ELF>@ @8 @hh99   p, p< p< W0s, < <  $$Std Ptd 44QtdRtdp, p< p< GNUGNU9MBTEF Ac$U=X! @,2 !!L 4 "n@DD04@$A!Ȓ&A C(VP19efX< " 0 (@`рIQP,APDdPD @0 @ đ` H@"Dk @CP @ @J"L0 @+h!"$%'(+,-./012456789:;<=>?ABCDEFHIJKLMNOPQRSTUVWXZ[]^`abcdefghijlmnqrsuvxyz{|}~     Tf-,_.,_WQ_g'3,_οM(hg (w|· ԦΕH Ԧ Ԧ J%{΅_J%_IJ%_(d6}(dkΑ3WK%; )dUSHUDe([D!s!sc75(g 6BH`RD\~8H RmXMIw~uJ^;> ՌܽH)ΑؔέXq3`%s@)y/=SNDm3vr+*s3vrmM?Pe߭[(՛#EG·i1;KY1;GVIVMVШTuRh uV u ( 9U 6H^C \0 Cb? (Qc gpnU@ 5U92S e 4C P[4 q y C Y{ 4 F *7c dl  ~ ( , L<?S #r 6 F"! c E Z<V Bp1P    pv   $ p X& 10=c  P OIQp P   @ ` p    V 8 c =p53@ F lI{@` A,Z@[ `:B L0 ' `I 0( ` ~ 0    =M P  uq`pI vV0V&   PC -` 85@ V?P @p *;p`7J0 I0z {p ZPyA7#`@ z[ @0*H`  `0 8MPn` %7w wZ;9 l`Bp"   `m,@  AD  Z  ^ 5P `$`  . 0 =` B@ /@L0)XpP`y+W@ V 0  ` L 'uP :+  (7h x  qp@   )H  n `     m 0+0 _5a x u1 c J1- `  :  W :BpX c mY B^pq  dP@I5P $`W !nbЩ r@N @o 0_m `p`p   5 W0  __gmon_start___ITM_deregisterTMCloneTable_ITM_registerTMCloneTable__cxa_finalize_Py_NoneStructPyDict_Size_PyDict_GetItem_KnownHashPyExc_TypeErrorPyErr_Format_Py_Dealloc__stack_chk_failPyObject_GetAttrPyCapsule_IsValidPyCapsule_GetPointerPy_EnterRecursiveCallPy_LeaveRecursiveCallPyErr_OccurredPyObject_CallPyExc_SystemErrorPyErr_SetStringPyMethod_TypePyNumber_AddPyNumber_InPlaceAddPyObject_SetAttrPyList_NewPyDict_TypePyDict_GetItemWithErrorPyTuple_NewPyObject_GetItemPyExc_KeyErrorPyErr_SetObjectPyTuple_PackPyObject_RichCompare_Py_TrueStruct_Py_FalseStructPyObject_IsTruePyUnicode_TypePyUnicode_FormatPyNumber_RemainderPyBool_TypePyLong_TypePyDict_NewPyLong_FromSsize_tPyDict_SetItemPySlice_NewPyLong_AsLong_PyType_LookupPyExc_AttributeError_PyThreadState_UncheckedGetPyObject_SizePyFunction_TypePyCFunction_TypePyType_IsSubtypePyFloat_FromDoublePyList_TypePyTuple_TypePyNumber_InPlaceTrueDividePyBaseObject_TypePyFloat_TypePyUnicode_ConcatPyNumber_SubtractPyObject_SetItemPyLong_FromLongPyNumber_FloorDividePyEval_SaveThreadPyEval_RestoreThreadPyLong_FromUnsignedLongPyObject_FormatPyObject_GetIterPyExc_ValueErrorPyExc_OverflowErrorPyErr_ExceptionMatchesPyErr_ClearPyFloat_AsDoublePySequence_ListPyNumber_MultiplyPyNumber_MatrixMultiplyPyList_AsTuplePyList_AppendPyNumber_NegativePyExc_UnboundLocalErrorPyNumber_Absolute_PyLong_CopyPyThreadState_GetPySequence_TuplePyNumber_OrPyLong_FromSize_tPyLong_AsUnsignedLongPyNumber_InPlaceMultiplyPyErr_PrintExPyUnicode_FromStringPyErr_WriteUnraisablePyMem_MallocmemcpyPyMem_FreePy_OptimizeFlagPyExc_AssertionErrorPyErr_SetNone_Py_EllipsisObjectPyObject_IsInstancePyUnicode_New_PyUnicode_FastCopyCharacters_PyUnicode_ReadymemsetPyExc_BufferErrorPyExc_StopIterationPyNumber_IndexPyLong_AsSsize_tPyExc_ZeroDivisionErrorPyErr_GivenExceptionMatchesPyBytes_FromStringAndSizePyBytes_FromStringstrlenPyBytes_TypePySlice_TypePyIndex_Check_PyList_ExtendPyExc_NotImplementedErrorPyGILState_EnsurePyUnicode_DecodeASCIIPyGILState_ReleasePyErr_NoMemoryPyObject_GC_UnTrackPyObject_GC_IsFinalizedPyObject_CallFinalizerFromDeallocPyObject_MallocmallocPyErr_FetchPyObject_FreePyErr_RestorefreePyObject_GenericGetAttrPyThread_allocate_lockPyObject_GetBufferPyBuffer_ReleasePyThread_free_lockPyObject_GC_Trackrandom_standard_uniform_fillrandom_standard_uniform_fill_frandom_uniformPyCode_NewPyObject_GetAttrStringPyDict_SetItemStringPyInterpreterState_GetIDPyExc_ImportErrorPyModule_NewObjectPyModule_GetDictPyCapsule_NewPyOS_snprintfPyErr_WarnExPyDict_GetItemStringPyModule_GetNamePyCapsule_GetNamerandom_laplacerandom_gumbelrandom_logisticrandom_triangularrandom_zipfPyInit__generatorPyModuleDef_InitPyObject_RichCompareBoolPyExc_NameError_PyObject_GenericGetAttrWithDictPyDict_DelItemPyType_ModifiedPyExc_RuntimeErrorPyFrame_NewPyEval_EvalFrameExPyImport_ImportModuleLevelObjectmemcmpPyErr_NormalizeExceptionPyException_SetTracebackPyExc_IndexError_PyObject_GetDictPtrPyObject_NotPyTraceBack_HerePyUnicode_FromFormatPyUnicode_AsUTF8PyCode_NewEmptyPyMem_ReallocPy_GetVersionPyFrame_TypePyUnicode_FromStringAndSizePyImport_AddModulePyObject_SetAttrStringPyUnicode_InternFromStringPyUnicode_DecodePyObject_HashPyLong_FromString__pyx_module_is_main_numpy__random___generatorPyImport_GetModuleDictPyType_ReadyPyImport_ImportModulePyCapsule_TypePyExc_ExceptionPyCMethod_New_PyDict_NewPresizedPyExc_DeprecationWarningPyErr_WarnFormatrandom_standard_uniform_frandom_standard_uniformrandom_positive_int64random_positive_int32random_positive_intrandom_uintrandom_loggamrandom_poissonrandom_binomial_btperandom_binomial_inversionrandom_binomialrandom_geometric_searchrandom_intervalrandom_bounded_uint64random_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_fillrandom_multinomialrandom_multivariate_hypergeometric_countlogfactorialrandom_hypergeometricrandom_multivariate_hypergeometric_marginalsnpy_sinlnpy_coslnpy_tanlnpy_sinhlnpy_coshlnpy_tanhlnpy_fabslnpy_floorlnpy_ceillnpy_rintlnpy_trunclnpy_sqrtlnpy_log10lnpy_loglnpy_explnpy_expm1lnpy_asinlnpy_acoslnpy_atanlnpy_asinhlnpy_acoshlnpy_atanhlnpy_log1plnpy_exp2lnpy_log2lnpy_atan2lnpy_hypotlnpy_powlnpy_copysignlnpy_modflnpy_ldexplnpy_frexplnpy_cbrtlnpy_sinnpy_cosnpy_tannpy_sinhnpy_coshnpy_tanhnpy_fabsnpy_floornpy_ceilnpy_rintnpy_truncnpy_sqrtnpy_log10npy_lognpy_expnpy_expm1npy_asinnpy_acosnpy_atannpy_asinhnpy_acoshnpy_atanhnpy_log1prandom_standard_exponentialrandom_standard_exponential_fillrandom_exponentialrandom_paretorandom_weibullrandom_powerrandom_rayleighrandom_standard_exponential_inv_fillrandom_standard_exponential_inv_fill_frandom_standard_normalrandom_standard_normal_fillrandom_standard_gammarandom_gammarandom_negative_binomialrandom_betarandom_chisquarerandom_frandom_normalrandom_lognormalrandom_standard_cauchyrandom_standard_trandom_noncentral_chisquarerandom_noncentral_frandom_waldrandom_vonmisesrandom_logseriesrandom_geometric_inversionrandom_geometricnpy_exp2npy_log2npy_atan2npy_hypotnpy_pownpy_copysignnpy_modfnpy_ldexpnpy_frexpnpy_cbrtnpy_sinfnpy_cosfnpy_tanfnpy_sinhfnpy_coshfnpy_tanhfnpy_fabsfnpy_floorfnpy_ceilfnpy_rintfnpy_truncfnpy_sqrtfnpy_log10fnpy_logfnpy_expfnpy_expm1fnpy_asinfnpy_acosfnpy_atanfnpy_asinhfnpy_acoshfnpy_atanhfnpy_log1pfrandom_standard_exponential_frandom_standard_exponential_fill_frandom_standard_normal_frandom_standard_normal_fill_frandom_standard_gamma_frandom_gamma_fnpy_exp2fnpy_log2fnpy_atan2fnpy_hypotfnpy_powfnpy_copysignfnpy_modffnpy_ldexpfnpy_frexpfnpy_cbrtfnpy_heavisidefnpy_rad2degfnpy_deg2radfnpy_log2_1pfnpy_exp2_m1fnpy_logaddexpfnpy_logaddexp2fnpy_heavisidenpy_rad2degnpy_deg2radnpy_log2_1pnpy_exp2_m1npy_logaddexpnpy_logaddexp2npy_fmodlnpy_fmodnpy_fmodfnpy_divmodfnpy_remainderfnpy_floor_dividefnpy_divmodnpy_remaindernpy_floor_dividenpy_heavisidelnpy_rad2deglnpy_deg2radlnpy_gcdunpy_lcmunpy_gcdulnpy_lcmulnpy_gcdullnpy_lcmullnpy_gcdnpy_gcdlnpy_gcdllnpy_lcmnpy_lcmlnpy_lcmllnpy_lshiftuhhnpy_rshiftuhhnpy_lshifthhnpy_rshifthhnpy_lshiftuhnpy_rshiftuhnpy_lshifthnpy_rshifthnpy_lshiftunpy_rshiftunpy_lshiftnpy_rshiftnpy_lshiftulnpy_rshiftulnpy_lshiftlnpy_rshiftlnpy_lshiftullnpy_rshiftullnpy_lshiftllnpy_rshiftllnpy_spacingfnpy_spacingnpy_spacinglnpy_nextafterfnpy_nextafternpy_nextafterlnpy_get_floatstatus_barrierfetestexceptnpy_get_floatstatusnpy_clear_floatstatus_barrierfeclearexceptnpy_clear_floatstatusnpy_set_floatstatus_divbyzeroferaiseexceptnpy_set_floatstatus_overflownpy_set_floatstatus_underflownpy_set_floatstatus_invalidPyException_SetCausePyObject_CallObjectPyEval_EvalCodeEx__vsnprintf_chk_Py_FatalErrorFuncPyDict_NextPyUnicode_Comparenpy_log2_1plnpy_exp2_m1lnpy_logaddexplnpy_logaddexp2lnpy_divmodlnpy_remainderlnpy_floor_dividellibm.so.6libc.so.6GLIBC_2.3.4GLIBC_2.14GLIBC_2.4GLIBC_2.2.5GLIBC_2.35GLIBC_2.27GLIBC_2.29 EPti O[ii fui p;|ui pp<  \x< [<  <  <  <  <  <  <  G G G  `H ؗ hH 8 pH H xH @ H 8 H 0 H ` H X H 8 H 0 H P H 8 H 0 H x H ( I  I 8 @I ؗ HI З PI 8 `I 8 hI 0 pI ( I Ș I  I  I ؛ I 8 I  I  I 8 J  J 8 J  (J  0J 8 @J  HJ  PJ 8 `J  hJ  pJ 8 J  J  J 8 J ؗ J 8 J ؗ J 8 J ؗ J 8 K 0 K ( K 8 K @ (K 8 @K 8 `K @ hK P pK 8 K @ K 8 K p K h K P K 8 K p K h K 8 L  L  L 8 L  (L 8 0L 0 8L ( `L  hL  pL 8 L 8 L 0 L ( L ` L X L 8 L  L  L 8 L  L  L x M @ M 8 M 0 (M ( 0M  8M 8 PM @ XM 8 pM ؗ xM 8 M @ M 8 M  M @ M 8 M  M @ M 8 N  N 8 N h (N  0N 8 8N x `N  hN  pN 8 N 0 N  N 8 N  N 8 N  N 8 N ( O إ O Х O ȥ O  (O  0O  8O  @O 8 `O  hO  pO  O h > D> 0k> D> ? D? @? DH? `h? Dp? ? C? ? D? P? D? L@ E@ 0@  E8@ X@ E`@ O@ E@ @ E@ PA 'EA  A 3E(A @A 8EHA `A DhA  A DA A ЉA }A A B }B `B DhB B DB B EEB  C ME(C 0@C DHC `C DhC C 0C }C @C CC }C @D BHD XD  hD YEpD  xD `D hED bD uED oE EE p E : (E v8E  - @E  HE 0\XE ' `E J hE _xE  E A E ~E   E V E E @ E EE PE  E  E pE  F {'F F  F '(F `O8F  @F 'HF XF  `F 'hF 0XxF @ F 'F F  F &F PF  F &F F  F 'F 0F  G &G  G  G '(G 8G  @G "'HG 0XG  `G -'hG  xG t G 6'G G e G ='G G X G E'G G K G K'G G > H S'H `H , H Z'(H 8H  ! @H c'HH XH   `H m'hH p xH  H v'H PH  H 'H )H @ H ,H H  H ,H `aH  I ",I eI  I *,(I g8I  @I X/HI jXI  `I S/hI ppxI  I /,I mI  I 9,I I  I /I UI  I F/I @I @m J /J P*J b J 0(J 8J  Y @J 0HJ {XJ T `J x0hJ xJ O J ")J 1J  J 9M K  K 3M 0K  8K M XK  `K L K @ K L K 0 K @L K  K L K Ц L K L H (L K HL H PL `A pL  xL HA L ` L  A L 8 L @ L  L @ M  M @ 8M  @M @ `M  hM `@ M  M @@ M  M  @ M 0 M @ N @ N ? (N  0N ? PN  XN ? xN Ъ N `? N  N @? N  N  ? N p N ? O ` O > @O  HO > hO  pO > O  O @> O  O > O  O = P  P = 0P  8P @= XP ` `P = P  P < P P P < P p P < P  Q p< Q  (Q P< HQ Ь PQ 0< pQ P xQ < Q ` Q ; Q  Q ; Q  Q ; R P R @; 8R p @R ; `R  hR : R 0 R : R  R P: R Ы R 0: S @ S : (S  0S 9 PS  XS 9 xS  S 9 S h S 9 S  S h9 S  S @9 T  T 8 @T  HT 8 hT  pT `8 T  T @8 T  T <8 T  T 08 U  U 8 0U  8U 7 XU  `U 7 U  U 7 U H U 7 U  U 7 U  V |7 V x (V p7 HV  PV 6 pV ( xV 6 V  V 6 V x V 6 V  V 6 W Ȩ W 6 8W ( @W `6 `W  hW  6 W  W 6 W p W 5 W  W 5 X x X @5 (X  0X  5 PX ب XX 5 xX  X 5 X  X 4 X  X `4 X p X Q4 Y h Y O4 @Y  HY N4 hY ؗ pY L4 Y  Y L4 Y h Y 04 Y x Y 4 Z  Z 3 0Z  8Z 3 XZ  `Z ]3 Z  Z Y3 Z  Z P3 Z 8 Z @3 Z  [ 33 [  ([ (3 H[  P[ 3 p[ @ x[ 3 [  [ 3 [  [ 3 [ 0 [ 2 \ ` \ 2 8\  @\ 2 `\  h\ 2 \ 8 \ 2 \ З \ 2 \  \ 2 ]  ] 2 (] ( 0] $ P] Ș X] $ x] H ] $ ]  ] $ ]  ] $ ]  ] # ^ @ ^ " @^  H^ " h^  p^ " ^  ^ " ^ ȣ ^ " ^  ^ " _  _ " 0_  8_  X_  `_  _ ( _ X _  _ X _  _ H _  ` 0 ` ` (`  H` h P`  p` 8 x`  `  `  `  `  `  ` p a ( a k 8a أ @a d `a ` ha ^ a  a P a  a E a  a  b Ю b  (b H 0b  Pb  Xb  xb @ b  b @ b  b 8 b  b h b  c ` c  @c p Hc  hc X pc  c ` c  c  c `c  c Md H d F0d 0 8d @Xd ȥ `d 0d 0 d !d  d !d  d d Ƞ e  e  (e  He P Pe pe  xe e  e e ȕ e e  e f  f 8f  @f `f 8 hf f Е f f Х f f  f g  g (g  0g xPg  Xg pxg  g pg  g ig H g `g X g @h  h 0@h  Hh hh x ph h  h h  h ph  h @i  i 0i X 8i Xi Ȝ `i i 0 i i ت i @i Ш i #i  j  j  (j Hj ا Pj pj  xj j  j j @ j j 0 j k  k 8k p @k ظ`k  hk k P k k  k k  k l  l (l  0l ЫPl  Xl xl  l l  l l X l l ( l m  m @m @ Hm hm 8 pm m  m m  m ̞m  m n  n 0n  8n Xn  `n n  n n  n n  n n p o  o x (o  Ho h Po  po Ȫ xo  wo  o wo ` o wo h o wp p p v8p Ч @p v`p ؠ hp vp 8 p vp  p vp  p vq  q v(q  0q pvPq  Xq @vxq  q 0vq x q 'vq Ȯ q vq  q ur  r u@r  Hr uhr 0 pr ur  r ur  r `hr  r Ts  s A0s  8s @Xs h `s @s x s @s h s @s ( s @s  t @ t  (t @Ht  Pt 4pt @ xt @4t ` t  4t 0 t 4t  t 4u  u 48u 8 @u 3`u P hu 3u H u 3u  u )u ( u v h v (v  0v  Pv  Xv  xv  v  v  v  v  v  v  v  w @ w p @w  Hw P hw X pw F w P w 0 w x w  w h w  x 8 x  0x  8x  Xx إ `x  x h x  x  x  x  x  x ( y   y  (y p Hy  Py @ py @ xy 8 y p y 8 y p y  y  y  z ȫ z  8z H @z `z X hz z x z  z P z z p z {  {  ({ 8 0{ P{  X{  x{ X { { ` { { P { {  { |  | @|  H| h|  p| p|  | `| 0 | P|  | 8}  } (0}  8} X} X `} }  } }  } } @ } } ج ~  ~  (~ H~  P~ p~ X x~ ~ h ~ ~ h ~ ~ ` ~ p p  X8 H @ P` X h H   8   ( ؛   0  ( ȧ 0 P  X x У   Ȁ  Ѐ  8  0 @ X H h P p      0  h  `0 ( 8 XX  ` R  R 8 MЂ  ؂ H ؕ C X ( 8H  P @p Ȭ x  X  ` ȃ Ц x     8  @ ` ` h     ؄     p(  0 @P Э X  x    ȅ ( Ѕ     @ ( H h x p    @      x0 ؔ 8 qX 8 ` h P `  YЇ 8 ؇  X   ( H  P ܁p  x Ձ Ж ρ H Ȉ  P    `x8  @ Ex`  h >x P 7x ث @l؉ H a ؝  a( О 0 aP  X ax  T  TȊ p Њ T  T  T@  H J  Ћ PPP  X m` m > > ، JX  p J  B A ȍ  A P P( @ 8 @? x E E ,8 } 0 ȏ `B  P 0 E B C C  D C h  C x B @-8 H P ,x h kВ p ؒ  Ђ D  @D H cX Ѐ E3ȓ `" 0 ` `4 ( D@  H J > > > > > > > )> /> > G> Q> S> T> V> W? ? ? ?  ? (? 0? 8? @? H? P? X? `? h? p? x? ? ? ? ? ? ? ? ? ? ? ? ? ? @  @ (@ 0@ 8@ @@ H@ P@  X@  `@  h@  p@  x@ @ @ @ @ @ @ @ @ @ @ @ @ @  @ !@ "@ #A $A %A &A ' A ((A *0A +8A ,@A -HA .PA 0XA 1`A 2hA 3pA 4xA 5A 6A 7A 8A 9A :A ;A <A =A >A ?A @A AA BA CA DA EB FB HB IB J B K(B L0B M8B N@B OHB PPB RXB U`B XhB YpB ZxB [B \B ]B ^B _B `B aB bB cB dB eB fB gB hB iB jB kC lC mC nC o C p(C q0C r8C s@C tHC uPC vXC w`C xhC ypC zxC {C |C }C ~C C C C C C C C C C C C C D D D D  D (D 0D 8D @D HD PD XD `D hD pD xD D D D D D D D D D D D D D D D D E E E E  E (E 0E 8E @E HE PE XE `E hE pE xE E E E E E E E E E E E E E E E E F F F F  F (F 0F 8F @F HF PF XF `F hF pF xF F F F F F F F F F F F F F F F F G G G G   G  (G  0G  8G  @G HG PG XG `G hG pG xG G G G G G G G G  HH? HtH5? %? hhhhhhhhqhah Qh Ah 1h !h hhhhhhhhhhqhahQhAh1h!hhhh h!h"h#h$h%h&h'qh(ah)Qh*Ah+1h,!h-h.h/h0h1h2h3h4h5h6h7qh8ah9Qh:Ah;1h<!h=h>h?h@hAhBhChDhEhFhGqhHahIQhJAhK1hL!hMhNhOhPhQhRhShThUhVhWqhXahYQhZAh[1h\!h]h^h_h`hahbhchdhehfhgqhhahiQhjAhk1hl!hmhnhohphqhrhshthuhvhwqhxahyQhzAh{1h|!h}h~hhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!hhhhhhhhhhhqhahQhAh1h!hhhhhhhh%/ D%50 D%m0 D%e0 D%]0 D%U0 D%M0 D%E0 D%=0 D%50 D%-0 D%%0 D%0 D%0 D% 0 D%0 D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%/ D%}/ D%u/ D%m/ D%e/ D%]/ D%U/ D%M/ D%E/ D%=/ D%5/ D%-/ D%%/ D%/ D%/ D% / D%/ D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%. D%}. D%u. D%m. D%e. D%]. D%U. D%M. D%E. D%=. D%5. D%-. D%%. D%. D%. D% . D%. D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%- D%}- D%u- D%m- D%e- D%]- D%U- D%M- D%E- D%=- D%5- D%-- D%%- D%- D%- D% - D%- D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%, D%}, D%u, D%m, D%e, D%], D%U, D%M, D%E, D%=, D%5, D%-, D%%, D%, D%, D% , D%, D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%+ D%}+ D%u+ D%m+ D%e+ D%]+ D%U+ D%M+ D%E+ D%=+ D%5+ D%-+ D%%+ D%+ D%+ D% + D%+ D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%* D%}* D%u* D%m* D%e* D%]* D%U* D%M* D%E* D%=* D%5* D%-* D%%* D%* D%* D% * D%* D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%) D%}) D%u) D%m) D%e) D%]) D%U) D%M) D%E) D%=) D%5) D%-) D%%) D%) D%) D% ) D%) D%( D%( D%( D%( D%( D%( D%( DUH5 1,Hy Hu H5q 1Hvw HtHJ H5x 1Hw HtH5 1Hqv HtH5 1HJv HqH5ʏ 1~Hu HNH5 1[Hu H+H5t 18Hyu HH5I 1H>u HH5 1Hu HH5 1Ht HH5Ȏ 1Ht H|H5 1HZt HYH-z 1HHH]Hs H-H5u 1:Hs H H5# 1H{ HH5 1Hm{ HH5͍ 1Hjz HH5 1H?z H~H5w 1Hz H[H5L 1hH9} H8H5! 1EH| HH5 1"Hk| HH5ˌ 1H@| HH5 1H| HHHH.H{ HH5W 1Ht{ HkH, H5 1qH { HAH5 1NHz HH5ϋ 1+HL~ HH5 1H!~ HH5y 1H} HH5N 1H} HH5# 1Hx} HoH5 1|H%} HLH5͊ 1YH H)H5 16H HH5w 1Ht HH H5 s 1H2 HH5" 1H~ HH5 1H܉ HsH5ĉ 1HWH5 1dH] H4H5u 1AH2 HH5J 1H HH5 1H HH5 1H) HH5Ɉ 1H HH5 1Hˁ HbH5{ 1oH H?H5P 1LH% HH5% 1)H HH5 1H? H H HHw H5և HHP1Hd HH5 1H9 HhH5a 1uHn HEH5f 1RH H"H5C 1/H HH5| 1 HHDo L ͆ VA11AQh5 5 RRPRRHPHH5o 1HT HsH5< 1H! HPH5 1]H H-H5օ 1:H H H5 1H HL p La 1H ;~ H<~ H5=~ HHn L Q11AAQj5~ 5 RRPRRHPH]AVIAUIHATUSDHt/H;i HuE1tHLL&AH_H AH8tE1[D]A\A]A^AVAUATUHQ*HxH* HuH Ht#H9tH2 H53pH8[E1L% Mt I$H5HIHtHLI^MtLIHAH ?HHH=xkAH .LHH*xHAH LHHpx%E1H LHH PxML^ZL]A\A]A^ATI1HU1QHŃHt*H5~ LHHx ]1]Z]A\AVIHAUIATUDSHHdH%(H$1IHH@u H LLH5QoH81OnML$ I9s#H ILLH5LoH81"AuHI9vCHPMLAQHoIH1b1H1Hy LE1\H$dH+%(tHL[]A\A]A^AWIAVIH5AUIATUSHAPHŃHLHIHu(L{LH5qpHH_ H81=gLHu9LLH9IMLHHv H5WpH81LLHHt H[1 H[Z[]A\A]A^A_AWIAVIH5AUIATUSHAPHŃHLHIHu(LLH5 pHHs H81QgLHu9LLHMIMLHH H5oH81 LL HHt HZ1 HZZ[]A\A]A^A_H=Ff 1ATIUQH5j KHHuE1Ht!LHAąxHxZDZ]A\HGH H9u 1ATUQJIHu-HxXHHL H0dt111H"KLZ]A\AWAVAUIATE1USHH5y HVH- H5y IHHHwH5y L JHD$HZH9H5wy HWHHHH5\y LIHH4H59y H9uH5-y LIHuVHrAąu1E11IH5x HIH5x AąyH9GHE1H5x LIHH H5tx L,IHt6IH5^x HxZIH5Ht9Hu/LwTHD$1E11E11 1E1E11HuH_ IUH54mH81AHtHWH|$u1E1H|$WMtLWHtHWMt&LWH5pw Hd H@'}@H9m H#' #Hc H'}Hj H&~}Hm H&12HCc H&Ha H&H` H&2H}` Hg&1 Hr HM&ʚ;Hj H3&11H=n Hz H&HwHi H%H 8uvHH#6Hyx H5c H=x yAL-AL%HLHtH=s CHtA.H9x LHyAL-AH=cs CHuAL-AeH=2s eCHb HtH=s MCHb HtH=r 5CHa HtH=r CHf HtH=r CHx HlH=r BHj HPH=jr BHm H4H=Fr BHl HH="r BHs HH=q yBHH=q dBHp HH=q HBHqp H!HL H-[ HHHs Hs Hs Hr Hr #H5Or H=Pv H#H)r Hk H#Y Hn HHrXHr #H5n H=Y #Hv#Htr H%W H=X#HH#H^q Hon HPU Hm HBnHH8q HuH2q H-H,q H%H&q HIH q H|Hq HHq "H5pm H=U "H"Hep H-R H57m HHl HH`p HH~H[ HPp HS /"H5el H=S p"H_"L%H-m LHH!AxLHH0fHHOLHH!A LHH $HHNLgHHF!A LHH7HBHL%'NLHH A`LHHbH] HAH LHHlHAi HA0LHH?Hd HAXLHHH[ HkALHHcHEALHHEHALHH.HVc HALHHlHALHHFHALHH HALHHHY HSALHHrH-ALHH\HALHH?HALHH#[HHL-KLHHvA`LHHxHrA@LHHHLHH5l IH-1HHHu HuH H5aH8LFKHALHHZdHHL-SKLHHgL%HHo HLH5B+lLHk H5H KH Hk HH5 &H~JH=aHHL%'aHX HLH5LHX H5HhLHkX H5HGLH:X H5H&nLH X H5~HMLHXX H5hH,LHGX H5SH LHX H5>HLHW H5)HH.ILHHuH 3`H_ HH5@H ^`H_ HH5cH HU HH5>L%l`HX HLH5LHW H5HH k`H,_ HH5H n`HwW HH5cH iaH^ HH5n>H \bH[ HH5NaL%bHZ HLH5-9LH] H5 HH}GH=U 11 @HHH5T H=V H[H;GH=\ 11?HH^H5x\ H=V HHFOHH:Ha HUHHHH=k 1a?IHHFH5a LN@HHLH5a H=V HHlFLdFH=5k 11>IHH5U H=U HBL"FxIHH=a IULHHH=j 1>HHLEH5a Hw?IHH5` H=EU HLEHEHHH` HUHHHH=$j =IHH>EH5_ L>HHH5_ H=T HHDLDHD$HD$HD$ 5HHH(Ht H;-tuHPHtHL`LhHtHEMtI$MtIEH=$HHrH5)HD$H|$ISDMuHgH5 H84H<I9Ft#HoH58`H8`LD1L LHQ CH Q HuH'H5H8= HP t& H5_HH81 HP w'H5`HH81VUAƅuHH5!`H81/.tHqH52`H81Ht2HC(HPH{XH0,u4AjcAMtLBMLBH jcH=IHL$ HT$HHt$?x=H5eh H=c 13IHt.H1A?LAcABAcA AcAHLLHpH|$HtBH|$HtAH|$ HFA<Hf IHuH_\ H5hf LHLAH=2\ HQ H'f H9Xu#L-f MtIE7H=P 1IH=P He H5e d;IMH5P Ly1HH LAHP HP He H9Xu#H-{e HtHE7H=aP \1HH=PP HQe H5Re :HHH5"P H0HH: H~@HP HxO Hd H9Xu#L-d MtIE7H=O 0IH=O Hd H5d V:IMH5N Lk0HH L?HO HN Hbd H9Xu#H-Md HtHE7H=SO N0HH=BO H#d H5$d 9HHH5O H/HHG Hp?HO HS Hc H9Xu#L-c MtIE7H=N /IH=N Hc H5c H9IM;H5N L]/HH L>HBc 11H=7J HR HHH58d H=IN H H>*HHHb H5b H}g Hb H5b H_[ Hxb H5ib HAO HJb H5;b H#C Hb H5 b H7 Ha H5a H+ Ha H5a H Ha H5a H Hda H5Ua H H6a H5'a HoHa H5` HQH` H5` H3H` H5` HH~` H5o` HHP` H5A` HH"` H5` HH_ H5_ HH_ H5_ HH_ H5_ HaHj_ H5[_ HCH<_ H5-_ H%wH_ H5^ HkH^ H5^ H_H^ H5^ HSH^ H5u^ HGHV^ H5G^ H;H(^ H5^ Hq/H] H5] HS#H] H5] H5H] H5] H Hp] H5a] HHB] H53] HH] H5] HH\ H5\ HH\ H5\ HH\ H5{\ HcH\\ H5M\ HEH.