH*H5*HPH*H5*H2H*H5*HHu*H5*HH*H5*HH*H5*HyHS*H5t*HlH*H5>*H~_Hw*H50*H`RHQ*H5 *HBEH+*H5L*H$8H=*H5n*H(H*H5x*HH5 *H=*HH u HCHP01H5H=sHHA H#*H5*HHuH u HCHP0H=*H5k*H=*1]H H=*H*HuHGP0H5+*H=*1%H H=e*H^*HuHGP0H5*H=d*1H H=%*H*HuHGP0H5*H=,*1H H=*H*HuHGP0H5k*H=*1}Hv H=*H*HuHGP0*H*H*H*H*H*H*xH*lH*`H޹1H*H=*H5H=HHH*H5*HHzH u HCHP0H=*1H5H=HHH*H5X*HHH u HCHP0H={*6H*1H=*HH<H5*H=*HGH HCHP0H HCH߽L%cAP0H=Z*tCH=H*tLDH=TH=3*Ht9HH *u)HGP0 HuH4(H5H81H=*AAA rE1A bI޽A RE1ЛABI޽қA2Ao%AAj 5AI $u ID$LP0L%%MIIFLP0H HCH߽L%AP0H 9޽ݛH=8L%A!SRL%A Q͛L%A:L%rA5#%L%[Ae EL%DAOL%-PL%QL% RL%SL%TL%UL%~VL%mWL%\XL%KYL%:ZL%r)[L%a\L%P]L%?^L%._L%`L% aL%bL%cL%dL%eL%nfL%]gL%LhL%;iL%s*jL%bkL%QlL%@mL%/nL%oL% pL%qL%rL%sL%tL%ruL%dvL%VwL%HL%2A4L%A% L% A L%I 4L%0A8L%A9L%A:L%A`L%AobL%AXdL%AAfL%A*qL%xAzL%aAL%JAL%3AL%A L%A țL%AL%AorL%A[L%A5DL%Aj-#L%{Ae3L%dACL%MAML%6L%AL%AL%AL%A {L%{A#dƜL%dA$ML%MA%6L%6AL%L%\AL%EAhL%.AH[]A\A]A^H=*HH9|(HtHf.H=*H*H9tH{(Ht H=*H5z*H)HHH?HHtH|(HtfD=Y*u/UH=|(Ht H=x(h1*]{f.HG8HHGHHGWl1~HGxHDf.AWAVAUATUSHL.M~HHIEătLHB1DyHD$HFH$HT$H4$HEDHII9uH[]A\A]A^A_Eu#1H;H/u0HWHR0IL9u1H HHII9u믐HII9uH[]A\A]A^A_DHu310Ht!H*HPHtz(HHPHHHz(1H5*8ATIUHSHHHt HՅu!H1Ht[LH]A\[]A\Ðf.UHSHHy(HHH]HHHt H/tCHHHHHHtH/tH1[]HGP0H1[]HGP0HGHHGHu#10HtH?y(HHPHfHy(1H5x*8HHHt H1DHx(HGHHWHtH(t 1HHPHR01HUHSHHx(HHH]HHHt H/HH} HH] HHt H/tbHH}(HH](HHtH/t5H}PHtHEPH/tH1[]DHGP0H1[]HGP0HGP0HGP0e@HGHHGHxHHxf.USHHGXHo`H$H_hHt$~D$HOh$GXHtH(tPHtHmt4HtH+tH[]f.HCHH@0H[]DHEHP0@HPHR0ff.HGE1LH@ uLGuLA1LAHL v(HLM9tMu HxHuLHxH:H@HHHtHHHtHHHtHÐHt;Ht6H9HOHVH9t(\Ht 1~\HtÐ1H91DF\f9G\u΋GX;FXuATUSH_H9^H_ H9^ tmH_(H9^(tZH^0H9_0tGH_8H9^8t4H^@H9_@t!H^HH9_HtHFPH9GP\St[]A\1^`9_`uH_LfHtcMtH;I4$Ht_HtHSI9T$u1tHcH@HHI H:H1Ht%Ht HAH9Bt1x1Mk1I<YSHGHL@t+Hs(LHōH81蛻u%H[fHs(HHH5DH81rH+t 1H[DHCH1P0fAVAUATUSHGLM|HIH趶L-s(p NH A;MhLHHAH茶H QP AEH=~9|HtTH[]A\A]A^ @9}Q@$[]A\A]A^{H=輵t1@蓶HHuHr(H5H8AVIAUATIUHSHH蓰1LLHIHtzIHA*H~%1I fDHHHHHH9u1LAE I,$tAm H[]A\A]A^IT$HD$LR0HD$1f.HGHHtfDUHHSH诳HtJHHEHHH@pPH+tH[]fHSHD$HR0HD$H[]D1ff.SHGHuGHwH{HtHCH/tHCH[H@HGP0HtGu蓲t[f.SHGHuoHH{HtHCH/t@HHtHǃH/tHCH[H@fHGP0HGP0HtGut[ÐSHH HGHOH|$HT$Ht$HH=p(H9{^H9{PH{0Hi*1H9=*JtyH;=*tgH;=*tUH;=*tCH9=*t1H;=*tH;=*t H;=*/H+H|$HT$Ht$H{HtHCH/thH{ HtHC H/tAH{(HtHC(H/tHCH@H [HGP0HGP0HGP0H{H/f.*9(H *HH4H4H<DH/HCPJHGP0>f.HG裯!fDnfD^fDNfD>fD.fDfDAWAVIAUIATUSH8HIDd$pL|$xHD$HH$AH~hIAA1IHD$(I@IHD$HAHD$ HFHD$AWLLHATLL$8LD$ HL$0Ht$(jLl$Lt$XZH9uH8[]A\A]A^A_f.HD$H~ H<$~I9tJ1H~LLd$L,$LHLH>MHLH9uH8[]A\A]A^A_ÐL9<$uH$H8HLH[]A\A]A^A_fAWMAVMAUATIUSH(L*Ht^H~CA1D$ HBHD$HFHD$L$ HT$LMHt$MHMH9uH([]A\A]A^A_H~1fLLLHNMH9uH([]A\A]A^A_AUfIATIUHSHH(HGXHT$Ht$HD$HG`GXHD$HGhHGhH|$HD$軪H{XHt$Ht H|$̨HD$HtHHD$HtHHD$Ht HHD$HT$IUI$HD$HEHH8HhHHT$HXHPHT$HPHtH/tMHtHmt1HtH+tH(1[]A\A]fHCHP0@HEHP0@HGP0H|$IEI$HEHt H/uHGP0H|$Ht H/uHGP0H|$Ht H/uHGP0H([]A\A]AWAVLvAUATUSH(Ll$`L9vLHG LL$MIHT$HHD$LːH|$LIHL$HLIL<LHHLHHLL)L9t$~Hi(HH([]A\A]A^A_@ATIUSHHHoH薭Ht9HPHHtHHHH[]A\fDHH[]A\Hj(LHD$H:!HD$f.AUATUSH^HH9=IHLo H1E11L莩HLHDHHHTHI9~H[]A\A]DAVAUAATIUSHt}H=>*HIHMHHLEHHL荬IH+t[L]A\A]A^HCHP0[L]A\A]A^f.1HHtrH=*uIHtMȬHHt@LEHHL HmItHEHP0H+fuDHmu HEHP0E1[]LA\A]A^UHSHHHt7HH}H/tH]H1[]@HGP0H]H1[]@Hqg(Df.ATIUHSHHHt HՅuAH{ Ht LՅu/H{(Ht LՅuH{P1Ht[LH]A\fD[]A\Ðf.USHHHoH$H_Ht$~D$HO$HtH(tJHtHmt.HtH+t H[]@HCHH@0H[]DHEHP0@HPHR0@H9t[HXHt/HJH~H;rt@1 @H9tt1HH9u1DHH9tHu1H;5ff(ff.H9t7HGH;`f(u>HGHuHtHe(H@HuuHUe(HH;Yd(t齤DfH*f.Gzt@ATUSHH9H.f(H9GHH9F ~ HSH9VHCHNH9@H@t H{ DF D8@ "H{HA |HN0HHA@HEȃpwDA9uEHH1u)1 L%Id(L9ut1H[]A\L9uuH߉EHHH;*d(H;c(uL9u<H+uHSD$HR0D$1H[]A\H舧fDHK0HH@HHDfHt$6Ht$>'HHt$ Ht$@HvHD@Dff.HGHHGG<4wiH bHcHfDf.f.f.f.f.HHa(@H52H81`1HfUSHHHGH;Tc(tjH;`(t1HhhHHEHt{Hy HH[]fHt HyHGtH;CsEHDHH[]DHt HyHGtH9CvHSHHH[]fHآHHt`HHEHmuHUHD$HR0HD$HUHFHt$Ht$HxHHE)fD1H`(Ht$H8舡Ht$tHt$HEHt$AWIAVIAUATUHSHHHHD$(LD$Ll$0HD$HD$8HD$(HD$0HD$8HD$@HT$Ht$LLSHHt$(H{H8HH9uf.H;2HHHuHFLIHFH9GII$Ht-H8H9uLHL$0H)I I<$KHt$(fDH9u&fDHFH9GHH9HEH8H9uHT$HH5|H^(H81聥HH[]A\A]A^A_DHT$0H)I蛣xDHt$(%;HtoxnHt$(oHH9VHHT$H5L|]HFH^(HT$H5{H81HH[]A\A]A^A_踡HHt$(DHHH@@AT1USHIHti1HH I,$Hu ID$LP0HtBHM@teHHڜf.HmuHE[H]A\H@0[]A\飜H](H5{H8Zf.H\(HH5'{H81蕣SHH$H$H$L$L$t@)$ )$0)$@)$P)$`)$p)$)$H\$ H$D$HD$HH$HL$HzD$ 0HD$ԚHܠff.AVAUATUSHHHWDl$@HjpHHEHIMtI0H[]A\A]A^DHL31HH1HO\(LgIHtH+u HCHP0MtFLLUI.ItHHL[]A\A]A^HA[(HRH5&zH81HE1[L]A\A]A^ÐIFLP0HL[]A\A]A^fDEu+L5[(=EuCH[(H5@1HL$H4$谜H4$HL$HIHoH舜IHt:H9[(LHNIHtH+t@t5H9tbHXHt4LIM~H;QtG1H;Tt;HI9uIM9u1@HDHH9tHuH;X(uθHHH?H9t}HXHtOHJH~H;rtb1 @H;ttSHH9uHW(HNH5vHWH81踝1+@HDHH9tHuH;5W(uH@H V(H5H8 1fDATIUHSHHHt HՅuQH{ Ht LՅu?H{(Ht LՅu-H{PHt LՅuHx1Ht [LH]A\[]A\Ðf.USHH"H-V(HxHEHxHtH/tMHH9t(Ht#H@@~BHǃtAHǃH1[]DHGP00˔H=Rt1HHtHǃH/uHGP0fUHSHHH=e*HGHHtHHt#HH[]f[HfDHV(HH56H81襛UHSHHGHHtHHtHH[]HfDH V(H8tHU(HH5ܟH812HG@HFH@t{H9t^HXHt2HJH~!H;rtC1H;tt1HH9u1DHH9tHu1H;5T(ff. t|ff.SHH HGH进HT$Ht$H|$[HHPHHH;T(HHB@HǃHǃHHH|$HT$HHt$HxHtHǃxH/tH HH [DHGP0HǃHG+뮐0hH=2q1HH:HǃH/%HGP0HHHOD_@GDHt?H1HLHEH9t8HQLNHLSHH( }w%IcLAGE=H }vу0 1CHK0p@ wDHTPp@ vHcHIW(|E1LeIG(HIG0AGEAGFCEo@AGDIG(9fDE1A8ODuE9o@uAGEA8GFuAGuIG(HIG0IG(fDKHSA<v dHADIG(IG({{HD$IG8HD$LfHCH|$AGDA)G0HD$JE1Ht$L8HHIL9l$uHD$H>IG85f.HAOE{:HCtfDH8:uHXfDI(}Lt`IGKHs1HLESX) VLHHuH5)fHD(H81蠋1fADtItLJtIL1LHIo8tAGDH}IO 1HHHtlHH)Io `L_IG(IG HAGEIG(IG0AGDAGFHD(H5d1H8XH(H[]A\A]A^A_fDHC(H5d1H8fDу0 HFvJЍVЀ wfDHLV0VЀ vtMA9~HcHcITH9,t ),ƒHH f A9uAGGH^IG(HB(DH5dH81萉Hb1ZH5<`HB(H5BdH81PHHwB(H5cH81.H]B(H56c1H8贂WH@B(H5Ad1H8藂:fUSHHHGHt)HkHUHH HcHHHH+tHH[]DHCHP0HH[]@kH+HcukCHH H+uH+kuDkCHH HH+uH8H+HlvfDH@`HtOHHtCHHt9H@H;FB(HH5HHu3HD裄HuH@(H5fH8H@f.AUATIUHSH趃HHEHHhID$HH觃L-@(p VP A;U1HLH{H QP AEH=~D9|JHH+tHH[]A\A]HCHP0HH[]A\A] @9}@$H1[H]A\A]1HL#HfDH=9Z\7@1bf+HHuH>(H5ZH81AWAVAUATUSHH8HGHsHT$(Ht$ H|$HC`HHH{H{8跀H+H|$HT$(Ht$ |H{PHtHCPH/tHH{XHtHCXH/t!HCH@H8[]A\A]A^A_DHGP0HGP0SheClLsuL{MDHS8H*H~K0Lc@E1tRID$DyH$HBHD$H$Ht$LE1DI,M4$L9uLsDIM4$L9tI>H/uHWIR0M4$L9u@H@t GpH#`fDATUSHHGHPhH7HJH*HHH=m>(H9~#HVHrHweHuHtsH;>(^H;w<(1HhhHHMHHHH[]A\fHuEHH fHHHCH;P>(t6H;;(uHCHH0HH9stHDHH[]A\fHSHH2H9vJHSHHH[]A\DHPH <(H5.^H81贂H1[]A\HfH}HHHHA|HmkHUHD$HR0HD$RDHSH]@SHPHCH;G=(H;:(HhhHfHMHH[HEHhHt$HHt$HSHHMDf.H8IHdHT}I,$HIHD$ID$LP0Ht$/@HCHHPfHtH}Hf.uUHH kfDHCHH;!<(tw1H;9(rHCH1H:(HH2hzt,|HEH5\\HPH;(H81诀1HSHHS11HD:(Ht$H8zHt$tHt$L|HMHt$@f.SHHGt~HGHtLHE1HHHu HCHP0HtHmuHEHP0@I fHIA AIGLP0H fDDH=A>HkE1fDLaIHH$\I/IIwLV0HHAFAVHH IDE1I/11AA*H f H=p=fDH1L_HHI.AA11IFLP0H=i3[fk\HuH(H5`3H8XfDH Ne2AH=<H+fD_HH eH=s<E1cH dH=N<ً0)H=61 IE`L8ZIAFAVHH HII.uAAH jdH=;f.HH %k*HWHH9HXHLFM~(H;N1fDH;LHI9u11lpxppphp`pXpPpHp@p8p0p(p ppp0'HHtCHfDHH9tHuH; x(NfHH=DMH c͆HD$H=:HD$Hff.AUATUSHHi*L-xZ*HCHHYL%(p VP A;$1HLHXH QP A$H=~\9|bHH߽]x袷H+t,H EbH=:H1[]A\A]DHCHP0@ @9}qX@$H1L[HHuYxfH=/W'f.XHuH(H5/H8U@f.AUATUSHHg*L- Y*HCHHWL%(p VP A;$1HLHWH QP A$H=~\9|bHH߽%xBH+t,H `H=8H1[]A\A]DHCHP0@ @9}W@$H1L;ZHHu!xfH=A.dV'f.;WHuH(H50.H8S@f.AVAUIHcATIUSHdUHIƅH-q(HE*VHHLl$~D$Lt$IEHk(H-f*D$CHELM1UL-(H QP A;UR1HHAHUp VP AEH= @9H1H+u HCHP0H}Lu HEHP0HH[]A\A]A^fH-(HEI.tRHmu HEHP0xH ^1H=6`HH[]A\A]A^T@$HfIFLP0HmuDx륐H1HXHHH+xuHCHP0ffH=+Tf.THuH_(H5+H8`Q@f.ATUSHHGH5]`*HH=HHHEH5_*HHHAHHEHH HEH~HCH5[*HHH0HHHHHHtSiSHH HhH="a*HWHH+t+H[]A\fHEHP0sHCHP0@HSHD$HR0HD$H[]A\AvDH \hH=4?H1[]A\@VHAvHEHuHEHP0DVHAvHHuHCHP0y{VHHEAvHHAvH뷐UHSHHHGHtAHKHHHQ0HtTH(tH(HH[]@HPHR0@oH y[H=3+H1[]fo@AVAUATUHSHHLfHD$HuDIH^HH}H/tH]1H[]A\A]A^DHGP0IMt0IuZHFHHD$KIMH\$fDHxKH5X*LIHVIQHD$HuLcHH (HZH53ATL YAH `ZH81'TX?fZH ;ZH=DZHT$LLL8ZH55);4f뱐HHGHt H@H YcHD$H=2vHD$Df.AUATUSHH_*L-pQ*HCHHOL% (p VP A;$1HLHOH QP A$H=~\9|bHH߽MebH+t,H YH=2H1[]A\A]DHCHP0@ @9}1O@$H1L[RHHuIefH=a&N'f.[OHuH (H5P&H8K@f.AUATUSHH^*L-P*HCHHNL%x (p VP A;$1HLHSNH QP A$H=~\9|bHH߽eH+t,H WH=0RH1[]A\A]DHCHP0@ @9}M@$H1LPHHuefH=%$M'f.MHuHo (H5$H8pJ@f.UHSHHGH5U*HHHHHEtYHCHHHHtIH HQHHtBHt H[]fDHSHD$HR0HD$H[]DHHPRdHu HCHP0H VH=/H1[]ÐPdf[PH,f.ATIUHS!LHHt H[]A\ÐH9 (H8IthK[LH]A\fATUSHGHHIHH5T*HHHHLHEHx7HP1HHt[]A\f.HCHP0[]A\HdHu HCHP0H TH=^.[]A\f.OHVdHPH(H5G.H81XNRUHSHHGH5S*HHHHHCHHH@pHtKH@HtBH HQHHtAHt H[]DHSHD$HR0HD$H[]DfdHu HCHP0H SH=-bH1[]fMH<dfAUATIUSHHHxIAD$lH-(HEzIHHLd$~D$Ll$I$Hk(H-Y*D$CHELMLIL%E(p VP A;$21HHAHIH QP A$H= @9HH+u HCHP0HH[]A\A]H-a(HEImtQHmu HEHP0cH $R1H=,,HH[]A\A]ÐkH@$hfIELP0HmuDc릐H1HsKHH6H+cvHCHP0gfH=aGf.[HHuH(H5PH8D@f.AVAUIATIUHSH H(H^HD$HT$M)HHHH)QH QHOH7RL 3PLOL@HH(SHPH5Z)H81JX>ZH P H=*A1H []A\A]A^fHV I|$HHHu E1H&J*AHH S*jQPjQHPj5bM*WH= (LW*HPHH+uHSHD$HR0HD$H []A\A]A^HHHL@H5L*LIHVIFHD$H MH|$HT$H+t2 H O>HD$H=)HD$fDHSHD$HR0HD$f.oFL)D$?H{HT$HLL9OH5)סUp><DHFLHD$?I$H5!J*LHVEHtHD$IF|I\$AVAUIATIUHSH HD(H^HD$HT$M)HHHHiNH YNHOHwOL sMLOL@HH!(SHRH5&H81GX];ZH "N H=6(1H []A\A]A^fHV I|$HHHu E1HfG*AHH P*jQPjQHPj5J*WH=2(T*HPHH+uHSHD$HR0HD$H []A\A]A^HHHL=H54J*LIHVI!DHD$H MH|$HT$H+t2+ H L;HD$H='PHD$fDHSHD$HR0HD$f.oFL)D$=H{HT$HLLPH5)UM;<DHFLHD$HPHH>HCHP0HH[]A\A]fDHV HFkHHt'H G ~:H=!DHCHP0@oFL)$7HHLPGHLH5)ʙD:nHFLH$7IMH5B*LHV=HtHD$IE낐H5F*LHV=HtH$I@f.AVfAUIATIUHSH0H'H^)D$HT$ MHHtyHH 1H='DHCHP0@oFH)$o.H~HL>HLH5)nv1q@HFHH$(.IMLH58*LHVy4HtHD$IFfH58*LHVM4HtH$I@f.AVAUATIUSH0H6*H 6*H-'H^HD$~D$HL$Hl$ D$)D$HpHV~$HHHV(HN LNfIHHHI$It$ AH='HUj5?*Pj57*QHj5:*SC*HHHPHHPHHtH0H[]A\A]A^@HCHP0H0H[]A\A]A^fDHDHH;IH ;H1<HIHH'I?SIH57H8L <A1R5X1ZH ;, H=1H0H[]A\A]A^ÐHIHcHtHMHF(HD$ HF LHD$HFHD$+IHHM~.H5?6*LHV1H HD$ IMLL$HL$HT$ Hb4*$DHH?+IM~H589*LHV1HtHD$IM~H55*LHVn1HWHD$IDHHt'H P: >1H=w誽DHCHP0@HVfHFHHD$*IaHT$HLL!:H5)y1 AVAUATIUSH0H23*H 3*H-,'H^HD$~D$HL$Hl$ D$)D$HpHV~$HHHV(HN LNfIHHHI$It$ AH='HUj5<*Pj54*QHj57*?*HHHPHHPHHtH0H[]A\A]A^@HCHP0H0H[]A\A]A^fDHDHHt8IH a8H8HIHH>'I?SIH5H8L ^9A11X0ZH 48H=1葻H0H[]A\A]A^ÐHIHcHtHMHF(HD$ HF LHD$HFHD$:(IHHM~.H52*LHV{.H HD$ IMLL$HL$HT$ H0*$DHH'IM~H5(6*LHV.HtHD$IM~H52*LHV-HWHD$IDHHt'H 6' 0H='*DHCHP0@HVfHFHHD$&IaHT$HLL6H5N)l0 AVAUATIUSH0H/*H /*H-'H^HD$~D$HL$Hl$ D$)D$HpHV~$HHHV(HN LNfIHHHI$It$ AH='HUj58*Pj51*QHj5a4*S<*HHHPHHPHHtH0H[]A\A]A^@HCHP0H0H[]A\A]A^fDHDHH4IH 4HD5HIHH'I?SIH57 H8L 5A1R.X/ZH 4eH=@1H0H[]A\A]A^ÐHIHcHtHMHF(HD$ HF LHD$HFHD$$IHHM~.H5?/*LHV*H HD$ IMLL$HL$HT$ Hb-*$DHH?$IM~H52*LHV*HtHD$IM~H5/*LHVn*HWHD$IDHHt'H P30H=誶DHCHP0@HVfHFHHD$#IaHT$HLL43H5)y/ AVAUATIUSH0H2,*H ,*H-,'H^HD$~D$HL$Hl$ D$)D$HpHV~$HHHV(HN LNfIHHHI$It$ AH=+'HUj55*Pj5-*QHj50*8*HHHPHHPHHtH0H[]A\A]A^@HCHP0H0H[]A\A]A^fDHDHHt1IH a1H1HIHH>'I?SIH5 H8L ^2A1*XV/ZH 41H= 1葴H0H[]A\A]A^ÐHIHcHtHMHF(HD$ HF LHD$HFHD$:!IHHM~.H5+*LHV{'H HD$ IMLL$HL$HT$ H)*$DHH IM~H5(/*LHV'HtHD$IM~H5+*LHV&HWHD$IDHHt'H /`/H= *DHCHP0@HVfHFHHD$IaHT$HLL/H5)B/ AWAVAUATIUHSH(L5'H^HD$Lt$HJH0HHV ML$HHHu H_(*H='AHH 1*AVjQPjQHPj51*5*H;HWHPHH=HHSHD$HR0HD$H([]A\A]A^A_I\$@HH<.H ,.HOHJ/L F-LOL@HH'SHr.H5mH81'X.ZH -H= T1H([]A\A]A^A_LIHHHHHH50*LIHVIn$HD$HMLL$HT$kHt3 H ?-.HD$H=& 虰HD$=HSHD$HR0HD$foFH)D$nH~HT$HLL1-H5\)gX.DHFHHD$I'H5'*LHVm#HtHD$IG|f.AWAVAUATIUHSH(L5'H^HD$Lt$HJH0HHV ML$HHHu H%*H= 'AHH .*AVjQPjQHPj5.*2*H;HWHPHH=HHSHD$HR0HD$H([]A\A]A^A_I\$@HH\+H L+HOHj,L f*LOL@HH'SH+H5H81$X=.ZH +HH=)t1H([]A\A]A^A_LIHHHHH8H5-*LIHVI!HD$HMLL$HT$kHt3H _*f.HD$H=n蹭HD$=HSHD$HR0HD$foFH)D$H~HT$HLLW*H5\)|X-.DHFHHD$?I'H5$*LHV HtHD$IG|f.AWAVAUATIUHSH(L5'H^HD$Lt$HJH0HHV ML$HHHu H"*H='AHH ,*AVjQPjQHPj5+*/*H;HWHPHH=HHSHD$HR0HD$H([]A\A]A^A_I\$@HH|(H l(HOH)L 'LOL@HH4'SH(H5H81!X-ZH 5(H=y蔫1H([]A\A]A^A_LIHHHHHXH5**LIHVIHD$HMLL$HT$kHt3CH '-HD$H=٪HD$=HSHD$HR0HD$foFH)D$H~HT$HLL'H5\)yX-DHFHHD$_I'H5!*LHVHtHD$IG|f.AWfAVAUATIUHSH8L5'H^)D$Lt$ H:H(HHV(ID$ ML$HHHu AH='HHAVj5+)*5*j5M%*Pj54$*,*HPHH+HSHD$HR0HD$H8[]A\A]A^A_I\$HH%H %HMHSH&L@H]'H5L &H81X5-ZH _%H=辨1H8[]A\A]A^A_DLIHS~qHtHYHF(HD$ ID$ LHD$ID$HD$lIHtlHHt8MLL$HD$HT$ oHHIH5"*LIHVuHD$HH5#*LHVTHD$HIM~H5n*LHV*Ht~HD$ ISH+t2H $^-HD$H=}hHD$fDHSHD$HR0HD$fHFHHD$?I@HT$HLL$H5κ)9v#-!HHT'H Z#H5jL v$AH#H81Y^-fAWAVAUATIUHSH(L5'H^HD$Lt$HJH0HHV ML$HHHu H?*H=x'AHH %*AVjQPjQHPj5#*)*H;HWHPHH=HHSHD$HR0HD$H([]A\A]A^A_I\$@HH<"H ,"HOHJ#L F!LOL@HH'SH"H5mH81X,ZH !5H=T1H([]A\A]A^A_LIHHHHHH5"*LIHVInHD$HMLL$HT$kHt3H ?!,HD$H=虤HD$=HSHD$HR0HD$foFH)D$nH~HT$HLLO!H5̷)gsX,DHFHHD$I'H5*LHVmHtHD$IG|f.AVAUATUHSH L%'H^Ld$H8HHHVHHHu E1L *H=i'HH#*HATjPAQjPAQjP&*HPHH+HSHD$HR0HD$H []A\A]A^fDHHH tHIHHH?L HLIL@HH8'SHH5H81X",ZH 9H= 蘢1H []A\A]A^fLIHHJHFHHD$YHbHT$f.H+t22H K,HD$H=}HD$hfDHSHD$HR0HD$HT$HLLH5I)px,fHIHXH5M*LHV HtHD$IF-fDAWfAVAUATIUHSH8L5q'H^)D$Lt$ H:H(HHV(ID$ ML$HHHu AH='HHAVj5 *5*j5*Pj5*F$*HPHH+HSHD$HR0HD$H8[]A\A]A^A_I\$HHH HMHSHL@H'H5>L H81]X+ZH H=1H8[]A\A]A^A_DLIHS~qHtHYHF(HD$ ID$ LHD$ID$HD$ IHtlHHt8MLL$HD$HT$ oHH IH5E*LIHVHD$HH5p*LHVHD$HIM~H5*LHVHt~HD$ ISH+t2H n+HD$H=uȞHD$fDHSHD$HR0HD$fHFHHD$ I@HT$HLLH5α)m+!HH'H H5.jL AHOH81>Y^+fAWAVAUATIUHSH(L5'H^HD$Lt$HJH0HHV ML$HHHu H*H='AHH /*AVjQPjQHPj5R* *H;HWHPHH=HHSHD$HR0HD$H([]A\A]A^A_I\$@HHH HOHL LOL@HHT'SHH5H81X+ZH U`H=贜1H([]A\A]A^A_LIHHHHHx H5A*LIHVIHD$HMLL$HT$kHt3H A+HD$H=HD$=HSHD$HR0HD$foFH)D$H~HT$HLLH5̮)jX+DHFHHD$I'H5*LHVHtHD$IG|f.AWfAVAUATIUSHH8L51'Hn)D$HD$ Lt$(HqHWHHV0HK(HC LKHI$It$ AH=O'HAVj5*QHj5]*Pj5D**HPHH+HSHD$HR0HD$H8[]A\A]A^A_@LPIH5*LIHV HD$HHk@HHH sHMHUH$L@HE'H5L pH81X*ZH GH=覙1H8[]A\A]A^A_DLHfIH&HcHHF0HD$(HC(HD$ HC LHD$HCHD$DIH ~2H Hu+MLL$HD$HL$ HT$(&HM~HT$HLLH5̫)gy*fH+t2[H .*HD$H=舘HD$fDHSHD$HR0HD$fH5*LHV H_HD$(ILfDHFLHD$/IH5*LHV HD$HtoIH5$*LHVh HD$ Ht IHH2'H 8H5jL TAHH81 Y^v*HH'AH5hjL H H8H1w _p*AXf.AWfAVAUATIUHSH8L5A'H^)D$Lt$ H:H(HHV(ID$ ML$HHHu AH='HHAVj5[*5 *j5u*Pj5\**HPHH+HSHD$HR0HD$H8[]A\A]A^A_I\$HHH HMHSHwL@H'H5L H81- X)ZH H=31H8[]A\A]A^A_DLIHS~qHtHYHF(HD$ ID$ LHD$ID$HD$IHtlHHt8MLL$HD$HT$ oHHOIH5*LIHVHD$HH5*LHVHD$HIM~H5 *LHVZHt~HD$ ISH+t2 H >#*HD$H=蘔HD$fDHSHD$HR0HD$fHFHHD$oI@HT$HLLH5)ic)!HH'H H5jL AH?H81 Y^)fAWAVAUATIUHSH(H *L5'H$H^HD$Lt$HWHEH#HHHH HOHL LOLDHH'SHeH5%H81K Xk)ZH lH=y1 H(H[]A\A]A^A_LML$HHHu AH='HHAVj5[*5*j5 *Pj5 **HHHPH[HPHHqHCHP0b@HV(ID$ mf.LIH[HtHHF(HD$ID$ LHD$ID$H$IHHM~.H5 *LHVRHHD$IML $HD$HT$@HHIH5u *LIHVH$HM~H5| *LHVHbHD$IOfHHt'H )H=w DHCHP0@H'VfHFHH$I`HLHLH5q)_X)vI\$AVAUATIUSH0H*H {*H-'H^HD$~D$HL$Hl$ D$)D$HpHV~$HHHV(HN LNfIHHHI$It$ AHH=*UjWPj5*QHjWH='6*HHHPHHPHHtH0H[]A\A]A^HCHP0H0H[]A\A]A^fDHDHH IH HHIHH'I?SIH5H8L A12X'ZH H=1H0H[]A\A]A^ÐHIHcHtHMHF(HD$ HF LHD$HFHD$IHHM~.H5*LHVH HD$ IMLL$HL$HT$ HB*$DHHIM~H5 *LHVtHtHD$IM~H5*LHVNHWHD$IDHHt'H 0  'H=芍DHCHP0@HVfHFHHD$_IaHT$HLLH5)Y\' AVAUATUHSHL *L%'H^L $Ld$H=H+HHHH4 IH ! HHIHH'I?SIH5wH8L  A1XZH H=1QHH[]A\A]A^ÐLHHG*HHu AH='HH *ATjQPjQHPj5*l*HHHPHHPHHuHCHP0HH[]A\A]A^HV LNnLIHHtmHHIHL $HT$%fHHt'H /H=:DHCHP0@oFH)$H~HLL HLH53)Zvq@HFHH$IMLH5m*LHV)HtHD$IFfH5*LHVHtH$I@f.AWfAVAUATIUHSH8L5a'H^)D$Lt$ H:H(HHV(ID$ ML$HHHu AH='HHAVj5{ *5)j5*Pj5T *6 *HPHH+HSHD$HR0HD$H8[]A\A]A^A_I\$HHH HMHSHL@H'H5.L H81MXZH H=1H8[]A\A]A^A_DLIHS~qHtHYHF(HD$ ID$ LHD$ID$HD$IHtlHHt8MLL$HD$HT$ oHHoIH5*LIHVHD$HH5 *LHVHD$HIM~H5)LHVzHt~HD$ ISH+t2H ^9HD$H=踇HD$fDHSHD$HR0HD$fHFHHD$I@HT$HLLH5~)V!HH'H H5jL AHaH81.Y^fAWAVAUIATUSHHHH*HHH*HEHH=)MHHHmHCH-R*HHHHIMI<$H+IEH5*LHH HHHCH5)HHHHHHHHHfL=*HP'HLH9EHHWHHmu HEHP0(IHI^IELH5J*HHHHHCH5)HHHIMH+HHLd$~D$Lt$I$Lx(D$@I,$u ID$LP0HH[]A\A]A^A_f.HEHP0HCHP0PID$LP0H+:fHCHP0HCHP0;AiAHfDHmu"E1HEHP0MtI/u IGLP0H DD1H=MfDH qMH=˃fDE1ARAl@Hq'H8)HAAUHHH[HCHP0L IH d1H=N1>@HAAff.HHH,AiAfDHAlAH2IGLHM'H8}H+HLAAq@II.tHAsAHIFLAAsP0HHHI.LAAvf.H'HH5AAUH81HH1HHEHH3'H8cH+)HCHP0I/uAvA"H vH=C&3UHSHHGH5)HHHHHCH5K)HHHHH HQHx-HtHѰ'HH[]fDHCHP0@ Ht&H H=lH1[]HCHP0@HL f;[fDUSHHGH5)HHHHHCH5N)HHHH HQHHt2Ht H[]fDHSHD$HR0HD$H[]DHu HCHP0H H=bH1[]fHIfUfDAUATUHSHHGH5`)HHhHHjHCH5o)HHH|IHHHMtH{HEH5p)HHHHHHEH5)HHHHHEHHHEHHCH5)HHHHHHHHHH=)H0HHEHxHHEH5s)HHHHHHu HCHP0HLHHEH.HHEu HEHP0I,$u ID$LP0HIHPHHHu HCHP0HL[]A\A]HCHP0vHEHP0HCHP0HEHP0HxHHAHCHP0H TDH=`|MiI$LE1HD;HH ^E1H= `|HL[]A\A]fI|`AHWH `H=|f.Hkm޺LH gH=x{I$E1HhcHOoHEHuHEHP0fD3H`rfDHuDH{H H=3{@f.AUATUSHHHH1HHLHcClM,M9sLI<$5HHtmHEH;E HHUHHHEH+u HCHP0IM9wHHHmuPHUHD$HR0HD$:@sHm@HEHP0<H |H= ,z1H[]A\A]HH=ZHmu HEHP0H+uHCH߻sP0<s<@HY)L-)HCHHIL%B'p VP A;$1HLHH QP A$H=~v9~HHJH+t7s:s<<fDHCH߻sP0: @{@$tfs[H1LHHXs:MH=HuH'H5H8fDAUIATIUSHH(HGH;'H;='HHI$L`IELh HCHLML-'p VP A;Ux1HHAI]H QP AEH=9MaHmH+tpH(L[]A\A]fHWBፁ)Ht$~D$1Ll$D$)D$LB uH{Ht$AIfDHCHP0H(L[]A\A]HEHP0H+\f. @9(]@$@Ht$~D$Ht$HT$D$)D$KIE11HHCIH=Y|t@E1KIHuH'H5=H8w1ɺAIvfDAVAUATUSHrL%'HHLfI$t Gh1tHExHC01tHHC81tHHC@1tHEpHC(HEHHElC$HE`HCHEXHCEhC HEH{H/tHkL9t!1[]A\A]A^fDHGP0HkL9uHmtHCHEHP0HC@H-q)L5)HELMaL%Z'H QP A;$1HLAH4H QP A$H=~}9HHA}qEHmtGH=H {D,tH{HH/t4HCfHEHP0@ @xDHGP0{@$]H1LHHMAyqWH=HuH''H5AyqH8"HV'H5WH8f.SHX)LOL9IXHLAM~&H;A`1@H;DMHI9uHWxHLH>HGHHFlM;LL9H9AD L9AH9AE D!M9AL9AE DGo )No)VPAo)tjob)f oi)n`Aop)tFoz )~0oy )~pAo` )t"oB0)F@oI0)AoP0)9tHJHcHpyH;0utHߗ'HHİ[f.HA'HHİ[@nH vHD$H=gHD$Hİ[+f.AVAUATIUSH HGH5f)HHHHHCH5)HHH HHHHHHYHEH5)HHHHHEHHHEH(H=y)Ld$HGH;!'H;ܗ'HWBHjE1 uLwKL-D'H QP A;UQLLH!H QP AE=@9HRIHlH\$~D$HHl$H=z)D$@HtoI,$uIT$HD$LR0HD$H []A\A]A^fDHCHP0HEHP029ZP@$LI,$uIID$LAdA?vP0DDH H=FYeH 1[]A\A]A^f.AdA?vfH=Ad@HA-vAeHHHu1HCHP0HvHEHHEHaHEHP0R@H+AdA7vu@LXSHHJoAdAvHAvAdBHAdA"v?HHt$I9H\HH'H5 H8LJE1 uLGHt$uLAH1ɺLAHAWAVAUATUSHHL-'HHL5)LnLgPIEM9L=.'M9|$M9~„_WA|$ A~ IT$I;VLePL5)M9M9|$M9~„A|$ A~ IT$I;VID$IvH9@H@t HA|$ EF D@@8v@ It$0IH@IDHA MN0IvHA@IEDAA DDE9HL$ E1HL$ AfM9L5)M9MM9~LLL$ IHH;'L$ AH;$'DM9L?I,$L$ AuID$L$ LP0L$ EUE6%M9uLLL$ /IHH;'L$ AH;'Du M9I,$EuID$L$ LP0L$ EEE%HEHHE HCE0C$HE8HC0HE@HC@HC8HEHC HC1tHE(HC(HEH{H/uHGP0HkL9t1H[]A\A]A^A_fHmHCID$IvH9@H@t HA|$ EF D@@8@ I|$HA cMN0IvHA@IEDAUADDE9KHL$ E1HL$ AHEHP0HCfM9uuM9LDZH-a)L5)HELMSQL%J'H QP A;$\1HLAH$H QP A$H=9aH3H/HmHEHP07bH=*H U^H{HH/uHGP0HCH[]A\A]A^A_LI,$L$ A(7b @<DIt$0IH@IDHI,$EL+@LL$ ,L$ 5aLL$ L$ fDIvHDD@$fLL$ L$  bLL$ L$ fDDD>IvHbI|$H7fDDD\H1L6HH3b H=<_@HuH'H55H8DDHߊ'H5H8fDAWAVIAUIATLUHSHHHD$XHt$pHD$`HD$hNHI~`HHHD$0L$@AIFHLLP0HH(NIHXIcVlHH9r<@HH9+H8xH~)L-)HCLMnH-g'p VP ;U%1HLAHBH QP EH=89>HH߽z+H+H AnH=m+ZfHT$`HD$XHHHt$XHD$`HD$hHHLpH@LhH $HL$hf~$H|$0L,$$)$eHfo$H8LxLhLpHt H/Mt I/iMt ImIHD$XH{XLs`LkhHCXHD$`HC`HD$hHChHt H/Mt I.Mt ImHD$XHD$`HD$hH $DH=>X1f.H-و'H}IN`AL{HL$ AMlD$E1 HiL5'H QP A;xL1IaH QP AH=9M^HHmu HEHP0H=s)H5T)HGHHHHbH=)Ll$HGH;D$H;8'HWBHZE1 uLL5'p VP A;^LLI~p VP A=29M[HEH;`~'L}MLuIIHmu HEHP0LLL I/Hu IGLP0Imu IELP0H]I.u IFLP0HLRHI$HRHPI$Hu ID$LP0H+u HCHP0H(H[]A\A]A^A_Ll$H;D$H;'CHUBHZE1 0L5)~'p VP A;'LLHH QP AH=9H I@@9f@$X@Lc @9S@$E@Le29@`@$2L} @9M-@$?@xL MhHHz'H5YH8ٻ@Ht$Q!IfLH:II/qI$LAHI$uHI$AH+sHCHP0dH=贽LH:H!DsHHy'H5dH8 @D賽@$f.LJE1 uLGHt$ LAId11HH IQfHt$HIfDLJ1 uH}Ht$AIkDH=Y|tSHHx'H5DAH8边]fHt$H.HfDH=@I/_E1DfD˼HuH?x'H5H8@fDLB1 uH}Ht$AHDH=Qt@1KHHuHw'H5=H8轸H1HHؾIH=Q{1ɺLAIU1ɺAI1ɺAHbID$LAP0H+T@f.AVAUATUHSHHLfHD$H8IH^HH}H/H]HCHH5)HH6IM8H5/L)QH5LIHAoEH5!)E AoMM0IE HE@HCHHHHHHH/t1H1I,$t1H[]A\A]A^fHGP0@HGP0ID$LP0H[]A\A]A^fDIMtIHFHHD$IMH\$kItH 4mH=cF8fH)L5)HCLM H-v'H QP ;U1HLAHݸH QP EH=S9[HHH+w>@ Hy fDHhH9zH /H=hE1fH舲H5A)LIHVI޸HD$HzLc@HHt'HH5ATL AH hH81/X?ZH qH=DfDHCHP0w! @D[@$H1L腺HHwH=讶&菷HuHs'H5H8wHT$LLLH5L[)GB4DAUIHATIUHSHHWH=o)ZHH`)H@IEI$HtHHH[]A\A]DӶHuH=)HHGHHt'HHuHst'HH5H81HAUATUSHH}訰HXXHHu1H[]A\A]Hs'H0H9fLm`LehHEhEXH+tgMtImtkMtI,$uID$LP0fH/uHGHt$P0Ht$Hq'HH5H81许Z@HCHP0@IELP0HCte@t\HFHt\@tSHXHtXHJH~"H;r1 H;tHH9uH蚱tHHHH9Hu1H;5.r'fAWAVAUIATUSHHHIEHIEH5)HCHFIELHHHHL=Eq'L9袳IHHEHHHhIHI,$H+HZ)H@H9)YHڲ)HHHƲ)HIHoIEHIEID$H:)HID$ M|$(I۲IHH$~$LL$$$@ILp(H+LHtHmu HUHR0HL[]A\A]A^A_DIohIGhfMw`AGXI,$u ID$LP0MtI.u IFLP0HtHmu HEHP0L=o'IH)ImH@L9H9)*Hs)HZHL=_)MıIHIEHIEID$H)HID$ HI\$(脱IHL<$~$L$$$@H+HSHR0f.I$$;gA HPI$HH+MtImu IULR04$H DLH=E1O>+f.HCHP0`ID$LP0H+tIf.H+$0gE1A fDHSHR0]$fE1AI$IHPI$H41IT$LR0HH+fDL8HKIHAn'MgXL0M9MEIFH>IT$A$@A@I$XHHJHL;r1@L;tHH9ufE1H H=ٚL|$0H|$8H/uHGP0I,$Ll$0HD$8u ID$LP0HD$0HtH+u HCHP0H$HtHH$HHu HCHP0HtHmu HEHP0IUMHBMI$I$Ht5HHL[]A\A]A^A_HH9tHuH; 8e'ID$LP0HpHhH4$HHId'H29@$AlHD$(H|$0Ht H/uHGP0HD$0H|$8Ht H/uHGP0L=)IUXHD$8I9HIGHVHJ@A@HXHHQH~"L;yt1 L;|eHH9uAIH\$~D$H8HP$LhHhHtH/uHGH$P0H$HtH*u HBHP0MtImu IELP0H|$(Ht H/uHGP0H|$0Ht H/uHGP0H|$8Ht H/uHGP0DDH ٮE1H=2I$HPIOHIwH9tb1 fDI;TtQHH9uE1HHL$LL$HT$ u*LL$HL$HT$IL9KtH9u@H ;DH=B1HL$(HT$8LHt$0L5a'H|$(IH/uHGP0HD$(H|$0H/uHGP0HD$0H|$8H/uHGP0IH\$HD$8~D$H8Lx$LhHhHt H/uHGP0MtI/u IGLP0Mt&ImuIELP0I$HP I$HPAlH|$(HH/xHGP0lfD1LH蓦HD$0H AmDH=z輢HL$HD$0Amz@{HuH^'H5pzH8fDA"mALH赟HI9uHI9Hu1L;=B`'f.AWAVIAUATIUSH(10HHH-_'H)fHCP)$HCHkHk Hk(Ml$HEHD$MII}IHH HMHAUHOL@HQ^'H5҃L |H81XfhZH YH=7.H+(16ID$(HD$AoL$I|$ Ml$) $HGHGHHHHcHfDE1DH|$HH;=s^'AH;=]'D|H9sAƃZIEH{H/uHGP0LkH)DH9CI9)(HS0H?ADHC8HS)Hʉq)HS0H蟠HC0H-)iiH ~H=0,yHsHDLH{PFHkPHE8HSp1:Ou 1z?HI\'1H58)8H.fIV~dItIwID$(HD$ID$ LHD$ID$H$IIt^It}Mt/MBL,$H|$@ML?IH5)LIHV蕞H$HH5)LHVuHD$HIM~H5_)LHVKHHD$IY@E1˝H_hHCH1P0h]fID$LH$OI/HUXHGf.HLaLLH5yA)$PhfH@`HHHIHHHL%JZ'L9IFHHHHcHEfAf.I.IFLP0AFAVHH HHcAH9tH Y'H5,{H8ԘI.IFLP0EfAFAVHH HcAH9wI.ALQ0zf.H5HIHgHHLԛHcAH9HA薛H3DHHW'H H5'}jL ϥAH|H817Y^FhAMl$,HHTW'H5H8蝗HA`Ht_HHtSLIHtFL9`u$Lh!I/A;IGLP0,H5~HIHu\蛚HNHV'H5ZH8 3HHt(Hs)HǀHPH5W'HHxHDHGtkHGHHHwUH5.HcHDGHËGWHH fG@GWHH HۙUSHH@`H"HHHHH@H;V'fDHCHHHH5~HcHHH5RHHHfH+uHSHD$HR0HD$H[]CHCSHH HfDCCSHH HH5xHHt-H@)fH記c~HHcH@`HHHHHHHU'H9EHEtHEHPHwkH PHcHHHmHUHD$HR0HD$EHӋEUHH HE뼋EUHH H輗蕗HfHS'H5TH8KHS'H59H8fDAWAVAUATUSHH+T'H9bLnHIIEHH/eLmID$HZH2H5q)HFPHEHHHqHVH(lHEH5))HHHfHHhHEH5)HHHZIHEHHEMH'Md$ I$IEH;Q'MuMImIHEIm'LLHyI.ILI,$Mt7HmI/HHH[]A\A]A^A_HGP0HmHEHP0xH H=GB"H1[]A\A]A^A_ÐHbHPHR0fID$LP0@HEHP0IGLP0;HEHP0I/&f.IELP0xH;J'9HUBbLrE1 uL}nHgI'H QP ;LLAIEH QP =/29tMH諌HIH~aHtHHF(HD$ HC LHD$HCHD$IHt^Ht}Ht.M9H|$Ld$7fHH߅IH5)LIHV5HD$HH5)LHVHD$HCIH5K)LHVHD$ HI^@DHHHH fH=FwP)f諊@$~f蛊@$&fH ÐH=vfDcHLH]I*DHAHH(E1HCHP0H DH=vPMfHLIzDHHHF'H5HH81% H 6H=vI$HP1{f{IHG'H8AQImu IELP0H ӒH=u{CfDffDfDѐAHFHHD$?IfDMt$MI\$IHI,$u ID$LP0HLH腝I.IIFLP0 fHt$HI+fDHt$LIfDLB1 uH}Ht$mAIDLB1 uI|$Ht$FAIQ@H=_4F@HEHAѐHLDHuHgC'H5^H8hfDH=^Ԇ@E1談IHuHC'H5^H8HE'HH5H81]H@`H&HHHHHPH-D'H9HkHMH H5jHcHH@`HEHH5LIH$H9hL! I.Hu IFLP0ImuIELP0DH+HSHR0kHcfDkCHH HfDkH+뷐kCHH H5H2HHxHPDHH`HT$HLL H5_*)FHHA'H ŎH5_gjL AHH81oY^BjHHA'AH5gjL ÏH jH8Hv1*_<AX$(HHPA'H5H8虁vHB`H/HHHIHH9huuIEImHEHH*HcHAmHcAmAEHH HAmAmAEHH H5HlIHpH+HCHP0L8Hr1ɺAIc1ɺAI H5HIHH*ÃHzH?'H5H84_蚃HuH?'H5]H8H DH=jo5H kH=HoAHHFH;>'t H;5@'u^Ht,H(tH?'HHfDHPHR0@gH H=n1H@HHH>'H5HH81腅g@f.H9AUATUSHHHHFHHHW@@HXHt6HJH~H;rte1 H;ltWHH9u1H[]A\A]HH9t2Hu1H;-?'H)~uIM9tJtH9u]H[]A\A]fHHH[]A\A]}LfM{HvH9t1 fDH;\tHI9uE1HCs@fHFHt[@tRHXHt^HJH>H;rG1DH9t3HH9uéHHH9Ht*HH9u@HH9Hu1H;5='fAWAVAUATUHSHXL%='L-)H<'H^H|$HD$ Ld$(Ld$0Ll$8HD$@HHHrHcHL=a<'Lm0HE(HD$H] HmHEHL9H=)Ll$ HGH;<' H;f='HWBHrE1H4$ uLw~H QP H;';) LLH$H$~p VP H;'=W29ZH<$1 HJ)H Sw)H9H H:w)H HL5&w)M IFH5ʇ)LHH IIHM IHuIFLT$LP0LT$H<$LֺLT$pzLT$HIIL HIu IBLP0L;5:;'L;5:'MM9DL~d I.uIVD$LR0D$L;=:'AL;=Z:'D&M9Lu~AAHD$HH1HLLH IAWHT$)_AXIIM*HIu IGLP0L9d$H $IMHHD$HHuH<$HGP0MtI.u IFLP0Hmu HEHP0H+u HCHP0HXL[]A\A]A^A_DL=Y9'Ld$L@L=A9'Ld$L=)9'@L~8LPvIH5~)LIHV|HD$ H H]@HHH tHOHL LOLHH88'SHH5]H81~XZH 9jH=oE1@9{@$@HEI,$u ID$LP0L5:~)IHm+HEHHLP0fDH)H5s)H9pHs)HPHLms)M|IBH5)LLT$HHLT$IIHMIHu IBLP0H<$LvIIMHIuIFLT$LP0LT$L;7'L;7'<M93LLT${LT$A I*u IRLR0EEL;=67'AL;=6'Du M9 EHD$H1HHLLH IARHT$LT$ |)Y^LT$HIIHIbIBLP0Sf.E LH$H<$pH E1H=^m<@HLDDHIHHcHHF8HD$@HE0HD$8HE(HD$0HE LHD$(HEHD$ rIH HwHcHfDIEH55'LtHHH;5'AH;5'Du H;L$EH)EImuIELP0EH~)H o)H9Hm Ho)H HH o)Hq HAH5փ)HHL$HH HL$IHHM\ HHuHALL$HP0LL$H!3'I9A LLLL$LL$IMIHML IHu IBLP0IGH5{)LHH HIHH IHuIGHL$LP0HL$H54)HϺHL$rHL$HIH HHu HAHP0L;=3'L;=:3'GM9>LVwAąk I/u IGLP0E&IEH1'I9ELLMIIHM/IHu IBLP0H $HHD$HHHt$  H$tH9)H="n)H9x H n)HHL5m)M IFH5~)LHH IIHM IHH<$LֺLT$oqLT$HIIPHIL;5?2'L;51'M9yLu#I.uIVD$LR0D$L;=1'AL;=_1'D[ M9R LzuAHD$H1HHLLH IAWHT$)IXIZMH1E1MArAH= {)Hl)H5l)I`fDH g}H=hH4$AHHD$HHE1@H=z)I@3wI1AAIHWIFHL$LLT$P0LT$HL$MHtHHHHu HAHP0DDH |AH=MgH4$HHD$HHfH=IrHAA IHuIBLP0zfrHQH+.'H5IH8,o6IA 1E1AHHAHP0H5/'LlnHH H;Q/'AH;.'DL9 HHL$rHL$As H)u HAHP0EH<)H i)H9H Hi)H HH i)H H5|)HHL$HL$HI H)u HAHP0H5y)L賻IH I/uHD$IGLP0LT$LֺLLT$HmLT$HH I*uHD$IBLP0HL$H; .'AH; -'DL9HHL$qHL$Ac H)u HAHP0ImpafIFLT$LP0LT$IBLP03HHL$#qHL$AIEIAHIEH I*A8LpAD$oDD$HCH=cH *yAH<$HHD$HHHGP0f.AA8I*E1fHFLHD$ /iIMH5x)LHVoHtHD$(IM~zH5s)LHVZoHtHD$0IM~TH5Py)LHV4oHtHD$8IM~.H5x)LHVoHHD$@IMHD$0Hl$ H\$(Ll$8HD$L|$@LJE1 uLGHt$  LAH$DH)H5rf)H9p HYf)H HLEf)Md H5 w)LLT$@LT$HIIm HIu IBLP0H<$LiIIMc HIuIFLT$LP0LT$L;*'L;*'0M9'LLT$*nLT$A\ I*u IRLR0EL;=B*'L;=)'\M9SLmA HD$H1HHLLH IARHT$LT$ ^})AZA[LT$HIIHAAf1H=Is)Hd)H5d)6IIfDD$OlDD$HZfH1E1MAAyE AA3coLT$IfDH=r)I@H1A6AH=ir)HZc)H5[c)VHfDAAEAAnHL$IjfDH=q)H@AA IIHdMQHII)uIALT$LHL$P0HL$LT$HLLHL$LT$HL$LT$IH) HALT$HP0LT$MAMAmH$DHIAA8IMAMA1HVIEMAMAH7EHT$ HLLvH5)Hz)H a)H9HHa)H0HL5a)MH5l)L蕳IIMHIuIFLT$LP0LT$H<$LֺLT$#eLT$HIIHIu IBLP0L;5%'L;5c%'M9LiI.uIVD$LR0D$L;=%'AL;= %'DM9L(iAƃ HD$HE1HHLLH IAWHT$x)AXAYIIMH1E1MAAR@]H=n)H`)H5`)菰I HAAFH=kn)vI1A_AjIM}MMUIIImuIELT$LP0LT$LLLLT${I/LT$IIGLT$LP0LT$AA. HAAbIAd1E1AHDIEMAMAHIEIAHAHw)H5^)H9pHl^)HHLX^)MH5i)LLT$sLT$HIIEHIu IBLP0H<$L bIIM:HIuIFLT$LP0LT$L;"'L;F"'DM9;LLT$]fLT$AI*u IRLR0E[L;=u"'L;=!'mM9dLfAǃHD$HHELLH IQ1ARHT$LT$ u)^_LT$HII-HAAfDD$dDD$HofH=Zk)H\)H5\)GIbHAAzH=#k).I9H1AAH=j)H[)H5[)H]IEMAMAHTAAiEIEIAHH=j)HIEMLMAAHIEAHC1ɺLAH$DIEMAMAHMD$&cDD$H 1Hbt)H [)H9HbHZ)HHL5Z)MH5Nf)LIIMWHIuIFLT$LP0LT$H<$LֺLT$^LT$HIIHIu IBLP0L;5f'L;5'gM9^LbI.uIVD$LR0D$L;='AL;='D7M9.LbAƃ HD$E1HHLLH IASAWHT$_r)IXIZM,H1E1MA"ALH=#h)HtY)H5uY)IH=h) IH1AAsHAAIA1E1AHADHAAH=og)zIRH=[g)HX)H5X)HI0Hq)H bX)H9HHIX)HHL5X)MfH5c)LLT$pLT$HII)HIu IBLP0H<$L\IIMHIuIFLT$LP0LT$L;'L;C'&M9LLT$Z`LT$AI*u IRLR0E3L;=r'L;='M9L`Aǃ(HD$1HHLLH IAPEARHT$LT$ o)AYAZLT$HII'HAAN^HnEH1AAAADIA1E1AHO^H_DHo)H5V)H9pHU)H#HL5U)MH5{i)L3IIMHIuIFLT$LP0LT$H<$LֺLT$YLT$HIIoHIu IBLP0L;5'L;5'M9L^vI.uIVD$LR0D$L;=6'AL;='DM9L]AƃHD$EHHLLH IQ1AWHT$um)^_IIMRH1E1MAzAH=Xc)HT)H5T)EIHAAxH1AA $H=b)IHHAAj)H1AgAXeH=b)IIAl1E1AHeH=sb)~IYH=_b)HS)H5S)LI7AA@^[HkwnESH5d)H<$}IHH;2'AH;'DM9LLT$[LT$ADI*u IBLP0EH=g)L9H$H@t H;d'H4$M^IMH=i)LLT$ LT$HIIt:HIu IBLP0L袸I.u.IFLP0KHAA.'H4$YIUH50\)H=Ai)1jIHtIHI,$u+ID$LP0AAEzH=`)H,Q)H5-Q)I5H1A>AIELT$LP0LT$XHK HAA;9XHH aDDH=LY@AWAVAUATIUSHHGUHMu\H贶H UaH=feH+t!:RH[]A\A]A^A_fDHCHP0@LH!%tDHVHDHL)1L\IHHHCH;3'H;'Ll$H;d'fHSBLrE1 VL%'H QP A;$LLAIVH QP A$H=9MIImtrMt-I,$L:I.td{DI,$u!ID$LP0U?fIELP0@IFLP0 ID$LP0^L{ @DU@$ fLHIfDL-g)IE-L{M5LcII$H+u HCHP0LLLjI/IIGLP0Ht$H葷ImQLB1 uH{Ht$uTAI>H=,>TE1"UIHuH'H5 ,H8Q1ɺAI@AWAVAUATUSHH@lAAAAHLgLP1DpHՐyOHA9DHtP)HcHI LH9H|PH1H0HLHtHHݐxH=c)H5Hu QH /]H=HNH1[]A\A]A^A_fH[]A\A]A^A_ff.USHHOHDOlHWxHH<$HLHL$E%LML9H9AD L9AH9A D!M9AI9AE DAIo D)L$o)T$PAo)$tsob)d$ oh)l$`Aop)$tLoz )|$0ox )|$pAo` )$t%oB0)D$@oH0)$AoP0)$DɃA9t$ANHcLTJHDPKHԐ>HHefH;I'HgHHHHtHH[]@HCHP0HH[]fL9AH9A H9L9AD AWAIJo*Dfv)l$o0)$)t$PtaoZ)\$ ox)$)|$`t@oj )l$0op )$)t$ptob0)d$@oX0)$)$DɃD9AN HcLLJHDŽԐHDPHHH Z<O1H=E*H YrH=E@H HL$HHL$PIH$A HJHL$HHHL$XIHH$AHJHL$ HHHL$`IHH$AHJHL$(HHHL$hIHH$AHJ HL$0HH HL$pIH H$AHJ(HL$8HH(HL$xIH(H$AXHJ0HL$@HH0H$IH0H$A-HR8HT$HH@8H$I@8H$f.Hi_)HHxH9HXHHqH~&H;Q1@H;THH9uHE 'HJH5J*HWH81QH+u HCHP0H W*r1H=CHH[]+H Wr1H=^CaHHPLDH HL$HHL$PHH$ALBLD$L@H$LD$XALBLD$ L@H$LD$`ALBLD$(L@H$LD$hAxLB LD$0L@ H$LD$pATLB(LD$8L@(H$LD$xA0LB0LD$@L@0H$L$A HR8HT$HH@8H$H$H'H5TH8IKH@HH9HuH;l 'AWAVI1AUI1ATIUSHHH=P)HHmIFH5 S)HLHEAHL|$;HBIMDI|$pzLIHSKIHL`H'I9FILLLL$OGLL$HII)tI.ZIGHHKL5z'H QP A;q 1LLHVKH QP AH=\9bHI/!I,$HCH;k'L%n'L9 HHKHPHHS HH9IEH9HC0@I9@HH)H!HHH)Hx1HHfoL ALHH9uHHH2H9'AT5HPH9PAT5HPH9PAT5HPH9PAT5HPH9PAT5HPH9PAT5HPH9PAT5HPH9PAT5HPH9v|PAT5HP H9vjP AT5 HP H9vXP AT5 HP H9vFP AT5 HP H9v4P AT5 HP H9v"P AT5 HPH9v@AD5HHHHu HCHP0I$HmLH+u HCHP0HHH[]A\A]A^A_fH/HHI|$p?IIH+HCH;'2LKMLcII$H+.HCHLL$P0ID$LL$H;'H;'LL$L$HLL$HI>MHcL$tLHHM|IMtID$HHELD$GL5'LD$H QA;P 1LLD$LHGLD$H QP A=H @9HI(u I@LP0I,$u ID$LP0HCH;'OH'L%'H:L9H5TL;Dp8I/ApE1I.uIFLL$LP0LL$MtI)u IALP0E1Mt'I,$u ID$LP0MtI/u IGLP0H /PD1H=<HmE1HEHLP0HT_ID$IFLP0IALP0I.f.ID$LP0IGLP0I,$f @9E@$@ID$@=^LL$~D$E1L|$HD$Lt$0Ht D$)D$ ID$LX@ uMT$LL$<HLALL$HHApI/H&H8%ApHHH|Q1IH5LApDE1H1LALL$HI1HHHHtHHЅt1HJ&HHÐ;Ht>H(uHPHR0@H 1JH=@61ff.AWIAVAUATUSHXHGt$,H1_:IHbHtC)HIGH;&t H;L&IME1HD$HcD$,E1E1L|$0D$ MHH|$D$HD$8vIEH @&I9MI9IEJ,IHEMtI)u IALP0HMtI/u IGLP0H{H;=K&H5B)6HCL@I\H{sH)LFHIE>IHHMHHuHCLT$HP0LT$H9-N) HuH&H9tH~`HNH@T$u H9D$It$ ID$HHH9H9HEIt$H,HID$H|$IILLL$LHD$LL$HHL|$L|$0>LD$LL$HH<&H2H9 LD$LL$I=LL$LD$f.L$ I|$ ID$H5{@)IIL9DH9;HI|$H4HID$D$ D$F=HtILxI H F&yE1H=KHXL[]A\A]A^A_fDH;=A&=H8IHHMD$yL|$0HAHHuHBLT$HP0MLT$tImuIELT$LP0LT$MI*u IBLP0t$H FDH=JJIE1HIuIGLP0MtI,$u ID$LP0HtH+u HCHP0HHmHEHP0I9KlIHEP@HFHHG`HIfDH|$0LT$)HIWH =)H HI9uLLLT$%:LT$HH(uHpLT$HV0LT$I/IGLT$LP0D$ LT$D$Xf.{CHH H~f.CsHH H)ZfDH=I)H;=&L|$0HFt H;&LT$s>LT$IMH=I)LD$HHGH;& H;f&DHWBHRE1HT$ uLwLT$LD$9LD$LT$P P H&;LT$LLHD$LD$I9LD$LT$P P Hk&=.29cMI(uI@LT$LP0LT$LLT$I.LT$uIFLT$LP0LT$ILE1D$zAH @HLLT$P8LT$`ILL|$0D$?zAE1HLLt$DH BH=UFAI/LLT$7LT$.D$ D$@DML|$0ImEH+ID$HHc\$,H)zD$:Hk&HLHT$LD$LL$h2LL$LD$HIHT$HT$LD$LL$d7LL$LD$HIHT$HT$~D$LLLl$D$@IHIDL9IH|H@HHD$HIHCLD$HLL$P0LD$LL$DIELD$LLL$P0LD$LL$DHLD$LL$>5LL$LD$HHzH ?H=:DLD$LL$LL$LD$LLILL|$0D$yAE1H=ILL|$0D$yAE1HfLT$LD$5LT$LD$@${>:XC|4ILLT$LD$貱LT$LD$IMBILMD$zAHLD$HILL$t/LL$LD$HH1H~*H 7)HHJH57)H4HH9uHLLD$ LL$HT$4HT$LL$HLD$ H(u(HHHT$ HLD$LL$Q0HT$ LD$LL$H*HBLD$HLL$P0LD$LL$LT$4LT$I>ILL|$0E1D$yAHjILE1D$zAHJIMLL|$0MD$yAH"Ht$HLT$LD$胖LD$LT$I<LZE1ɨ uLOLT$Ht$HLD$LALD$LT$IH=  3LD$LT$LT$LD$3LD$LT$HH:&H5 H8;0LD$LT$UzHLLAD$zHHHVE1E1 D$zAH*uLLE1ImLLE1HH11D$pyAHHAE1E1D$ry뜾izHLD$LL$舚LL$LD$>HHE1LLD$yAH6HLLAD$rzH1ɺLALD$LT$I<H W;LyAH=?I/t11?f.AWAVIAUATUSHHT$HHGhIwlLHHH;&HPHHhLh HEIEH(I,$tjL%&L9AH;- &Du}H;-V&ttH,2ÅyjH a:2lE1H=&Hm$ID$LP0H@HP0I,$qfAIFHt$LPIHL9L;=T&1L;=&$Ll1IVLLHRpHHBHHHIFLHLPIH~H+u HCHP0I,$u ID$LP0HE1HHEuHEHP0MtI/u IGLP0Imu IELP0H[]A\A]A^A_HT$LLP IHH(tHmuE1@H@LP0HmuDHHBHHHH;5&IFHT$HLPIHAnlH+Au HCHP0H 8DDH=$跻HmH)>)L5/)HCLM).H-"&H QP ;U 1HLAH-H QP EH=C9HH誌H+k@ lH 17H=#MۺfDH1&ME11H5z#A$lAH8f*H+u HCHP0H 6DDH=j#}Hnf.~Hx.HH7H 5HEH&H5"H81K0MAlE11AlfHCHP0k @DH1&H5H81/떐AlAAHR ~&HHHHflHfHH9uHHH9WH5}&H4HHHI,$ ID$LP0DIH.HI,$H$ ID$LP0L@H &H?!H5_ H8&HIH81vDH&H]"H5s H81M@HHGH$rfDL0AEAUHH f@H5#)LMLI fDH&H5 H8o_H ;H=Z I,$ID$LP0fq_@Lo2IGLL$LP0LL$DHCHP04IALP0GHEHP0fH5)H|$j{H)HH{PH/uHGP0H)S0H{@LC8HCPdHcHH)H$H HHIHHH9uNfDDGII L<$fDGII IL<$zfDID$LHD$ n IH5<)LHVHD$(HIH5)LHVHD$0HIMHl$ H|$(DH5)LHVeHtHD$8IM~H5)LHV?HtHD$@IM~HT$ HLLH5ӵ(mz _zLx IH5N)LIHVHD$ HIl$fD$x`E1E1D$fDH5a)H= )1YHHHwoHm\HEHP0_H5 )H=R )1{YHHH'oHmHEHP0_s@hHy&HCXH&H5gH8 `L%N&)fL;59&H,&H{XHH/uL%&#fH51)H=Z)1XHH,HWnHm HEHP0D$$1`nfDC@$f._qL(IM$s`D$$M`E1D$HEE1HHE@E1E1$d`D$rfD_$f`D$$h`D$D$$1`l@$p`D$@Ht$ pIf_QIIEHE1D$$_@LJE1 uLGHt$ eLAI|H= $s`E1D$2 HLHQ&H5H8R 1H$D$$`@_zD$$ax H@D$$a__ H &&HH^&HJH5jL AH GH81^z_^]D$$`sD$$-`_H=)H;=&HD$H@t H;Z&Ht$BHHH=)HEIHKHmu HEHP0LjI/IGLP0D$$`H5# )H=T)1THHHQjHmHEHP0D$$CahHH&L 'AjHH5`H8H 1AX^AYz1D$$`D$$CaMHHHw&HMH5}H81D$$_E1$`D$ED$$?aD$$`qHt$w H@1ɺLAID$$_6fDAWfAVAUATIUSHHa&HnH.)H|$H$Hr ))$HD$ H$H )H$HD$XH$H Ht%~0HtH HF8HD$XID$0HD$ ID$(H$fH&H$H Il$ Mt$IHEHHPHL%)I9H_&H9CI9D${ f#A|$ r#HSI;T$HCIL$H9@H@t H{ At$ 8@ AH{H@ #IL$0IH@HIDȃ%&DA92Ht E1H(AfE HHHHu HCHP0H)H )H9HV H)Hv HL)MZ I@H5/)LLD$HHW LD$IM\ I(u I@LP0H&I9G#LL@MHD$H|$lI*u IBLP0I.u IFLP0H)H5)H9pH)Ht HL)M0IBH5])LLT$HHLT$IMI*HH&I9G' HL?MIM I(;Hm H$H;&w H8)H5A)H9pw!H()H"HL=)M IGH5)LHH}"IM!I/H<$HGImIH'H4$L<$HH@H0H|$H5 )HGHHO!IM LLD$tLD$HI!I(u I@LP0I8"ID$H5l )LHH"IM"LLD$LD$HH"I(u I@LP0H H5)H=)1MIHdHcI/u IGLP0D$ @LA 1E1E1E1HD$E1E1HD$(HD$@HD$HHD$PHD$8HD$0HD$2f.H;&c[H5L)H[H5)H[Aą*"H+u HCHP0EH5)H=)1LHH$H&HdbHH$HHPD$ ?1E1E1HD$E1E1E1HD$(HD$@HD$HHD$PHD$8HD$0HD$H$Lt$A @Mt$I)uIAHL$`LL\$XP0HL$`L\$XMtI+uICHL$XLP0HL$XHtH)u HAHP0t$ DH m H=NE1ΏH<$tH $HHH$HHuHAP0MtI/u IGLP0MtI,$u ID$LP0MtImu IELP0HL$HtHH$HHu HAHP0HT$0HtHH$HHu HBHP0H\$8HtHH$HHu HCHP0Ht$PHtHH$HHu HFHP0HL$HHtHH$HHu HAHP0HT$@HtHH$HHu HBHP0H\$(HtHH$HHu HCHP0Ht$HtHH$HHu HFHP0HT$HH$HHu HBHP0Hmu HEHP0HĸL[]A\A]A^A_fL;%&uLH/IH12H;&AH;&Du L;=ҽ&I/E.IGLP0ELt$D$ j?HD$1E1A HD$(E1E1E1HD$@E1E1E1HD$HHD$PHD$8HD$0HD$H$HH+HCHL$xHL\$pLL$hLT$`LD$XP0HL$xL\$pLL$hLT$`LD$XDIBLP0fHEHP0I@LP0HmfIGLP0$L1&@Imu IELP0H$H;f.L(HH5&)LHHV~H$HIl$HHSH CHMHUHr LD@H&H5L ? H81X1?ZH  H=E1rVDHwIHHcHHF8H$ID$0H$ID$(H$ID$ LH$ID$H$HHH<HcHH=)H(H5(FIfDLt$1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?1H$H5L1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ ?mDH=)cI@CLD$IfDHD$1E1E1HD$(E1E1E1HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?Lt$A @Mt8I(u2I@HL$pLL\$hLL$`LT$XP0HL$pL\$hLL$`LT$XMzI*pIBHL$hLL\$`LL$XP0HL$hL\$`LL$XCH5q)LEHd111HHD$-VLD$HI I(u I@LP0H5%)LEIH7!11ҾHHD$ULD$HI!I(uHD$I@LP0LT$LֺLLT$LT$HIi#I/uHD$IGLLT$(P0LT$(LD$I*uIBLD$LP0LD$L;&@L;2&@vL;w&iLLD$DLD$#I(u I@LP0H|$H5(HGHHIM?111LLD$TLD$HII(uHD$I@LP0LT$ID$H5(LT$LHHLT$IM111LLT$(LD$2TLD$LT$(HIZI(uI@LT$LP0LT$L׺LLT$mLT$HII*uHD$IBLP0LD$I/uIGLD$LP0LD$L;&L;&L;յ&LLD$LD$D$`I(u I@LP0L$`(H<$HWHBpH0H@H#H5(HT(HHD$LD$HIII(u I@LP0H5(H|$BIH111HRIHgImuHD$IELP0LD$IW IGHHH9H9IIWLHIGI(u I@LP0H5(H|$AIH H!