ELF>`@' @8 @[kkh~[kk888$$Ptd000,,QtdRtd[kk((GNU.{a9_)4a MA@04"@AVD( 6!Q@!PQ "TT@ ((#D a& @0DBD 1"lP fxQX|k/+*P{`D\~`cos\ zK?FSS( }\=ڃ1KE T W WW)|[  p[ F D-   8  R"\ 1  ! &  PJ"l 3oO w&X-k7  <   { z ahn 4V y G    D  c  :   %  p S#h $L D) p=  8n,  ;u @Ms p+ @/$ No  RD  k p~w nZ ' `  <%  8 p/] c&+  @Kg M$ `ij @2S  =  U  +'  @=Y A `C (2 Q M~  @0& Qo  P3E JG 0tI |` @@I/ J O" Q3 @QN OBV -W M  p 0  2 @P M) r` Ay) ,y dW +B /   `R  (/ ,?U  p8 H  bx p,H CI p0/8 `  pE\  kV  >I  <I[ 0K  OC w 2 j  ?- fCh  `8`  `M( y{; @a  ~  30 z  <  n PO/  K9  `j_ EID U& 9d tE P,  m  Pi NS e` @.__gmon_start___init_fini_ITM_deregisterTMCloneTable_ITM_registerTMCloneTable__cxa_finalize_Py_NoneStructPyBaseObject_TypePy_EnterRecursiveCallPy_LeaveRecursiveCallPyErr_OccurredPyObject_CallPyExc_SystemErrorPyErr_SetStringPyObject_GetAttrrandom_intervalmemcpy_Py_DeallocPyExc_DeprecationWarningPyErr_WarnFormatPyExc_TypeErrorPyErr_FormatPyExc_OverflowErrorPyLong_AsLongPyLong_TypePyObject_GC_UnTrackPyObject_GC_IsFinalizedPyObject_CallFinalizerFromDeallocPyErr_FetchPyErr_NormalizeExceptionPyException_SetTracebackPyErr_SetExcInfoPyModule_GetDictPyDict_NewPyImport_ImportModuleLevelObjectPyList_NewPyObject_GetAttrStringPyDict_GetItemStringPyModule_GetNamePyExc_ImportErrorPyCapsule_IsValidPyCapsule_GetNamePyCapsule_GetPointerPyDict_SetItemStringPyExc_AttributeErrorPyErr_ExceptionMatchesPyErr_ClearPyThreadState_GetPyInterpreterState_GetIDPyModule_NewObjectPyDict_GetItemWithErrorPyExc_KeyErrorPyErr_SetObjectPyTuple_PackPyExc_ValueErrorPyOS_snprintfPyErr_WarnExPySlice_NewPyLong_FromSsize_tPyTuple_NewPyDict_NextPyUnicode_ComparePyLong_FromLongPyUnicode_AsUTF8AndSizePyCode_NewEmptyPyErr_RestorePyUnicode_TypememcmpPyObject_RichCompare_Py_TrueStruct_Py_FalseStructPyObject_IsTrue_PyUnicode_ReadyPyExc_StopIterationPyExc_NameErrorPyCFunction_TypePyType_IsSubtypePyList_TypePyTuple_TypePyObject_GetItemPyNumber_IndexPyLong_AsSsize_tPyErr_GivenExceptionMatchesPyExc_IndexError_PyDict_GetItem_KnownHash_PyObject_GetDictPtrPyObject_NotPyFrame_NewPyTraceBack_HerePyUnicode_FromFormatPyUnicode_AsUTF8PyObject_SetAttrPyMem_ReallocPyMem_Malloclegacy_waldPyDict_Sizelegacy_rayleighlegacy_lognormalrandom_logisticrandom_gumbelrandom_laplacelegacy_powerlegacy_weibulllegacy_paretolegacy_vonmiseslegacy_standard_tlegacy_standard_cauchylegacy_noncentral_chisquarelegacy_chisquarelegacy_noncentral_flegacy_flegacy_gammalegacy_standard_gammalegacy_normallegacy_gaussPyDict_SetItemPyMethod_Typelegacy_standard_exponentiallegacy_exponentiallegacy_betarandom_standard_uniform_fillPyObject_GetIterPyNumber_AddPyNumber_InPlaceAddPyExc_RuntimeErrorPy_GetVersionPyBytes_FromStringAndSizePyUnicode_FromStringAndSizePyImport_AddModulePyObject_SetAttrStringPyUnicode_InternFromStringPyUnicode_DecodePyObject_HashPyFloat_FromDoublePyLong_FromString__pyx_module_is_main_numpy__random__mtrandPyImport_GetModuleDict_Py_EllipsisObjectPyType_ReadyPyCapsule_NewPyImport_ImportModulePyErr_GetExcInfoPyCapsule_TypePyExc_ExceptionPyType_ModifiedPyCMethod_New_PyDict_NewPresizedPyDict_Copylegacy_logserieslegacy_random_geometriclegacy_random_zipflegacy_random_poissonlegacy_negative_binomialPyObject_IsInstancePySequence_ContainsPyFloat_TypePyFloat_AsDoublePyObject_SizePyObject_SetItem_PyType_LookupPyEval_SaveThreadrandom_positive_intPyEval_RestoreThreadPyNumber_LongPyDict_Typerandom_uniformlegacy_random_hypergeometricPySequence_Tuplelegacy_random_multinomialrandom_triangularlegacy_random_binomialPyBool_TypePyUnicode_FormatPyNumber_RemainderPyUnicode_New_PyUnicode_FastCopyCharactersPyObject_FormatPySequence_ListPyNumber_MultiplyPyList_AppendPyList_AsTuplePyNumber_InPlaceTrueDividePyNumber_SubtractPyInit_mtrandPyModuleDef_Initlogsqrtpowexplog1p__isnanrandom_binomial_inversionrandom_binomial_btpefloorrandom_loggamrandom_geometric_searchceilacosfmodrandom_standard_uniform_frandom_standard_uniformrandom_standard_uniform_fill_frandom_standard_exponentialrandom_standard_exponential_fillrandom_standard_exponential_fexpflog1pfrandom_standard_exponential_fill_frandom_standard_exponential_inv_fillrandom_standard_exponential_inv_fill_frandom_standard_normalrandom_standard_normal_fillrandom_standard_normal_frandom_standard_normal_fill_frandom_standard_gammarandom_standard_gamma_fpowflogfsqrtfrandom_positive_int64random_positive_int32random_uintrandom_normalrandom_exponentialrandom_gammarandom_gamma_frandom_betarandom_chisquarerandom_frandom_standard_cauchyrandom_paretoexpm1random_weibullrandom_powerrandom_lognormalrandom_rayleighrandom_standard_trandom_negative_binomialrandom_noncentral_chisquarerandom_noncentral_frandom_waldrandom_vonmisesrandom_logseriesrandom_geometric_inversionrandom_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_fillmemsetrandom_bounded_bool_filllibstdc++.so.6libm.so.6libgcc_s.so.1libpthread.so.0libc.so.6GLIBC_2.14GLIBC_2.2.5&00ui ;ui ;k0k0kk` 0 p       H Ȗ           @  H  P  X H        ` H      ( H 0  @  H  P  `  h  p  x     Ș  И      (  0  @  H  `  h    ` H  ` ș H Й  ` H       H (  @  H H P  `  h  p p x  ` H  ` Ț H К       H (  @  H H P  X  p H x       `   x   0 0 0 @  H h  p   @  `  0   0  8 X  `     Н  ؝     ( PH  P Pp  x      Ȟ @M  `   M8  @ ` x h  p @ h L؟ `  X  L( P 0  P H X x @ @L 8 LȠ 0 Р ` ( K    @  H h  p       `   0  8  X  `   `  Т  آ      ( H  P `p  x     ȣ `  K    8  @ ` x h  p  h ؤ ` @ X  ( P 0 P H X x @  8 `ȥ 0 Х  ( `   @  H h  p     @K  @    0  8 pX  `     Ч  ا     ( KH  P p  x   J  Ȩ l  h   P8  @ ` x h F p  h dة ` c X  &( P 0 @P H X x @  8 \Ȫ 0 Ъ x ( !  @  H h  p     p     0  8 ;X  `      UЬ  ج   :  ( :H  P 9p  x   h  ȭ h  `   08  @ @` x h  p  h /خ ` N X  N( P 0 !P H X 0x @  8 ȯ 0 Я  (   `@  H h  p   G  ~  ~   $0  8 $X  `     б  ر     ( \H  P @p  x     Ȳ      X8  @ ` x h  p   h  س ` @ X  ( P 0 P H X  x @  8 ȴ 0 д X ( 9  @  H h  p          0  8 TX  ` @  @  ж  ض   @  ( PH  P 2p  x 2  @  ȷ      P8  @ P` x h  p { h +ظ `   X  ( P 0 P H X x @  8 ȹ 0 й  (   @  H h  p       x   0  8 X  ` P    л  ػ L  L  ( HH  P Hp  x     ȼ      8  @ ` x h @ p  h ؽ ` H X  H( P 0 P H X x @ h 8 hȾ 0 о @ (   @  H `h  p p  p  `  D   0  8 X  `           ( H  P 8p  x   x  x     8  @ ` x h ` p ` h  `  X  ( P 0 P H X x @  8  0   (    w@  H h  p          0  8 X  `   `k      ]  ( H  P p  x N  u  0  0   8  @ M` x h ` p  h @ `  X  ( P 0 XP H X x @  8  0  (   =@  H h  p       (   (0  8 X  `   @      /  (  H  P @p  x P         @8  @ ` x h  p  h  `  X  ( P 0  P H X  x @ @$ 8  0  ( @  @  H h  p        p   p0  8 X  ` `          ( HH  P Hp  x     A     8  @ ` x h 8 p  h  `  X  ( P 0 P H X x @  8  0  (   @  H h  p   `  `     0  8 X  `     @{    |  ( |H  P Pp  x     ~     8  @ ` x h p p  h  ` l X  0( P 0 0P H X ax @  8  0 Y (   @  H  h  p v  v  0     80  8 X  ` (  3      /  ( yH  P pp  x   +    p   p8  @ ` x h g p q h  `  X  ( P 0 pP H X x @  8  0  (   @  H h  p   b  b  M   ]0  8 ]X  `         @  ( H  P  p  x j  X  X  h U     0 5X  @   1 7   (  J8 1  p1 `3  @ ̨( @ ٨H `  h x   `d   0  o   h   @ c  y W( 8  o @ NH X j ` ӧh  x e   Z  0 `W  K I   `; W  6 \( 8 - @ ܧH 6X " ` >h 0x  G p  /   8 P  -   " Ѐ @ (  ~8  @  H zX  ` h @xx   u   r   Po  ڣ l z ԣ i l ̣( 0f8 ^ @ ţH pbX K ` h ^x  ?  Z @-  @X  #  T  E -  ʧ pR  ( 8 `@ H X `` h x  %     `        9    /( p8 @ H pX ` h ux }   `u  0 u   `v    y  ( PW8  |krxormnnnn n(n0n8n@nHnPnXn`nhnpnxnnn!nnn.n0nn:n<n>n?n@nFnGnInoNooo[ o(o0o8oc@oHoPoXon`ohopooosooo|oooooooooooop p(p0p8p@pHpPp Xp`p hppp xp ppppppppppppppppq q"q#q$ q%(q&0q'8q(@q)Hq*Pq+Xq,`q-hq/pq1xq2q3q4q5q6q7q8q9qq;qq=q>qAqBqCqDrErHrJrK rL(rM0rO8rP@rQHrRPrSXrT`rUhrVprWxrXrYrrZr\r]rr^r_r`rrarbrrrdresfsgshsi sj(sk0sl8s@sHsmPspXsq`sthsupsxsvswsxsyszs{ss}s~sssssssstttt t(t0t8t@tHtPtXt`thtptxtttttttttttttttttuuuu u(uHH5HtH5%@%h%گh%үh%ʯh%¯h%h%h%hp%h`%h P%h @%h 0%h %zh %rh%jh%bh%Zh%Rh%Jh%Bh%:h%2h%*hp%"h`%hP%h@% h0%h %h%h%h%h %ڮh!%Үh"%ʮh#%®h$%h%%h&%h'p%h(`%h)P%h*@%h+0%h, %zh-%rh.%jh/%bh0%Zh1%Rh2%Jh3%Bh4%:h5%2h6%*h7p%"h8`%h9P%h:@% h;0%h< %h=%h>%h?%h@%ڭhA%ҭhB%ʭhC%­hD%hE%hF%hGp%hH`%hIP%hJ@%hK0%hL %zhM%rhN%jhO%bhP%ZhQ%RhR%JhS%BhT%:hU%2hV%*hWp%"hX`%hYP%hZ@% h[0%h\ %h]%h^%h_%h`%ڬha%Ҭhb%ʬhc%¬hd%he%hf%hgp%hh`%hiP%hj@%hk0%hl %zhm%rhn%jho%bhp%Zhq%Rhr%Jhs%Bht%:hu%2hv%*hwp%"hx`%hyP%hz@% h{0%h| %h}%h~%h%h%ګh%ҫh%ʫh%«h%h%h%hp%h`%hP%h@%h0%h %zh%rh%jh%bh%Zh%Rh%Jh%Bh%:h%2h%*hp%"h`%hP%h@% h0%h %h%h%h%h%ڪh%ҪhAWIAVIH5AUIATUSHAPHLHHuIHu(LELH5HH9H81qLH*u9LLHIMLHHhH5H81)LLHHtHMAuHvHMuHfAZD[]A\A]A^A_AWIAVIH5AUIATUSHAPHLHHxIHu(LHLH5&HH<H81qLH-u9LLHIMLHHkH5 H81)LLHHtHMAuHyHMuHiAZD[]A\A]A^A_AVIAUIHATUSDHt5H;HuE1tHLLAHMu)HHAH8`tE1[D]A\A]A^AVAUIATUQhHx?HHHuH;Ht#H9tHH5AH8E1L%R7 Mt I$H5L IHtH=IHuL=HtHIHAH HLHxkAH LLHxHAH |LLH{ix%E1H oLLHkIxIHMHZL]A\A]A^AVIHAUIATUDSHHHIH@u#HLLH58H81.ID$(ML$ Ht HHLLH9v#HfILLH5H81BuOL9sJHl$PLMAQHdIH11H1ZYyI $uLE1HL[]A\A]A^AWAAVMAUAATAUSLHHHT$8Ht$0HL$H|$(fHIcHHD$HHH5H1HD$H HH5H^1oH$HHH5H0Ic@IHHH5H1IHHH5kHIcIHHH5JHHT$H57HnHH5%HqTLH5HW:H$H5H8H$H5HH$H5HH$H5HH$1IHH$1IH~$HLHmH5kHHD$LT$HIt61*LT$HIt"HHL/LT$HHt I uFE1I1HPIHu0L&1E1kH$1E1E1HD$E1E1HMuHIH|$H|$H<$ M0M>MLMt IuLH|$(HBHT$8Ht$0.H$1E1E1E1E1e1E1E1E1E1R1E1E1E1B1E181E1.1E1E1!HD$HHPHD$HD$HHHN HD$HHPHD$HD$HHHH$HHPHD$H$HHHIMLI $LILHHH[]A\A]A^A_AVAUATUHHHW0 Ht)1H9[HКH5aH8Y[ Ll$A IH L11ALAu 0 v8uHHuD$ 11D$$ALL$ .uBtILd$(1LH LLJ1L%Pr@ v뷹AHHu1H/ HO1H=,H. HO1H=bH. HOHEHH-. H. HOHH=H. HOHH=H. HOHH. H5A`H={. AH-& LeMt~E E!H}t:}"t I$:HEHUHpHt 1I$pI$HEHp}I$HEH8HNHNH(yWH HhNH HKNH H.NH\ HN1 H|H5>$ 1H HYH5  1oH H6H5 1LH HH5 1)H HH5 & 1Hg HH5% 1H< HH5% 1H HH5& 1H HdH5# 1zH HAH5 1WH HH5 14He HH5 1H: HH5 1H HH5 1H HH5 1H HoH5A 1H HLLLLHH H.H5 1DHE H HU$ H5 1H HH5 1H HH5 L1H HH5" 1H HuH5 1Hd HRH5 1hH9 H/Ha# H5R! 1>H HHwH5h 1H HH5 1H HH5: 1H HH5 1HyH{ H5 1HVL# H # ARMQh5a 5 APAPPeH@H Lu# 11H ]# AQMQh5 5 APAPAPH@HHw H5 1HL # H " WQMh5 5  APAPPH@HXH2 H5 1gH5L" H " V1QMh5l 5 APAPPHH@HH H5A 1HL5" H &" R1QMh 5 5L APAPPH@HHF H _H! HqNfHnH=e fHnH PflH ) Bx]H=! 11L-4 HHt>H5 HL HExHHEu,H"HHEuHA`AH5 H=J! H xH H=4H XHHAHHFH5HoHMuHH=HHWA HHH5SHHMuHH=HHA HHH5`HHMuHdH=uhHHA`HHUH5HH HxAH HH*H5HNA0HH H5XHI HAPHHH5'H HAHHH5HAHHH5aHAHH|H57H HgAHHEH5qH=AHHH5ܾGHAHHH5HAHHH5H HAHHH5WHAHHH5-HdAHHqH5nH:AHHQH5ٽDHHMuHH=-HHA`HHEH5HA@HHH5ֽHHH5 IHt{1H$IHu HuHcH5H8IMuLMt6AHH{H5L@HtHMu)HHMuHA`AH=EHH=H BHv HH5?y<H HI HH5T<H H HH5/<HMuHH=HHH H) HH5˼gH H HH5BvH nH HH5QH IH HH5,H $Hu HH5jH HH HH5PH ڢH HH57H H HH5dH H HH5?sHMuH|H==HH[H HV HH5Ż(H ȢH) HH5H H HH5H ֢H HH5H H HH5`H Hu HH5|;xsH HL HH5`xRH H# HH5Dx1H ѤH HH54xHMu)HHMuH A`A0H= 11LHH:H5u H= H4HMuHH= 11`LHH9H5 H=V H4HMuHhHH9H HHH HEHH= 1KIHh4HMuHH5 LVHH#=H5m H= H.54HMuHIMuLH= 11jKHH8H5 H=` H<HMuHrIH8H LHHx IEHH=: JHH{<IMuLH5? HUIHn3H5, H= H5d3IMuLHMuHHT$Ht$HH$HD$HD$HD$HD$ HD$(H=~IH|H5HIMHuLBHuHH5rH8'<HH9Et-HvH5H8HMH1HHMH uHH HuHH5H8= Ht t& H5\HH81 H< w' H5pHH81EMAƅuHH5hH8(tHjH5{H8 H<$Hu%-H~H8Gu`WAHuH|$H$Ht HuHD$H|$H HxWH=H ζ]HT$(Ht$ H|$2Fx@H5_ H= 1YAHHt0HYKHMu/HA,W%WA(WA ,WAH<$HT$Ht$vH|$Ht HuH|$ Ht HuH|$(H3H33H HH4H H5 HH/HMuH.H= H}H= HH uUHH4H5 H= H3/HMuHH H  H9Hu)H Ht HH-{ .H=J PHH=9 HZ H5[ 6[HH 4H5 H;@IH /HMuHAH5 H= L{7IMuLH H H9Hu)H Ht HL- .H= ]OIH= H H5 ~ZIMb3H5 L?HH7IMuLH5 H=K H<.HMuH]H& H  H9Hu)H Ht HH- .H= NHH= H H5 YHH2H5 H>IH-HMuHH5 H= L F6IMuLHn H G H9Hu)H2 Ht HL-# .H=" MIH= H H5 YIM2H5 L>HH5IMuLH5 H= HS,HMuHH H  H9Hu)Hj Ht HH-[ .H=j 5MHH=Y H: H5; VXHHm1H53 H[=IHv,HMuHaH5 H=# L4IMuL5H H H9Hu)H Ht HL- .H= }LIH= Hr H5s WIM0H5 L<HH4IMuLH5 H=k H+HMuH}HF H H9Hu)H Ht HH- .H= KHH= H H5 VHH0H5 H;IH++HMuHH5 H= L+3IMuLH H ' H9Hu)H Ht HL- .H=B KIH=1 H H5 .VIMx/H5 L3;HH>3IMuL9H5 H= Hs[*HMuH H H _ H9Hu)HJ Ht HH-; .H= UJHH=y H H5 vUHH.H5 H{:IH)HMuHH5 H=C Ln2IMuLUH H H9Hu)H Ht HL-s .H= IIH= HR H5S TIM*.H5  L9HH1IMuLH5 H= H)HMuHHf H H9Hu)H Ht HH- .H= HHH= H H5 THH-H5 H 9IH(HMuHH5Z H= LK&1IMuLH H  H9Hu)H