\ H5\ H'H\ H5[ H H[ H5[ HH[ H5[ HHv[ H5g[ H{H5H[ H=!I HoHL%|j91LH=6HH HZ H5Z HH@0H 9H=Z DH5u^ H=V 1*H H=Z HZ 8H54^ H=V 1^*Hf H=VZ HOZ 8H5] H=|V 1-*HM H=Z HZ 8H5] H=KV 1)H4 H=Y HY P8H5q] H=V 1)H H=Y HY H]Y L-f98GU H;Y H7Y H3Y H/Y H+Y H'Y H#Y H1HY H=T LLHH~ H~C H5Y HHpxHP7H=QC t1LLHHD HlU H5X HH>H6H=?U "HX 11H= B HH H5U H=YF HH6AL--AH6MtL|6H=|[ t=H=F tLDH=Dc=H=T[ Ht2HD[ 76 HuH<H5FH8eH=[ AL-A]E1A,L-tABMA1AL-S'E1APL->A ML-)ARAE1AbL-AMAgAL-E1AL-AE1AL-AAڣL-AmAL-[AL-uIAL-c7AL-Q%AL-?AL--AL-AL- AL-AL-AL-AL-AL-AL-qAL-_AL-yMAL-g;AL-U)AL-CAL-1AL-AL- AL-AL-AL-AL-AL-AL-uAL-cAL-}QAL-k?AL-Y-AL-GAL-5 AL-#AL-AL-A L-A L-A L-A L-A L-yAL-gAL-/AOAL-A%7AL-AAL- DDH ArH=~L-A(8He1HX1HK1A/A`A>A RAMADAeA6A}A(AAAA AʣAqML-dEAL-RA5AL-:AAL-"AAL- AAL-AAL-AAL-AAL-AAL-AuAL-zA]AL-bAEAL-JA-A'L-2AA<L-A AHL-AA]L-AA{L-AAL-AAL-AAL-AmAL-rAUAȣL-ZAq=AأL-BA%AL-*AL-AA%L-AA3L-AAAL-~A AOL-fA#A]L-NA$AL-6A%kAL-ASAL-AAL-FA)AL-.AAL-AAL-AAL-AH$dH+%(tH[]A\A]A^A_f.H=8 H8 H9tHHt H=8 H58 H)HH?HHHtHHtfD=8 u+UH=Ht H=艳dm8 ]wAWAVAUATUHSHH(L5LndH%(HD$1H$HD$Lt$H~I\IHV(HE LMHHHsH=AHEHAVj5: 5: j5: Pj5: : IHEHPMHHEHD$dH+%(H(L[]A\A]A^A_HH5: LIHVIH$HLmDIH HAHMEIHHSHmH5]AUL .H81¾XZH J&H=pE11*LHHII[qM IHFHH$IH59 LHVHD$H3IML $HD$HT$,f.IHF(HHD$HF HD$HFH$脱IM~1HL QMH5vLβypHHEt&H %LH=F0@H(fDH A`fDH58 LHVH`HD$IMfDHF HHD$HFH$诰IHH\H eH5fjL 7AHYH81込Y^f蝳H &Aff.AVAUATUSHH L 7 L-'dH%(HD$1HnL $Ll$HQH?H%HHHIH HۿHIHHrI?UIH5{H8L KA1޻X ZH fQH=E1.HD$dH+%(H L[]A\A]A^MHH6 HHsH6 H=zAHEHAUjRPjRLPj5V6 6 IHEHPM|HHEbHְULV LNvMIHHteHH脮IHL $LT$%fHHEH 13H=z-HF HHD$HFH$H~1HL IH5Ldy_fDHFHH$ЭIMTH55 LHVHtHD$IF느H訯AH54 LHV譵HtH$I詰fUHHGH55 HHHHHEH5S5 HHHHMHQHUHt-HtH]fHHD$HD$H]=HtVH H=#,H1]fD諹HHa;뻐苹hfDHt$tt$ff.AVAUATUHSHH LndH%(HD$1HD$HDIHnH{HEH/HEHkHH5q4 HHAIMCL-LLaYLLIHAoEH53 CAoMK(IE HC8HEHHHHHHH/t=HE1I,$tLHD$dH+%(H D[]A\A]A^ ffDLfDIMDIzHFHHD$詪H>Hl$fD蛷IH ZAH=R),H-2 L-2 HEHHsH=ǻz1HLHhHn1H5HmH H=»A')f{H fDH訶H=@HxH5 2 LIHV蒱HHD$IFHLmfHHHgH5AUL AH H81WXZH ߸H=ݺAB(DHt$ T$تT$t$ H1LnHHfLHuHH59H81HL$MLL H5販0kff.AWAVAUIATUSHHGH5p0 HH0HH:HCH9]YL}MLLuIIHm%LLBI/IM|I.RH=/ H5/ HGHH HHH=/ LIHH9]L}MLuIIHmLLLI/HImHI.HL5II$MHI$H+HL[]A\A]A^A_H萨HIIDLHIImHaLLTL8L(CHHVL<L>Lا˲HHAeDE1H vH=$I$AvHI$uLlHtHmuHXfDKHI$AxHI$uf.I.As^LQHI$tFH+tA3DHئfDI$LAHI$AIL訦fDAVAUATUHHHGH5, HHsIMvID$H5, LHHII$HI$M|HcHEH5, HHHHHwHEH5,, HHHIHEHMHEHID$H5+ LHHHI$HHI$HH=+ HŤIHiHmH5o+ L蟤HHI,$HLMIH1HmImIMHPIIHHL]A\A]A^HȤ L踤L訤/H蘤EI,$uL|H KME1H=!IEHaL@HL]A\A]A^ÐL(LHIMH H="!HL]A\A]A^軮IrHH vE1H=̳ fADME1H ;H= IEH^@CH;AHmuH#말I? HqkfDAALТ$ff.ATIUHHGH5/) HHHHHEH5( LHHHHMHQHUxHtIHBHH]A\fDHt6H H=fH1]A\@HfDHt$ t$ fD}fˬH1;PfDAWAVIAUATUSH1HIH' IT$LHHH=x' >HHI,$H5X' H(IHH8HmIFH5' LHHHHHEH5' HHHIHEHMHEHJH& HLHI9D$HH*HI,$qIHIFIl$LH5& HHHHHEH59& HHHIMHm5HIULHJIML` Lp(LhIHHHL[]A\A]A^A_fDH萟tH耟Hm^LhHXAHI,$uLT$ T$ fH ߬DH=JMAIELE1HPHH'ҞDLPH谞H wE1H=-E1A=DAHmBHT$ FT$ -D3H7H=H [IELE1HP<IAs苟HHIAHCT$ HHIH8ѝT$ HA0[H&I,$AD+I"I,$t)I.t@H H=FLAI.LAŜAKLA蚜31tT$ HHDHH*H8貜HmT$ #HKT$ fAWAVAUATIUSH8HL" dH%(HD$(1HnH\$LD$H\$ H1HgqHHIHFHD$H=! LCHHHx! H $ H9HL5$ MGIIFH58! LHH=IIHMIH,LH IH I/fL;5L;5u I9DI.wEM$HT$It$MH=2ILu H<I.Hm7HT$(dH+%( H8[]A\A]A^A_@HVLn(LF fH\$IH H z# H9HL=a# M`IIGH5 LHHIIHMIHLH襚IH!I.L;=L;=fSI9JL肠AƅI/EM$HT$It$MH=IL H:I/LHD$HD$oLLADžAAI.H DDH=1L記LHD$蓘HD$fLF I@Lh|HHD$SHD$fDHH HIHHI?IAHHHH5æUL H81)XZH H=1,IHHHcHFHHD$yIM~TH5j LHV莝HtHD$IM~.H5< LHVhHXHD$ IMFHD$LD$Ll$ HD$R@H ҤAfDHuHF(HHD$ HF HD$HFHD$返If.L踖H H=h1@H=Q H H5 IMAAzAA^H= IIL.AAI/LDLЕLNH= |IMA!AA!Af.AA ~H=! H H5 IfA!A+IHIHH5 LHVHWHD$IDf.HF HHD$HFHD$薒I>fDAA!H=! H9HEt H;H螟IMH= LIHI.tl1LA$A;I/*@A"A1HL$ILL ͡H5'L譓HIPA$AAA$褔@AWHAVAUATIUHSHHHFH5 dH%(HT$81L-p LpH\$Ht$Ll$ H\$(H/IH JcH>@H$Lh(Hp H@HD$H= _HHcH H  H9HL5 M3IIFH5T LHHIIHMIHLH(IHI/L;5;L;5I9LADžI.EH H  H9HL5 MOIIFH5h LHH]IIHM0IHLHLIH0I/L;5_L;5 2I9)L)ADžI.lE H5D LL L$It$M$ILHSIHT$/H=hoAHeI.HmHT$8dH+%(HH[]A\A]A^A_@H$H\$fH$sH$_HH0H $GI.D2LE*H5J LL L$It$M$ILHSIHT$H=XAHAAL蘏;L舏QMHBMH /HiHIHH I?AVIH5H8L A1tXZH H=e 1wfDDHH$H$LISH JcH>DHP0HT$(HP(HT$ HP H@HHT$HD$脌III~aIu*M~.H5W HHV胔H$HD$(IMHD$Ht$Ll$ HD$HD$(H$KfDM2IuM~H5 HHVHtHD$IM~H5 HHVHTHD$ IA@LH$謍H$L蘍H=XAHMAAd%fH /H=H 1@L0H= HJ H5K ~IMkAAfDH DDH=ܝ' 1.H=iWAHMAAh@L蠌AAI.uL~@sIH=1 ,II@AAI/JL2=DAAf.H@HHD$IpH؉IMH5 HHVHHD$IfH=y H H5 IMAA*fH=9 4IA,A;IAA/A1A{H=) H9HEt H;LHIMH=[ LCIHI.1LAA跄I/>AA<oAAWAIA1HL$MHL #H5vLZHOIAAAA9H/tD髉ff.AWAVAUATIUSHXH|L= H|$LndH%(HD$H1H%H\$(HD$ H\$0L|$8HD$@H I HJcH>@HٵHD$M|$0ID$(H$Il$ Ml$IEHEH9a H= LbIH H H  H9HL5 MIIFH5 LHH|IIHMIHk L޺LL\$&L\$HI]I+[ L;54L;5I9LVI. H H  H9H L M4IICL\$LH5 HHL\$IIHM IH LL?IHI. L;RL;} I9t LL\$L\$AI+mEH H  H9H|L5 MIIFH5 LHHzIIHMIHqL޺LL\$TL\$HII+qL;5bL;5u I9I.KH H  H9HL MIICL\$LH5 HHL\$IIHMIH~LL~IHI.L;L;?I9LL\$VL\$AI+E H H H9H_L5 MMIIFH5 LHHIIHM*IHL޺LL\$蓅L\$HI I+EL;5L;5O<I93LkI.@H" H H9HL MIICL\$LH5 HHL\$IIHMIHLL謄IHI.L;L;m?I96LL\$脊L\$AuI+EH2 H H9HrL5 MIIFH5 LHHIIHMIHHL޺LL\$L\$HI I+/L;5ϯL;5}I9L虉I.HP H  H9HL MLIICL\$LH5 HHL\$IIHMIHgLLڂIHI.L;L;dI9[LL\$貈L\$AbI+6E:H` H  H9H$L5 MIIFH5 LHHIIHMfIHL޺LL\$L\$HI$L6L>D$LD$ H|$AHD$HLLLHIQ1AVHT$ ^_IHzLHD$Lt$fH!H$HHD$OHH$HD$-HHD$HF8HD$fI.LD$D$HL$H; ߬AH; DpH9gH|$襆A HD$H1HLLLHIARHT$LT$_ IXZMLT$d I*H9$ILI,$MI.~LImuvL HmunH~afDLL\$~L\$~fL~H bH=ImtHmtHD$HdH+%(HXL[]A\A]A^A_HP|IH5v HIHVfHD$ H,Ml$MHH HNHL LNLHHAUHH5H81XZH H=E1fD1LI}M_DIEH+ L%r I$ImLMnEIHHJcH>f.HF8HD$@ID$0HD$8ID$(HD$0ID$ HHD$(ID$HD$ zIIHJcH>fL|4LHD$|L\$@fIH5LH}IH H;eL;L;$LL$*L$ I( H5EL|IHH;L; I9 LL$軂L$(I( eHn H  H9H<L MII@L$LH5 HHL$IM0I(IAL $LH5q HHL $IMuI)zL޺LL${L$HI.I+L;ЧL;~) I9 LL$薁L$uI([I/H.fDH9 H  H9HL M II@L$LH5r HH L$IIHM IHHFI9C LLL$eL$IIHM IHIAL $LH5 HH0 L $IIHM/ IHH5 LǺL$3zL$HI I(L; BL; I9LL $L $ I)IH5I9G LLXHIHH IHI,$IkfDH=Y H H5 IMwH /H=I,$E1f.HL$H; 4AH; DH9H|$~AC HD$H1HLLLHIASHT$L\$ A[IXL\$MI+TLwGLwE1yAI.LL$L\$WwLt$L$fDMtI(t-H DH=buI,$zE1fLwfDI|H= Ii@A|I+uLv}f.LvY~AE1E1f.Lx};AE1E1f.EYL(vuHFHHD$ sIMHD$0Ll$ Hl$(L|$8H$HD$@HD$fD$w{DD$H E1@H wH=҆L$L4$cH5 HHV{HtHD$(IMVH5G HHVc{HtHD$0IM,H5 HHV9{HtHD$8IMH5 HHV{HtHD$@IM1HL$ MHL ڂH5tE1A!LLMrtDH`tMLL$DtL$FI/L$#t$LL\$ tL\$xLL $sL $9LsLL$sL$XH= H H5 IMHL$H; AH; 9DEH9<H|$RzA4HD$H1HLLLHIARHT$LT$ AYAZHLT$IE1E1MֻAtfDLD$rD$DyLH$rL $j}L\$IfDH=Q LI@ALXrcE1ADAE1XDELlj$r$DAE1E1MֻAeLqLHD$qL\$!fH=i H H5 IM>Mn5H=) HB H5C IMrHL$H; ,H; ڝH9H|$wAHD$H1LHLLHIASHT$L\$~ _AXHL\$IA fLpGpA5-k{L$INH=! I@+{I~H= I@E1ADA5MCMIKIHI+WLLHLD$H $LD$L$II(LL\$L $oL $L\$A[zL $IMMA5fMA5PD,MME1A5Llj$n$CLL\$nL\$ MCMA4IAD$RtDD$HE1MûDI/ I+ A4LKnH= H H5 IM)iHL$H; H; ÚlH9cH|$tAHD$H1HLLLHIARHT$LT$W Y^HLT$I8E1E1Mֻ<A"LD$dmD$DMxL\$IvH= IAOMGMID$H5 LHHIMs&jIHHeHHH57 HLkHmH H { H9HH-b HYHEHEH5 HHHfIM@HmH5 LLjI.YIGH- HH6H=qB`jLHLIgMfI/$I,$ IFH5N LHH HHI.3HTH9]*L}MLeII$HmUH LLI/INMI,$IFH5 LHHIIHIMHI9_MwMMgII$I/ LLI.HHlI,$HEHXpH(H{L]cIH%H=ŒHHjI,$IMLHSImDHUHJHHMHfH[]A\A]A^A_DnLImHHIHHDnFII DnFII IIudIHgH l9H=Op21eDnAMc@HPHmH5FpH81hHUHBASHEHDH VlH=oH1[]A\A]A^A_nASHuLH^fDH8^AR7E1I/hI,$HtHmtYMlI.bHcI@H]H5 HIFIfDH]fDH]kI,$AS1E1DL`]NLM}H0DL8]L(]L]ARF1fD)AR\gI_L\I/uL\AR3fL\E1AR5[DHp\L`\LP\JLHLeIHRfL\HHD$\HD$fH= HB H5C FHHE1ASAL[H=i dHASCVkfILX[LHD$C[HD$fL([\AS fHARPE11`HH1ARPDLZeI,}fDH@`H`HHPHHH?H@H;LoIEHHNnHcH>DoH/ ZH/YYE1@DoAMcH/f.DoGII IDDoGII yH5ɊHHdH@$H|$:_H|$IH|$VH|$IsHARPH5hH8Z^HHH5&gH8Zf.AVAUIATIUHLwL_HHt?H@HHtHLLH]A\A]A^HEHH]A\A]A^@HLH8XHH]A\A]A^AUATUSH^HH9=IIH@H1E11LHLHDHHHTHI9~H[]A\A]DAWAVAUATUHSH(H$H|$(HL$@LD$0L$dH%(H$1H HHDŽ$HH IHDŽ$HHDŽ$HDŽ$HDŽ$HDŽ$H9Pr,L= M:-IL$IGH5k LHH(-HH*-I/([H$IH-HEHh^H$IH`.HyH5 H_ HCLM H=fJUr2LLHAI\M3L$H+sI/YHDŽ$Im2HmHDŽ$ IFH5 LHH2IL$M2H5 L9tIGH;4IO L-AIEL$I/L$d HDŽ$Iml HDŽ$EIFH5 LHH=IL$M4=IEH$LH@pH5H@H5IL$M=ImY2H5 HDŽ$L9tIGH;@I L$IEL$Im3HDŽ$H= H H9PDHm H$HFHH$H<$H5 HGHHFIL$MFH$HHD$HH-HDŽ$H+I9E)HH$H$DH$IHDŽ$H$MJGH//-H5 L9z-IEH;LZI}^-H$AHH$H$H$Im;HDŽ$H$HHD$HH<HDŽ$EgL%X L- ID$HHbnH=UbQO|1LLIXM|L$1LLImeHDŽ$E1AHD$HHD$xHD$(E1E1AHD$`E1HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$LQ[PHD$HDHHHt H;~HPHuLhL`HMtIEMtI$H3 H H9P4L= M,:IL$IGH5 LHH9H$H$H$H:I/{.IFH5 LHDŽ$HH:IM:H}H$I9G(M_M(MOIII/W0LLL\$LL$L\$LL$H$II+)Mt:I).H$H$H9P<H$LL$薺L$H$II(}.HDŽ$H$M<H//HDŽ$HDŽ$Ht H+V)Mt Im`)Mt I,$C)H5 LKPH$IH\=H;$L;- |u L;-/|'Im;/HDŽ$ H H9 H9P0)L- MTIEL$IEH5 LHHTH$H$H$HSImT8H$H$HDŽ$H9PVH$H$H$IHDŽ$MUH$H/P8H5. L9].IEH;R{^I}A.H$AHH$H$H$Im9HDŽ$H<$HHD$HH;HDŽ$E L% L-4 ID$HHpH=];Lo~1LLISM}L$1LfGImSeHD$HE1E1H$HD$xE11E1HD$(AAHDŽ$HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$bffA.GzH!yL-yE1HHTyH9yL$AI/H$LLHDŽ$ImLKLLHUH$IHHD$HE1E1H$HD$xE1E1IHD$(AAHD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$ffA.Gz ifDHw1L-wHH$H9wJDLhL`HQHD$HIH$E1HD$xE1E1AHD$(E1E1AHD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$@Ht H/Ht H+H$Ht H/;H$Ht H/tMt I,$Mt I+%Mt I*H eWDDE1H=/[HD$HHt!H;avtH@@ ((Hm4Mt I/uMt ImH $HtHHD$HHHH\$HtHH$HHHL$HtHH$HHHt$HtHH$HHH\$ HtHH$HHHL$xHtHH$HHHT$8HtHH$HHHt$XHtHH$HHH\$PHtHH$HHHL$@HtHH$HHHT$hHtHH$HHHt$pHtHH$HHH|$`HtHH$HH7H\$(HtHH$HH)I.H$HH$HHHL$0HtHH$HHH$dH+%(gH(L[]A\A]A^A_ÐHGLD$D$FD$D$=LL$D$D$L$FL$D$D$L$LD$D$L$`FD$D$L$L$D$D$L$FL$D$D$L$fDHL$D$D$L$EL$D$D$L$L$D$D$L${EL$D$D$L${fDL$D$D$L$+EL$D$D$L$BfDHDLDLDHDLD~H$DCfDLDfLDHpD.cDfDHPDH@D$H0D2H D@HDNHDHCHCHCHC,HC:HCHHCVHD$0H;tp!HIHD$H2H0 Hi H9P2L-P MC4IEL$IEH5 LHH4H$H$H$H4ImNH H5 HDŽ$H9p5L- Mi8IEIEH52 LHH98IMo7ImTHG H` H9PeL-G M%IIEIEH5 LHHHHD$H|$>HIm+H^nH$I9D$KHt$LvH$IHt$HHD$HH+MJI,$+IEH5* LHH]LHD$H|$KIm..H$H$HDŽ$H9PPHt$H$ͫH$IHt$HDŽ$HH$HH-MOH$H/.HD$0Hf HDŽ$HDŽ$H@H9t9HXH-HqH1@HH9H;TuH] H5f H9p\HM H$H`^HH$H<$H5 HGHH^IM\]H4$HHD$HH+JH|$0H5 HGHH]H$H$H$H_I|$A1H;$ cH;=lOH5,mH9tLL$ILL$tID$@=dL~DIHmH$HtICHDŽ$H4$HcH ItCHHITID$HDŽ$HHbH=NL$=L$TxL\$1LLHH$EHL\$xH$H$I+VI,$+OL$L;$L;%Kk.2L;%ik!2LcEÅidI,$OHDŽ$ZH H  H9HH H$HsHH<$H5 HGHH?IMH$HHD$HHvaH H ~ H9HHe H$HHH$H<$H5 L\$HGHHL\$HD$HD$H$HQH$HHD$HHdH|$0L\$H5 HGHHL\$H$H$H$H HD$H$H9PH4$H$L\$KL\$IH4$HHD$HHfHDŽ$MH$H/MfID$L\$LH5 HH9L\$H$H$H$HI,$eH$I9CH4$LL\$葦L\$HD$H$H$HHD$HHeH|$HDŽ$שI+kIEH|$L}HDŽ$H|$HHD$HH"AHDŽ$ZH<$H5F HGHHZHD$HD$H$HZLH9HD$H$H^Ht$HHD$ HHHH|$H$HDŽ$H9H;=e*H;=e*?ÅJ_HT$HHD$HHIHDŽ$`Ht$L H% H5 D$H9pbH߿ HD$HeHH$H|$H5I HGHHeIMdHT$HHD$HHnJD$L\$M;L\$HHD$H$eH$I9ClHt$LL\$ L\$ HD$H$HL$HHD$ HH JH|$HDŽ$VoI+OHT$H$H9AH;cDTBH;cGBH=Aą>lH|$HHD$HH-OHDŽ$E{uH HT H9XdvL; M>wIICL\$LH5 HHwL\$HD$HD$H$HwI+ \H|$H5[ HGHHwIMvHt$HHD$HH[H5 H<$1L\$6L\$HHD$H$߈H$I9C QMcMQI[I$HI+dHT$LH_IH$I,$HD$eHT$HHD$ HH\H|$HDŽ$I+\HT$H$H9AH;aD MH;aMH;AąH\$HHD$HHcHDŽ$ED$\,IfTdH1ٷH3E1HDŽ$A} HD$HAHD$xHD$(E1E1HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$sHD$HE1E1E1HD$xAm AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$r%HD$ H$鍶Ha龵H-B H5FC H9pHH-C HD$H EHH$H5> H|$"IHjDH|$HD$72L\$HHD$8H$CH$HT$8L\$HHBsL\$HHD$H$BHiA H rB H9HBHYB HD$HAHH$H5H@ H|$L\$ !L\$ HI1AH|$L\$ i1H5A H|$LHDŽ$YL\$ k@LL\$/1HT$Ht$8H|$"L\$HHD$ ?L1H|$80H|$HDŽ$0fHD$  'H*D$PYHDŽ$H@HD$pf/9?H,HH5*A HH HHH HHH HHH HHH HH H HD$hL`H? H9p>H@ HD$H>HH$H58< H|$n HD$8H$HS=H|$/LHD$H$H<HIH<HL? H5@ H9PL;H@ HD$H$\HH$H5= H|$L\$HL\$HHHo[H|$L\$HL/HL$8L\$HHDŽ$HGH9AH$ZH$HxH;=HD$D$@D`McYH55H9tL\$8L\$8tHD$H@@=WLL\$8~L\$8HI+WH$HtIBHDŽ$HcD$@1H|$HDŽ$LT$I|Hƍ@H|$HM\FLHI\LT$HI6VL!.H$@HD$8H|$ .H|$814L;%HDŽ$UH9 ID$H9UHXHYUHq1H;TkUHH9LL1Ҿ=LLD$$LD$HHIHl`IHaH$L$H$H$0H$H.HDŽ$H$ HtH$0HDŽ$8HDŽ$HDŽ$HDŽ$HDŽ$'H$H@HJH$HBHBHx\StHspX9HHHHW`L L9IHX~3HGxH8~HHIH9IHHI1HT$8H۾.2HH$@H$HHD$HHzGL+HD$(H5*8 H\HD$HFHD$(H57 H\H$HHFH$H9GFLGMtFHGILD$HH$g+H$Ht$H|$IH+MEH$2+LHDŽ$+HL$PIIHD$XLIHHkH)HHHD$pHD$HD$(HxH;\$XGHE111LT$@H|$8K_HT$hLT$@H|$8H!H9DHH#T$hI H1HuHHt$HHH56 LIHdH9 H=: H9xsdH: HD$(HdHH536 H|$(L$qL$HH$H$cH|$(L$)褸L$HIOcHD$ LT$(HID$(LT$(HHcH(9 H9 H9PbL9 MlbIH546 LL$L$L$L$HH$HD$(bLL$)HT$(H58 HL$aH|$(L$(H$HL]L$HHD$(H$baH$(LHDŽ$(H(LL$HH`HD$(L$HCҹL$HHD$(HH$`H$H55 -L$Y`HT$(LHL$L$HI`L'H'H|$('HD$(gH5z6 H|$(HD$(tHVDHD$HE1E11HD$xE1H$AHD$(AHD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ fHD$HE1E11HD$xAn E1H$HD$(AHD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$!fH5 3 H=6 1HD$HH$HaH1)H1&E1HDŽ$A[ HD$HAHD$xKH貯黽H襯鈽L蘯aHL$(H; OH;$ H; h I,$uL\H|$(JÅH$E1E11A{%ALE1HD$(eH 顼HD$HE1E11HD$xA- H$AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$`dHD$HE1H$E1HD$x1AAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ cHD$HE1E11HD$xE1H$AHD$(AHD$`HD$pHD$hHD$@HD$PHD$8HD$ ^cHD$HE1E1E1HD$xAAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$ cH|$HD$8|H$E1E1AHD$HAHD$xHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$0H$WbH$E1E1AHD$HAHD$xHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$0H$aH<$裶ImH=d1 _H$H<$vmHD$HE1E11HD$xE1H$AHD$(AHD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$`HD$HE1H$E1HD$x1AAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$/`H|$5H=/ H1 H51 H$H|$0L\$H$lHD$HE1H$E1HD$x1AAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$H$G_H<$NL\$HD$kL$$E11AHD$HAH$HD$xHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$^H=`. L\$VL\$H$H$H$HjHD$HE1H$E1HD$x1AAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$]H=- H{/ L\$H5w/ L\$H$HD$HE1E11HD$xA' H$AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$\HD$HE1H$E1HD$x1A AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$i\HD$HE1H$E1HD$x1A AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$[H|$(ٰI;HD$HE1E1AHD$xAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$H$K[HXHH\hH@HL\$HH$OH$H$HH+L\$I6hHL\$ܤL\$hH$E1E1E1HD$HA1AHD$xHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$0H$@ZHD$HE1H$1HD$xAAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$YHD$HE1A* AHD$xHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$$YL\$H$fHD$HE1H$1HD$xAAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$OXHD$HE1H$E1HD$x1A% AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$ HD$HD$WHD$HE1E1AHD$xAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$H$LWH$E1AE1HD$HE11H$HD$0AHD$xHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H${VHD$HE1H$E1HD$x1AAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8VHD$HMH$E1HD$x1AAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$H$YUH=2% H& H5& IMWHD$HE1H$1HD$xA( AHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$T諩L\$ HD$߫H=e$ `IL $#L $krHD$HE1E11HD$xE1H$AHD$(AHD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$H$i=HH5H8L\$H$E1E1E1HD$H1H$AVHD$xAHD$(HD$`HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$HD$0H$HPHrH5rWL\$H81PL\$HD$HE1E1Af#HD$xA*HDŽ$HD$(HD$pHD$hHD$@HD$PHD$XHD$8HD$HD$VHD$HE1H$E1HD$x1Ad#A*HD$(HD$pHD$hHD$@HD$PHD$XHD$8HD$HD$HD$HE1E11HD$xE1H$AN#HD$(A&HD$pHD$hHD$@HD$PHD$XHD$8HD$6A#H|$L$E11D$aH$E1HD$HHD$xA&D$HD$(L$HD$pHD$hHD$@HD$PHD$XHD$8HD$HD$LH转IH$IA"/HD$HE1E11HD$xE1H$A"HD$(A&HD$pHD$hHD$@HD$PHD$XHD$8HD$HD$HE1E11HD$xE1H$A"HD$(A$HD$pHD$hHD$@HD$PHD$XHD$8HD$HD$ZHD$HE1E11HD$xA"E1H$HD$(A"HD$pHD$hHD$@HD$PHD$XHD$8HD$ HD$HD$f.AWAVAUIATUHSHHhL%nL5ndH%(HD$X1L L~HD$ HD$HLd$(Lt$0Ld$8LL$@HIQHUJcH>fHn@Hl$HLK8LL$@LC0LD$8HK(HL$0HS HT$(L{L|$ HL9H;-muuL9tpHLL$HL$LD$H$GH$LD$HL$LL$tsHLLSY^HT$XdH+%(KHh[]A\A]A^A_MCMLL?fML)DEH$LD$HHL$LL$otVL{MHMH MHNHhnL {oLNLDHHulAWH#NH5}NH81IXZH sMVH=AQܼ1DItHSJcH>DHF@HD$HHC8HD$@HC0HD$8HC(HD$0HC HHD$(HCHD$ =HIH~SJcH>@Hf.AVAUIATUSHH H L%jdH%(HD$1HU HnH$HD$Ld$MHHHMLNHHHsH  H=AHEATjQRLj5 PjQH IHEHPMHHEtLHD$dH+%(H L[]A\A]A^HLV(HF UfH;fDIM5DHF M HH IHIIHHI?IAHHVhHPH5`JUL 2kH81EX)ZH NI H=PE1贸H IAfH^HHIHFLH$8IM~TH5V LHVAHtHD$IM~.H5 LHV@H HD$IML $HD$LT$H H.HF(LHD$HF HD$HFH$T8IHHEt&H Hk)H=O職@H :fDL8IHSH5U LHV@HH$IHF LHD$HFH$7I1HL OIH5sL8):fAWAVAUATUHSHH(L5YfLfdH%(HD$1H H$Lt$HD$HcIQI7IMHFH FHNAL hHkgELNODHH~eH.NATH5GH81BX=+ZH |FH=:NE1HD$dH+%(H(L[]A\A]A^A_@LLMHHHsH=kAHEHAVj5 5 j5u Pj5d  IHEHPMHHEbH7UHV(HE nLIIKIMIHFHH$5IMjL $HD$HT$IHF(HHD$HF HD$HFH$45IM~1HL LMH5pL~6y*+cHHEtFH Df+H=L9R@H DAL nfH6fDH4H5 LIHVIX+ZH *BH=JE1萱*LH(4II[qM IHFHH$1IH5 LHV9HD$H3IML $HD$HT$,f.IHF(HHD$HF HD$HFH$d1IM~1HL HMH5lL2y+HHEt&H Am+H=Hi@H3fDH @A`fDH5ٸ LHV8H`HD$IMfDHF HHD$HFH$0IHH<_H E@H5FAjL bAH#GH81fHF0HD$HC(HD$HC HHD$HCH$5.IItlI~.IML $HD$HL$HT$M*IH5 HHV6HD$HqIH5ٹ HHV5HD$HIzfDHHEH a=,H=jEŬH5a HHV}5HtHD$IM#1HL CMH5hH.R,fDH/qHFHH$,IH,IH5 HIHV4H$HLcHHM[H V<H5W=jL (^AH)CH818Y^H,HH[AH5=jL ]H ;H8HB1j8_B,AXH/AWAVAUATUHSHH(L5ZLndH%(HD$1H$Lt$HGI5IMHi;H k;HNH\L ,]LNL@HH'ZAUH6BH5/<H817X,ZH %;H=cCE1苪HD$dH+%(H(L[]A\A]A^A_DLV LMHHz HHsHc H=TAHEHAVjRPjRLPj5 C IHEHPMHHE^H,Q@MtIIIMHH*H5Q LIHVI^2H$HML $LT$f.HHEt&H 9-H=B9@H+fDHF HHD$HFH$)H~1HL n@MH5dL*d,,fHFHH$`)I7H5Q LHVm1HtHD$IGLm~\,ff.AWAVAUATUHSHH(L5WLndH%(HD$1H$HD$Lt$H~I\IHV(HE LMHHHsH=AHEHAVj5 5 j5z Pj5a  IHEHPMHHEHD$dH+%(H(L[]A\A]A^A_H(H5 LIHVI0H$HLmDIH 7H7AHMEIHHcVHi>H5m8AUL >YH813Xy-ZH Z7H=?E1*LHX)II[qM IHFHH$'IH5$ LHV(/HD$H3IML $HD$HT$,f.IHF(HHD$HF HD$HFH$&IM~1HL M=MH5aL'yg-HHEt&H 56O-H=>虥@H8(fDH 5A`fDH5 LHV%.H`HD$IMfDHF HHD$HFH$%IHHlTH u5H5v6jL GWAHU<H811Y^]-(H 65Aff.AVAUATUSHHL->THndH%(HD$1L,$H5H$HHH4H 4HHHHVH?L mUHLHL@HHSUH;H55H810X-ZH 4TH='=E1HD$dH+%(lHL[]A\A]A^HVHHHsE1Hӫ L ī HEH=HAUjPAQjPAQjP IHEHPMt`HHEsH&fDLIHHHFHH$#HGH$P@HHEt&H }3.H=<@H%1HL D:IH5h^L$y-fH8#IHwH5- LHVI+HtH$IFMB&fAWAVAUATUHSHH(L5QLndH%(HD$1H$Lt$HGI5IMHi2H k2HNHSL ,TLNL@HH'QAUHP9H5/3H81.Xx.ZH %2H=;E1苡HD$dH+%(H(L[]A\A]A^A_DLV LMHHz HHsH H=dAHEHAVjRPjRLPj5 C IHEHPMHHE^H#Q@MtIIIMHH!H5Q LIHVI^)H$HML $LT$f.HHEt&H 0.H=99@H"fDHF HHD$HFH$ H~1HL 7MH5|[L!dh.,fHFHH$` I7H5Q LHVm(HtHD$IGLm_#ff.@AWAVAUATUHSHH(L5NLndH%(HD$1H$HD$Lt$H~I\IHV(HE LMHHHsH=8AHEHAVj5 5 j5R Pj5A  IHEHPMHHEHD$dH+%(H(L[]A\A]A^A_HH5 LIHVI'H$HLmDIH .H.AHMEIHHcMH5H5m/AUL >PH81*X/ZH Z.H=h7E1*LHX II[qM IHFHH$IH5 LHV(&HD$H3IML $HD$HT$,f.IHF(HHD$HF HD$HFH$IM~1HL }4MH5FXLy.HHEt&H 5-M0/H=>6虜@H8fDH ,A`fDH5 LHV%%H`HD$IMfDHF HHD$HFH$IHHlKH u,H5v-jL GNAH3H81(Y^.H 6,Aff.AWAVAUATUHSHH(L59KLndH%(HD$1H$Lt$HGI5IMH+H +HNHiLL |MLNL@HHwJAUH2H5,H81'X/ZH u+RH=4E1ۚHD$dH+%(H(L[]A\A]A^A_DLV LMHHʢ HHsH H=AHEHAVjRPjRLPj5y  IHEHPMHHE^HQ@MtIIIMHH59 LIHVI"H$HML $LT$f.HHEt&H %*/H=^3艙@H(fDHF HHD$HFH$H~1HL 0MH5TLDd{/,fHFHH$I7H5 LHV!HtHD$IGLm~ff.AWAVAUATUHSHH(L5IHLndH%(HD$1H$Lt$HGI5IMH(H (HNHyIL JLNL@HHGAUH/H5)H81$X0ZH (H=1E1HD$dH+%(H(L[]A\A]A^A_DLV LMHHڟ HHsHß H=AHEHAVjRPjRLPj5  IHEHPMHHE^HQ@MtIIIMHH5I LIHVIH$HML $LT$f.HHEt&H 5'80H=0虖@H8fDHF HHD$HFH$H~1HL .MH5|QLTd/,fHFHH$I7H5 LHVHtHD$IGLm~ff.AWAVAUATUHSHH(L5YELndH%(HD$1H$Lt$HGI5IMH%H %HNHFL GLNL@HHDAUH,H5&H81 "X0ZH %H=3/E1HD$dH+%(H(L[]A\A]A^A_DLV LMHH HHsHӜ H=dAHEHAVjRPjRLPj5  IHEHPMHHE^HQ@MtIIIMHH5Y LIHVIH$HML $LT$f.HHEt&H E$w0H=-詓@HHfDHF HHD$HFH$H~1HL +MH5lNLdd0,fHFHH$I7H5 LHVHtHD$IGLm~ff.AVAUIATUSHH H L%dBdH%(HD$1H HnH$HD$Ld$MHHHMLNHHHsH=wAHEHATj5b RLj5& Pj5 O IHEHPMHHEtIHD$dH+%(H L[]A\A]A^DHLV(HF UfHXfDIM5DHF M HH !H!IHHI?IAHH@H )H5"UL CH81&X(1ZH !