&I9EqImHdMUHEIImuIELT$LP0LT$LLHLT$ HmLT$HD$uHELT$HP0LT$H|$v I*u IBLP0Hl$H5(H2AIH!HEHD$HHEuHELT$HP0LT$H5h(H|$LT$@LT$HI*!111HQLT$HIt!ImuHD$IELP0LL$LT$IBH;ֱ&M#MjMe#IjIEHEI*]D$HE1ҿD$H;&H;̳&LL$(4LL$(HI#MtLhHcD$H>(LHHHITHcD$1L\$ML%>L\$HHD$$I+u ICLP0Hmu HEHP0H5*)L?IH"H)H (H9H#Ho(Hp$HL[(M3#H5[(LLT$(L\$1?L\$LT$(HI#I+uICLT$LP0LT$HB&I9BvIJHiMBHII*uIBLD$(LHL$P0HL$LD$(HLLHL$(LD$HL$(LD$HH)uHALD$HP0LD$ImuIELD$LP0LD$HR$I(u I@LP0I,$u ID$LP0H5(HJK$!RIHp*Ho(LLD$HI@HX(HH=(1GLD$HI*I(uHD$I@LP0LT$H5(LLT$ZLT$HI+H8uH@LP0LT$I*u IBLP0IEH&I9E.HL+MIMl-I(uI@LT$LP0LT$IBH;e&$,IRH-IBHD$IB HD$0IB(HD$8HD$HHD$0HHD$8HI*3!IBLP0HD$@HD$HHD$PH5u(H|$ I(HD$(H56(HsI.Lt$ H5(LLI/t"H5;(L.Iu9-H5a(H I.H)H@$H;(9H(H8HL (M8H5c(LLL$(~;LL$(HI[8I)uHD$(IALP0L\$(H\)H5(H9P7H(H7HL(M7H50(LL\$hLT$(:LT$(L\$hHI:I*uHD$(IBLP0L\$hLD$(H|$8Y:H5F(H|$8L\$hLD$(:LD$(L\$hHIt=H|$0=Ht$0HL\$xLD$pHD$hLT$hLD$pHHD$(L\$x=I*uIBL\$pLLD$hP0L\$pLD$hE1Ha&I9@L=T$`I@H;ͬ&Hc8H;&7HL\$xLT$pLD$hLD$hLT$pHHL\$xEMtLPHcD$`Ht$(1LL\$pHtHt$8@HHL$hHHtHLD$`7LD$`HL$hHHD$(L\$p%EH)uHAL\$hHP0L\$hLD$`I(uI@L\$`LP0L\$`L\$`L\$`HIDHD$(LD$(I@HEIh LD$(L\$`HI:DLt$XH5b(HL\$hLD$`LHD$(LL$(LD$`L\$h01H5(LLL\$`LD$XLL$(LD$XL\$`DLLLLL$hLD$`L\$X6L\$XLD$`HHD$(LL$h+<I+uICLL$`LLD$XP0LL$`LD$XI(uI@LL$XLP0LL$XI)u IALP0HL$(H; &H; &n"H; F&a"H|$(&D&1H5_(HDn&H|$@)HD$@HHH5(H=7H})H|$HHD$ LD$ HI.I(uHD$ I@LP0L\$ H|$LL\$ L\$ HI/I+u ICLP0HL$HHD$HHu HAHP0L|IH/ID$LLD$LH5(HHa/LD$2%I(u I@LP0I$MH\$LI/AfD@HK0LCHIDHfHCHP0HFLH$\HH5J(LHVH$H>HHLL$H$H$H$H$H$HD$ H$HD$XHfH5(LHV=HtH$HH~H5((LHVHtH$HH^H5C(LHVH$H$HHH#H$HLLH5c(NF?L1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ m@UL1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @Lt$1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?Lt$D$ p? fDMgMMWI$II/uIGLT$LP0LT$LLLLT$I,$LT$HD$ID$LT$LP0LT$zfD$`@1E1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?T@H=(H2(H53(~0I-fDLT$IKfD1E1E1E1HD$E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?H=(MI@MoMMGIEII/uIGLD$LP0LD$LHLLD$1ImLD$IIELD$LP0LD$wD1E1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?@H8L~ It$HfDL1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?@H=(H(H5(.IfDL1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ ?@Af+ILL1E1E1E1E1E1HD$E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ ?H=(JI`@I{D$ A@LA 1E1E1E1HD$E1E1E1HD$(HD$@HD$HHD$PHD$8HD$0HD$NfDH5!(H=(1 ,IHHAI/IGLLA P0D$ Q@%DDV@@IgL1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ c@fLD$ e@A HH(LH0HVHHL$H$fLt$D$ v?`LLMHD$fDD@L1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ @I;H&HRLA H5mH81+D$ @&fDL1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ @4LT$IL1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ @=LD$ Q@A L1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @~IBLL$(LP0HE1ҿD$LL$(D$LL1E1E1E1E1E1HD$A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @HE@=H(HHHL$LL$L$H$$LL$HHD$MtImuIELL$LP0LL$I)%IALP0L1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ @LD$ @A Lt$1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ ?H5(H=(1&IHpH;I/IGLLA P0D$ @"LLLT$LT$HML1E1E1E1E1E1A HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @cH5(H4ZIHH(LLT$HIBH(HH=(1D0LT$HI5I*uHD$IBLP0L\$H5R(LL\$ECL\$HHHD$PH8u H@P0L\$I+u ICLP0HD$PHI&HH9PHHI5MIMNI*uIBL\$LP0L\$ICH;&tISHrICHD$0IC HD$HD$0HHD$HI+ICLE1P0HD$@HD$HHD$8L1E1E1E1E1E1A HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ o@KL1E1E1HD$E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @LL1E1E1E1E1E1HD$E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ r@dLL1E1E1E1E1E1HD$A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @LL1E1E1E1E1E1HD$A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ t@LLLD$LD$LD$ @1E1A E1E1E1HD$E1HD$(HD$@HD$HHD$PHD$8HD$0HD$H(HHHL$LL$L$H$:LL$HHD$9ILL1E1E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ A@LL1E1E1E1E1E1HD$A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ w@L1E1E1HD$E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @.LD$ z@A L1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0D$ @2LD$ @A HHD&H5ŷL ojAH 4HrH81^>_H\$L1E1E1E1E1E1HD$A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @L1E1E1HD$E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @LL1E1E1E1E1A HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @EL1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ M@IH6H:(LL\$HICH#(HH=Y(1'L\$HII+uHD$ICLP0LL$H5(LLL$:LL$HHHD$HH8u H@P0LL$I)u IALP0HD$HH&HH9PHHI MHD$@H|$@I+u ICLP0HD$PE1HD$8HD$0HD$LպE1D$D$LD$D$L1Ld$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ ;AILL1E1E1E1A HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ !AL1Ld$E1HD$(E1A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ =AaH=%(H(LT$H5( LT$I-LL$0LD$HD$@HD$HHD$PL1Ld$E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ ?AH=g(LT$m7LT$I}I1LE1HD$(E1E1A HD$@HD$HHD$PHD$8HD$0HD$D$ ,A"L1E1E1HD$E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ @6L1Ld$E1E1ME1A HD$(HD$@HD$HHD$PHD$8HD$0HD$HD$D$ OABD$ \ALd$A 1E1E1E1HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$ZLt$1E1E1E1E1E1E1HD$A HD$(HD$@HD$HHD$PHD$8HD$0HD$H$D$ j?H5(H%H|$ H(H*(H9X !H(H HL(M!H5](LLD$ (LD$ HIO!I(uHD$ I@LP0L\$ H|$0 H\$0H&H9C HCHHH &I9CLHL\$ L\$ HMH+uHCHL$ HP0HL$ H<I,$uID$HL$ LP0HL$ H|$HHL$ HL$ HH$H)HAHP0H;(H(HHL (MsH5(LLL$(LL$(HII)uHD$(IALP0LD$(H|$0H|$XLD$(LD$(HIIH|$01HLD$`HD$X>LL$XLD$`HHD$(I)uIALD$XLP0LD$XH}&I9@Ht$(LLD$XiLD$XHMHt$(HHD$(HHuHFHL$(HP0HL$(H I,$uID$HL$(LP0HL$(H; &AH; &DH; :&HHL$(HL$(A!H)u HAHP0Eb H&HHD$(HD$(H; &H;& LL\$ L\$HI1"I+uHD$ICLP0LD$I@LLD$LALD$HIULLD$(HD$0ALD$(LT$0HHD$0LAԾHLD$(LT$0?!I(uI@LT$(LP0LT$(LT$0HD$@E1HD$HHD$8L1E1E1HD$E1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$D$ @Ld$D$ AA >D$ BLd$A 1E1E1HD$HD$(Ld$1E1E1HD$(E1A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ fASLd$1E1E1HD$(E1A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ kA(H|$H(HF(H9X>H-(HHH (H<H5(HHL$ THL$ HIH)uHD$ HAHP0LD$ H|$0<H^&I9@Ht$0LLD$ JLD$ ILMH+uHCL\$ HP0L\$ H|$LL\$ ZL\$ HHI+SICLP0DLd$1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ nAH\$A 1E1D$ BDE1E1}ELd$A 1E1D$ CE1HD$ H;&WLLT$(LT$(HHD$I*u IBLP0Ht$HFHLAIHHD$(H|$ALD$(HI_Lt$HD$0LALD$(LL$0HHD$81LAԾHKLD$(LL$0IHD$HItIFLL$(LLD$P0LL$(HD$@HD$HLL$0HD$P2Ld$1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ AH H! HHQH PLT$XHEH&H5*H81Ld$A 1E1D$ AE1LT$XHD$HD$(HD$@HD$HHD$PHD$8HD$0HD$PMeMRMEI$IImuIELD$LP0LD$LHLLD$I,$LD$IID$LT$(LLD$P0LT$(LD$H5U(H=.(1W HHD$(HSH!HHD$ HH)HCHA P0Ld$D$ BLd$1E1E1H\$A D$ .DʿH;&HH5=A H81 Ld$1E1E1D$ CHD$邿D$ BLd$A 11E1E1HD$HD$(VLd$D$ BA aD$ BLd$A ILL1E1E1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ ALd$1E1E1HD$(A HD$@HD$HHD$8HD$0HD$HD$D$ BL`HMLPI$IHHD$HHuHFLT$HP0LT$LHLLT$I,$LT$IID$L\$(LLT$P0L\$(LT$Ld$1E1E1HD$(A HD$@HD$HHD$8HD$0HD$HD$D$ AvLd$1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ ALd$1E1E1HD$(A HD$PHD$8HD$0HD$HD$D$ qB鸼L`HMLXI$IHHD$HHuHAL\$HP0L\$LHLL\$[I,$L\$HD$@ID$L\$LP0L\$yH=|&HHD$(Ld$1E1E1HD$(A HD$@HD$PHD$8HD$0HD$HD$D$ VBLd$1E1A HD$(HD$@HD$HHD$PHD$8HD$0HD$HD$D$ SB[Ld$1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ NBLd$1E1E1HD$(E1A HD$@HD$HHD$PHD$8HD$0HD$HD$D$ A鈺Ld$1E1E1H\$A D$ 0DHHHH.H -L\$XHEHy&H5H81Ld$A 1E1D$ BE1L\$XHD$HD$(HD$@HD$HHD$8HD$0HD$鹹ISHHICHH@HT$0HD$D$ CLd$A 1E1HD$HD$(Ld$1E1H\$D$ 3DA @蓼LD$1E1E1H\$D$ @DA Ld$A 1E1D$ CE1HD$Ld$D$ BA Lt$HHLMHD$@Ll$PHLMIH5)(H|$ z g HE(H5(H9p- H(H HH ѷ(H H5Ѿ(HHL$ HL$ HI H)uHD$ HAHP0LD$ H5(1LLD$ LD$ HI I(uHD$ I@LP0LT$ I*IBLP0Ld$1E1E1HD$A D$ 3C鶾Ld$1E1E1HD$(A HD$D$ 1C鄾Hx&LD$XHH5A H81肽Ld$1E1E1E1HD$(LD$XHD$D$ 0C&Ld$1E1A HD$(HD$D$ -C鷶Ld$1E1A HD$(HD$D$ +C鈶H=( IAH=(H(H5(ILd$A 1E1D$ AE1HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$gHu&LT$XH5H810Ld$1E1A HD$(HD$D$ B锵H=տ(L\$(L\$(IH=(H(L\$(H5(L\$(ILd$1HD$(A HD$D$ BLd$1E1A HD$(HD$D$ BH=0(;IH=(H(H5( ILd$D$ BvIRHIBHHT$HPH@HT$0HD$8pI@@=BHD$8LT$hL~D$hL\$xH$HcD$`D$(LT$pHkLD$`)$HlLD$`LT$pHIL\$xMt.I*u(HD$hIBLL\$pLD$`P0L\$pLL$hLD$`HT$(HHD$(HHu(HBL\$hHLL$`LD$(P0L\$hLL$`LD$(LL$(HD$8LT$hL~D$hL\$xH$HcD$`D$(LT$pHkLD$`)$HLD$`LT$pHIL\$xMt.I*u(HD$hIBLL\$pLD$`P0L\$pLL$hLD$`Ht$(HHD$(HHAHFL\$hHLL$`LD$(P0L\$hLL$`LD$(H t&L\$`A HH5LD$XH81ոLd$1E1E1D$ BLD$XHD$(L\$`HD$wLd$1E1A HD$(HD$D$ B酹ME1Ld$E1E1D$ BCA HD$(HD$MpMbM`II$I(u I@LP0HT$(LLI.HEIFHL$XLP0HL$X,MsMtMcII$I+u ICLP0HLL2I.HWIFHL$ LP0HL$ >HL\$ HD$0xL\$ HHLd$1E1A HD$D$ CLd$E1E1HD$D$ DA 隰H=ۺ(HH=Ǻ(H(H5(HLd$E1E1HD$D$ DA =H|$0˰L\$ HH;HXH\$(Ld$1HD$(HD$A D$ BH\$(Ld$1HD$(HD$A D$ BHD$`H#q&HH5%LD$XA H81Ld$1E1HD$(LD$XHD$LT$`D$ BL\$h鐶Ld$1E1A HD$(HD$D$ BaLd$1E1A HD$D$ DMPMM`II$I(uTI@L\$pLLT$hP0MLT$hL\$pD$`fLd$1E1A HD$D$ BǵMD$`0ME1Ld$D$ CHD$A PLd$E1E1HD$D$ CA )Ld$1E1A HD$D$ CCLd$1E1E1HD$(A HD$@HD$HHD$PHD$8HD$0HD$D$ A!L\$8H\$1E1Ld$A D$ AHD$HD$(HD$@HD$HHD$PHD$8HD$0HD$lAH\$HHD$HHuHCLL$ HLD$P0LL$ LD$LL$ LD$DLD$LL$ IHLLL$`H H5LD$XA HEHCl&H81Ld$1E1D$ AE1LD$XHD$LL$`HD$(HD$@HD$HHD$PHD$8HD$0HD$mE1E1Ld$A 1E1D$ AE1HD$HD$(HD$@HD$HHD$PHD$8HD$0HD$Ld$1E1E1HD$A D$ wCѲLd$E1E1HD$D$ lCA jLd$E1E1HD$D$ jCA CH=̱(HH=(H(H5(]HH5(H=(1HHtAH H+u HCHP0Ld$A 1E1D$ CE1HD$鯪Ld$1E1E1HD$A D$ C醪Ld$A 1E1D$ `CE1HD$]Ld$1E1E1ID$ DA HD$qHk&H<H5A H81tLd$1E1E1D$ DHD$HZk&HH5\A H81,Ld$1E1E1D$ CHD$顩H=(IH=γ(H(H5(IH|$0L\$ 质L\$ H_Hj&L\$XH%H5A H81药Ld$1E1HD$L\$XD$ CLd$1E1E1HD$A D$ CLd$1E1E1HD$A D$ C鮨E1I(uI@LT$LP0LT$LT$LT$IHXLLT$XH OH59A HEHg&E1H81裮Ld$1E1D$ 4BE1LT$XHD$HD$(HD$@HD$HHD$8HD$0HD$ZLd$A 1E1D$ 4BE1E1HD$HD$(HD$@HD$HHD$8HD$0HD$M`M*IXI$HI(u I@LP0HT$0LH9I,$I ID$L\$ LP0L\$ Hh&LD$XHH5A H81WLd$1E1E1E1HD$LD$XD$ DLd$1E1A HD$(HD$D$ CխLL$(Ld$1HD$(HD$A D$ CdLd$E1E1HD$D$ CA }H\$(Ld$E1HD$(HD$A D$ BBLd$A E1E1D$ ECHD$HD$(ۥLd$1E1E1HD$A D$ B鲥Ld$D$ CLd$A 1E1D$ UCE1HD$wLL$Ld$1E1D$ ,BA E1HD$HD$(HD$@HD$HHD$8HD$0HD$SLd$1E1E1HD$(A HD$@HD$HHD$8HD$0HD$HD$D$ $B鴤Ld$A 1E1D$ BE1HD$HD$(HD$@HD$HHD$8HD$0HD$UHc&L\$XH5H81蛪fDAWfAVAUATIUSHL5qd&HnH~(H|$)D$pL$H$H;HHH HHxH hHMHUHLDH9c&H5L dH81٩XHZH ;sH=OE13HĘL[]A\A]A^A_fH^0ID$(HD$ ID$ Il$HD$HD$PHD$XHD$`HD$hHEHL%G(H\$HI9 Hd&H9CI9D${ I#A|$ m#HSL|$HI;T$HCIL$H9@H@t H{ At$ 8@ H{H@ #IL$0IH@HIDȃ$%DA9uGH( E1H֣A L9uMuM9{ I߄p H5)(LAą!H|$HH/uHGP0HD$HE HD$HfHL8Mt M9, HPHuLhL`M MtIEMtI$H(H (H9HH(HfHH=(H|$PHHGH5r(HHZHD$XH:H|$PH/uHGP0H|$XH_&HD$PH9G Ht$H|$PHD$HHtH/~ HD$HHD$PHH|$XH/uHGP0HD$XH|$HH/uHGP0HD$HMt I/2 Mt ImB Mt I,$" H5(H|$1_IHD$PHH;?`&L;-_&u M9DImHD$PEtHMIHD$PHH|$HڞIHD$XH9 H|$PH/HD$PH|$XH;=_&H;=_&L92H|$XAą H/uHGP0HD$XE H|$HD$(H UHD$H HL M9t M HPHuLhLxMtI$MtIEMtIH(H @(H9H H'(H!HH=(H|$PH HGH5b(HH !HD$HH H|$PH/uHGP0H|$HH\&HD$PH9G!HH|$PHD$XHtH/uHGP0HD$XHD$PH|$HH H/uHGP0HD$HHD$XHmHD$u HEHP0H|$H5(HD$XHGHH HD$XH H50(H9?HPH;]&v!HxuxDH Q]&HHHT$HH(u HPHR0HD$XH|$HH;= ]&@H;=\&@u L9$@H/uHGP0HD$HH5K(H|$ HD$HHLH5l(HǺ跛H|$HHD$XH#H/uHGP0HD$HH|$XH;=z\&H;=[&L9H|$X…$H/uHGT$0P0T$0HD$XHk(H@OH9x(r'Hc(H'HH=O(H|$HHk'H5( H|$HHD$PHp'H/uHGP0H5(H|$HD$HHD$HH'Hί(H כ(H9H(H(H(HH (H'H5:(HHL$0uHL$0HHD$`(H)u HAHP0LD$PI@H;Y&(1H;Z&%H;[&Y HL$0t$<HL$0HHD$h*Hɋt$HL%.(Ld$XMH5(LIHD$PHH|$XH/uHGP0HD$X&IHD$XHIEHD$XLh2IHD$`H_H(H5(H9p%Hz(HwHL%f(Ld$hMIH5a(LIHD$HHlH|$hH/uHGP0HT$HH5r(HD$hH|$`H|$HH/uHGP0HT$`Ht$XHD$HH|$PhIHD$HHH|$PH/uHGP0HD$PH|$XH/uHGP0HD$XH|$`H/uHGP0H|$Ld$HHD$`HD$HH5͓(LHD$HHl H|$%HD$`H H|$HH誓HD$XHk H|$HH/uHGP0HD$HH|$`H/uHGP0HD$`H|$XlHD$ H H|$XH/uHGP0HD$XM91 ID$H5e(HHD$0HD$H5O(HuIH HD$H5M(HQHD$`H HD$HH I&H9HHPHT$HHH@HHH|$`HD$`H/uHGP0Ht$HHOH|$`LH|$HHD$XHt H/uHGP0HD$HH|$XH|$`H/uHGP0HD$`H|$XH/uHGP0HD$X萏H|$LAPHt$8LL$0H LD$8HL$(HT$rH߉D$vH5(1LHIHIAYAZHu IFLP0H; H+u HCHP0|$NI$LImPI,$MoVH=(H(H5(>H fDTIH|$HHt H/uHGP0HD$HH|$PHt H/uHGP0HD$PH|$XHt H/uHGP0HD$L5X(HD$XHPXI9ZHIFHHJb @U J A@< HXHHQH~'L;q1DL;tHH9uAHD$LLLE1E1H&L|$HMHI/>IGLP0/INH~IvH9t^1 fI;TtQHH9uE1HLL$ HL$HT$vu*LL$ HL$HT$IL9JKtH9u@H H=>H|$HL$PHT$XHt$HH5(H=(1HH1HH+R HCHAIP0H=Y(H@VIfDSfD蓊H|$HHcHl$CJAHt;H/t1HGHL$P0HL$HD$HHtH)u HAHP0H|$`Ht H/uHGP0HD$`H|$hHt H/uHGP0HD$hH|$PHt H/uHGP0HD$PH|$XHt H/uHGP0HD$L M(HD$XHPXI97HIAHHJO@B7A@)HXH; HQH~,L;I1f.L;LHH9uHD$LLLE1E1H L|$HfIGLP0L|$HfDL計ImAj[DIAjHGP0@HGP0@@HK0LCHIDHfHAHP0HGP0@HGP0x@HGP0P@HFLHD$pHH5ݍ(LHVYHD$xH HHH5o(LHV+H3H$HL蘀HH5Ƒ(LHHVHD$pHtIl$QL|$HKAfDE1MLaJA@Hl$6f.@DH B&HHIIHIqH9to1 I;TtaHH9uE1HLD$(HL$ LL$HT$u5LD$(HL$ HT$LL$IL9xKtH9uf.H DH=|HH|$HL$`HT$hHt$X'H|$XHt H/uHGP0HD$XH|$hHt H/uHGP0HD$hH|$`Ht H/uHGP0HD$LLLHD$`HH>IHD$`HH=Ə(HnIHD$hHeH|$`H/uHGP0Ht$hH=(HD$`IHD$`HH|$hH/uHGP0HD$hH|$`H|$`H/uHGP0L|$HKE1E1HD$`A DHD$HeIDHGHD$PHHWHHH|$XHT$XH/uHGP0Ht$PH|$XHHT$L|$HIA H5ф(H=ڒ(1IHD$PH HH|$PH/uHGP0L|$HIAE1HD$PL|$H IE1E1ADL|$HIAkH|L|$HIE1E1ADLP|fDIt$HfDHL|$HE1IA7L|$HIE1E1ADH葀IHD$XH H=(HIHD$PH H|$XH/uHGP0Ht$PH=B(HD$XdIHD$XH8 H|$PH/uHGP0HD$PH|$XH|$XH/uHGP0L|$HIE1E1HD$XA7LH}-KA HL|$HIE1E1AH=(Hj~(H5k~(HCfDIArH|$H JAD@H=Q(\H@Hl$"JA ӃfDLH|1JAHGHD$PHHWHHH|$HHT$HH/uHGP0Ht$PH|$HHHPH|$HHl$>JAfD#fDD@L|$HE1Hl$KAJH=1(H|(H5|(IfDH= (I@裂IL|$HE1Hl$KAIAE1E1fH;:&HzHD$HH@JAXL|$HE1Hl$KAbDIL|$HE1Hl$KAH/HGHP0L$@I9uۉ3H$@H$p4$jH$H$8L$L$@yL$5L$Y^M1IHHL$@L9uHDŽ$H%H5qH84E1J@HH?H$H~1H@HHHH?HDHHكHDHL$@I9L$Ll$PL$Ll$XL$Ll$`tpL$Ll$ht]L$Ll$ptJL$Ll$xt7H$L$IŃtH$H$tH$H|$P/HDŽ$H=WF(1H$ LD$HE1ҺwJ@HH?Hm1H$H@HHHH?HDHHكHD:11lH$pHH?H1DSMcJCqH$2H$@HH?H1H$H$(E1HH$HAF AALL$@HfDLl$L/H$@H$p4$SH$H$8L$L$@L$2L$A[A\M9gDSL$HL$H$EuVL$0HcIIH|$HHL$LHD$HL$HH"HtHv%H2H9!ImH7IELP0(I$ID$xHJ4I$JH(J<f.MIAjH]EHH fHHHL$!H HHL$nHHL$;%HL$HtHHL$HD$O HT$HL$HH*uHBHL$HP0HL$vH9IlHHE@HFHHL$HL$H/@L`IHrH=5.(H$HHhI,$H=0(H\$(HGH;%zH;%HWBLbE1 1 L=*%H QP A;HLAI H QP A=@9MH+u HCHP0L~I,$ID$LAOAP0eD1]EHH fDID$LP0H%H5"AAH8NfDH (DDAjH=>dImu IELP0H (DE1H=-8HtHmu HEHP0H8L[]A\A]A^A_LhIHH==,(H"HHI,$VH='.(H\$(HGH;%hH;%jHWBSLbE1 uLw9L=2%H QP A;`HLAIH QP A=29MH+u HCHP0L|I,$9ID$LA AP0mDHHHH%H5uAAH8y,@Lw!29c@@$UID$LP0HIH?H@HHD$H{HYHt$(xIMAAJH+~HCHP0oAOAWA A?@pDHhI{Aj,;@$EAEAAGAI,$ID$LP0HIMAA H X%jE1H=@LRE1ɨ uLOHt$(ILAIH=EAJAHrH7%H5H88WAAAAHt$(}IeHH%H5VH8H=!DAAHӂYHAjdAj1RLRE1ɨ uLOHt$(u(LAIT1ɺLAIQ1ɺLAI*AW1AVIAUATUSHH$@H|$HHH3%8y HD$HE*(HxH9 HXH HqH~#H;Q 1 H;T HH9uHD$HHXHHD$pHpxL@lH|$(H\$xHHl$pL$H$L H}PLLHt$(HLkHl$8~D$HD$(H\$IFH;_%D$)$@t H;%IE1HD$ 1H|$ D$D$D$`IFH %I9NP I93IFN$II$HtHmu HEHP0ID$HP`HHxH; % ID$HPH HA\$H(HcD$HL$8HLHhPH@H9=L$H݅HcD$HMx T$`4H$HL8L$HD$H|$ LLHD$ IHHtH}%H2H9I.HHD$H'(H@H91 HXH HqH~,H;Q 1f.H;T HH9uHD$11I싈$$$$$$$$$$$$$$$$$$$$$$$$$$$0HHHHDH;N%eH5Y&(Hx{MI޽;}fDL;%%HH5M(LHIMsH%H-}%I9I9qL;=%dLTWIHHD$0IHD$0ID$H5(LHHIMI9I9UL;=6%HL;IHHD$@IoHD$@ID$H5.(LHH[ IM= I9I9L;=%|LDoIHXIH1ID$H5(LHH IM I/u IGLP0ID$H5n(LHH HH H)uHAHHL$HP0HL$HID$H5/(HL$HLHHW HL$HHH4 HmuHEHL$HHP0HL$HHcD$Ht$8H;-%HHLL@PLHII?AHuNH="(H5 HL$hD\$dLL$XLD$PLT$HPLT$HLD$PLL$XD\$dHL$h0 L;=%EIAHHD$0IH; %HHDH;-%tHLHHHH9 HpHHHHcD$`LĐHPLD$HcD$HD$`\$MI؃D$`\$jf.H%I9G5 IGHPHd H AwHt$0H~I/~IGLP0oDHV%I9GI IGHPHx H AH|$@HI/IGLP0DHcD$`HDŽPHHDŽĐHDŽ\$`XA\$AD$HH HHHcD$HL$8HLHhPH@HyHHT$H=" (H5!}A@ DH H= 膝߻| f.I.IFLP0{HD$H;d% HD$HH~H\$8HHD$(H%I9GIGHPH H A_HI/bIGLP0SfT$H=(H5ɣ"AH$H~I9OdII$@H|$0L;L$0H|$0IH; I%.HD$0I)HEHIDIf{|I.I/u IGLP0H 1H=ÛH|$(t HL$(HHD$HHu HAHP0MtI,$u ID$LP0HH[]A\A]A^A_H\$0H|$0qH: HD$0U|3IFLP0M*5HD$0L1H; %5H|$@9HD$@I9HINH+t$0I9tHD$0H/E1I9xLLL$01I럐H$H; H}PLL HL ~D$Hl$8!c|Ige|+K HLH\$8HcD$HHLHhPH@fLxHHtH HmHHEHP0HD$plwH%1H8 H ]{H=jDy|I HHL H@HDHH9HuH;t%"HD$H;%HHH9HuH;=%yH%HJH5HWH81H t{1H=x| IHH9tHuH;%fH|$(HD$I싈HD$(HH$$$$$$$$$$$$$$$$$$$$$$$$$$$0AHHHH;G%fH5R(HxoNI޽ }H\$@H|$@a HHD$@EA_AGHH @HH HHf| I( HU|H\$8HcD$H1LHhPH@@HD$p~D$HD$(HD$8f.LX IHHH@HHD$ HILL$@HL$0LH)HMHHIE1LL$0GHD$01@| HL$HHfDA\$AD$HH IFLP0E2dL HH^HHmH>HEHP0/HD$@PFH }H=襔߻|$E1{H%H1IH5H81~ H |H=A@HH9HXH:HqHRH;Q1H;THH9u(Lw HwHHD$@HT$@HD$0H*JHBHP0;HHAGAWHH HD$0HD$0}L HsHHD$HHT$HHD$@H*FHBHP07HHAGAWHH HD$@wH(HA_AGHH 1 |GLHD$0BAGAWHH HHD$0&LQHD$@AGAWHH HHD$@pL#HH%H5 H8{E1 }3i| |1}|Hl:I{AHfH;5(AVAUATUSH HslHHaL5%L9HPHL`Lh I$IEH(t{L;%|%L;%%uM9unHCLHHHSHHR(HHI,$tcImu IELP0H[]A\A]A^ÐH@HP0vLyH wkH=<臐I,$uID$LP0fDHH[]A\A]A^HLHHTkH j H=f.RkH D 1H=f.HA%H5ikH8HmuHEHP0f>ZkHxHH: H 9 HEH߾%H5H81@H%ZkH5XH81n@k fkAWfAVAUATUHSHhH$%H^H|$)D$@HD$PHD$XH Hu HHF0HD$L}(H] LuHO(H(hE111AHLHH H8u H@HP0H(L(hE111AHHAIH H8u H@LP0H(L(hE111AHLAIH H8u H@LP0AD$;EHQ(H5'H9p(H'H@HH'HHCH5 (HHH&IM(H+H5(HHH&IGH;%5H;n%H\$@H;$%FIWBHrE1Ht$ uMwp VP H~%;HLHD$HD$_HT$p NH HJ%=p4@9` HwMH+uHCHT$HP0HT$HI.uIFHT$LP0HT$H;x%H;% H;4% HHT$HT$H*u HBHP0 Hc(H5'H9pH'HHL='MIGH5 (LHHHHI/u IGLP0H5+(LIHH%H9C>LHI6HI/ HI.# H;D%AH;%Du H;% AƉD$H* L$ HG(H5'H9pH'HHH'HHBH5 (HHT$ HHHT$ IMH*u HBHP0H(H5'H9pH'HHH'HHCH5(HHHIHHMHHi HJ(H5'H9pBHz'HyHH f'HJHAH5(LT$(HHL$ HH[HL$ LT$(IMZH) IAH;%W11ɿH;`%H;%= HL$0LL$(LT$ t$L IH5V(LIHVvHD$@HTH]@HHSH CHMHSHL@H%H5L @H81X;ZH 0 H=S1tHhH[]A\A]A^A_fHD$fDHCHP0A;EnfL&^IHI H^HHd L]HD$H} Iɚ; Hɚ; IH9D$4 HD$LLILT$ GLT$ HHD$( H,LT$ L\$(HI H|$ LT$ L\$(HHnHt$HE1LjAH=(%5"'H Pj5'AWj5c'ASHT$`L\$hLT$Xt(HPLT$L\$HHI*`I+6I/ H+uHT$HCHP0HT$HmtI$HHI$ufDID$LP0MImIELP0fD$ E1E11D$0<1E1E1E11Mt.I/u(IGHT$(LHL$ LL$P0HT$(HL$ LL$Ht.H+u(HCHT$(HHL$ LL$P0HT$(HL$ LL$Ht$H*uHBHL$ HLL$P0HL$ LL$HtH)uHALL$HP0LL$MtI)u IALP0MtI.u IFLP0T$t$H H="EHt$Hm1HEHT$HP0HT$HMI,$sHIHHcHHF0HD$XHE(HD$PHE LHD$HHEHD$@IH~7HHu0MlHD$XLt$@H\$HL|$PHD$HHM~HT$@HLLH5'bPy;@1fH='H;="%H5'HFt H;ٱ%HH4H=c(HHT$,HT$HI)H*u HBHP0LRI/u IGLP0D$ 11E1D$c=!E111E1D$ D$?<11E1D$ D$N<@HT$HT$@$IGHT$LP0HT$IFHT$LP0HT$IGHT$LP0HT$ICHT$LP0HT$HD$IBLL\$P0L\$HT$}DHCLT$ HP0LT$ ~HBHP0_fHALL$(HLT$ P0LL$(LT$ DHHT$ HT$ D$D$ 1E1D$O=IA@=HL$ ~D$ Ht@HHl$ LLT$0HL$(D$ Ld$PLL$ )D$@}7LL$ HL$(HILT$0HH)HAL\$0HLL$(LT$ P0LT$ LL$(L\$0fH5'LHV}HHD$XIfDD$ 1E1D$2=fDHLulH6DHFLHD$@IH5 'LHVHD$HH IH5`'LHVHD$PH ID$=D$ DI*B 1E1IBL\$(LHL$ LL$P0L\$(LL$HL$ M IHI1HICHT$(LHL$ LL$P0LL$HL$ HT$(vfD1E1D$ D$=NfDH='H'H5'n8HfDH=Y'dVH@I1D$ D$=Dn1D$ D$!=fIOHMwHII/uIGHL$LP0HL$HHLHL$ZHL$HH)$HAHT$HP0HT$ 1MD$ D$/=fDHt$@LPHfDIB@=>LL$~D$Ht@HL\$LLL$(L\$ D$Ll$PLT$)D$@3LT$L\$ HHLL$( Mt.I)u(IAL\$(LHL$ LT$P0L\$(HL$ LT$I+9ICHL$ LLT$P0LT$HL$ f.LB1 uIHt$@@ AHDH=9\N@1f+HuHD$H%H5H8HT$ZHD$ 11E1D$p<ffDHD$ 11E1D$z<6fDHuD$ 11E1D$<fDH='H;=R%H5'$HFt H; % IMH='L$HIH'HIu IGLP0HQJH+u HCHP0D$ 11E1D$<UDH5q'H="'1K4IHHII/u IGLP0D$ 11E1D$<H=1'H2'H53'4IofD11D$ D$9=H='H'H5'3HCfD1E1D$ D$u=nfDE1E11E1D$<D$ XHL$ ~D$ Ht@HHl$ LLT$0HL$(D$ Ld$PLL$ )D$@LLL$ HL$(HILT$0E1D$=D$ H=' QIM@1D$ D$;=f{HHD$<1D$ IE1E11HIpH='PH@1D$ D$w=! HT$ IfD1D$ D$>=1D$ D$z=fH= 'H'H5'1HfDD$<D$ %LL$~D$Ht@HL\$LLL$(L\$ D$Ll$PLT$)D$@JLT$L\$ HHLL$(D$=I*AD$ `*D$=D$ ILH="'-OH D$ D$|=IE1E111 HKHLsHIH+uHCHL$HP0HL$LHLHL$]HL$HH){HAHT$HP0HT$bLE1D$ D$L='H=`'H1'LT$ H5-'H0LT$ HE1E1D$=D$ H='LT$ NLT$ HtLT$(HL$ IE1D$=D$ IIHaIYHHI)6HCI1һ|MJM7MrIII*1IFM1ҿD$Q1MD$ D$=MGMMwIII/uIGHL$ LLD$P0LD$HL$ HLLHL$ LD$LD$HL$ HI(I@HL$LP0HL$1D$=D$ jHH%H5xL "jAH HKH81^;_D$ 1E1D$=D$=H5'H='1-IHHCI/u IGLP0D$ 11E1D$=E1D$=D$ HH%L PAjH5H H8Ho1AX;AYIALHL$(LT$ P0HCI1һLT$ HL$((IBLL\$(LL$ P0IFM1ҿLL$ L\$(D$I11D$ D$<H11D$ D$<uLHE1D$ D$<1E1D$ D$\=<1D$ D$^=%18!DRIH3HtHHF(HD$ HF LHD$HFHD$蚫IHHcM~.H5'LHV۱HHD$ IMLd$Lt$Ll$ {@HH'IM~H5'LHV|HtHD$IM~H5r'LHVVH_HD$ILHt$HH='HZ'H5['IfDAMH='L=R'HGLHHIMaH&n%LMH5AH81'fOAu蛳I:H=ARI/BIGLP03@ïHH3k%H5H84kH=a'lI@ALIE1 uLGHt$&LAHHfAyHbAa@H=ѵ'H'H5'IyfDA{KIHFHHD$菨IH~A@H=A'H9HEt H;,l%HHHH='HIHMHHu HCHP0Lq I,$uoID$L续AP0VAHT$HLL=H5K' B #I5AH3HH-AH H=:EIHA1ɺLAHfAWAVAUATUHSHHi%H^H|$ HDŽ$H$H_HeHLn H]HDŽ$HHDŽ$HDŽ$HD$HH' H(hHE1AHHH$HH$HH$H/uHGP0Hh'H$HDŽ$HDŽ$H5'H9pHȩ'HHL5'L$MIFH5'LHH=HH$H$HH/uHGP0H'H5D'HDŽ$H9pH'H6HL5 'MIFH5o'LHHHHI. HEH;qf%E1E1H;g%H;h%d L$HcL$HI MtLpHB'IHK\HIDHELM˩L=f%H QP A;1LHAI蟩P P A=X 29MDL$I,$u ID$LP0Hmu HEHP0H$HGH;Ne% L$H;f%L$H;tg%vHWBgHjE1 uLwL=e%H QP A;LLH躨P P AH= @9HH$H$H/uHGP0L$HDŽ$MH$H/uHGP0H$H;=e%HDŽ$AH;=0e%D H;=ue% JH$D$H/uHGP0HDŽ$L$qHCL;-&e%HD$8+fHD$(H HH(Ht H;-d%X HPHuL`H@HD$HF MtI$HD$HtHH'H'H9PHz'HAHH=f'H$HHGH5'HH2H$H$HH/uHGP0H$Hb%HDŽ$H9G]LH$H$HtH/NH$HDŽ$H$H H/uHGP0H|$HDŽ$H$OH$H ޥH$OIH$HB ~$HDŽ$HDŽ$$HDŽ$@Ht HmMt I,$Ht$HtHHD$HHeDH9'H 'H9H8H'HxHL5u'L$M4IFH5'LHHHH$H$H8H/uHGP0H'H 'HDŽ$H9HH'HOHL5̢'L$M[IFH5Ȯ'LHHIH$MH/uHGLT$P0LT$H$HDŽ$HAH;"`%4H;a%H;Xb%BLT$LT$HIH$HtIFHDŽ$HcD$I1LM|MTH$HH$HoI._ H$H/- H$HDŽ$HH$HD$HEuHDŽ$H} HD$H'H|$HD$0~HCH5k'HHHIL$MH^%I9D$ Ml$M ID$IEHH$H$H/uHGP0H$LH$Imu IELP0L$MH$H/uHGP0H5'1H$HDŽ$讞IH$HH$H/uHGP0H$H;=n_%HDŽ$H;=^%eH;=_%XH$Aą\H/uHGP0HDŽ$E HG'H'H9PHw'HHL%c'L$MrID$H5֫'LHHIMH$H/uHGP0HDŽ$H\%I9D$=HLMIH$L$HtH/uHGP0L$HDŽ$MI.u IFLP0L$L;%]%HD$IL$fHDŽ$HT$8HHL$HHD$PXHHuHD$ H5x'HIHHD$ H5|'HIHUHDŽ$H[%I9FUIFH$H@IVHHI.uIFHT$(LP0HT$(H$H HHT$(nHT$(H$H$HtH/H$HDŽ$HH*u HBHP0H$H/uHGP0HDŽ$觡H|$0HD$XHD$Lt$ L|$`HD$(HHl$ Hl$PI HD$@HD$Ll$hLl$HLd$HLd$8H\$8HfDHtKE1CLCLT$T$f(Y\YB II9uHt$Ht$(HD$(TH\$@H9D$0Hl$ L|$`Ll$hLd$HH\$8H|$XĘH55'1LImIu IELP0M3I.u IFLP0Ll$IEH+2L<H]fDHH4H $HOHBL >LOL@HHY%SH*H5eH81苠XNZH KH=1J*HH[]A\A]A^A_DIIFLP0]HE@=Lt$~D$HHH\$Hџ'HH$D$)$IH$HH$E1ID$YOD$D@DL`H@HD$HEfAljD$ IHs H H`H(H5Y'LIHVI~H$H#MTH$L$Ls@$f.D$OE1E11D$E1E11HD$H$E1@Ht H/QH$HtH/uHGLT$ P0LT$ H$HtH/uHGLT$ P0LT$ MtI.uIFLT$ LP0LT$ MtImuIELT$ LP0LT$ MtI*u IBLP0T$t$H NH=_'HtH+E1HCHLP0HtHmu HEHP0MtI,$u ID$LP0MtI/u IGLP0HL$HHHD$HHHAHP0HGLT$ P0H$LT$ f.1Vf苙@$fH|$fHH$H8IH$HH$HDŽ$HDŽ$IGLt$~D$HHH\$H>'HH$D$)$yIH$HMMbI.XIFLP0IOHt H/< H$HDŽ$HtH/uHGt$0P0t$0H$HDŽ$HtH/uHGt$0P0t$0H H=HDŽ$\%H|$(H$H$H$*HT%I9EJIEH|$貖IHlHD$0臗LT$0HILPHLIHpImu IELP0I.u IFLP0H$Ht H/uHGP0H$HDŽ$Ht H/uHGP0H$HDŽ$Ht H/uHGP0HD$(HL$LHHDŽ$HLgE1E1E1E1D$)OH$D$HD$E11xfH$E1E1E1D$8OD$DH='H”'H5Ô'IfDHGP0@HGP0@IFLP0HFHP0ID$LP0\HEHP0=H=1mT MIE1E1H$D$jOHDŽ$D$L菘IHD$*PE1E1D$HD$(HL$LHLT$ E1E11HH$LT$ HD$L IH$E1E1E1E1HD$E1D$:OD$˘HH= 'Ip@HA@=H$HHHL$LT$H$L$H$LT$HHH$HtH/uLT$HGP0LT$HDŽ$I*IBLP0fH$E1E11HD$E1E1D$=OD$@H='H'H5'IfDLHHD$ H5'HIH.HD$ H5'HIHeHDŽ$H=O%I9F IFH$H~ MnHIEI.u IFLP0H$HS LH$H$HtH/uHGP0H$HDŽ$H Imu IELP0H$H/uHGP0HDŽ$*H|$0HD$X1HT$HD$8Hl$`L|$hLl$HLd$pHD$@HHHHT$PHH\$xHD$H1HH|$HD$ M|Lt$8H\$(Ld$@fLHh AHIHL$赍L$CXM9u%H|$H\$(^f(f(HT$H1fҐfAfYAHH9uHD$PH9D$tHIDYH\$H9\$0,Hl$`L|$hLd$pH\$xH|$XH5o'1LI,$Iu ID$LP0M I.A9*f.E1E1E1E1H$D$?OHD$D$ffD胔HH='I@LuM LeII$Hmu HEHP0ID$L1ҿAfDHoHHGHEHH$H$H/uHGP0H$H$HH$Hm)HEHP01E1E1E1D$|OD$H$HfDDD$OH$D$E1E1E1E1HD$E11fIE1E11H$D$_OHD$D$onH)$HH$HLLH52' NcHFHH$輈I@H5'H=2'1[HH$H HH$H/uHGP0H$D$OHDŽ$D$1LHIH$HSH$MIE1D$jOD$fLJE1 uLGH$f LAHfDHGt$0P0H$t$0@諍HHDŽ$YH=qd蔌2@1dH$HHHL$LT$H$L$XH$LT$HHH$]E1E1E1HD$D$vPD$HHHgHXH%H5cH8Y@H5A'LHVHzH$IE`H=Y'HJ'H5K'FIfDH$E1E1D$]PD$)DH= 'I@E1E1E1E1HD$D$_PD$VfDsHrH=A'H'H5'H8fDH$OfDH$E1E1D$bPD$iDH=I'H*'H5+'6IJfDH=',H@O2fD諎fD1MfE1E1E11HD$D$dPD$?[IH='I@HGH$HHWHHH$H$H/uHGP0H$H$HLLٞG@HqH$HHAHHH$H$H/H$1ҿD$HAlHGHT$(P0H$HT$(0LHHT$(HT$(H$E1E11HD$D$PD$H$E1E1HD$E1D$PD$D$PE1E1E1D$`E1E1E1E1H$E1HD$D$OD$`ML H$E1ID$QOD$H$E1E11E1E1HD$D$OD$I9H$E1E1E1D$PD$HGLT$P01ҿLT$H$D$HAH$E1E1E1D$PD$jH=ێ'H'H5'I5H$E1E1E1D$PD$"E1D$,PD$H={'IH$E1E1E1D$PD$IH$E1E1E1D$PD$D$.PD$SH$E1E1D$PD$eID$H$HMt$HII,$u ID$LP0H$HaHLQImu IELP0H$D$E1E1ID$LQH$E1E1E1D$QD$E1E1E1E1H$E1HD$D$OD$D$QI,$u ID$LP0H$E1E1E1D$sMD$QH@%H5\H8舁QH$E1E1D$QD$(H$E1E1E1D$RD$E1E1E1HD$D$PD$1ɺLAHaIM@f.AW1AVIAUATA̹USHHH|$@HT$H|$ HHLD$0DL$)|$PAo)$tsow)t$ o~)|$`Aop)$tLo )|$0oF )D$pAoH )$t%og0)d$@on0)$AoP0)$ȃ9t"HH`IHaIHD$HH.Lp.cIHbHr'H _'H9H{H_'H.HL%_'M@ID$H5Zj'LHHHD$PHI,$u ID$LP0HT$PH5ik'LcH|$PH/uHGP0Ht$HH|$@LHD$PlIHD$PHH|$@H/uHGP0HD$@H|$HH/uHGP0HD$HI.u IFLP0Hl$PHD$PH;l$HEIH5Bj'HD$0pHD$(HIH5Dj'OIHHD$HH%I9FIFHD$HHMfHI$I.u IFLP0Ht$HHLTH|$HHD$PHtH/uHGP0HD$PHD$HH0I,$u ID$LP0H|$PH/uHGP0HD$PYHL$hHT$`Ht$XHHL4$IIfMIE Hl$8Hl$0MH$HD$ H\$ IHMLl$H<$L`HL$LHHL4L]LLL]LLH]I)IuH\$ Hl$8HD$H8HD$Pu HH@P0HD$PH|$XHt H/uHGP0HD$XH|$`Ht H/uHGP0HD$`H|$hHt H/uHGP0L|$(H5_'1HD$hL虦HD$hIH$HIu IGLP0H|$hH H/uHGP0HD$hHD$HHmL|$-Ld$1 ]@$ fHD$PHtpHLL|$xHD$pL$yHD$@HH|$PHt H/GHD$PI,$ID$LP0$=H|$@A/S1E1E1Ht9H/uHGP0MtI.u IFLP0MtI,$u ID$LP0H|$HHt H/uHGP0H|$PHt H/uHGP0$H eDH=WAHtHmuME1HEHLP0HtH+u HCHP0M*I/ IGLP0fD1@H|$@1A9S$AH=b'H:Z'H5;Z'覤ImfDH=b'HZ'H5Z'~IfD$AA;SH|$@1HHIHPH%HH9HD$H;-%ADA H;-6%A Hu HWR0HD$@EL[AąL|$AS$CE1_fHGP0H;%r HIVHHD$PHH|$@AV$fIFHHHD$@Q]IA=S$AfDH=!a',I@Hk'HKH9%HH5^a'HHHHl$HHH%L=%H9HD$L9H;-%HZHl$HAąHmu HUHR0HD$HEH5^'HѣHHD$HHH;D$L9 H;-C% HZHl$HAąHmu HUHR0HD$HEFHkj'H $W'H9HH W'H3HL5V'MIFH5\'LHHILd$PMI.u IFLP0LL$PIAH;%E1E1H;~%r H;9%t$ $WHHD$@HMHc $Hct$tLpHZ'IHJ\HHT$@HDHD$1HHt$@HDH|$P{H|$@HHD$HH/H/uHGP0HD$@H|$PH/uHGP0HD$PLd$HH+u HCHP0Hh'H U'HD$HH9HH_U'HHH-KU'Hl$PHH5fa'HvHD$@H~H|$PH/uHGP0ID$H5:X'LHD$PH@pH H@H HHl$PH|$@HHM%H9GpHCHD$HH|$PH/uHGP0Hl$HH|$@HD$PHH/uHGP0IH5M`'HD$@Hl$HHD$HiIHIH5?`'JHD$@HHD$PH %H9H HPHT$PH H@HHH|$@HD$@H/uHGP0Ht$PHD$@H HBH|$PHD$HHtH/uHGP0HD$HHD$PHH|$@H/uHGP0HD$@H|$HH/uHGP0HD$HtOHL$XHT$`HHt$hHD$(LUHHXH6IE L|$ H$H<$HVIH911HLeHD$HHtH5%HH5NH|$HH/uHGP011HLHD$HHHD$HHID$L=^%L9;H@hH1H@(H$LLЅH|$HH/uHGP0HD$HID$L9 H@hHH@(HHHLЅHH|$hHt H/uHGP0HD$hH|$`Ht H/uHGP0HD$`H|$XHt H/uHGP0H5MU'1LHD$XHD$XI.u IFLP0H|$XH\H/uHGP0HD$XH%HD$HHL|$LHALS$AKfDHGHD$HHHWHHH|$PHT$PH/uHGP0Ht$HH|$PHHgIIVHT$PHInHHEI.u IFLP0HEIA1ҿIFHAnS$A{fDDL1LTIHD$@HtAyS$A1fHFHHD$pKIooNH)L$pnKHVHT$pHLL_H5,&g0R.DH=1(TPHt$HD$@AySE11$A8 QHAyS$A1HD$@D@H5a'H{萵H|$@L|$AS1$Gf.H|$@1L|$AS$HvfSIH5!\'LHVPHHD$xIFfDIA@=L4$HR'HtpH~$H$LH$HD$$H$)D$p蹕HHD$HHMI.IFLP0fD1MHHHL2LHmI4HEHP0%ff.AL= %HZ %E1II9LHD$AOH H|$@L|$AS1$H@kRfD$IASL|$1L|$AS$H1RHL|$AS$I}fHl$fDHD$@I>E;NHIL|$ASE1$IDH=T'H:L'H5;L'ƖIfDH|$@1L|$AS$Nf.M%L|$ HD$H$I H\$0Hl$MMILILLMOILH0H3HuH0HHEL)IuH$Hl$D MLD{H=S'I@PIL|$AS$N1H|$@1L|$AS$NHCJHLLHAU!HD$HHD$@HD$PyAUnAUcLl$HHl$P>LLHHAVŃHD$@HD$PHD$HhAUAVRAVGAV<E1AT$YH|$@L|$AV$gH|$@L|$AV$iH|$@L|$LAU$[H$H9HD$H;-3$DH;-u$HG>Hl$PAąH|$@L|$AV1$fMfAWfAVAUIATUSHH$H^H|$()D$pH$H:H8 H Hn(IEI] HD$HD$`HHD$hHD$pJ=H$HHN'L(hE111AHH|$AIHoH8u H@LP0HM' L(hE1ɹAHHAIHH8u H@LP0ID$H5A'LHD$ _L'nH$H|$ Hp;L'f/?Aw H5C'LL'H;-$HL'H 8'H9HFHg8'H.HL5S8'MJIFH5E'LHHHIHHIHu IFLP0c:IH?HEHhy'HHnHHl$pHtHHt$zI'HHD$`HH|$pH/uHGP0HD$pH|$`H/uHGP0HD$`IEH5='LHHHHl$`HH<$7HHD$pHt7HHSHD$pH|$`HHEHD$ps3HHD$pHH|$`H/uHGP0HD$`Hmu HEHP0HD$pH4'HD$pHD$HeI'H9P H4'HHH-4'Hl$pHHEH5:'HHHzIMH|$pH/uHGP0HD$p6HHD$pHHL$HHD$pHH8HHD$`H HH'H 4'H9Pc H3'H HH3'Ha H5@'HMHHD$hH H+u HCHP0HT$hH50A'H|$`F9H|$hH/uHGP0HT$`Ht$pLHD$h1HHD$hIHv HIu IFLP0H|$pH/uHGP0HD$pH|$`H/uHGP0HD$`Lt$hIH\$hHD$hH5@'HCHD$8IEHD$0HD$(H#HHHD$(H5?'HHD$`HHD$pHI$H9PHPHT$pHyH@HHH|$`HD$`H/uHGP0Ht$pHD$`HCHmH|$pHD$hHtH/uHGP0HD$hHD$pH|H|$`H/uHGP0HD$`H|$hH/uHGP0HD$h97H|$0HD$@,H $HD$(E1Ld$HL\$8Ld$0L|$8HHPHH Lt$0HL$(IIHl$PLH\$XL@I0MHLHL$ L$H0H0 1AMIE 1,xH(H0H0H@(A;u}NHcILH0H@H0Pt8V H(HR8HcR H0A;u|HHl$(I9=L|$8Ld$HLt$0Hl$PH\$XH|$@.H55'1H1|HmHuH$HEHP0H $HH)u HAHP0ILg I]@HH<H {<HMHSHAL@HM$H5L x=H815XDZH O< H=c.1謿HĘH[]A\A]A^A_HH;-$ ,HD$8HHL(Mt L;-n$HPHuHHH@HL$HD$0M~HD$HtHHD$0HtHHC'H .'H9HH.'H@HH=.'H|$hHHGH5;'HHHD$pHH|$hH/uHGP0H|$pHF$HD$hH9G[H3jH|$hHD$`HtH/z HD$`HD$hH{H|$pH/uHGP0HD$pH<$/HD$pH90HD$HD$hHHD$h~D$`HD$`D$p@HD$pMt Im HL$HtHHD$HHk HT$0HtHHD$HH< @HA'H :-'H9H H!-'H HH- -'Hl$`H HEH5L3'HHH HHl$pH|$`HY H/uHGP0HD$`a/HHD$`HHT$HHD$`HPi1HHD$hHPH!A'H Z,'H9HHA,'HbHH-,'H4HCH58'HHHDHHHHHH=H59'H|$hH1aHmFHT$hHt$`H|$pwHHMH|$pH/7HD$pH|$`H/?HD$`H|$hH/HCsHD$hH{ HHD$8H@'H|$HD$0^HD$HfH=5'H*D$>'HD$(L5E8'LLIUHHT$/HHxH@HT$HHHLHH]HD$(H5 8'HHD$hHHD$`H [$H9HQ HPHT$`H? H@HHH|$hHD$hH/uHGP0Ht$`HD$hH H fIH|$`Ht H/uHGP0HD$`MH|$hH/uHGP0HD$hI.u IFLP0]/HD$HD$0HH<$HHHD$(H $E1L|$(Ll$8Ld$0HPHHH Hl$8H\$@LLIIIIfDL$HL$ HMHt$LHLE)I9uL|$(Ld$0Hl$8H\$@H|$&H5C.'1HtHmHD$u HEHP0H|$8HT$HH$HHu HBHP0HHIE1HD$HD$@HHH@HL$HD$0IEx@HH~yHtHaHF(H$IE HHD$xIEHD$pc%IHtsHHt?MHD$pH\$xH$HD$ifDHH%IH52'HIHVe+HD$pHH50'HHVD+HD$xH^IMwH5Z/'HHV+H7H$IA@HD$E111$=E1E1E1HD$H|$`HD$D$ DfHt H/uHGP0H|$hHt H/uHGP0H|$pHt H/uHGP0HtHmu HEHP0$t$ H \31H=s%辶MtI,$u ID$LP0HtH+u HCHP0MtI/u IGLP0HL$HtHH$HHu HAHP0MtImu IELP0Ht$HtHH$HHu HFHP0H\$HtHH$HHu HCHP0MFI.D$ E_H<$'HHD$hH3'HHD$HD$`HHD$`HD$hHD$hHBfDH|$`E1E11HD$1HD$HD$$?D$ Ef.HCHP0HGP0z@HGP0@HEHP0HGP0@HGP0@HGP0@HGP0@HGP0@IFLP0HCHP0tHBHP0HAHP0IELP0WHE-H58'HxLl$pH|$`E1E11HD$1$MD$ E}X*HE1E111HD$HD$$^D$ FRu;HP0H;0HHP0H0H0H0:*LcJE11HD$$ND$ EH$H|$`E11$PD$ EH='H'H5'aHHD$8HL$0L1HT$E1HnIHD$HD$H$cHD$D$ UGFH|$`E1E11HD$HD$HD$$DD$ CEI~H='H9H|$`11$PD$ ED$ WGHD$8HL$0LE1HT$HmI$cHD$HD$HHD$uI1$PD$ EHXHH$AH50jL !H H8H%1?_DAXL1D$ \G?I1$PD$ EHI$P11D$ FHH|$`IE11HD$$lD$ =HmH='H'H5'_HI1$PD$ EH^H=#%11H ?# [E1:H|$`HD$HD$HD$$ID$ wEH=' }H I$PD$ EHH1$PD$ FH|$`$UD$ FHmu HEHP0H|$`$U1HD$ LFH|$`1$UD$ F#AWfAVIAUATUSHHQ$LfH|$)D$pH$HjI0 I.Lf(Mn MvH&' HD$@HD$HHD$PL(HD$XHD$`HD$hhE111AHLAHHH8u H@HP0H &'D{H(hE111AHLHHH8u H@HP0EED$Lf.3D$LyIHH=F'D$j$'YfH=A'I*F$'L;%$H%'H 'H9HH'H0HL5'Lt$HMIFH5'LHH)ILt$@MH|$HH/wH$'H y'HD$HH9HHW'H*HL5C'Lt$HMIFH5'LHHIMH|$HH/:HL$@H54$HD$HAHAHt$ H9H;$H;O$HHHD$HHt HAHD$HHcD$I$1HHL$LdIcLtH|$@ZHL$HIHD$P\H)u HAHP0H|$@H/uHGP0HD$PHD$@HH|$PH/uHGP0Lt$@H#'HD$PHD$@AvI~ H5{'HD$IFHt$8HD$0HD$LIT$HHT$(2IH)H@HT$(HHLLIH HD$H5-'H1dHD$PH HL$ H9HL`MH@I$HH|$PHD$PH/uHGP0H|$PLFIHD$@I,$u ID$LP0HD$@HH|$PH/uHGP0HD$PH|$@H/uHGP0HD$@H|$HD$ ~iHD$Ld$0L|$Ht$Hl$(HHHH H\$LI4IIHD$LLLHXHEH9uL|$H\$Hl$(H|$ 6 H5'1LUXHD$hI/u IGLP0L|$hMI/u IGLP0HD$hI>QME1fIFLMP0MtI/u IGLP0HtH+u HCHP0HHmHEHP0wMfDIHCH 3HMHATH\L@H$H5L /H81X5ZH  H=J E1bHĘL[]A\A]A^A_DH5'H'ZH5'H'\L;%%$ H'H 'H9H'H 'HW HL5 'MIFH5('LHH-HHD$@IHIu IFLP0H'H 1 'H9HH 'HHL5 'M#IFH5'LHHUIIHMIHu IFLP0LL$@H $IAHL$ H91E1H;$? H;$L$t$5 IHD$HHMt$HcL$tLpHcI$1LHMdMlH|$@-UIH H|$HH/ HD$HH|$@H/uHGP0Lt$@I>uIFLP0Lt$@HD$@E1H'AvI~ HHLL |'HD$1AIHH;$HD$H+Mt I,$ IGH5'LHD$HHH4HD$HH LH'HD$@HjH|$HH/uHGP0HD$HH|$@H/uHGP0HD$L-9'HD$@LLIT$HHT$( HD$HnH@HT$(HH5 LH|$HD$HHD$L-'LLIT$HHT$( H<HpHT$(HH; LHHD$HHjHpHD$PH;t$ % HPHT$PH H@HHH|$HHD$HH/Ht$PH|$HH BH|$PHD$@HtH/HD$@HD$PHH|$HH/<HD$HH|$@H/HD$@E1 HD$ HD$HHHH H|$Lt$0MLd$IH\$HHl$(LDH0HLL0H@H0H8H0H6 1IMHE %pH(H0H0H@(;}}MHcHLH0H@H0Pt8H(HR8HcR H0;}|IM94IH\$Hl$(Lt$0H|$ Ll$H5d '1LQHD$hIEHD$HIEH|$hHH/~HD$hI>MI/H+Iu;HP0H;0HHP0H0H0H0HcLMH(M9(iI@(H0H(H)0+I@(H0H(H)0PH0HcLMH(M;(}ILL(H0H(H0H@0H0H@(H0H(H+0H0II3 ~qItIHF(H$IF LHD$xIFHD$psIItkIMt7M%Lt$pLl$xL$9MOH'IH5 'LIHV}HD$pHH5h 'LHV\HD$xHIMH5r 'LHV.HH$II@D$61E1E1A f.H|$@E1Ht#H/uHGP0MtImu IELP0H|$HHt H/uHGP0H|$PHt H/uHGP0t$H DH=MI.u E1E1IA@= Lt$~D$HtpHLd$LL$D$)D$pJIH5MtI.u IFLP0Imu IELP0ME1E1A D$#6H 'HH1IHH;$HD$@(Hy'H &HD$@H9H)Hp&HaHL5\&M IFH5'LHH ILl$HMI.u IFLP0ID$H5N'LHH_HH8H'H5&H9pFH&HHL-&MNH5B 'LHL$mMHL$HIpImuIELP0HL$LL$HH5$IAHt$ H9RE1E1H;ܿ$8 H;$ HL$t$HL$HHD$PMHct$tLhI1JLLtH|$HHJIHD$@HaH|$PH/uHGP0HD$PH|$HH/uHGP0HD$@HD$HHH|$@H/uHGP0HD$@Lt$HHD$HID$LP0HGP0R@HGP0}@HGP0@HGP0@H|$HGP0)HGP0@HGP0)@HGP0J@HA@=HD$HHtpHHLd$xHD$pL$FHD$PHH|$HHt H/uHGP0HD$HI.IFLP0DHGP0v@H=y'gID$8E1A H)u+E1HAHP0MtIHIHu IFLP0H|$@E1H?