Ht HL- .H=b -HIH=Q H H5 NSIM,H5{ LS8HH0IMuLYH5R H= H'HMuH-H H ? H9Hu)H* Ht HH- .H= uGHH= H H5 RHH5,H5 H7IHJ'HMuHH5r H=c L/IMuLuH> H w H9Hu)Hb Ht HL-S .H= FIH= H2 H53 QIM+H5 L6HHf/IMuLH5 H= H#z&HMuHH H H9Hu)H Ht HH- .H=: FHH=) Hj H5k &QHH*H5 H+6IH%HMuH1H5 H= Lk.IMuLH H H9Hu)H Ht HL- .H= MEIH=q H H5 nPIM@*H5 Ls5HH.IMuLyH5 H=; H/%HMuHMH H  H9Hu)H Ht HH- .H= DHH= H H5 OHH)H5C H4IH$HMuHH5 H= LN-IMuLH^ H W H9Hu)HB Ht HL-3 .H= CIH= H H5 NIM(H5 L4HH,IMuL H5 H= HC#HMuHH H H9Hu)Hz Ht HH-k .H=Z %CHH=I HJ H5K FNHHK(H5; HK3IHi#HMuHQH5 H= L,IMuL%H H H9Hu)H Ht HL- .H= mBIH= H H5 MIM'H5; L2HH+IMuLH5 H=[ H"HMuHmH6 H H9Hu)H Ht HH- .H= AHH= H H5 LHH&H5; H1IH"HMuHH5 H= L*IMuLH~ H 7 H9Hu)H" Ht HL- .H=2 @IH=! H H5 LIMV&H5k L#1HHF*IMuL)H5B H= HcN!HMuHH H o H9Hu)HZ Ht HH-K .H=z E@HH=i H* H5+ fKHH%H5 Hk0IH HMuHqH5r H=3 Lv)IMuLEH H H9Hu)H Ht HL- .H= ?IH= Hb H5c JIM%H5[ L/HH(IMuLH52 H={ H HMuHHV H H9Hu)H Ht HH- .H= >HH= H H5 IHHa$H5 H.IHHMuHH5b H= L;.(IMuLH H  H9Hu)H Ht HL- .H=R >IH=A H H5 >IIM#H5 LC.HH'IMuLIH5 H= HHMuHH H O H9Hu)H: Ht HH-+ .H= e=HH= H H5 HHH#H5 H-IH=HMuHH5 H=S L&IMuLeH. H H9Hu)Hr Ht HL-c .H= <IH= HB H5C GIMl"H5 L,HHn&IMuLH5 H= HmHMuHHv H H9Hu)H Ht HH- .H=* ;HH= Hz H5{ GHH!H5 H,IHHMuH!H5 H= L[%IMuLH H H9Hu)H Ht HL- .H=r =;IH=a H H5 ^FIM!H5 Lc+HH&%IMuLiH5 H=+ H"HMuH=H H / H9Hu)H Ht HH- .H= :HH= H H5 EHHw H5C H*IHHMuH豿H5 H=s LV$IMuL腿HN H g H9Hu)HR Ht HL-C .H= 9IH= H" H5# DIMH5{ L)HH#IMuLH5R H= H3HMuH;H H H9Hu)H Ht HH-{ .H=J 9HH=9 HZ H5[ 6DHH)H5 H;)IH\HMuHAH5 H= L{#IMuLH H H9Hu)H Ht HL- .H= ]8IH= H H5 ~CIMH5 L(HH"IMuL艽H5 H=K HÿHMuH]H& H  H9Hu)H Ht HH- .H= 7HH= H H5 BHHH5K H'IHHMuHѼH5" H= L !IMuL襼Hn H G H9Hu)H2 Ht HL-# .H=" 6IH= H H5 BIM4H5k L'HHN!IMuLH5B H= HSAHMuHH H  H9Hu)Hj Ht HH-[ .H=j 56HH=Y H: H5; VAHHH5[ H[&IHHMuHaH52 H=# L蛽~ IMuL5H H H9Hu)H Ht HL- .H= }5IH= Hr H5s @IMH5 L%HH IMuL詺H5b H=k HHMuH}HF H H9Hu)H Ht HH- .H= 4HH= H H5 ?HH?H5 H$IH{HMuHH5 H= L+6IMuLŹH H ' H9Hu)H Ht HL- .H=B 4IH=1 H H5 .?IMH5 L3$HHIMuL9H5 H= HsHMuH H H _ H9Hu)HJ Ht HH-; .H= U3HH=y H H5 v>HHH5 H{#IH*HMuH聸H5 H=C L軺IMuLUH H H9Hu)H Ht HL-s .H= 2IH= HR H5S =IMJH5 L"HHvIMuLɷH5 H= HWHMuH蝷Hf H H9Hu)H Ht HH- .H= 1HH= H H5 =HHH5 H "IHHMuHH5 H= LKIMuLH H  H9Hu)H Ht HL- .H=b -1IH=Q H H5 N<IMH5 LS!HH.IMuLYH5 H= H蓸HMuH-H H ? H9Hu)H* Ht HH- .H= u0HH= H H5 ;HHUH5 H IHHMuH衵H5 H=c L۷^IMuLuH> H w H9Hu)Hb Ht HL-S .H= /IH= H2 H53 :IMH53 LHHIMuLH5 H= H#HMuH轴H H H9Hu)H Ht HH- .H=: /HH=) Hj H5k &:HHH5C H+IH7HMuH1H5 H= LkI $uLH 11H=c ^IHH5[ H= HI $uL足Hg 11H= IHbH5 H=U H͵I $uLgH 11H= IH$H5 H= H~\I $uLH 11H= qIHH5 H= H/I $uLɲHz 11H= "IHH5 H=h HI $uLz5 IHwH HID$H HH HID$Hu HPH2 HID$H# HPH HID$H HPH HID$H HP H HID$H HP(Hz HID$Hk HP0HH HID$H9 HP8H HID$H HP@H HID$H HPHH HID$Hs HPPHX HID$HI HPXH HID$H HP`H HID$H HPhH HID$H HPpH` HHV ID$HPxH. HID$H HH HID$H HH\ HID$HM HH' HID$H HH HID$H HH] HID$HN HH( HID$H HH HID$H HHN HID$H? HH HID$H HH HID$H HH HID$H HH HID$H HH HID$H HH HID$H HHs HHi ID$HHF