|H=t+E1H z!AfH^HHIHFLH$]IM~TH5 LHVrHtHD$IM~.H50 LHVLH HD$IML $HD$LT$Ht H.HF(LHD$HF HD$HFH$IHHEt&H } Q1H=>*@HfDL`IHSH5 LHVqHH$IHF LHD$HFH$I1HL 'IH5AJLY1fAVAUIATUSHH H; L%>dH%(HD$1H HnH$HD$Ld$MHHHMLNHHHsH=tAHEHATj5 RLj5v Pj5  IHEHPMHHEtIHD$dH+%(H L[]A\A]A^DHLV(HF UfHfDIM5DHF M HH CH3IHHI?IAHH=Hb%H5UL ?H81vX1ZH H='E1dH AfH^HHIHFLH$ IM~TH5 LHVHtHD$IM~.H5 LHVH HD$IML $HD$LT$HĔ H.HF(LHD$HF HD$HFH$ IHHEt&H C 1H=&1@HfDL IHSH5 LHVHH$IHF LHD$HFH$_ I1HL n#IH5qFL 1bfAVAUIATUSHH H L%;dH%(HD$1HU HnH$HD$Ld$MHHHMLNHHHsH=jqAHEHATj5 RLj5ƒ Pj5=  IHEHPMHHEtIHD$dH+%(H L[]A\A]A^DHLV(HF UfH fDIM5DHF M HH HIHHI?IAHHV9H!H5`UL 2<H81XR2ZH NH H=t$E1贉H AfH^HHIHFLH$ IM~TH5V LHVHtHD$IM~.H5Б LHVH HD$IML $HD$LT$H H.HF(LHD$HF HD$HFH$T IHHEt&H  {2H=>#聈@H fDL IHSH5U LHVHH$IHF LHD$HFH$I1HL IH5BL >2 fAVAUIATUSHH Hۏ L%T7dH%(HD$1H HnH$HD$Ld$MHHHMLNHHHsH= AHEHATj5R RLj5~ Pj5m ? IHEHPMHHEtIHD$dH+%(H L[]A\A]A^DHLV(HF UfHH fDIM5DHF M HH HIHHI?IAHH5HH5UL 8H81X2ZH  H= E1H jAfH^HHIHFLH$MIM~TH5 LHVbHtHD$IM~.H5 LHV<H HD$IML $HD$LT$Hd H.HF(LHD$HF HD$HFH$IHHEt&H m 3H=ф@HpfDLPIHSH5 LHVa HH$IHF LHD$HFH$I1HL IH5>LI2fAVAUATUSHH L  L-3dH%(HD$1HnL $Ll$HQH?H%HHH+IH H~HIHH2I?UIH5H8L 5A1^Xo3ZH  H=lE1LHD$dH+%(H L[]A\A]A^MHH? HHsH( H= AHEHAUjRPjRLPj5֊  IHEHPM|HHEbHVULV LNvMIHHteHHIHL $LT$%fHHEH H 3H=2HF HHD$HFH$H~1HL IH5L<Ly^3_fDHFHH$PIMTH5E LHVa HtHD$IF느H(AH5q LHV- HtH$I)fAWAVAUATUHSHH(L50LndH%(HD$1H$HD$Lt$H~I\IHV(HE LMHHHsH=AHEHAVj5Ј 5ˆ j5 Pj5  IHEHPMHHEHD$dH+%(H(L[]A\A]A^A_HH5 LIHVIH$HLmDIH eHgAHMEIHH3/HH5=AUL 2H81 X3ZH *M H=E1*LH(II[qM IHFHH$IH5< LHVHD$H3IML $HD$HT$,f.IHF(HHD$HF HD$HFH$dIM~1HL MH58Ly3HHEt&H  '4H=i~@HfDH A`fDH5ن LHVH`HD$IMfDHF HHD$HFH$IHH<-H EH5FjL 0AHH81 Y^3}H Aff.AWAVAUATUSHXL- L5p H|$HndH%(HD$H1H,Ll$0Lt$8HD$@HHNH'H1 H,HD$LnH H(hE111HALIHk H8yH‚ H(hE111HALHH H8eEA;D$7H H q H9H LX M IIBLT$LH5P HH LT$HIHH IHH{E1E1AH;=*m H;=+pH5+H9tLD$JLD$tHSBፁ LLD$<LD$HIK MtL@IcHEAIlHCMcI$OdLM8 H=i 1LHAI MhI/H+|H IL0IH H8bHC HĈ H9X L M" IIBLT$LH5 HH LT$IM I*Hׂ HH H9X H/ H HHCH5o HHHd IM~ H+H(I9B LLLT$gLT$HHJI*H(I9GHLfIH+IHMIHL;,)L;(u L;(~I*1 iHL$H] MHAH=VLHqIL5(jP5 j5( AUjPHT$XLT$` HPLT$HI&I* I,$MHmMt Im#Mt I.HD$HdH+%(GHXL[]A\A]A^A_fDHfHF(Lv HD$C@HzLf.D$zLf.jz\D$He H D$ H9XL M IIBLT$(LH5 HH LT$(IIHM] IHID$ IH HS&I9G LLLD$(qdLD$(II(?M& I/L;-&L;-s&u L;-&GIm Lt$D$MIHHD$ IHSH~ LILH5!&AHH=8TjP5~ j5 AUjPHT$X~ IHPH I/=H+ImE1E1QDHEA;D$DH%HD$Hq%Lv HD$L`'HPE1LLD$8Ld$HDLLiLHLD$0L)LD$Ht0Hl$8Ld$@LD$'HIMI(LfDH ZAHH0$H H5:UL 'H81X&ZH (]H=E1tfIHHTHHFHHD$0IM~TH5| LHVHtHD$8IM~.H5| LHVH HD$@IM HD$@Ll$0Lt$8HD$HH CH3IHHI?IA@HHF(HHD$@HF HD$8HFHD$0IfDHwHE1E11E1E1Һ&E1MtI*tJMtI(tgMtI/t|H H=vsMuI,$UE1bDLLD$T$t${LD$T$t$@LljT$t$XT$t$|LT$t$8T$t$gE1E1E1E1Һ&%DLHLT$LT$=L-LLKHLT$LT$ZfLLLT$sLT$i (E1jfLHÅM2'E11E1E1fDI/B LT$t$LT$L|$IE1t$T$fH E1E1&E1E1D$ UD$ HE1E1'E1E1H=Ay H~ H5~ cIME1E1E1'Rf.LLT$#LT$fLGH=x HZ~ H5[~ VcIME1E1E1'fLL'ME1E1E1I(tE1E1fLljT$E1t$LT$Xt$T$E1LT$^H= x YII@#LT$H`fDLCMLkIIEH+I}LAAW@'MIE1E1%1LHKIHE1E1'SLHD$cLD$fLD$01L)Hl$8Ld$@Ht0LR uH{LD$LALD$Iƾ'M?-E1E1Һ'LLD$E1E1LD$OHLD$LD$LHD$LD$fH=Iv H{ H5{ `IME1'`;LT$IfDH=u VIMкE1Ҿ'H=u H2{ H53{ N`HHE1Ҿ'HIHH5t LHVHHD$0IfHF HHD$8HFHD$0VIfDSIH=u VHY@Iؾ'I/LLD$ T$t$LT$LT$t$T$LD$ ^H=t UIq@'#fLT$(IifDMBMIJIHI*LLHLD$ HL$=[LD$ LT$HI(LLT$9LT$'f'E1E1E1E1E1M 'MGMlIOIHI/@LHHLD$ HL$uZLD$ L|$II(5LLT$qLT$1۾(/IHHxH5H8E1Ҿ'MwM I_IHI/LLHLD$(YI.LD$(IILLD$(LD$(fDE1E1M/'L= w H5v IGHHhH=Ht$Ht$1LHD$LT$M1LLT$BLT$I*(E1E1fDL-qv L5rv IEHHH=O*1LLHD$LT$Mj1LLT$LT$I*ZE1E1B'E1E1o6(YE1E1E1_'HLD$LA I}LD$AyfDE1E1i'MIE1E1s'1HL$0ILL H5%&5LHL$ LD$|LD$HL$ LHL$ LD$[LD$HL$ LLD$(?LD$(PL-E1E1Ҿ(0LL1ALD$I1LIHE1(1LLqIH%E1E1E1>'(E1E1E1E1>'E1E1LT$HzHD$HH5H8LT$LD$(>LT$H^HD$HOE1E1H5H8LT$LD$>'ME1IE1E1E1IںAWAVAUATUHSHhH^H|$dH%(HD$X1HsHD$0HD$8HD$@HD$HHH HN HF0HD$H](Le HmHk L(hE111HAHAIH6H8 HEk L(hE111HALAIH H8O Hk L(hE111HAHAIHDH8 AFA;E Hm H >r H9HH-%r HHEHEH5p HHHIMHmHl H q H9HH-q HHEHEH5 p HHHIHEHMHEHJI{1E1H;=H;=pBH5H9LT$ L\$#L\$LT$ HLT$ L\$L\$LT$ HIMtLPHcIEMlICHcIMtHHpH=ILL$ L\$zL\$LL$ 1LLHHL\$LL$ I)I+ HI9D$MHLOIHm MI,$ L;)L; L; LLT$LT$Hc؅ I* xHj H o H9HL%o MxI$ID$H5n LHHDHHI,$ H4j H %o H9HsL% o MI$ID$H5m LHHnII$HMCI$H IyE1H;=>AH;= H5NH9tLL$ L\$L\$LL$ tIAPуፁ+LLL$ L\$L\$LL$ HI[MtLXHcIMtIAHcIM|HHxH=LT$ LL$LL$LT$ 1LLIBMLL$LT$ I*I) HH9ELH?MII,$ MHmL;L;D L;b LLT$WLT$I*.Hh H l H9HsH-l HHEHEH5vk HHHIM}HmHg H tl H9H H-[l HHEHEH5k HHHPIM.Hm^IzE1H;= H;=fH5H9tLT$ LL$LL$LT$ tIBPуፁHLT$ LL$LL$LT$ HIMtLHHcIEMlIBHcIM|HHH=/LT$ L\$`L\$LT$ dL\$ 1LLLT$HHLT$L\$ II+I*pH I9D$HLJIHm.MI,$L; L; ` L; S LLT$LT$I*uHD$HML e H=CLHpILjAQAWjAQAVjHT$PLT$Hh H@LT$HHI*L4LIH5h LIHV&HD$0HH]@HHH HNHSH L@Hm H5~L QH81X4ZH m H={E1[HD$XdH+%(HhL[]A\A]A^A_DHI HD$H8KHL\$#L\$fHLLT$0H)LT$ 5Ht0L\$Ll$8Lt$@L\$LT$ HHMRI*HLL\$L\$1f.ISBፁWLT$01H)Ll$8Lt$@Ht0LB uI{LT$ L\$HALT$ L\$HŻ5HWMI*1E1ɺ I+Mt I,$Ht HmMt I)H H=YMu7Mt I.(M I/L}ImE1HXoHHA;GHf.f(zHL$ f.L$ D$zLL$ f.L$ f(zf/f/T$ f.L$z H\$f(T$ LIL\$L\$HIw T$ HD$(f(L\$LL$(HI= D$LL$ L\$LL$ HH H` H5LjHAPHH=?UjPATjPHT$`LL$hL\$XV` HPL\$LL$HHD I+I)I,$eHmJImt5I.IfDLhHX1LICfDVL߉T$LL$T$LL$ LT$LL$T$LL$HT$LL$LL$T$LωT$T$HIHDHcH>f.HF0HD$HHE(HD$@HE LHD$8HEHD$0TIHtsH~5H|MRHD$HHl$0Ld$8H\$@HD$@HH>H5b LHVHD$8H0IH5yb LHVHD$@H IsfDE1E1E1ɺ 4E1E1ɺ 4HLT${LT$=Lh E11L׉T$LL$L\$@L\$LL$T$M.3@E1ɺ 4>fDLLT$LT$H5] LHVHtHD$HIM1HL$0ILL H5 \4@MSM$I[IHI+ H{I۽HFLHD$0/I 6E1MJI*@L׉T$LL$T$LL$!LL`LL\$L\$@fH=i\ Ha H5a FIMtLWfDE1ɺ 5fDE1ɺ 5efDH= \ =I#ILLL$ LL$ H=[ H` H5` VFHHE1ɺ 5LLL\$0H)L\$ Q6Ht0LL$Lt$8L|$@zLL$L\$ HIpMI+LLL$TLL$uf.H8L(LsLLL$LL$OfLLT$LT$8fLH=Z ;H@IIE11 5I*E1@HLT$cLT$f1LLLL$ L\$L\$LL$ HH 6@LI\$HIL$HHI,$cHHHHL$@H+Ld$IvHLT$LT$_fDE11 6@D$L$HD 5E1t@L$ L$ H> 5E1D@L$ D$(L$ T$(H* #5E1H-9Y L%R\ HEHH H=7r 1LHIM 1LI,$B ;5E1H-X L%[ HEHH^ H= 1LHIXM 1L%I,$! &6E1HH-)X L%:[ HEHH9 H='b 1LHIM 1LI,$ [5E1xLLL$3LL$TfH=V H[ H5[ vAIM_E1ɺ 865fDH-9W L%BZ HEHH H=7r 1LHIM* 1LI,$ {5E1E1ɺ :6MfD3HH=U 6I@HLT$LT$f1E1 5@H=U HZ H5Z .@IM|E1ɺ =6 61HcӸLLL$0H)HLL$ 6Ht0LT$Ll$8L|$@?LT$LL$ HHMI)LLT$LT$f.1 5I)E1HLT$LT$LM E1?6IH=eT `5ILL$6L\$$L\$LL$ HaHH5T$H8L\$LL$T$4 5,LLT$LT$ _6 ME1MYMIYIHI)=H{IA1LLLT$ LL$MLL$LT$ HI j6 5mL\$01H)Lt$8L|$@Ht0H@ uIyLL$ L\$# LLL$ L\$IĻY6MLDME1E1ɺ |6H]H HMHHHm(LHHHL$}9H+Hl$IHLT$~LT$ 6E1?LLT$PH{ILT$H-R L%U HEHHpH=71LHIBM.1LI,$ 6E1LLT$LT$fH=tQ HUV H5VV <IMuYE1ɺ 6LHL$jHL$E1ɺ 6mII[H= Q 2ILL;5 E1>HHH H5jL AH"H81Y^z4+L&6 E1H=xP HIU H5JU ;HHE1ɺ 6 j6E1L[5X E1{I 16F2IH=O 0HLT$j6LL$LL$LT$ HgH.H5T$H8cLL$LT$T$:L{5 E11LLL\$ LT$?LT$L\$ HH*MѺ 6sMJMIZIHI*H{IڽHcLL$01H)Ll$8L|$@Ht0H@ uIzLT$ LL$HLT$ LL$Hź 6H6HHaAH5ljL =H TH8H1_t4AXE11 6qI\$H+IL$HHI,$%HHHHL$ 5H+Ld$IHLT$ LT$L1HIHhE1ɺ 75DL1HsIHE1ɺ "6D 6E1kLL\$A{H{IL\$L1HIHE1ɺ W5jf.H-M L%P HEHHH=1LHI8M1LI,$ 6E1L1H[IH E1ɺ w5DHHL${HL$E1ɺ 7<1 6H1AL\$LT$ H L绎6' E1JMѺ 61L\$6LT$LL$L\$ HHH5|T$H8PL\$LL$T$ 75E1 W5E1475E1ɺ HHH5H8E1ɺ { "6E1f"6E1ɺ HKH\H5H8E1ɺ ( w5E1W5E1ɺ HH H5mH8EE1ɺ L1HKIHE1ɺ 6LLL$kH{ILL$w5E1ɺ HjH{H5H8E1ɺ GLHL$ HL$L1L\$LL$ IL6 E1 6E1fHHH5O6H8"E1ɺ L1H(IHE1ɺ 6H1LL$LT$ Hh 6E1ZHuHaH56H8E1ɺ (E11 fAWAVAUATIUSH8HL%M dH%(HD$(1HnH\$LD$H\$ H1HgqHHIHFHD$H=;H L,HHHH H yM H9HL5`M MGIIFH5G LHH=IIHMIH,LHIH I/fL;5L;5mu I9DI.wEM$HT$It$MH=ILG H<I.Hm7HT$(dH+%( H8[]A\A]A^A_@HVLn(LF fH\$IHF H L H9HL=K M`IIGH5aF LHHIIHMIHLHEIH!I.L;=XL;=SI9JL"AƅI/EM$HT$It$MH=ILE H:I/LHD$軿HD$oL訿LADžA(AI.H FDDH=<1LHLHD$3HD$fLF I@L|HHD$HD$fDHH HIHHI?IAHHYHH5cUL 5H81X(ZH QH=;1,IHHHcHFHHD$IM~TH5 D LHV.HtHD$IM~.H5C LHVHXHD$ IMFHD$LD$Ll$ HD$R@H rAfDHuHF(HHD$ HF HD$HFHD$_If.LXH (H=X:1@H=B HRH H5SH ~-IMAA(zA(A^H=B #IIL証.AA(I/L肼DLpL`NH=!B #IMAA)AA)f.A(A~H=A HG H5G N,IfAA)+IH萹IHH5A LHVHWHD$IDf.HF HHD$HFHD$6I>fDA)AH=yE H9HEt H;lH>IMH={@ Lc%IHI.tl1LA A5)۴I/*@AA)1HL$ILL H5ǹ(LMH裿IPA A.)A0)A D@AWAVAUATIUHSHHHH5:D dH%(HD$81LuHD$H\$Ht$ H\$(HIIIIHE H$HEHD$H=V? $HHH6? H wD H9HL5^D MIIFH5> LHHIIHMIHjLHʹIHI/L;5L;5u I93DI.EH> M$HIt$H= H=AILAUjRPjRPj5B LL$XHT$P= HPH:I.Hm=HT$8dH+%(+HH[]A\A]A^A_fILm0Hu([fH$IN@H= H B H9HL=B MIIGH5Q= LHHIIHMIH%LH5IHI.7L;=HL;=SI9JLAƅI/ELM$It$L (A H=ILAUjLD$HT$@ Y^HI/yLH$蝶H$d@L舶LxADžAo*AI.gH &DDH=31L(OLH$H$Hu(In@LFHH$ܵH$DI&IH`JcH>fHE0HD$(HE(HD$ HE LHD$HEHD$dIII~aIu&M~jH57; LHVcHtHD$(IM~D1HL$MLL zH5py &*fMtFIuMTHD$Ht$ Ll$(HD$HD$H$Mf.L訲IH58 LIHV辺HD$HuLuMHFH HHNAL HELNOD@HHHAVH5 H81wX<*ZH qH=uh11H AL f.LH \*H= 11@H=9 H> H5> $IMYAAh**Aj*AH=19 ,IKI-L8AAm*I/LDLLH=8 IM,AA*hHELHD$藰I8H58 LHV襸HtHD$IMH5W8 LHV{HHD$ I@AA* H=7 H"= H5#= ~"I-fDA*AIZAA*A*ANH=; H9HEt H;umHʼIMtjH= 7 LIHtdI.t61LAA*oI/AA*LHmIAA*A*AAVAUATIUHSHH HLndH%(HD$1H$HD$HT$M=IIIH .H0AHMEIHHHH5AUL H81kX;ZH  H=qE1Y-HD$dH+%(H L[]A\A]A^DHV(HM HEHHHsE1H=AHEj5(5 55 j5< QHj5< P< IHEHPMHHE\H]OIlMSIHFLH$IH53 LHV+HD$HnIMH$HL$HT$ DIHF(LHD$HF HD$HFH$蜬IM~1HL hMH5Ly;THHEt&H 5 ;H=+;@H8fDH AfDH5 4 LHV%HXHD$IEfDLЫH5; LIHVIH$HLm7HF LHD$HFH$臫IHH4H =H5>jL AH0H81薷Y^;&uH AAUATIUSHH(H HndH%(HD$H\2 HT$H$M:HHHHHIH yHHIHHSI?UIH5\H8L ,A1迶X'<ZH G H=E1)HD$dH+%(H(L[]A\A]HV HFHHLSE1H51 H 1 AHEH=EjVQjVLQHj 58 P8 IHEHPMtaHHEfH軫YfDHHt`HLIHH$HT$BfHHEH 1 P<H=(HF LHD$HFH$H~1HL IH5\Ld|<zfHFLH$ШIMTH50 LHVHtHD$IE느H訪AH57 LHV議HtH$I詫fAVAUATIUHSHH HHLndH%(HD$1H$HT$M>IIMHȷH ʷHNHxL LNL@HHAUHH5H81X<ZH  H=jE1&HD$dH+%(H L[]A\A]A^fDHV H}HHHsE1H . H. AHEjQPjQHPj5. WH=b$6 IHPHEMHHE`HSDIIML賦H5T. LIHVIɮH$HMH<$HT$DHHEt&H Ee <H=&%@HHfDHF LHD$HFH$H~1HL MH5<Ldi<=fHFLH$ХI<H5- LHVݭHtHD$IFLm̨ff.AVAUATIUHSHH HhLndH%(HD$1H$HT$M>IIMHH HNHL LNL@HHAUHH5H81X/=ZH j H=E1 $HD$dH+%(H L[]A\A]A^fDHV H}HHHsE1H + H+ AHEjQPjQHPj5z3 WH=D3 IHPHEMHHE`HSDIIMLӣH5) LIHVIH$HMH<$HT$DHHEt&H e X=H=n"@HhfDHF LHD$HFH$7H~1HL NMH5<L脤i==fHFLH$I<H5* LHVHtHD$IFLmff.AVAUATIUHSHH HLndH%(HD$1H$HT$M>IIMHH HNHL LNL@HHAUHH5βH81LL$ 1LLLT$H躟HLT$LL$ #I)I*HI9D$oHLHH+H=I,$uH;-H;- H;- HԟÅHmHD$HML T% H=UHHpHEHj5B% AWj50% AVjHT$P% HH@HTE1' ?LHL$3HL$fHHDA;GHE= LEI@H HHcH>f.DE@ID$ Il$HEHH HԳHcH>fAl$HC LKIAH HHcH>@DK@Iɚ; Hɚ; I(L9 H\$LLL$ LILT$ϚLT$HHD$H贚LT$L\$HIH|$ 蔚LT$L\$HH^HHLE1jH=SHA5(# Pj5# ATj5# ASHT$`L\$hLT$XF# HPLT$L\$HHOI** I+` I,$= HmImtEI.It]f.I/+LΕfH踕91LHL$衕Ll$@MtI.u L腕DMDHNIHܱHcH>fHF0HD$XHE(HD$PHE LHD$HHEHD$@IHtsH~5HDMHD$XHl$@Ld$HH\$PHD$@HHH5! LHVښHD$HHIH5! LHV赚HD$PH'IsfDE1E111E1 >E1E11Mt I,$Ht Hm(Ht H)YHtH+tnMt I(MtI)t4H ӡH=FAMPIm-E1=DLωt$T$t$T$fDHLD$ LL$t$T$薓LD$ LL$t$T$aLLL$t$T$cLL$t$T$BLLD$(HL$ LL$t$T$1LD$(HL$ LL$t$T$DHLD$(HL$ LL$t$T$LD$(HL$ LL$t$T$DHLD$ LL$t$T$趒LD$ LL$t$T$vH=i H;=H5S WHFt H;ٿ=讝HH9H= HHL$HL$HItH)1LII,$! 5?11E1HE111E1 >g11E1 >Jf.DEEII IILD$+LD$H B>11E1DEEII DDEAMc@Al$HcHLD$躖LD$H L>11E1fDAl$AD$HH Hf.Al$AD$HH Bf.DKCII II]LL$ LD$&LD$LL$ H; V>11E1DDKCII DDKAMc@HHL$ HL$ D$ !?1E1fDH5 LHVHtHD$XIM1HL$@ILL H5=@HLD$(LT$ vLD$(LT$ LLLT$(H)LD$ Ht@H\$@Ll$HLt$PE1LD$ LT$(HIþe?$ tuHwH+mHL\$0LD$(LT$ LT$ LD$(L\$0BfHFLHD$@诌IE1E11E1侶> I*1E1LL\$(LD$ LL$t$T$mL\$(T$t$LL$MLD$ I+LLD$(HL$ LL$t$T$T$t$LL$HL$ LD$(HHL$HL$fL؍ Hȍ H= H H5 IMtLWfD11 >DLHL$kHL$H=) $ICI1ɺ >NHHL$HL$fLL\$HD$ތL\$HL$LHL$軌HL$LHL$裌HL$fHLT$ 背LT$ -f1ɺ >HPHILLD$LD$HMD$M11E19fL舉~H- L%: HEHHH="{ 1LHI舏M 1LUI,$- >11E1@H4H= H H5 FIM11 ?WH=y H H5 IMu11$ G?Lv1ɺ ?1$ I?;HYH= II@HL$ I9fDH= IY@1> IE1E11HIw3LL\$0LD$(LT$ 衇L\$0LD$(LT$ D1ɺ ?H=9 HJ H5K HH1ɺ$ L?f> E$ N?IE1E111IE1 ?H= H낾> LLT$蹆LT$bLEM H]IHHm,LLHLD$XLD$HHI(LHL$VHL$H= H LT$ H5 LT$ IMuSE11۾Q?$ qLT$ IE1ɾS?$ MH= LT$ LT$ ILLLD$踅LD$IXHIhHHEI(H}IA1LLLT$0LL$(LD$ LD$ LL$(HLT$0I1۾~?$ H\$@1H)Ll$HLt$PHt@H@ uIxLD$(LT$ LLD$(LT$ IE1ɾm?$ M%MBMFIZIHI*/D$H{Iڽ1LLLL$ LT$LT$LL$ HHE1?$ Hct$LD$@1L\$HH)L|$PH@Ht@ uIzL\$(LD$ LT$HL\$(LD$ LT$Hþ?H1ɺ$ ?bMD$MIL$IHI,$LHHLD$ HL$dLD$ Ld$HI(JLe=HHH5L jAH HDH81g^=_U$ ?1E1HKL)Ls?$ 6H-: L% HEHHLH=8s1LHIوM1L|I,$% ?11E1LM:H@LL\$ LT$)L\$ LT$ ~?$ 1nHHL AjH5H H8H1AX=AYLD$LT$LL$^LL$LT$HLD$HWHԭH55H8 LL$LT$~?LD$$ HLD$RLD$?$ E11LT$LL$ՆLL$LT$HHD$HNH5H8臂LL$LT$?LD$$ >fL1HsIH11 >DLLT$ 藀H}ILT$ AօI{LL\$(LD$ WD$H{IL\$(LD$ 11 r>聅H1E1! .?L E1t>LHL$ LD$LD$HL$ 3 >11E1Y1! 0?HL0FHHH˫H5,H81ɺ >L1HIH11% ?1LLT$ LD$(IC1HLT$LD$ L\$(H% ?11E1萄HHuHH5zH8R1ɺ% ?Q1R11IAWAVIAUATUSHH- LfH|$dH%(H$1HVHDŽ$HDŽ$H$H H$HD$PH$H HD$XH$H ImII Mf(Mn MvIH5* HIEHE EHmzHk H\ H9P L=C MIIGH5 LHHHHI/HH9C7LHH$H<$H+I.H H5 H9paL= MIIGH5  LHHHHI/HH9C LH HD$H|$Z H+ImL;%FID$&yIH(H@I$L MH<$H5F HGHH2!IM LlIHg!I/I "H|$H5HGHH"IM"LIH#I/ I H- L- HEHH4H=ˊz1LHIlM>1L9uI//Ll$AWBHD$1E11HD$(1E1HD$@HD$8HD$HHD$0HD$ HD$HD$DIHF8HD$XIF0HD$PH5 H<H5 HÅp"HmsH-v L%O HEHH#H=txi*1LHIM1LsI,$L4$AZAHD$1E11HD$( 1E1HD$@E1HD$8HD$HHD$0HD$ HD$HD$+L%!G@HyyHQ H9HXHt,HqH~K1HH9t7H;TuHH9HuH;pfDI$fDH=YTIM HD$1E11HD$(1E1 HD$@HD$8HD$HHD$0HD$ HD$HD$L4$AlA@Ht H)Mt I)Ht H.+H DH=MtE1I,$MMt I/Ht H+HL$HtHHD$PHHH\$HtHHD$HHHT$ HtHHD$HHHL$0HtHHD$HHH|$HHtHHD$HHH\$8HtHHD$HHwHT$@HtHHD$HHhHt$(HtHHD$HHYHL$HtHHD$HHJH$HHD$HHQIm6H$dH+%(DHL[]A\A]A^A_f.LvHvI.LuLLT$PuLd$P*fLu#Hu"Hu1Hu@HuOHxu^kupfDHXu|HHuH8uH(uLuH<$ufHωT$`Ht$XLL$PtT$`Ht$XLL$PLωT$XHt$PtT$XHt$PHT$PtT$PLtLt&Hpt I%IHJcH>f.HF8H$IF0H$IF(H$IF LH$IFH$qHIHJcH>fDH AHHpHH5zATL KH81}XAZH g} H=EE1"LXqHH5&LHHVnyH$HMfIH HAHNEOD@FLs>L4$A)MkIIEI+I}AH;=%6H5H9tLT$ LD$tLD$LT$ tIEPуፁ'LLT$oLT$HIh,MtLPHHcHIDCHT$HITIEHHh&H=yLL$(iLL$.1LLHpHLL$ /I)"ImH|$H5*HGHH(IM(HHH9PT)HqHD$HW*HH|$L\$H5HGHH*L\$HD$ H|$ )HT$HHD$HH[HI9C_MKMRMCIII+ HT$ LLLL$LD$LL$LD$II)d$HT$ HHD$HHM(*I(dHT$HHD$HH=H5-H+(fHD$HHU2HpHVHHH=V17IH4HT$HHD$HH(H5*LL\$ L\$HHD$4H8*I+*HD$HBHH9PH5H|$L_LD$IM5I(+ICH;6I{HM6ICHD$IC HD$ IC(HD$0HD$HHD$ HHD$0HI+ ,HD$@HD$8HD$HH5H|$Pv2H5m H8H5KLt$PH5qL8t"H5L?4?H5!HL<H]H@+H;)BHHD$(HAHH|$(H5HGHHAIMjAH|$(LL$XHLL$HH9PAH|HD$(H@HLt$(H5LL$hLiLL$hHHD$h@LLL$(H|$0LL$(?H5IH|$0LL$("LL$(HIBH|$ AHt$ HLL$hHD$(eL\$(LL$hHIKLHD$hLL$(fHD$E1HoLL$(LT$hH9P=KHD$HxD$`H;=DpMc>H5cH9t'L\$pLT$hLL$(nLL$(LT$hL\$ptHD$H@@=vGLL\$pLT$hLL$(iLL$(LT$hHL\$pHCMtLXHcD$`HL$01LL$hH|$Ht$`LT@HHHLHt$`LL$hHHD$(ItFHLL$(?LL$(LH|$HL$`LL$(#hLL$(HL$`HHD$I FHHIELh lkLL$(HHEHT$XH5?HLL$`HD$(lHL$(LL$`g?HT$XH5HlHL$(LL$`FHLLHL$`LL$XLL$XHL$`HHD$(FLHL$HL@H|$6H|$(<>H559Hg4)H|$@:HD$@HHD$H|$H5NHGHH:HH8HH)nIH8Hm-H<$LLL$PaLL$PHI8I){/H+Z/LLT$P`LT$PHH5IBH53HLHH:LT$P.Hm2IL1I_IHLl$1E11HD$1E1AAHD$( HD$@HD$8HD$HHD$0HD$ HD$HD$#L4$M1E1HD$1E1E1HD$(E1AnA HD$@HD$8HD$HHD$0HD$ HD$HD$1kHXL`L4$AZAfHFLH$\^HH5LHVvfH$H`HH\H$L$L$L$HD$PH$H$HD$X(DIGLI H5LHVeHtH$HHwH5LHVeHtH$HHJH5LHVeH7H$HH+H1H$MLL -yH5^@~dIH*L4$1E11HD$1E1AVAHD$( HD$@HD$8HD$HHD$0HD$ HD$HD$Ll$ASB@‰D$`f@L4$I1E1HD$11E1HD$(E1A}A HD$@HD$8HD$HHD$0HD$ HD$HD$fLCMHKIHH+-LLHLD$HL$LD$H\$H$I(L]xH]nH=IH*H5+IMHD$1E11HD$(1E1AAHD$@ HD$8HD$HHD$0HD$ HD$HD$DH=Iq@gHM1E11HD$E1E1AAHD$( HD$@HD$8HD$HHD$0HD$ HD$HD$IGLILL\$#\L\$fI1E11HD$E11E1HD$(E1AA HD$@HD$8HD$HHD$0HD$ HD$$LsM HKIHH+LLHHL$JI.H\$HD$LN[fL4$ABA;Ll$1E11HD$1A BHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$H<$eIIGLILL\$ LT$~ZL\$ LT$Ll$LA BHD$1E11HD$(E1E11HD$@E1HD$8HD$HHD$0HD$ HD$HD$bDLYLLT$YLT$lM_IH-L-HEHHH=iRX 1LHI_Ms1LSI/Ll$ABGfDL(YD\H|$dIhfDLl$1E11HD$1A/BHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$fMWILl$LA1BLHIIfH8XHHʒLH0HV3^HHH$L$L$L$HD$PH$HD$XLLT$WLT$fL4$AHAHHL$L$WL$HL$IBHHOLhWLH<$ZbIfLl$1E11HD$1AiBHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$IGLXILL\$ LT$VL\$ LT$lHVH9HRH5hABH81`Ll$ pL1H`IHBHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$H|$)WLT$IvHLD$LLD$$HL\$KL\$HD$1E11HD$(E1L1HD$@HD$8HD$HHD$0HD$ HD$Ll$E1A;BHD$LnKj1LT$ NLT$ HHD$ILHLT$(LLT$(HIEHD$ HIEH|$LT$ KLT$ 1LHTIHLl$1E11HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$oHoJLbJLl$1E11HD$E1AB HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$nLT$^MLT$HHHLKH+LT$IbHHD$ IL\$ LT$FLl$1E11HD$1ACBHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$Ll$1E11HD$ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$H|$SHLLL$LD$HLD$LL$YHD$1E11HD$(L1HD$@HD$8HD$HHD$0HD$ HD$Ll$E1A@BHD$LHHD$(Ll$LAFBLl$1E11HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$Ll$1E11HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$QLT$ILl$1E11HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$H<$zQL\$I1LHPIHALl$1E11HD$1ASBHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$HHrH5TL nujAH }SH_H81O^@_;CIHHpIQLLL$HHH=f1GLL$HHOI)1H5-HHL$(HL$HHD$8H8H)HD$8H]qHH9PLt$8LLwMHD$@H|$@I)HD$HHD$0HD$ HD$HD$]LLD$DLD$1L\$LT$GLT$L\$HHLHL\$ LT$FH+LT$HD$L\$ H#DL\$ LT$A1ME11LLLL$MLL$HHM1Ll$1HD$ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$Ll$ABmLABNCLl$PLl$1E11HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$ZH Ht$L$1H$H$HcH@H)H̀ uI}LT$rLLT$HHHL$M1Ll$HD$E1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$}A1MGHHHnH5}QABH8OCLl$1E1HD$1ɺHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$PH|$NLIULl$1E11HD$ACHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$HD$1E11HD$(E1E1LHD$@HD$8HD$HHD$0HD$ HD$H=EHL\$H5ͰL\$HD$H|$Ll$1E11HD$A CHD$(HD$@HD$8HD$HHD$0HD$ HD$sLL$Ll$11HD$A CHD$(HD$@HD$8HD$HHD$0HD$HD$ H|$sJL\$HD$ H=-L\$#L\$HD$HL\$7?L\$L%?Ll$1E11HD$E1AB HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$Ll$1E11HD$MACHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$HL$M1Ll$HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$OHD$1E11HD$(E1LHD$@HD$8HD$HHD$0HD$ HD$Ll=qH_=L\$U1LHFIH#Ll$1E11HD$1ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$A(C HD$1E11HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$]BHHHhH5KASBH8=Ll$1E1HD$1ɺHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$LL\$;L\$MABLl$1HD$11HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$LB;HD$@HD$8HD$HLL$@LL$HHD$'HFgH5JLL$PABH8t6pLLL$,6LL$3HD$1E11HD$(A2C!HD$@HD$8HD$HHD$0HD$ HD$黽L5H5LL$iAC#?;HHGHaH5%EABH86Ll$1E1HD$1ɺHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$HD$1E11HD$(E1LHD$@HD$8HD$HHD$0HD$ HD$uALD-HD$1E11HD$(uHHD$Pp4LL$PL^4HD$@HD$8HD$0HD$Hv`H5C1H85L4$1E1HD$1AVA HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$驻Hl$1E11HD$E1A7C!HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$LAFG1E11E1E1HD$1E11HD$(A:C!HD$@HD$8HD$HHD$0HD$ HD$RHLT$P2LT$PLHD$P2LT$PnLHHML@ILL$ILD$詨Ht$H|$L7LL$LD$II)yLLD$HD$.2L\$LD$XHD$1E11HD$(E1LAUCHD$@"HD$8HD$HHD$0HD$ HD$/Ll$1E11HD$ABHD$(HD$@HD$8HD$HHD$0HD$ HD$HD$ξx2HHHD$1E11HD$(A^C"HD$@HD$8HD$HHD$0HD$ HD$fH;o^VLL\$9L\$HHD$hLLT$IBLT$ LHHD$LT$ HHD$LHD$LT$ HHLHD$0Lt$ALT$ HL$0HHD$($LHL$ LT$A־H诫LT$HL$ LL$(LLL$+LL$HD$@HD$8LL$0HD$HH|$HHH9PuH޼HD$H,HLt$H5LHHL蜥H|$ H[H9EHt$ H™IMHLL$YLL$H|$LLL$P/LL$PHHD$L'@HLT$P.LT$PHD$11ADHD$((HD$@HD$8HD$HHD$0HD$ HD$HD$^ARDHD$1E11A{E<3HD$1E11HD$(AC$HD$@HD$8HD$HHD$0HD$ HD$HD$˵A^D.