[E1E1H H=A D$6HD$@DE1E1A D$N6GE1E1A D$W6'HD$HfLt$~D$HtpHLd$LL$D$)D$paIHA D$6HHD$Hf.H|$HFH9HD$@L|$H5 'I0RIH IH5 'RHD$HH HD$PH [$H9HHPHT$PHH@HHH|$HHD$HH/uHGP0Ht$PHD$HH{H 7H|$PHD$@HtH/uHGP0HD$@HD$PHN H|$HH/uHGP0HD$HH|$@H/uHGP0HD$@;HL$XHT$`HHt$hHD$ CHL$D$LHy HQH4H,IHD$@H HD$ HL$XHD$@HT$`Ht$hHS1LL|$XH5Y& FHD$`I.u IFLP0Lt$`M I.u IFLP0HD$`Lt$XHD$XH+E1A D$k6Zf.H='Hz&H5{&EIfDD$m6HA E1E1?#fDHFHHD$pgI`H=A'H&H5&.EIfDE1A D$p6f. %HE1E1A D$7c{IHE1E1A D$7)fH='bI0@A D$r6DE1E1A D$7INMqM~ IAIHH|$@HD$@H/uHGP0LL$@1ҿIAE1E1A D$7WH q[MH= A 蚇D$7HD$H@H5 'Hxp_L|$HfDA D$+7DfDHD$HHtpHHLd$xHD$pL$Y\HD$PH-IA D$8H[A D$-7DA D$6mDH='H*&H5+&BI9fDE1A D$8fH9$LA H8HD$:7E1A D$6fH=)'4`I@E1A D$8fIIA@=bLl$~D$HtpHHL$LHL$L$D$)D$p>HL$HHD$@MtImuIEHL$LP0HL$H)u HAHP0I.VIFLP0GH $LH8D$<7HD$HHL$A HHD$HHHAHP0~H=&HX&H5Y&@IE1A D$8FD$J7H=&^IE1A D$8-IHqHt$HHHAHHH|$@HD$@H/uHGP0HL$@A1ҿHAD$HT$pLLLH5D&/S5pH$Ht$8A H8D$8VHD$8I/uEIGLA E1P0#IA D$8HA D$7E1A D$ 9Ll$~D$HtpHHL$LHL$L$D$)D$pgXHL$HHD$@sD$6HMA HHMD$6A XH5'HxZLd$@H=&H@&H5A&>I/IMLA D$6H=W&b\IHH$H5(L jAH HH811^5_ME1A D$6|HH=&Hi&HL$H5e&=HL$IMD$6A BA D$K9H=&HL$[HL$IbMD$6A MiMIAIEHH|$HHD$HH/LL$H1ҿAIAuE1A D$8D$8I.uIFLE1E1A P0YE1E1A HD$8D$6H|$HHt H/uHGP0HD$HH|$PHt H/uHGP0H  )8HD$PH=)DH|$ HL$PHT$HHt$@DJHL$PHT$H1Ht$@xH1HLHD$(;I.HL$(HuHD$0IFLP0HT$0HL$(H)uHAHT$(HP0HT$(HH;$H;Z$hH;$[HHT$(mHT$(AH*u HBHP0EHt$@EHtH.u HFHP0HD$@H|$HHt H/uHGP0HD$HH|$PHt H/uHGP0HD$ HL$XHD$PHT$`Ht$hHG3MD$7A e1MD$ 7A HGHL$AP0LL$H1ҿHL$IAA D$6E1A D$8H*DIA D$8HlE1E1A D$6iD$@8HD$ HL$XE1A HT$`Ht$hHF3E1D$6Ll$PLd$HHt$Ht$LLH5D$U8HD$@HD$HHD$P_D$M8RD$I8ED$D88A D$8A D$:7sD$<7fDAWIAVAUATUSH8Hج$L5I&H<$HnHD$H\$Lt$ H\$(HHF~,HHLn0Mw(Mg f.IIHIGHD$H=P'Lt$HGH;p$ H;+$HWBHjE1 H QP H$;LLHrH QP Hb$=@9HH'H &H9HHw&HHL5c&MIFH5G&LHHIIHMIHu IFLP0LHMIHI/u IGLP0L;5"$L;5$u I9hDI.u IVLR0EH $HH&AH=$LHq IH&LAUjRPjRLPj5A&LL$X&HIHPHHI.fDHmu HEHP0H8H[]A\A]A^A_DL29N@$@IdHI&H &H9HH&HHL=&MIGH5&LHHIIHMIHu IGLP0LHIIMHIu IFLP0L;=N$L;=Ĩ$I9LAƅI/u IWLR0EdH$LH=6$LHp ILL }&AUjLD$&Y^HHI/FIGLP07LPADžI(A_HIHu IFLP0H D1H=xfIFLP0LIH5&LIHVNHD$H4Io@HH,H HOH:L 6LOLD@HH$UHH5\H81Xj(ZH H=X1Aw%@ILeHHH ^(1H= vDHIHHcHHF0HD$(IG(HD$ IG LHD$IGHD$IH~aHHu*M~.H5&LHVHHD$(IMHD$Ld$Lt$ Ll$(HD$Hu5f.Ht$QKHfH=&H&H5&1IsfD(A_[It(A_]H=&L=&HGLHHIM HV$L(H5A_H81WfH=AdJA_(I/IGLP0@HH$H5H8H=&NI@HFLHD$IMAH54&LHVHtHD$IMH5&LHVHHD$ I(AeLJE1 uLGHt$LAHH=&H&H5&/I1fDH(A`q@(Aeo;I"Ae(kH(Ae)@H=)&H9HEt H;,$HHHH=&HIHMHHu HCHP0LqDI,$uoID$L)AjP0Af(HT$HLLH5&A@T(#I)Aj:H3HH-(AjH j(H=/rIH)Aj1ɺLAHfAWAVAUATUSHH(HFv LfIT$HH HcHHCH5&HHHIMHHHHHH5&HHmHmu HEHP0H&H &H9HH&HDHH-&HHEH5}&HHHBIMHmu HEHP0H5$&LH9Imu IELP0IFL-&HHL=$H QP A;oHLLIH QP AH= @9M_I.u IFLP0H+u HCHP0IEH5&LHHIMImu IELP0IFH;W$M~M I^IHI.u IFLP0H&LHI/Hu IGLP0HH+u HCHP0HEH5&HHHHHEHHEH"Hu HEHP0HCH;$LsMLkIIEH+u HCHP0Lt$IEH;$jH;$IUBHZ1 uIm L=$H QP A;HLHH QP A=29HI.DImu IELP0HELhpMI}L#HHH=Н$HHH+Iu HCHP0MuLHAUI,$uIT$HD$LR0HD$HUHJHBHMHuHUHD$HR0HD$H([]A\A]A^A_fDDfFII LIl$HHIHHeDDfFII IIuIH/H HH=l1fDfAMc@DfmH="@I.u1IFLP0HELDHPH$H5vH814HUHBHEHtSAADDH zH=+kH(1[]A\A]A^A_DHt3AADHEHAAP0HEHAAP0H-)&H;$Hl$H;h$:IVBHZE1 uMnL=К$H QP A;HLHH QP A=29HLH;$H;̛$HKQE1 HiuLk9L=2$H QP A;1LHH QP AH= @95IH9\HuA H+AE11sfDHHI@$fAAsINE1AAI.u IFLP0H+u HCHP0HtHmu HEHP0MsImhIELP0YI.u IFLP0AA8E1AAk@@!DAA1DDH=&H&H5&$HfDE1AA@H=&BH@AA#I@ED@$nf.LHL"IH/fD1AAdD@$f.I/AAeH=ٱ}E1AA1fDHHgH0$H5AAH8%H=q>@LAf.@$fHHLUH*DHL-b&HL=$H QP A;1LHK@LLeUHI.HMHt$L;HfDXLB1 uI~Ht$AHjD11HIy;HHt$L^;HdfD+HH$H5H8LB1 uI}Ht$aAHDH=\H H$H5H8H@`HdHHTHIHCHPH;$f`MeIL$HH5HcHEeAMcImIULR0@EeAEII I@EeImtnEeAEII ImLE1R08f.H52HIHLHP@L`IBL1HIHH=$HH*$H5H8sHW$H5حH8XpHB`HHHLHHH$H9EHEt#LeID$HHYHcHHLPIHmHHEHP09DeAMcDeEII IDeDeEII HH5 HH[ImIELP0DHIg1ɺAHHuHő$H5\H81ɺAHf.AWAVAUATUSHXL-3&L%&H-$H^H|$Ll$~D$Ld$HD$@D$)D$0H H ~0HHHF(HD$Lf LnfHD$HH+& H(hE111AHLHH H8H& H(hE111AHLHHT H8zE;CH&H&H9P>H&HHL-&M IEH5&LHHIIEHM>IEHI@H;>$xE1E1AH;$ H;^$LD$LD$HIMtLxIHEKlHK\I@LM!L\$ LD$LD$L\$ P L5~$A;P o 1LL\$(LLD$ AHD$PLD$ L\$(P P A=@9 H|$=I+uICLD$ LP0LD$ I(u I@LP0HD$HHH&0IHn H8u H@LP0H&H&H9Pf H&H HL={&Mj IGH5&LHH IM I/u IGLP0H&H &H9P H&HHL5&MIFH5_&LHHIMI.uIFLL$ LP0LL$ IAH;Ɍ$ H;D$Ld$0H;$ IQBHrE1Ht$( uMyLL$ _L5X$LL$ p VA;P LL$ LLHD$(I'LL$ P P A=k29 MMMI.u IFLP0H$I9E LLMII/uIGL\$ LP0L\$ IHMIHu ICLP0L;5$L;5$u L;5Ό$DI.u IVLR0EHT$HIAH=$LHr IEH&L5r$jP5a&jPATjPHT$`&IHPHImu IELP0H+LHmu HEHP0MtI,$u ID$LP0HL$H1HHD$HHHAHP0 fDH@HP0CH@HP0E;C|tLZf.D$vL>f.~\D$H&H&D$ H9PH&H HL-&M IEH5&LHH IIEHM IEHu IELP0D$ IH H[$I9F1 LLMKII/u IGLP0M Imu IELP0L;%$L;%$u L;%O$yDI,$u ID$LP0E HD$D$LIL\$(ZL\$(HID$ L\$9L\$HIHt$HLMHB&5$AH H=È$jP5&jPATjPHT$`L\$X&HPL\$HIIbHIUIm:I,$HD$E1 fDHD$[fDHH|IH iHHIHHF$I?SIH5H8L fA1X$ZH <H=1XHXH[]A\A]A^A_IELD$LP0LD$I@@= L|$~D$Ht0HHl$LLD$ H\$@D$)D$0'LD$ HHD$D$%@MIME1E1HD$HD$Vf2?D$FE11E1HD$E1D$$E1MtImu IELP0MtI/u IGLP0MtI.u IFLP0T$t$H H=.WHtH+OE1HCHLP0HHD$GfDHH3HtHHF(HD$@HF HHD$8HFHD$0IHHM~.H5?&HHVH HD$@IM HD$@Ll$0Ld$8HD$H/H?IM~H5h&HHVHtHD$0IM~H5&HHVnHWHD$8ID1fHD$E1E1E1D$GD$% fL\$(LD$ 1L\$(LD$ @$fDL|$~D$Ht0HHl$LLD$ H\$@D$)D$0*LD$ HHD$ MI/IGLD$ LP0LD$ D@DLAŅ)MD$\E1D$7&&fLAŅxI$ME1E1D$`%E1D$MHHD$IHE1ICLP0MIHIHcIGLE1P0fID$LE1P0HD$@IELP0ICLP0HvHD$E1E1E1D$JD$,%DZD$ D$ H@HD$E1E1E1D$KD$6%fH=aLD$L\$ sIME1E1D$%HD$VHD$@H=&H&H5&IfDE1GHD$E1E1D$VD$% @H=&HR&H5S&IfDHFfLL$ LL$ @$D$VD$%ME1E1E1HD$fDH=&$,IA@I\MxMM`II$I(u I@LP0ID$MA1ҿATfDE1E1D$[D$%DH=&H"&H5#&n IfDE1D$\D$&D$%;LLLL$ ?LL$ IH= &+I)@IDD$\D$&G1LLL\$(LD$ LD$ L\$(HHD$IEMMD$$&D$\E1H9fLL$ƾLL$HIEHMME1D$$&D$\fD$^D$d&IE1E1HD$D$%HD$QDIE1HD$D$%HD$RMHE1HD$MD$%D$PnHT$0HHLH5\&?$D$%D$]E1E1D$G&IEMD$$&E1D$\HAE1Hx$H5\L\$E1LD$H8ϹL\$LD$D$%D$VIMHvE1D$]D$C&BHD$E1E1D$ND$l%HSx$H5ԓLL$H8OIELL$H1ɺALL$ IHD$E1E1MD$VAWI1AVAUATUH1SHxH=A&HD$8HD$@HD$HAHD$8HH\Iw`HPHD$8IH6H x$IHHHHT$H9t HHPHuHHH|$LpHL$t HD$HHD$HtHMtIHCH5ƾ&HHHCILD$@M@Ip?IHHD$@HHH; v$}LHMH@IHH|$@HD$@H/XHD$@E1ۿHHH; w$sH; x$LL$ LD$t$(LD$LL$ HHD$HMHct$(tLHHLcLDI$Ht$HNdLD$@I@HHLD$ Ht$ϹHt$LD$ H QH v$P ;HL$1LH虹H QHL$P =@9HHl$8H|$HH/uHGP0HD$HH|$@H/uHGP0IpHl$8HD$@HD$8HkHEII~D$H8D$HPLhLpHtH/uHGHT$P0HT$HtH*u HBHP0MtImu IELP0H+uHCHP0MtI,$u ID$LP0HtHmu HEHP0HxL[]A\A]A^A_fHHLpHL$fDHGLL$ LD$P0HD$@E1ۿLD$LL$ HHH; u$LL$~D$HJtPLD$HLL$(LD$ D$Ld$`)D$PLLD$ LL$(HHD$84MtI)uIALD$LP0LD$I($I@LP0@HPR⍁LL$~D$JtPE1LD$Ld$`D$)D$PHPLRR uLXLL$ LD$ZHLALL$ LD$HD$8H?D$odfD29E谶@$7HEH;u$H;r$H@hHH@H1HIL|$8MHD$8?fD$VoD$ I1E1H|$@Ht H/uHGP0H|$HHt H/uHGP0T$ t$H }H=Բ/CHNH+A>/D$boD$ {裹ID$~oE1H|$8Ht!H/uHGLL$(LD$ P0LL$(LD$ HD$8H|$@Ht!H/uHGLL$(LD$ P0LL$(LD$ HD$@MtI(uI@LL$ LP0LL$ MtI)u IALP0H|$HHt H/uHGP0IEXfH5K&HHD$8IE`AEXHD$@IEhIEhHD$HHCHHHHH|$8AH9Hmu HEHP0HL$8I}XIU`ImhIMXHL$@IM`HL$HIMhHtH/uHOHT$ Q0HT$ HtH*u HJHQ0HtHmu HUHR0HD$8HD$@HD$HEt$H RH=@HL$8HT$@LHt$H H-y&H5&HELMHt$dHt$H QH Rp$P ;!HL$1HAI2H QHL$P H=\9dMLI/D$p1D$ DD$ 1I~D$H8D$LxLhLpHt H/uHGP0MtI/u IGLP0MtImu IELP0H|$8HH/I*HGP0fDE1D$oA1E1ɿfDA1fHEL8IIGL1P0D$pD$  @D裱@$H1LALD$LL$ D$oD$p1D$ fL1膴IHD$8HE1D$oDH=HL$(蟰Ht$LD$ HL$(KD$oE1E1{DSHuHl$H5HH8ȭfDL}ISD$oD$ 胴HHG@HEH@HXHt;HJH~&H9j1@H;lAHH9uE1@HH9HuE1H;-m$AZ1iHHHHD$ЬHT$IH*0HBHP0!D$oD$ D$o1H色IH:D$p1D$ XH=HL$ Ht$蠮Ht$HL$ wHHuHj$H5iH8D$pD$ AKH豫A;tHA$HD$8D$oD$ IDff.HHHtHHt&HÐKHu𾲂f.H H=0S;1Df.AWAVAUATIUSHXH- &H^H|$HD$0Hl$8H HHHHhH XHOHvL rLOL@HH j$SHH5H81述XWZH !H=1~:HXH[]A\A]A^A_@Hn ID$H$H&HEH X&H9HN H?&H HL%+&M ID$H5F&LHH HI$HH I$HH<$HGH+AHB&H@H9&aH&H!HL%~&MID$H5&LHHFII$HMI$HHg$IFHT$H9H$H;gi$HT$04H;j$IVBHZE1 uM~茫L%h$H H A; $H4$LHaH H A$=) 29>HMHI,$u ID$LP0H|$H5;&HGHHIMHD$I9D$ Ml$M IL$IEHI,$uID$H $LP0H $HHLH $葿ImH $IuIEH $LP0H $MH)u HAHP0I.u IFLP0HHPHHHuAIE1E1HCHLP0MtI/u IGLP0MtImu IELP0HmHEHP0ID$LP0@HئH+D$H{&L$H@-H9Ĩ&&H&HHL%&M*ID$H5&LHHIM%I,$"H 3e$IFHL$H9 H$H;f$HT$0H;Tg$IVBHZE1 uMnèL%e$P JH A; $H4$LH蘨P JH A$=p29eHMIEHHKIEHu IELP0H&H g&H9HHN&HMHL%:&MYHCH5&HHHIMID$H;D$1AE1H;Qe$ H; f$&L$ uL$ HIuMtLxHcHEMc1HLLD$(IlLOtqLD$(HHD$ vI( I,$u ID$LP0Hmu HEHP0HCH5Ʈ&HHHIMH5&L9 ID$H;d$ I|$uA|$ L c$H-2d$E1II9MAI,$AIt$LV0EuxL谧ADžyiIEHl$ E1E1$XE1HD$D$HCHP0oID$LP0ID$LP0Im H&HREH9ʤ&H&HHL5&MIFH5&LHHIIHM}IHu IFLP0HCH5&HHH6IMIFHt$ LH@pHZH@HMIIHMIHuIFL$LP0L$L$謤L$HI$L@æIH:H&H &H9HHp&H@HL=\&MbIGH5Э&L$LHHL$IMI/uIGL$LP0L$H5&LLL$ L$I.uIFL$LP0L$LLLL$L$HII,$uID$L$LP0L$ImuIEL$LP0L$I(u I@LP0H5&H|$?IHHD$I9D$ Il$HMD$HEII,$uID$L$LP0L$LLHL$ HmL$IuHEL$HP0L$MRI(u I@LP0I.u IFLP0H5{&HIH轜IHHR&LLL$HI@H9&H9L$HII(u I@LP0I.u IFLP0Ht$ LL9LɜIHHCLH@pHH@HL$HL$IMI(u I@LP0H+Hl$ IH# HHHH5&LHHVH6HD$0HH HD$0Hl$8H$1@HLMLIf@D@DD$E1E11$WfT$4$H H=+.HH+uE11H=&H&H5&IfDID$@=L|$ ~D$ Ht0HHl$ LLt$@D$ )D$0HD$ H{MtI/u IGLP0I.IFLP0H=9&DI @ӣH%$WME1E1D$E1IEHE1IELD$LP0MLD$IHIH}IGLD$LE1E1P0LD$MtI,$uID$LD$LP0LD$MtI.uIFLD$LP0LD$M*I( I@LP0E1E1H9& H&H HL%Н&MA ID$H5&LHH II$HM I$HuID$LD$(LP0LD$(I@H;D$ AE1H;[$ H;\$LD$(LD$(HI MtLxLc|$H$Ic1HLIL$K\HIDLL$HI I,$uID$LP0L$I(u I@LP0I9L;5Z$L;5Z$L趞Ņ I.u IFLP0 H&H F&H9HPH-&HbHL&M H5y&LL$L$HIcI(u I@LP0HD$I9D$HLLIMdH)uHAL$HP0L$H+uHCL$HP0L$H|$H5&L$HGHH L$IM HD$I9D$JI\$H<IL$HHI,$uID$HL$LL$P0L$HL$LHHLD$H $߰H+H $ILD$uHCHL$HL$P0HL$L$M H)uHAL$HP0L$I.uIFL$LP0L$IHl$ LHPI#IULR0lI@LP0 MfM*MnI$IEI.u IFLP0H$LLI,$HID$LP0M~MMfII$I.u IFLP0H$LL觯I/HhIGLP0YHt$0LH9fDHt$0LHfA.D$H-W$AHEIfD @$f@$fH4$LHp@H4$LH@L|$ ~D$ Ht0HHl$ LLt$@D$ )D$01HD$ H^I$ME1M$|XMHD$H=&H&H5&IfDE1E11D$$KX4@E1E11D$$W@H=9&H*&H5+&&IfDoNH)L$0H^HT$0HLL"H5<&8W5DHFHHD$0ǒHSIME1E11$XD$|fH=i&tI@E1E11E1D$$MXfH=)&4Ii@ÛITE1E1D$$iXfDH=&H&H5&I\fDE1E1D$$XnfDCIIH5y&LHVHoHD$8HEXfDII$ME1E1$kXHD$&H=!&$I@E1E1E1ID$$*XELJ1 uI~Ht$0 AH(D`LJ1 uI~Ht$08 AHDM|$MkMl$IIEI,$u ID$LP0IEM1A@H=1mTJME1E11D$E1E1$XDI@@=[L|$H$Ht0H~D$H\$LL$HD$@D$)D$0L$HI+ MI/~IGL$LP0L$gf.H=al脔CIM$\XE1D$E1E11H;IH;IME1E1E1$\XHD$H8ME1E1E1E1D$$XE1E1Hl$ D$$XIfI%I$M$XMD$MH>LLL$LL$IDI$ME1E1$XHD$H;O$HL)II$MD$HM$XHl$ E1E1E1ILL]MIGL|$H$Ht0H~D$H\$LL$HD$@D$)D$0L$HI$XMNIHl$ E1E1D$HCIH=&H&H5&IE1E1Hl$ D$$IYE1E1Hl$ D$$XoH=&HY&H5Z&IuE1E1E1E1D$$XE1E1E1$\XD$CME1E1Hl$ $KYD$!ɕIDH= &IH=&IMD$$X}IMxMJM`II$I(u I@LP0ID$MA1ҿD$E1E1Hl$ E1D$$NYIMHl$ E1$PYD$.I/Hl$ LIMHl$ E1$XE1E1HD$E1Hl$ E1D$$SY&$XIEE1Hl$ $XYD$HIEHl$ $_YD$H{E1E1Hl$ D$$XIEE1Hl$ $ZYD$H8H=.&H&H5&L$L$IGI\$ E1E1LHl$ D$$Y芓L$I0ImHl$ $\YD$f.KL$IH=&L$L$ILE1E1Hl$ E1ID$$(YE1E1Hl$ D$$XH=%&H֍&H5׍&IH=&IIEE1Hl$ $aYD$HE1E1Hl$ D$$XPE1Hl$ D$$pYgE1E1Hl$ E1ID$$YIl$HOIL$HEHI,$uID$H $LP0H $HHHH $ڢHmH $IHEHL$HL$P0HL$L$1AE1Hl$ D$$~YaHmI$H5dMH8kIE1E1$\XHD$HD$ME1E1H%I$H5dE1H8#D$LD$$XHL$> L$IE1Hl$ D$$YE1Hl$ D$$YxI$ME1M$XMHD$Hl$ D$$Y7Hl$ D$$YpHl$ D$$YWHl$ D$$Y A$X1ɺAH1ɺAHE1~H-bI$HpI4$I9@L;-H$D L;-I$@ AH%Hl$ E1D$nAWAVAUIATUSH8HH$L=Q&L%Ž&HnH\$L|$Ld$ H\$(H H?H HcHH$Lf(L~ HFHD$H=\&L|$HGH;|H$H;7I$iHOARHiE1 uLw覊H H HG$; LLH~x WP HnG$=a29dH H&&H &H9HHf&H HL5R&MaIFH5V&LHH# IIHMIHu IFLP0LH\IH I/u IGLP0L;51G$L;5F$tI9kLÊADž I.u IVLR0EH(&H q&H9H HX&H7 HL5D&M IFH5`&LHH IIHM IHu IFLP0LH^IH I/u IGLP0L;53F$L;5E$I9LʼnADž I.u IVLR0E L=‹&M9HF$I9D$I9GA|$  A  IT$I;WvID$IOH9@H@t HPA|$ EG D8,@ { IL$0IH@IDHA U IO0IwHA@HEȃ  DA9Ht#1H=MIu L$HT$H=eC$ILؗ&HHD I.IFLP0H$H\$f.H$H$HF0H$I9WOMIu L$HT$H=\C$IL/&HHI/u IGLP0Hmu HEHP0H8H[]A\A]A^A_DD@9@$@HHIH HXHIHHnB$I?UIH5gH8L A1X:ZH d5H=1E@L=و&M9HC$I9D$I9GA|$ *A IT$I;WLID$IOH9@H@t H&A|$ EG D8@ IL$0IH@IDHA IO0IwHA@HEȃDA9H1HP}fDI9uuwI9uunLL辀IHH;A$L;%A$I9L5I,$u ID$LP0iMIu L$HT$H=A$ILx&HHAA`@DLhHHH %[Z1H=€}H6IHHcHHF0HD$(HF(HD$ HF LHD$HFHD$4}IH~eHHu*M~.H5&LHVmH HD$(IMHD$L|$Ld$ HD$HD$(H$mHufDHt$HfI9uLL~IHtIH;?$L;%?$I9L.I,$u ID$LP0AbDA\fDH wD1H=IH=Ɉ&Hy&H5y&IDfDMIu L$HT$H=?$IL&HH0IA^H#H=Y&dI@hA\IHAIFLP02f˄IH=!XD[A\kI/IGLP0@HHc<$H5WH8d}ImA\HDHFLHD$zIMH5&LHVЀHtHD$IMgH5"&LHV覀HHD$ I LzIM"H5&LHVaHuHD$I^f.LIE1 uLGHt$ LAHH=&HZw&H5[w&vIfDAahIH=9&DI@AaAaIAaHfH=&H9HEt H;<$H螂HHH=+&HcIHMHHu HCHP0LI,$uyID$L"AgP0WI,$DI,$DAeIAcHAg"HT$HLL H5 &#Lw:Llw0mIwHI|$HLDwA]x\L$wIwHVI|$H2DqDNDYD6H|HHAAhIHAgH EhH=y 1ɺLAHDAW1AVIAUI͹ATULSHHHH|$@H$0HHH$HIH&LL$H v&HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$H9H%H4v&H&HL= v&L$M%IGH5&LHH &IL$H$M%H/HDŽ$zIH$H&HHX|IH$H'H7$H5&H}H$H$H$IH$H H$H/uHGP0H$HDŽ$H/uHGP0H$HDŽ$H/uHGP0H+H$HDŽ$HD$xu HCHP0H|$xH5&HDŽ$HGHH5IL$M5H5|&L9tIGH;n7$x6I H6$HHD$(H$I/ H$H;|$(HDŽ$H;=@6$H6$H\$u H9 H/o HDŽ$sH$H$HH$HSH&H s&H9Hs9Hls&H;:HH=Xs&H$H/7H5(&CH$HR9H$H/uHGP0H5&H|$xHDŽ$H$H9H 04$HL$HH9H)%L`M%H@I$HH$H$H/1H$LH$I,$1H$H9H$H/uHGP0H$H$HDŽ$HD$HH9G1:HyH$H$Ht H/J1H$HDŽ$H/uHGP0H$H$HDŽ$6:H/0H$L$HDŽ$HDŽ$Ht H/0H$HDŽ$Ht H/0H$HDŽ$Ht H/0H5y&LHDŽ$rIH$HKH;D$(L;%D3$u L;d$j0I,$ 0HDŽ$ Hч&Hp&H9X;1Hp&H UHL%}p&L$M SH5|&LuH$IH$H3TH/uHGP0H$HD$HHDŽ$H9GVLqIH$L$HtH/uHGP0L$HDŽ$MUH$H/uHGP0H5;x&1H$HDŽ$XIH$HWH$H/uHGP0H$H;|$(HDŽ$H;=1$2H;|$v2uH$Å ZH/uHGP0HDŽ$6 H5v&H= &1IIH$HHH$H/uHGP0H$E1E11H\$xE1HDŽ$E1HD$0D$D$(jHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ qHGP0S@IGLP0H5ix&H|$xIH$H1H@Ht$LH@pH0H@H0IL$M3H$H/uHGP0L$H5u&HDŽ$HDŽ$L9tIGH;s0$}4ILd$(I$L$I,$u ID$LP0HDŽ$- H&Hl&H9X IHl&HRJHLl&L$MIHH5Bx&L貼IH$H{IH$H/,H$H-$HDŽ$H9G`ML诪IH$L$HtH/uHGP0L$HDŽ$H$MuMH/S,H5|t&1H$HDŽ$IH$H,H$H/uHGP0H$H;|$(HDŽ$H;=-$.H;|$.rH$ÅOH/uHGP0HDŽ$wH5(s&H=a&1芹IH$HSrH.H$H/uHGP0H\$xE1E11HD$0E1E1H$HDŽ$D$/D$(oHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ fHGP0HDŽ$EffA.Gz fDHi,$H,$HH\$(@pH$ÅH$E1E11H\$xE1E1E1HD$0D$D$(cHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ HD$0E1E11HD$E1E1H+$HD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(bHD$HH/ HGL$L$HL$xP0H$L$L$HL$xHt7H/u1HGL$L$HL$xP0L$L$HL$xH$Ht7H/u1HGL$L$HL$xP0L$L$HL$xH$Ht7H/u1HGL$L$HL$xP0L$L$HL$xHt*H)u$HAL$HLL$xP0L$LL$xMtI(uI@LL$xLP0LL$xMtI)u IALP0T$(t$H >vH=jHD$HD$0HL$H9tHtH@@(5(MtI/u IGLP0MtI,$u ID$LP0HT$ HtHHD$HHu HBHP0H4$HtHH$HHu HFHP0MtImu IELP0HL$8HtHH$HHu HAHP0H$Ht H/uHGP0HT$PHtHH$HHu HBHP0Ht$pHtHH$HHu HFHP0HL$hHtHH$HHu HAHP0HT$@HtHH$HHu HBHP0Ht$HHtHH$HHu HFHP0HL$`HtHH$HHu HAHP0HT$XHtHH$HHu HBHP0Ht$HtHH$HHu HFHP0H+u HCHP0MtI.u IFLP0HtHmu HEHP0HD$H[]A\A]A^A_ffA.Gz_H&$1L%&$HHD$(H9&$EH5!l&H=y&1賲H$H$HHVH$H/uHGP0D$H\$xE1HDŽ$HD$0D$(Lt$ Lt$HD$E11E1HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$fH$H&$E1E11H$E1HD$0E1D$E1D$(bHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ HD$+H;l$5HiH$H>Hy&Hyb&H9XvCH`b&H#HL%Lb&L$MBH5l&LdIH$H@DH$H/uHGP0HDy&Ha&HDŽ$H9XEHa&H(GHLa&L$M EH5|q&LܱHHGH$H/uHD$HGP0HL$Hx&HSa&HDŽ$H9XIH.a&HKHL%a&L$MIH5p&LHL$MHL$HIKH$H/uHGHL$P0HL$H^"$HDŽ$HD$HH9AMHLHL$;HL$HH$H$Ht H/uHGP0HDŽ$I,$u ID$LP0L$MLH+u HCHP0H5;p&H$nIHNH$H/uHGP0H$HD$HHDŽ$H9G7PLrH$H$Ht H/uHGP0HDŽ$I,$u ID$LP0L$MFOH$H/uHGP0H}H5Av&HDŽ$L$HDŽ$̻Hv&H_&H9X=XH^&HRcHH^&HD$ H$HYYH5m&H,HH}[H$H/uHD$HGP0HL$H5 o&HHL$HDŽ$HL$HHD$ H$cHu&HH^&H9XhH/^&HhHH^&HD$ H$HgH5m&HHL$iHL$HIBgH$H/uHD$ HGHL$P0LD$ HL$HDŽ$HAH;D$Hlh1H; $UH;!$ZNLD$8t$0HL$bHL$t$0HHD$ LD$8H$|vH$HtHT$ HDŽ$HBH$HcHt$ HHHL$HT1LDHDŽ$觫HL$HHD$ H$wH$H/u HGP0HL$HDŽ$H)u HAHP0H$H;|$(H;=7$4EH;|$)ETcH$ÅqH/uHGP0HDŽ$Hs&H\&H9XVH[&HHH [&HH5f&HHL$QHL$HHD$ H$rH)u HAHP0H/s&H[&H9X(Ho[&HHL[[&M[H5{k&LLD$֫LD$HHD$ H$I(u I@LP0H5k&H蜫HD$ HH$HD$HHDŽ$H9GXHt$ 豙HH$HtH/uHGHL$P0HL$H\$ HDŽ$HHD$HHuHCHL$HP0HL$HRH$H/uHGHL$P0HL$H5j&HHL$HDŽ$輪HL$HHD$ H$>H)u HAHP0H$HD$HH9GH$轘H$H$H/uHGP0H$HDŽ$HD$ HH$H/uHGP0HDŽ$LI$H$L$[HHH;D$(H;$dH;L$dHHL$-`HL$H)u HAHP0H$HH$H$H$H/uHGP0H$HDŽ$H/uHGP0H$H$HH$H/uHGP0HDŽ$H$I,$u ID$LP0HDŽ$IfDHo& H(hE111AHHHD$ H$HLH$HH$H/uHGP0HmH$HDŽ$u HEHP0HDŽ$H;\$MH5e&HHkHD$ H$HENH5o`&HǺ袴HD$ H$HPH$H/uHGP0H$H;|$(HDŽ$H;=$3H;|$3 ^H$…$RH/uHGT$P0T$HDŽ$UZH5a&H?HD$ H$H]HǺLXH$HD$ H$H^H/uHGP0H$H;|$(HDŽ$H;=$>;H;|$3;0]H$…aH/uHGT$P0T$HDŽ$qcH4$Hl&Hqm&H U&$H9HMeHU&H=mHHU&HD$ H$HdH5d&H H$HD$ H$H-nH/uHGP0$HDŽ$YHD$ H$HlH$HD$HH9G{Ht$ H$H$H/uHGP0H$HDŽ$HD$ H&|H$H/uHGP0H$H;|$(HDŽ$H;=d$FH;|$F[H$ŅzH/uHGP0HDŽ$MHk&H T&H9H#HS&HHHS&HD$ H$HH5nb&HvHD$ H$HH$H/uHGP0H$H5<_&HDŽ$+H$HD$ H$HqH/uHGP0H5q\&1HHDŽ$UHD$ H$HH$HD$HH9GVHoHVHGHEHH$H$H/uHGP0H$H$HmH$Hmu HEHP0H$H/uHGP0H$HDŽ$HD$ HSH$H/uHGP0H$H;|$(HDŽ$@H;=&$@KH;|$KBYŅhH$H/uHGP0HDŽ$چ$\*\fTj\eVHD$ H$HtHǺLTHD$ H$HH$H/uHGP0H$H;|$(HDŽ$H;=I$ZH;|$ZfX…/H$H/u HG$P0$HDŽ$H݅H5[Y&H=f&1HD$ H$HH|$ 腵H$H/uHGP0HH$E11H\$xE1E1HDŽ$HD$0D$5D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ fH=)]&HP&H5P&IvfDH!$E1E11HD$0E1E1H$HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(bHD$fD YIHi$E1E11HD$0E1E1HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(bHD$fDH=[&贻I@H$E1E11HD$0E1E1H$HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(bHD$fD0`$H=/1kH$E1E11H$E1E1HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(bHD$fDHhH$!E1IL;t$ Hd&HL&H9X;4HL&HN6HHL&H$H$H.4H5X&H/H$H$H6H$H/uHGP0HDŽ$QH$H$HS8ILp TH$H$H9Hc&HK&H9X;HK&H,>HH K&H=;H5[&HH $iH $HI >H)uH$HAHP0L$H5J\&H$LL$VTL$9I(u I@LP0H$H$H$=HD$H?H$H/uHGP0H$HDŽ$H/uHGP0H$HDŽ$H/uHGP0HDŽ$I.u IFLP0L;l$(L;- $rL;l$gLQ4H;l$(H5[&HH$H$H?HA $HDŽ$H\$HH9XZ2HH@H$HB2HSHHH$H$H/uHGP0H$H$H<$H1ˈH$H$Ht H/uHGP0H<$H$HDŽ$AH/uHGP0H<$1ɺHHDŽ$^IHCH<$HHD$ MLD$ HIH$EI(u I@LP0H $H$HH$HHu HAHP0H5T&H|$@HDŽ$腙IHDEHDŽ$HD$HI9E:IEH$H:MEHIImuIEL$LP0L$H$H:LLL$4cL$IH$L$HtH/Z6L$HDŽ$M:OI(u I@LP0H5:S&HL$HDŽ$莘HD$8H$HP~MIHSIELhH$OL$HHD$8H$WHR&H5xR&HL$LPL$j?H$H$LL$DH$L$HHD$8H$rH/uHGL$P0L$HDŽ$I(u I@LP0H$H/uHGP0H$H F&HDŽ$HDŽ$HD$8Hi^&H9HqH`F&HqHHLF&HD$PH$H'qH5X&HHD$PH$HpH$H/uHGP0HDŽ$KHD$PH$HqHL$8HH$HHMIHrH$H5$W&HH$NL$-rHu]&H nE&H9HHUE&HHHAE&HD$PHH$HH5T&L$H$L$HHxH/uH$HGLD$ P0LD$ H $H5U&HLHL$ L$HDŽ$ML$HL$ H)uHAL$HP0L$H$H$LL$轓L$HHD$PځH$H/uHGL$P0L$H$HDŽ$H/uHGL$P0L$HDŽ$I(u I@LP0HL$8HH$HHu HAHP0HD$PH$HD$`HD$8HD$HHD$@HD$hHD$pHD$PHD$0HD$XL;t$7H5bQ&H|$xhHHjH5L&H98H@H;a$*qHy7Hq$1Lh$HHD$(H9X$@H)uBHAL$Hωt$ P0L$t$ uLLD$ NJLD$ uI(uI@t$ LP0t$ JL;t$SH5}P&H|$8胓IHH5K&H9SH@H;|$4IxSH$HHI(Z1H;L$(H)u HAHP0{SHZ&HcA&H9X8HJA&HHH 6A&H6H5R&HHL$HL$HIH)uHD$HAHP0LD$LD$ILD$HHH5R&H|$xLD$ HD$`HL$LD$ HI/H5SR&HHLD$(HL$HD$ YJHL$LL$ LD$(„I)uIALD$ LHL$P0LD$ HL$H5wI&HLHL$(LD$ "LD$ HL$(HHD$ I(uI@HL$ LP0HL$ H)u HAHP0H|$xHt$8HGH@pHɎH@HIMsH5KX&H|$LLL$ @LL$ I)u IALP0HD$H\$xLt$ Lt$HHD$fHGP08@HGP0?@ID$LP0NID$LP0HGP0@HGP0e@HGP02@HGP0@LFL$ÅH$E1E11H\$xE1HD$0E1D$D$(iHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ HGP00@HGP0@HH&HD$HI."H=L& IH=K&HO?&H5P?&ۍIE1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(nWHL$0HHD$HHHAHP0GIH$E1E11H$E1HD$0E1HD$H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(cHD$!H;)#L?IH$HXH;$E1E11H$E1HD$0E1HD$H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(cHD$KkI3H$E1E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(muDD$1H$Ht H/uHGP0H$HDŽ$Ht H/uHGP0H$HDŽ$Ht H/uHGP0H$HDŽ$Ht H/uHGP0H{XH5P&HDŽ$蠾D$(ft$H eJfH=A>H$H$HH$蔑H5C&H=.P&1WH$H.HH$H/uHGP0HDŽ$D$D$(hHE1E1E1H$H$H$@H\$xE1E1H$1HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ [H=F&H9&H59&nHfDD$3H$E1E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(mDH=9E&败H@D$wD$(gFH$D$6H$D$DH;)#L:IH$H^H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(nZf.HGH$HHWHHH$H$H/uHGP0H$H$H$HpQnD$TsH|$L>9HH$H$Hb;HD$(H9H;#pHL$H9bH=Åg<H$H$H $HH$HHu HAHP0HDŽ$cH5L?&H|$18HH$H$HCHD$(H9H; #HD$H9H<ÅEBH$H$H $HH$HHu HAHP0HDŽ$"DH;l$3H3M&H 5&H9XPH4&HHZHH4&H$H$HYH5kF&HӅIHZH$H/uHGP0H5#>&HHDŽ$b7H$H$HYH#HD$HI9EdH4$LsMH$H$H/uHGL$P0L$H$HDŽ$H$H9cI(u I@LP0H$Ht$16H$IHdH/uH$HGP0L$L;D$(HDŽ$L;#)L;D$(LL$;L$3dI(u I@LP00jH<&HHH5D&H$@H$H$HjHDŽ$H\$HHH9XKH@H<$H$HJH$HSHHH$H$H/uHGP0H$H$HJqHD$8H$Ht H/uHGP0H|$8H$HDŽ$oH/uHGP0HDŽ$Hmu HEHP0HIJ&H2&H9XnH1&H|nHL1&MmH5;&LLD$ LD$ HH$H$rI(u I@LP07IH1sILpHD$ 9LD$ HH$H$zHI&HF1&H9XyH-1&HH|HH 1&H{H5 A&HLD$0HL$ /HL$ LD$0HH$H$zH)uHAL$HP0L$H$H5A&L$H$:L$yH$H/uHGL$P0L$H$H$LLD$ HDŽ$H$LD$ HH$H$߀H/uHGL$P0L$HDŽ$I(u I@LP0H$H/uHGP0H$H5<&HDŽ$HDŽ$HHD$PH$H$HH$H\$HH9XlHXHlH@HHH$H$H/uHGP0H$HnH$H+u HCHP0H$H$H~H$H/uHGP0HD$`11E1H$H$HDŽ$HDŽ$H$L|$ Ld$0HD$p1L$IHt$H$11HH$HH;D$(AH;#D(VH;L$VH6AąH$H/uHGP0HDŽ$EH5s:&H|$@AHH$HH$H|$8IH4HH$HSH$L`HD$HH9G}LgM}HGI$HH$H$H/uHGP0H$H$LHH$I,$u ID$LP0H$H/uHGP0H$HDŽ$HІH$H/uHGP0H$HDŽ$HD$hHtH+u HCHP0H5i6&H$HDŽ$/HH$HօH;D$(H;#iH;L$iH4ÅjH$H/uHGP0HDŽ$HhD&H,&H9XH+&HڙHH +&H$H@H5=&H }H$HH$H/uHGP0H$HD$HHDŽ$H9GHt$8 kH$H$Ht H/uHGP0H$H$HDŽ$H9H/uHGP0HDŽ$H$MtImu IELP01ɺHHHDŽ$lHH$HQHH/H$HH$H/uHGP0HDŽ$L$H+u HCHP0H5<6&LHDŽ${H$HД0HH$H4H\$hHH$HX2HH$HH5&H5}5&HU3H$H$H$PyIHEH$H/uHGP0H$HDŽ$H/uHGP0H$HDŽ$H/uHGP0HDŽ$HtHmu HEHP0HA&H)&H9XH)&HnHL%(&MאH583&L0zHH$H3I,$u ID$LP0/IHILx'1HH$HHT$(H54&H1GH$H$LwH$HH$H/uHGP0HDŽ$I,$u ID$LP0H$H/uHGP0H$HDŽ$HGH;#ՈHWHHGH_ H$H$HHH$H/uHGP0HDŽ$H$HD$XMtI.u IFLP0HL$`HDŽ$HtHH$HHu HAHP0H52&HqxHHÆHDŽ$HD$HH9AHAH$HLaHI$H)u HAHP0H$HXLOfH$H$Ht H/uHGP0H$HDŽ$HI,$u ID$LP0H$H/uHGP0H50&LHDŽ$~wHHHDŽ$HD$HH9AHAH$HkLaHI$H)u HAHP0H$H@HL9AH$H$Ht H/uHGP0H$HDŽ$HlI,$u ID$LP0H$I/u IGLP0H50&HHDŽ$vH$HAH$H$'IHH$H/uHGP0HD$pHDŽ$H@LxpMxImHT$H$L.HHHHH$H|$pAWH $AH)u HAHP0E}I,$u ID$LP0H5/&HuIHƭH$H&H$HCI,$u ID$LP0H$H$H$H/uHGP0H\$`Lt$XHDŽ$H\$h?DH$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(iлH$E11E1HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(n0H=Y0&H#&H5#&FrIH$E1E11H\$xE1HD$0E1D$ND$(sHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ sE1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$ D$(n'H=/&I饵H5 1&H|$@frHH1H$\'H $HI8H*&HL$0LD$ HI@IMx V)LD$ HL$0HH$H$K;H5(,&HLLD$0HL$ *HL$ LD$07H8&H &H9XIH &HJHH &H$H$H4JH5B0&HLD$0HL$ hqHL$ LD$0HH$H$pIH$H/uHGLD$ H $P0LD$ H $H$H5#1&LD$ H$H $HDŽ$!)H $LD$ H$sHH/uHGLD$ H $P0LD$ H $H$LHLD$0HL$ HDŽ$nHL$ LD$0HH$H$RH)uHAL$HP0L$I(u I@LP0H$H/uHGP0H$E1HDŽ$HDŽ$HD$8HD$`HD$HHD$@HD$hHD$pHD$PH$3H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$XD$(uyH=+&H&H5&mI郼HGH$HHWHHH$H$H/uHGP0H$H$HIL8IDE1E11E1HD$0H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(n`H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$ZD$(uȴE1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(i|H$E11E1HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$]D$(u阳H=(&H&H5&jI*E1E11E1HD$0H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(iH=(&IH\$xE1E11HD$0E1E1D$D$(nHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ 7xH=X'&cIϸE1E1E1E1H$H\$xHD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$_D$(u邱E1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(i6HGH$H\HWHHH$H$H/uHGP0H$H$HLk3IH=%&H]&H5^&gH$H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8Lt$ D$eD$(ԯH<$;H$H=$&H &HL$H5&fHL$IԶE1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$bD$(uH$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(i`H=#&蔃H$H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8Lt$ D$gD$(鿭Lt$ H\$xE1E1Lt$1E1H$HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$&H=O"&HL$UHL$IFE1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$dD$(urH$E1E11H\$xE1HD$0E1D$D$(iHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ ׫H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8Lt$ D$jD$(KHAH$HHYHHH)u HAHP0H$HLH-ƲHE1E1E1H$H\$xHD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$tD$(u`E1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8Lt$ D$oD$( HGL$P0L$L$Lt$ H\$xE11H$D$vE1HD$0D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$'H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$wD$(u鏨E1E1E1Lt$ H$H\$xHD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$qD$(H=&&H&H5&_HE1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(u鏧HGH$HHWHHH$H$H/uHGP0H$H$HrL)mHA@=~$HHHL$$LD$0HL$)$ZHL$LD$0HHD$ H$UH$Ht!H/uLD$ HL$HGP0LD$ HL$H$HDŽ$H/uLD$ HL$HGP0LD$ HL$HDŽ$I(I@HL$LP0HL$tMLLL$LL$IouH=&zHE1E1Lt$ H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$sD$(H$LE1HD$`HD$HHD$8HD$@HD$hHD$pHD$PH$H\$8H5#&H{iH5&HH]IH|IH}#I9@MHMIXIHI(uI@LL$ LP0LL$ Hi&LHLL$ 'LL$ HI)uIAHL$ LP0HL$ H&@H+uHCHL$ HP0HL$ H\$8HHD$ HHZHCHL$ HP0HL$ HL$8D$D$(hff.A LD$(IH$E1E11HD$0E1H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$Lt$ D$xD$(E1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$CD$(z顢Lt$ E1Lt$E1HD$01E1H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(H5 &H{t-HH$E11H\$xE1E1HD$0D$RD$(|HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ HH$E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$\D$(~Lt$ E1Lt$E1HD$01E1H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(H$E11H$Lt$ H\$xLt$D$HD$0D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8g~$HHHL$$LD$0HL$)$oHL$LD$0HHD$ H$HD$0E1E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(w˞H=&H&H5&UHD$ 鿧@RHE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$^D$(~ Lt$ E1Lt$1H$E1HD$0HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(2H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(w飜HE1H\$xE1HD$01E1D$aD$(~HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ Lt$ E1Lt$1HD$0H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(ћH$E1E11Lt$ H\$xLt$H$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(E1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(w]H|$nHD$ HNLlnHD$PHN$95H0 &HD$XHLl$PI'=H|$ H$H$HxLIHxHt$XHHD$0 LD$0HH$H$gwI(u I@LP0H$H$H$IHvH/uH$HGP0L$H$HDŽ$H/uHGL$P0L$L;D$(HDŽ$AL;#Ds@L;D$h@LL$L$A}I(u I@LP0EH&ff H*L$PHD$8H~HD$8H5&HXHD$@HYH$H$H}HD$@H5&HYHH$H/H#HDŽ$HD$HH9GJHGH$HJHWHHH$H$H/uHGP0H$H$HaJ>IH$Ht H/uHGP0HDŽ$M:{H$H/uHGP0HDŽ$Imu IELP0Ht$PH|$@HIŸHH+T$ HHN"YLH$Ht\H5>&1MH$H$H/uHGP0H<$HDŽ$/{H$HH$HHu HCHP0Ht$L H$H$H ~H$VE1E1jH|$H11dHD$H$_AXHV}H$H/uHGP0H$H5s&HDŽ$NH$H$H$H|H/uHGP0H$HD$HHDŽ$H9GO|HGH$H:|HWHHH$H$H/uHGP0H$H$H{iHL$ LL$(HH$L$GIH$H/uHGP0L$LL$(HL$ H$H$L$H5%LL$(HL$ HDŽ$HL$ H$LL$(L$HH/u+HGL$LL$(HL$ P0L$LL$(HL$ H$HLL$HL$(LD$ HDŽ$\;LD$ HL$(HH$L$GI(uI@LL$(LHL$ P0LL$(HL$ H)uHALL$ HP0LL$ H$H/uHGLL$ P0LL$ LL$ HDŽ$LL$ HIH$RwH$LL$ HDŽ$I@vLL$ HIH$vHT$H5%H9LL$ vH$H$LLL$ /:LL$ HHD$vI)u IALP0H$H/uHGP0H$HDŽ$H/uHGP0Lt$ H\$xLt$HDŽ$HD$DH$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(jH\$xE1E11HD$0E1D$D$(wHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ 酁Lt$ H\$xE1E1Lt$H$HD$0D$KD$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$Lt$ E1Lt$E1H$H\$xHD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$D$AD$(]yHD$8ZH=w%UHD$ 齒HE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(I(uI@HL$ LP0HL$ H;L$(H; u#qH;L$fHHL$ HL$ ҤLt$ H\$xE1E1Lt$H$D$#D$(HD$HE1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(~H$E1HD$0E1HD$H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(w}Lt$ E1Lt$E1H$H\$xHD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$ID$(p}E1E1E1H\$xHD$0H$HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(w/}H= %Hm%H5n%3H$}Hy%H %H9H>H%H^HL%MO^H5%LLD$0 5LD$0HH$H$]I(u I@LP0IH\HD$LD$0HI@ LD$0HH$H$3\H%H 1%H9H[H%H[HH%H$H$H$H[H5$%H<$LD$0N4LD$0HHbH\$0H{X~?HCxHHcHHcH9rcHD$0HH4cHL$0HYHHA@u H|$`uHH $HHD$`HHu H<$HGP0HD$@H5%H6H$H$H aHD$@H5%H6HH`HD$HH9B`HJH`LBHIH*uHBHL$HHL$P0L$HL$HHLHL$HL$HL$HL$H)uHQHD$HHL$R0HD$HL$H_I(uIPH$LR0H$H(u HPHR0L\$hLL$PIH$HD$@HL+L$ H LMMHD$HMH;\$P]H|$HHE111LT$pLL$`LL$`LT$pHL!H9b]HL!HtHHuHI$HILt$ E1Lt$E1H$1HD$0E1HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$( sH=3%>HH$鲥MLt$ E1Lt$H$1HD$0E1HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(drLt$ E1Lt$E1H$1H\$xHD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$D$D$(qHoHԄHGHEHH$H$H/uHGP0H$H$H_H$HmHEHP0還HH$E11H\$xE1E1HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ pELt$ E1Lt$E1H$H\$xHD$D$#D$(~pHE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(8pH=%H%LD$0H5%HL$ &LD$0HL$ H$ӶLt$ H\$xE1E1Lt$D$QHD$0D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$)oLt$ E1Lt$E1HD$0H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$ND$(nLt$ E1Lt$E1H$H\$xHD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$LD$(nH=:%LD$0HL$ ;CHL$ LD$0H$H$HH$H$}H$E1E11Lt$ H\$xLt$H$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$D$D$(GmH$E1E11Lt$ H\$xLt$H$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$D$D$(lH=%AHD$P2H=%Hk%H5l%#HD$PH$E11H\$xLt$ Lt$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$D$D$(lH$E1E11Lt$ H\$xLt$H$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$D$D$(kH$E11H$Lt$ H\$xLt$D$HD$0D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$P kH$E11H$Lt$ H\$xLt$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$D$D$(jH=%H%H5%!HvH;j#HϺHL$ HL$ HI?Lt$ H$E1Lt$H\$xHD$D$#D$(iH="%-?HD$ }H= %H%H5% HD$ }E1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(x)iLt$ E1Lt$1HD$0E1H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(hI]HכMEHIImuIEL$LP0L$H$LHL$2L$H$H+HCL$HP0L$酛Lt$ E1Lt$1HD$0E1H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$D$D$(gLt$ H\$xE11Lt$H$E1HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$gHAHLD$P0HCH1һLD$}qHE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(fHH$E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(eLt$ E1Lt$E1HD$0H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$SD$(leLt$ H\$xE11Lt$H$D$#D$(HD$dH%HD$XHH5B%H=%1HD$ H$Hz@H|$ \1H$H/uHGP0HH$E11H\$xE1E1HDŽ$HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ cHD$ILt$E1H$H\$xHD$D$#D$(HD$ cHH$E11H\$xE1E1HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ cH5n%H=%1HD$ H$HKH|$ /H$H/uHGP0HH$E11H\$xE1E1HDŽ$HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ bHE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(aH5%H=D%1mH$HSH$H.HHD$HHy Lt$ H\$xE1E1Lt$1E1H$HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$`Lt$ E1Lt$E1HD$01E1H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(a`H=>%I5HCl逅E1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(xc_HGH$HlHWHHH$H$H/uHGP0H$H$HQlHT$ HLlH$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(xk^H$E11E1HD$0H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(x]H$E11E1HD$0H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(xJ]H=s%~2IiH=_%H@%H5A%LIiE1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(x\Lt$ Hl$8E11Lt$E1H$HD$0HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PH$HD$8HD$D$D$([H=% 1I{H=%HR%H5S%IYLt$ E1Lt$E1HD$01E1H\$xHD$XHD$`HD$HHD$@HD$hHD$pHD$PH$HD$D$D$(.