HID$H7 HH HID$H HH HID$H HH HID$H HH HID$H H H HID$H H(H HID$H H0Hs HID$Hd H8H. HID$H H@H HID$H HHH HID$Hu HPHO HID$H@ HXH HID$H H`H HID$H HhH0 HID$H! HpH HH ID$HxH H= HID$Hp HHJ HID$H; HH HID$H HH HID$H HH HH ID$H5e HLήI $uLh-螭IHvH H5D H茮He H56 LnH H5 LPH H5 L2uH H5 LbHE H5 LOH H5 Lح<H H5R L躭)H H5 L蜭H H5f L~H H58 L`HQ H5 LBHc H5 L$H H5 LH? H5 LH H5B LʬH H5, L謬~H] H5 L莬kH H5 LpXH9 H5r LREH H5 L42H= H5 LH H5` L H H5 LګH H5 L輫Hm H5 L螫H7 H5 L耫H H5 LbH H5 LDHM H5n L&H H5 LtH H5J LaH[ H5D L̪NH H5^ L讪;H H5 L萪(Hi H5j LrH H5 LTH5 H5 L6H H5x LH H5: LHS H5 LܩHM H5 L辩Ho H50 L蠩Ha H52 L肩}H{ H5d LdjH5 H= LFWI $LܦA+`AA7`AE1AF`AMAAK`E1Aj`A}MAAl`iA`AXA`AGE1A`A3A`A"E1A`AA`AE1A`AA`AE1AaAAaAE1AaAA0aAE1AbAAE1AJbA-A\bAE1AhbAAzbAE1AbAAbAE1AbAAbAE1AbAAbAE1AbAwAbAiE1AbAXAcAJE1AcA9A.cA+E1A:cAALcA AXcAHMuHMI $LӢHMHA`A谢H DAw`H=Ai(麲A`A驲A_A防A_A釲A_AvA_AeA_ATA_ACA_A2A `A!A`AA)`AA5`AAA`AݱAV`A̱Ab`A黱A`A骱A`A陱A`A鈱A`AwA`AfA`AUA`ADA`A3A`A"A aAAaAA+aAA:aAްAIaAͰAXaA鼰AgaA髰AvaA隰AaA鉰AaAxAaAgAaAVAaAEAaA4AaA#AaAAaAA bAAbA߯A*bAίA9bA齯AHbA鬯AWbA雯AfbA銯AubAyAbAhAbAWAbAFAbA5AbA$AbAAbAAbAAbAA cAϮAcA龮A)cA魮A8cA霮AGcA鋮AVcAzAecAiAqcAXA}cAGAcA6AcA %AcAAJd MAAI`IAX`AMAAg`MAA`MAA`MAA`tMAA``MAA`LMAA`8MAAa$MAAaMAA!aMAA-aMAA?aMAAKaMAA]aMAAiaMAA{aMAAapMAAa\MAAaHMAAa4MAAa MAAa MAAaMAAaMAAaMAAbMAAbMAA/bMAA;bMAAMblMAAYbXMAAkbDMAAwb0MAAbMAAbMAAbMAAbMAAbMAAbMAAbMAAbMAAc|MAA chMAAcTMAA+c@MAA=c,MAAIcA[cAAgcAAscAAcAAcAAcA ABdAALdAMdANdAOduAPdjAQd_ARdTASdIATd>AUd3AVd(AWdAXdAYdAZdA[dA\dA]dA^dA_dA`dAadAbdAcdAddAedAfdxAgdmAhdbAidWAjdLAkdAAld6Amd+And AodApd AqdArdAsdAtdAudAvdAwdAxdAydH]A\A]A^H= 鑘H= H H9tHF>Ht H=ɹ H5¹ H)HH?HHHtH>HtfD= u/UH=>Ht H=;ݒha ]{f.Hq>GPHGXHDATIUHSHHHt HՅu!H1Ht[LH]A\[]A\ff.HGHHGHu310Ht!H HPH=HHPHHH<H5 18AUIATIUHSHHGHHt]H=M葐u:LLHIMtHL[]A\A]IHt"HE1[L]A\A]fH[]A\A]H;H5MH8+fHGHHtfDcAWL~AVAUATUSHL$M~qLHG HIHHD$MLfH|$LH $HLIL4L虖HHL苖HHL}L)IuH5<HH[]A\A]A^A_f.UHSHHt:HHH}H/tH]H1[]@H]H1[]fDH;ff.@UHGHL@t3H ;LHeLH81諔AHEu&]fDHY:HHH5LH81蚎Hmt 1]Hp1ff.UHHGHGHHHtHcH1H]GH]WGHH HcʉH9tH:H5SLH8ӓ@WGHH HHcʉH9uH]GH]pHcH9fHuHtLf.H@`HtkHHt_HHtUH9H9Eu,@HHmHD$ D$ HH5p HHuxgHjH8H50lH8ȒOUHGHuoHwH}HtHEH/t@HHtHDžH/tHEH]H@f[fKfHt!zH!j]AWAVAUATIUHSHH(L|$Lt$Ll$LLL LLLΎIHutHt$HtH|$蠐x\HD$HtHHD$HtHHt$Ht HHt$H|$HT$I<$HuHH(1[]A\A]A^A_fH|$I$HEHHtH/t9H|$HtH/tIH|$HtH/t)H([]A\A]A^A_DfۏfˏfUSHHH-x7HHEHkHHEHt H/tAHEHHHHEHtH/t H1[][H1[]fKfAVAUIATUSHtvH= IIH莑HHLALHLIHmt[L]A\A]A^HȎ[L]A\A]A^@1aIHthH=z }IHtEHHt8LALHL脋I,$I{L^Hmo끐I,$uLAE1[]LA\A]A^fATUHHqIHtHHL]A\f.{HuHEHuH4H819HHtHH4H8迉HmuH蠍ff.AVAUATUHSH HH_pHHCHHtH1H H[]A\A]A^f.EL-4E1HH2H4L貋IMtI.uLMtCLHSImIt\H L[]A\A]A^@H3HWH5HH81ćH E1[L]A\A]A^ÐEu[H\4Hr@L而fD1DL$HT$Ht$BHt$HT$HDL$IIfHIHt8H3LH辊IMtI.t8I,$ LM;I.1L݋LЋff.AUATUHHH@uvt@tv1IHtO1HHImItpMt3IL$@tmLHbfI,$u]LA\A]<@]A\A]fH]A\A]3H1]H5GA\A]H8DLfDH1HH5GH81AWAVAUATIUHSHHHHT$L|$0Lt$(LD$Ll$8HD$(HD$0HD$8@LLLL/H Ht$(HHHfDHPHHtH92uHT$0H|$H)HHFVH%f.HFH9GHJHHt5H9H9uHL$0H|$HH)H H:?Ht$(f.H9u&fDHFH9GHH9HEH8H9uHT$HH52FH0H81iHH[]A\A]A^A_DHT$HT$xJHt$(+D蓄HuHt$(HT$ 