(HD$11A=DHD$()HD$HHD$0HD$ HD$HD$hLpHMiLHIILL$芣H|$LLI.LL$HD$@GLLL$-LL$0H;ZLLL$5LL$HHD$pLHt$HFHHHT$IHHT$H|$HD$HT$ L\$HHD$Lt$HT$ LҾHL\$HL$PLHL$萢L\$HL$HD$@HD$8L\$ HL$HD$0HD$1LLT$H1E11LA FGݳHD$11ACHD$(&HD$@HD$8HD$0HD$ HD$HD$L@HMbLXILD$IL\$螡Ht$H|$L,LD$L\$II(3LL\$HD$#+L\$LL$HH HD$&11HD$(ACHD$@HD$8HD$0HD$ HD$HD$靲HD$11ACHD$($HD$@HD$8HD$HHD$0HD$ HD$HD$зAD111AEFHVHHD$(HD$1AD(HD$(HD$@HD$8HD$HHD$0HD$ HD$HD$霱11E1E1AEF1E11AEFkI{HICHHT$HPH@HT$ HD$0H5˵H|$P T HHH9P HgHD$H HLt$H5yLyHHD$P LH5LH|$P1耐LT$PHHD$I8 LҞLʞH5۴HD$1E1A"DHD$((HD$@HD$HHD$0HD$ HD$HD$HD$11ACHD$($HD$@HD$8HD$0HD$ HD$HD$齯H|$2H6HDUHAH5&CAEH8111E11HD$=lH5=H=֭1/1HD$(IHHt!LAtDv/AdD.LT$P,LT$P2IyHIAHH@HT$ HD$"HD$0HcT$`LL$pH|$L$H$H)LL\$hHĀL$LT$("LT$(L\$hHLL$pHMt&LLT$hHL$`LL$(蛜LT$hHL$`LL$(LHL$`LL$(zLL$(HL$`4HRLL$PH57H81I0HD$LL$P11HD$(AC&HD$@HD$8HD$0HD$ HD$HD$頭H9SLL$PH?H5AADH81/Hl$11LL$PE1E1Һ2HD$HD$(òL\$(11ADHD$(2鼲H=LL$LL$HD$(H|$(#11AD2HD$H=HbLL$H5^9LL$HD$(HL$(12HD$HD$(AD錬H|$(/IRH=KFHD$(H|$(HD$1E11AD2=H=HϱH5б蛔HD$(HD$1E11A!E5Hl$AD21E1E1HD$HD$(JHD$1E11AE>魫HD$XHAQHDSH5#?LL$PADH81-Hl$11LL$PL\$XE1HD$HD$(2ɰHl$11E1HD$AD2HD$(雰L\$P HD$1E11HD$(L\$PA^C"HD$@HD$8HD$HHD$0HD$ HD$HHQOL\$PH54H81,uAE4E11HD$HD$(bLE11E1E1AE4HD$HD$(鯯LPHMsLpILT$ITLT$HT$(LLLT$XڎH|$XHD$+Ht$?Hl$1E1E1E1AD4HD$-Hl$1E1E1E1AD4HD$HD$(HOHQH5<ADH81+Hl$1E11E1E1Һ4HD$HD$(騮HL$(1E1ADHD$(4H|$(,HD$H=ŦHD$(H|$(HD$1E11AD4鷨H=H)H5*HD$(Hl$1ɺ2HD$HD$(ADH|$ LL$P LL$PHD$H|$11AEB4HMLL$PHOH5;AEH81R*LL$P11HD$BHl$11E1E1AEBHD$CH=蒆HD$H|$1E11AEB钧Hl$12HD$HD$(ADLL$PWYL1E11E1E1ACE7鷬Ht$E11A8EHD$7H=HD$H|$1E11A6E7H=HEH5FAHD$H5KH=,1腆1HD$IHtPHLA^EД:1E1HD$1nHD$1E11A,E6NE11AZE:9HD$1E11HD$(AxC"HD$@HD$8HD$HHD$0HD$ HD$rHl$1E1E1HD$AD2HD$((HD$(1AD2鐥Hl$1E1E1HD$AD2HD$11ACHD$(&HD$@HD$8HD$0HD$ HD$HD$0HcT$`LL$pH|$L$H$H)LL\$hHĀL$LT$(1LT$(L\$hHLL$pH?Hl$12HD$HD$(ADHl$1E1E1HD$AD2ܩHl$AD^11AED;HD$11E1E1AED镩LuMHMIHHHL$:HL$HT$ LHHL$PLHD$Hl$PLL$HUIHXKH577AEH81%1E11HD$E1E1ҺDHt$E11AEHD$DTH=% HD$H|$1E11AED H=HbH5c~HD$HHH 5H5z6AEH81"%1E11HD$D1E1LHL$HL$thHl$AC1E1ɺ"E1E1HD$HD$(HD$@HD$8HD$HHD$0HD$ HD$鵧LHL$PACHl$1E1HD$E1E1HL$PHD$("HD$@HD$8HD$HHD$0HD$ HD$;LXMLpIHLT$pILL$hL\$(֏Lt$LT$pD$`LL$hL\$(rHl$11ADHD$2HD$(鿦E11AEB-HD$1AEBMqMIiILHE8HT$LHȅLHD$PHt$PIHLL$PHD$ xLL$PHD$GE11ApD/HD$}H=NHϥH5ХۈHD$HEHc2H53AEH81|"1E11HD$BHl$AC1E1ۺ"{A:E1H|$L\$+&L\$t_HD$1E11HD$(AD&HD$@HD$8HD$0HD$ HD$HD$LL\$PAD[L\$P1E1HD$&1HD$(HD$@HD$8HD$0HD$ HD$HD$靤Hl$AC1E1HD$&E1HD$(HD$@HD$8HD$0HD$ HD$HD$H|$ LL$PHHD$;HXfAWAVAUATUSHHLfH|$dH%(H$1H=CHDŽ$HDŽ$H$HIIXLv(Lk LcH] HD$hHD$pHD$xH(hE111HALHHH8H]L(hE111HALAH$HH8<uH$@D$ H5fH`gH5@H<$A^L;5AHĚHeH9XHLHHHCH5$HHHzIHL|$hHMIHHHTHH9XH̢HHHCH5<HHH IHHMHHL-i@M9oLd$hA1E1I|$H;=AZH5fAH9LD$ LT$L\$L\$LT$LD$ LLT$L\$L\$LT$HHD$pIMtLXHcIMtCHMTID$HHH=" 71LLHsHI/I,$&H;E1HYsHD$hH{ H $HHL 2HD$1AIHH;?HD$pMt I,$IGH5LHHILd$pMHLdHD$hIH'I,$kHD$pI.KHD$H5HD$hLHt$IVHHT$ Ht$HIH@HT$ HHTLLIH )HD$H5LHt$IVHHT$ Ht$HHH@HT$ HHLHD$pHH(H@HD$xL9LoLl$xMLWIEILT$pH/LL{HD$hIImHD$xM4H|$pH/lHD$pI.cHD$hE1HD$HD$HH@HH|$HHI8Ld$IHl$ LIH\$(HI0LLL0I@H0I8H0H64I$AGIG ~jI0LH(H@(H0HL9t;HHH@tՀ8xH(HHR8HcR H0L9uHH9l$:Ld$Hl$ H\$(H|$1ID$L-ՑLMH=q1LLAI~M2I,$ImH;IM I/ Hm H H5HIHVIH$HLcIH eHgAHMEIHH3;HU(H5=ATL >H81X7ZH * H=`*1葋H$dH+%($HĨH[]A\A]A^A_DL5;@H VIL$Aƃ捁-HcӾL$1H)L$L$HLI uI|$LT$L\$"LALT$L\$HHa"Mt I+@I*FLN 9fu;HH0H0H;0HH0HH0H0+HcH0H.H+(HB(HH0LB(L;(}HcIH(LD(H0DH(H+0H@0H@(H0LHf.&D$zID$ID$HD$HH H(HcH>AD$HD$DH=!D$BfH=H*D$ʘL;58 HuH֙H9XHHHH\$pHCH5ЍHHHIL|$hMH+ H HZHD$pH9XH8HHH\$pHCH5HHHIMH+ HD$pL-7M9oH\$hAH{H;=7H58H9tLD$(lLD$(tHCPуፁOLcIH@HD$pHt IAHD$pHcD$ IMt@HMdHCLMH=LL$ LL$ LL$ 1LHAIMLL$ zL|$xI)H+H\$xH;vHsHD$xHD$hH{ L%HD$ HCLHD$0HD$LIVHHT$(MIHaH@HT$(HHLLIH HD$L%pLLIVHHT$(HHRH@HT$(HHwLHD$xHHH@L9W LwMJ LWIILT$xH/=LLrHD$hII.MH|$xH/HD$xImHD$hE13H|$ HD$(HD$Lp@Lh~YL|$L|$ Hl$ Hl$H\$LLd$0D$LHLܰIHI9uL|$Hl$ H\$H|$(VIGL%LM H=?I1LLAI M I/I,$lH;4H$'HHI xMIHFHH$IH5ȊHHV H$H IM+ L$L$L$IHF(HH$HF H$HFH$;IM~1H$MHL  H5Azn7fDH$E11AU A7H|$hE1HtH/Mt I*H|$pHt H/H|$xHt H/H iDDH=!тHu\MtI/taHu4H<$ H $HHD$HHH@HmuH)H+tb1fDLfDL1LT$LT$ @ fD$fDE1HL)DHE11AW A7HLT$pLT$NH LJH$H1IH3HD$hL9dHHHD$hH9XHaHHHCH5YHHHIHLT$pHMHH ID$LT$LH5HHLT$IMHRHÑH9XL-MIEIELT$LLL$H5,HHLL$LT$HHYImx L-Z/M9jXLt$pAE11I~H;=/F H5X0H9t&LL$ L$LD$ LD$HcL$LL$ tIFDPDփ捁LLL$ LD$L$LD$LL$ HHD$xIiML$tL@HcMLAHI\IFHHQH=s1LLHQH(H\$hI/HD$xI. H\$hH;HAD$AT$HH HD$AD$HHD$H|$0HE11Ar A9yAD$AT$HH HHD$H= hHHfE1A^ A=81L*aE11H mH=A` A8.~HD$hI,$ HT$hAx9Ag HHE1HAx9Ag HHHcӸLL\$H)LLT$HĀL$L$L$;LT$L\$HHE1LA^ AV8HtH+tH\$h1Ho/HLT$LT$HLLLE11A[ A 8E11A\ A)8H N AHr5I$HH|$pPLCH|$xdhHD$hIH|$pMhHD$hI.LHD$mH)LLNHD$HKLHMl$LIH H@HH LLLIHSHD$HLHMl$LjHHH@HH LLHD$pHHH@H;n* LoLl$xMLGIEILD$pH/ LLghHD$hIIm HD$xMH|$pH/ HD$pI,$ *HD$0H@HL L;%c*t M HPHuHXLhH\$(MtI$HD$(HtHMtIEHD$Ht$D$HP@Hx+HHD$hHHHD$0HL$(HH8L@L LpHHLhHt H/ Mt I( Mt I. IGL%~LMu H= C1LLAIMHI/Q I,$BL25(Ll$xMLT$pH=فHzH5{flHHcT$ L|$pHL$H)LL$HĀL$HD$xH Mt I/L HD$pI,$LxH5dHHVHH$IA^ A?8E1E1*I~H}H aHF HH$HFH$I9H=H:H5;6kHHcE1A^ AB8H=ojaHE11As A9LwHD$HE11Aq A9aIH5H8wI/L{HFMAa HD$hA8DHD$hM1Aa A8HHH9HuH;HH5pMAe A8H8>L1E1Aw &E1LA^ A_81LAL\$LT$HeHHH9HuH;2HH5E1A` A8H8HH5+H8I,$ HH5H8LL$A8-A:HD$0HL$(HH8L@L LxHHLhHt H/Mt I(2MI/E1Aw I+D%LL\$XL\$XLL\$XL\$Xy1L HtE1LHD$XHH5H8L\$X'Ht$81LIA:LD$LD$ #HL$@HT$HA':Ht$PH6PHD$hHD$pHD$xA:LA:qA::A9OAg A 9HH5H81LLD$LL$,A9#LAw E1/B1Aw A9/A~ A:H Af@AWAVAUATUHSHLnH|$dH%(H$1HHDŽ$H$H I I HF HD$HmHD$hHHD$pHD$xHD$H]Hj L(hE1ɹAHƺHAHD$hIHH8%HlHvHD$hHD$pH9XH-|vH{HEHl$hHEH5pHHHkILl$xMHm HFlHvHD$hH9XLL%uMI$ID$H5tLHHxHHzI,$W H}E1E1AH;=HH;== H5^H9tL$L$tHUBፁxLL$L$HIMtL@IcIAM|HtjMcHKDHELMH= 1LHAH$tL$MTLD$hI,$Hm HMI9EkHt$hH|$xlOHD$pHH|$hH/ HD$hHH|$xH/ HD$xH;-H;-R H;-p Hj$OHm\ HD$p $HD$H;,I_jIH fHL0L;5t M8 HPHuHpH@Ht$ HD$(M& HD$ HtHHD$(HtHHiH JsH9H H-1sHHEHl$pHEH5KgHHHwILd$hMEHmHD$pHI9D$Ht$H|$hMHD$xIHD$pH|$hMH/H|$HHD$hHD$0HD$pHHHT$0L`HD$xHP HD$hHD$pMtI.u L@HL$ HtHHD$HHu H]DHL$(HtHHD$HHu H5DHhH qH9HLqMILD$hI@LD$LH5dHHLD$ILl$pMI(HgH ,qHD$hH9HL qMILD$hI@LD$LH55gHHLD$IMI(HD$hH I9EfLd$pAI|$H;='H5H9YLT$LT$BLLT$LT$HIfHD$hHt I@HD$hHc$HEIl@HMTID$LMH= L$@L$L$1LLAIML$ Ll$xI(I,$vHD$pLl$xIEIEAuI} HD$xHD$0HZcH|$HD$( IGH5hnLHHILd$pMH I9D$Mt$MI|$IHH|$pI,$KLIHD$xII.MH|$pH/H5m1L HD$pIHI,$HD$xL; L; L; LL$L$AI(;HD$pEU HvdH5nH9pL%mM!I$Ld$xID$H5lLHHIMVI,$HD$xH I9@NLLL$HL $HD$pIHD$xMI)aL;% HD$IT$fHD$pHHT$@HD$HXHsHD$H5`LHt$ IVHH$H$Ht$ HHD$\HD$H@HHwH|$LHD$Hs HD$H5_LH4$IVHHT$ H4$HHH@HT$ HHMLHH3 H@H; LwLt$xMLGIIH/+LLL$GL$HD$pII.\HD$xM7I(I)HD$p\H|$(HD$PHD$Lt$L|$XHD$ HLl$ILl$@HD$8HD$0Ld$@Ld$HHl$HHDHafHnMtNE1@CLBL$$f(YBL I\YM9uHL$HL$ HL$0HD$ Hl$8TH9D$(wLl$Ld$@Hl$HL|$XH|$PHD$L5]H@HHH=H|$1LH$@L$M_H\$HH$HHI(G IELI/MfDII# I MMH6H 8HNHL LNL@HHAUHH5H81jXUQZH H=8E1XWH$dH+%(HĨL[]A\A]A^A_@HHD$LHpH@Ht$ HD$(IDID$PуፁHc4$LD$h1H$H)L$L$H̀H@ uI|$LD$L$LLD$L$ILl$xM}Mt I(|HD$hI*LfD‰$vDHH>LH]LHL$L$H$L)L$HĀRL$HHD$hIfM=I(3L2&DH IfDZfDHQ/E1E11E1E1E1HD$@H|$hHt H/H|$pHt H/H|$xHt H/Mt I(HL$HtHH$HHMt I*H H=`TMt I/LE1MtImt=MtI,$tAHtHmtEMImL@LfDLfDHfDT$LD$L$T$LD$L$@LljT$L$pT$L$HωT$L$PT$L$L׉$5$DT$LD$L$T$LD$L$X@T$LD$L$T$LD$L$D@1LIfDH|$HD$pIHeHD$xHHLhHD$pHD$x?E1E11QHD$0E1E1fL8,H=ZHdH5dEHHl$hHTE1E1Q3HD$f.HH5cLIHVIH$HMH$H$HD$5L$HHD$xE1E1S?HD$HD$H5VLH4$IVHHT$8H4$HI H@HT$HHLLIHHD$H5WVLH4$IVHHT$H4$HHH@HT$HHLHHH@HD$xH;/LOLL$xMLGIIH/LLLL$L$=LL$L$HD$pII)HD$xMI(I. HD$pH|$(HD$ HT$HD$0L|$HHD$H HLl$0HL$HLd$8IIHl$@HM~iHD$fE1LpBL $ $BDIXM9uH^DHY@I9uHL$Ld$HD$HIH9D$(rLl$0Ld$8Hl$@L|$HH|$ ID$L5HTHHH=QP1LLH$L$MI,$I(ALq@E1E11E1HD$Q3DH= W8H@#ILQH=VH`H5`^AIME1E11E1HD$Q3HaHME1E1E1HD$Q3@H=9V47Iq@LLT$KLT$;?fDLEM9LeII$HmI|$LAAfHc$LD$hLL$H)LLD$ HĀL$LT$H$LT$LD$ HHD$xIzE1E1R?HD$1LHHD$hIHAHl$M1Q3f.H 1TL$1L$H$LJL)H̀ uH}L$OLAL$ILT$hMHl$Q13ifDL}E1E1Q3HD$5fMeMI}I$HH|$xImHT$hL);HD$pHI,$bL.UfDVQ3HD$E1E11E1E1*L7HD$xIDHF HH$HFH$H1H$MLL dH5I EQOfHFHH$LI@H-SL%[HEHH H=~1LHIXMLl$p1L ImpHD$pR4DHl$E11E1Q3H5RLHVHH$IFLwAH=;RH[H5[<ILD$hMHE1һR?HD$Hl$MQE1HD$h31H=Q2IE1E1E1RHD$?LD$InL$H HD$hP[H=gOH [H5![;HHl$pHH|$hURHt H/HD$hH|$pHt H/HD$pH|$xHt H/H ;HD$xH="EHHL$xHT$hLHt$pn>HD$H :H9HHH|$IHIHD$0LT$0HILd$LPHHD$0LLD$0HHE I$HD$0HI$lI(UH|$pHt H/DHD$pH|$hHt H/1HD$hH|$xHt H/IHt$ HD$xHL$(H8LhL0L`HpHHHt H/Mt ImMeI,$ZLMH=FOHXH5X9ILD$hME1һR?HD$LL$+L$LH|$pWRLH$L$IH=L/HWLD$IkH=N/IJE1E1E1RHD$?LLtfRIl$Hl$pH#I|$HEHH|$hI,$t5HT$H5HD$xIHmHLHl$pHH|$hLH|$LD$0LD$0{IEHD$hHMeHA$I$Ld$pImiLLT$wLd$pLT$M1LLL$L$HHD$xID3iRHL$LAI|$L$AHl$Q13HD$HH|$hkRSL]LL $L $>HL)LtH|$xL$aL$Lt$xMLL$s/L$HD$pIR<E1E1HD$IHL$ H8HhHHHL$(LhL0HHHt H/Ht Hm%Mt:ImLT$1E1LD$E1L$L$LD$T$1E1LL$~L$oLmI$sE1E1R?HD$E1E11E1HD$)R8]eL hLLH^E1E1E1E1HD$+R8IE1E1BSaHD$S?E1E1HD$xE1HD$L$CLL$xL$MLL$U-L$HD$pILL$HD$ LL$ L$H=IHOSH5PSK4ILd$xMWE1E1bSiHD$H|$IHE1һR=HD$bE1E1PSaHD$E1R=2H\$1LHIHH$HHuH|$L$L$MAE1һ;TpHD$0E1E1dSiHD$T$LD$L$T$LD$L$HT$LD$L$T$LD$L$E1E1SSaHD$dI H=%H )Ip1LHHD$pIH0E1E11E1HD$Q4GR=E1MȻsSiHD$MpLt$xMMHIII(tDLLLL $v.L $HD$pII.zLL $tL $eLL $_Lt$xL $MuM(fIUSaE1E1HD$E1WE1һR=H^SH8AE1E1p!LZDMIt$I9HXHHJH~1L;DHH9uHVIHvSE1H%H5fH81iE1E1HD$LL$ H(hE1ɹAHƺLHD$hHHH8 L{H5JHHD$hL|$(FHHD$LHp;>f/H|$O H5GG H;-H~@H5_KH9p"L=FKM!IIGH5<LHH!IIHM!IH"HIHP"HEHhHD$xIH)$H?HJH9P$L5JM%ILt$hIFLD$LH5>HHA%LD$ILd$pM %I._H5?LLHD$hLD$I,$LD$OIEHH"H=LD$ LD$LD$LLLIeMLD$(Lt$pImI/I(Ht$L1HD$xH ,<IIH&H;HD$p(HHPHH5<LH(ILl$pM|(LLpEHD$xIHA*Im HD$pI(ID$H5]<LHH%ILl$xM$H|$ݻHD$pIH)%HD$ݼIH%LD$HLL@ HD$HD$pH'ImHD$xI/HM=H5HHD$pH9p)L-GM*IELl$pIEH59LHH*IMn*ImHD$pIH+HD$HIE蝾HD$xIHP,H<H5NGH9p,H-5GH-HEHELD$0HH5|;HH,LD$0ILL$hM,Hm"H5B<LLLL$8LD$0荿$LL$8LD$0I)#IGHH++H=LD$0̴LD$0-LD$8LLLHD$0%LT$0LD$8M-LT$hI/x$ImL$HD$pI("$IBH- 8LT$8HD$xHD$HID$HIHD$0HD$ HD$hLMxLD$@LPLT$8HI-H@LD$@HH0'LT$8LLLLT$8HI1HD$ H-a7LT$8LHMxLD$@LٻLT$8HH,H@LD$@HH'LT$8LLLT$8HHD$xH?1H@H;T%HoHl$pH-LGHEILD$xH/%HLLT$8LT$8HD$hIHm*HD$pM,H|$xH/'HD$xI/&LT$8HD$hH|$0LT$8HD$@ Ht$HD$ Ll$PE1Hl$HLl$(LT$(H HP@HLt$HHL$ I$8IHL$8H\$XHHfLD$MLHI$0HH0H0lAD$ID$ eHL$8I$0L 'H(H@(H0HI9(HHH@tр8H(HR8HcR H0@H蘰H5?HIHVI许H$HZLcfIH %H'AHMEIHHHH5ATL H81bXFZH JH=`1Q/H$dH+%(,HĨH[]A\A]A^A_DH-@H;- NHD$@HHL0L;5t MHPHuHPH@HT$0HD$8MHD$0HtHHD$8HtHH7HAH9P|L=AM0IL|$hIGH54LHHILL$pMI/HD$hH$I9AH|$pHEHD$xIHD$hM3H|$pH/H|$rHD$pIHwHD$HD$hHLxLh HD$xHD$pHD$hMtI.uLfDHL$0HtHHD$@HHu HDHt$8HtHHD$0HHu HͯDH5H5*@H9pL-@MIELl$xIEH5K2LHHrILt$pMBIm*`HD$xIHHD$HIEHD$hHHH4Hn?H9P>H=U?HHHGH53H|$0HHH|$0IMH/H54LHLD$0LD$0I(e IFLMH=5oHLLAI虴M I.^ HD$pIm: HD$xHm IGAwHD$hI IHD$@HQ1HD$8ID$9ID$HD$0HH HHcH>HPH@HT$0HD$8I!DHxHh.HI{}MI)HFHH$IH5|<HHV0H$HIML$L$H$DIHF(HH$HF H$HFH$芪IM~1H$MHL YH5ϫu|FBDH/IFH9tIH1E1E1E1E1JE1E1 HHD$pE1HD$I/tJ1E1/HH0H0H;0HH0HH0H0@LLD$(1E1t$ T$֟T$t$ LD$(d@IH\$ L9|$0LLt$HLl$PLT$(H\$XH|$@LT$IIELT$L=!HHJH=LT$(LT$ LT$ 1LLHD$肥LD$LT$ MImI(IM1@HcH0H&H+(HB(HH0LB(L;(}HcIH(LD(H0DH(H+0H@0H@(H0l1E1E1E1FE1E1HD$L(LLLL$LL$fH AfDӝBfDH)!Ht$1IHaE1H9nL-*M|HpI9QHXHHJH 1fHH9L;luE1 @LD$0&LD$0i@H5 #HHV%HjH$ITDHF HH$HFH$訚I1E1E1E1GE1E1HD$'1E1E1E1JE1E1E1HHD$E1L0HLLD$LD$$LLD$LD$fH=!HJ,H5K,F ILl$xM/1E1E1E1E1JH裛H|$hLd$xIMHLx}H=IHuHպH56H8ImuLo1E1E1JE1E1qJ(H=HH5ILl$pM1E1E1oH1E1E1HD$JHIEE1IILD$ t$T$赍LD$ t$T$LLD$ t$T$莍LD$ t$T$HLD$ t$T$gT$t$LD$ %GAHD$HHܹH5L jAH ϚHH81>^rF_LLT$8LT$8LT$8ьLT$8IIHH|$x/1E1E1qH0H|$h聗I[H=B=I:IEHD$pHImHHEImH}IA1LLLD$•LD$HI1E1E1E1HD$QG~LHD$ыLD$!Ht$H +1Ld$pH$IcH$H)L$LBH̀ uI}HAIMMt I,$HD$pHT$HHD$HHEH(8E1IIE1tHLHLLLLD$0LD$0HHD$hIE1E1Hd@E1E1VGHD$E1E1f.HhH@H$HaI/HD$@H1HD$ \L ÅA/LE11E1As DL}L}H}H}H}AFHD$f.+|HH HH(Ht H;-^@HPHuLpH@HD$HtHEMtIHD$HtHHH [ H9HXL%B MI$Ld$HID$H5;LHHILD$@MI,$HD$HHI9@H|$@L#H$HD$PH<$H|$@HD$HH/|HD$@ImrH<$H5(HGHH HH|$PHH5zH9HGH;F0!Hu HAHILl$@H/HD$PImHD$@EH<$H5HGHHILl$@M#H5uL@|HD$PIH#ImLHD$@L; DL; }L; pLLL$(LL$(A{$I)HD$PHHPEaH9 'L- M(IELl$@IEH5LHH'ILL$HM&ImQ!H<$H5PLL$(HGHH(LL$(IL\$@My(HI9A(Ll$HE1I}H;=0!H5H9t'L\$8HT$0LT$(ЃLT$(HT$0L\$8tI}Oȃ={)HLT$(L\$0~LT$(HHD$XI*ML\$0tLPIcHM\AD$HHITIEHD$@LM)H=͈LL$(xLL$(1LL$(1LLAI^MLL$(0Ld$PI)$HD$XIm HD$HLd$PL;%إAL;%D)L;%LAŅ*I,$ HD$PHNE H H9H'L-mMg'IELl$HIEH5LHH%'ILd$XM'ImJ H5H<$1xHHD$HHD$(*H{HD$@IH*H<$HxIH*ImLL$(V"LLLT$0LL$(/|LT$0HHD$@Ip*LL$(I)[#HD$HI*;#HI9D$Y0H|$XLHD$PIIm"HD$@H|$XM+H/#HD$XL;%AL;%DuL;%£tL}'.AI,$"HD$PA vfDLxvhkvGfDLXvLHvIHHDJcH>HF0HD$xHC(HD$pHC HHD$hHCHD$`sIII~cIu*MH5HHV{HtHD$xIM~k1HL$`MHL H5styGBKYMIuH5JHHV~{HD$hH IMHD$pLl$`Lt$hL|$xHD$ ZDLpH@HD$fDAKE11E1Ak@H|$@Ht H/H|$HHt H/H|$PHt H/Mt I. H|$XHt H/H `DDH=MtI,$pE1Ht Hm5M@Im5L9t(H H5SL:t Hɚ;H5LtHH9D!H9\$ HD$ H;HH H9H#L%zM=#I$Ld$HID$H5;LHH#ILt$@M'"I,$EHD$HHI9F$Ht$ H|$@HD$PIHD$HH|$@Mx$H/AHD$@L;%L;%L;%LyŅ$I,$HD$P]HD$ H H9He*HHD$@H<$sHD$PIH&]wHD$HIHU&L`Hl$ HHqHD$PIHs&HEHD$(HHE"HD$@I.q"HD$HHD$PHH H9H"H-H{$HEHl$PHEH5^HHHD$HD$ HD$ HD$HH $HmIlvHD$0HD$PHq&I$L`xHD$(HD$@H:&HH H9H&H-H%HEHl$XHEH5HHH(IMy(HmH5H|$(LHD$Xy!I.<#HD$ H@HH\$H=;od HT$(Ht$0H|$ HvHK'HL$ HHD$8HH"HD$HHL$0HHD$ HHu"HD$PHL$(HHD$ HHl"HD$@H<$HEH5HHHh/IMI/H<$8pHD$@H!/HLHD$(|HD$ HD$PH.I.LD$(P&I(9&HD$@HD$ H@q-HD$ H@H%H%H%-H|$ wHD$(H|$(,HL$ HHD$0HH%HD$PH;--HEH5LHD$ .L|$H5I#IH?,IH5fHD$@HH,HH9G+LGM+HGILD$0HHD$@LH|$@Ht$0H|$0HD$PI+Mz+H|$@LHD$@HD$PvH|$HAPIt$(LL$8HLD$(HL$ HT$'LD$?lH51L>LIAYAZM[*L<$)HEHHHEI,$t8HmIHmD $mD $mfD1LIlfDD $lD $fDD $lD $fDD $lD $fDLD $lD $HxlH-L%zHEHHH={k1LHIxrMLd$@1L@fI,$AKE11E1HD$@AlHtHmuHkMtI.u LkDHL$HtHHD$(HHu HkD HIH zH9HL%aMI$Ld$@ID$H5*LHHJHHl$XHI,$pHD$@IHoH$HIFrHD$PIHHH H9HL-MyIEIEH5LHHIHH\$HHImFH5LHLs H+HEHHH=yhLLHITpMwLl$HHm HD$XI. HD$@I,$v HD$PH $HHD$HHG IEH5LHHILd$HMID$HH $q!ID$HpHLHR L5qH$H<$I,$ HD$HL;-IE1H<$HD$LD$H<$11HIDHHH9|HIH)H9~HlHD$HIHRH=bHsHD$PHHI,$H=HHD$HIHHmJHD$P1LHbI,$HD$HAgNE1E1AfH|$@1HH=HH51HH|$HAKHH|$@Ht H/HD$@H|$HHt H/kHD$HH|$PHt H/:Lt$H5DL$AoHD$PI~XiDL$H toDH=[HL$HHT$PLHt$@L%UL=ID$LM H=Rve 1LLAIlM01LAq`I,$ALuLDL$ofDL$f.HD$H $HH8LpHL`HhHHHt H/wMt I.SMt'I,$&E11E1AOAE11LL$eL$pH|$I7H=HAKAFHHD$H|$2kHA{LE11E1AwpAFAVHH HHD$AFAVHH HD$ALApLeHFHHD$`bIHbIH5 HIHVjHD$`HVLcCfDHdlLLHMnHD$HIHL,$E1E1ANAH<$goI jHHD$@E11AKAlVfxf.G.(HؐE1L-ΐHH H9AczLcML$LL$HMlI|$IHH|$PI,$t<LLLL$pLL$HD$@II)9Lqc,LdcLL$HMH|$PAKE1A.LAs@+cHD$PZLcXH-yL% HEHH H=wra| 1LHIiMPLd$H1L\I,$THD$HA>LE1E1AtL}boAKE11E1AkoE1APLAu[LhHD$E1E1ARLAu2DL$bDL$DL$bDL$DL$aDL$PLD $E11aE1D $LD $aD $D $aD $wDATLE11E1AuHaeHD$PIHH=YHlHD$HIHvI,$ H=LHD$PIHI.HD$H1L([I,$AiLE11E1HD$PAvLHD$&H=HH5ILd$HML,$ALA|H|$@HH/DL$\`H|$XDL$HD$@fHt H/yHD$XH|$HHt H/pHD$HH|$PHt H/gH{XH5\DL$HD$PYbDL$H mDDH=@}HL$@HT$XHHt$P,`H|$PHt H/HD$PH|$XHt H/HD$XH|$@Ht H/HHL$HD$@H8LhH(L`LpHHHt H/OMt ImIMt I,$"HbHD$@IHH=H jHD$XIHI.bH= L#HD$@IHoI,$ HD$X1LXI.r L,$AME11HD$@AE1XAMAHHL$H8LhH(L`LpHHHt H/Mt ImMtI,$L,$E11E1L,$E11DL$]DL$sDL$]DL$|DL$]DL$Lt]jH5`HHV|cHHD$pIhL,$H|$XALA|&hIKH=InM`Ld$HMXIxI$HH|$@I(t,LLHD$PI,$H$7L\*L\Ld$HMH|$@L,$ALA|H|$X8L,$ANE11AE1kHS\LF\}L9\\H,\LWWE1A+NAbI/]HHD$HL,$ANE1E1HD$HALL%M IMI9*HXHcHrH~1L;dHH9uHHQE11MD$H5rAE1H81L`A8NH5HHH@\H H$dALA\HHH5eH8WL8VH<$VHD$PHHq [HD$@IH@ HD$ HID$Il$ HD$PHD$@AyNE11E1ALU H<$EVHD$HIH/ ZHD$PIH LpHD$HHD$PHD$HA:LE1E1AtNLLL$0HD$(,ULL$0LT$(HUANE11E1AH-`L%HEHH H=^dS1LHIZMLd$H1LNI,$HD$HANE1E1A|LdTnHHH5cL jAH aH6qH81f^A[2K[LTLTLLT$(SLT$(LSLSWLSALA2H=yHH5ILl$@MH|$XALA/YH E1A:LHD$HAtT@;SBH-L%HEHHH=bQ 1LHI>YM Ld$H1LMI,$=HD$HAOE1E1A]IH=gbIuVHD$HIH: H=EH]HD$PHH I,$ H=HʼHD$HIH Hm HD$P1L;LI,$ HD$HANE1E1AE11A_LAvE1AaLAvH|$XALA`H<$\LL$(IMMQM_MiIAIELl$HI)ELL\$8HT$0LT$(?QLl$HL\$8HT$0LT$(1E1AHOA4ASMA[IHP&H=詷ILl$HMAQMA9H=HH5 IH=aHrH5sILd$HME1AFOAiT[IH=IH|IcLT$`L\$hHD$pH)Ht`HG1 uI}L\$0LT$(q L\$0LT$(HD$PAMHYH|$@Ht H/P HD$@AM%I*H|$XXLDL$OH|$XDL$ LGM1 HGILD$0HHD$@]H|$@Ht$0螴H|$0HD$PI~JHE1ANHD$HAH;qAH;[qE L;-{q H/D$(uoDHD$P|$(LSKAąALA}HxpH5SH8EWIHyHpH5RH8E^H@`HUHHEHIH4H9X LI.H$LD$HI{MtL@HcI$MdCHIMtIGLMbH=HLT$7LT$S1LLAH/?HLT$I*I/H9H;xe5 H;e( H?Ņ4H+,HEHfH9XLMMII@LD$LH5}HHeLD$IMxI(HD$I9ALLLL$yLL$IM[I)I,$IEH5LHH IMHD$I9@~IhHqMhHEIEI(9LHL5HmIM|ImI)IHl$MHPIZfLLL$7LL$EfLLL$6LL$_H=jeIM Hl$1E1A_CfA.@DH-Y\HEHfI,$5E1XDLljT$LL$/LL$T$fH߉T$1E1LD$LL$.LL$LD$T$fLωT$.T$LT$.T$IH-[LHI9IHufDLLjL$ T$X.L$ T$H;[H;6[@@H!5…Hl$A3_E1E1E1fE1LLD$-Ld$fDH߉T$-T$pLLL$-LL$H=iHH5IM1E1E1A^~fH=)HH5趝IMT1E1E1A^>fL-H=ٲԓI7LD$I-fDLE1E1E1A^@7LD$IfD1E1E1E1A^|DH=QLI@M1E1E1E1ɺA^1fMyMIYIHI)LLHI/IILLL$+LL$yMyM5IYIHI)yLLH菘I/IILLL$+LL$LE1E1E1ɺA^D@HF HHD$8HFHD$06)H1HL$0MLL aIH5rg}*8^HFHHD$0(H5I1E1A^H=1HʼH5˼IfD1E1A^A^ln5ICM1E1E1A^:LKM?L{IIH+ILAAH5LHV0HHD$8HE1LH3HD$HE1E1A_Hl$1E1A._4IHLD$)LD$Lo)HcӸLLD$0H)LLD$Ht0Ld$8Lt$@+%LD$HHMZI(PL)CLLLD$芓Ll$IL(#LLD$(LD$MH;"UELǺLD$)LD$HIA0_fDI(y LljT$1E1e(Hl$T$dHH(A_A_E1E1-HDHD$HKTH57A_H8~)LD$E1ɺLLLD$\L|$IHLD$ LL$'LD$ LL$rHt$LLD$ LL$LD$ LL$I{Lm'?LLL$['LL$LI''L<'zL/'cL"'LLL$'LL$LLD$&LD$L&L&L&EH&.L&Hl$A_yH=bHsH5tIH=@HqH5r͖IE1A_"HLD$LLL$-&ILD$ALL$AHl$E1E1ɺA_0IZA@_V0LD$IL 0LD$IhMGMI_IHI/H{IALh%xHLL$V%LL$Hl$1E1E1A_LLL$"%LL$#HcӾLD$01H)Ld$8Lt$@Ht0LQ uILD$LALD$HHAZ_M.1LLLT$L.LT$HHHl$E1E1MкAk_\LHl$E1ɺA_2Hl$Ap_6L5$ Hl$E1E1ɺA_L $L$LL$Hl$E1ɺA_LLHLD${-LD$HIHl$E1ɺA_Hl$1E1MA_.IH=AHBLD$H5>ɓLD$IMtLHl$A_H=LD$LD$IHLL$#LL$HHl$A_-LD$IA`_MMHl$1E1E1A_Hl$M1E1Ak_E1nLLD$r"LD$uH=1HRH5S辒IMHl$1E1A{_DLsMLT$'LT$H|HD$HCNH51E1LT$Ak_H8n#Hl$LD$LL$D,LD$IH=wrIOHl$1E1MǺA}_MHl$1E1E1ɺA_LIiHWMqHEII)LHL HmMI9HLL$ !LL$"+IHl$1ۺA_1LLL$LD$ HD$Hl$A_E1LD$R&LD$HHD$HLH510A_LD$H8!Hl$LL$LD$5DHl$1ۺA`FAR_mHl$1ۺA`3+ICuIHl$1ۺA`LLD$IH{LD$A Hl$A)`Hl$A`Hl$A3`Hl$A5`vHLD$A_LL$QHl$LL$LD$L0V 1LALD$HAHl$1E1 Hl$1E1ff.AWAVAUATUSH(L-KLfdH%(HD$1Ll$HMILfH5_HFID$LHHHHPH/vID$H5LHHHHH5+H"HmAH=ݮELtIHAaA)DDE1H +H=<MH^+H `+HHLH+MH?L LMLHL@HHJATH;H5 ,H81'X`ZH +H=,<E1|HD$dH+%(uH(L[]A\A]A^A_fDk!H=ܭID$H9t*HXHt.HJH~E1 HH9t7H;|uI$fHH9tHuH;=ItfDHaHRH9XH-9HHEHHH9EFLH׆IHEHMkHEHH=L视IIEMeHIEL@MRHfDHM<IHFHHD$HLd$fDH)HH5 :H8HXXLx`HHPhHtHMtIHtHHUhH]XL}`H$qH=9#H}XH$H]XLM`LEhL}`IHUhHt H/Mt I)Mt I(ML'I.H=LH=H]{%HH=LH-H5.ɊHHAOaA/pL4fLL$#L$BLD$L $ LD$L $1HL$MHL .8H5#V>F`@HEA]aA/HHHA`aA/IELH`HHH5MHHVqHJHD$HCf.