[H<$TH$d8HD$Lt$E11H$H\$xHD$D$#D$(HD$ ZH$E11H$Lt$ H\$xLt$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$D$%D$(TZH$E1H$Lt$ H\$xLt$D$#HD$0D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PYH$E1H\$xLt$ Lt$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$D$ D$(TYH$E11H\$xLt$ Lt$HD$0H$HD$XHD$`HD$HHD$@HD$hHD$pHD$D$D$(#YH=%L$.L$HD$PH5{H=%H`%H5a%L$L$HD$PH{E骿HE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(8XHE1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(bWLt$ Hl$8E11Lt$E1HD$0HD$XH\$xHD$`H$HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(WLt$ Hl$8E11Lt$E1H$HD$0HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PH$HD$8HD$D$!D$(4VHH$E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(ULt$ E1Lt$E1HD$01E1H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D${D$(]UH=:%H %H5 %' HzLt$ E1Lt$E1H$H\$xHD$D$#D$(TH=%)HrzLt$ H$Lt$H\$xD$#D$(HD$kTLt$ H$Lt$H\$xHD$D$#D$(6TLt$ H$E1Lt$H\$xHD$D$#D$(SLt$ H$E1Lt$H\$xHD$D$#D$(SH<$E1HGP0Lt$ H\$xE1Lt$E11H$HD$0D$D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$"SH_HKaHGHHH$H$H/uHGP0H$H$HH$H+ aHCHP0`E1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(x&RLt$ E1Lt$E1HD$01E1H\$xHD$XH$HD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$(QE1E1E1H\$xH$HD$0HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ D$D$(xPE1E11E1HD$0H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(xPHD$0E1E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$D$(w2PH=%HP%H5Q%L$L$HLt$ Hl$8E11Lt$E1H$HD$0HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$&D$()OLt$ Hl$8E11Lt$H\$xE1HD$0H$D$-D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$NLt$ Hl$8E1E1Lt$H\$xHD$0H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$*D$(MLt$ Hl$8E1E1Lt$H\$xHD$0H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PH$HD$8HD$D$(D$(dMH=%L$"L$H駃FI隵HH$E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$"D$(LHH$E11HD$0E1E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$$D$(LrHܱLt$ H\$xE1E1Lt$1E1H$HD$0D$ D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$pK˹HLt$ H\$xE1E1Lt$1E1H$HD$0D$ D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$JHE1E11HD$0E1H\$xHD$H$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$lD$(JHD$Lt$ E11Lt$H$D$#HD$H\$xD$(IH|$x"7I7qLt$ H$E1Lt$H\$xHD$D$#D$(I\H,H?NLt$ Hl$8E1E1Lt$1E1H\$xH$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$8HD$D$LD$(IH H$@HH$E11H\$xE1E1HD$0D$&D$(HD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ bHLt$ Hl$8E1E1Lt$1E1H\$xH$HD$0HD$XHD$`HD$HHD$@HD$hHD$pHD$8HD$D$>D$(GLt$ Hl$8E11Lt$E1HD$0HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$/D$(JGLt$HL|$ E1E1L$Ld$0L,$IHl$@Hl$8H\$hH\$xLt$ Lt$H$HD$0HD$XHD$8HD$D$ZD$(FLt$HL|$ E1E1L$Ld$0L,$1Hl$@E1Hl$8H\$hH\$xLt$ H$Lt$D$[HD$0D$(HD$HD$XHD$8BFLt$ E1Lt$1H$H\$xHD$D$#D$(FH\$xE1E1E1H$D$HD$0D$(xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$HD$ rEH;{o#XLǺLD$ ӯLD$ HHLt$ H$E1Lt$H\$xHD$D$#D$( EAQE1A1jH$1H$tH$HAZA[HH%H|$8H訬x&H$H/uHGP0HDŽ$Lt$HL|$ E1E1L$Ld$0L,$1Hl$@H\$xE1Lt$ Hl$8Lt$H$HD$0D$D$(HD$HD$XHD$8DLt$HL|$ E1E1L$Ld$0L,$IHl$@H\$xLt$ Hl$8Lt$H$HD$0HD$XHD$8HD$D$D$(CH=%H%H5%H~Lt$HL|$ E1E1L$Ld$0L,$1Hl$@H\$xE1Lt$ Hl$8Lt$H$HD$0D$D$(HD$HD$XHD$8BLt$HL|$ ME1L$Ld$0L,$1Hl$@IHl$8H\$hH\$xLt$ Lt$D$jH$HD$0HD$XHD$8HD$D$(BLt$HL|$ IE1L$Ld$0L,$1Hl$@E1Hl$8H\$hH\$xLt$ Lt$H$HD$0HD$XHD$8HD$D$hD$(ALt$HL|$ E1E1L$Ld$0L,$IHl$@Hl$8H\$hH\$xLt$ Lt$H$HD$0HD$XHD$8HD$D$fD$(aALt$ 1Lt$E1H$H\$xHD$0HD$`HD$HHD$@HD$hHD$pHD$PH$HD$D$#"D$(@I@@=vHcLLD$0~$~$Hk$$)$H)$LD$0HH$H$HtH/uLD$0HGP0LD$0H$HDŽ$H/uLD$0HGP0LD$0H$HDŽ$H/uLD$0HGP0LD$0H$HDŽ$H/uLD$0HGP0LD$0HDŽ$9Lt$ E1Lt$1H$E1HD$0HD$`H\$xHD$HHD$@HD$hHD$pHD$PHD$D$"D$(9?Lt$HL|$ E1E1L$Ld$0L,$IHl$@H\$xLt$ Hl$8Lt$H$HD$0HD$XHD$8HD$D$D$(>Lt$HL|$ E1E1L$Ld$0L,$IHl$@1Hl$8H\$hH\$xLt$ Lt$HD$0HD$XHD$8HD$D$|D$(>H$E1E11HD$0E1H\$xHD$HD$XHD$`HD$HHD$@HD$hHD$pHD$PHD$8H$D$1D$(=H=ز%H%H5%IHD$PL|$ L,$E1Hl$@Ld$0Lt$HHl$8L$HH\$hHD$8gaLt$ E1Lt$1H$H\$xHD$D$ $D$(=Lt$ H$Lt$H\$xHD$D$$$D$(Hl$@Hl$XME1L|$ Ld$0L,$IL$H\$`Hl$HH\$xLt$ Hl$8Lt$H$HD$0HD$XHD$8HD$D$v D$( Hl$@Hl$XIE1L|$ Ld$0L,$1L$H\$`E1Hl$HH\$xLt$ Hl$8Lt$H$HD$0HD$XHD$8HD$D$s D$(Hl$@Hl$XIE1L|$ Ld$0L,$1L$H\$`E1Hl$HH\$xLt$ Hl$8Lt$D$q HD$0H$HD$XHD$8HD$D$(EHHL,$L$HH aHl$@Hl$XHDHPL|$ E1HP:#H5{L$Ld$0H81H\$`E11Hl$HH$Hl$8Lt$ H\$xLt$HD$0D$v L$D$(HD$HD$XHD$8.H9LHl$@Hl$XIE1L|$ Ld$0L,$1L$H\$`E1Hl$HH\$xLt$ Hl$8Lt$H$HD$0HD$XHD$8HD$D$ D$(Hl$@Hl$XME1L|$ Ld$0L,$1L$H\$`E1Hl$HH\$xLt$ Hl$8Lt$H$HD$0D$v D$(HD$HD$XHD$8 Lt$ 1Lt$E1H$H\$xHD$0HD$`HD$HHD$@HD$hHD$pHD$PHD$D$1"D$( Lt$ E1Lt$E1H$1HD$0E1HD$XH\$xHD$`HD$HHD$@HD$hHD$pHD$PHD$8HD$D$D$( Lt$ E1Lt$1H$H\$xHD$D$B$D$( Lt$ H\$xE11Lt$H$D$A$D$(HD$ Lt$ 1Lt$H\$xD$?$D$(H$HD$ Lt$ H$1Lt$H\$xHD$D$:$D$( Df.AWIAVAUATUHSHHHL%B7#L c7#L5 }%LnHD$HD$8Ld$LL$ Ld$(Lt$0HIGH8JcHH^@H\$8Lu8Lt$0LE0LD$(HM(HL$ Lm Ll$HmHl$H L9H;<6#uML9tHHHL$L$WzL$HL$tWHHMLSLY^HH[]A\A]A^A_MkMLMofMLYDyL$HL$HtHrIH5V%HIHV.yL 5#HHD$\LmDMHH HOHL LOLDHH4#AUH܇H53ZH81Y{XZH H=uHH1[]A\A]A^A_fIfHsJcHfDHF@HD$8HE8HD$0HE0HD$(HE(HD$ HE HHD$HEHD$qIL 4#I~H'JcHfD'H5|%HHVwL 4#HtHD$IM H5|%HHVwL k4#HtHD$ IMH5}%HHVywL :4#HtHD$(IMH5̂%HHVHwL 4#HtHD$0IM~zH5{%HHVwL 3#HHD$8IuuHl$Ll$HHL$ LD$(Lt$0HFHHD$WpL 3#IMHl$H\$8Ll$HL$ LD$(Lt$0H~HT$LHLH5%x*Hl$H\$8Ll$HL$ LD$(Lt$0L 3#5DHHH?Pf *Y0HÐf.HGH?H~3ATL$UHSHDH}HUCI9u[]A\fDH~CATL$UHSHDH}HUf *YCI9u[]A\AUL-wATL%nUH-fSHH(H;SfHH HH*@AY HH9DH;HC@A\L$HcAD\$\$L$D$f(fW%f(`o$YT$\$L$Xf/MH(f([]A\A]@ x\f(q H([]\A\A]f(fDH~3ATL$UHSHDHHoCI9u[]A\DAVIAUL-xATL% |UH-SHI>AVfɉ *@AY H9DI>IF@A\L$HcAD\$ \D$L$(W- (Vrf\$ L$*YZYT$X/KH([]A\A]A^f& *Y \s H[]\A\A]A^(@f.H~3ATL$UHSHDHHrCI9u[]A\DH~KATL$UHSHDH}HU u\f(ofWPCI9u[]A\ÐDf.H~kATL$UHSHDH}HUf u *YZ\f(oZWCI9u[]A\Df.AWAVIAUL-ATL%UHSHH(I>IFL$Hc\$\$L$%D$Yf(Yk$YT$\$L$Xf/wAI>AVfII HLH!H*AYLtfW žI94CH(f([]A\A]A^A_I>IF5't\f(m I>Y $AV=s\f(m $fWHf(XYf/vX vAffW Y@H~3ATL$UHSHDHHqCI9u[]A\DAVIAUL-қATL%ɟUH-SHI>IFTHcDT$ \D$%fZL$Y f(YifT$ *Y ?YL$XZf/wCI>AVf H*AYDD$t WD$A98D$H[]A\A]A^DI>IFf5 *Y\(oI>YT$AVf *Y l\SoT$(W S(YX/oXGT$@W!T$.fDH~3ATL$UHSHDHHnCI9u[]A\Df.PqRff.4SHH0=,qD$f/@H;SHD$i pl$T$\f/r. pD$f(^Sj\$f/rH0[ÐD$pL$ \^D$-jt$L$ f(f(T$Y\f( gp^iT$\$\f/,H0[fDf(\=Y|$(ff.Q5p^t$fDHmff(D$YXof/sf(L$H;YYD$SL$t=of(YYY\f/wZL$ iD$D$hL$ =0Ho\T$YXYT$(YXf/L$0D$(YD$H0[fDfDgL$znL$Df..=f.fSHH 5D$t$5/fH;SfH *YT$T$ k Ƕl$T$ \/r. D$ (^j\$ /rH [D$vL$\^D$Mkt$L$((T$ Y\( 6^9jT$ \$\/ H [f.(\- fYl$.Q-=ε^|$=|$fHjf(D$YX/s(L$H;YYD$ SL$fw =T(*YYD$YY\/wYL$jD$D$ jL$5$\T$ YXYT$YX/L$%D$YD$ H [fDwhL$jhL$Df.HHH?PHHÐf.HHH?PHff.HHH?PHHÐf.HGH?f.kE„f.]D„USH(5If/?f(1f(=k--D$Y#L$^YXY\ YX Y\YXY\YXY\Yf(T$BdT$XܳL$=~f/|$^f(\XYXX\r@H~;\ PjT$Hf(L$cT$H9L$\~H(f([]f(\H,fH*Xfff(HL$$gL$YX$HHD$aYD$HfDHHH?L$$PL$YX$H@HL$`L$HYfHL$ bL$ HYfSHH 5iD$f/L$rf/sBD$Hj`H$D$W` $H [X^f(ff/wnH;SH;$S$ h^L$D$f(bT$ mh^L$$f(aX$=Lhf/r@<$H [^f(DHY_HXfDSHH $D$xe $Hf(Yf($Ze\$$H[Y^f(f.SHHeHD$eL$H[^f(ÐHD$_^D$f^\VgHfff.zu fHD$9_ g^L$H`fHD$_fWY] ff(^L$H\f(Q`SHH$L$ ff/wFH;Sf/xr v\\f(`YD$$H[\f(@X_YD$X$H[@f.SHH$L$DH;Sf f\f/f(v_fWW_YD$$$H[\f(fSHH$L$DH;Sff/v e\^#_YD$X$H[fHbHn\@f.HHH?$P >e\f(^Yyff.Qw$HYL$dL$fSHH D$bL$HY D$f(L$G\L$ff(f.Qw"f.L$YQw5f(H ^[D$f(T$df\$T$f(T$L$cT$L$Uf(HSHXf/{$stf.Ԭzu1HXH[]@$fW[1Z cD$@H $H}U $Yf/L$wHXH[]DQff.l$$]D$(D$YҬXҬf(D$YȬf(\ ̬\5,f(\%^ Ĭf(d$0Xl$HXD$8^\L$@@H}UH}f(\T$UT$ D$f(fTb\D$H^L$XD$YX$X%0aL$f/ H,\$r|$@f/?H\-f/v f/Df(L$ [D$D$8[L$ l$0D$D$Y^X{[L$HCXL$\fH*L$fH*YL$(\ $L$]L$\f/L$HXH[]f(af.`aSH\^f(#]H[Yf.AWfAVH*AUIATUHSHHD$xt H9rx `l$xLkf(k\l$Hf/|$@$ \$HT$Yf(fD$@CXL$[(f($A_L$$YL$@L,fT$Lc0f(L$`Qf. Y -`5f(L$@Y \f(f(fTf.1f-ݨI*f(Xl$ Xf(L$XK8f({@$\Xf(L$pKP ~t$0^tsHXL$HYfD(D$8CXf(\f(\f(^YXYf(Y|$@\f(L$X^K`Yf(XYfA(AXXfA(^L$hKhYT$SpD^Xf(D$PCxfA(XD$@H}UL$H}Y $U $f/L$f(nf/L$"l$|$8fI*\Yf(^f(A^XT$(XD$0\X\$ fTg^^\f/ $X\ $T$(H,IM)LH?HL1H)H~#\$`Y\$ fH*\f/IEfd$H^d$@H*YL96pf/I)l$xf/l$ IGHH[]A\A]A^A_fDfAf.DDf/L$Pwkf(L$($V^D$XXD$0[H,HPEGL$(\L$$Y\YL$Xf(L$($LVl$p^D$h\f(S[H,I9E$L$(\L$PYg\YL$hi@ID$H9fff(H*H^f(\YH9}DHCL9ff(H*H^f(\^L9~UDf(^t$`LIXf(XYHX^^XT$ YfH*^T$($f(U$T$(f(f(\f/Xl$(f/HCfEfML*ID$fM)H*fEIGL*IEH)H*fE(D$EYf(f($A^fE(D$YEYfD(,$DYD$$D$D$TD$$D$fA(^<$f(S\$@4$D$Yt$H$AYf(^S=D $l$(fA(f(5D$D^D$D$D$D$\f(f(^\f(%^f(\f($^f(fD(D\fI*XT$ Y$D^$Y$XfI*YfA(Xf(A^E^\f(f(A^DXfA(A^\f(f(A^E^\f(f(A^\f(f(A\\A^f(f(A^A^\f(f(A^\A^\f(f(A^\A^\f(A^\^$A^A^AXA^XXf/h$D$Y\f(XVH,df.H,ffUH*f(f(fT\f(fV@f(|$H\l$@f.B}wz8rj Lb0|$z@Yt$H5&$zHl$@|$0zPYt$ |$pzX|$8z`|$XzhL$`|$hzp|$zx|$P|$f(\$ $KVT$\$ $PDAUIATIUHSHHD$t H9r_T$-aVfMeI*AE\AUAm f(l$L$OL$Y MT$L$d$f(AEfYYAUXX%Uf.QY-$Xf/VH,I]0\$H}U\$1f/f(vvfDHBH9}(\$H}U\$f/vN1f(HBH9|Lf\H)H*YT$YfH*YT$^f/v H@1HH[]A\A]f.Bb ZHZ0d$1f(T$8\$0l$(L$ d$WTl$(Y- T$8L$ d$\$0Xf/f(T$ \$T\$T$ fH,fDHff.EujfH*f/rYOf/rITS@TSH\Yf( f/r"%SH)H[D1D[KSKH)H[Ðf.SHH D$f(L$CJL$ff.gSd$f/v[\HL$f(PHD$?QL$ff(f.Q|Xf(YXD$H [Y ؛Hf(4KfHHH*XD$H [UPDH [fCD$H H['PD$f(\$"R\$T$^SHH $f(D$TP$$Hf(Yf( $O\$ $H[Y^f(ff(f(SHXf(H ^L$\$,$O\$L$Y f(YYYf(YXff.QwU\Y$\$H;X$S\$$f(f(X^f/s Y^f(H f([\$T$d$P\$T$d$SHH@D$8f(L$ GQf/D$ w?t$ f/Pfl$D$^XD$00f(^f(/JXD$L$\f/D$syH;SYtJ\$0H;f(YXXL$f(L$ ^\d$(YL$SL$\Y\f/T$^H;SD$D$(WJ ǘf/L$vfW?T$8 )Xf(fTPT$X~Il$T$\f/v.fWH@[fDH;SX\fOYH@[H@[f-8Of|$Yl$YXf.QwYXL$t$f(Xf.QwN\t$ f(f(X^f(YXXL$^L$0:L$FNL$L$(T$-NL$(T$Df.SHH05hND$ \f(GD$(H;Sf/D$ D$H;SYD$(.ENT$\f(Yf/r~f(T$\$G\$D$f(tGL$^MXyLH,HaT$ff.EфEH0[fDf/Ӹr@f.H\MHH?D$\$PL$$f/vf(fYHXf/wHfHHH?$PL\f(yFL\$D$f(]FL$^f(jLHH,f/hrIBf.SHH \xLf(D$NED$H;SH;$S%?L\$$D$ AVAME9wI IHL[]A\A]A^A_@IFHEuTLbIHHI9v+H1HIIH9sfDI>AVIHI9wHL$DI>IFL!H9rL$pfDI>IFD!D9wL$PfDЉI>@I0fDAWAVAUATUSHteIẢAH?IFEucJL$ Dd$ IA9v%1AA9sI>AVIA9wH DH[]A\A]A^A_fDI>IFD!9rA\ff.AWAVAUATUSHH\$PfLAIE f AAEDrE T$ H?AUT$ E3EAfA9vY™AAf9r$FD3Ex}3AfA9v$uI}AUE3AfA9wA H[]A\A]A^A_Ð+EDHDMD!fA9s"EuI}AUED!fA9rHD[]A\A]A^A_fEuCH?AUEH[]DA\A]A^A_fD+Ex}+mDf.AWAVAUATUSHH\$PLAIE AAEDzE"T$ H?AUT$ E DA8vVE™AA@8r? ExD} A8v#uI}AUDE A8wfA4H[]A\A]A^A_f.+EDHDMD!A8s!EuI}AUED!A8rHD[]A\A]A^A_@EuCH?AUEH[]DA\A]A^A_fD+Ex}+mDf.USHHl$ tAtmA)EH[]@HLH?PEfAWAVAUATUSHHu]H~DHHH4$~$LHHflLʐHH9uHHH9tI4H[]A\A]A^A_DLIHIH9EH~IDbH$MЉD$ I?AWID9s(D$ 1AA9sfDI?AWIA9wH HHHKH9$uP@HEH0ILbIH$DI?AWIIIL9s%L1III9sI?AWIII9wLHHHCH;$uDHM$I?HAWHHCL9uf.HM$I?HAWHHCL9u`@III LHI LHI LHI LHI LH I HM4I?AWL!I9rHHHCL9u@III LHI LHI LHI LHI HM4f.I?AWD!D9wHHHCI9ux1mDAWAVAUATUSHH~uHAHHʉ4$fn $LHHfpLfHH9uHHH9t(HxHA4H9~HAtH9~AtH[]A\A]A^A_ÐLAՉIEH~IADbH$MDl$ I?AWIA9v(D$ 1AA9sfDI?AWIA9wH HKH;$uH[]A\A]A^A_@HGM$I?HAWCL9uH[]A\A]A^A_@III LHI LHI LHI LHA HM4I?AWD!A9rHCL9uH[]A\A]A^A_1uf.AWAVAUATUSHfHHAHf4$HLHHfn$LfafpHH9uHHH9tcHxHfA4AH9}QHxfAtH9~BHxfAtH9~3HxfAtH9~$HxfAtH9~HfAt H9}fAt H[]A\A]A^A_fDALIfE H~DrIIE1H$A1D$ EI?AWAfA9vAD$ Af9s2EtE1fA9vI?AWȉfA9wAHDfKH;$'EE1fHI,IHDfCH9I?AWATfHSH9u@AHHH HHH HHH HH HM$I11t@1!fA9sI?AW!fA9rDHfSL9uH[]A\A]A^A_1DuH@HLT5@AWIAVAAUATUSLHE7HI, DjE1ɉD$ 1f.EI?AWAD@A8vk+D$ EݙAA8r3QI?AWD@A8vDA@A8v EtDA@A8wfHDKH9YH[]A\A]A^A_fH~I, 11HA KH9tuI?HAWASH9uAHHH HHH HH HlM, 11$I?AW!A8sH!A8st!A8r@DHSI9uH[]A\A]A^A_fDHSI9uf.AL@ADH~StWAT1I1UI, SL˅uI<$AT$#HH9t#HH9u[]A\D@HL2AWIAVAUATM`UHSH(H|$LD$M~lIM1 A9L$A\LHI9tBADH|$LLL$^y7I)HDMH([]A\A]A^A_H~HD$L|H([]A\A]A^A_f.DAWAVAUATUSHxMAML$AuHu1Hx[]A\A]A^A_fILLD$IH<HL$@HH$4IHE11H<$LL$@MffDI4H~RHEHHT$~D$KHHflHDHH9uHHIH9tIIHH9uHH|$H?HIHHD$ H9H$HH\$(HK>> from numpy.random import Generator, PCG64 >>> rg = Generator(PCG64()) >>> rg.standard_normal() -0.203 # random See Also -------- default_rng : Recommended constructor for `Generator`. Gets the bit generator instance used by the generator Returns ------- bit_generator : BitGenerator The bit generator instance used by the generator __pyx_capi__name__loader__loader__file__origin__package__parent__path__submodule_search_locationsMissing type objectname '%U' is not definedcannot import name %Sendunparseable format string'complex double''signed char''unsigned char''short''unsigned short''int''unsigned int''long''unsigned long''long long''unsigned long long''double''complex long double''bool''char''complex float''float'a structPython objecta pointera string'long double'an integer is requirednumpy/random/_generator.c%s (%s:%d)stringsourceexactlyView.MemoryView.Enum.__init__at leastat mostlogseries_generator.pyxzipfpoissonnegative_binomialwaldrayleighlognormallogisticgumbellaplacepowerweibullparetovonmisesstandard_tstandard_cauchynoncentral_chisquarenoncentral_fbetaBitGenerator__cinit____pyx_unpickle_EnumtupleExpected %.16s, got %.200s%d.%d%sbuiltinscython_runtime__builtins__4294967296complexnumpyflatiterbroadcastndarrayufuncnumpy.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_fillfloat_fillvalidate_output_shapecontdisccont_fcont_broadcast_3discrete_broadcast_iiinumpy.core._multiarray_umath_ARRAY_API_ARRAY_API not found_ARRAY_API is NULL pointer__init__.pxdnumpy.import_arraygetbuffer(obj, view, flags)init numpy.random._generatorintegersView.MemoryView._err'NoneType' is not iterableView.MemoryView._unellipsifydefault_rngformatmultivariate_normalvhumultivariate_hypergeometricView.MemoryView._err_dimDimension %d is not directView.MemoryView._err_extentsView.MemoryView.memview_sliceIndex out of bounds (axis %d)memviewsliceobjtriangularstandard_normalrandomdirichletshufflemultinomialnumpy.PyArray_MultiIterNew1numpy.PyArray_MultiIterNew2numpy.PyArray_MultiIterNew3standard_gammauniformpermutationstandard_exponentialassignmentdeletionbuffer dtypeBuffer not C contiguous.choicebase__reduce_cython____setstate_cython__Tstridessuboffsetsndimitemsizenbytesis_c_contigis_f_contigcopycopy_fortrannumpy.random._generator.Enumnumpy.random._generator.arraymemview__getattr___bit_generator__getstate____setstate____reduce__IJJIJJJJJIIJJIJJIIIJJJJJJJJJJJJJJJJIIIJIIIIJJIJJJIIJI [l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\l\Zl\l\,[l\l\l\l\l\<[L[l\l\\[l\l\l[|[[l\l\[l\l\l\l\l\l\l\l\l\l\l\l\l\[[[l\[ \,\<\l\l\L\l\l\l\Z[l\ZXZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ|WZZWZZZZZDX4XZZ$XZZXXXZZXZZZZZZZZZZZZZXXXZtXTXWWZZWZZZWWZW]]]]]]]]]]]]]]]] [ []]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]][\\[\\\\\\\\\\\\ZZZ\\\\\\\\\\\\\\\\[[\\x\H\\\\\\\\\[Z\[Z[[Z[[[[[\\[[\[[\\\[[[[[[[[[[[[[[[[ZZ\[]\\\[[\[[[Z\[Z`]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]`]]]]]]x_]]]]]]]]]]]]]]]]]P_]]@_`8^@_]8^]]]]]8^8^]]8^]]8^]8^]]^]]]]]x^]]]@_]]]8^8^8^]8^8^8^8^]]8^]]]8^8^]]]]]]`]]]]8`pa0aa`aHa 1. The Zipf distribution (also known as the zeta distribution) is a continuous probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. Parameters ---------- a : float or array_like of floats Distribution parameter. 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. Notes ----- The probability density for the Zipf distribution is .. math:: p(x) = \frac{x^{-a}}{\zeta(a)}, where :math:`\zeta` is the Riemann Zeta function. It is named for the American linguist George Kingsley Zipf, who noted that the frequency of any word in a sample of a language is inversely proportional to its rank in the frequency table. References ---------- .. [1] Zipf, G. K., "Selected Studies of the Principle of Relative Frequency in Language," Cambridge, MA: Harvard Univ. Press, 1932. Examples -------- Draw samples from the distribution: >>> a = 2. # parameter >>> s = np.random.default_rng().zipf(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy import special # doctest: +SKIP Truncate s values at 50 so plot is interesting: >>> count, bins, ignored = plt.hist(s[s<50], ... 50, density=True) >>> x = np.arange(1., 50.) >>> y = x**(-a) / special.zetac(a) # doctest: +SKIP >>> plt.plot(x, y/max(y), linewidth=2, color='r') # doctest: +SKIP >>> plt.show() weibull(a, size=None) Draw samples from a Weibull distribution. Draw samples from a 1-parameter Weibull distribution with the given shape parameter `a`. .. math:: X = (-ln(U))^{1/a} Here, U is drawn from the uniform distribution over (0,1]. The more common 2-parameter Weibull, including a scale parameter :math:`\lambda` is just :math:`X = \lambda(-ln(U))^{1/a}`. Parameters ---------- a : float or array_like of floats Shape 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 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: >>> rng = np.random.default_rng() >>> a = 5. # shape >>> s = rng.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(rng.weibull(5.,1000)) >>> x = np.arange(1,100.)/50. >>> scale = count.max()/weib(x, 1., 5.).max() >>> plt.plot(x, weib(x, 1., 5.)*scale) >>> plt.show() vonmises(mu, kappa, size=None) Draw samples from a von Mises distribution. Samples are drawn from a von Mises distribution with specified mode (mu) and dispersion (kappa), on the interval [-pi, pi]. The von Mises distribution (also known as the circular normal distribution) is a continuous probability distribution on the unit circle. It may be thought of as the circular analogue of the normal distribution. Parameters ---------- mu : float or array_like of floats Mode ("center") of the distribution. kappa : float or array_like of floats Dispersion of the distribution, has to be >=0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mu`` and ``kappa`` are both scalars. Otherwise, ``np.broadcast(mu, kappa).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized von Mises distribution. See Also -------- scipy.stats.vonmises : probability density function, distribution, or cumulative density function, etc. Notes ----- The probability density for the von Mises distribution is .. math:: p(x) = \frac{e^{\kappa cos(x-\mu)}}{2\pi I_0(\kappa)}, where :math:`\mu` is the mode and :math:`\kappa` the dispersion, and :math:`I_0(\kappa)` is the modified Bessel function of order 0. The von Mises is named for Richard Edler von Mises, who was born in Austria-Hungary, in what is now the Ukraine. He fled to the United States in 1939 and became a professor at Harvard. He worked in probability theory, aerodynamics, fluid mechanics, and philosophy of science. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] von Mises, R., "Mathematical Theory of Probability and Statistics", New York: Academic Press, 1964. Examples -------- Draw samples from the distribution: >>> mu, kappa = 0.0, 4.0 # mean and dispersion >>> s = np.random.default_rng().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`. Parameters ---------- low : float or array_like of floats, optional Lower boundary of the output interval. All values generated will be greater than or equal to low. The default value is 0. high : float or array_like of floats Upper boundary of the output interval. All values generated will be less than high. The default value is 1.0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``low`` and ``high`` are both scalars. Otherwise, ``np.broadcast(low, high).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized uniform distribution. See Also -------- integers : Discrete uniform distribution, yielding integers. random : Floats uniformly distributed over ``[0, 1)``. Notes ----- The probability density function of the uniform distribution is .. math:: p(x) = \frac{1}{b - a} anywhere within the interval ``[a, b)``, and zero elsewhere. When ``high`` == ``low``, values of ``low`` will be returned. If ``high`` < ``low``, the results are officially undefined and may eventually raise an error, i.e. do not rely on this function to behave when passed arguments satisfying that inequality condition. Examples -------- Draw samples from the distribution: >>> s = np.random.default_rng().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. 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. 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.default_rng().triangular(-3, 0, 8, 100000), bins=200, ... density=True) >>> plt.show() standard_t(df, size=None) Draw samples from a standard Student's t distribution with `df` degrees of freedom. A special case of the hyperbolic distribution. As `df` gets large, the result resembles that of the standard normal distribution (`standard_normal`). Parameters ---------- df : 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. 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? We have 10 degrees of freedom, so is the sample mean within 95% of the recommended value? >>> s = np.random.default_rng().standard_t(10, size=100000) >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 Calculate the t statistic, setting the ddof parameter to the unbiased value so the divisor in the standard deviation will be degrees of freedom, N-1. >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> import matplotlib.pyplot as plt >>> h = plt.hist(s, bins=100, density=True) For a one-sided t-test, how far out in the distribution does the t statistic appear? >>> np.sum(s>> rng = np.random.default_rng() >>> rng.standard_normal() 2.1923875335537315 #random >>> s = rng.standard_normal(8000) >>> s array([ 0.6888893 , 0.78096262, -0.89086505, ..., 0.49876311, # random -0.38672696, -0.4685006 ]) # random >>> s.shape (8000,) >>> s = rng.standard_normal(size=(3, 4, 2)) >>> s.shape (3, 4, 2) Two-by-four array of samples from :math:`N(3, 6.25)`: >>> 3 + 2.5 * rng.standard_normal(size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random shuffle(x, axis=0) Modify a sequence in-place by shuffling its contents. The order of sub-arrays is changed but their contents remains the same. Parameters ---------- x : array_like The array or list to be shuffled. axis : int, optional The axis which `x` is shuffled along. Default is 0. It is only supported on `ndarray` objects. Returns ------- None Examples -------- >>> rng = np.random.default_rng() >>> arr = np.arange(10) >>> rng.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] # random >>> arr = np.arange(9).reshape((3, 3)) >>> rng.shuffle(arr) >>> arr array([[3, 4, 5], # random [6, 7, 8], [0, 1, 2]]) >>> arr = np.arange(9).reshape((3, 3)) >>> rng.shuffle(arr, axis=1) >>> arr array([[2, 0, 1], # random [5, 3, 4], [8, 6, 7]]) rayleigh(scale=1.0, size=None) Draw samples from a Rayleigh distribution. The :math:`\chi` and Weibull distributions are generalizations of the Rayleigh. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. 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. 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 >>> rng = np.random.default_rng() >>> values = hist(rng.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 = rng.rayleigh(modevalue, 1000000) The percentage of waves larger than 3 meters is: >>> 100.*sum(s>3)/1000000. 0.087300000000000003 # random random(size=None, dtype=np.float64, out=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` by `(b-a)` and add `a`:: (b - a) * random() + a Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. Returns ------- out : float or ndarray of floats Array of random floats of shape `size` (unless ``size=None``, in which case a single float is returned). Examples -------- >>> rng = np.random.default_rng() >>> rng.random() 0.47108547995356098 # random >>> type(rng.random()) >>> rng.random((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) # random Three-by-two array of random numbers from [-5, 0): >>> 5 * rng.random((3, 2)) - 5 array([[-3.99149989, -0.52338984], # random [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) power(a, size=None) Draws samples in [0, 1] from a power distribution with positive exponent a - 1. Also known as the power function distribution. Parameters ---------- a : float or array_like of floats Parameter of the distribution. 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 < 1. Notes ----- The probability density function is .. math:: P(x; a) = ax^{a-1}, 0 \le x \le 1, a>0. The power function distribution is just the inverse of the Pareto distribution. It may also be seen as a special case of the Beta distribution. It is used, for example, in modeling the over-reporting of insurance claims. References ---------- .. [1] Christian Kleiber, Samuel Kotz, "Statistical size distributions in economics and actuarial sciences", Wiley, 2003. .. [2] Heckert, N. A. and Filliben, James J. "NIST Handbook 148: Dataplot Reference Manual, Volume 2: Let Subcommands and Library Functions", National Institute of Standards and Technology Handbook Series, June 2003. https://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf Examples -------- Draw samples from the distribution: >>> rng = np.random.default_rng() >>> a = 5. # shape >>> samples = 1000 >>> s = rng.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 = rng.power(5, 1000000) >>> rvsp = rng.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('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 + Generator.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)') permutation(x, axis=0) Randomly permute a sequence, or return a permuted range. 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. axis : int, optional The axis which `x` is shuffled along. Default is 0. Returns ------- out : ndarray Permuted sequence or array range. Examples -------- >>> rng = np.random.default_rng() >>> rng.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) # random >>> rng.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) # random >>> arr = np.arange(9).reshape((3, 3)) >>> rng.permutation(arr) array([[6, 7, 8], # random [0, 1, 2], [3, 4, 5]]) >>> rng.permutation("abc") Traceback (most recent call last): ... numpy.AxisError: axis 0 is out of bounds for array of dimension 0 >>> arr = np.arange(9).reshape((3, 3)) >>> rng.permutation(arr, axis=1) array([[0, 2, 1], # random [3, 5, 4], [6, 8, 7]]) pareto(a, size=None) Draw samples from a Pareto II or Lomax distribution with specified shape. The Lomax or Pareto II distribution is a shifted Pareto distribution. The classical Pareto distribution can be obtained from the Lomax distribution by adding 1 and multiplying by the scale parameter ``m`` (see Notes). The smallest value of the Lomax distribution is zero while for the classical Pareto distribution it is ``mu``, where the standard Pareto distribution has location ``mu = 1``. Lomax can also be considered as a simplified version of the Generalized Pareto distribution (available in SciPy), with the scale set to one and the location set to zero. The Pareto distribution must be greater than zero, and is unbounded above. It is also known as the "80-20 rule". In this distribution, 80 percent of the weights are in the lowest 20 percent of the range, while the other 20 percent fill the remaining 80 percent of the range. Parameters ---------- a : float or array_like of floats Shape of the distribution. 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. 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.default_rng().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() numpy.core.multiarray failed to import normal(loc=0.0, scale=1.0, size=None) Draw random samples from a normal (Gaussian) distribution. The probability density function of the normal distribution, first derived by De Moivre and 200 years later by both Gauss and Laplace independently [2]_, is often called the bell curve because of its characteristic shape (see the example below). The normal distributions occurs often in nature. For example, it describes the commonly occurring distribution of samples influenced by a large number of tiny, random disturbances, each with its own unique distribution [2]_. Parameters ---------- loc : float or array_like of floats Mean ("centre") of the distribution. scale : float or array_like of floats Standard deviation (spread or "width") of the distribution. 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. 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 :meth:`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.default_rng().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 N(3, 6.25): >>> np.random.default_rng().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. 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. Notes ----- When calculating the power of an experiment (power = probability of rejecting the null hypothesis when a specific alternative is true) the non-central F statistic becomes important. When the null hypothesis is true, the F statistic follows a central F distribution. When the null hypothesis is not true, then it follows a non-central F statistic. References ---------- .. [1] Weisstein, Eric W. "Noncentral F-Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/NoncentralF-Distribution.html .. [2] Wikipedia, "Noncentral F-distribution", 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. >>> rng = np.random.default_rng() >>> dfnum = 3 # between group deg of freedom >>> dfden = 20 # within groups degrees of freedom >>> nonc = 3.0 >>> nc_vals = rng.noncentral_f(dfnum, dfden, nonc, 1000000) >>> NF = np.histogram(nc_vals, bins=50, density=True) >>> c_vals = rng.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. 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. 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 >>> rng = np.random.default_rng() >>> import matplotlib.pyplot as plt >>> values = plt.hist(rng.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(rng.noncentral_chisquare(3, .0000001, 100000), ... bins=np.arange(0., 25, .1), density=True) >>> values2 = plt.hist(rng.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(rng.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]. Parameters ---------- n : float or array_like of floats Parameter of the distribution, > 0. p : float or array_like of floats Parameter of the distribution. Must satisfy 0 < p <= 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. 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 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. http://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.default_rng().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.default_rng().