誁xnHt$(rHH9YHHT$H5E`HF Hj/HT$H5EH81謃HH[]A\A]A^A_H!Ht$(AUAATUSHH9H/H9GHIH9F A|$ HUI9T$HEIL$H9@H@t Hu ED$ D8u}@ H}HA UIt$Hȃ\cDA9uAHHS1Au%1H/H9ut1AH[]A\A]@I9uuHDLiHHH;F.H;-$.uH9u8HmuHD$ چD$ @1AH[]A\A]@H谀fDHM0H}H@HE@ ^:fLGfDIL$0It$HA@HED@D@ATHHu萁E1HuHHDA\fDH/tRHc,HAH53CH81HDA\H,H8!t؅fDHt$薅Ht$A|@AUIATLHH=3 HGHHtIMtHLA\A]D蛀IMuH$-LH5`H81(HLA\A]ff.AUIATHHGHHtIMt#HLA\A]#IMuH,H8tH*LH55`H81HGtkHGHPHwUH fHcHDG@GWHH fGWHH HGH~UHH@`HtvHHtjHHt`H+H9EuHHHmt)H]HH57bHHuHHHD$苃HD$~HuH+*H5]H8dfATIUSH謃HtHHHXID$HHtsH=t;~uH1LHI虀Mt$Hmt L[]A\DHL[]A\[~IHt-E1fDE1[]LA\DL1H+IH')H5;H8萃ff.ATIH5*UHSHH9t tMHUBtBHZ ujHmH=:*}uCLHIMt!L[]A\f[LH]A\{}IHtE1[]LA\1H](H5>:H8Ƃ@AUIATIUSHHHIEHLhHHX ID$I$HHH=9P|uq1HLI~MtMHmtI,$t&HL[]A\A]fHI,$uL HL[]A\A]s|IHt8E1fDHE1[L]A\A]1HL;InH4'H59H8蝁SAUH5(ATUHHH9t }tOHUBtDLb HmH=88{uq1HAI}MtOL]A\A]fDHEL-ͼ LMtqH=R8zu&1LHAIv}Mu[{IHtE1]LA\A]1eH:&H58H8裀ofDLH1]A\A]fDATHHGHLgID$Hw|H`HcHfDgAMcfDH/t HLA\ HLA\ÐDgf.DgGII DgGII IH|$yH|$IfH@`HtgHHt[HHtQH@H;.&H5\HHuNIFH|$H|$I)fyHuH$H5XH8 H@DAUATIUHHHWHBpHtH@Ht H]A\A]HBhHHHHHg%I9D$$ID$HpHYHAt$HH;2$4H;$t3LbhMIL$HHHH]A\A]HUHHH9HDHH]A\A]At$AD$HH fHHHUH;#H;+$bHyHUHHHq#HRH56:H81w1yHfHh}IHHHwI,$BLHD$Y|HD$+HUHH9vHUHHHyHUH2fwHHEH;"SH;:#RL`hM:IL$HH2I$HfHt$HHt$H"IL$HAL}IHlHxImHiLHD$Y{Ht$RH$HAt$AD$HH fDH;!1H;R"HU1H"HH2ut&{ID$H5z8HPH!H81u1LwHHoHHU1:HD"Ht$H8uHt$tHt$zIL$Ht$DAUIATUHHSHHHWH= yIH H@HEL#MtI$HL[]A\A]DcuHuH= LHGHHt'IMuH!LH51UH81ttIAWAVIAUIATAUSHHvH=* HHD$(L|$8Ht$0HLH$Ht$nwH= sHHtfH8H H9GLǛ MFH' I9tL;3 LsHT$8Ht$0H|$(u2H=} H5F HGHHHHH|$rH|$HHHBHH9tDHT$8Ht$0H|$(DuH մ EAHD DǃHcHD9|1D}7p9}%)HcHTD9~؉9|A9A9HHHD9xL(IEHd H1LWpHHt7X(H4wImuLewHmt.HH[]A\A]A^A_ImuL9wH(wHH[]A\A]A^A_fHT$8Ht$0H|$(,uELLsIH|$(MHT$8Ht$0sH=W EH7 rA:HcHD;| 1.AL$A9D)AAAAAIcHDA9}EHT$8Ht$0H|$(VtDHQL1H=QpIH?HwHHLrI,$ILuuHH5B H=k tHT$8Ht$0H|$(rH # HD AE1DǃNQ1GE1fA9AD9McIIE;x 9HcHHDoHcHHD9ExM(m IE8I,$u Lt@H|$(Ht H/H|$0HtH/tyH|$8H0H/&RAԋ 9Dp@IcHnnHHMcċٰ Hڰ ID5˰ ID9>r ttfH5 HL$HVsHL$H IHH@H Ht$H<$LAquHH{H9 DxH& L(IEMcIIHHHIHD$LD$TsLD$HD$FI8M(H/0sLdHEvnHqff.AWAVAUATUHSHH(L5LnH$HD$Lt$HrIHIHV(HE LMHHHsHH=AHEHAVj5 5F j5 Pj5 / IHEHPMHHEH(L[]A\A]A^A_HnH5A LIHVIqH$HLmDIH MHMAHMEIHHsHMH5/AUL QH81lX5ZH zM^ H=/E1>LHXqII[qMIHFHH$mIH5 LHVpHD$H3IML $HD$HT$8f.IHF(HHD$HF HD$HFH$dmIM~HL{LLLH5C y4fHHEt&H ML *5H=n.@H8pfDH LAXfDH5 LHVoH`HD$IMfDHF HHD$HFH$lIHHtH KH5-jL OAHKH81jY^4H [KAAVAUATUSHHL * H-LfL $Hl$HMI;I!MMHJMH JHJHIHHI?ATIH5,H8L NA1iXr4ZH J H=,E1HL[]A\A]A^ILHHs HW H=AI$H LUjRPjRLPj5ߞ ! II$HPMHI$xLnHL[]A\A]A^LV LNfIIIIt]MHjIHL $LT$fHI$H IY 4H=+oFH)$/jH~HLYILLH5? ~ya4vHFHH$iIMdH5 LHVlHtHD$IF뒐LlQH5 LHVmlHtH$Iff.AUIATUSHH(H Hr HLffHnfHnflHD$)$MIMIHILNLHHsHH=AI$Pj5T QLj5 RLj5, ֧ II$HPMHI$t0H(L[]A\A]fIujLV(HHV lfDLkH(L[]A\A]f.HQII8fDH9HV IDH 1GAHHHCGH5)ATL 3KH817fX3ZH G H=)E1uH(L[]A\A]IM?IHFLH$}gHH~TH5 LHV2jHtHD$HH~.H5 LHV jH<HD$HH*L $HT$LT$H $ H MH FHFMHII?IAIHF(LHD$HF HD$HFH$fHr@HI$t&H E 4H=.(!