+$HLx%H=LuMH]IHHmt?LLHH軅I.IL}H=HWHff.AWAVAUATUHSHLvH|$dH%(HD$x1HPEHD$`HD$hHD$pHTIL%#EH^Ld$HHH I$H9HQ L-M1IEIEH51LHHHIEHH0 IEHHCH9E2HHH$H<$pHm}H+cL;%\D ID$H;ҚdHH5~LHIMH<$H5YHGHHIMLLHHkI.ImH;-CH;-}CH;-CHÅ+Hm HIHZH9XL5AMIIFH5٧LHHIIHMIHIHI$L`H$LT$HIB LT$HI_HPH5AHLT$IEHHH=%LT$OLT$LLLHHLT${ImI*|I.bHmGHD$H;BHHH9XLΧMIHH9XH-HHEHELT$HH5HHLT$IMHm H@I9G%LLLT$~LT$HHhI/ Iz1E1H;=@AH;=;A H5AH9tLT$LT$tIJAƃ捁LLT$LT$HI MtLxHL$HcÍSHcHILIBIlHHjH=#LT$NLT$G#1LLIMLT$#ImI*<LtŃI./ID$l$\LH5|HHIMIFIFHD$HHYHo2HcH>DHH5aLIHVIHD$`HLu@HH?H0H5'!AVL @AH ; H81XlUZH  H=E1E1mHD$xdH+%(*HĈL[]A\A]A^A_@]L LH=xIMH$E11UDH _H=0ˎHt HmE1Mt I.IH $HHD$HHhMI,$L$L`HHvIM,IZHFHHD$`IML-NMIML-Ld$8LT$@IBL-HH@ H=LT$( LT$e"LT$1LLIMLT$$I* I.L I$Hm%H$MHHD$HHfDH<$ f.H HH<$H5HGHH HH>IHH#H5HdHEL5HHH= @LLHIbMHmMs Imx H7H7HHD$HHH H 3H=yHH5{IfDH$ME1UI.tfE1E1E1MtImtqMtI*t.MI/LT$/ T$fDL׉T$ T$fDLT$E1E1 T$E1DLLT$T$ LT$T$pfHHLT$ LT$fLLT$ LT$3fHD$HcLLT$L|$`jWHD$hH)LHl$pE1Ht`$LT$HItJMt I/KHmHLT$LT$UE1E1E1DHm$HLT$1T$T$LT$G@AFHD$I. ID$H5ŒLHHiHHhHEH5Hct$\H9 H;4J H@hH}H@HpHIM{Hmm IEIEHD$HHH'HcH>AFAVHH HD$AFHHD$H|$E H1WE1AFAVHH HHD$AEHD$fDImID$H5LHHIMIBHct$\H9~ H;g3H@hHmH@H`LT$ LLT$ IMI*J IEIEHD$ HHH[&HcH>AEHHD$H|$ H ME11WE1AEHHAEH HHD$AEHHAEH HT$AEHHD$ H|$  HM1WE1Ff.AEAUHH HHD$ AEAUHH HD$ fDLp^LmMLuIEIHmHLLLrImH$LfDH$IE1E1E11UH-L-HEHHeH=O1LHI M1LI.;E11U1VKI1 VH<$I8H(H@8Hc@ H0PHLLwLLT$LT$Sf?VI*K1D VVE1E1E1E1mDH<$H5HGHHIMH5lL׺LT$LT$HII* L;5/L;5/L;5/vL ŅI.u L;%/AD$@bHڅ 1LhIH.H;G/HD$H5-H@HHH|$IMrIFLT$LH5HH=LT$HHLT$ LT$HIHvH5gH LT$HEL=υHH]H=LT$FLT$LLHIMLT$HmImH-I9BYIZHLMjHIEI*LHLmH+HI/HImHmHLp H- u 6I$M#fDH-L-HEHHH="f1LHIM1LUI. E11VwLLH=HH56pIMn10VLLLK LT$HH9@VLxH=94fILE1E1E12VeD3 IE11E15VDHU0H0H;0 HH0HU0H0,@HpHEL,IEfDHHLp{LE1E1һ=VxLHHT$ HHD$(HXLt$H\$Hl$LLd$0IMIf.LLIHHH0I4$IuH0I$IEI)IuIMHl$Ld$0fDLLT$LT$CfIBL,IEfDL`LPfL1E1U6@H=HH5~mIM1EWaLlIEIHH@HT$(HHLHIHHD$L5HLHSHHT$(HHHHHT$(HHHHHHHH; ( H_H} LwHIH/HLfH+IMuI./I(HD$8HEH9E H\$ HD$Ld$HLt$Ld$Ll$@HHIHIHD$(Lt$HL$ H0HD$IHL$(HxnHl$H\$HH\$ fDH|$HHgHL$LLIL4L\HLLNHLL@L)HuHl$HEUHHHMH(HE(H0H9M >Ll$@Ld$HH|$8jIEL5}HHH= S"1LLIMIm9LMM(XD8tH(H@8Hc@ H0D@rHU0H0H;0HH0HU0H0@HT$ HHD$(HH\$Lt$Hl$LLd$0IMIfLLI豣HHH0I4$I7H0I$II)IuIMHl$Ld$0VHMPH=HLT$H5iLT$HHFE1E1GWH=e_IIE11IWdLT$I!H=~LT$v_LT$HLLH>IHyE1E1һUL5HD$MlIEMoMI_IEHI/ LLHLT$dImLT$IHLLT$LT$E1XW}LT$@VwLT$H.H#H5[H83LT$E1UH<$~HLLT$(iLT$(,MzM(IZIHI*7 H{IA1LLLT$LT$HI1E1WE1E1E1UHT$L|$`1Hl$pHT$hHcLYH)Ht` uIzLT$LALT$IE1sWMQLf~1LHIH1ULHD$L1E1VyL\HcH0HD)H+(H@(HH0"Hx(H;(}HcHH(H|(H0HcH0HD*@H+(H@(HH0Hx(H;(}HcHH(H|(H0U1W|H1WE1vIH|$H5xHGHHq IMO H!I9EImHMUHEIImLLHLT$FaHmLT$I M I*I.I$HI$+L~LHD$`HE1WWLHD$ HLLT$LT$H)IH HHRI.IlL}_sIRLcHD$H(H+0HE0HE(H0H(H+0HE0HE(H0CE1zW!E11WUUE1E1E1LT$0H|$([Lt$(LT$0IH|$([Lt$(II1W ,HHD$HE1H5 UH8LT$i1LHIH1VH|$H5uHGHHf IMD ID$LT$LH5tHH LT$IM LT$hLT$HI HH59HLT$IGH-uHH^ H=LT$LT$ LHLIxMLT$ I/ImHVI9B^MzMQMjIIEI*LLL]I/H I.H# Im] HmVHI$HIHLT$ LT$ HILHI/LT$ IwL4LT$ eE11WE1L HD$ E1 W_E11E1WE1LT$vLT$HIHH5W1WH8(LT$$L-M HpI9HXHI HJH~1L;lHH9uHVIMIWHH5 1E1H81O DL-}L5JIEHHBH=g1LLIM 1LI.iE1 WLj'1UHHH5H8E1W+IIIElLLXMIME1WqIHHLLLD$(LD$(LLT$LT$cLLT$(~LT$(LLT$(gLT$(H<$TI0E1WE1E11[VLT$(LT$(3 L1E1]VE11_VE1TLLLT$HWLl$HV1VHHH53H8 Li.L\HLT$(JLT$(HLD$(3LD$(.1LLLT$LT$II*uLMOXYfLLT$LT$PLL55oM(It$I9HXHHJH~1L;tHH9uHVIN1E1H H5LjVH81LLT$+LT$HLT$LT$VE1 L'H%H7L XE1H8*HLqXE1H8LLLT$ULl$H$HuHH5H8IEE1HIEL75LWE1" I*XLME1VI*E1OX:H#LLT$H8LT$I XHIE1HLT$LT$*ImXvL`_HLH8IEsXHIEE11VLLT$ LT$LLT$LT$L1E11VcE1V]LT$H1VIE11W1LALT$IH|$]I6H5kHv_1VE1LLHLT$HIV$H LE1V1LLsImIh[1E1"WCL1L?IH WE11WE11WH|$BIL2XLT$H/HVH5H8LT$I*L׻OXIE1NE1MVcgVLT$HHH5ALT$T$H8T$LT$VE1HH5IW1H8 L.tLE1OXE11VLHLLT$HI1VfE11V^LT$IE11VH|$IE1E1MVE11VLT$HHaH~H5VH8LT$HHI9IHuL;-,7W HAHH5sT$H8GT$HHI9HuL;5HH51jVH8H^1VH81L1,E1 XE1qXdf.AWAVAUATIUHSHH<$HdH%(H$1HHHD$hHD$`HD$XHD$PHD$H/HfH}H9t;HXHHqH 1fDHH9 H;TuHhH uH9H L-uM"IELl$`H uH9H!L5uM|"ILt$XIFH5fLHHi"IL|$PM/"I.=HD$XHHD$(I9G"H|$PHLHHD$XH>#H|$PH/ HD$PHD$(I9E%Lt$`E1I~H;=%?H5H9tDD$HT$HT$DD$tI~Oȃ=&HDD$HD$XIH7)HD$PDD$Ht IGHD$PIcI$MdA@HI\IFLM &H=%)1LLAIMy)Ll$hI/HD$XI.I,$HD$hHD$`HD$H}HfdHD$hH9 HH5dHH['HH\$hHU'H5dH9B HCH; )H{u { H~ L=/ E1HD$0II9AL|$`H+1HD$hI/HD$`H}HcELd$H9HH5eHH$2IL|$`M2H5dL9IGH; ^IH Hh E1HD$0HH9AH\$PI/&HD$`H+6'HD$PE'H*eH qH9H6L-qM6IELl$`IEH5dpLHHR6IL|$XM,6Im*HD$`H3 HD$(I9GR8Ll$XAI}H;= *H51 H9tLL$LL$tIUBፁe7LHD$hHH;HD$`Ht HCHD$`IcHEHBcHlAFHHHTAFHI$LdIELM7H=9@1HLAI+M<@L|$PH+s3HD$hImL.HmHD$PHD$XHD$!.HbcH oHD$PH9H7L=oM9IL|$XIGH5kLHH2:HHl$hH:I//H|$H5PnHGH@pH3H@H3HH\$XH9HD$`HD$(H9E8H|$hHJGHD$PIHD$`H+b/HD$XH|$hML9H/1H$H-^HD$hHD$PLHMuL[HH??H@HH&3HLLHHDH$H-^LHMuLHH9H@HH2LLHD$hHHlCH@H;D$(0LoLl$XM0LGIEILD$hH/0LLFHD$PHIm7HD$XH=H|$hH/O2HD$hHm.2HD$PHD$8H fDHH(Ht H;-0HPHuHHH@HL$(HD$ HtHEHD$(HtHHD$ HtHH|$5H<HHHH$Hl$IH$Lt$HLd$HD$@H|$@L苄HL9FH7HD$PHH<L`LHX IFHH@pH7H@H6LL$HLLL$HILd$PM>I)6HHHD$hHD$H)>HD$XHH=HT$HHX HLHHPLHD$h:,<Hmu6HD$XI,$:6LkHHD$PHD$H=nHD$XHHK=HT$HHX HLHHPLHD$P<HmC:HD$XIH$Hl$Ld$HtH$NHD$(HtHNHD$ HtHNH5)Z1H?HHMHl?HMHHl$HD$ -Ds1f.CL=^AIL|$0f.Ld$HXHt$HqH~N1 fHH9t?H9TuHHH9HuH; fDH\HiH9XHiH> HH\$hHH,AŃ H+HD$hEH5\H hH9H%HhHq'HH\$hHCH5aHHH='IL|$PM'H+9HD$hHH8(HgH}H5\HHCHGHHGILl$XMl(IEH;(IELl$`MIEHHfIvHD$XAN HHH@%*)Ls HHS(IHzAE IU0IMH@HDHT$E1HcHl$1Ld$ LIAL|$88DHz0LRH@LED91L~LHHHTz LBMtIM1J  uLRHfLHAYAP1E1H|$hHt H/H|$`Ht H/Ht H+H|$XHt H/H|$PHt H/H DDH=QMt I/E1HtHmtBMtI,$tFH$dH+%(;7HĨL[]A\A]A^A_f.HHfDL8fD+]fDfD fDH fD1LI0fDMcLl$PLL)Ld$xHtpLl$pH$HD$hH\Mt ImHD$PH+OH`BLPZL@,HDHH9HuH;LfHH5WHH"IL|$`M"HL;|$0I9H\$H L;= LAą"I/oHD$`EHfHD$ H9yHUHEH5bHHD$HHIL|$`MIGH; H;~pH@hHH@H1LHH\$hHnI/ HD$`HCLcID$H$HHcH>L|$8Hl$Ll$`H+Ld$ uHlrHD$hHH&LhHfaHHC HD$`IH&HSH52aHb!IGLM&H=(LHLAIM4-Ll$XI/ HD$PH+ HD$hI. HD$`ImuLxH5yTL9 ID$H;#I|$ L=SAIL|$0L|$XI/EHD$XE#H$HQLHMnLIH$H@HHLLLIH6H$H(QLHMnLIHF%H@HH`LLLHD$`IH6H@HD$hH;MhLl$hMIxIEHH|$`I(L8HD$XIIm HD$hM$H|$`H/HD$`I.HD$XYH HD$(HHL$ HL0I9t MHPHuHHH@HL$HD$MtIHD$HtHHD$HtHH\$HHH$L|$L4$HLd$8IHHH;wIID$H;H;_iH@hH H@HLLILt$XMID$H;JH; H@hHsH@HfHLIL|$`MID$H;H@hHH@(HLHLЅI.HD$XID$H;H@hHH@(HLLLAE@I/HD$`HLL4$L|$Ld$8Mt I.1HL$HtHH$HHE/H\$HtHH$HH4/H5M1L3I/H{.H2H+.HD$ HI{H HD$hDcf.H+HEH5PHHH- HH\$hH HCH5\HHH@IL|$`M:H+HD$hIGMoIEH! HGHcH>DcAMcI9H+Ld$A Z1E1A]DcCII IDcCII EofDI/HOHX\HD$`H9XL=6\MIL|$`IGH5KLHH[HH\$hHfI/L9HD$`IH>HD$XIHLxHD$`IHHNH [H9HH=l[HHHGLD$@H|$8H5MHHH|$8LD$@IL|$PMgH/H5tNLLLD$8L I/LD$8HCLMH=LD$8LD$8NLD$8LLHAIbMLD$8nL|$PH+D HD$hI. HD$XI( HD$`HD$PL;|$ HEMwHH5MHHHH\$PHHCH5XHHH\ILD$`MgH+u HD$PL;D$0L;D$HL;D$ LLD$0 LD$0I(P H$H5eIHD$`LHt$0IPLD$@HT$8HׅHt$0HHH@HT$8LD$@HHBHLHH.H$H5HLHt$8IPLD$@HHT$0RHt$8HHIH@HT$0LD$@HHLHD$PHH,H@H;D$(LGLD$XMA!L_IIL\$PH/SLLLD$(A0LD$(HD$`II(nHD$XMJH|$PH/]HD$PI)=H$HL$HD$`HHH$IfHHHl$0IHD$H\$L|$(HI@H<$L4oHL$LLIH,H)HLHHLL L)IuH\$L|$(Hl$0HCL%GLM H=#^ 1LHAIMH+ImLd$DHD$ H\$ HI/ZL|$ EoAMcIJH[AoĸIHLD$8语LD$8}Ld$A\ALL4$L|$A]Ld$8LkJL^^1AZHD$hA]H51E1A\ALA1E1H;;LILt$`ME1A\A¬HHHT$H\$IHl$H$HL<$L,MMMILHVIHH0I4$I7H0I$IM)IuH\$L<$Hl$LH;LHD$XIH(1A2]ALs HHS(辷IHPHHHT$H\$IHl$H$HL<$L,MMMILHUIHH0I4$I7H0I$IM)IuH\$L<$Hl$L-]<L5^<IEHHH=úN1LLIdM L|$X1L,I/0 HD$XAC]A1E1ΪĪA躪Ll$XLD$hM*H|$hHD$PH11LH3H+IuHrM Ld$1AZAf)HHLHD$(6LL$(OLHD$(LL$({HHH@HL$(HD$ .HWH1AV]AH8.LHL{HD$XIH1E1A]A|@HH51AZA\H8 Ld$JE1A]ANLd$A]AHHAX]H8耩HD$`!H|$H#Af]HHHH|$h!LϨ 1Ad[As貳I1HAZAkH8֨Ld$TE18H==.8IL|$`MM(H=.H:H5:I+HHHI9HuL;=2[HAA|ZH8#HD$PH+Ld$Af1HAzZAfH8Ld$aHAZH8ħHD$PH+Ld$Ak1'Ab[As.AZiHD$XA]A1E1IcH\$`Hl$x1H ,L$H\$pH$LBH)Htp uI} LAIL|$PM H+H+!H薦H|$PLIGHD$`HMoHAAIELl$XI/LLL$3Ll$XLL$cHs1HL躯HD$PIH,1A[AsfDH=+H:8H5;86IL|$XM^Hl$1A[AtuHHuHH55H8 H+Ld$AZAf13A]K L> THL$HxXHX`HHXH $LxhLhhHH`Ht H/Ht H+EMt I/CHD$`A]HD$XHD$hpH踤-L諤0A]KH|$P>AZ1HuHH5H8H+Ld$A[Ak1LmLl$`MkHEIEHHD$hHmt0H|$hHLHD$PIImDL7HLl$`Mu fHl$E1A[AtH=){ IHAk1蓣Ld$aA].Lv1LHH+ILd$A[AsHl$1A[AtLd$Hl$1A[At_H襨HpHD$8H(H5H8aLD$8HAf1谢Ld$~1HLIIU\A|Ld$H$Hl$A<\A|Ld$ H$Hl$AQ\A}Ld$kH$Hl$A1\A|Ld$H$Hl$A9\A|Ld$-H$Hl$A/\A|Ld$HHH9dHuH;BRL`H;SH5LFILƙH$Hl$A%\A{Ld$HH5³A_ZAdH8Ld$N1LLHD$XIHRhHH5H8͚DAZHnH5ϨH8觚IEHA|Z01A~[As1LAIHH9HD$0AH;AEA H;A I/uL蔘HD$`E HzAƅAC[Ao1E11AX]MA[\E1A]AHl$1A\Au1AV]AHl$A[AuHI9HD$0AL;=EL;=L譞AŅsA4]A1E1dHH5.H8Ld$AzZAf3HdI9HD$0AL;=AEA L;=$A H+uHHD$hEʶLAŅLd$AY1E1AXLd$AZAkfAVAUIATUHSHH HLfdH%(HD$1H$HT$HIIMHH HNHL LNL@HHATHvH5ޤH81LXKYZH ԣ%H=B=HD$dH+%(=H 1[]A\A]A^f.HV HsHD$dH+%(H L[]A\A]A^鰱IItPMH_H5('HIHVIuH$HM{H4$HT$HF HHD$HFH$H~1HL _MH5lHTy;YfDHFHH$IH5HHV՚HtHD$IFLc:ǕHG8H@HGHWl1~HGxHff.LGxLHHGHH>lHF~>1f.IHTIHTPHHtHHƐH9ff.fxyHcHTHt\H:~HHLPII?~T1H9~HH|Hc~HDPHHIHDHHHHHH9D@CfE11ɅCÐIuH1H9D@Cf.AA)xGHcHDf.H0HcHH HqHpHHqPHHуD9uE~IAPHGHT@HHOPHHH8H9uDHǐHff.@ATIUHSHHHt HՅuH1Ht[LH]A\[]A\ff.HGHHGHHHHHt1DATIUHSHHHt HՅu=H{ Ht LՅu+H{(Ht LՅuH{P1Ht[LH]A\f[]A\ff.HGHHu/10HtHCHHPHHHaH5Z18Hu'10HtHHHPHfDH H518AWAVAUATUSHL6M~JAHIIE1At@IqIUEHLL$ $LL$ $HI]I9uH[]A\A]A^A_ÐH;HEt HH@HHu‰L$L $"L $L$ATIUHHHGHtDHHEHLP0HHt\H(tHǼHH]A\軏fDvH #H= H1]A\fNv@USHHH-THHEHkHHEHt H/tMHEHHHHEHtH/tH1[]H1[]f.fUHSHHHHt/HH/t H]H1[]諎H]H1[]fDHff.@HHGHt HH ؜jHD$H=ۯ HD$H@HLGHHGMtI(t1HL1HDUSHHH-ԺHHEHkHHEHt H/t}HEH{ HHk HEHt H/tnHEH{(HHk(HEHtH/t_H{PHtHCPH/tH1[]SH1[]f.;yfD+ffAUATUSHHL-HHAHPLnH5-"IED%-HEHHE HCE0C$HE8HC0HE@HC@HC8HEHC HC1AtHE(H{HC(HEH/tRHkL9tW1H[]A\A]@H}PH5}!;DADPDHkL9uHmHCD#iH=H D H{HbH/HCH[]A\A]H-L- HELMH="1HLAH臑H1HTHmmiAHT$ t$t$T$ $fHAiˊ'H1LiHHDiiDHuHжH51H8 HH5)H8ATL%{UH-SHEHHtzH=~1HLHTHto1HAKlHmt(DH bH=[1]A\fDH踉fDH1LSHHuAGl;HuHǵH5(H8ff.ATL%UH-SHEHHtzH=Ø~1HLHdHto1HAl/Hmt(DH rH=*-[1]A\fDHȈfDH1LcHHuAlKHuH״H58H8ff.AUATUSHHL%JHHI$Lft Gh1tHExHC01tHHC81tHHC@1tHEpHC(HEHHElC$HE`HCHEXHCEhC HEI,$tHkL9t'1H[]A\A]fL蘇HkL9uHmHCf.H- L-"HELMH=ז1HLAHwH1HAx>HmtgH=xH }D<H{H&H/t$HCfH踆/諆fH蘆H1L9HH_DAxfHuHH5AxH8ׇ;HH5H8輇nfATL%UH- SHEHHtzH=辄~1HLH$Hto1HA[Hmt(DH 2H=J[1]A\fDH舅fDH1L#HHuAW HuHH5H8Іff.ATL%UH- SHEHHtzH=΃~1HLH4Hto1HA~Hmt(DH BH=[1]A\fDH蘄fDH1L3HHuAHuHH5H8ff.ATL%UH- SHEHHtzH=ނ~1HLHDHto1HA~Hmt(DH RH=ʦ [1]A\fDH訃fDH1LCHHuA+HuHH5H8ff.ATL%UH-SHEHHtzH=~1HLHTHto1HA}Hmt(DH bH=[1]A\fDH踂fDH1LSHHuA;HuHǮH5(H8ff.AUATUSHH5HtIŋClL%I$(HHHLe(L%EH]ID$Lm HHH=i褀1LHI MHmt*HL[]A\A]@L%I$eHxHL[]A\A]f.ImI,$AkDE1H H=vHL[]A\A]軆HtkHmAkuHL1H苊IHDAkyDL谀gL蠀FHܬH5=H8zHHcl_Ht HDGH ؎{HD$H=HD$H@ATIUHHGH5[HHHHID$t_HELHHHtOHMHQHUHtFHtH]A\fHHD$HD$H]A\fDLH腊kHtFH H=+H1]A\@kf;H-Ht$$t$ff.ATIUHHGH5'HHHHHELHH@pHtNH@HtEHMHQHUHt:Ht H]A\DHHD${~HD$H]A\fD붐kHtFH ӌH=3H1]A\@+HHVk뻐Ht$~t$ff.AVAUATUHSHHLndH%(HD$1H$HuYIH^H}HH/t-H]1HT$dH+%(H[]A\A]A^fDk}fIMt(IuZHFHH$2{HH$H{H5ILIHV2HH$IFHuLkfDHHHH5AUL [AH H81XumZH 5H=1HL MH5L{7jm}@ATH HGH9t.LXMtbMHM~I1DHI9t7I;LuH;=H?LtLA\{LA\DILA\fHDHH9tHuH; uH;=tHHH9tHuH; ҨtHPHIH5H2H81踅H  E1H=uMHttHIfH nH=cH1[]A\A]A^A_fI.3LAMnHsIEA fnHIELHD$sHD$@1,nAHMMHQHUHE1HHD$csMHD$}I,$wL=sjI.OLsBfH/&s@'nH kH=p+IEMHP1hfH=HH5IMH nH=IEHP1IUH7H=qHH5IMwA dnlf.H=1Ii@H=ILnARnMMH @nH=;AWAVAUIATUHHGH59HHHHsHEH5HHHIHEHM_HEHID$H5NLHHkHI$HHeI$HtdH=%LeIHiuIHcHhH=XHL` {Ht7ImtpH]A\A]A^A_fLxpfDHhpAImAu}AdtMDDH ~H=H1]A\A]A^A_@LHD$pHD$H]A\A]A^A_LofDAS}AdfzHgAU}AdHmlHo_@zI[zHAX}I,$Ad$Lio@Ac}AefHmtAm}HAm}3o믐AUATUHHGH5HHHIM ID$H5.LHH+HI$HH%I$HHEH5HHHIHEHMHEHtKsHHL`H=HyH Hmt2H]A\A]H(nfDLn[HHD$nHD$H]A\A]@A}DH d|hH=H1]A\A]xIxHA}I,$uLmf{xIA}HmuHWmhfA}A}ATIHUHpHtEHID$LHH@pPHmt H]A\ÐHHD$lHD$H]A\fD1ff.ATIHUH*pHtEHID$LHH@pPHmt H]A\ÐHHD$slHD$H]A\fD1ff.HH`oHt HDKH z{HD$H=^HD$H@AWIAVAUATIUSH(HFH;MLwHt H;p/I1HD$E11H|$f.IGH 5I9OcH9IGH,HHEHt H+HH9EHEHPH?H>]HIAD$lIL$`HVID$XHHHHHyI|$xJOH9HIHxIIIMH|$HHt$HD$LHt$HH3pHHtHܗH0H9nI/H7LNj*fI$ID$xI$J JHH Jf.H]EHH fHHHt$yoHHt$HHHt$iHt$HϖH9E4DHHt$ctHt$HHtHt$HD$mH|$Ht$HH/cHt$6iHt$O@H9IlHHE@HHHt$!mHt$H@LXlIHH=H tIHI,$H=L'IH)Im1L绅AbI,$L%vDLH=ҍrI/$LH=ύE1oHt HmH(L[]A\A]A^A_D1]EHH }AI,$hLg[HH5܇AH8'i2fLkIHH=}HrIHI,$H=OLIHIm1LVAKaI,$CHHHH̓H5EAH8bhmDLfLfULfkHoIHH@HHD$HHAImLDfH0fI/rL%tLt$T$eT$t$LecrL%Vt{A]H :tqE1H=JLA(NAAQHG@HFH@HXHt8HJH~1H;tHH9uHݾqL%lsHH9kHuH;5Yq1L%)sef+땩tAWIAVAUATIUSHHHdH%(HD$81HHD$0HD$(HD$ HxHFH>H9t3HXHtgHqH~~1f.HH9tgH;TuHGI$LI<$HXHT$8dH+%(qHH[]A\A]A^A_DHH9tHuH;АtfD+bH|H$H HH(H9t HHPHuLpLhHMtIMtIEA@f@ϘHcgHD$0HHA`HHHL$HT$(HT$gHT$HL$HHD$ IL I$IO IAMgIW(HD$0HHIH= rHL$LL$>aLL$HL$1LLHD$hHD$HeHD$(I/I,$HtHmuHH$bH$@MtI.uLH$aH$DMtImuLH$aH$@H8IDLpLhHE\fDHYHLH$taH$LH$\aH$3LH$DaH$1LLjHD$(HA7tDAtHD$0H|$(Ht H/IHD$(H|$ Ht H/6H$H5eHD$ HxX3cL(oL uDLLLT$LL$H<$HL$0HT$ Ht$(LL$LT$IH|$0HH/H|$(H/H|$ H/H$HH8L@H(LxLpLhHt H/Mt I(\Mt I/ZI$HxHL?nL H$HH8L@H(HXLpLhHt H/Mt I(1Ht H+IH|$0Ht H/\H|$(Ht H/lH|$ Ht H/|LLDlI$Hx1(AXtIA,tH|$0HH/^dHuH H5mnH8E`HD$(A7t^^^R^9w^MLL$LT$LD$$[^LL$LT$LD$$LLL$LT$$/^LL$LT$$HLL$LT$$^LL$LT$$LL$LT$$]LL$LT$$LL$LT$$]LL$LT$$pLL$LT$$]LL$LT$$`L]Lu]I$HxHL$\]L$^~^ff.AW1AVAUIATI1USHxH=dH%(HD$h1HD$@HD$8HD$HHHeIu`L\HD$HIHT_[HHHL0Mt L;5HPHuHHH@HL$H$MtIHD$HtHH$HtHHEH5/HHH8HHL$@H2I}pHL$aIHm HL$H[H9AILYM<LIIAD$ILL$@H)IyH;=Ј H53H9LT$(L\$ LL$~eLL$L\$ LT$(LLT$ L\$LL$(p`L\$LT$ HHD$8I MLL$(tLXHcD$MT@HI$MdIALMH=jLT$LL$YLL$LT$ LL$1LLAHD$aHD$LL$H HD$HI/HD$8I)I}pL|$HHD$@HD$Hs[H9ILHHt$H $H:HZL2LjHrHJHt H/mHt H+FMt ImHmI,$`Mt I/9HT$hdH+%( Hx[]A\A]A^A_fDIAH΃捁1Hc|$L\$PLT$XH)Ld$`1H@HtP DLT$ L\$LL$ LLT$ L\$LL$HD$HH^ Mt I+I*LLL$YLL$qfDHpH@Ht$H$yfLH$XH$HH$XH$H$XH$fDHcL$LLL\$PH)L\$ HtPLT$XLT$LL$Ld$`]TLL$LT$HHD$HL\$ AvH|$@LHis@IGH;ŅH;H@hHH@H1LHD$HHAwLUfL}KDLW LH$WH$LH$WH$AvHE1LeE1L^}H|$@Ht H/H|$8Ht H/LDLH$tHH$t+HEHP1HUHuHH$VH$MfHLT$ L\$VLL$@LT$ L\$f.D$LL$@AE1DLLL$VLL$fAvL\$LT$HD$$IV$HD$LT$L\$L\$LT$HD$$VL\$LT$HD$$`HH|$HE1AvHtL0I#H II/$wjI.Mt I)tMtE1ImMt I/b4$H DW1H=%oI,$E1fLMHH!HxHLLLL$H)L $HtL|$Lt$ 5DL $HHMt I)3I/.LH!LH1wH iVH=Nn)M;sMH]$wE1ɻI/E1LLL$GMLL${LpGL`GLPGrI/3Im$wME1LLL$GLL$DLG E1^LF1LLPHH1wwfQHQ{QI$wE1ɻ$wIgIA1E1f1LLL$OL$HH$xML뭹LL$1H)L|$Lt$ HtLR uI}L $^LAL $HH$xLL $NL $HIe$wIA1$wE1HrL-ArH:L9HHHx1H5ve_O$wHmCHE6$xGJLL<$$xH{HqH5uTH8MF`Haq$wHHHew1H5dN$xY1LAL $HHD$HpH5SH8E$wLL$WEI$wDM$wE1\HHIHHtLЅt5HpHHDL`HHt8H(uC&H RH=(j1H@:@AUATUSH1H@H/LkxIHcClI\I9sHI}FHHtyID$I;D$ IT$HEH,HID$Hmt;IL9wLjAHI,$uTLHD$BHD$@fDHBfDAzI,$tsH Q4DH=eiȿ1H[]A\A]HLFcI,$tfDHmAzuH3BL)BLBfDAyzrAz`f.AVAUATUSHHBH1h?IHLHcClI\I9sHI}EHHtvID$I;D$ IT$HEH,HID$Hmt8IL9wL?HeI,$ubLHD$:AHD$NH(AfDAzI,$A<u LADH }ODDH=h71H[]A\A]A^HLEENI,$uL@HmA<AzuH@두H-L%HEHHtcH=O6?uh1HLHFHtm1HA:Aze:Hm*A<AzH1LIHHuA:AzAzEHuHlH5vOH8NAff.AWAVAUATIUSH8HGt$*H1=HHL=tID$IH;lt H;kE I$M1HD$HcD$E1E1Ld$ D$HHH|$$HD$(SIEH lI9M~H9IEL$HI$Mt I*IMt I.IwH;5kLIGHxH8H_AwH)LFHIERBIIHMoIHL9%4L5jM9t$tLL\$ @L\$^$u 1M9t$$Hu HEHHH9H9HuI$L$HHEH|$MMMLT$HD$LLT$HILT$Ld$ 1CLT$HIHjH0L9 LT$ALT$fLL\$3=L9%$L\$L$H} HEH5.IIL9H9H}HH4HHED$$ fDL<L<H %KH=\۹I,$AH8L[]A\A]A^A_fAHtI$L`IH J\H=\sE1fDLHL\$@L\$5LuLd$ E1LLd$ E1M̀I+Im#MI(H JH=[ݸI,$AOHLH/Ht HmlHH+HL$=;L$LMɁI+u@E1E1L߉T$LD$4$:M4$LD$T$)M4H [IH=8[I,$HE1LH/AL$:L$/H;5f&LL9IH9MdHI$@H>HHF`LLIfDH|$ L$@L$HuLt$(I)II~7L$HIiM~!H@JIHH@HHD$HHLL$<9L$HIāH DH=UL$MزH$LLd$ E1IAXG8ILLT$=5LT$LLT$&5LT$1HLT$HI2LT$HIIM~#H@J H HHHPH9uLHLT$L\$8L\$LT$HHH(I+rLLT$4LT$[L$9L$ILLd$ IwLLd$ `KQhL\$LT$3L\$LT$QM1E1ME1E1 LME1LMgI@A@HFHts@tjIXHt:HQH~1H;tHH9uE1MLMÀCML9MuH;5'`ĩtLLT$CLT$iLLT$74LT$AWAVAUIATUSHH:GhHI$5wlLIHHH;T_HPHXLxHX IHH(I,$H-^L;=%_AI9DL;=^L8AąyzAhsAE1H M@DDL$H=YL$I/A7M"fDH+.HD[]A\A]A^A_@EIEE?LLPIH2H;S^I9RL;#^ELLT$8LT$ IUHRp.H HBHLT$HLLT$HHGIELT$LHLPLT$IHEMHHEI,$I/AI*Lb0H+HP0H@0I,$L-0LHLP IHH(I/dL/WIUHRpHHBHLT$HLLT$HH1H;-\<IELT$LHLPLT$HIHEDHP/MH-L-rHEHH{H=>-1HLHP5H1H)Hm1sH U=H=VLA 'CsH !=sL$H=PVAͫI/L$$LLT$T.D$$LT$fDHZH5UAZsH8/HmrH <DLH=UAE`vAKsHxHHz]H V\HEHfZH5UH817LLT$-LT$fH-LT$fDHZH5E?AKsH817,Ht$$A-$t$]DAsAOArsA7A~sAHLLT$HLT$HHLLT$(LT$H H1LN6HHf-sAsAAsAHHFH}H9HXHRHqH~1H;TyHH9uHXHJL$H5GHWAsH816L$fDHmAHL$+L$fHHEHL$+H 9sH=6S蹨L$AsHGH5RAHPHWH81r50HH`WH5:H8,-s"H;WL$H5EAsH8j,L$ HHH92HuH;W H*f.AWAVAUATUSHLg L;%kWt!I$HL[]A\A]A^A_f.L=HoxIHcWlIHLHHL$IH9(1H}I-HHtcMtI.tHHL4IHI/tHH9l$ML)fDL)fD|VH 8H=QE1ئIHIHt9MI.L^)fI޾|W룐L8)fDI$I} IHI$H/tI$Me MHPI$f(LE1Mff.AWAVAUATUSH(HULl$`Ht$HT$I9 LcLiUIMƿ-IHHH-I\$HHID$ HEMt$(LM+H=7'71HLAHt.H I,$oD$`oL$po$o$o$o$o$o$o$o$o$o$o$ (8HXhMtIE@SWIEH5ϻLHHIMHxH/ AuhIUpLxII}XIM`HHIuhH8HHD$hD}lIHEHAHUpJ>H]PHH}XHHM`H]xHHDžH9s!HfDHH9v H8xHN,;HMXE1L9H;M)IHMt I,$H}X^)IH LH0IIMHIHRI9D$IT$HBHHEt$HI,$HLuXI9DHD$IHHD$HHEHPHUfDL)RIMsDL%H9RI9D$UL/HHt[HD$)H|$IH/u9$2L$HIEH(L[]A\A]A^A_ÐII.*HIfLX$lLH$33H 2H=@LMpHEE1HHEHt$M`I/VL#I@H#fD#fDM?fDHAAItFH 2AH=KϠZf.I.1E1>ALN#H 1DH=SK膠H E1DH1L,HHI,$4H r1IH=J(E1-H"Ht L&I#E1:AD$AT$HH IH 0>H=J賟뉐[-ITH 0hAH=JJ}HmH"rD'HHCNH51H8|#0_H=R51; IEIAE11L!VAD$AT$HH HII,$t"H /DH=I贞?DAfHH͵HGH9t8HXHLAM1HI9H;TuLHlvxvpvhv`vXvPvHv@v8v0v(v vvv6LHHtFHfHH9LHuH;|M:fD1E18fDMH .HD$H=KHVHD$Hff.ATUHHDdH%(H$1L$LAeHHMlELE`H%OA8P}8%ZYHfo$foL$LHfoT$ fo\$0fod$@fol$P)$fot$`fo|$p)$fo$)$fo$fo$)$fo$)$fo$)$ )$0)$@)$P)$`)$p)$)$^~Ht)H$dH+%(u-Hĸ]A\Ð~|H ,H=F賛1f.WUHHDdH%(H$x1H$辉HHANjMlEHPLE`AXPҧ#ZYHfo$foL$HH$foT$ fo\$0fod$@fol$P)$fot$`fo|$p)$fo$)$fo$fo$)$fo$)$fo$)$ )$0)$@)$P)$`)$p)$)$Ht)H$xdH+%(u-HĀ]H C+H=HE1f.WAWAVAUIATAUHSHtHELAdHsLH1^(IHJIc* IH^HLs'IIEMcHIEI,$HEHFHH9EdLHIfII/tmMI,$1LAzImtzH *DH=<טHmtxDH[]A\A]A^A_@LXfDLHIL8I,$3L yLRH{A_KDL-IEImLAafHAcIEuLI,$LH]HLeHI$Hmt/LHL1H+IqH<dH(AayAuxAWAVIAUATUSH(H|$ HT$pHdH%(H$1HHF8 HD$ H?HxH9t:HXHHqH 1DHH9 H;TuHD$ H;`FHD$ HHHL$(HD$8H\$(HHD$pHCHD$xIFH;Ft H;E IE1HD$1H|$D$D$D$IFHwFI9^ L9IFN,IIEHt HmLÅyL;-E IEH5LHH HH( L%EH;-ZEL9u H;-.E(H^EH9EHEHPH.HPMHL$0Ho HmDIEH5ILHH HH H;-DL9H;-DH HEHHD$HHEHD$HIEH5LHH HH H;-0DL9H;-DHHEHHD$@HEHD$@IEH5?LHH IM I,$IEH5LHH>IM I)IELL$PLH5HHLLL$PIMKI(- HcD$HL$(L;CHLLPPHhSHD$@HH?HcL;%BrHEHD$0HD$0HIL; BHHH\$@HHHHIHH9rHh1HHHHcD$HHĀLD$HcD$HD$\$MI؃D$\$@HQBI9E IEHPH H A]HHcD$H|$(HLHhPH@H9L$\HcD$HHMxQT$HD$xL Ld$x6HcD$HDŽĀHHDŽHDŽL$D$H|$LOHD$LIHhHHtHoAH0H9fI.x HD$ HhH@H9t;HXH HqH 1fDHH9 H;TuH|$8HD$ I틈HD$8HH$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$HIH H;?HΔHDHpH9HXHHyH1 HH9H;TuIHEHHD$0HEHD$0/HIAI.I(fDDH J E1H=X2H|$8tHL$8HHD$HHMt ImH$dH+%(\H(L[]A\A]A^A_A]AEHH HHHcD$H|$(HLHhPH@HLHCT$H=H51],AۄA@DDH JH=9AL|"H=H9E HEHPH H5 MHL$HHHmHDCHf=H9E HEHPH H@ }H|$@HOHmH H H|$ H$@^{Ht$(HD$8SHT$H=}H5.8#AXAHHl$xfL9AOlIIEq@H|$0HD$0H9a8LHL; ;7 Hl$01M9fDHD$ pl5H8H8yH aH=o,MAHIAD3HVH\$(HcD$HHLHhPH@DL8HHtHhHmHH" DHDHH9HuH;,8.HD$ H;7zHHH9gHuH;7UH`7HJH5&HWH81H H=+ʇE1fkH3ŃA+HHLdHHH9HuH;H7nHD$ 11I틈$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$8$VHIHH;6aHWHaIT$H9CHXHH~H1 HH9H;DuIIۃAvH~H9'HXHHqH1f.H;THH9uA2IHD$0HxH9IHM9rUHl$0y|HD$01PLL$PIAHl$HHH$A]AEHH IHL; 4HHEHL$0ILl{E1 AM9HD$01H#HHHD$@M H|$@HD$0H/H\$(HcD$H1LHhPH@H3HEUHH HD$0HD$0HHHHD$@ H|$@HD$HH/dHHEUHH HD$HHD$HHIHH: I,$HD$@_LRHHDeEII Ld$@HD$@Hl$0H HD$0$EUHH HHD$0 H HD$HEUHH HHD$HHy HD$@DeEII ILd$@CAHU1H54H8bIAFA Hg2H%E1IH5C H81 H )4H=7% AE1WIAIA˃lgAvHG@HFH@HXHt2HJH~1H;tHH9uIA'HH9HuH;50HHH9FHuH;04H0HJMIHVH5KHA H81~ Hz/H5YMHA H8;HHHH9-HuH;0I&HHHRMIHv/qH5AH81 H.H5MqAH8W{H;5MAVAUATUHSHulIHUH/H9mHPHLhLp IEIH(L;-.L;-.uI9uoHELHHHHEHP(IHImteI.tm[L]A\A]A^DIH[L]A\A]A^DL`yrH H=*A~ImuL*I.uL[L]A\A]A^HLHIHKrDrH PE1H=*~ HY-H5(rH8%I,$uL@>rHxHH 0H .HEH,H5'H81 @H,rH5H81^ @rfrAWAVAUATIUHSHHEAMuZ1HH 3H= |HmtDH[]A\A]A^A_HxfDLhHH 1Lb IHHEH+H9ELHIiII/MtPI,$1L+I.