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]] >>> rng = np.random.default_rng() >>> x = rng.multivariate_normal(mean, cov, (3, 3)) >>> x.shape (3, 3, 2) We can use a different method other than the default to factorize cov: >>> y = rng.multivariate_normal(mean, cov, (3, 3), method='cholesky') >>> y.shape (3, 3, 2) The following is probably true, given that 0.6 is roughly twice the standard deviation: >>> list((x[0,0,:] - mean) < 0.6) [True, True] # random multivariate_hypergeometric(colors, nsample, size=None, method='marginals') Generate variates from a multivariate hypergeometric distribution. The multivariate hypergeometric distribution is a generalization of the hypergeometric distribution. Choose ``nsample`` items at random without replacement from a collection with ``N`` distinct types. ``N`` is the length of ``colors``, and the values in ``colors`` are the number of occurrences of that type in the collection. The total number of items in the collection is ``sum(colors)``. Each random variate generated by this function is a vector of length ``N`` holding the counts of the different types that occurred in the ``nsample`` items. The name ``colors`` comes from a common description of the distribution: it is the probability distribution of the number of marbles of each color selected without replacement from an urn containing marbles of different colors; ``colors[i]`` is the number of marbles in the urn with color ``i``. Parameters ---------- colors : sequence of integers The number of each type of item in the collection from which a sample is drawn. The values in ``colors`` must be nonnegative. To avoid loss of precision in the algorithm, ``sum(colors)`` must be less than ``10**9`` when `method` is "marginals". nsample : int The number of items selected. ``nsample`` must not be greater than ``sum(colors)``. size : int or tuple of ints, optional The number of variates to generate, either an integer or a tuple holding the shape of the array of variates. If the given size is, e.g., ``(k, m)``, then ``k * m`` variates are drawn, where one variate is a vector of length ``len(colors)``, and the return value has shape ``(k, m, len(colors))``. If `size` is an integer, the output has shape ``(size, len(colors))``. Default is None, in which case a single variate is returned as an array with shape ``(len(colors),)``. method : string, optional Specify the algorithm that is used to generate the variates. Must be 'count' or 'marginals' (the default). See the Notes for a description of the methods. Returns ------- variates : ndarray Array of variates drawn from the multivariate hypergeometric distribution. See Also -------- hypergeometric : Draw samples from the (univariate) hypergeometric distribution. Notes ----- The two methods do not return the same sequence of variates. The "count" algorithm is roughly equivalent to the following numpy code:: choices = np.repeat(np.arange(len(colors)), colors) selection = np.random.choice(choices, nsample, replace=False) variate = np.bincount(selection, minlength=len(colors)) The "count" algorithm uses a temporary array of integers with length ``sum(colors)``. The "marginals" algorithm generates a variate by using repeated calls to the univariate hypergeometric sampler. It is roughly equivalent to:: variate = np.zeros(len(colors), dtype=np.int64) # `remaining` is the cumulative sum of `colors` from the last # element to the first; e.g. if `colors` is [3, 1, 5], then # `remaining` is [9, 6, 5]. remaining = np.cumsum(colors[::-1])[::-1] for i in range(len(colors)-1): if nsample < 1: break variate[i] = hypergeometric(colors[i], remaining[i+1], nsample) nsample -= variate[i] variate[-1] = nsample The default method is "marginals". For some cases (e.g. when `colors` contains relatively small integers), the "count" method can be significantly faster than the "marginals" method. If performance of the algorithm is important, test the two methods with typical inputs to decide which works best. .. versionadded:: 1.18.0 Examples -------- >>> colors = [16, 8, 4] >>> seed = 4861946401452 >>> gen = np.random.Generator(np.random.PCG64(seed)) >>> gen.multivariate_hypergeometric(colors, 6) array([5, 0, 1]) >>> gen.multivariate_hypergeometric(colors, 6, size=3) array([[5, 0, 1], [2, 2, 2], [3, 3, 0]]) >>> gen.multivariate_hypergeometric(colors, 6, size=(2, 2)) array([[[3, 2, 1], [3, 2, 1]], [[4, 1, 1], [3, 2, 1]]]) 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``. Parameters ---------- n : int or array-like of ints 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. Examples -------- Throw a dice 20 times: >>> rng = np.random.default_rng() >>> rng.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: >>> rng.multinomial(20, [1/6.]*6, size=2) array([[3, 4, 3, 3, 4, 3], [2, 4, 3, 4, 0, 7]]) # random 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. Now, do one experiment throwing the dice 10 time, and 10 times again, and another throwing the dice 20 times, and 20 times again: >>> rng.multinomial([[10], [20]], [1/6.]*6, size=(2, 2)) array([[[2, 4, 0, 1, 2, 1], [1, 3, 0, 3, 1, 2]], [[1, 4, 4, 4, 4, 3], [3, 3, 2, 5, 5, 2]]]) # random The first array shows the outcomes of throwing the dice 10 times, and the second shows the outcomes from throwing the dice 20 times. A loaded die is more likely to land on number 6: >>> rng.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: >>> rng.multinomial(100, [1.0 / 3, 2.0 / 3]) # RIGHT array([38, 62]) # random not like: >>> rng.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. Parameters ---------- p : float or array_like of floats Shape parameter for the distribution. Must be in the range (0, 1). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logarithmic series distribution. See Also -------- scipy.stats.logser : probability density function, distribution or cumulative density function, etc. Notes ----- The probability mass function 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.default_rng().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. 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. 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: >>> rng = np.random.default_rng() >>> mu, sigma = 3., 1. # mean and standard deviation >>> s = rng.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. >>> rng = rng >>> b = [] >>> for i in range(1000): ... a = 10. + rng.standard_normal(100) ... b.append(np.product(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, 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). 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. Notes ----- The probability density for the Logistic distribution is .. math:: P(x) = P(x) = \frac{e^{-(x-\mu)/s}}{s(1+e^{-(x-\mu)/s})^2}, where :math:`\mu` = location and :math:`s` = scale. The Logistic distribution is used in Extreme Value problems where it can act as a mixture of Gumbel distributions, in Epidemiology, and by the World Chess Federation (FIDE) where it is used in the Elo ranking system, assuming the performance of each player is a logistically distributed random variable. References ---------- .. [1] Reiss, R.-D. and Thomas M. (2001), "Statistical Analysis of Extreme Values, from Insurance, Finance, Hydrology and Other Fields," Birkhauser Verlag, Basel, pp 132-133. .. [2] Weisstein, Eric W. "Logistic Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LogisticDistribution.html .. [3] Wikipedia, "Logistic-distribution", https://en.wikipedia.org/wiki/Logistic_distribution Examples -------- Draw samples from the distribution: >>> loc, scale = 10, 1 >>> s = np.random.default_rng().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() integers(low, high=None, size=None, dtype=np.int64, endpoint=False) Return random integers from `low` (inclusive) to `high` (exclusive), or if endpoint=True, `low` (inclusive) to `high` (inclusive). Replaces `RandomState.randint` (with endpoint=False) and `RandomState.random_integers` (with endpoint=True) Return random integers from the "discrete uniform" distribution of the specified dtype. If `high` is None (the default), then results are from 0 to `low`. 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 0 and this value is used for `high`). 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 np.int64. endpoint : bool, optional If true, sample from the interval [low, high] instead of the default [low, high) Defaults to False 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. Notes ----- When using broadcasting with uint64 dtypes, the maximum value (2**64) cannot be represented as a standard integer type. The high array (or low if high is None) must have object dtype, e.g., array([2**64]). Examples -------- >>> rng = np.random.default_rng() >>> rng.integers(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) # random >>> rng.integers(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: >>> rng.integers(5, size=(2, 4)) array([[4, 0, 2, 1], [3, 2, 2, 0]]) # random Generate a 1 x 3 array with 3 different upper bounds >>> rng.integers(1, [3, 5, 10]) array([2, 2, 9]) # random Generate a 1 by 3 array with 3 different lower bounds >>> rng.integers([1, 5, 7], 10) array([9, 8, 7]) # random Generate a 2 by 4 array using broadcasting with dtype of uint8 >>> rng.integers([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7], [ 1, 16, 9, 12]], dtype=uint8) # random References ---------- .. [1] Daniel Lemire., "Fast Random Integer Generation in an Interval", ACM Transactions on Modeling and Computer Simulation 29 (1), 2019, http://arxiv.org/abs/1805.10941. hypergeometric(ngood, nbad, nsample, size=None) Draw samples from a Hypergeometric distribution. Samples are drawn from a hypergeometric distribution with specified parameters, `ngood` (ways to make a good selection), `nbad` (ways to make a bad selection), and `nsample` (number of items sampled, which is less than or equal to the sum ``ngood + nbad``). Parameters ---------- ngood : int or array_like of ints Number of ways to make a good selection. Must be nonnegative and less than 10**9. nbad : int or array_like of ints Number of ways to make a bad selection. Must be nonnegative and less than 10**9. nsample : int or array_like of ints Number of items sampled. Must be nonnegative and less than ``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 -------- multivariate_hypergeometric : Draw samples from the multivariate hypergeometric distribution. scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{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. The arguments `ngood` and `nbad` each must be less than `10**9`. For extremely large arguments, the algorithm that is used to compute the samples [4]_ breaks down because of loss of precision in floating point calculations. For such large values, if `nsample` is not also large, the distribution can be approximated with the binomial distribution, `binomial(n=nsample, p=ngood/(ngood + nbad))`. References ---------- .. [1] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [2] Weisstein, Eric W. "Hypergeometric Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", https://en.wikipedia.org/wiki/Hypergeometric_distribution .. [4] Stadlober, Ernst, "The ratio of uniforms approach for generating discrete random variates", Journal of Computational and Applied Mathematics, 31, pp. 181-189 (1990). Examples -------- Draw samples from the distribution: >>> rng = np.random.default_rng() >>> ngood, nbad, nsamp = 100, 2, 10 # number of good, number of bad, and number of samples >>> s = rng.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 = rng.hypergeometric(15, 15, 15, 100000) >>> sum(s>=12)/100000. + sum(s<=3)/100000. # answer = 0.003 ... pretty unlikely! gumbel(loc=0.0, scale=1.0, size=None) Draw samples from a Gumbel distribution. Draw samples from a Gumbel distribution with specified location and scale. For more information on the Gumbel distribution, see Notes and References below. Parameters ---------- loc : float or array_like of floats, optional The location of the mode of the distribution. Default is 0. scale : float or array_like of floats, optional The scale parameter of the distribution. Default is 1. 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 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: >>> rng = np.random.default_rng() >>> mu, beta = 0, 0.1 # location and scale >>> s = rng.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 = rng.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. Parameters ---------- p : float or array_like of floats The probability of success of an individual trial. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized geometric distribution. Examples -------- Draw ten thousand values from the geometric distribution, with the probability of an individual success equal to 0.35: >>> z = np.random.default_rng().geometric(p=0.35, size=10000) How many trials succeeded after a single run? >>> (z == 1).sum() / 10000. 0.34889999999999999 #random gamma(shape, scale=1.0, size=None) Draw samples from a Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, `shape` (sometimes designated "k") and `scale` (sometimes designated "theta"), where both parameters are > 0. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. 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. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", 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.default_rng().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. 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. 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.default_rng().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. choice(a, size=None, replace=True, p=None, axis=0, shuffle=True) Generates a random sample from a given 1-D array Parameters ---------- a : {array_like, int} If an ndarray, a random sample is generated from its elements. If an int, the random sample is generated from np.arange(a). size : {int, tuple[int]}, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn from the 1-d `a`. If `a` has more than one dimension, the `size` shape will be inserted into the `axis` dimension, so the output ``ndim`` will be ``a.ndim - 1 + len(size)``. Default is None, in which case a single value is returned. replace : bool, optional Whether the sample is with or without replacement p : 1-D array_like, optional The probabilities associated with each entry in a. If not given the sample assumes a uniform distribution over all entries in a. axis : int, optional The axis along which the selection is performed. The default, 0, selects by row. shuffle : bool, optional Whether the sample is shuffled when sampling without replacement. Default is True, False provides a speedup. Returns ------- samples : single item or ndarray The generated random samples Raises ------ ValueError If a is an int and less than zero, if p is not 1-dimensional, if a is array-like with a 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 -------- integers, shuffle, permutation Examples -------- Generate a uniform random sample from np.arange(5) of size 3: >>> rng = np.random.default_rng() >>> rng.choice(5, 3) array([0, 3, 4]) # random >>> #This is equivalent to rng.integers(0,5,3) Generate a non-uniform random sample from np.arange(5) of size 3: >>> rng.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: >>> rng.choice(5, 3, replace=False) array([3,1,0]) # random >>> #This is equivalent to rng.permutation(np.arange(5))[:3] Generate a non-uniform random sample from np.arange(5) of size 3 without replacement: >>> rng.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'] >>> rng.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. 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.default_rng().chisquare(2,4) array([ 1.89920014, 9.00867716, 3.13710533, 5.62318272]) # random bytes(length) Return random bytes. Parameters ---------- length : int Number of random bytes. Returns ------- out : str String of length `length`. Examples -------- >>> np.random.default_rng().bytes(10) ' eh\x85\x022SZ\xbf\xa4' #random binomial(n, p, size=None) Draw samples from a binomial distribution. Samples are drawn from a binomial distribution with specified parameters, n trials and p probability of success where n an integer >= 0 and p is in the interval [0,1]. (n may be input as a float, but it is truncated to an integer in use) Parameters ---------- n : int or array_like of ints Parameter of the distribution, >= 0. Floats are also accepted, but they will be truncated to integers. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized binomial distribution, where each sample is equal to the number of successes over the n trials. See Also -------- scipy.stats.binom : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the binomial distribution is .. math:: P(N) = \binom{n}{N}p^N(1-p)^{n-N}, where :math:`n` is the number of trials, :math:`p` is the probability of success, and :math:`N` is the number of successes. When estimating the standard error of a proportion in a population by using a random sample, the normal distribution works well unless the product p*n <=5, where p = population proportion estimate, and n = number of samples, in which case the binomial distribution is used instead. For example, a sample of 15 people shows 4 who are left handed, and 11 who are right handed. Then p = 4/15 = 27%. 0.27*15 = 4, so the binomial distribution should be used in this case. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics with R", Springer-Verlag, 2002. .. [2] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [3] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [4] Weisstein, Eric W. "Binomial Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/BinomialDistribution.html .. [5] Wikipedia, "Binomial distribution", https://en.wikipedia.org/wiki/Binomial_distribution Examples -------- Draw samples from the distribution: >>> rng = np.random.default_rng() >>> n, p = 10, .5 # number of trials, probability of each trial >>> s = rng.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(rng.binomial(9, 0.1, 20000) == 0)/20000. # answer = 0.38885, or 38%. When method is 'count', sum(colors) must not exceed %dGenerator.permutation (line 4256)Generator.multinomial (line 3548)Axis argument is only supported on ndarray objects wald(mean, scale, size=None) Draw samples from a Wald, or inverse Gaussian, distribution. As the scale approaches infinity, the distribution becomes more like a Gaussian. Some references claim that the Wald is an inverse Gaussian with mean equal to 1, but this is by no means universal. The inverse Gaussian distribution was first studied in relationship to Brownian motion. In 1956 M.C.K. Tweedie used the name inverse Gaussian because there is an inverse relationship between the time to cover a unit distance and distance covered in unit time. Parameters ---------- mean : float or array_like of floats Distribution mean, 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. 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.default_rng().wald(3, 2, 100000), bins=200, density=True) >>> plt.show() standard_gamma(shape, size=None, dtype=np.float64, out=None) Draw samples from a standard Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, shape (sometimes designated "k") and scale=1. Parameters ---------- shape : float or array_like of floats Parameter, 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. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", https://en.wikipedia.org/wiki/Gamma_distribution Examples -------- Draw samples from the distribution: >>> shape, scale = 2., 1. # mean and width >>> s = np.random.default_rng().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, dtype=np.float64, method='zig', out=None) Draw samples from the standard exponential distribution. `standard_exponential` is identical to the exponential distribution with a scale parameter of 1. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. method : str, optional Either 'inv' or 'zig'. 'inv' uses the default inverse CDF method. 'zig' uses the much faster Ziggurat method of Marsaglia and Tsang. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. Returns ------- out : float or ndarray Drawn samples. Examples -------- Output a 3x8000 array: >>> n = np.random.default_rng().standard_exponential((3, 8000)) standard_cauchy(size=None) Draw samples from a standard Cauchy distribution with mode = 0. Also known as the Lorentz distribution. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- samples : ndarray or scalar The drawn samples. Notes ----- The probability density function for the full Cauchy distribution is .. math:: P(x; x_0, \gamma) = \frac{1}{\pi \gamma \bigl[ 1+ (\frac{x-x_0}{\gamma})^2 \bigr] } and the Standard Cauchy distribution just sets :math:`x_0=0` and :math:`\gamma=1` The Cauchy distribution arises in the solution to the driven harmonic oscillator problem, and also describes spectral line broadening. It also describes the distribution of values at which a line tilted at a random angle will cut the x axis. When studying hypothesis tests that assume normality, seeing how the tests perform on data from a Cauchy distribution is a good indicator of their sensitivity to a heavy-tailed distribution, since the Cauchy looks very much like a Gaussian distribution, but with heavier tails. References ---------- .. [1] NIST/SEMATECH e-Handbook of Statistical Methods, "Cauchy Distribution", https://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm .. [2] Weisstein, Eric W. "Cauchy Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/CauchyDistribution.html .. [3] Wikipedia, "Cauchy distribution" https://en.wikipedia.org/wiki/Cauchy_distribution Examples -------- Draw samples and plot the distribution: >>> import matplotlib.pyplot as plt >>> s = np.random.default_rng().standard_cauchy(1000000) >>> s = s[(s>-25) & (s<25)] # truncate distribution so it plots well >>> plt.hist(s, bins=100) >>> plt.show() poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. Parameters ---------- lam : float or array_like of floats Expectation of interval, must be >= 0. A sequence of expectation intervals must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. Notes ----- The Poisson distribution .. math:: f(k; \lambda)=\frac{\lambda^k e^{-\lambda}}{k!} For events with an expected separation :math:`\lambda` the Poisson distribution :math:`f(k; \lambda)` describes the probability of :math:`k` events occurring within the observed interval :math:`\lambda`. Because the output is limited to the range of the C 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. http://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 >>> rng = np.random.default_rng() >>> s = rng.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 = rng.poisson(lam=(100., 500.), size=(100, 2)) laplace(loc=0.0, scale=1.0, size=None) Draw samples from the Laplace or double exponential distribution with specified location (or mean) and scale (decay). The Laplace distribution is similar to the Gaussian/normal distribution, but is sharper at the peak and has fatter tails. It represents the difference between two independent, identically distributed exponential random variables. Parameters ---------- loc : float or array_like of floats, optional The position, :math:`\mu`, of the distribution peak. Default is 0. scale : float or array_like of floats, optional :math:`\lambda`, the exponential decay. Default is 1. 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. Notes ----- It has the probability density function .. math:: f(x; \mu, \lambda) = \frac{1}{2\lambda} \exp\left(-\frac{|x - \mu|}{\lambda}\right). The first law of Laplace, from 1774, states that the frequency of an error can be expressed as an exponential function of the absolute magnitude of the error, which leads to the Laplace distribution. For many problems in economics and health sciences, this distribution seems to model the data better than the standard Gaussian distribution. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] Kotz, Samuel, et. al. "The Laplace Distribution and Generalizations, " Birkhauser, 2001. .. [3] Weisstein, Eric W. "Laplace Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LaplaceDistribution.html .. [4] Wikipedia, "Laplace distribution", https://en.wikipedia.org/wiki/Laplace_distribution Examples -------- Draw samples from the distribution >>> loc, scale = 0., 1. >>> s = np.random.default_rng().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. 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 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, http://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.default_rng().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") Unsupported dtype %r for randomGenerator.triangular (line 2597)Generator.standard_t (line 1589)unable to allocate array data.probabilities do not sum to 1a must an array or an integerGenerator.logseries (line 3263)Generator.lognormal (line 2348)Generator.geometric (line 3069)Generator.dirichlet (line 3915)Generator.chisquare (line 1376)Generator.vonmises (line 1683)Generator.rayleigh (line 2460)Generator.logistic (line 2268)Generator.binomial (line 2697)nsample must be nonnegative.itemsize <= 0 for cython.arraya and p must have same sizeGenerator.weibull (line 1864)Generator.shuffle (line 4114)Generator.poisson (line 2917)Generator.laplace (line 2064)Generator.integers (line 362)nsample must be an integermean must be 1 dimensionalRange exceeds valid boundsGenerator.uniform (line 762)Generator.pareto (line 1766)Generator.gumbel (line 2149)probabilities contain NaNnsample must not exceed %dInvalid shape in axis %d: %d.Generator.random (line 155)Generator.power (line 1963)Generator.normal (line 939)Generator.gamma (line 1132)Generator.choice (line 518)Generator.zipf (line 2989)Generator.wald (line 2529)Generator.bytes (line 489)Cannot index with type '%s'p must be 1-dimensionalnumpy.random._generatornumpy.core.multiarrayGenerator.f (line 1210)normalize_axis_indexNotImplementedErrornsample > sum(colors)ngood + nbad < nsamplecline_in_traceback__pyx_unpickle_Enumascontiguousarraymay_share_memorysum(pvals[:-1]) > 1.0standard_normal__setstate_cython____pyx_PickleError_poisson_lam_maxdtype_is_objectallocate_bufferView.MemoryView__generator_ctorRuntimeWarning__reduce_cython____pyx_getbuffer_generator.pyxcount_nonzerobit_generatorOverflowErrorstringsourcesearchsortedreturn_index__pyx_checksumdefault_rngcheck_validPickleErrorMemoryErrorImportError__pyx_vtable____pyx_resultnumpy.dualmode > rightlogical_orless_equalleft == rightissubdtypeempty_likeValueErrorIndexError__reduce_ex____pyx_statemarginalsleft > modeenumerateTypeErrorGeneratorwarningsswapaxessubtract__setstate__reversed__reduce____pyx_typepickleoperatoritemsizeisscalarisnativeisfiniteintegers__getstate__floatingendpointcholeskyallcloseEllipsistobytesstridesshuffle__rmatmul__reshapereplacensamplememviewinteger__imatmul__greaterfortranfloat64float32capsule at 0x{:X}asarrayalpha <= 0updateunpackuniqueuint64uint32uint16structreducerandom_picklenamemethod__matmul____import__ignoreformatencodedoublecumsumcompatcolorsastypearangezerosuint8statestartsigmashapescalerightravelrangeraisepvals_pcg64numpyngoodkappaisnanint64int32int16indexflagsfinfoerrorequal__enter__emptydtypedfnumdfdencount__class__arrayalphaPCG64ASCIIwarn__test__takestopstepsqrtsortsizesideseedrtolprodpacknoncndimnbad__name__modemean__main__longlocklessleftitemintpint8highfull__exit__eigh__dict__copybool_baseaxisatolzigtolsvd__str__outobj__new__maxlowloclamepsdotcovanyalladd?333333?C?/*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..&$ŗ: Aޓ=?~)@ lѿ3 ; @UUUUUU?"@m{??@@5gG8?SˆB?AAz?<ٰj_?$+K?88C?J?llf?UUUUUU?dg?$@= ףp=@n?[ m?h|?5?333333 @r?$~?B>٬ @r鷯?Q?Q?9v?(\@ffffff@0C.@4@x&?@?UUUUUU?a@X@`@|@@MA>@@-DT! @h㈵>-DT!@3?r?q?0@9B.? * ?,|l @yD@:5/?@@R2B@96SC@wz*E@r4dF@OOfq]@Ob^@+NT_@ݭC#`@~{`@kbba@YSȐa@n b@1Ib@5ca c@c@ͦ3 d@\>d@nz e@s9Je@FGGʪ f@yyuf@IJC g@Y&g@oFh@·h@aQL i@ai@ F~x*j@&Pj@7k@!+k@VFl@ l@tVm@pZNm@k9ihn@HQOUn@a,~|o@b4nʼnp@+e Ip@cp@)Vp@*q@6Gaq@q@>m#FJq@FK.5r@b)C|r@Wrr@V] s@rRs@GIqs@ >6qs@jB*t@ A=rt@fIw|t@d'-u@X+{ Mu@# u@ZGDu@;#(v@b%rv@iv{Իv@w@Ow@\&әw@}6-#w@h͙.x@k?7yx@–'x@_*y@Yy@1*y@^TTy@,{Lp??@8AP BBRBBA A(D0 (C ABBF 80T?RBBA A(D0 (C ABBF @lx@ADD0f AAG X AAF B CAB 4A40AGBDD Q ABB YGBDHA BAA ^ ABK M ABH z ABK @DBADD0_ AAF X AAF B CAJ LB"BBI A(D@ (D ABBD U (D ABBB DBBE D(D0DPkXH`YXFP^ 0A(A BBBC TXa`AhApBxADBFAQPg 0A(A BBBI dFBBE D(D0DPkXH`YXFP^ 0A(A BBBC TXa`AhApBxADBFAQPg 0A(A BBBI IBBE D(D0DPkXH`YXFP^ 0A(A BBBC TXa`AhApBxADBFAQPg 0A(A BBBI ||HKBBD D(D@dHLP`HF@^ (D ABBC SHaPAXA`BhApHxEFAP@c (D ABBG xMHBFE D(D0D`KhDpkhF`^ 0A(A BBBA YhZpFxFBFAEFAJ`g 0A(A BBBC hWp_hA`,P^BFB B(D0D8DpKxZBFFBFABFJpk 8A0A(B BBBA ixDkxFp^ 8A0A(B BBBF xW_xApDRBBB A(D0D@dHLP`HF@^ 0D(A BBBB UH^PBXA`AhBpAxDBFP@c 0D(A BBBD UyBBB D(A0D`h_pBxFABFAEFP`Y 0D(A BBBE N 0D(A BBBG khLp`hF`^ 0D(A BBBB pWyBBB D(A0D`h_pBxFABFAEFP`Y 0D(A BBBE N 0D(A BBBG khLp`hF`^ 0D(A BBBB ZyBBB D(A0D`h_pBxFABFAEFP`Y 0D(A BBBE N 0D(A BBBG khLp`hF`^ 0D(A BBBB ]yBBB D(A0D`h_pBxFABFAEFP`Y 0D(A BBBE N 0D(A BBBG khLp`hF`^ 0D(A BBBB \x`BBB B(D0D8D`JhepBxAABADBFQ`m 8A0A(B BBBA ~hHpYhF`^ 8A0A(B BBBD bBBB B(D0D8D`JhepBxAABADBFQ`m 8A0A(B BBBA ~hHpYhF`^ 8A0A(B BBBD eBBB B(D0D8D`JhepBxAABADBFQ`m 8A0A(B BBBA ~hHpYhF`^ 8A0A(B BBBD $ Pg^BFB B(D0D8DpKxZBFFBFABFJpk 8A0A(B BBBA ixDkxFp^ 8A0A(B BBBF xW_xAp jBBB B(D0D8D`JhepBxAABADBFQ`m 8A0A(B BBBA ~hHpYhF`^ 8A0A(B BBBD h!LlZBBB A(D0DP{Xe`BhApBxBABBAJPk 0A(A BBBG yXH`YXFP^ 0A(A BBBJ ! n^BFB B(D0D8DpKxZBFFBFABFJpk 8A0A(B BBBA ixDkxFp^ 8A0A(B BBBF xW_xAp"pBBB B(D0D8D`JhepBxAABADBFQ`m 8A0A(B BBBA ~hHpYhF`^ 8A0A(B BBBD 8#sBFB B(D0A8GpOx`BFAEFABFJpk 8A0A(B BBBE UxDkxFp^ 8A0A(B BBBF xW_xApNxV`xGp#Tv^BFB B(D0D8DpKxZBFFBFABFJpk 8A0A(B BBBA ixDkxFp^ 8A0A(B BBBF xW_xAp|$yPBBB B(D0D8D`hHpYhF`^ 8D0A(B BBBI ShZpBxFFBFABFP`%{yBBB D(A0D`h_pBxAABFAEAW`Y 0D(A BBBH N 0D(A BBBG khLp`hF`^ 0D(A BBBB %~BBB A(D0D@dHLP`HF@^ 0D(A BBBB UH^PBXA`AhBpAxDBFP@c 0D(A BBBD H&܀^BFB B(D0D8DpKxZBFFBFABFJpk 8A0A(B BBBA ixDkxFp^ 8A0A(B BBBF xW_xApH&BBB J(A0A8DP 8D0A(B BBBK 4@'$ADD n AAG x CAD @x'AAD0a AAG X AAF r CAJ L'BBA D(D0 (D ABBD  (D ABBC 8 (XBBA A(D@ (A ABBH LH(jBED A(GP (D ABBC n (D ABBI <( BBB A(A0 (A BBBG (|A D b F 0(hAJ0 AD R AE nA0)ĝ0D)AJi AK R AE nAx)\X)X BBB D(A0DP 0A(A BBBG { 0C(A BBBK d)BBB B(A0A8DPu 8A0A(B BBBC   8F0A(B BBBD LP*tBBE E(D0D8G 8A0A(B BBBH L*BBA A(D@ (D ABBF ~ (G ABBF H* BBB E(A0A8D` 8D0A(B BBBI d<+[BBB A(D0G@ 0C(A BBBC o 0C(A BBBG HWP_HF@8+BHD D(D0s (D ABBF 8+TBBA A(D@` (A ABBH H,BBB J(A0A8DP 8D0A(B BBBF 8h,,WBBA A(D0T (A ABBD L,PBBB B(D0A8D 8D0A(B BBBD d,iBBE B(D0A8D`hEpkhF`c 8D0A(B BBBJ hWp_hA`\-;Dv(t-yAD0 AAH d-BBB B(A0A8DP 8D0A(B BBBA B 8C0A(B BBBB .|BFB B(A0A8Gp 8A0A(B BBBA NxV_xFp 8A0A(B BBBF xW_xApNxV`xGp.Du G s E t.|>KBA A(G0 (A ABBD B (F ABBC D (G ABBH F0$0/7daAA ~AAHX/sBEE B(A0A8DP8D0A(B BBB@/d)BBB A(A0JI)0A(A BBB/'BBB B(A0D8DYLB 8D0A(B BBBF HYFYQA5YODzYRB\MB DQASBODERB%DLAH0LBBB B(D0A8GPF 8F0A(B BBBG \41BBB B(A0A8D@ 8C0A(B BBBC D8F0A(B BBBD1PAAG DAE Q DAJ B DAI L1% BBG G(D0A8Du 8D0A(B BBBC ,21Dj B LH2|1BEB B(A0A8D3 8D0A(B BBBG H2 @CBBE B(A0A8DP' 8C0A(B BBBD T2G) BBB B(A0A8D`hHpYhF`^ 8D0A(B BBBE |<3QBBE B(D0A8DDlFh 8D0A(B BBBJ #W_F-VaGp3hfbBFB B(D0A8G8 8D0A(B BBBC =DlF(W_F04;BFB B(D0A8GpDlFb 8D0A(B BBBC zGIB|W`FGDAH4BBB B(D0D8FP 8F0A(B BBBE 5@7BHB E(A0A8I: 8D0A(B BBBE D_B] ISBA}A-B{COA|D 5,>t BBB B(D0A8J)GGGGGGGGGDDDDDDDDDDDDDDDCGGGGGGGGGGDDDDDDDDDDDDDDDCLX 8A0A(B BBBG H6I BBB E(D0A8Dp 8D0A(B BBBA ((7RBIE B(A0A8GBGGGGGGGGGGGGGGGGGGGGGGGGGS 8D0A(B BBBH lGGGGGGGGGGGGGGGGGGGGGGGGGSLT8gIBB A(A0 (A BBBB ` (A BBBI 8iEBFB B(A0D8DeFBBFBBRyDkF^ 8D0A(B BBBC HSEBFBBFBYjW_FVaGx9Qz%BFE B(A0A8GEkFb 8D0A(B BBBF W_F>,v BEB B(A0A8Dp$xoBAABADBFUpq 8D0A(B BBBF mxBLxApxHYxFpd?BBB B(A0A8G` 8A0A(B BBBG F 8C0A(B BBBF ?ЏBBB B(A0A8DpBAFBABBARSSAFBABBATL`F^ 8D0A(B BBBH L<@ BGB B(A0F8D 8D0A(B BBBJ @HaDZ B X@x"BBB B(D0A8DpHYF^ 8D0A(B BBBE XA;BBB E(A0A8Dp< 8D0A(B BBBF SxL`xFp`A1BDE J(A0D8J 8A0A(B BBBA eH\BHQMdBc1HRAA`BEB B(A0D8GJIAD 8A0A(B BBBA IYF\ 8C0A(B BBBJ lB BP%G]Bh ,Bd9GED ^ABG,BtIGED qABDHCBIH H(GP (E ABBE c(A AFB,\Ch9GED _ABFTCx2BEI H(H0D@ 0D(A BBBH r0A(A FBB,C`9GED _ABF,DpQGED {ABB,DDqGED UABHHtDBBE I(H0K8K` 8E0A(B BBBI ,DD9GED _ABF@DTBEI H(H0D@ 0A(A BBBF ,4E9GED _ABF8dE]G@y AB x AG (AG[@8ED[G0 AD t AK ;AD[0EGJEGJ FGJ$F 08FrAD@C EAH hlFh(D cFD UF,D gFD UFD U(FAG0J AM AGtDQ(G|SAG }AHG/AG ]AhG'D bG>\ ]G ?D n(G4AG R AM XAGgAG UAGNAG DAHDI4H ^D B E PHdAG0e EA DtHAHDpa DAE M DAF ! DAA H&IXLH BFG E(A0D8J 8D0A(B BBBG 8(I:BED D(Dpk (A ABBD dI YiFPL<I| /AG0 AD m AJ L AC P DE Il WAG AA I IO0 EA 4 Jx AGP AG ^ AA L AC DJAG@ AG hJWD RJ$`D VJl JxNAG0?F J L@ I r E 0JMGG T ABF hH(K ZBBB B(D0A8DP} 8D0A(B BBBE HtK BBB B(A0A8FPm 8C0A(B BBBG xKBDB B(A0A8DP 8A0A(B BBBB @ 8D0A(B BBBC \ 8A0A(E BBBG x>> rng = np.random.default_rng() >>> rng.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) # random >>> rng.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) # random >>> arr = np.arange(9).reshape((3, 3)) >>> rng.permutation(arr) array([[6, 7, 8], # random [0, 1, 2], [3, 4, 5]]) >>> rng.permutation("abc") Traceback (most recent call last): ... numpy.AxisError: axis 0 is out of bounds for array of dimension 0 >>> arr = np.arange(9).reshape((3, 3)) >>> rng.permutation(arr, axis=1) array([[0, 2, 1], # random [3, 5, 4], [6, 8, 7]]) shuffle(x, axis=0) Modify a sequence in-place by shuffling its contents. The order of sub-arrays is changed but their contents remains the same. Parameters ---------- x : array_like The array or list to be shuffled. axis : int, optional The axis which `x` is shuffled along. Default is 0. It is only supported on `ndarray` objects. Returns ------- None Examples -------- >>> rng = np.random.default_rng() >>> arr = np.arange(10) >>> rng.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] # random >>> arr = np.arange(9).reshape((3, 3)) >>> rng.shuffle(arr) >>> arr array([[3, 4, 5], # random [6, 7, 8], [0, 1, 2]]) >>> arr = np.arange(9).reshape((3, 3)) >>> rng.shuffle(arr, axis=1) >>> arr array([[2, 0, 1], # random [5, 3, 4], [8, 6, 7]]) 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. 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 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, http://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.default_rng().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") multivariate_hypergeometric(colors, nsample, size=None, method='marginals') Generate variates from a multivariate hypergeometric distribution. The multivariate hypergeometric distribution is a generalization of the hypergeometric distribution. Choose ``nsample`` items at random without replacement from a collection with ``N`` distinct types. ``N`` is the length of ``colors``, and the values in ``colors`` are the number of occurrences of that type in the collection. The total number of items in the collection is ``sum(colors)``. Each random variate generated by this function is a vector of length ``N`` holding the counts of the different types that occurred in the ``nsample`` items. The name ``colors`` comes from a common description of the distribution: it is the probability distribution of the number of marbles of each color selected without replacement from an urn containing marbles of different colors; ``colors[i]`` is the number of marbles in the urn with color ``i``. Parameters ---------- colors : sequence of integers The number of each type of item in the collection from which a sample is drawn. The values in ``colors`` must be nonnegative. To avoid loss of precision in the algorithm, ``sum(colors)`` must be less than ``10**9`` when `method` is "marginals". nsample : int The number of items selected. ``nsample`` must not be greater than ``sum(colors)``. size : int or tuple of ints, optional The number of variates to generate, either an integer or a tuple holding the shape of the array of variates. If the given size is, e.g., ``(k, m)``, then ``k * m`` variates are drawn, where one variate is a vector of length ``len(colors)``, and the return value has shape ``(k, m, len(colors))``. If `size` is an integer, the output has shape ``(size, len(colors))``. Default is None, in which case a single variate is returned as an array with shape ``(len(colors),)``. method : string, optional Specify the algorithm that is used to generate the variates. Must be 'count' or 'marginals' (the default). See the Notes for a description of the methods. Returns ------- variates : ndarray Array of variates drawn from the multivariate hypergeometric distribution. See Also -------- hypergeometric : Draw samples from the (univariate) hypergeometric distribution. Notes ----- The two methods do not return the same sequence of variates. The "count" algorithm is roughly equivalent to the following numpy code:: choices = np.repeat(np.arange(len(colors)), colors) selection = np.random.choice(choices, nsample, replace=False) variate = np.bincount(selection, minlength=len(colors)) The "count" algorithm uses a temporary array of integers with length ``sum(colors)``. The "marginals" algorithm generates a variate by using repeated calls to the univariate hypergeometric sampler. It is roughly equivalent to:: variate = np.zeros(len(colors), dtype=np.int64) # `remaining` is the cumulative sum of `colors` from the last # element to the first; e.g. if `colors` is [3, 1, 5], then # `remaining` is [9, 6, 5]. remaining = np.cumsum(colors[::-1])[::-1] for i in range(len(colors)-1): if nsample < 1: break variate[i] = hypergeometric(colors[i], remaining[i+1], nsample) nsample -= variate[i] variate[-1] = nsample The default method is "marginals". For some cases (e.g. when `colors` contains relatively small integers), the "count" method can be significantly faster than the "marginals" method. If performance of the algorithm is important, test the two methods with typical inputs to decide which works best. .. versionadded:: 1.18.0 Examples -------- >>> colors = [16, 8, 4] >>> seed = 4861946401452 >>> gen = np.random.Generator(np.random.PCG64(seed)) >>> gen.multivariate_hypergeometric(colors, 6) array([5, 0, 1]) >>> gen.multivariate_hypergeometric(colors, 6, size=3) array([[5, 0, 1], [2, 2, 2], [3, 3, 0]]) >>> gen.multivariate_hypergeometric(colors, 6, size=(2, 2)) array([[[3, 2, 1], [3, 2, 1]], [[4, 1, 1], [3, 2, 1]]]) 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``. Parameters ---------- n : int or array-like of ints 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. Examples -------- Throw a dice 20 times: >>> rng = np.random.default_rng() >>> rng.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: >>> rng.multinomial(20, [1/6.]*6, size=2) array([[3, 4, 3, 3, 4, 3], [2, 4, 3, 4, 0, 7]]) # random 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. Now, do one experiment throwing the dice 10 time, and 10 times again, and another throwing the dice 20 times, and 20 times again: >>> rng.multinomial([[10], [20]], [1/6.]*6, size=(2, 2)) array([[[2, 4, 0, 1, 2, 1], [1, 3, 0, 3, 1, 2]], [[1, 4, 4, 4, 4, 3], [3, 3, 2, 5, 5, 2]]]) # random The first array shows the outcomes of throwing the dice 10 times, and the second shows the outcomes from throwing the dice 20 times. A loaded die is more likely to land on number 6: >>> rng.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: >>> rng.multinomial(100, [1.0 / 3, 2.0 / 3]) # RIGHT array([38, 62]) # random not like: >>> rng.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. 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. method : { 'svd', 'eigh', 'cholesky'}, optional The cov input is used to compute a factor matrix A such that ``A @ A.T = cov``. This argument is used to select the method used to compute the factor matrix A. The default method 'svd' is the slowest, while 'cholesky' is the fastest but less robust than the slowest method. The method `eigh` uses eigen decomposition to compute A and is faster than svd but slower than cholesky. .. versionadded:: 1.18.0 Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. Notes ----- The mean is a coordinate in N-dimensional space, which represents the location where samples are most likely to be generated. This is analogous to the peak of the bell curve for the one-dimensional or univariate normal distribution. Covariance indicates the level to which two variables vary together. From the multivariate normal distribution, we draw N-dimensional samples, :math:`X = [x_1, x_2, ... x_N]`. The covariance matrix element :math:`C_{ij}` is the covariance of :math:`x_i` and :math:`x_j`. The element :math:`C_{ii}` is the variance of :math:`x_i` (i.e. its "spread"). Instead of specifying the full covariance matrix, popular approximations include: - Spherical covariance (`cov` is a multiple of the identity matrix) - Diagonal covariance (`cov` has non-negative elements, and only on the diagonal) This geometrical property can be seen in two dimensions by plotting generated data-points: >>> mean = [0, 0] >>> cov = [[1, 0], [0, 100]] # diagonal covariance Diagonal covariance means that points are oriented along x or y-axis: >>> import matplotlib.pyplot as plt >>> x, y = np.random.default_rng().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]] >>> rng = np.random.default_rng() >>> x = rng.multivariate_normal(mean, cov, (3, 3)) >>> x.shape (3, 3, 2) We can use a different method other than the default to factorize cov: >>> y = rng.multivariate_normal(mean, cov, (3, 3), method='cholesky') >>> y.shape (3, 3, 2) The following is probably true, given that 0.6 is roughly twice the standard deviation: >>> list((x[0,0,:] - mean) < 0.6) [True, True] # random logseries(p, size=None) Draw samples from a logarithmic series distribution. Samples are drawn from a log series distribution with specified shape parameter, 0 < ``p`` < 1. Parameters ---------- p : float or array_like of floats Shape parameter for the distribution. Must be in the range (0, 1). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logarithmic series distribution. See Also -------- scipy.stats.logser : probability density function, distribution or cumulative density function, etc. Notes ----- The probability mass function 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.default_rng().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``). Parameters ---------- ngood : int or array_like of ints Number of ways to make a good selection. Must be nonnegative and less than 10**9. nbad : int or array_like of ints Number of ways to make a bad selection. Must be nonnegative and less than 10**9. nsample : int or array_like of ints Number of items sampled. Must be nonnegative and less than ``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 -------- multivariate_hypergeometric : Draw samples from the multivariate hypergeometric distribution. scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{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. The arguments `ngood` and `nbad` each must be less than `10**9`. For extremely large arguments, the algorithm that is used to compute the samples [4]_ breaks down because of loss of precision in floating point calculations. For such large values, if `nsample` is not also large, the distribution can be approximated with the binomial distribution, `binomial(n=nsample, p=ngood/(ngood + nbad))`. References ---------- .. [1] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [2] Weisstein, Eric W. "Hypergeometric Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", https://en.wikipedia.org/wiki/Hypergeometric_distribution .. [4] Stadlober, Ernst, "The ratio of uniforms approach for generating discrete random variates", Journal of Computational and Applied Mathematics, 31, pp. 181-189 (1990). Examples -------- Draw samples from the distribution: >>> rng = np.random.default_rng() >>> ngood, nbad, nsamp = 100, 2, 10 # number of good, number of bad, and number of samples >>> s = rng.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 = rng.hypergeometric(15, 15, 15, 100000) >>> sum(s>=12)/100000. + sum(s<=3)/100000. # answer = 0.003 ... pretty unlikely! geometric(p, size=None) Draw samples from the geometric distribution. Bernoulli trials are experiments with one of two outcomes: success or failure (an example of such an experiment is flipping a coin). The geometric distribution models the number of trials that must be run in order to achieve success. It is therefore supported on the positive integers, ``k = 1, 2, ...``. The probability mass function of the geometric distribution is .. math:: f(k) = (1 - p)^{k - 1} p where `p` is the probability of success of an individual trial. Parameters ---------- p : float or array_like of floats The probability of success of an individual trial. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized geometric distribution. Examples -------- Draw ten thousand values from the geometric distribution, with the probability of an individual success equal to 0.35: >>> z = np.random.default_rng().geometric(p=0.35, size=10000) How many trials succeeded after a single run? >>> (z == 1).sum() / 10000. 0.34889999999999999 #random zipf(a, size=None) Draw samples from a Zipf distribution. Samples are drawn from a Zipf distribution with specified parameter `a` > 1. The Zipf distribution (also known as the zeta distribution) is a continuous probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. Parameters ---------- a : float or array_like of floats Distribution parameter. 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. Notes ----- The probability density for the Zipf distribution is .. math:: p(x) = \frac{x^{-a}}{\zeta(a)}, where :math:`\zeta` is the Riemann Zeta function. It is named for the American linguist George Kingsley Zipf, who noted that the frequency of any word in a sample of a language is inversely proportional to its rank in the frequency table. References ---------- .. [1] Zipf, G. K., "Selected Studies of the Principle of Relative Frequency in Language," Cambridge, MA: Harvard Univ. Press, 1932. Examples -------- Draw samples from the distribution: >>> a = 2. # parameter >>> s = np.random.default_rng().zipf(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy import special # doctest: +SKIP Truncate s values at 50 so plot is interesting: >>> count, bins, ignored = plt.hist(s[s<50], ... 50, density=True) >>> x = np.arange(1., 50.) >>> y = x**(-a) / special.zetac(a) # doctest: +SKIP >>> plt.plot(x, y/max(y), linewidth=2, color='r') # doctest: +SKIP >>> plt.show() poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. Parameters ---------- lam : float or array_like of floats Expectation of interval, must be >= 0. A sequence of expectation intervals must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. Notes ----- The Poisson distribution .. math:: f(k; \lambda)=\frac{\lambda^k e^{-\lambda}}{k!} For events with an expected separation :math:`\lambda` the Poisson distribution :math:`f(k; \lambda)` describes the probability of :math:`k` events occurring within the observed interval :math:`\lambda`. Because the output is limited to the range of the C 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. http://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 >>> rng = np.random.default_rng() >>> s = rng.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 = rng.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]. Parameters ---------- n : float or array_like of floats Parameter of the distribution, > 0. p : float or array_like of floats Parameter of the distribution. Must satisfy 0 < p <= 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. 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 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. http://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.default_rng().negative_binomial(1, 0.1, 100000) >>> for i in range(1, 11): # doctest: +SKIP ... probability = sum(s>> rng = np.random.default_rng() >>> n, p = 10, .5 # number of trials, probability of each trial >>> s = rng.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(rng.binomial(9, 0.1, 20000) == 0)/20000. # answer = 0.38885, or 38%. triangular(left, mode, right, size=None) Draw samples from the triangular distribution over the interval ``[left, right]``. The triangular distribution is a continuous probability distribution with lower limit left, peak at mode, and upper limit right. Unlike the other distributions, these parameters directly define the shape of the pdf. Parameters ---------- left : float or array_like of floats Lower limit. mode : float or array_like of floats The value where the peak of the distribution occurs. The value 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. 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.default_rng().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. 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. 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.default_rng().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. 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. 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 >>> rng = np.random.default_rng() >>> values = hist(rng.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 = rng.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. 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. 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: >>> rng = np.random.default_rng() >>> mu, sigma = 3., 1. # mean and standard deviation >>> s = rng.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. >>> rng = rng >>> b = [] >>> for i in range(1000): ... a = 10. + rng.standard_normal(100) ... b.append(np.product(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, 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). 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. Notes ----- The probability density for the Logistic distribution is .. math:: P(x) = P(x) = \frac{e^{-(x-\mu)/s}}{s(1+e^{-(x-\mu)/s})^2}, where :math:`\mu` = location and :math:`s` = scale. The Logistic distribution is used in Extreme Value problems where it can act as a mixture of Gumbel distributions, in Epidemiology, and by the World Chess Federation (FIDE) where it is used in the Elo ranking system, assuming the performance of each player is a logistically distributed random variable. References ---------- .. [1] Reiss, R.-D. and Thomas M. (2001), "Statistical Analysis of Extreme Values, from Insurance, Finance, Hydrology and Other Fields," Birkhauser Verlag, Basel, pp 132-133. .. [2] Weisstein, Eric W. "Logistic Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LogisticDistribution.html .. [3] Wikipedia, "Logistic-distribution", https://en.wikipedia.org/wiki/Logistic_distribution Examples -------- Draw samples from the distribution: >>> loc, scale = 10, 1 >>> s = np.random.default_rng().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. 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 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: >>> rng = np.random.default_rng() >>> mu, beta = 0, 0.1 # location and scale >>> s = rng.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 = rng.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. 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. Notes ----- It has the probability density function .. math:: f(x; \mu, \lambda) = \frac{1}{2\lambda} \exp\left(-\frac{|x - \mu|}{\lambda}\right). The first law of Laplace, from 1774, states that the frequency of an error can be expressed as an exponential function of the absolute magnitude of the error, which leads to the Laplace distribution. For many problems in economics and health sciences, this distribution seems to model the data better than the standard Gaussian distribution. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] Kotz, Samuel, et. al. "The Laplace Distribution and Generalizations, " Birkhauser, 2001. .. [3] Weisstein, Eric W. "Laplace Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/LaplaceDistribution.html .. [4] Wikipedia, "Laplace distribution", https://en.wikipedia.org/wiki/Laplace_distribution Examples -------- Draw samples from the distribution >>> loc, scale = 0., 1. >>> s = np.random.default_rng().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. 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 < 1. Notes ----- The probability density function is .. math:: P(x; a) = ax^{a-1}, 0 \le x \le 1, a>0. The power function distribution is just the inverse of the Pareto distribution. It may also be seen as a special case of the Beta distribution. It is used, for example, in modeling the over-reporting of insurance claims. References ---------- .. [1] Christian Kleiber, Samuel Kotz, "Statistical size distributions in economics and actuarial sciences", Wiley, 2003. .. [2] Heckert, N. A. and Filliben, James J. "NIST Handbook 148: Dataplot Reference Manual, Volume 2: Let Subcommands and Library Functions", National Institute of Standards and Technology Handbook Series, June 2003. https://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf Examples -------- Draw samples from the distribution: >>> rng = np.random.default_rng() >>> a = 5. # shape >>> samples = 1000 >>> s = rng.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 = rng.power(5, 1000000) >>> rvsp = rng.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('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 + Generator.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}`. 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 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: >>> rng = np.random.default_rng() >>> a = 5. # shape >>> s = rng.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(rng.weibull(5.,1000)) >>> x = np.arange(1,100.)/50. >>> scale = count.max()/weib(x, 1., 5.).max() >>> plt.plot(x, weib(x, 1., 5.)*scale) >>> plt.show() pareto(a, size=None) Draw samples from a Pareto II or Lomax distribution with specified shape. The Lomax or Pareto II distribution is a shifted Pareto distribution. The classical Pareto distribution can be obtained from the Lomax distribution by adding 1 and multiplying by the scale parameter ``m`` (see Notes). The smallest value of the Lomax distribution is zero while for the classical Pareto distribution it is ``mu``, where the standard Pareto distribution has location ``mu = 1``. Lomax can also be considered as a simplified version of the Generalized Pareto distribution (available in SciPy), with the scale set to one and the location set to zero. The Pareto distribution must be greater than zero, and is unbounded above. It is also known as the "80-20 rule". In this distribution, 80 percent of the weights are in the lowest 20 percent of the range, while the other 20 percent fill the remaining 80 percent of the range. Parameters ---------- a : float or array_like of floats Shape of the distribution. 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. 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.default_rng().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. Parameters ---------- mu : float or array_like of floats Mode ("center") of the distribution. kappa : float or array_like of floats Dispersion of the distribution, has to be >=0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mu`` and ``kappa`` are both scalars. Otherwise, ``np.broadcast(mu, kappa).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized von Mises distribution. See Also -------- scipy.stats.vonmises : probability density function, distribution, or cumulative density function, etc. Notes ----- The probability density for the von Mises distribution is .. math:: p(x) = \frac{e^{\kappa cos(x-\mu)}}{2\pi I_0(\kappa)}, where :math:`\mu` is the mode and :math:`\kappa` the dispersion, and :math:`I_0(\kappa)` is the modified Bessel function of order 0. The von Mises is named for Richard Edler von Mises, who was born in Austria-Hungary, in what is now the Ukraine. He fled to the United States in 1939 and became a professor at Harvard. He worked in probability theory, aerodynamics, fluid mechanics, and philosophy of science. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] von Mises, R., "Mathematical Theory of Probability and Statistics", New York: Academic Press, 1964. Examples -------- Draw samples from the distribution: >>> mu, kappa = 0.0, 4.0 # mean and dispersion >>> s = np.random.default_rng().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`). 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. 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? We have 10 degrees of freedom, so is the sample mean within 95% of the recommended value? >>> s = np.random.default_rng().standard_t(10, size=100000) >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 Calculate the t statistic, setting the ddof parameter to the unbiased value so the divisor in the standard deviation will be degrees of freedom, N-1. >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> import matplotlib.pyplot as plt >>> h = plt.hist(s, bins=100, density=True) For a one-sided t-test, how far out in the distribution does the t statistic appear? >>> np.sum(s>> import matplotlib.pyplot as plt >>> s = np.random.default_rng().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. 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. 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 >>> rng = np.random.default_rng() >>> import matplotlib.pyplot as plt >>> values = plt.hist(rng.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(rng.noncentral_chisquare(3, .0000001, 100000), ... bins=np.arange(0., 25, .1), density=True) >>> values2 = plt.hist(rng.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(rng.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. 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. 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.default_rng().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. 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. Notes ----- When calculating the power of an experiment (power = probability of rejecting the null hypothesis when a specific alternative is true) the non-central F statistic becomes important. When the null hypothesis is true, the F statistic follows a central F distribution. When the null hypothesis is not true, then it follows a non-central F statistic. References ---------- .. [1] Weisstein, Eric W. "Noncentral F-Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/NoncentralF-Distribution.html .. [2] Wikipedia, "Noncentral F-distribution", 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. >>> rng = np.random.default_rng() >>> dfnum = 3 # between group deg of freedom >>> dfden = 20 # within groups degrees of freedom >>> nonc = 3.0 >>> nc_vals = rng.noncentral_f(dfnum, dfden, nonc, 1000000) >>> NF = np.histogram(nc_vals, bins=50, density=True) >>> c_vals = rng.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. 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. 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.default_rng().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. 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. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", 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.default_rng().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, dtype=np.float64, out=None) Draw samples from a standard Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, shape (sometimes designated "k") and scale=1. Parameters ---------- shape : float or array_like of floats Parameter, 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. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/GammaDistribution.html .. [2] Wikipedia, "Gamma distribution", https://en.wikipedia.org/wiki/Gamma_distribution Examples -------- Draw samples from the distribution: >>> shape, scale = 2., 1. # mean and width >>> s = np.random.default_rng().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]_. 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. 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 :meth:`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.default_rng().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 N(3, 6.25): >>> np.random.default_rng().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, dtype=np.float64, out=None) Draw samples from a standard Normal distribution (mean=0, stdev=1). Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. 