Ӓ1LT$ t$t$T$ f.I,$ΒLT$ t$t$T$ fDLfLpffL= IH]HLeHI$Hmt(LHLjH+IHHff.fAWAVAUATUSHH@lAAAA~gHLgLP1DpDHtP)HcHI LH9H|PH1H0HLHtHHݐy/HՐy$HA9H[]A\A]A^A_H=H5&5uH RH=&LzUH1[]A\A]A^A_@ATIUHdH%(H$1HdhL\HH;)HHt;HEIHEHt\H$dH+%(HL]A\fDH +yE1H=&xyHEHHEHuHDL% <LH=%8yLE1*yH=%yTHAHHxH9HXHHqH~1H;THH9uH(HJH5RHWH81Hmt%H *yE1H=%uxHHW'H56H8HHH9UHuH;(Ceff.AWAVAUATUSHL%'L92LnHHHIEH/mHCLmHSH5H5HFiHEHHHHHlH/HEH57HHH|HHVHEH5HHH`IHEHHEMHuLs H&II9EI]HImHHEImtsLHHMfH+IhI.&MHm:I/uL4I$HL[]A\A]A^A_DfDLfDAAHt=DDH ^H=#vH1[]A\A]A^A_HmAAuHLLL"cI*f.{TfDH%H5z"ApA H8bfH8~L(AA (HHH$H5rAA H8V+fDHnAAH|sIff.SHL$HFH9t H;#uZ}HHt%H(tHH[fDfnH H=h!Ct1[HHH#H5#HH81nff.AWMAVAUMATUSHH(HL6HD$tYM~BAHE1D$HBHHD$L$HT$HMMHIH\$M9uH([]A\A]A^A_M~1fHLLHNH\$I9uH([]A\A]A^A_DAWAVAUIATIUHLSH(dH%(H$1Ht$@HD$8HD$0HD$(7yHVH}`HL$E1HHEHLLP0HHUH(HHHcUlHH9wfHH9H8xH-f{L%߉HEHHH=d1HLHH1HA(HmIL%DALH=qAuDL$HT$0Ht$8HHHHHH(H@LhHH@H $HL$(gDL$LDL$HH $DL$H8LPH(LpLhHHHt H/UMt I*1Mt I. HD$8H{XLk`HkhHCXHD$0HC`HD$(HChHt H/Mt ImHtHm{LDH=Dp1YL%! I<$LU`EmlH{HSPHsMMD>LI$LH$dH+%(H([]A\A]A^A_D;IIH-AA`uL%/HM'H:LU`EmlL%NH{HSPLT$HkH|$H$7H$HE1H|$DD$\a|$sH{H$MLD$DH1FL{AMD$HCH~9AtFHSXHsLHD$A $`HD$ $HL{PH9uNj|$DIHLxUkHH0L%L1H@A9uAL%H1LHH[A$dDAuAL%aqHD $D $sLD $D $KD $D $)LD $D $LD ${D $LT$D $dLT$D $HCXHCXHD$8HC`HC`HD$0HChHChHD$(HHQH5A$H8^f.AWAVAUATUSHH8LfdH%(HD$(1HD$HD$HD$ HI)L~H~ HF(L|$H|$HD$ HG4LoIEH9HHcH>f.GWHH IIHwIHH.IT$LHHH=1]HHI,$ZH5H^IHH8EHmJLzHHH=HIHHmI$HI9D$LLLXII.>HEHMHEHP1L)Im&H kH=$kI,$u LD1HT$(dH+%(&H8[]A\A]A^A_HH5فHIHVIHD$HLc@HHHSH5ATL AH #H81gXZH H=XjCHmH rH=(jDoLt$ I1 H=ˀH5̀HGHHHHHHH9EQH]HDLeHI$Hm<LHL&YH+IAI$HMI$HL;5 t2IFH;wLL6HHH(IULIUHLHD$HD$DDoAMcIGHIofDoGII DL`HPH@H06LHIUII_DLHI3MHE8HD$PHE0HD$HHE(HD$@HE LHD$8HEHD$0XIIH JcH>f.IEHIEHH5oH|$ M H5pH|$ LdmH4qIPHH/"MG@I8IWPAW0"HcHH)HT$HHIHHH9uD$,IW HH5oAGh;IHlH;XH;SI9JL"AŃjI,$D$,EoltQMg LIGHHEt0LHH|$H~!H3HH DHHH9uHHHmHfD{HHHHmHD$H_f.HL@H H5gHHKH81bKDHH-H5H819"@H H|HD$"DLHT$HT$H1fDHQH5ZH8"fH H=VMf@GHH‹GH HT$fDL8kHynIPHH/IcO0I@IWP1Iw8HT$}HHHH9u^GHH‹GH HHD$HELHD$0IH5jLHVHD$8HIH5^LHVHD$@HIMILl$0H|$8L2IH5x\LIHVHHD$0HrLeH5aLHVHtHD$HIM~H5kLHVHtHD$PIMg1HL$0MLL H5?CfzfL%]L-bkID$HHH=!1LLIMV1LTI,$fLt$T$T$t$jfL%Y]L-jID$HHH=V1LLIM1LI,$fkf.LhL绮gE1E1ML8I9MGXHHI(RI_XHH5HH8;g^fDDHL\$L\$fLL\$L\$fLLD$LD$fLh8gE1ImI.ALT$E1L\$'L\$T$E1仚ggLg댻gQgBg-IHiIHgAL1LIH5AAfH DDH=QL1LIHAAf HVigSfzLhL1LwIHcgkWHMhCHT$ g5H='gH9HD$ H@t H;`Ht$ IMnH=YL=IHAIm1L/h,I.LT$T$AAfHgHH5`H88LNAAfH2HH,HH56jL AH H81^z_3fXL%`L-eID$HHfH=k1LLIDM_1Lyh I,$rcgHnHH5QT$H8%T$KHH1L AjHH5,H8H 1AX-fAYz[L8(^IIHHL\$)gHHH5MH816Hl$Ht$ =IE1*hA(hsL1LWIHuhK7uhH3gLT$L\$OL\$T$MfUHH dH%(HD$HGHHHt$H|$HE`HEH~H}H}8/HmH|$H$Ht$7H}PHtHEPH/tpH}XHtHEXH/tIHD$dH+%(HEHH@H ]Uh}Elu.H}%h;f+fM0HU@E1Hu8H}I>H@t HHD$dH+%(uH ]DAUATUHHGHHIIH5cHHHLLHHEx=HAHEtHD]A\A]H(HD]A\A]HlHEtIH H=A;KHD]A\A]fDHW l븐Ht$ t$ HPHH54AH81BAUIATUHIHt L]A\A]HH8tLH]A\A]MAWAVIAUATUHSH(dH%(HD$110IMHHy`ID$PH$LmID$HI\$I\$ I\$(HD$HD$MIIIH 2H4AHMEIHHHbH5 AUL H81oXoZH YH=E`II,$5E1DHE(HD$H} LmH|$L,$^ŃH|$E1Ht'H;=AH;=wDu H9.EI|$IEH/Ml$HMA$I9D$'I9^~sIT$0H4LIT$H$HLmHHH H5jL AHH81Y^|oH wAQff.HcHt(HZHǀHPHiHHxHÐUHH dH%(HD$HGH}H|$HHt$HEH=H9}H9}P;H}0HtSlZ1H ZfH9HcH;DH+HHnHOH2HHÅHuH^HDÅHyHaHDHHHHH|HHHHÅH;HHDHNHMfH~?AUIATIUHS1HfH}UADHI9uH[]A\A]fDH~OAUIATIUHS1HfI}AUf *YCAHH9uH[]A\A]HHH?$L$PYD$X$H@ATIUHSHuHxHt[LH]A\fD[]A\USHHnH-OHxHEHxHtH/tIHH9t(Ht#H@@~>Hǃt=HǃH1[]Ðf0 H=R1;HHtHǃH/u覣@HIw@LODHHH8Mt?IHLHL9t8IQLHH5K+1HLLHHLIIL1H5*HOff.@GDE1AWAVAUATUHSHHWL:ILJMDBXHH5+E1H81sHEu@L8EDDL%]fUFDI?@^<4wtH=UHcH>u@AfDHHH HHHHHHuDLxHHHML9tHEfDIL9uHEHukEDE1E@HD[]A\A]A^A_HAUH &#9uIHHH1Au*1!HѺH9ut1AH[]A\A]@I9uuHDL聎HHH;H;-LuH9u8HmuHD$ bD$ @1AH[]A\A]@H8fDHM0HH@HE@ÊH:L訊9fDIt$HfDM>@M>ff.H;=H;=cu H;=u {ff.AUIATIUHSHH(dH%(HD$HGXHGXHHt$HD$HG`HG`HD$HGhHGhH|$H$dH{XH4$H|$HtH4$HtHH|$HT$HtHHtHHIUI<$HuLL`HhHHxHpMtI(tAMtI,$tEHtHmtI1HT$dH+%(H([]A\A]L8Mu뼐L(Hu븐HfDH|$IEI$HEHtH/t0H|$HtH/t8H<$HtH/tX賊ɐ諊f蛊ċ@ATUSHHWHBhHHxHHHH9FHFHPHw3H:vHtAHH[]A\HvCHH fHHu蒏HHtHHH2DtL+HCH5pHPHŶH81ۓ%fH9HRH5H81贓H1[]A\fHHIHeHtI,$HILHD$)Ht$2Ht"H:Hf1fvCHH ff.HHtCHH9t*HXHtNHJH~m1 HH9t_H;tuHfDHH5H81HHDHH9tHuH;5xtfDHHNH5HWH81X1@HHHH |H58HEHH81ff.SH:HHzAHH~U1f.HHTPH9LGxH1fIHTHHtHHÐH9HGHH;HCHG@1D t[fDH[f.HW`x3LcLWxJI)IDHHTPIHL9uKHYH5H8:HHC[@LVM1 HI9tH9|ufE1DJTHBtv@tmH9tHXHt,LAM~S1HI9t?H;TufDHDHH9tHuH;rf.IM9k1ATUSHOHxXHu1H[]A\HHH0H9ufLc`HkhHCXHC`HChH/t0MtI,$u LL@HtHmuH4f+fHGta@tXHFHt_@tVHXHtXHJ1H~H;t?HH9uH|$(H|$ݩtHHH9HuH;5@ATUSHHHxXHHu1H[]A\HH0H9Lc`HkhHCXHC`HChH/twMtI,$u LۃHtHmuHăfH/t*HSHH5H81ߍH[]A\ÐHt$膃Ht$sHGt_@tVHFHt]@tTHXHtVHJ1H~fDH;tHH9uH|$xH|$ݩtMHHH9HuH;5@AWAVAUIATAUHSH(.H=IHH@XMw`IGXIG`H$IGhIGhHD$HHH8H#H9G)L MHڮI9tL; L(E1L EAMJDDLDA9,HHLD;pIH(HEHb1HL]IHXdH6HmuHWI,$LH([]A\A]A^A_襅DHH5E1H=H蓆H$IXMW`MOhMw`IGXHD$IGhHt H/Mt I*,Mt I)5L EM|DLDD9_ADHH1H=}IHH衈HHLkI,$HHL %EMKDDLDAA9HcHLD;vD9AJDHcHLoHcƒHHAA9|ADvH.DHEHm$HH([]A\A]A^A_WEHLjHfAD9%Ej@LIcH舁IHkIcDHHD-HE9 ;LL$L$~LL$L$@LL $~L $L~H=H5RHGHHLHHH|$vH|$HHHHH9+ L~H5HL$HV HL$HIHH@HfFHH@HH5DpH(HE@IcHLHHtH%IHD$LL$K}LL$HD$*H>H.H/'}L +Hܩ HDAWAVIAUL-ATIUSHA$T7>xIcDL>AFE=AT$IT~fDs?xx}ujLII^8stvAFDHtoIN 1HHHt^HH)IN RJH% &IT@t]fDH5YHH818E1HL[]A\A]A^A_JwHu!9uAVEI1A8VDLLtIF(IIF0AFEAFFAD$A^@AFDIF(~1@B< ,AT$IL$r@ w+fD0HҍBr@ vHIIF(HH5E1H8l|A~DtI~tLI~LE1A|$:ID$t H8:uL`I~(LkjIFIt$E1HH8DX@)!Ȅ w@IF8I^(IF(A|${HD$LAFDIl$IF0IF8H#HLIHIL9uHD$HIF8@AT$ID$Jv dIĻA8VDf.A9^@AFEA8FFA~GIF(IIF0IF(2LIF(IF IAFEIF(IF0AFDAFFHiH5 E1H8Gzf oJЀ VHFɍr@ w,0 Hҍ Jr@ vKE9}IcHcHTH9,t)ue,AH4E9~tAFGLfIF(HH5 H8vyZH5mHiH5 H81Hf.IH1DDH5t H81躁HH5 H81蠁HH5@ E1H8xKHt7Ht2H9t}HOHVH9t(\Ht 1~\HtÐ1H91DF\f9G\uDGXD;FXuATUSE~11HA9~H\H9\t1[]A\@À\Su^`19_`uH_HnHt_HtH;HE11Ht9HH0HtH@H9ButAIcH@HHH:HuH|p1Hcff.AWIAVAUA͹ATIUSHHT$H|$0LD$DL$$dH%(H$1H|$(HHHD$E1fHA9IǐxHH5 H81foD$0foL$@foT$Pfo\$`fod$pA$fo$fo$AL$fo$fo$AT$ fo$fo$A\$0fo$Ad$@fo$Al$PAt$`A|$pA$A$A$A$A$H$dH+%(HL[]A\A]A^A_IcHyHHH1H$fHDHH;$tqI|wIHuHmHD$0HD$8MfI.Ls{IcxHHgDHD$H|$HXp=wIHqHIyHHH|$tHHH|$1IH:L$qxL$HIkHELp L5) HX(HhL@0IFHHH=L $qL $/1LLICyML $I)I>I$PLt$$H蕲HHzHT$(D$DD$$$$$$$$$$$$$$$$$$$$$$$$$AAAAAAAAAAAwxAwpAwhAw`AwXAwPAwHAw@Aw8Aw0Aw(Aw AwAwAwA7$HĠHm%HqLq9Lq"vL $HuH1H5H8jrL $I)AlH CDE1H=5tH|$0HfH/tHD$0HD$8Hkp1LLL $zL $HIbmAloAlI.tzH+tEMTI(JLp=LILo AlHL$oL$I.AlLoLL$oL$qpDATHGIHHt.HHH81zu.LA\HHIHH5H81gyI,$tE1LA\LofDUHHGHGHtDHt&HtxGH]vD1H]GWH]HH DGH]HaH5BH8pHfH@`HttHHthHHt^HzH9Eu.@H(Hm_HHD$nHD$HHH57gHHuH%sHuHH5{H8SoҐUHHGHGHHHHcH>1H]GH]WGHH HcʉH9tH"H5H8n@WGHH HHcʉH9uH]GH]rHcH9fHurHtLf.H@`HtkHHt_HHtUHڙH9Eu,@HHmHD$ qlD$ HH5HHuxqHjHH5OzH8mOHGtgHGHPHwQH "HcH>G@GWHH fGWHH HGH{qUHH@`HtvHHtjHHt`HH9EuHHHmt)H]HH5HHuHHHD$+kHD$pHuH×H5+yH8lfATHHGHLgID$HwuHHcH>DgAMcfDH/t HLA\jHLA\ÐDgf.DgGII DgGII IH|$#pH|$IfH@`HtGHHt;HHt1H@H;Vu:&H|$H|$I@oHt&I0H5>HHtH@HiH5wH8:kHGtgHGHPHwQH ʡHcH>G@GWHH fGWHH HGHn#HHH?Pf *YHHGH?HHH?PHHHHH?PHHHH?PHHHGH?f.|E„af.If(D„GUSH(-af/f1X|H>& &HP^YHYXH9uf(d$L$T$jT$L$f(5d$^f(\zf/YX X\vFH~AfD\{L$Hf(T$7jL$H9T$\}H(f([]f(f\H,H*Dff(AWAVATUSHH@f/ D$|$f.=zAuH@L[]A\A^A_ÐD$fWrE1r zD$f.IL$H;SL$Yf/L$wH@L[]A\A^A_fDf(ff.)Qf(D$L$hL$Y ?X ?D$09Yf(L$ \ 3\-f('\%^ +f(d$8XfI~XfI~^\fH~fDH;SH;f(\CT$ST$ ,D$f(fTy\fIn^L$XD$ YXD$XnL$f/ L,\$rfHnf/'M_5_f/v f/Gf(L$(gD$fIngL$(t$8D$D$ Y^X_g|$ID$X|$f(f(\fI*YL$0\L$D$fH*L$L$\f/L$H@L[]A\A^A_D$of(fAWfAVH*AUIATIUSHH$t H9r xMe$AEf(\Aed$Pf/l$H/ L$P|$HT$YAMA} X\$AM(f(L$l\$Y\$HH,fL$T$f.Im0f(\$h Qf(Yu\$HYo%Ow-\f(fTf.A=Xf(|$(f(D$AE8ff(H*XXXf(Au@$\f(\$xA]Pd$8^f(AeH\XfD(\$@A]X\$PYf(f(\^f(YXYf(Yt$H\\$`^A]`Yf(XYfA(AXXL$pAMhYf(D$AEpfA(^D^Xf(D$XAExfA(XD$ AMI)IEH$@H;SL$ H;YL$SL$f/L$f(ff/L$"|$l$@fH*\Yf( 9u^XL$0XD$8\X\$(fT+u^\f/T$[jT$L$0L,MI)LHIHH~#D$hYD$(fH*\f/ID$fd$P^d$HH*YI9=f(sf/$M)f/d$(MGH[]LA\A]A^A_fDf.ADEf/L$XwkL$0D$b^D$`XD$8iL,MSEJL$0\L$T$Y sYT$`fL$0D$b\$x^D$p\f(hL,M9ET$L$0\L$XY YsYT$pifDHEf(I9ff(H*H^\YI9}IFH9ff(H*H^\^H9~Vff(^ LfX t$hIf(YXX ^HH*^^f(XL$(YL$0\$`\$L$0f(f(\f/XT$0f/IFffH*fEHEH*$fL*ID$L)H*fD($$DYfD(DYfA(fA(D$^AYfD(l$DYD$D$$D$_D$$D$fA(^D$_|$Y|$P$\$Hf($Y^_5k$-bfD(D$D$f(D$^f(D$D$DY$D$$\%f(^\fD(f(^D\D^ f(A\fEM*DXL$(DY$f(T$0A^DEXfEM*EYfD(EXA^AXfD(E^E\fD(E^E\fD(E^E\fD(E^fD(E\E^E^AXfD(E^A^E\fD(E^\A^E\fD(E^\A^E\fD(E^\A^E\D^\^L$A^E^DXDXfA/u$YT$\XdL,qf.H,ffUH*f(fT\fVf(f(l$P\t$Hf.Blfb8jz r`d$b@YHj0t$`rp$bHl$P-d$8bPYt$d$xbX|$Hd$@bht$ d$pbx\$hd$Xl$(D$hT$L$PdT$L$Rff.@AUIATIUSHH8$t H9rL$=_mfMeI*AE\A]A} f(T$|$[T$YT$d$T$L$D$YAEfYA]XX lf.Q%f(YXf/-H,Im0H;Sd$E1f/f(wWH;SL$E1f/vAI@H9|Lf\L)IH*Y$YfH*YT$^f/wH8L[]A\A]@f.Br Hj0t$rt$Lf(\$(T$ L$mb%\$(T$ L$YXf/f(d$ \$,b\$d$ f(@YXH,fDHf.EunfUHH*f/rYf/rI]{k\Yf/f(r&QIH]L)fD1D]rfkIH]L)HjHH?D$\$PL$$f/vf(fDYHXf/wHÐ1HATIUHSHHH HHH HHH HHH HHH HH H HH uI<$AT$!H9r[]A\I<$AT$H!H9rDff.@AWAVAUATIUSHHtpHHH?IIH9wmHEAEDjAME9v(D1AAA9sH}UAME9wI IHL[]A\A]A^A_@HEHEuQLbIHHI9v+L1HIIH9sfDH}UIHI9wHL$DHEH}L!I9rL$oDHEH}D!D9wL$ODЉI>@I0fDAWAVAUATAUSHtiHGIH?AEudJL$ Dl$ AME9v)1AAA9sDI>AVAME9wI AHD[]A\A]A^A_@IFI>D!9rD$fAfAWAVAUATUSHt$Lt$PfAIALf$AEDjH?AT$AAAfA9AAǙAf9r1@t$ I<$AT$t$ AAfA9vptA.I<$AT$AAD!f9stA.AD!f9rfD$H[]A\A]A^A_D$H[]A\A]A^A_A.AtA.AD$fADH?AT$Aff.AWAVAUATUSHt$ Ld$PAIALˀ&AEDrH?AUA$DA$$A8AAAǙ8r-I}AUA$DA$$@8votA,$DI}AUA$A$D!@8stA,$A$D!@8rD$ H[]A\A]A^A_fDD$ fH[]A\A]A^A_fDA,$A tA,$AD$ A$@H?AUA$ff.USHHl$ tAHL˅tmAEH[]fDH?QEAWAVAUIATIUSH(LL$Hu.LIH~fL#HH9uH([]A\A]A^A_DHHH9FEH~ĉDrE1Dt$D$@H;St$I9s$D$19sfDH;SI9wHD$H LJ IM9uVf.HEHLrE1Hl$H!DH;SIHHI9v-HD$1IHH9sf.H;SIHH9wHt$HLJIM9ufH1DH;SHt$LHHI9uH1DH;SHt$LHHI9uZfDIII LHI LHI LHI LHI LH I HE1H;SL!H9rHt$LJIM9uIII LHI LHI LHI LHI HE1H;SD!9wHt$LJIM9uufAWAVAUIATAUSHL $u0LIH~@D#HH9uH[]A\A]A^A_DHՃEu}DrE1Dt$l$ H~fDH;St$I9v$D$ 19sfDH;SI9wH$H DB IM9uH[]A\A]A^A_ÐIII LHI LHI LHI LHA H E1H;SD!9rH$DBIM9uH[]A\A]A^A_H1DH;SH$DHI9uH[]A\A]A^A_AWAVAUAATUSLHfu1IIH~fD+HH9uH[]A\A]A^A_@IfCEH~DpI4I1H4$A1D$ uz@I?AWfD9s@D$ Af9s1t1fA9vI?AWȉfA9wۿHDfKH;$;t1뒐AHHH HHH HHH HH HM4I11t@1!fA9sI?AW!fA9rDHfSI9uH[]A\A]A^A_@HI,I11'DI?AWALHfKH9ct1ff.AWAVIAUATUSH@t$ uRH~8H@HL[]A\A]A^A_`Kf@t$ H@sH9uUH[]A\A]A^A_fDIEH~I, DzLˉD$ 11Af.8I}AUDA8nD+D$ EϙAA@8rWDl$D|$l$D$fD$l$l$f|$l$D$fD$l$l$fl$D|$l$D$ fD$l$l$ff.l$w9D'=7G'DFgB7FDAG@7A;<GABCl$l$t@>wAGAGDD=A@?DfTNNf(fTf.v3H,f%eNfUH*f(fT\fVf(f.DNf(fTf.v3H,ff(%NfUH*fTXfVf(f.MDf(fTf.vXfU\fVf(f Mf(fTf.vH,ffUH*fVf(ff.wQCg=;D=wB7<g8D7CG6:AUIATL%NUH-FSH>H(I}AUfHH HH*YLH9IEI}AL$HcAT$\$L$D$f(fW%Gf(CT$L$f($YD$Xf/KH(f([]A\A]fDfW YH([]\A\A]f(fDH~?AUIATIUHS1HfLADHH9uH[]A\A]DHD$}YD$HfHD$]^D$B\JHDf(ff.zuHT$ JT$H^.6ff.HD$fWB PJf(^L$H\f(5ff.fHD$fXf.wQYD$H@fH~WAUIATIUHS1HfI}AUfW fWAHH9uH[]A\A]fff.@H~oAUIATE1UHSHHI}AUf *YmWZZWBDIL9uH[]A\A]AWAVIAUL-ATL%UHSHH(IFI>L$HcT$\$L$%dD$Yf(Y@T$L$f($YD$Xf/wAI>AVfII LH!H*AYLtfW iI94?H(f([]A\A]A^A_@IFI>fW72 I>Y $AVfW  $fWf(XYf/vX fAzfW mff.H~?AUIATIUHS1HfL8ADHH9uH[]A\A]DUHH0f.FD$z^|$ff.zftv=Ff/|$@H}UHD$ Ft$T$\f/r= }FD$f(^2\$f(f/rH0f(]D$:FL$ \^D$4|$L$ D$f(Y\f( E^1T$\$f(\f/H0f(]f|$\=Y|$(ff.Q5E^t$@Hff(D$YX^Ef/sf(L$H}YYD$UL$3f(YYYE\f/wbL$ v3D$D$e3%D\d$f(L$ f(X5YD$(YYXf/D$'L$(YL$H0]f(fH0];ff.HL$MYD$HfY|$Yfl$Xf.tQXL$t$f(Xf.dQ\l$ f(f(X^f(YXXL$^L$0-@f(^f(W,XD$L$\f/D$s{H}UY{,\$0H}f(YXXL$f(L$ ^\d$(YL$UL$\Y\f/T$\H}UD$D$(, f/L$vfWT$8 FXf(fT=T$Xs+|$T$\f/vfW1H@]5<f(t$t$ ^Xft$D$0H?UX\v<YH@]f\<^D$ f.weQD$HFYD$ `XD$8f/vXTf/,E\>H@]"~2D$f(m2f(~f(L$U2L$f(fDSHH0D$ fWD$(H;Sf/D$ D$H;SYD$(3R;T$\f(Yf/~f(T$\$)\$D$f()L$^:X 0L,M]T$ff.E„AH0L[f/ArAff.HD$~T$$fWf(~ $f(fWf(^0HH,Ðf/$r mDG#w#,,w%fT $fT9fV&G#G)W-'#)W%$G-T@t(T.v,,f%U*(T\V(@$(T.v,,f(%U*TXV(@x(T.vXU\V(ff.@ 8(T.v,fU*V(ff.f.wQI)f(&"+(g'W"#%&AVAUIATL%̅UH-čSHHI}AUfɉ *YL9IEI}AL$HcAT$ \D$L$A(W-A ( T$ L$(fA*YkYD$X/HH([]A\A]A^f *Y3Wt +H[]\A\A]A^(H~?AUIATIUHS1HfLADHH9uH[]A\A]DAWAVIAUATL%JhUH-BdSH:`HIFI> HcL$ \D$%fAnfZAYA f(Y,L$ f(fA*YYD$XZf/wDI>AVfAA A*AYfA~t WfA~D9l/HfAn[]A\A]A^A_IFI>f *YW0 I>YL$AVf *YLWL$W{(XY/vX /AfA~RW JfA~AH~?AUIATIUHS1HfLADHH9uH[]A\A]DUHH .D$zd$f.z f=n/|$%Yd$DH}UfH *YT$T$  "t$T$ \/r5 D$ (^#&\$ (/rH (]@D$ʼL$\^D$!|$L$D$ (Y\( ^%T$ \$(\/H (]t$\5ZfRYt$.5Q=%^d$|$H8f(D$YX߻/s(L$H}YYD$ UL$ƻf (*YYD$YY\/w_L$R D$D$ A =Y\|$ (L$(XWYD$YYX/D$L$YL$ H ](fH ]A"ff.HL$ YD$ Hfg(#w#T TVf.w%$.(z$f.z(tf/w@YYܹHSYǹHfDYff.@.(zt"(fH\/w/sIH@Xh(@WiT$ NT$ HXf.L$ %L$ HXÐ.(zt"(fH\/w/sIH@X(@WٺT$ T$ HXf.L$ L$ HXÐf(f(f.z&ff.zt"ff/w-ff.@YYHYHfDYԹoff.@f.f(zt(f(fH\f/w"f/sDHXf(fWT$pT$HXÐL$PL$HXÐf.f(zt(f(fH\f/w"f/sDHX,f(fWT$eT$HXÐL$EL$HXÐH(l$0l$@-Ƹvw6@zrz t.@|$@|$0H(Dz, z|$<$\,$l$|$<$Dl$,$|$<$,,$l$vH~+f(̷fTf.v fTf.w f.zjf. ztHDf.z*f.zL$$L$$HL$$$L$L$$pL$$vH(=T.v T. w.zp. ztHl!@.z3.zL$ D$L$ D$H8!L$ D$D$L$ 뮐L$ D$L$ D$nff.S(HH.h(L$\$L$f\$(.z(.\^z//8t X\%.z/u-^(T$((T$H[(T$d$<d$T$\/%qvXKH[Ð(\$ d$L$\$ d$fL$(T.ݺ(^E„Q.HD$T$ T$D$&L$D$ L$\$vff.HdH%(HD$1. zt-H|$#D$HD$dH+%(u&HHD$dH+%(u HfH(fdH%(HD$1.zt-H|$(HD$dH+%(uU(H((.ú^ET$ u.z T$  T$ fDSf(HH f.vf(L$\$L$f\$f(f.zf(f.\^zf/f/8t X\%&f.z-u+f(T$^f(T$H [@f(T$d$d$T$\f/%ðvXI&H [f(\$d$L$U\$d$fL$f(KfDf.ݺf(^E„Kf.AD$T$& T$D$DL$D$/ L$\$hff.HdH%(HD$1f. ȱzt,H$HD$dH+%(u(Hf.HD$dH+%(u HfH(fdH%(HD$1f.zt,H|$f(HD$dH+%(uSf(H(Df(f.ú^ET$uf.z T$@ T$fDl$z,zl$tw @Dff.l$-ff.@l$-ff.@AuA1ADuDDADÐATASHAtD1AAHD[A\ff.IHHu I1ILHuL@ILfATISHHIHtL1IHIHL[A\IHHu I1ILHuL@ILfATISHHIHtL1IHIHL[A\AAAHAHuDIHIIHHIHIHIIHHIH#AAAHAHEDIHIIHHIHIHIIHHIH@w@ljD1D@w@ljD1D@w@ljD1D@w@ffwljfD1DfwljfD1DfwljfD1Dfw@fÐ1 Bff.@1 Bff.@1 Bff.@Gff.H1HH@HBff.H1HH@HBff.H1HH@HBff.HHH?HH?HGfDH1HH@HBff.H1HH@HBff.H1HH@HBff.HHH?HH?HGfD(T .  cv (Df~f~(%=#u-D$L$. zNuLL$\(fDBt5fnȁ(YT$T$@ fDXff(fT pf.  Xvf(ffH~fH~H ȉ%=~Z-f( кu;%==f(YL$L$H H fHn\f(@ uDHD$L$f. zuL$\DHfHn\fDgq%=df(XH8dH%(HD$(1l$@-v$nHD$(dH+%( H8DHD$Hl$@|$l$@z?u=HD$D$l$<$,$zu,$l$@wDHL$@qt$H L$uP% ‰T$l$tIu<$,$HT$@HH ul$@z?D$l$Yfl$'' SHHJ HtD$H[ fDHdH%(HD$1H|$D$HT$dH+%(uHff.@ATeAąuDA\vDA\HdH%(HD$1H|$D$HT$dH+%(uHOff.@"ffffAVIAUATUHHLoI@@1 IHtY1HL!ImI#Mt9IL$@MMHu9LLMt I,$H]A\A]A^fE1HtH;-1twHEHt @ud@t}HEHL} Hq1H52H8H]A\A]A^7HL]A\A]A^D1@H1 HHu4fH 1H5 H8DL8H0LH5OH81UAWAVAUIATIUSHHLw0LHo8H=rAVuA~$CtVHKHHS@HtHEHHEE1LQLLR1PUj H0I HL[]A\A]A^A_Ht+HEMu IcVH9t/HKHHS@E1IcVL9t!HKHHS@yHuLLkILLLkIvATUHSHHGH;0ttH;M/t;HXhHHCHt HHH[]A\Ht HyHGtH;EsEHDHH[]A\Ht HyHGtH9EvHUHHH[]A\HIHt`HHI,$uLHD$HD$f.HH@Ht$Ht$HxHHC#1H.Ht$H8UHt$tHt$HCHt$@UHH$H$H$L$L$t@)$ )$0)$@)$P)$`)$p)$)$dH%(H$1Hl$ IH$HHD$LWH$$D$0HD$6HH= gAWAVAUATIUHSJHHHL$L|$(Lt$ LL$Ll$0dH%(HD$81HD$ HD$(HD$0LLLLOH Ht$ HHHfDHPHHtH92uHT$(H|$H)HHFfH%f.HFH9GHJHHt5H9H9uHL$(H|$HH)H:H ?Ht$ f.H9u&fDHFH9GHH9HEH8H9uHT$HH5H+H81Y HT$8dH+%(HH[]A\A]A^A_ÐHT$VHT$x:Ht$ DsHuHt$ HT$xdHt$ bHH9IHHT$H5ǠPHFH+HT$H5vH81.HHt$ UDG<4w5HHcH>f.f.HHu*@H5bH811Hff.f.f.H9+HATUHSHHFHHW@@HXHt7HJH~i1f.HH9tRH;luH[]A\fHH9tHu1H;-)H|$'H|$uIL9uXH1[]A\f.HH[]A\H^H~1 HH9tH;|ukfE1JtH9OHGm@`HFHtR@tIHXHt\HJ1H1M~HLLH5LLd$I9uH8[]A\A]A^A_fL;l$uLH8HLI[]A\A]A^A_fHt$t$KHfl$|$黽ff.Hl$ l$ztwsPHÐHD|$H<$JH<$辽l$0H H|$ H<$H<$芽l$@H Hff.Hl$ l$ztwsPHÐHD|$H<$*H<$l$0H H|$ H<$H<$zl$@H Hff.SHH l$@l$0|H<$|$Pt$Ht$H H l$@z l$0z @8tzDuH<$l$0H |$<$kl$ H  f.;H [|$H<$|$fl$XZԕl$v;H [<$H |$|$`<$H ,$l$@Al$0l$0E„OU|$<$il$,$?Dl$@tH(dH%(HD$1l$0l$@zt6HH |$<$l$ H HD$dH+%(u*H(@HD$dH+%(u|$@|$0H(fH8dH%(HD$(1l$PztDfDH|$H<$t$Xt$XwH HD$(dH+%(uGH8fDl$@<$E„uz#,$fDA,$EHHat leastat mostbeta_generator.pyxBitGeneratorexactlyrandomstandard_exponentialintegersan integer is requiredassignmentbuffer dtypeBuffer not C contiguous.choiceuint64_t__init__.pxdstringsourcetype.pxdbool.pxdcomplex.pxdbit_generator.pxd%.200s() takes %.8s %zd positional argument%.1s (%zd given)numpy.random._generator.Generator.betanumpy.random._generator.Generator.exponentialnumpy.random._generator.Generator.__getstate__ while calling a Python objectNULL result without error in PyObject_Callnumpy.random._generator.Generator.__init__numpy.random._generator.Generator.__repr__numpy.random._generator.Generator.__str__numpy.random._generator.Generator.__setstate__numpy.random._generator.Generator.__reduce__numpy.random._generator.Generator.randomnumpy.random._generator.Generator.standard_exponentialnumpy.random._generator.Generator.integersnumpy.random._generator.Generator.bytes'%.200s' object is unsliceablenumpy.random._generator.Generator.choicetoo many values to unpack (expected %zd)'%.200s' object does not support slice %.10sBuffer has wrong number of dimensions (expected %d, got %d)Item size of buffer (%zu byte%s) does not match size of '%s' (%zu byte%s)Buffer exposes suboffsets but no stridesBuffer and memoryview are not contiguous in the same dimension.Buffer not compatible with direct access in dimension %d.Acquisition count is %d (line %d)[[ \lY\X`8`]]]nkk lc$lmu)ooo omuvvvPܑ4ƫhxX pg^ULȬ/]?333333?Cnoncentral_fnoncentral_chisquarestandard_cauchystandard_tvonmisesparetoweibullpowerlaplacegumbellogisticlognormalrayleighwalduniformtriangularstandard_normalstandard_gammanumpy.random._generator.Generator.normalnumpy.random._generator.Generator.gammanumpy.random._generator.Generator.fnumpy.random._generator.Generator.noncentral_fnumpy.random._generator.Generator.chisquarenumpy.random._generator.Generator.noncentral_chisquarenumpy.random._generator.Generator.standard_cauchynumpy.random._generator.Generator.standard_tnumpy.random._generator.Generator.vonmisesnumpy.random._generator.Generator.paretonumpy.random._generator.Generator.weibullnumpy.random._generator.Generator.powernumpy.random._generator.Generator.laplacenumpy.random._generator.Generator.gumbelnumpy.random._generator.Generator.logisticnumpy.random._generator.Generator.lognormalnumpy.random._generator.Generator.rayleighnumpy.random._generator.Generator.waldnumpy.random._generator.Generator.uniformnumpy.random._generator.Generator.triangularnumpy.random._generator.Generator.standard_normalnumpy.random._generator.Generator.standard_gammat 8 13000negative_binomialpoissonzipflogseriesmultivariate_normalvhunumpy.PyArray_MultiIterNew2Missing type objectnumpy.PyArray_MultiIterNew3numpy.random._generator.Generator.negative_binomialnumpy.random._generator.Generator.poissonnumpy.random._generator.Generator.zipfnumpy.random._generator.Generator.geometricnumpy.random._generator.Generator.logseriesnumpy.random._generator.Generator.hypergeometriclocal variable '%s' referenced before assignmentnumpy.random._generator.Generator.multivariate_normalCannot convert %.200s to %.200snumpy.random._generator.Generator.binomialRRL LlRRR@L8LRRdSlLdLLS0DTBN9N0NoPoDo8o,ooŃH9dirichletmultinomialnumpy.PyArray_MultiIterNew1multivariate_hypergeometricnumpy.random._generator.Generator.dirichletnumpy.random._generator.Generator.multinomialnumpy.random._generator.Generator.multivariate_hypergeometriccan't convert negative value to size_t A@E>A>A|\|ldYd|||kkk?permutationdefault_rngpermutedshufflenumpy.random._generator.Generator.permutationhasattr(): attribute name must be stringnumpy.random._generator._check_bit_generatornumpy.random._generator.default_rngnumpy.random._generator.Generator.permutednumpy.random._generator.Generator.shufflejoin() result is too long for a Python string (x08%$'0&$View.MemoryView.Enum.__init__'NoneType' is not iterableExpected %.16s, got %.200sView.MemoryView._unellipsifyView.MemoryView._err_dimView.MemoryView.memview_sliceIndex out of bounds (axis %d)memviewsliceobjView.MemoryView._errtuple__pyx_unpickle_Enum__cinit__formatView.MemoryView._err_extentsDimension %d is not directView.MemoryView.memoryview.setitem_indexedView.MemoryView.array.memview.__get__PyObject_GetBuffer: view==NULL argument is obsoleteView.MemoryView.array.__getbuffer__View.MemoryView.array.__reduce_cython__View.MemoryView.array.__setstate_cython__View.MemoryView.memoryview.__getbuffer__View.MemoryView.memoryview.__reduce_cython__View.MemoryView.memoryview.__setstate_cython__View.MemoryView._memoryviewslice.__reduce_cython__View.MemoryView._memoryviewslice.__setstate_cython__View.MemoryView.array.get_memviewView.MemoryView.memoryview.ndim.__get__View.MemoryView.array.__getattr__View.MemoryView.array.__getitem__View.MemoryView.get_slice_from_memviewView.MemoryView.memoryview.is_c_contigView.MemoryView.memoryview.is_f_contigView.MemoryView.Enum.__reduce_cython__View.MemoryView.memoryview.__repr__View.MemoryView.memoryview.__str__View.MemoryView.memoryview.itemsize.__get__integer division or modulo by zerovalue too large to perform divisionView.MemoryView.pybuffer_indexView.MemoryView.memoryview.get_item_pointerView.MemoryView.memoryview.is_sliceView.MemoryView.memoryview.convert_item_to_objectView.MemoryView._memoryviewslice.convert_item_to_objectView.MemoryView.memoryview.assign_item_from_objectView.MemoryView._memoryviewslice.assign_item_from_objectView.MemoryView.memoryview.shape.__get__View.MemoryView.memoryview.strides.__get__need more than %zd value%.1s to unpack'NoneType' object is not iterableSubscript deletion not supported by %.200sView.MemoryView.memoryview.__setitem__View.MemoryView.memoryview.size.__get__View.MemoryView.memoryview_fromsliceView.MemoryView.memoryview_copy_from_sliceView.MemoryView.memoryview.copyView.MemoryView.memoryview.copy_fortranAll dimensions preceding dimension %d must be indexed and not slicedView.MemoryView.slice_memviewsliceStep may not be zero (axis %d)View.MemoryView.memoryview.__getitem__Cannot transpose memoryview with indirect dimensionsView.MemoryView.transpose_memsliceView.MemoryView.memoryview_copyView.MemoryView.memoryview.T.__get__'NoneType' object is not subscriptableView.MemoryView.__pyx_unpickle_Enum__set_stateView.MemoryView.Enum.__setstate_cython__View.MemoryView.assert_direct_dimensionsView.MemoryView.memoryview.setitem_slice_assign_scalarView.MemoryView.__pyx_unpickle_EnumArgument '%.200s' has incorrect type (expected %.200s, got %.200s)View.MemoryView.array.__cinit__Argument '%.200s' must not be Noneobject of type 'NoneType' has no len()expected bytes, NoneType foundView.MemoryView.array.__setitem__View.MemoryView.memoryview.__cinit__View.MemoryView.memoryview.suboffsets.__get__View.MemoryView.memoryview.nbytes.__get__View.MemoryView.memoryview_cwrapperView.MemoryView.copy_data_to_tempView.MemoryView.memoryview_copy_contentsView.MemoryView.memoryview.setitem_slice_assignmentd6) y'bool''char''signed char''unsigned char''short''unsigned short''int''unsigned int''long''unsigned long''long long''unsigned long long''complex float''float''complex double''double''complex long double''long double'a structPython objecta pointera stringendunparseable format stringname__loader__loader__file__origin__package__parent__path__submodule_search_locations__pyx_capi__name '%U' is not definedcannot import name %Snumpy/random/_generator.c%s (%s:%d)%d.%d%sbuiltinscython_runtime__builtins__4294967296complexnumpyflatiterbroadcastndarraygenericnumberunsignedintegerinexactcomplexfloatingflexiblecharacterufuncnumpy.random.bit_generatorSeedSequenceSeedlessSequencenumpy.random._commondoubleLEGACY_POISSON_LAM_MAXMAXSIZE_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 pointernumpy.import_arraygetbuffer(obj, view, flags)init numpy.random._generatorbase__reduce_cython____setstate_cython__Tstridessuboffsetsndimitemsizenbytesis_c_contigis_f_contigcopycopy_fortranmemview__getattr___bit_generator__getstate____setstate____reduce__numpy.random._generator.Enumnumpy.random._generator.arrayb&%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%&%%Z&%%%%%&&%%&%%z&r&j&%%(&%%%%%%%%%%%%% &&E&%0&&&&%%&%%%&&%& *)) *)))))+*))*))***)))))))))))))))) * *F+)3+++*))*))) **) *L)D(D(L)D(D(D(D(D(,*D(D(*D(D(***D(D(D(D(D(D(D(D(D(D(D(D(D(D(D(D(L)L)*D(**,*D(D(*D(D(D(L)*D(L)NMMMMMMMMM3LMM3LMMMMMMMMMMMMMMMMMM3LMMMMMMMjNMMMMMMMMMMMMMMMMMKNM.L,MMZZlZ|ZZ\ \[[[T]]]4]D]`^x^@^H^P^ 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() zigzerosyou are shuffling a 'x 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() warningswarn 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() 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() updateunpackunique 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. high - low must be non-negative. 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. 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() unable to allocate shape and strides.unable to allocate array data.uint8uint64uint32uint16>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.default_rng().triangular(-3, 0, 8, 100000), bins=200, ... density=True) >>> plt.show() toltobytes__test__takeswapaxessvdsum(pvals[:-1].astype(np.float64)) > 1.0. The pvals array is cast to 64-bit floating point prior to checking the sum. Precision changes when casting may cause problems even if the sum of the original pvals is valid.sum(pvals[:-1]) > 1.0sum(colors) must not exceed the maximum value of a 64 bit signed integer (%d)sumsubtractstructstringsourcestrides__str__stopstepstatestart 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? Our null hypothesis will be the absence of deviation, and the alternate hypothesis will be the presence of an effect that could be either positive or negative, hence making our test 2-tailed. Because we are estimating the mean and we have N=11 values in our sample, we have N-1=10 degrees of freedom. We set our significance level to 95% and compute the t statistic using the empirical mean and empirical standard deviation of our intake. We use a ddof of 1 to base the computation of our empirical standard deviation on an unbiased estimate of the variance (note: the final estimate is not unbiased due to the concave nature of the square root). >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> t -2.8207540608310198 We draw 1000000 samples from Student's t distribution with the adequate degrees of freedom. >>> import matplotlib.pyplot as plt >>> s = np.random.default_rng().standard_t(10, size=1000000) >>> h = plt.hist(s, bins=100, density=True) Does our t statistic land in one of the two critical regions found at both tails of the distribution? >>> np.sum(np.abs(t) < np.abs(s)) / float(len(s)) 0.018318 #random < 0.05, statistic is in critical region The probability value for this 2-tailed test is about 1.83%, which is lower than the 5% pre-determined significance threshold. Therefore, the probability of observing values as extreme as our intake conditionally on the null hypothesis being true is too low, and we reject the null hypothesis of no deviation. standard_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 * rng.standard_normal(size=...) rng.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 standard_normal 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() stacklevelsqrtsortsizesigmaside shuffle(x, axis=0) Modify an array or sequence in-place by shuffling its contents. The order of sub-arrays is changed but their contents remains the same. Parameters ---------- x : ndarray or MutableSequence The array, list or mutable sequence 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]]) shuffleshape__setstate_cython____setstate__seedsearchsortedscalesafertol__rmatmul__rightreversedreturn_indexreshapereplace__reduce_ex____reduce_cython____reduce__reduce 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 ravelrange 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]]) randomraise__pyx_vtable____pyx_unpickle_Enum__pyx_type__pyx_state__pyx_result__pyx_getbuffer__pyx_checksum__pyx_PickleErrorpvalsprodprobabilities do not sum to 1probabilities contain NaNprobabilities are not non-negative power(a, size=None) Draws samples in [0, 1] from a power distribution with positive exponent a - 1. Also known as the power function distribution. Parameters ---------- a : float or array_like of floats Parameter of the distribution. 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)') _poisson_lam_max poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. Parameters ---------- lam : float or array_like of floats Expected number of events occurring in a fixed-time interval, must be >= 0. A sequence must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. 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)) pickle_pickle permuted(x, axis=None, out=None) Randomly permute `x` along axis `axis`. Unlike `shuffle`, each slice along the given axis is shuffled independently of the others. Parameters ---------- x : array_like, at least one-dimensional Array to be shuffled. axis : int, optional Slices of `x` in this axis are shuffled. Each slice is shuffled independently of the others. If `axis` is None, the flattened array is shuffled. out : ndarray, optional If given, this is the destinaton of the shuffled array. If `out` is None, a shuffled copy of the array is returned. Returns ------- ndarray If `out` is None, a shuffled copy of `x` is returned. Otherwise, the shuffled array is stored in `out`, and `out` is returned See Also -------- shuffle permutation Examples -------- Create a `numpy.random.Generator` instance: >>> rng = np.random.default_rng() Create a test array: >>> x = np.arange(24).reshape(3, 8) >>> x array([[ 0, 1, 2, 3, 4, 5, 6, 7], [ 8, 9, 10, 11, 12, 13, 14, 15], [16, 17, 18, 19, 20, 21, 22, 23]]) Shuffle the rows of `x`: >>> y = rng.permuted(x, axis=1) >>> y array([[ 4, 3, 6, 7, 1, 2, 5, 0], # random [15, 10, 14, 9, 12, 11, 8, 13], [17, 16, 20, 21, 18, 22, 23, 19]]) `x` has not been modified: >>> x array([[ 0, 1, 2, 3, 4, 5, 6, 7], [ 8, 9, 10, 11, 12, 13, 14, 15], [16, 17, 18, 19, 20, 21, 22, 23]]) To shuffle the rows of `x` in-place, pass `x` as the `out` parameter: >>> y = rng.permuted(x, axis=1, out=x) >>> x array([[ 3, 0, 4, 7, 1, 6, 2, 5], # random [ 8, 14, 13, 9, 12, 11, 15, 10], [17, 18, 16, 22, 19, 23, 20, 21]]) Note that when the ``out`` parameter is given, the return value is ``out``: >>> y is x True 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]]) _pcg64 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() packp must be 1-dimensionalpout must have the same shape as xout must be a numpy arrayoutorderoperator' object which is not a subclass of 'Sequence'; `shuffle` is not guaranteed to behave correctly. E.g., non-numpy array/tensor objects with view semantics may contain duplicates after shuffling.objnumpy.random._generatornumpy.linalgnumpy.core.umath failed to importnumpy.core.multiarray failed to importnumpy.core.multiarraynumpynsample > sum(colors)nsample must not exceed %dnsample must be nonnegative.nsample must be an integernsamplenpnormalize_axis_index 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)) 0.0 # 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() noncno default __reduce__ due to non-trivial __cinit__ngood + nbad < nsamplengood__new__negative dimensions are not allowed 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 mumode > rightmodemethod must be one of {'eigh', 'svd', 'cholesky'}method must be "count" or "marginals".methodmemviewmemory allocation failed in permutedmean must be 1 dimensionalmean and cov must have same lengthmeanmay_share_memorymax__matmul__marginals__main__lowlong 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() logical_orlocklocless_equallessleft == rightleft > modeleft laplace(loc=0.0, scale=1.0, size=None) Draw samples from the Laplace or double exponential distribution with specified location (or mean) and scale (decay). The Laplace distribution is similar to the Gaussian/normal distribution, but is sharper at the peak and has fatter tails. It represents the difference between two independent, identically distributed exponential random variables. Parameters ---------- loc : float or array_like of floats, optional The position, :math:`\mu`, of the distribution peak. Default is 0. scale : float or array_like of floats, optional :math:`\lambda`, the exponential decay. Default is 1. 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) lamkappaitemsize <= 0 for cython.arrayitemsizeitemissubdtypeisscalarisnativeisnanisfiniteintp 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. integersint8int64int32int16index__import____imatmul__ignoreid 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! high - low range exceeds valid boundshigh - lowhighhasobject 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() greatergot differing extents in dimension %d (got %d and %d)__getstate__ 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 _generator.pyx__generator_ctor 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() fullfortranformatfloat64float32flagsfinfo 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. __exit__errorequalepsenumerate__enter__endpointencodeempty_likeemptyeighdtype_is_objectdtypedoubledot 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") __dict__dfnumdfdendfdefault_rng (line 4570)default_rngcumsumcovariance is not positive-semidefinite.cov must be 2 dimensional and squarecovcount_nonzerocountcopytocopycompatcolors must be a one-dimensional sequence of nonnegative integers not exceeding %d.colorscollections.abccline_in_traceback__class__cholesky choice(a, size=None, replace=True, p=None, axis=0, shuffle=True) Generates a random sample from a given 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. Default is True, meaning that a value of ``a`` can be selected multiple times. p : 1-D array_like, optional The probabilities associated with each entry in a. If not given, the sample assumes a uniform distribution over all entries in ``a``. 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 Notes ----- Setting user-specified probabilities through ``p`` uses a more general but less efficient sampler than the default. The general sampler produces a different sample than the optimized sampler even if each element of ``p`` is 1 / len(a). 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 uniform random sample from a 2-D array along the first axis (the default), without replacement: >>> rng.choice([[0, 1, 2], [3, 4, 5], [6, 7, 8]], 2, replace=False) array([[3, 4, 5], # random [0, 1, 2]]) 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 check_valid must equal 'warn', 'raise', or 'ignore'check_validcastingcapsulec bytes(length) Return random bytes. Parameters ---------- length : int Number of random bytes. Returns ------- out : bytes String of length `length`. Examples -------- >>> np.random.default_rng().bytes(10) b'\xfeC\x9b\x86\x17\xf2\xa1\xafcp' # random both ngood and nbad must be less than %dbool_bit_generator 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 39%. basebaxisatol at 0x{:X}astypeascontiguousarrayasarrayarrayarangeanyalpha <= 0alphaallocate_bufferallclosealladda must be a sequence or an integer, not a must be a positive integer unless no samples are takena cannot be empty unless no samples are takena and p must have same sizea)(When method is "marginals", sum(colors) must be less than 1000000000.When method is 'count', sum(colors) must not exceed %dView.MemoryViewValueErrorUserWarningUnsupported dtype %r for standard_gammaUnsupported dtype %r for standard_normalUnsupported dtype %r for standard_exponentialUnsupported dtype %r for randomUnsupported dtype %r for integersUnable to convert item to objectTypeErrorTSequenceRuntimeWarningRange exceeds valid boundsProviding a dtype with a non-native byteorder is not supported. If you require platform-independent byteorder, call byteswap when required.PickleErrorPCG64OverflowErrorOut of bounds on buffer access (axis %d)ONotImplementedErrorMemoryErrorKInvalid shape in axis %d: %d.Invalid mode, expected 'c' or 'fortran', got %sInvalid bit generator. The bit generator must be instantiated.Insufficient memory for multivariate_hypergeometric with method='count' and sum(colors)=%dIndirect dimensions not supportedIndexErrorIncompatible checksums (%s vs 0xb068931 = (name))ImportErrorGenerator.zipf (line 3098)Generator.weibull (line 1972)Generator.wald (line 2637)Generator.vonmises (line 1791)Generator.uniform (line 861)Generator.triangular (line 2705)Generator.standard_t (line 1685)Generator.standard_normal (line 965)Generator.standard_gamma (line 1137)Generator.standard_exponential (line 389)Generator.standard_cauchy (line 1620)Generator.shuffle (line 4389)Generator.rayleigh (line 2568)Generator.random (line 235)Generator.power (line 2071)Generator.poisson (line 3025)Generator.permuted (line 4236)Generator.permutation (line 4498)Generator.pareto (line 1874)Generator.normal (line 1035)Generator.noncentral_f (line 1394)Generator.noncentral_chisquare (line 1540)Generator.negative_binomial (line 2949)Generator.multivariate_normal (line 3453)Generator.multivariate_hypergeometric (line 3822)Generator.multinomial (line 3658)Generator.logseries (line 3372)Generator.lognormal (line 2456)Generator.logistic (line 2376)Generator.laplace (line 2172)Generator.integers (line 442)Generator.hypergeometric (line 3229)Generator.gumbel (line 2257)Generator.geometric (line 3178)Generator.gamma (line 1228)Generator.f (line 1306)Generator.dirichlet (line 4038)Generator.choice (line 598)Generator.chisquare (line 1472)Generator.bytes (line 569)Generator.binomial (line 2805)GeneratorFewer non-zero entries in p than sizeEmpty shape tuple for cython.arrayEllipsisConstruct a new Generator with the default BitGenerator (PCG64). Parameters ---------- seed : {None, int, array_like[ints], SeedSequence, BitGenerator, Generator}, optional A seed to initialize the `BitGenerator`. If None, then fresh, unpredictable entropy will be pulled from the OS. If an ``int`` or ``array_like[ints]`` is passed, then it will be passed to `SeedSequence` to derive the initial `BitGenerator` state. One may also pass in a `SeedSequence` instance. Additionally, when passed a `BitGenerator`, it will be wrapped by `Generator`. If passed a `Generator`, it will be returned unaltered. Returns ------- Generator The initialized generator object. Notes ----- If ``seed`` is not a `BitGenerator` or a `Generator`, a new `BitGenerator` is instantiated. This function does not manage a default global instance. Examples -------- ``default_rng`` is the recommended constructor for the random number class ``Generator``. Here are several ways we can construct a random number generator using ``default_rng`` and the ``Generator`` class. Here we use ``default_rng`` to generate a random float: >>> import numpy as np >>> rng = np.random.default_rng(12345) >>> print(rng) Generator(PCG64) >>> rfloat = rng.random() >>> rfloat 0.22733602246716966 >>> type(rfloat) Here we use ``default_rng`` to generate 3 random integers between 0 (inclusive) and 10 (exclusive): >>> import numpy as np >>> rng = np.random.default_rng(12345) >>> rints = rng.integers(low=0, high=10, size=3) >>> rints array([6, 2, 7]) >>> type(rints[0]) Here we specify a seed so that we have reproducible results: >>> import numpy as np >>> rng = np.random.default_rng(seed=42) >>> print(rng) Generator(PCG64) >>> arr1 = rng.random((3, 3)) >>> arr1 array([[0.77395605, 0.43887844, 0.85859792], [0.69736803, 0.09417735, 0.97562235], [0.7611397 , 0.78606431, 0.12811363]]) If we exit and restart our Python interpreter, we'll see that we generate the same random numbers again: >>> import numpy as np >>> rng = np.random.default_rng(seed=42) >>> arr2 = rng.random((3, 3)) >>> arr2 array([[0.77395605, 0.43887844, 0.85859792], [0.69736803, 0.09417735, 0.97562235], [0.7611397 , 0.78606431, 0.12811363]]) Cannot take a larger sample than population when replace is FalseCannot index with type '%s'Cannot create writable memory view from read-only memoryviewCannot assign to read-only memoryviewCan only create a buffer that is contiguous in memory.Buffer view does not expose stridesAxis argument is only supported on ndarray objectsASCIIA9B.? * ?,|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@,{L?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<̀<<<.~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{xI4_h2z3+3@3aQ3i`3{am3Ay3i3*353=3r333|ϡ3ڍ3+333^33׶3iż3-¿3c3%3uY3<3L3gv3;3k3-3$3!333P3P33<3p~3չ3^3J3I34V?4=3@4A4A4qB4C4D4udE4-CF4K"G4H4H41I4J4vK4\fL4HM4+N4aO4O4bP4ٽQ4R4ԊS4crT4ZU4CV4-W4ZX4Y4UY4Z4[4(\4_]4^4_4C`4va4alb40cc47[d4~Te4Of4Jg42Hh4Fi4Fj4Hk4Kl4MPm4Vn4^o48hp4sq4r4s4 t4u4v4Cw4x4 z42{40S|4u}4~44v4@ 4L4>4ق4v444lV44R4F44p4 I44"4_44Ќ4l4L4`4ԏ4坐4y4ݖ4%44r&4k44(4444.4Q4N4t44\۶4H94̻4p44~X4w4p_4~444?yjD?l[T?w'??o?Wp?xI?-3?x^j??N?R:e?4:>?l?*?%z?PՋt?4?e;?$"?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?j%= BT ~Q~U~KD Ga7\%aFOaSuzpD(|Wc %WM$ t`K[oT`gtSwf# Wl`0H7[z1z(zK^2#9MM0MFPrOxS왎2ȩn{TH,ҭ^p .]M[\}r;/4d6dcNQp.t@e$oX%L(xo6xu{fUY>9>{ppCBwS(:5^dܓAN}8) YfHqն&|s f2,2Ztզޗ .n ZR'ӯB)[l@u Pҍ'TȈt(5wI'L/$;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: L< Ŀk<4xV<=A[<'?}y<NG<~;[ ?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$4l44x 4_444{4 4EP4±4{:4귴4);4nķ4S444<*4տ44A4.44ע44f4RW4R4*Y4Fk44δ444444g44k4<444y44u4_45555@5ó 5 5]5^555q5v 5!5%5V*5s/5;S55:<5D5NO5^5Nv5?7E?P?'{{?*!?bv?mU?9U1T?/v?x]?&1$-?~ 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?$ka?%> T+Y? O?K 2=?Ɨ$'R~1[} 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$y1gWr-ެ @樫(afoeW-|&aY +M]d<A]X`<+M[Ij<[5q&<.8eG< h#ឪV <;LC%K<ꆭh>> from numpy.random import Generator, PCG64 >>> rng = Generator(PCG64()) >>> rng.standard_normal() -0.203 # random See Also -------- default_rng : Recommended constructor for `Generator`. 4K?N@Si@>Aޓ=?.eB5<;?r1??@0C?:0yE>qh?8?5gG@dg?$@= ףp=@n?[ m?h|?5?333333 @r?$~?B>٬ @r鷯?Q?Q?9v?(\@ffffff@.@4@x&?@?UUUUUU?a@X@`@|@@MA>@v@UUUUUU?dg?3?r?q?0@@~)@ lѿ3 ; @UUUUUU?"@m{?-DT! @h㈵>.A-DT! -DT!@cܥL@9RFߑ?+eG?9B.?calling %R should have returned an instance of BaseException, not %Rraise: exception class must be a subclass of BaseExceptionexception causes must derive from BaseException%s() got multiple values for keyword argument '%U'%.200s() keywords must be strings%s() got an unexpected keyword argument '%U'_______________________________________________________pyx_fatalerrorý.@5?;48JPYxYi(?r?:s?t@tH@uAvHAwTBw|B(xCx8C{8HH8(<$p,@h@hX<("h&(*H.8814HH78:<=@LCFLxJ(N<QU,hX(\4sxH x (D!!ȴ,""##Eh$ht$؟8%X%(@&x=&C&xxT''((0(D(xX((l((((8(H(h(,)@)H\)x)H))x4*l*(*h*H*,+\++8+(+(,X,,,(-HL-X`-h--h .H.p..(`/(//$0XD000,1L1h11h1<2)212x3 3;l3h<3>,4>@4?x4?4HB4Y6h\h6x^6x_7aD7d7e7hf08xl8(v8v9Xw89(99؎<:(t:::$;(t;h;8;(P<(<H=X >ȵ >ȶ4>p>x>>>(@X@AA(AB(BhB(BB(CC$DtDD8D8$EHEEHFPFdFFFHGGDGXGGGXHHHI,JXPJJ8JXJ K84KhLKx`KxKKKKxK`LHLxL( M8M(lMMH N hN N NDO8O(FBB E(A0D8DElF} 8D0A(B BBBH W_FVaG,+hR{EG0 AB  AA P+S)FBA D0b  DBBI L  DBBD q  DBBG 4,TPFEA T BBD _BBd@,TFBE B(A0D8D`hWpRhF`7 8D0A(B BBBD hWp_hA`,Y?Hv,,YEG0@ AC  AA L,[FBA A(G@ (A ABBG  (C ABBD D@-];FID d BBE L BBH @ BBD ,-^EG0 DL _ AH `-D`FBH A(C0 (D BBBH a (D BBBE J (D BBBD .aFBB G(D0A8G 8D0A(B BBBE GYAJSAFsB .LsFEB E(A0A8GFFFFFFFFFCCCCCCCCCCCCCCCCFFFFFFFFFFCCCCCCCCCCCCCCCBLo 8A0A(B BBBH /Hw/Dwp/w80xIKED D(F0](A ABBG8<0xYKED D(F0p(A ABBDx0x,H c40x=FDD Y GBH AAB(0xEAG m CAB \0) AKQ E(F0F8A@AHAPAXA`NQ B(F0F8A@AHAPAXA`NG8T123xBEH A(A0\(D BBB81n3UFBB A(D0;(D BBB$14MBIC {AB1xDX L dL24BHE A(D0JHhn0D(A BBBdd2xQBB B(A0D8DPo 8D0A(B BBBD @HPD24BEL E(A0A8E@8A0A(B BBBD35BEL E(A0A8E@8A0A(B BBB(\3}EG Z AI XA$3}qFFG0MAK3<~VEG HA 3|~P@ I z E (3hZFAG@CFB 4k544%$H4S5QBDA FAB0p4|tFEG e EBB lEB44ȀzFAD  ABC D DBF $45fbAA ~ABH5N5BBB E(D0A8DPo8D0A(B BBB\P5FED D(D0| (D ABBD Q (D DBBC D (A ABBE <5FEB D(D0u (D BBBD L5\!FAD ~ AEA ~ ABG I FDE aIB@6<0\T6XJFBB A(G@ (D BBBE ^ (D EBBE U (D EBBF 86HFEA D(GP (D ABBA `6 FBE A(A0V (D BBBA j (A EBBD I (D BBBA (T7xFED ^ EBD H7 |FHA D(G0s (D ABBJ N(D ABBL7@TFEA A(D@< (A ABBE m (A ABBE 8P5808|FED D(GP (A ABBI @l8 FAA D0W  NABL   CABJ 8HA G \ D 885$8d<E G E K {9|0$98\FAA D0T  AABD @X9dFAA D0`  AABH   FABB d9БFBB E(D0D8F`N 8A0A(B BBBA  8A0A(B BBBL H:2^*FBB B(A0D8G;*8A0A(B BBBHP:̖FBE I(D0A8DP 8D0A(B BBBD 4:@$^AA hABEF  :8{FEB J(D0A8G_ 8D0A(B BBBA MGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGDDDDDDDDDDDDDDDCS ;Fz H p H <<?ED l AJ F AI J AM G AH <\<ED } AI G AH V AI G AC <`yD F AA (<\XFD P EA I EB <}=)K]= ,=KJD=KJ\= KJt=( 0=$zAD@ EAH `l=FBB A(A0Gpp 0D(A BBBB V 0D(A BBBG / 0D(A BBBA L,> FFG E(D0A8J 8A0A(E BBBG 8|>*FED A(G`S (D ABBE 0>oa H nJHA G L>_H V0?HQDD X ABD XH8?ZFBB B(D0A8D@y 8D0A(B BBBE H?ȺFBB B(D0A8DPq 8D0A(B BBBE `?<FDB B(A0A8DP  8A0A(B BBBI L 8A0A(D BBBH `4@xFDB B(A0A8DP 8A0A(B BBBG M 8A0A(D BBBG (@PECD k AAG H@ؽFBB E(D0A8D`h 8A0A(B BBBF A\FBB E(D0A8DPh 8A0A(B BBBF  8A0A(B BBBB e 8A0A(B BBBA m8A0A(B BBB`AXFDB E(A0A8GPg 8A0A(B BBBE A 8A0A(B BBBE xBIFBE B(A0A8DPR 8K0A(B BBBE _ 8A0A(B BBBG L 8A0A(B BBBJ <ByKJE D(E0F(A BBBH\BFIB B(A0D8D`j 8A0A(B BBBA S8A0A(B BBBd$CH=FBB E(A0A8Dt 8A0A(B BBBE u 8C0A(B BBBG C h fHCFBB F(D0D8G 8D0A(B BBBB LC|FBB B(A0G8D& 8A0A(B BBBH @D TD hD |D D D D D'D'D E$E0E DE XE lE E E E E E E E  F  F 4F HF \F pFF F F F F F F G $G 8G LG `GVtGVGD7Gp=GG G G H H (H >> 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 an array or sequence in-place by shuffling its contents. The order of sub-arrays is changed but their contents remains the same. Parameters ---------- x : ndarray or MutableSequence The array, list or mutable sequence 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]]) permuted(x, axis=None, out=None) Randomly permute `x` along axis `axis`. Unlike `shuffle`, each slice along the given axis is shuffled independently of the others. Parameters ---------- x : array_like, at least one-dimensional Array to be shuffled. axis : int, optional Slices of `x` in this axis are shuffled. Each slice is shuffled independently of the others. If `axis` is None, the flattened array is shuffled. out : ndarray, optional If given, this is the destinaton of the shuffled array. If `out` is None, a shuffled copy of the array is returned. Returns ------- ndarray If `out` is None, a shuffled copy of `x` is returned. Otherwise, the shuffled array is stored in `out`, and `out` is returned See Also -------- shuffle permutation Examples -------- Create a `numpy.random.Generator` instance: >>> rng = np.random.default_rng() Create a test array: >>> x = np.arange(24).reshape(3, 8) >>> x array([[ 0, 1, 2, 3, 4, 5, 6, 7], [ 8, 9, 10, 11, 12, 13, 14, 15], [16, 17, 18, 19, 20, 21, 22, 23]]) Shuffle the rows of `x`: >>> y = rng.permuted(x, axis=1) >>> y array([[ 4, 3, 6, 7, 1, 2, 5, 0], # random [15, 10, 14, 9, 12, 11, 8, 13], [17, 16, 20, 21, 18, 22, 23, 19]]) `x` has not been modified: >>> x array([[ 0, 1, 2, 3, 4, 5, 6, 7], [ 8, 9, 10, 11, 12, 13, 14, 15], [16, 17, 18, 19, 20, 21, 22, 23]]) To shuffle the rows of `x` in-place, pass `x` as the `out` parameter: >>> y = rng.permuted(x, axis=1, out=x) >>> x array([[ 3, 0, 4, 7, 1, 6, 2, 5], # random [ 8, 14, 13, 9, 12, 11, 15, 10], [17, 18, 16, 22, 19, 23, 20, 21]]) Note that when the ``out`` parameter is given, the return value is ``out``: >>> y is x True 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, *, method='svd') 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 Expected number of events occurring in a fixed-time interval, must be >= 0. A sequence must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. 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 39%. 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? Our null hypothesis will be the absence of deviation, and the alternate hypothesis will be the presence of an effect that could be either positive or negative, hence making our test 2-tailed. Because we are estimating the mean and we have N=11 values in our sample, we have N-1=10 degrees of freedom. We set our significance level to 95% and compute the t statistic using the empirical mean and empirical standard deviation of our intake. We use a ddof of 1 to base the computation of our empirical standard deviation on an unbiased estimate of the variance (note: the final estimate is not unbiased due to the concave nature of the square root). >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> t -2.8207540608310198 We draw 1000000 samples from Student's t distribution with the adequate degrees of freedom. >>> import matplotlib.pyplot as plt >>> s = np.random.default_rng().standard_t(10, size=1000000) >>> h = plt.hist(s, bins=100, density=True) Does our t statistic land in one of the two critical regions found at both tails of the distribution? >>> np.sum(np.abs(t) < np.abs(s)) / float(len(s)) 0.018318 #random < 0.05, statistic is in critical region The probability value for this 2-tailed test is about 1.83%, which is lower than the 5% pre-determined significance threshold. Therefore, the probability of observing values as extreme as our intake conditionally on the null hypothesis being true is too low, and we reject the null hypothesis of no deviation. standard_cauchy(size=None) Draw samples from a standard Cauchy distribution with mode = 0. Also known as the Lorentz distribution. 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() 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)) 0.0 # 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 * rng.standard_normal(size=...) rng.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. high - low must be non-negative. 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. 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 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. Default is True, meaning that a value of ``a`` can be selected multiple times. p : 1-D array_like, optional The probabilities associated with each entry in a. If not given, the sample assumes a uniform distribution over all entries in ``a``. 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 Notes ----- Setting user-specified probabilities through ``p`` uses a more general but less efficient sampler than the default. The general sampler produces a different sample than the optimized sampler even if each element of ``p`` is 1 / len(a). 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 uniform random sample from a 2-D array along the first axis (the default), without replacement: >>> rng.choice([[0, 1, 2], [3, 4, 5], [6, 7, 8]], 2, replace=False) array([[3, 4, 5], # random [0, 1, 2]]) 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) b'\xfeC\x9b\x86\x17\xf2\xa1\xafcp' # 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]]) Construct a new Generator with the default BitGenerator (PCG64). Parameters ---------- seed : {None, int, array_like[ints], SeedSequence, BitGenerator, Generator}, optional A seed to initialize the `BitGenerator`. If None, then fresh, unpredictable entropy will be pulled from the OS. If an ``int`` or ``array_like[ints]`` is passed, then it will be passed to `SeedSequence` to derive the initial `BitGenerator` state. One may also pass in a `SeedSequence` instance. Additionally, when passed a `BitGenerator`, it will be wrapped by `Generator`. If passed a `Generator`, it will be returned unaltered. Returns ------- Generator The initialized generator object. Notes ----- If ``seed`` is not a `BitGenerator` or a `Generator`, a new `BitGenerator` is instantiated. This function does not manage a default global instance. Examples -------- ``default_rng`` is the recommended constructor for the random number class ``Generator``. Here are several ways we can construct a random number generator using ``default_rng`` and the ``Generator`` class. Here we use ``default_rng`` to generate a random float: >>> import numpy as np >>> rng = np.random.default_rng(12345) >>> print(rng) Generator(PCG64) >>> rfloat = rng.random() >>> rfloat 0.22733602246716966 >>> type(rfloat) Here we use ``default_rng`` to generate 3 random integers between 0 (inclusive) and 10 (exclusive): >>> import numpy as np >>> rng = np.random.default_rng(12345) >>> rints = rng.integers(low=0, high=10, size=3) >>> rints array([6, 2, 7]) >>> type(rints[0]) Here we specify a seed so that we have reproducible results: >>> import numpy as np >>> rng = np.random.default_rng(seed=42) >>> print(rng) Generator(PCG64) >>> arr1 = rng.random((3, 3)) >>> arr1 array([[0.77395605, 0.43887844, 0.85859792], [0.69736803, 0.09417735, 0.97562235], [0.7611397 , 0.78606431, 0.12811363]]) If we exit and restart our Python interpreter, we'll see that we generate the same random numbers again: >>> import numpy as np >>> rng = np.random.default_rng(seed=42) >>> arr2 = rng.random((3, 3)) >>> arr2 array([[0.77395605, 0.43887844, 0.85859792], [0.69736803, 0.09417735, 0.97562235], [0.7611397 , 0.78606431, 0.12811363]]) D0kDDD`DCDPDLE EEOEEP'E3E8ED DЉ}}DDEEME0HDD0}@C}B YE `hEbuEoEp: v -  0\' J _ A ~  V @ EP p {' '`O ' '0X@ ' &P & '0 &  ' "'0 -' t 6'e ='X E'K K'> S'`, Z' ! c'  m'p  v'P ')@ , ,`a ",e *,g X/j S/pp /,m 9, /U F/@@m /P*b 0 Y 0{T x0O ")1 9M  3M  M 3 L $@ L 70 @L & L =Ц K H K BH `A  HA ` A #8 @ & @  @  @  `@  @@  @ 0 @ @ ?  ?  ? Ъ `? % @?  ? p ? ` >  >  > " @> 2 > * = ( = + @= #` =  < P < "p <  p<  P< Ь 0< P < ` ;  ; & ; *P @; %p ; % : !0 : ! P: Ы 0: @ :  9  9  9 h 9 2 h9  @9 " 8 [ 8 ? `8 0 @8  <8  08  8  7  7  7 H 7 ) 7  |7 x p7  6 ( 6  6 x 6  6 Ȩ 6 ( `6 ! 6 " 6 p 5 . 5 )x @5 ( 5 ب 5  5  4 7 `4 Fp Q4 h O4  N4 ؗ L4  L4 h 04 x 4 . 3 9 3 ) ]3  Y3  P3 8 @3  33  (3  3 @ 3  3  3 0 2 ` 2  2  2 8 2 З 2  2  2 ( $ Ș $ H $  $  $ ) # @ "  "  "  " ȣ "  "  " 4   ( X  X  H  0 ` h  8 T   p ( k أ d ` ^  P  E  %Ю )H  @ @ 8 h ` p X `  `0  MH F0 @ȥ 00 ! ! Ƞ   P      ȕ     8  Е Х    x p p iH ` X @ 0 Kx   6 p @)  X Ȝ  0 &ت @Ш #   ا    @ 0  p ظ   P       Ы      X (  @ 8    ̞         p  x d h Ȫ w  w` wh w p v Ч v ؠ v8 v v v v# pv @v% 0vx 'vȮ v' u2 u u 0 u u `h<  T A @h @x @ h @( @ @ @ 4B @ @4$` 40 4 4 48 33P 3H 3 ) ( = h       @ p  P X F P 0 x 'h "8   إ h    (  p  @ "@ 8 p 8 p   ȫ H X x P p  8   X #` P     p `0 P 8  (   X   @ ج #  X  h h  ` pp XH PX H 8  ( ؛ 0 ȧ   У   8  0 X  P   0 h `B( X R R8 M Hؕ CX 8  @Ȭ tX  ` Цx  O `     p @!Э    (    ( Nx  @   x ؔ q8 h P ` Y8 Y X   ܁ ՁЖ ρH P & `x  Ex >xP 7xث @l H aD ؝ aО a a  T  Tp T T T J  PPD mm> > J JB A A DPP@ @? EE,}D0`B P0EpBC C DC  C B @-H ,hkDp ЂD @D cЀE3`"0``4 D J 394dfb874254454620cb41a7630824553d58e3.debug<#/.shstrtab.note.gnu.property.note.gnu.build-id.gnu.hash.dynsym.dynstr.gnu.version.gnu.version_r.rela.dyn.rela.plt.init.plt.got.plt.sec.text.fini.rodata.eh_frame_hdr.eh_frame.init_array.fini_array.data.rel.ro.dynamic.got.plt.data.bss.gnu_debuglink  $1o ; 2C(B(BKo^^6XoccgccwqBpp{v  ` P;,,     4 \p< p, x< x, < , 8 < , > . `@ 0 G 7 L  h  h 4