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. Notes ----- For random samples from :math:`N(\mu, \sigma^2)`, use one of:: mu + sigma * gen.standard_normal(size=...) gen.normal(mu, sigma, size=...) Examples -------- >>> rng = np.random.default_rng() >>> rng.standard_normal() 2.1923875335537315 #random >>> s = rng.standard_normal(8000) >>> s array([ 0.6888893 , 0.78096262, -0.89086505, ..., 0.49876311, # random -0.38672696, -0.4685006 ]) # random >>> s.shape (8000,) >>> s = rng.standard_normal(size=(3, 4, 2)) >>> s.shape (3, 4, 2) Two-by-four array of samples from :math:`N(3, 6.25)`: >>> 3 + 2.5 * rng.standard_normal(size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random uniform(low=0.0, high=1.0, size=None) Draw samples from a uniform distribution. Samples are uniformly distributed over the half-open interval ``[low, high)`` (includes low, but excludes high). In other words, any value within the given interval is equally likely to be drawn by `uniform`. Parameters ---------- low : float or array_like of floats, optional Lower boundary of the output interval. All values generated will be greater than or equal to low. The default value is 0. high : float or array_like of floats Upper boundary of the output interval. All values generated will be less than high. The default value is 1.0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``low`` and ``high`` are both scalars. Otherwise, ``np.broadcast(low, high).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized uniform distribution. See Also -------- integers : Discrete uniform distribution, yielding integers. random : Floats uniformly distributed over ``[0, 1)``. Notes ----- The probability density function of the uniform distribution is .. math:: p(x) = \frac{1}{b - a} anywhere within the interval ``[a, b)``, and zero elsewhere. When ``high`` == ``low``, values of ``low`` will be returned. If ``high`` < ``low``, the results are officially undefined and may eventually raise an error, i.e. do not rely on this function to behave when passed arguments satisfying that inequality condition. Examples -------- Draw samples from the distribution: >>> s = np.random.default_rng().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, axis=0, shuffle=True) Generates a random sample from a given 1-D array Parameters ---------- a : {array_like, int} If an ndarray, a random sample is generated from its elements. If an int, the random sample is generated from np.arange(a). size : {int, tuple[int]}, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn from the 1-d `a`. If `a` has more than one dimension, the `size` shape will be inserted into the `axis` dimension, so the output ``ndim`` will be ``a.ndim - 1 + len(size)``. Default is None, in which case a single value is returned. replace : bool, optional Whether the sample is with or without replacement p : 1-D array_like, optional The probabilities associated with each entry in a. If not given the sample assumes a uniform distribution over all entries in a. axis : int, optional The axis along which the selection is performed. The default, 0, selects by row. shuffle : bool, optional Whether the sample is shuffled when sampling without replacement. Default is True, False provides a speedup. Returns ------- samples : single item or ndarray The generated random samples Raises ------ ValueError If a is an int and less than zero, if p is not 1-dimensional, if a is array-like with a 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 -------- integers, shuffle, permutation Examples -------- Generate a uniform random sample from np.arange(5) of size 3: >>> rng = np.random.default_rng() >>> rng.choice(5, 3) array([0, 3, 4]) # random >>> #This is equivalent to rng.integers(0,5,3) Generate a non-uniform random sample from np.arange(5) of size 3: >>> rng.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: >>> rng.choice(5, 3, replace=False) array([3,1,0]) # random >>> #This is equivalent to rng.permutation(np.arange(5))[:3] Generate a non-uniform random sample from np.arange(5) of size 3 without replacement: >>> rng.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'] >>> rng.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3]) array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'], # random dtype='>> np.random.default_rng().bytes(10) ' eh\x85\x022SZ\xbf\xa4' #random integers(low, high=None, size=None, dtype=np.int64, endpoint=False) Return random integers from `low` (inclusive) to `high` (exclusive), or if endpoint=True, `low` (inclusive) to `high` (inclusive). Replaces `RandomState.randint` (with endpoint=False) and `RandomState.random_integers` (with endpoint=True) Return random integers from the "discrete uniform" distribution of the specified dtype. If `high` is None (the default), then results are from 0 to `low`. 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 0 and this value is used for `high`). 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 np.int64. endpoint : bool, optional If true, sample from the interval [low, high] instead of the default [low, high) Defaults to False 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. Notes ----- When using broadcasting with uint64 dtypes, the maximum value (2**64) cannot be represented as a standard integer type. The high array (or low if high is None) must have object dtype, e.g., array([2**64]). Examples -------- >>> rng = np.random.default_rng() >>> rng.integers(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) # random >>> rng.integers(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: >>> rng.integers(5, size=(2, 4)) array([[4, 0, 2, 1], [3, 2, 2, 0]]) # random Generate a 1 x 3 array with 3 different upper bounds >>> rng.integers(1, [3, 5, 10]) array([2, 2, 9]) # random Generate a 1 by 3 array with 3 different lower bounds >>> rng.integers([1, 5, 7], 10) array([9, 8, 7]) # random Generate a 2 by 4 array using broadcasting with dtype of uint8 >>> rng.integers([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7], [ 1, 16, 9, 12]], dtype=uint8) # random References ---------- .. [1] Daniel Lemire., "Fast Random Integer Generation in an Interval", ACM Transactions on Modeling and Computer Simulation 29 (1), 2019, http://arxiv.org/abs/1805.10941. standard_exponential(size=None, dtype=np.float64, method='zig', out=None) Draw samples from the standard exponential distribution. `standard_exponential` is identical to the exponential distribution with a scale parameter of 1. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. method : str, optional Either 'inv' or 'zig'. 'inv' uses the default inverse CDF method. 'zig' uses the much faster Ziggurat method of Marsaglia and Tsang. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. Returns ------- out : float or ndarray Drawn samples. Examples -------- Output a 3x8000 array: >>> n = np.random.default_rng().standard_exponential((3, 8000)) exponential(scale=1.0, size=None) Draw samples from an exponential distribution. Its probability density function is .. math:: f(x; \frac{1}{\beta}) = \frac{1}{\beta} \exp(-\frac{x}{\beta}), for ``x > 0`` and 0 elsewhere. :math:`\beta` is the scale parameter, which is the inverse of the rate parameter :math:`\lambda = 1/\beta`. The rate parameter is an alternative, widely used parameterization of the exponential distribution [3]_. The exponential distribution is a continuous analogue of the geometric distribution. It describes many common situations, such as the size of raindrops measured over many rainstorms [1]_, or the time between page requests to Wikipedia [2]_. Parameters ---------- scale : float or array_like of floats The scale parameter, :math:`\beta = 1/\lambda`. 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. 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. 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. random(size=None, dtype=np.float64, out=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` by `(b-a)` and add `a`:: (b - a) * random() + a Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result, only `float64` and `float32` are supported. Byteorder must be native. The default value is np.float64. out : ndarray, optional Alternative output array in which to place the result. If size is not None, it must have the same shape as the provided size and must match the type of the output values. Returns ------- out : float or ndarray of floats Array of random floats of shape `size` (unless ``size=None``, in which case a single float is returned). Examples -------- >>> rng = np.random.default_rng() >>> rng.random() 0.47108547995356098 # random >>> type(rng.random()) >>> rng.random((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) # random Three-by-two array of random numbers from [-5, 0): >>> 5 * rng.random((3, 2)) - 5 array([[-3.99149989, -0.52338984], # random [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) x+P+H+X++x++x+x+P++H++P+x+P++X+x+ +@++++P+x+x+P+H+H++x++x+P+H+++x+++x++++x++x+++x+x++x+++x+X+x+X+x+X+x+H++x+H++x+H++x+++x++x+++x+p+++x+`+@+x+`+@+x++x+X+x+@+x++@++x+@+x+++x+H+++`++x+++x+++x+++++++++++H+X++H+++++++@p3+$+7+&+@=++B+8 +#+`&+ +`++ + + ++x+ p+ h+ %`+`X+@P+@H+ @+` 8+p"0+2(+* +`(+ ++#+++o"+ +++ ++&+`*+ %+%+!+!+++`+x+@p+ h+2`+ X+`"P+[H+?@+08+0+п (+  +@+++@)+P++ +++ؾ++ +`!+ "+` +.+)+(+ ++o7x+Fp+h+`+X+P+H+@+@-8+00+8(+ ++( +++ +++z++++ +f+a+++\+`a+@+@+Vx+)p+_sh+`+X+P+H+ @+48+W0+@Kq (+  + +p +++ T++++Q+j+0++%+)++ +++d+d+^x+^p+H h+ 1 `+X+P+XH+@+A8+A0+R(+p  +y+ +H + ++;+5+8 + > +/+)+++ +r+++1+3 ++ x+,Jp+ h+`6`+X+`)P+,H+@+8+k0+ (+`  +#++++'++ + +"+ + + + +` ++ +++++x+` p+h+ `+PX+P+@ H+@+8+0+0 (+`d  +`+` + ++ + +P +++++ #+++E+'+2++  ++x+ < p+h+`+X+P+ H+@@+8+0+(+B  +`$++++`+ 3+++ +{= +l+++++++`+@++`x+l'p+"h+ `+X+P+ H+@+8+0+(+ +]++`X+8+{++p+K +#+@+++++P+++ + +p +x+p+h++`+@D#X+P+H+ ; @+$8+` 0+(+ +x+p+ +P +++` ++++p ++@ +0+++X+7+++x+p+h+`+X+tP+ { H+ @+@.8+`! 0+(+ +++x+ +z!++P+` ++0 +@N++s+s+  ++ +H+o+Y +x+p+h+`+X+P+&H+  @+8+0+(+   +pD +++ +  +++k+@ +`+A*D'(+@+'Y^Vp0U@  ++0m+D0+ +P P% nY!`]P[[ZpYW0TM^ip h @D@+o^ p>p>`+@+0u0+ ++`@0vu0` q uH^rp`qhP @D+@+`@p' 6CPg X+ `@S+ L+6F+A9+ 8+RG++!+W@+`0@ +`*@* ***P** * *С * *@|*o*@]*Q*@*`7*~@-*L#*up*l` *d*_@)|@))Fy)z`))P )n)x T))%/g0).UGCC: (GNU) 4.4.7 20120313 (Red Hat 4.4.7-23)GCC: (GNU) 8.3.1 20190311 (Red Hat 8.3.1-3)H1384X 8 P    8 p{)x{){){)h})))@+ L*7 0K @ f P p @X +8+# E[ y p  H  = 0*  N w 0 ( _ `  ~  ,+L @ f ` dz y H+P++H+++)@+88+a0+q0+(+@+ +0+ +H+C+S+x++++++)+9 +e+u++++++h++,+B+R`+h+x+++++++D+T++8+@++@++% x+5 p+E +` h+p + `+ + X+ + P+ x+E H+U +s @+ + 8+ 0+ + (+ x+  +, p+X + + + + P+ + X+0 +@ +l +| + + + + + +0 +@ +l +| + + + + + + X+$+4+D+T+c0+u++++++ +,+<+]x+m+p++++++ +)+A sW P  ++@ G * > pz v [ gx+ SP+# 4 {N ;b >w  X+ Ab) Q e 8 fH |c pdz    Y [8 P Q P!o " $h %d % & 'uA (a (mvH+~ @)p ) * , / 6^- ;P =hk > @ D0 E `G H J M* Np=+NP+V@+kp+h++ 0URP+  VR= W|a+m pY<x+ ZI [I P[  `]V= ^X+s+ fT hR iR jB+1 0mJ+Z +| o o_H+0f+ qA `qR, rRT t+ u 0uG u  v  w"" y^+n`+|8+0+@+d+ | d+# @ZP+fX+r`d+ ++@d++ HJX+V`+b d+ y ^++8+++c+ * `ec+| y++`c+  y* @+8 H+F @c+ ]  y  c+  y c+  6!b+M! !b+! С!b+! ^)"+9"+F"+S"+c"b+ z" "+"+"Pb+" Z*#@b+A# P^#+#+# b+ # #a+$ O$+_$x+o$+$a+($+$ ^$a+ $ P)%+9%+I%a+ `% 0y% a+ % %`+& `^<&+H&+T&_+ k& &+&+&+&(+&+' P' ' 'p+'h+'  (`+#( pj<( W( ( ( 0( @5) R)  l)X+})(+) @)`+)+) * * @ U*+c*+v*+* `[*X+*8+*e+* + p 1+ _+p+n+++++++++++ W, 0, i3,+L,X+Z,+j,H+,@f+ , ';,+, ', *-0+- /S-+i-+|-h+-+-+-f+ - >. p?>@. da. t.p+.(+.8+.0+.+.+. d)/(+#/f+@=4/+D/+X/+e/+r/+/+/X+/+/@+/+/+/H+0+)00+?08+Y0+l0+00+0p+0(+0+0+1 +1`+/1+H1+y1+++1+1+1`+1+1X+1+2+82+L2+i2+z2 +82  2+2 t 2  "3 t K3+8i3+3 a3 p3+3+4+64+g4x+4p+4+5+85+p5+5+5+6+K6+6+6h+6`+ 7X+J7P+v7H+7@+7(+7 +/8+f8P+y8+8+8+8+8+8+8+9`+9+.9+G9+e9+w9x+9+9+9p+9+9+9+ :h+:x+/:p+M:`+Y:h+r:`+:`+:+ :X+:`+;+B;X+l;+;+;`+;p+<+9<0+d<+<H+<+<@+=+1=+\=+=P+=+=+>+'>+R>+s>+>+>8+>+?+I?+u?`+?+?8+?+ @(+K@+s@ +@+@+@+A+AA+fAx+AX+AX+Ap+B(+3BP+[B +BH+BH+B+C +,C+PC+{C+C+C+C(+#D +OD+yD+D+Dx+Dx+Ex+CEH+nEh+E+E@+Eh+F(+EFx+pF8+Fp+F0+Fh+G+EG+pG+G+G+G+G+H +@$H+>H+ tH A'H`+HX+H+H+IP+IH+;I+TI+rI+I@+I8+I+Iй+Iȹ+J+J+ JP+/JP+?J+RJ0+kJ(+J+J``+0J +J+Jh+J+K+.Kp+?K+XK+vKx+K+K+K`+K+Kع+L+L+'L+SL pihL mL nL+L(+L L+L CM 0) GMP+YM+rM+M`f+M `Mh+M8+M +M+ N+N+*N+=N+LN+fNf+0}N@+N `bN+N+ O8+O+4Ox+ROp+kOh+O`+OX+O+O +OH+O+P+"PP+;PH+YPX+jP+xPh+P8+Pp+P@+P8+P+P0+Q(+/Q`+=Q+IQ+XQ+gQ+uQH+Q+Qe+8Q@+Q(+Q+Q+ Rh+R +1R+OR+]R+vR+R+R+R+R P;(S+c+Wcؾ+ucо+cȾ+c+c+c+c+d+3d+Qd+jd+d+d@e+ d pz%dP+eH+!e@+:e8+Xe+qe+e+ex+ep+eh+e`+fX+4fd+ Kf  fм+fȼ+f+f+f+(g@a+(?g wg+gp+gh+g+g+g h+4hx+Rh+ehp+~hh+h +h`+hX+h+h+ i++i`+ Bi8+Si+ci x"i+i+i+i+j+ j+2j+Kj+ij+j+jp+jh+j`+jX+k+k+6k+Tk+ok+kx+ke+k 6;l+l++l+Il+bl+l+l `+(l @E1lP+mH+3m+Bm@+[m8+ym+m0+m(+m+m +m+n+n+v _v Bv 8 v #v `&v  v `"w Cw  iw w  w w w  x :x %cx `x @x @x  x ` y p"Dy 2my *y `(y +y #z 3z Vz o"~z z z z  { /{ &X{ `*{ %{ %{ !{ !!| D| i| `| | @|  | 2 }  } `"H} [q} ?} 0} } п } ~ @8~ T~ ^~ @)~ P~ ~  ~ ~   ؾ" ,  > `!g " `  . )  (6  I a o7 F  Ȁ Ӏ ݀  @-* 0P 8y   (   ȁ  ؁   z  + :  J fW ad n \{ `a @ VƂ ) _s ! 1  E 4n W @Kq   σ p ݃   T0 ? ] } Q j 0  % )    / : dH ^V H c 1    X ˅ A؅ R p  y   H '  9 E ;S 5a 8 n >  / ) † ҆   r  1 3 7 N d ,J   `6Ƈ և `) ,  3 > kM0+_  o(+ `  #   Lj 'Ԉ      "*  ; I  Z  k ` ~     Ή ډ `     P3 @ @ S _ l 0  `d  `ϊ `       /+@ P O [ t  # ͋ ݋ E ' 2> K  ^ i <       @  ! B I `$r ~  ` 3Ѝ ݍ   {= - lU r }   Ԏ ` @  `= l'e "    ͏  ޏ    ! ]I W `X 8 { Ɛ pސ K  #/ @Q w   P  Ց     p  % 8 F +U @D#}   ;  $В `     x p)  > P O ]+o `    p   @ ͓ 0  X 7+ 8 F S ` m  t {   @." `! J X f s  x  z!Ε  P `   0 0 @NY q s}      H oĖ Y     $ 2 X &    Ǘ ֗   pD % 2  C  k u  k @ + Ř`+0ۘ+`@+P+P9+X+v+ +@+``+P֙@+0+  ++P1+e@+xX+#Ӛ@S+@ L+5JF+t9+ Λ 8+s++q A!+ |+@ +`* ;@* t * *= *&* n* *  * + *e * @|* ٟo* @]*)NQ*d @*Ǡ7* @-*D ;#*Y y*` *B *5) m@)J)) ))p)<  ))1 ;)v)) p  @+,x{)S _p{)~ ` ` ` ` ` `ǥ `ѥ ` ۥ ` ` ` `,DS  [m i{)v{) (+)ɦۦ _)>P\ ЁGy x`ǧ ^>֧ ]S )6DSo 0Lʨը d& @^/,>Rczԩ O92?Vc Ty(+ d Ǫ`+! pw2DV `^f L `\, a ƫګ  1M] x׬ p )BU `Nes LIϭ p|Z  Z!2 WJ z \ }{ή ^?ۮ K psW$? Pqfyϯޯ 0& P3DPdr°ΰ \ڰ  @\ *BThw  \α L%  PZ+ = =Kat \ `yNβ @N22 p^'@Ogxh+}˳ Y  xW%3 `EW tg  R Z @L ɴش(+ \( ]%0AM`v N9 {ѵ sݵ @r/* PAP OQu p` ~۶ q);J @Z Vg| _g PL9˷ PT9 *7 oPeq Po:ȸ  :(7HZ @yk|Źҹ" 8 @R93K^kѺ/usr/lib/../lib64/crti.ocall_gmon_start_generator.c__pyx_array___len____pyx_MemviewEnum___repr____pyx_memoryview___len____pyx_memoryview_refcount_objects_in_slice__pyx_tp_new_5numpy_6random_10_generator_Generator__pyx_vtabptr_5numpy_6random_10_generator_Generator__pyx_empty_tuple__pyx_tp_traverse_5numpy_6random_10_generator_Generator__pyx_tp_clear_5numpy_6random_10_generator_Generator__pyx_getprop_5numpy_6random_10_generator_9Generator_bit_generator__pyx_tp_new_Enum__pyx_tp_traverse_Enum__pyx_tp_clear_Enum__pyx_tp_clear_memoryview__pyx_getprop___pyx_memoryview_base__pyx_getprop___pyx_memoryviewslice_base__Pyx_ErrRestoreInState__Pyx_PyCFunction_FastCall__Pyx__ExceptionSave__pyx_typeinfo_cmp__Pyx_PyNumber_IntOrLongWrongResultType__Pyx_PyObject_Call__Pyx_PyFunction_FastCallNoKw__pyx_pyframe_localsplus_offset__Pyx_PyObject_GetAttrStr__pyx_sq_item_array__Pyx_InitCachedConstants__pyx_kp_u_Invalid_bit_generator_The_bit_ge__pyx_tuple___pyx_kp_u_Providing_a_dtype_with_a_non_nat__pyx_tuple__8__pyx_int_4294967296__pyx_int_0__pyx_tuple__9__pyx_kp_u_a_must_an_array_or_an_integer__pyx_tuple__10__pyx_kp_u_a_must_be_a_positive_integer_unl__pyx_tuple__11__pyx_kp_u_a_cannot_be_empty_unless_no_samp__pyx_tuple__12__pyx_kp_u_p_must_be_1_dimensional__pyx_tuple__13__pyx_kp_u_a_and_p_must_have_same_size__pyx_tuple__14__pyx_kp_u_probabilities_contain_NaN__pyx_tuple__15__pyx_kp_u_probabilities_are_not_non_negati__pyx_tuple__16__pyx_kp_u_probabilities_do_not_sum_to_1__pyx_tuple__17__pyx_kp_u_Cannot_take_a_larger_sample_than__pyx_tuple__18__pyx_kp_u_negative_dimensions_are_not_allo__pyx_tuple__19__pyx_kp_u_Fewer_non_zero_entries_in_p_than__pyx_tuple__20__pyx_tuple__21__pyx_tuple__22__pyx_kp_u_Range_exceeds_valid_bounds__pyx_tuple__23__pyx_kp_u_left_mode__pyx_tuple__26__pyx_kp_u_mode_right__pyx_tuple__27__pyx_kp_u_left_right__pyx_tuple__28__pyx_kp_u_ngood_nbad_nsample__pyx_tuple__29__pyx_kp_u_method_must_be_one_of_eigh_svd_c__pyx_tuple__30__pyx_kp_u_mean_must_be_1_dimensional__pyx_tuple__31__pyx_kp_u_cov_must_be_2_dimensional_and_sq__pyx_tuple__32__pyx_kp_u_mean_and_cov_must_have_same_leng__pyx_tuple__33__pyx_slice__34__pyx_kp_u_check_valid_must_equal_warn_rais__pyx_tuple__35__pyx_builtin_RuntimeWarning__pyx_kp_u_covariance_is_not_positive_semid__pyx_tuple__36__pyx_tuple__37__pyx_kp_u_sum_pvals_1_1_0__pyx_tuple__38__pyx_kp_u_method_must_be_count_or_marginal__pyx_tuple__39__pyx_kp_u_nsample_must_be_an_integer__pyx_tuple__40__pyx_kp_u_nsample_must_be_nonnegative__pyx_tuple__41__pyx_kp_u_When_method_is_marginals_sum_col__pyx_tuple__42__pyx_kp_u_nsample_sum_colors__pyx_tuple__43__pyx_kp_u_alpha_0__pyx_tuple__44__pyx_tuple__45__pyx_kp_u_Axis_argument_is_only_supported__pyx_tuple__46__pyx_kp_u_numpy_core_multiarray_failed_to__pyx_tuple__47__pyx_kp_u_numpy_core_umath_failed_to_impor__pyx_kp_s_Empty_shape_tuple_for_cython_arr__pyx_tuple__49__pyx_kp_s_itemsize_0_for_cython_array__pyx_tuple__50__pyx_kp_s_unable_to_allocate_shape_and_str__pyx_tuple__51__pyx_kp_s_unable_to_allocate_array_data__pyx_tuple__52__pyx_kp_s_Can_only_create_a_buffer_that_is__pyx_tuple__53__pyx_kp_s_no_default___reduce___due_to_non__pyx_tuple__54__pyx_tuple__55__pyx_kp_s_Cannot_assign_to_read_only_memor__pyx_tuple__56__pyx_kp_s_Unable_to_convert_item_to_object__pyx_tuple__57__pyx_kp_s_Cannot_create_writable_memory_vi__pyx_tuple__58__pyx_kp_s_Buffer_view_does_not_expose_stri__pyx_tuple__59__pyx_tuple__60__pyx_int_neg_1__pyx_tuple__61__pyx_tuple__62__pyx_kp_s_Indirect_dimensions_not_supporte__pyx_tuple__63__pyx_tuple__64__pyx_tuple__65__pyx_n_s_seed__pyx_empty_bytes__pyx_n_s_default_rng__pyx_kp_s_generator_pyx__pyx_kp_s_strided_and_direct_or_indirect__pyx_tuple__68__pyx_kp_s_strided_and_direct__pyx_tuple__69__pyx_kp_s_strided_and_indirect__pyx_tuple__70__pyx_kp_s_contiguous_and_direct__pyx_tuple__71__pyx_kp_s_contiguous_and_indirect__pyx_tuple__72__pyx_n_s_pyx_result__pyx_n_s_pyx_PickleError__pyx_n_s_pyx_state__pyx_n_s_pyx_checksum__pyx_n_s_pyx_type__pyx_n_s_pyx_unpickle_Enum__pyx_kp_s_stringsource__pyx_tp_dealloc_Enum__pyx_tp_dealloc_5numpy_6random_10_generator_Generator__pyx_tp_dealloc_memoryview__pyx_memoryview_thread_locks_used__pyx_memoryview_thread_locks_copy_strided_to_strided__pyx_memoryview__slice_assign_scalar__Pyx__GetException__pyx_f_5numpy_6random_10_generator_9Generator__shuffle_raw__Pyx_PyObject_LookupSpecial__pyx_f_5numpy_6random_10_generator_9Generator__shuffle_int__Pyx_SetVtable__pyx_n_s_pyx_vtable__Pyx_setup_reduce_is_named__pyx_n_s_name__Pyx_ImportType__Pyx_copy_spec_to_module__Pyx_ImportVoidPtr__Pyx_ImportFunction__Pyx_Import__pyx_m__pyx_pymod_createmain_interpreter_id.24801__pyx_setprop_5numpy_6random_10_generator_9Generator__bit_generator__pyx_tp_traverse_memoryview__Pyx__ExceptionReset.isra.11__Pyx_IsSubtype__Pyx_PyInt_NeObjC.isra.31__Pyx_PyUnicode_Equals__pyx_getprop_5numpy_6random_10_generator_9Generator__bit_generator__Pyx_BufFmt_TypeCharToAlignment.constprop.77__Pyx_GetItemInt_Fast.constprop.80__Pyx_ParseOptionalKeywords.constprop.81__Pyx_Raise.constprop.82__pyx_fatalerror.constprop.84__Pyx_PyObject_GetSlice.isra.36.constprop.86__Pyx_PyFunction_FastCallDict.constprop.88__pyx_sq_item_memoryview__Pyx_PyErr_GivenExceptionMatchesTuple__Pyx_TypeTest.isra.16__pyx_tp_traverse__memoryviewslice__pyx_tp_clear__memoryviewslice__Pyx_GetBuiltinName__pyx_b__Pyx_ImportFrom__Pyx_PyErr_GivenExceptionMatches.part.43__pyx_tp_dealloc__memoryviewslice__Pyx_BufFmt_RaiseExpected__Pyx_BufFmt_ProcessTypeChunk__Pyx_BufFmt_CheckString__Pyx_PyInt_As_Py_intptr_t.part.21__Pyx__PyObject_CallOneArg__pyx_tp_dealloc_array__Pyx_PyObject_GetIndex__Pyx_PyInt_As_size_t__Pyx_PyInt_As_int64_t.part.22__Pyx_PyInt_As_long__Pyx_PyInt_As_int__Pyx_PyInt_As_Py_intptr_t__Pyx_PyObject_CallNoArg__Pyx_AddTraceback__pyx_code_cache__pyx_d__pyx_cython_runtime__pyx_dict_version.25690__pyx_dict_cached_value.25691__pyx_n_s_cline_in_traceback__pyx_pw___pyx_memoryviewslice_3__setstate_cython____pyx_builtin_TypeError__pyx_pw___pyx_memoryviewslice_1__reduce_cython____pyx_getprop___pyx_memoryview_size__pyx_int_1__pyx_getprop___pyx_memoryview_nbytes__pyx_n_s_size__pyx_getprop___pyx_memoryview_itemsize__pyx_getprop___pyx_memoryview_ndim__pyx_getprop___pyx_memoryview_suboffsets__pyx_getprop___pyx_memoryview_shape__pyx_memoryview_fromslice__pyx_memoryviewslice_type__pyx_n_s_base__pyx_memoryview_copy_object_from_slice__pyx_pw___pyx_memoryview_3__setstate_cython____pyx_pw___pyx_memoryview_1__reduce_cython____pyx_memoryview_new__pyx_memoryview_type__pyx_memoryview___str____pyx_n_s_class__pyx_kp_s_MemoryView_of_r_object__pyx_memoryview_setitem_indexed__pyx_MemviewEnum___init____pyx_n_s_name_2__pyx_pyargnames.22301__pyx_getprop___pyx_array_memview__pyx_pw___pyx_array_3__setstate_cython____pyx_pw___pyx_array_1__reduce_cython____pyx_array___pyx_pf_15View_dot_MemoryView_5array_8__getattr____pyx_n_s_memview__pyx_array___getattr____pyx_tp_getattro_array__pyx_mp_ass_subscript_array__pyx_array___getitem____pyx_array_get_memview__pyx_pw_5numpy_6random_10_generator_9Generator_83logseries__pyx_float_0_0__pyx_kp_u__5__pyx_n_u_p__pyx_f_5numpy_6random_7_common_disc__pyx_n_s_p__pyx_pyargnames.19840__pyx_pw_5numpy_6random_10_generator_9Generator_79geometric__pyx_pyargnames.19585__pyx_pw_5numpy_6random_10_generator_9Generator_77zipf__pyx_n_u_a__pyx_n_s_a__pyx_pyargnames.19530__pyx_pw_5numpy_6random_10_generator_9Generator_75poisson__pyx_float_1_0__pyx_n_u_lam__pyx_pyargnames.19473__pyx_n_s_lam__pyx_pw_5numpy_6random_10_generator_9Generator_73negative_binomial__pyx_n_u_n__pyx_n_s_n__pyx_pyargnames.19415__pyx_pw_5numpy_6random_10_generator_9Generator_67wald__pyx_n_u_scale__pyx_n_u_mean__pyx_f_5numpy_6random_7_common_cont__pyx_n_s_mean__pyx_n_s_scale__pyx_pyargnames.18871__pyx_pw_5numpy_6random_10_generator_9Generator_65rayleigh__pyx_pyargnames.18813__pyx_pw_5numpy_6random_10_generator_9Generator_63lognormal__pyx_n_u_sigma__pyx_n_s_sigma__pyx_pyargnames.18751__pyx_pw_5numpy_6random_10_generator_9Generator_61logistic__pyx_n_u_loc__pyx_n_s_loc__pyx_pyargnames.18688__pyx_pw_5numpy_6random_10_generator_9Generator_59gumbel__pyx_pyargnames.18625__pyx_pw_5numpy_6random_10_generator_9Generator_57laplace__pyx_pyargnames.18562__pyx_pw_5numpy_6random_10_generator_9Generator_55power__pyx_pyargnames.18506__pyx_pw_5numpy_6random_10_generator_9Generator_53weibull__pyx_pyargnames.18451__pyx_pw_5numpy_6random_10_generator_9Generator_51pareto__pyx_pyargnames.18396__pyx_pw_5numpy_6random_10_generator_9Generator_49vonmises__pyx_n_u_kappa__pyx_n_u_mu__pyx_n_s_mu__pyx_n_s_kappa__pyx_pyargnames.18338__pyx_pw_5numpy_6random_10_generator_9Generator_47standard_t__pyx_n_u_df__pyx_n_s_df__pyx_pyargnames.18282__pyx_pw_5numpy_6random_10_generator_9Generator_45standard_cauchy__pyx_pyargnames.18230__pyx_pw_5numpy_6random_10_generator_9Generator_43noncentral_chisquare__pyx_n_u_nonc__pyx_n_s_nonc__pyx_pyargnames.18173__pyx_pw_5numpy_6random_10_generator_9Generator_41chisquare__pyx_pyargnames.18117__pyx_pw_5numpy_6random_10_generator_9Generator_39noncentral_f__pyx_n_u_dfden__pyx_n_u_dfnum__pyx_n_s_dfnum__pyx_pyargnames.18056__pyx_n_s_dfden__pyx_pw_5numpy_6random_10_generator_9Generator_37f__pyx_pyargnames.17996__pyx_pw_5numpy_6random_10_generator_9Generator_35gamma__pyx_n_u_shape__pyx_n_s_shape__pyx_pyargnames.17935__pyx_pw_5numpy_6random_10_generator_9Generator_31normal__pyx_pyargnames.17786__pyx_pw_5numpy_6random_10_generator_9Generator_17exponential__pyx_pyargnames.16467__pyx_pw_5numpy_6random_10_generator_9Generator_15beta__pyx_n_u_b__pyx_n_s_b__pyx_pyargnames.16409__pyx_pw_5numpy_6random_10_generator_9Generator_11__reduce____pyx_n_s_generator_ctor__pyx_n_s_pickle__pyx_n_s_bit_generator__pyx_n_s_state__pyx_n_u_bit_generator__pyx_pw_5numpy_6random_10_generator_9Generator_9__setstate____pyx_pw_5numpy_6random_10_generator_9Generator_7__getstate____pyx_pw_5numpy_6random_10_generator_9Generator_5__str____pyx_kp_u__2__pyx_kp_u__3__pyx_getprop___pyx_memoryview_strides__pyx_builtin_ValueError__Pyx_PyObject_Call2Args__pyx_memoryview_getbuffer__pyx_memoryview_get_slice_from_memoryview__pyx_memoryview___pyx_pf_15View_dot_MemoryView_10memoryview_18is_f_contig__pyx_memoryview_is_f_contig__pyx_memoryview___pyx_pf_15View_dot_MemoryView_10memoryview_16is_c_contig__pyx_memoryview_is_c_contig__pyx_memoryview___repr____pyx_builtin_id__pyx_kp_s_MemoryView_of_r_at_0x_x__pyx_array_getbuffer__pyx_n_u_c__pyx_n_u_fortran__pyx_memoryview_setitem_slice_assign_scalar__Pyx_PyObject_CallOneArg__pyx_pw_5numpy_6random_10_generator_9Generator_3__repr____pyx_n_s_str__pyx_kp_u_at_0x_X__pyx_n_s_format__pyx_pw_5numpy_6random_10_generator_9Generator_1__init____pyx_n_s_capsule__pyx_n_s_lock__pyx_pyargnames.16097__Pyx__GetModuleGlobalName__Pyx_IternextUnpackEndCheck__pyx_pw___pyx_MemviewEnum_1__reduce_cython____pyx_n_s_dict__pyx_dict_version.22362__pyx_dict_cached_value.22363__pyx_int_184977713__pyx_dict_version.22367__pyx_dict_cached_value.22368__Pyx_IterFinish__pyx_memoryview_is_slice__pyx_tp_new_memoryview__pyx_vtabptr_memoryview__pyx_n_s_obj__pyx_n_s_flags__pyx_n_s_dtype_is_object__pyx_pyargnames.22426__pyx_tp_new__memoryviewslice__pyx_vtabptr__memoryviewslice__Pyx_PyInt_As_int64_t__pyx_unpickle_Enum__set_state__pyx_n_s_update__pyx_pw_15View_dot_MemoryView_1__pyx_unpickle_Enum__pyx_n_s_PickleError__pyx_n_s_pickle_2__pyx_kp_s_Incompatible_checksums_s_vs_0xb0__pyx_MemviewEnum_type__pyx_n_s_new__pyx_pyargnames.24283__pyx_pw___pyx_MemviewEnum_3__setstate_cython____Pyx_PyErr_ExceptionMatchesInState.isra.44__Pyx_PyObject_GetAttrStrNoError__Pyx_setup_reduce__pyx_n_s_getstate__pyx_n_s_reduce_ex__pyx_n_s_reduce_2__pyx_n_s_reduce_cython__pyx_n_s_setstate__pyx_n_s_setstate_cython__pyx_pymod_exec__generator__pyx_empty_unicode__pyx_string_tab__pyx_float_0_1__pyx_float_1eneg_8__pyx_int_20__pyx_int_50__pyx_int_1000000000__pyx_n_s_main__pyx_n_s_range__pyx_n_s_ValueError__pyx_n_s_id__pyx_n_s_TypeError__pyx_n_s_OverflowError__pyx_builtin_OverflowError__pyx_n_s_RuntimeWarning__pyx_n_s_MemoryError__pyx_builtin_MemoryError__pyx_n_s_reversed__pyx_n_s_NotImplementedError__pyx_builtin_NotImplementedError__pyx_n_s_ImportError__pyx_builtin_ImportError__pyx_n_s_enumerate__pyx_n_s_Ellipsis__pyx_builtin_Ellipsis__pyx_n_s_IndexError__pyx_builtin_IndexError__pyx_type_5numpy_6random_10_generator_Generatorgenericindirect_contiguous__pyx_vtable_5numpy_6random_10_generator_Generator__pyx_n_s_Generator__pyx_vtable_array__pyx_type___pyx_array__pyx_ptype_5numpy_6random_10_generator_Generator__pyx_vtabptr_array__pyx_type___pyx_MemviewEnum__pyx_array_type__pyx_vtable_memoryview__pyx_memoryview_get_item_pointer__pyx_type___pyx_memoryview__pyx_memoryview_setitem_slice_assignment__pyx_memoryview_convert_item_to_object__pyx_memoryview_assign_item_from_object__pyx_vtable__memoryviewslice__pyx_type___pyx_memoryviewslice__pyx_memoryviewslice_convert_item_to_object__pyx_memoryviewslice_assign_item_from_object__pyx_ptype_5numpy_dtype__pyx_ptype_5numpy_broadcast__pyx_ptype_5numpy_ndarray__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_float_fill__pyx_f_5numpy_6random_7_common_validate_output_shape__pyx_f_5numpy_6random_7_common_cont_f__pyx_f_5numpy_6random_7_common_cont_broadcast_3__pyx_f_5numpy_6random_7_common_discrete_broadcast_iii__pyx_n_s_operator__pyx_n_s_warnings__pyx_n_s_numpy__pyx_n_s_np__pyx_n_s_normalize_axis_index__pyx_n_s_numpy_core_multiarray__pyx_n_s_PCG64__pyx_n_s_pcg64PyArray_API__pyx_n_s_poisson_lam_max__pyx_dict_version.24846__pyx_dict_cached_value.24847__pyx_n_s_float64__pyx_k__4__pyx_dict_version.24849__pyx_dict_cached_value.24850__pyx_k__6__pyx_dict_version.24852__pyx_dict_cached_value.24853__pyx_n_s_int64__pyx_k__7__pyx_dict_version.24855__pyx_dict_cached_value.24856__pyx_k__24__pyx_dict_version.24858__pyx_dict_cached_value.24859__pyx_n_s_numpy_random__generator__pyx_mdef_5numpy_6random_10_generator_1default_rng__pyx_k__25__pyx_kp_u_random_size_None_dtype_np_float__pyx_kp_u_Generator_random_line_155__pyx_kp_u_standard_exponential_size_None__pyx_kp_u_Generator_standard_exponential_l__pyx_kp_u_integers_low_high_None_size_Non__pyx_kp_u_Generator_integers_line_362__pyx_kp_u_bytes_length_Return_random_byte__pyx_kp_u_Generator_bytes_line_489__pyx_kp_u_choice_a_size_None_replace_True__pyx_kp_u_Generator_choice_line_518__pyx_kp_u_uniform_low_0_0_high_1_0_size_N__pyx_kp_u_Generator_uniform_line_762__pyx_kp_u_standard_normal_size_None_dtype__pyx_kp_u_Generator_standard_normal_line_8__pyx_kp_u_normal_loc_0_0_scale_1_0_size_N__pyx_kp_u_Generator_normal_line_939__pyx_kp_u_standard_gamma_shape_size_None__pyx_kp_u_Generator_standard_gamma_line_10__pyx_kp_u_gamma_shape_scale_1_0_size_None__pyx_kp_u_Generator_gamma_line_1132__pyx_kp_u_f_dfnum_dfden_size_None_Draw_sa__pyx_kp_u_Generator_f_line_1210__pyx_kp_u_noncentral_f_dfnum_dfden_nonc_s__pyx_kp_u_Generator_noncentral_f_line_1298__pyx_kp_u_chisquare_df_size_None_Draw_sam__pyx_kp_u_Generator_chisquare_line_1376__pyx_kp_u_noncentral_chisquare_df_nonc_si__pyx_kp_u_Generator_noncentral_chisquare_l__pyx_kp_u_standard_cauchy_size_None_Draw__pyx_kp_u_Generator_standard_cauchy_line_1__pyx_kp_u_standard_t_df_size_None_Draw_sa__pyx_kp_u_Generator_standard_t_line_1589__pyx_kp_u_vonmises_mu_kappa_size_None_Dra__pyx_kp_u_Generator_vonmises_line_1683__pyx_kp_u_pareto_a_size_None_Draw_samples__pyx_kp_u_Generator_pareto_line_1766__pyx_kp_u_weibull_a_size_None_Draw_sample__pyx_kp_u_Generator_weibull_line_1864__pyx_kp_u_power_a_size_None_Draws_samples__pyx_kp_u_Generator_power_line_1963__pyx_kp_u_laplace_loc_0_0_scale_1_0_size__pyx_kp_u_Generator_laplace_line_2064__pyx_kp_u_gumbel_loc_0_0_scale_1_0_size_N__pyx_kp_u_Generator_gumbel_line_2149__pyx_kp_u_logistic_loc_0_0_scale_1_0_size__pyx_kp_u_Generator_logistic_line_2268__pyx_kp_u_lognormal_mean_0_0_sigma_1_0_si__pyx_kp_u_Generator_lognormal_line_2348__pyx_kp_u_rayleigh_scale_1_0_size_None_Dr__pyx_kp_u_Generator_rayleigh_line_2460__pyx_kp_u_wald_mean_scale_size_None_Draw__pyx_kp_u_Generator_wald_line_2529__pyx_kp_u_triangular_left_mode_right_size__pyx_kp_u_Generator_triangular_line_2597__pyx_kp_u_binomial_n_p_size_None_Draw_sam__pyx_kp_u_Generator_binomial_line_2697__pyx_kp_u_negative_binomial_n_p_size_None__pyx_kp_u_Generator_negative_binomial_line__pyx_kp_u_poisson_lam_1_0_size_None_Draw__pyx_kp_u_Generator_poisson_line_2917__pyx_kp_u_zipf_a_size_None_Draw_samples_f__pyx_kp_u_Generator_zipf_line_2989__pyx_kp_u_geometric_p_size_None_Draw_samp__pyx_kp_u_Generator_geometric_line_3069__pyx_kp_u_hypergeometric_ngood_nbad_nsamp__pyx_kp_u_Generator_hypergeometric_line_31__pyx_kp_u_logseries_p_size_None_Draw_samp__pyx_kp_u_Generator_logseries_line_3263__pyx_kp_u_multivariate_normal_mean_cov_si__pyx_kp_u_Generator_multivariate_normal_li__pyx_kp_u_multinomial_n_pvals_size_None_D__pyx_kp_u_Generator_multinomial_line_3548__pyx_kp_u_multivariate_hypergeometric_col__pyx_kp_u_Generator_multivariate_hypergeom__pyx_kp_u_dirichlet_alpha_size_None_Draw__pyx_kp_u_Generator_dirichlet_line_3915__pyx_kp_u_shuffle_x_axis_0_Modify_a_seque__pyx_kp_u_Generator_shuffle_line_4114__pyx_kp_u_permutation_x_axis_0_Randomly_p__pyx_kp_u_Generator_permutation_line_4256__pyx_n_s_test__pyx_n_s_pyx_getbuffer__pyx_t_4.24830__pyx_n_s_View_MemoryView__pyx_mdef_15View_dot_MemoryView_1__pyx_unpickle_Enum__pyx_pw_5numpy_6random_10_generator_9Generator_21integers__pyx_dict_version.16678__pyx_dict_cached_value.16679__pyx_n_s_int32__pyx_n_s_low__pyx_dict_version.16685__pyx_dict_cached_value.16686__pyx_dict_version.16748__pyx_dict_cached_value.16749__pyx_n_s_array__pyx_dict_version.16691__pyx_dict_cached_value.16692__pyx_n_s_int16__pyx_dict_version.16741__pyx_dict_cached_value.16742__pyx_n_s_compat__pyx_n_s_long__pyx_n_s_high__pyx_n_s_dtype__pyx_n_s_endpoint__pyx_dict_version.16697__pyx_dict_cached_value.16698__pyx_n_s_int8__pyx_pyargnames.16621__pyx_dict_version.16703__pyx_dict_cached_value.16704__pyx_n_s_uint64__pyx_dict_version.16709__pyx_dict_cached_value.16710__pyx_n_s_uint32__pyx_dict_version.16715__pyx_dict_cached_value.16716__pyx_n_s_uint16__pyx_dict_version.16721__pyx_dict_cached_value.16722__pyx_n_s_uint8__pyx_dict_version.16727__pyx_dict_cached_value.16728__pyx_n_s_bool__pyx_n_s_isnative__pyx_kp_u_Unsupported_dtype_r_for_integers__pyx_memoryview_err__pyx_memslice_transpose__pyx_getprop___pyx_memoryview_T__pyx_n_s_struct__pyx_n_s_pack_unellipsify__pyx_kp_s_Cannot_index_with_type_s__pyx_mp_ass_subscript_memoryview__pyx_pw_5numpy_6random_10_generator_1default_rng__pyx_n_u_capsule__pyx_dict_version.21666__pyx_dict_cached_value.21667__pyx_pyargnames.21628__pyx_tp_new_array__pyx_n_s_c__pyx_kp_s_Invalid_shape_in_axis_d_d__pyx_n_s_encode__pyx_n_s_ASCII__pyx_n_s_fortran__pyx_n_b_O__pyx_n_s_itemsize__pyx_n_s_mode__pyx_n_s_allocate_buffer__pyx_pyargnames.21930__pyx_kp_s_Invalid_mode_expected_c_or_fortr__pyx_pw_5numpy_6random_10_generator_9Generator_85multivariate_normal__pyx_n_u_svd__pyx_n_u_warn__pyx_n_u_eigh__pyx_dict_version.19973__pyx_dict_cached_value.19974__pyx_dict_version.19984__pyx_dict_cached_value.19985__pyx_dict_version.19996__pyx_dict_cached_value.19997__pyx_n_s_integer__pyx_n_u_cholesky__pyx_n_s_standard_normal__pyx_n_s_reshape__pyx_n_s_astype__pyx_dict_version.20041__pyx_dict_cached_value.20042__pyx_n_s_double__pyx_n_s_svd__pyx_n_s_numpy_dual__pyx_n_u_ignore__pyx_n_u_raise__pyx_dict_version.20102__pyx_dict_cached_value.20103__pyx_n_s_allclose__pyx_dict_version.20105__pyx_dict_cached_value.20106__pyx_n_s_dot__pyx_n_s_T__pyx_n_s_rtol__pyx_n_s_atol__pyx_n_s_cov__pyx_n_s_check_valid__pyx_n_s_tol__pyx_pyargnames.19899__pyx_n_s_eigh__pyx_n_s_cholesky__pyx_dict_version.20146__pyx_dict_cached_value.20147__pyx_n_s_sqrt__pyx_dict_version.20126__pyx_dict_cached_value.20127__pyx_n_s_any__pyx_dict_version.20157__pyx_dict_cached_value.20158__pyx_dict_version.20138__pyx_dict_cached_value.20139__pyx_n_s_warn__pyx_pw_5numpy_6random_10_generator_9Generator_89multivariate_hypergeometric__pyx_n_u_marginals__pyx_n_u_count__pyx_dict_version.20556__pyx_dict_cached_value.20557__pyx_n_s_index__pyx_dict_version.20590__pyx_dict_cached_value.20591__pyx_n_s_asarray__pyx_n_s_ndim__pyx_dict_version.20608__pyx_dict_cached_value.20609__pyx_n_s_issubdtype__pyx_dict_version.20611__pyx_dict_cached_value.20612__pyx_dict_version.20630__pyx_dict_cached_value.20631__pyx_n_s_method__pyx_dict_version.20674__pyx_dict_cached_value.20675__pyx_n_s_ascontiguousarray__pyx_dict_version.20677__pyx_dict_cached_value.20678__pyx_dict_version.20699__pyx_dict_cached_value.20700__pyx_n_s_isscalar__pyx_dict_version.20711__pyx_dict_cached_value.20712__pyx_n_s_zeros__pyx_dict_version.20714__pyx_dict_cached_value.20715__pyx_n_s_exit__pyx_n_s_enter__pyx_n_s_nsample__pyx_n_s_colors__pyx_kp_u_colors_must_be_a_one_dimensional__pyx_kp_u_nsample_must_not_exceed_d__pyx_pyargnames.20488__pyx_kp_u_sum_colors_must_not_exceed_the_m__pyx_kp_u_When_method_is_count_sum_colors__pyx_kp_u_Insufficient_memory_for_multivar__pyx_memoryview_err_dim__pyx_memoryview_copy_contents__pyx_kp_s_got_differing_extents_in_dimensi__pyx_kp_s_Out_of_bounds_on_buffer_access_a__pyx_memview_slice__pyx_n_s_start__pyx_n_s_stop__pyx_n_s_step__pyx_memoryview___getitem____pyx_pw_5numpy_6random_10_generator_9Generator_81hypergeometric__pyx_dict_version.19732__pyx_dict_cached_value.19733__pyx_dict_version.19744__pyx_dict_cached_value.19745__pyx_dict_version.19757__pyx_dict_cached_value.19758__pyx_dict_version.19760__pyx_dict_cached_value.19761__pyx_n_s_less__pyx_dict_version.19763__pyx_dict_cached_value.19764__pyx_n_s_add__pyx_n_u_ngood__pyx_n_u_nsample__pyx_n_u_nbad__pyx_n_s_ngood__pyx_pyargnames.19642__pyx_kp_u_both_ngood_and_nbad_must_be_less__pyx_n_s_nbad__pyx_pw_5numpy_6random_10_generator_9Generator_69triangular__pyx_dict_version.19018__pyx_dict_cached_value.19019__pyx_dict_version.19021__pyx_dict_cached_value.19022__pyx_n_s_greater__pyx_dict_version.19044__pyx_dict_cached_value.19045__pyx_dict_version.19047__pyx_dict_cached_value.19048__pyx_dict_version.19070__pyx_dict_cached_value.19071__pyx_dict_version.19073__pyx_dict_cached_value.19074__pyx_n_s_equal__pyx_pyargnames.18931__pyx_n_s_left__pyx_n_s_right__pyx_pw_5numpy_6random_10_generator_9Generator_29standard_normal__pyx_dict_version.17744__pyx_dict_cached_value.17745__pyx_dict_version.17753__pyx_dict_cached_value.17754__pyx_n_s_float32__pyx_n_s_out__pyx_kp_u_Unsupported_dtype_r_for_standard_2__pyx_pyargnames.17704__pyx_pw_5numpy_6random_10_generator_9Generator_13random__pyx_dict_version.16367__pyx_dict_cached_value.16368__pyx_dict_version.16376__pyx_dict_cached_value.16377__pyx_kp_u_Unsupported_dtype_r_for_random__pyx_pyargnames.16327__pyx_pw_5numpy_6random_10_generator_9Generator_91dirichlet__pyx_dict_version.20869__pyx_dict_cached_value.20870__pyx_dict_version.20872__pyx_dict_cached_value.20873__pyx_n_s_less_equal__pyx_dict_version.20897__pyx_dict_cached_value.20898__pyx_dict_version.20939__pyx_dict_cached_value.20940__pyx_dict_version.20942__pyx_dict_cached_value.20943__pyx_n_s_max__pyx_dict_version.20969__pyx_dict_cached_value.20970__pyx_n_s_empty_like__pyx_n_s_alpha__pyx_pyargnames.20797__pyx_memoryview_copy_new_contig__pyx_memoryview___pyx_pf_15View_dot_MemoryView_10memoryview_22copy_fortran__pyx_memoryview_copy_fortran__pyx_memoryview___pyx_pf_15View_dot_MemoryView_10memoryview_20copy__pyx_memoryview_copy__pyx_pw_5numpy_6random_10_generator_9Generator_93shuffle__pyx_dict_version.21119__pyx_dict_cached_value.21120__pyx_dict_version.21121__pyx_dict_cached_value.21122__pyx_n_s_x__pyx_n_s_strides__pyx_dict_version.21153__pyx_dict_cached_value.21154__pyx_n_s_empty__pyx_dict_version.21156__pyx_dict_cached_value.21157__pyx_dict_version.21213__pyx_dict_cached_value.21214__pyx_n_s_swapaxes__pyx_dict_version.21229__pyx_dict_cached_value.21230__pyx_pyargnames.21066__pyx_n_s_axis__pyx_pw_5numpy_6random_10_generator_9Generator_87multinomial__pyx_n_u_pvals__pyx_dict_version.20315__pyx_dict_cached_value.20316__pyx_dict_version.20318__pyx_dict_cached_value.20319__pyx_dict_version.20329__pyx_dict_cached_value.20330__pyx_dict_version.20332__pyx_dict_cached_value.20333__pyx_dict_version.20370__pyx_dict_cached_value.20371__pyx_dict_version.20410__pyx_dict_cached_value.20411__pyx_dict_version.20413__pyx_dict_cached_value.20414__pyx_n_s_pvals__pyx_pyargnames.20232__pyx_pw_5numpy_6random_10_generator_9Generator_71binomial__pyx_dict_version.19336__pyx_dict_cached_value.19337__pyx_dict_version.19339__pyx_dict_cached_value.19340__pyx_dict_version.19213__pyx_dict_cached_value.19214__pyx_dict_version.19216__pyx_dict_cached_value.19217__pyx_dict_version.19234__pyx_dict_cached_value.19235__pyx_dict_version.19237__pyx_dict_cached_value.19238__pyx_pyargnames.19134__pyx_pw_5numpy_6random_10_generator_9Generator_33standard_gamma__pyx_dict_version.17893__pyx_dict_cached_value.17894__pyx_dict_version.17902__pyx_dict_cached_value.17903__pyx_kp_u_Unsupported_dtype_r_for_standard_3__pyx_pyargnames.17850__pyx_pw_5numpy_6random_10_generator_9Generator_23bytes__pyx_n_s_integers__pyx_dict_version.16819__pyx_dict_cached_value.16820__pyx_kp_u_u4__pyx_n_s_tobytes__pyx_pw_5numpy_6random_10_generator_9Generator_27uniform__pyx_dict_version.17629__pyx_dict_cached_value.17630__pyx_n_s_subtract__pyx_dict_version.17645__pyx_dict_cached_value.17646__pyx_n_s_all__pyx_dict_version.17648__pyx_dict_cached_value.17649__pyx_n_s_isfinite__pyx_dict_version.17611__pyx_dict_cached_value.17612__pyx_pyargnames.17542__pyx_n_s_unpack__pyx_n_s_error__pyx_pw_5numpy_6random_10_generator_9Generator_95permutation__pyx_dict_version.21444__pyx_dict_cached_value.21445__pyx_dict_version.21450__pyx_dict_cached_value.21451__pyx_n_s_arange__pyx_n_s_shuffle__pyx_dict_version.21469__pyx_dict_cached_value.21470__pyx_dict_version.21478__pyx_dict_cached_value.21479__pyx_dict_version.21532__pyx_dict_cached_value.21533__pyx_dict_version.21536__pyx_dict_cached_value.21537__pyx_n_s_intp__pyx_dict_version.21497__pyx_dict_cached_value.21498__pyx_n_s_may_share_memory__pyx_dict_version.21512__pyx_dict_cached_value.21513__pyx_pyargnames.21404__pyx_pw_5numpy_6random_10_generator_9Generator_19standard_exponential__pyx_n_u_zig__pyx_dict_version.16571__pyx_dict_cached_value.16572__pyx_dict_version.16583__pyx_dict_cached_value.16584__pyx_kp_u_Unsupported_dtype_r_for_standard__pyx_pyargnames.16526__pyx_pf_5numpy_6random_10_generator_9Generator_24choice__pyx_dict_version.16972__pyx_dict_cached_value.16973__pyx_n_s_copy__pyx_dict_version.16984__pyx_dict_cached_value.16985__pyx_n_s_item__pyx_dict_version.17022__pyx_dict_cached_value.17023__pyx_n_s_prod__pyx_dict_version.17038__pyx_dict_cached_value.17039__pyx_dict_version.17050__pyx_dict_cached_value.17051__pyx_dict_version.17053__pyx_dict_cached_value.17054__pyx_n_s_finfo__pyx_dict_version.17056__pyx_dict_cached_value.17057__pyx_n_s_eps__pyx_dict_version.17074__pyx_dict_cached_value.17075__pyx_dict_version.17077__pyx_dict_cached_value.17078__pyx_n_s_floating__pyx_dict_version.17096__pyx_dict_cached_value.17097__pyx_dict_version.17099__pyx_dict_cached_value.17100__pyx_dict_version.17139__pyx_dict_cached_value.17140__pyx_n_s_isnan__pyx_dict_version.17151__pyx_dict_cached_value.17152__pyx_n_s_logical_or__pyx_n_s_reduce__pyx_dict_version.17167__pyx_dict_cached_value.17168__pyx_dict_version.17170__pyx_dict_cached_value.17171__pyx_n_s_cumsum__pyx_n_s_random__pyx_n_s_searchsorted__pyx_n_u_right__pyx_n_s_side__pyx_dict_version.17200__pyx_dict_cached_value.17201__pyx_dict_version.17203__pyx_dict_cached_value.17204__pyx_dict_version.17451__pyx_dict_cached_value.17452__pyx_dict_version.17225__pyx_dict_cached_value.17226__pyx_n_s_count_nonzero__pyx_dict_version.17246__pyx_dict_cached_value.17247__pyx_dict_version.17249__pyx_dict_cached_value.17250__pyx_n_s_ravel__pyx_dict_version.17276__pyx_dict_cached_value.17277__pyx_dict_version.17297__pyx_dict_cached_value.17298__pyx_n_s_unique__pyx_n_s_return_index__pyx_n_s_sort__pyx_n_s_take__pyx_dict_version.17213__pyx_dict_cached_value.17214__pyx_dict_version.17462__pyx_dict_cached_value.17463__pyx_dict_version.17465__pyx_dict_cached_value.17466__pyx_dict_version.17376__pyx_dict_cached_value.17377__pyx_dict_version.17379__pyx_dict_cached_value.17380__pyx_dict_version.17386__pyx_dict_cached_value.17387__pyx_n_s_full__pyx_dict_version.17389__pyx_dict_cached_value.17390__Pyx_TypeInfo_nn_uint64_t__pyx_pw_5numpy_6random_10_generator_9Generator_25choice__pyx_n_s_replace__pyx_pyargnames.16873__pyx_moduledef__pyx_k_ASCII__pyx_k_Axis_argument_is_only_supported__pyx_k_Buffer_view_does_not_expose_stri__pyx_k_Can_only_create_a_buffer_that_is__pyx_k_Cannot_assign_to_read_only_memor__pyx_k_Cannot_create_writable_memory_vi__pyx_k_Cannot_index_with_type_s__pyx_k_Cannot_take_a_larger_sample_than__pyx_k_Ellipsis__pyx_k_Empty_shape_tuple_for_cython_arr__pyx_k_Fewer_non_zero_entries_in_p_than__pyx_k_Generator__pyx_k_Generator_binomial_line_2697__pyx_k_Generator_bytes_line_489__pyx_k_Generator_chisquare_line_1376__pyx_k_Generator_choice_line_518__pyx_k_Generator_dirichlet_line_3915__pyx_k_Generator_f_line_1210__pyx_k_Generator_gamma_line_1132__pyx_k_Generator_geometric_line_3069__pyx_k_Generator_gumbel_line_2149__pyx_k_Generator_hypergeometric_line_31__pyx_k_Generator_integers_line_362__pyx_k_Generator_laplace_line_2064__pyx_k_Generator_logistic_line_2268__pyx_k_Generator_lognormal_line_2348__pyx_k_Generator_logseries_line_3263__pyx_k_Generator_multinomial_line_3548__pyx_k_Generator_multivariate_hypergeom__pyx_k_Generator_multivariate_normal_li__pyx_k_Generator_negative_binomial_line__pyx_k_Generator_noncentral_chisquare_l__pyx_k_Generator_noncentral_f_line_1298__pyx_k_Generator_normal_line_939__pyx_k_Generator_pareto_line_1766__pyx_k_Generator_permutation_line_4256__pyx_k_Generator_poisson_line_2917__pyx_k_Generator_power_line_1963__pyx_k_Generator_random_line_155__pyx_k_Generator_rayleigh_line_2460__pyx_k_Generator_shuffle_line_4114__pyx_k_Generator_standard_cauchy_line_1__pyx_k_Generator_standard_exponential_l__pyx_k_Generator_standard_gamma_line_10__pyx_k_Generator_standard_normal_line_8__pyx_k_Generator_standard_t_line_1589__pyx_k_Generator_triangular_line_2597__pyx_k_Generator_uniform_line_762__pyx_k_Generator_vonmises_line_1683__pyx_k_Generator_wald_line_2529__pyx_k_Generator_weibull_line_1864__pyx_k_Generator_zipf_line_2989__pyx_k_ImportError__pyx_k_Incompatible_checksums_s_vs_0xb0__pyx_k_IndexError__pyx_k_Indirect_dimensions_not_supporte__pyx_k_Insufficient_memory_for_multivar__pyx_k_Invalid_bit_generator_The_bit_ge__pyx_k_Invalid_mode_expected_c_or_fortr__pyx_k_Invalid_shape_in_axis_d_d__pyx_k_MemoryError__pyx_k_MemoryView_of_r_at_0x_x__pyx_k_MemoryView_of_r_object__pyx_k_NotImplementedError__pyx_k_O__pyx_k_Out_of_bounds_on_buffer_access_a__pyx_k_OverflowError__pyx_k_PCG64__pyx_k_PickleError__pyx_k_Providing_a_dtype_with_a_non_nat__pyx_k_Range_exceeds_valid_bounds__pyx_k_RuntimeWarning__pyx_k_T__pyx_k_TypeError__pyx_k_Unable_to_convert_item_to_object__pyx_k_Unsupported_dtype_r_for_integers__pyx_k_Unsupported_dtype_r_for_random__pyx_k_Unsupported_dtype_r_for_standard__pyx_k_Unsupported_dtype_r_for_standard_2__pyx_k_Unsupported_dtype_r_for_standard_3__pyx_k_ValueError__pyx_k_View_MemoryView__pyx_k_When_method_is_count_sum_colors__pyx_k_When_method_is_marginals_sum_col__pyx_k__2__pyx_k__3__pyx_k__5__pyx_k_a__pyx_k_a_and_p_must_have_same_size__pyx_k_a_cannot_be_empty_unless_no_samp__pyx_k_a_must_an_array_or_an_integer__pyx_k_a_must_be_a_positive_integer_unl__pyx_k_add__pyx_k_all__pyx_k_allclose__pyx_k_allocate_buffer__pyx_k_alpha__pyx_k_alpha_0__pyx_k_any__pyx_k_arange__pyx_k_array__pyx_k_asarray__pyx_k_ascontiguousarray__pyx_k_astype__pyx_k_at_0x_X__pyx_k_atol__pyx_k_axis__pyx_k_b__pyx_k_base__pyx_k_binomial_n_p_size_None_Draw_sam__pyx_k_bit_generator__pyx_k_bool__pyx_k_both_ngood_and_nbad_must_be_less__pyx_k_bytes_length_Return_random_byte__pyx_k_c__pyx_k_capsule__pyx_k_check_valid__pyx_k_check_valid_must_equal_warn_rais__pyx_k_chisquare_df_size_None_Draw_sam__pyx_k_choice_a_size_None_replace_True__pyx_k_cholesky__pyx_k_class__pyx_k_cline_in_traceback__pyx_k_colors__pyx_k_colors_must_be_a_one_dimensional__pyx_k_compat__pyx_k_contiguous_and_direct__pyx_k_contiguous_and_indirect__pyx_k_copy__pyx_k_count__pyx_k_count_nonzero__pyx_k_cov__pyx_k_cov_must_be_2_dimensional_and_sq__pyx_k_covariance_is_not_positive_semid__pyx_k_cumsum__pyx_k_default_rng__pyx_k_df__pyx_k_dfden__pyx_k_dfnum__pyx_k_dict__pyx_k_dirichlet_alpha_size_None_Draw__pyx_k_dot__pyx_k_double__pyx_k_dtype__pyx_k_dtype_is_object__pyx_k_eigh__pyx_k_empty__pyx_k_empty_like__pyx_k_encode__pyx_k_endpoint__pyx_k_enter__pyx_k_enumerate__pyx_k_eps__pyx_k_equal__pyx_k_error__pyx_k_exit__pyx_k_f_dfnum_dfden_size_None_Draw_sa__pyx_k_finfo__pyx_k_flags__pyx_k_float32__pyx_k_float64__pyx_k_floating__pyx_k_format__pyx_k_fortran__pyx_k_full__pyx_k_gamma_shape_scale_1_0_size_None__pyx_k_generator_ctor__pyx_k_generator_pyx__pyx_k_geometric_p_size_None_Draw_samp__pyx_k_getstate__pyx_k_got_differing_extents_in_dimensi__pyx_k_greater__pyx_k_gumbel_loc_0_0_scale_1_0_size_N__pyx_k_high__pyx_k_hypergeometric_ngood_nbad_nsamp__pyx_k_id__pyx_k_ignore__pyx_n_s_imatmul__pyx_k_imatmul__pyx_n_s_import__pyx_k_import__pyx_k_index__pyx_k_int16__pyx_k_int32__pyx_k_int64__pyx_k_int8__pyx_k_integer__pyx_k_integers__pyx_k_integers_low_high_None_size_Non__pyx_k_intp__pyx_k_isfinite__pyx_k_isnan__pyx_k_isnative__pyx_k_isscalar__pyx_k_issubdtype__pyx_k_item__pyx_k_itemsize__pyx_k_itemsize_0_for_cython_array__pyx_k_kappa__pyx_k_lam__pyx_k_laplace_loc_0_0_scale_1_0_size__pyx_k_left__pyx_k_left_mode__pyx_k_left_right__pyx_k_less__pyx_k_less_equal__pyx_k_loc__pyx_k_lock__pyx_k_logical_or__pyx_k_logistic_loc_0_0_scale_1_0_size__pyx_k_lognormal_mean_0_0_sigma_1_0_si__pyx_k_logseries_p_size_None_Draw_samp__pyx_k_long__pyx_k_low__pyx_k_main__pyx_k_marginals__pyx_n_s_matmul__pyx_k_matmul__pyx_k_max__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_memview__pyx_k_method__pyx_k_method_must_be_count_or_marginal__pyx_k_method_must_be_one_of_eigh_svd_c__pyx_k_mode__pyx_k_mode_right__pyx_k_mu__pyx_k_multinomial_n_pvals_size_None_D__pyx_k_multivariate_hypergeometric_col__pyx_k_multivariate_normal_mean_cov_si__pyx_k_n__pyx_k_name__pyx_k_name_2__pyx_k_nbad__pyx_k_ndim__pyx_k_negative_binomial_n_p_size_None__pyx_k_negative_dimensions_are_not_allo__pyx_k_new__pyx_k_ngood__pyx_k_ngood_nbad_nsample__pyx_k_no_default___reduce___due_to_non__pyx_k_nonc__pyx_k_noncentral_chisquare_df_nonc_si__pyx_k_noncentral_f_dfnum_dfden_nonc_s__pyx_k_normal_loc_0_0_scale_1_0_size_N__pyx_k_normalize_axis_index__pyx_k_np__pyx_k_nsample__pyx_k_nsample_must_be_an_integer__pyx_k_nsample_must_be_nonnegative__pyx_k_nsample_must_not_exceed_d__pyx_k_nsample_sum_colors__pyx_k_numpy__pyx_k_numpy_core_multiarray__pyx_k_numpy_core_multiarray_failed_to__pyx_k_numpy_core_umath_failed_to_impor__pyx_k_numpy_dual__pyx_k_numpy_random__generator__pyx_k_obj__pyx_k_operator__pyx_k_out__pyx_k_p__pyx_k_p_must_be_1_dimensional__pyx_k_pack__pyx_k_pareto_a_size_None_Draw_samples__pyx_k_pcg64__pyx_k_permutation_x_axis_0_Randomly_p__pyx_k_pickle__pyx_k_pickle_2__pyx_k_poisson_lam_1_0_size_None_Draw__pyx_k_poisson_lam_max__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_pvals__pyx_k_pyx_PickleError__pyx_k_pyx_checksum__pyx_k_pyx_getbuffer__pyx_k_pyx_result__pyx_k_pyx_state__pyx_k_pyx_type__pyx_k_pyx_unpickle_Enum__pyx_k_pyx_vtable__pyx_k_raise__pyx_k_random__pyx_k_random_size_None_dtype_np_float__pyx_k_range__pyx_k_ravel__pyx_k_rayleigh_scale_1_0_size_None_Dr__pyx_k_reduce__pyx_k_reduce_2__pyx_k_reduce_cython__pyx_k_reduce_ex__pyx_k_replace__pyx_k_reshape__pyx_k_return_index__pyx_k_reversed__pyx_k_right__pyx_n_s_rmatmul__pyx_k_rmatmul__pyx_k_rtol__pyx_k_scale__pyx_k_searchsorted__pyx_k_seed__pyx_k_setstate__pyx_k_setstate_cython__pyx_k_shape__pyx_k_shuffle__pyx_k_shuffle_x_axis_0_Modify_a_seque__pyx_k_side__pyx_k_sigma__pyx_k_size__pyx_k_sort__pyx_k_sqrt__pyx_k_standard_cauchy_size_None_Draw__pyx_k_standard_exponential_size_None__pyx_k_standard_gamma_shape_size_None__pyx_k_standard_normal__pyx_k_standard_normal_size_None_dtype__pyx_k_standard_t_df_size_None_Draw_sa__pyx_k_start__pyx_k_state__pyx_k_step__pyx_k_stop__pyx_k_str__pyx_k_strided_and_direct__pyx_k_strided_and_direct_or_indirect__pyx_k_strided_and_indirect__pyx_k_strides__pyx_k_stringsource__pyx_k_struct__pyx_k_subtract__pyx_k_sum_colors_must_not_exceed_the_m__pyx_k_sum_pvals_1_1_0__pyx_k_svd__pyx_k_swapaxes__pyx_k_take__pyx_k_test__pyx_k_tobytes__pyx_k_tol__pyx_k_triangular_left_mode_right_size__pyx_k_u4__pyx_k_uint16__pyx_k_uint32__pyx_k_uint64__pyx_k_uint8__pyx_k_unable_to_allocate_array_data__pyx_k_unable_to_allocate_shape_and_str__pyx_k_uniform_low_0_0_high_1_0_size_N__pyx_k_unique__pyx_k_unpack__pyx_k_update__pyx_k_vonmises_mu_kappa_size_None_Dra__pyx_k_wald_mean_scale_size_None_Draw__pyx_k_warn__pyx_k_warnings__pyx_k_weibull_a_size_None_Draw_sample__pyx_k_x__pyx_k_zeros__pyx_k_zig__pyx_k_zipf_a_size_None_Draw_samples_f__pyx_methods__pyx_moduledef_slots__pyx_methods__memoryviewslice__pyx_getsets__memoryviewslice__pyx_tp_as_sequence_memoryview__pyx_tp_as_mapping_memoryview__pyx_tp_as_buffer_memoryview__pyx_methods_memoryview__pyx_getsets_memoryview__pyx_methods_Enum__pyx_tp_as_sequence_array__pyx_tp_as_mapping_array__pyx_tp_as_buffer_array__pyx_methods_array__pyx_getsets_array__pyx_methods_5numpy_6random_10_generator_Generator__pyx_getsets_5numpy_6random_10_generator_Generator__pyx_doc_5numpy_6random_10_generator_9Generator_12random__pyx_doc_5numpy_6random_10_generator_9Generator_14beta__pyx_doc_5numpy_6random_10_generator_9Generator_16exponential__pyx_doc_5numpy_6random_10_generator_9Generator_18standard_exponential__pyx_doc_5numpy_6random_10_generator_9Generator_20integers__pyx_doc_5numpy_6random_10_generator_9Generator_22bytes__pyx_doc_5numpy_6random_10_generator_9Generator_24choice__pyx_doc_5numpy_6random_10_generator_9Generator_26uniform__pyx_doc_5numpy_6random_10_generator_9Generator_28standard_normal__pyx_doc_5numpy_6random_10_generator_9Generator_30normal__pyx_doc_5numpy_6random_10_generator_9Generator_32standard_gamma__pyx_doc_5numpy_6random_10_generator_9Generator_34gamma__pyx_doc_5numpy_6random_10_generator_9Generator_36f__pyx_doc_5numpy_6random_10_generator_9Generator_38noncentral_f__pyx_doc_5numpy_6random_10_generator_9Generator_40chisquare__pyx_doc_5numpy_6random_10_generator_9Generator_42noncentral_chisquare__pyx_doc_5numpy_6random_10_generator_9Generator_44standard_cauchy__pyx_doc_5numpy_6random_10_generator_9Generator_46standard_t__pyx_doc_5numpy_6random_10_generator_9Generator_48vonmises__pyx_doc_5numpy_6random_10_generator_9Generator_50pareto__pyx_doc_5numpy_6random_10_generator_9Generator_52weibull__pyx_doc_5numpy_6random_10_generator_9Generator_54power__pyx_doc_5numpy_6random_10_generator_9Generator_56laplace__pyx_doc_5numpy_6random_10_generator_9Generator_58gumbel__pyx_doc_5numpy_6random_10_generator_9Generator_60logistic__pyx_doc_5numpy_6random_10_generator_9Generator_62lognormal__pyx_doc_5numpy_6random_10_generator_9Generator_64rayleigh__pyx_doc_5numpy_6random_10_generator_9Generator_66wald__pyx_doc_5numpy_6random_10_generator_9Generator_68triangular__pyx_doc_5numpy_6random_10_generator_9Generator_70binomial__pyx_doc_5numpy_6random_10_generator_9Generator_72negative_binomial__pyx_doc_5numpy_6random_10_generator_9Generator_74poisson__pyx_doc_5numpy_6random_10_generator_9Generator_76zipf__pyx_doc_5numpy_6random_10_generator_9Generator_78geometric__pyx_doc_5numpy_6random_10_generator_9Generator_80hypergeometric__pyx_doc_5numpy_6random_10_generator_9Generator_82logseries__pyx_doc_5numpy_6random_10_generator_9Generator_84multivariate_normal__pyx_doc_5numpy_6random_10_generator_9Generator_86multinomial__pyx_doc_5numpy_6random_10_generator_9Generator_88multivariate_hypergeometric__pyx_doc_5numpy_6random_10_generator_9Generator_90dirichlet__pyx_doc_5numpy_6random_10_generator_9Generator_92shuffle__pyx_doc_5numpy_6random_10_generator_9Generator_94permutation__pyx_doc_5numpy_6random_10_generator_default_rngcrtstuff.cderegister_tm_clones__do_global_dtors_auxcompleted.7182__do_global_dtors_aux_fini_array_entryframe_dummy__frame_dummy_init_array_entrydistributions.cfe_doublewe_doubleke_doublefe_floatwe_floatke_floatwi_doubleki_doublefi_doublewi_floatki_floatfi_floatrandom_mvhg_count.crandom_mvhg_marginals.crandom_hypergeometric.clogfactorial.clogfact__FRAME_END____dso_handle_DYNAMIC__GNU_EH_FRAME_HDR__TMC_END___GLOBAL_OFFSET_TABLE_PyUnicode_FromFormatPyNumber_NegativePyObject_SetItemPyList_Newrandom_laplace_PyUnicode_ReadyPyExc_SystemErrorPyDict_SetItemStringfree@@GLIBC_2.2.5PyDict_Sizerandom_buffered_bounded_boolrandom_geometric_inversionPyException_SetTracebackPyExc_NotImplementedErrorrandom_weibullPyMethod_Typerandom_f_ITM_deregisterTMCloneTablePyGILState_ReleasePyFloat_TypePyTuple_TypePyErr_RestorePyList_AsTuple_PyThreadState_UncheckedGetPyModuleDef_Initrandom_multivariate_hypergeometric_countPyEval_RestoreThreadPyFrame_NewPyMem_Freerandom_negative_binomialrandom_standard_cauchyPyCFunction_NewEx__isnan@@GLIBC_2.2.5PyCapsule_GetNamePyNumber_InPlaceAddPyBuffer_Releasevsnprintf@@GLIBC_2.2.5exp@@GLIBC_2.2.5PyNumber_AddPyObject_GetAttrStringPyExc_BufferErrorPyImport_AddModulePyBytes_FromStringAndSizerandom_standard_exponential_fill_f_PyObject_GenericGetAttrWithDictPyBytes_TypePyObject_SetAttrStringPyErr_WarnExrandom_standard_gamma_edatarandom_binomial_btpePyModule_NewObjectPyErr_NoMemoryPyErr_SetObject__pyx_module_is_main_numpy__random___generatorPyErr_NormalizeExceptionPyNumber_Absoluterandom_logseriesPyNumber_MultiplyPyLong_FromSize_trandom_rayleighrandom_standard_exponentialPyObject_RichComparerandom_uniformPyCode_Newrandom_poisson_finistrlen@@GLIBC_2.2.5PyImport_GetModuleDictrandom_bounded_uint64_fillPyObject_GC_TrackPyExc_RuntimeErrorPyErr_GivenExceptionMatchesPyErr_SetStringrandom_bounded_uint16_fillPyObject_IsInstancePyObject_GetItemPyExc_ExceptionPyExc_ValueErrorPyExc_DeprecationWarningrandom_multinomialPyObject_MallocPyExc_TypeErrorPySlice_TypePyGILState_EnsurePyInterpreterState_GetIDPyEval_EvalFrameExrandom_logisticPyErr_PrintEx_PyLong_Copymemset@@GLIBC_2.2.5PyMem_ReallocPyErr_SetNonerandom_standard_uniform_fill_fPyErr_ExceptionMatchesrandom_bounded_uint64pow@@GLIBC_2.2.5random_positive_intlog@@GLIBC_2.2.5random_standard_gamma_frandom_triangularrandom_buffered_bounded_uint32PyOS_snprintfPyTraceBack_Herefmod@@GLIBC_2.2.5PyObject_CallFinalizerFromDeallocrandom_powerrandom_bounded_uint8_fillPyObject_NotPyObject_Freerandom_noncentral_fPyNumber_InPlaceTrueDividerandom_standard_exponential_inv_fill_fPyLong_FromSsize_tPyFloat_FromDoublePyType_Readyacos@@GLIBC_2.2.5PyLong_FromLongmemcmp@@GLIBC_2.2.5_PyList_ExtendPyLong_AsSsize_tPyObject_RichCompareBoolrandom_buffered_bounded_uint8logfactorialPyModule_GetNamePyErr_Clearmemcpy@@GLIBC_2.2.5PyList_Append_Py_CheckRecursiveCall_Py_CheckRecursionLimitPyCapsule_IsValidPyExc_KeyErrorPyNumber_Orrandom_beta_Py_FalseStruct__gmon_start__random_exponentialexpf@@GLIBC_2.2.5PyTuple_NewPyObject_GenericGetAttrPyThreadState_GetPyExc_OverflowErrorPyDict_DelItemrandom_hypergeometricPyNumber_RemainderPyType_Modifiedrandom_gammaPyObject_SetAttrPyBytes_FromStringPyErr_Occurredrandom_standard_uniform_f_Py_EllipsisObjectrandom_loggamPyInit__generatorPyLong_AsLongPyImport_ImportModulesqrtf@@GLIBC_2.2.5_PyDict_GetItem_KnownHashrandom_gamma_fPy_OptimizeFlagrandom_zipfPyDict_GetItemStringPyEval_EvalCodeExpowf@@GLIBC_2.2.5PyObject_Sizemalloc@@GLIBC_2.2.5random_standard_exponential_frandom_pareto_Py_NoneStructPyExc_ZeroDivisionErrorPyFloat_AsDouble_endPyObject_IsTrue_PyType_LookupPyImport_ImportModuleLevelObjectPy_FatalErrorPyExc_AssertionErrorrandom_positive_int64PyThread_allocate_lockrandom_geometric_searchPyObject_Hashrandom_standard_tPyUnicode_Comparerandom_vonmisesrandom_bounded_uint32_fillrandom_standard_normal_frandom_positive_int32random_standard_uniform_Py_TrueStruct__bss_startlogf@@GLIBC_2.2.5PyFunction_Typerandom_normalrandom_chisquarePyDict_NewPyExc_IndexErrorPyBool_TypePyObject_GetBufferPyLong_AsUnsignedLongPyDict_TypePyDict_NextPyBaseObject_Typerandom_standard_exponential_fillrandom_intervalrandom_waldPyLong_FromUnsignedLongrandom_noncentral_chisquarePyLong_TypePyFrame_Typerandom_standard_normalPyCapsule_Typerandom_standard_exponential_inv_fill_PyObject_GetDictPtrPyErr_Fetchrandom_lognormalPyUnicode_FromStringrandom_buffered_bounded_uint16PyObject_GetIterPyEval_SaveThreadPyUnicode_InternFromStringrandom_binomialPyExc_ImportErrorPyDict_SetItemrandom_uintPySequence_TuplePyExc_AttributeErrorrandom_gumbelrandom_standard_uniform_fillPyExc_StopIterationPySequence_Listrandom_standard_normal_fill_ffloor@@GLIBC_2.2.5PyUnicode_TypePyCapsule_NewPyUnicode_DecodePyErr_Formatrandom_bounded_bool_fillPyCapsule_GetPointerPySlice_NewPyExc_NameErrorPyUnicode_FromStringAndSizePyModule_GetDictrandom_binomial_inversion_ITM_registerTMCloneTablePyNumber_InPlaceMultiplyrandom_multivariate_hypergeometric_marginalsPyNumber_IndexPyObject_GetAttrsqrt@@GLIBC_2.2.5random_geometricPyCFunction_Type_PyDict_NewPresizedceil@@GLIBC_2.2.5PyUnicode_FormatPyLong_FromStringPyMem_MallocPyErr_WarnFormat__cxa_finalize@@GLIBC_2.2.5_initPyNumber_Subtractrandom_standard_normal_fillPyNumber_MatrixMultiplyPyThread_free_lockPyTuple_PackPyUnicode_DecodeASCIIPy_GetVersionPyObject_GC_UnTrackPyExc_UnboundLocalErrorPyErr_WriteUnraisablePyDict_GetItemWithErrorPyList_TypePyNumber_FloorDivide.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.comment$.o8 @HHHo11&Uo33@d8484 vnBXXx88sPPP ~D |   8 8 tPp{)p{ x{)x{ {){ {){ h})h} ) ) h3 @+( ( 0( Y h b