@LifDLPfHH#H5 LHViHH$HHF LHD$HFH$eHHL/ELLH5p; K3fAUIATUSHH(H H H LffHnfHnflHD$)$MIMIHUILNLHHs H=(AI$Pj5 QLj5 RLj5  II$HPMHI$t0H(L[]A\A]fIujLV(HHV lfDLgH(L[]A\A]f.HII8fDHyHV IDH qCAHH@HCH5Z%ATL sGH81wbXU3ZH GCG H=%E1H(L[]A\A]IM?IHFLH$cHH~TH5 LHVrfHtHD$HH~.H5 LHVLfH<HD$HH*L $HT$LT$H d HMMH VBHGBMHII?IAIHF(LHD$HF HD$HFH$bHr@HI$t&H A ~3H=$a@LefDLbHH#H5m LHVAeHH$HHF LHD$HFH$?bHHLyALLH57 A3fAUIATUSHH(H  H H[ LffHnfHnflHD$)$MIMIH ILNLHHs H= AI$Pj5ԝ QLj5 RLj5, V II$HPMHI$t0H(L[]A\A]fIujLV(H HV lfDL(dH(L[]A\A]f.H II8fDH HV IDH ?AHH H?H5!ATL CH81^X2ZH ? H=e"E1H(L[]A\A]IM?IHFLH$_HH~TH5Ɠ LHVbHtHD$HH~.H5 LHVbH<HD$HH*L $HT$LT$H H MH >H>MHII?IAIHF(LHD$HF HD$HFH$$_Hr@HI$t&H 5>B 2H=!@L bfDL^HH#H5 LHVaHH$HHF LHD$HFH$^HHL=LLH53 2fAUIATUSHH(H A H2 HLffHnfHnflHD$)$MIMIHILNLHHs H= AI$Pj5 QLj5` RLj5l  II$HPMHI$t0H(L[]A\A]fIujLV(HGHV lfDLh`H(L[]A\A]f.HII8fDHHV IDH ;AHHH<H5ATL ?H81ZX+2ZH ;m H=E15H(L[]A\A]IM?IHFLH$=\HH~TH5 LHV^HtHD$HH~.H5@ LHV^H<HD$HH*L $HT$LT$H HMH :H:MHII?IAIHF(LHD$HF HD$HFH$d[Hr@HI$t&H u: T2H=v@L`^fDL[HH#H5 LHV]HH$HHF LHD$HFH$ZHHL :LLH5/ 2fAWAVAUATUHSHHL5mLnH$Lt$HIAILV LMHHHsHH1 H=AHEHx HAVjRPjRLPj5@  IHEHPMHHEH\III9MMH8H 8HOH<L ~8LOL@HHLAUH8H5dH81WX1ZH Z8 H=E1HL[]A\A]A^A_fDMHHEtvH 8h 1H=6yHXH5 LIHVIn[H$HML $LT$Kf.H[fDoFH)$OXH~HL7LLH5C- y1HFHH$XIwH51 LHVZHtHD$IGLmD@AWAVAUATUHSHHL5LnH$Lt$HIAILV LMHHHsHH H=AHEHȓ HAVjRPjRLPj5 J IHEHPMHHEHHZIII9MMH5H 5HOH:L 5LOL@HHAUH6H5H81TX1ZH 5 H=E1HL[]A\A]A^A_fDMHHEtvH ]5 ;1H=HVH5a LIHVIXH$HML $LT$Kf.HXfDoFH)$UH~HL4LLH5s* y1HFHH$`UIwH5 LHV XHtHD$IGLmD@AWAVAUATUHSHHL5 LnH$Lt$HIAILV LMHHHsHHх H=AHEH HAVjRPjRLPj5  IHEHPMHHEHWIII9MMH>3H ?3HOHS7L 3LOL@HHAUH]3H5H81*RX0ZH 21 H=xE1hHL[]A\A]A^A_fDMHHEtvH 2 0H=&HXSH5 LIHVIVH$HML $LT$Kf.HHVfDoFH)$RH~HLR2LLH5' >y~0HFHH$RIwH5х LHV]UHtHD$IGLmD@AWAVAUATUHSHH(L5]LnH$HD$Lt$HrIHIHV(HE LMHHHs H=|AHEHAVj5\ 5 j5N Pj5 ߐ IHEHPMHHEH(L[]A\A]A^A_HQH5 LIHVI>TH$HLmDIH m0H^0AHMEIHH#H0H5=AUL V4H81ZOX 0ZH *0H=E1>LHTII[qMIHFHH$PIH5 LHVHSHD$H3IML $HD$HT$8f.IHF(HHD$HF HD$HFH$PIM~HL{/LLH5$ `y/fHHEt&H ., 30H=i@HRfDH .AXfDH5 LHVERH`HD$IMfDHF HHD$HFH$?OIHH$H E.H5>jL W2AH.H81NMY^/H .AAWAVAUATUHSHHL5LnH$Lt$HIAILV LMHHHsHHq H=AHEH؊ HAVjRPjRLPj5Ј Z IHEHPMHHEHXQIII9MMH,H ,HOH1L ,LOL@HHAUH--H5H81KX{/ZH ,eH=E1(HL[]A\A]A^A_fDMHHEtvH m,/H=>HMH5 LIHVIOH$HML $LT$Kf.HPfDoFH)$LH~HL",LLH5! yk/HFHH$pLIwH5 LHVOHtHD$IGLmD@AVAUATUSHHL-"LfLl$H8MIMH+H +HILH*H?L /MLIL@HHATH9+H5 H81IX.ZH *H=E1)HL[]A\A]A^fHVLHHsHL ^| H=E1I$H LAUjPAQjPAQjP0 II$HPMtxHI$uL6NHL[]A\A]A^fDL|HMIHFHHD$JHNHT$>f.HI$t&H )b /H= @LMHT$LHL)H5 y.f.HJIHhH5=} HHVLHtHD$IF=fDAWAVAUATUHSHH(L5LnH$HD$Lt$HrIHIHV(HE LMHHHsHH=$AHEHAVj5̅ 5fz j5 Pj5Ń O IHEHPMHHEH(L[]A\A]A^A_HHH5 LIHVIKH$HLmDIH 'H'AHMEIHHH1(H5 AUL +H81FX|.ZH 'H= E1>LHxKII[qMIHFHH$GIH5|~ LHVJHD$H3IML $HD$HT$8f.IHF(HHD$HF HD$HFH$GIM~HL'LLH5 пyj.fHHEt&H m&.H= @HXJfDH !&AXfDH5)z LHVIH`HD$IMfDHF HHD$HFH$FIHHH %H5jL )AH&H81DY^`.H {%AAWAVAUATUHSHHL5=LnH$Lt$HIAILV LMHHHsHHw H=BAHEHH HAVjRPjRLPj5@ ʄ IHEHPMHHEHHIII9MMHn$H o$HOH(L N$LOL@HHAUH$H54H81ZCX-ZH *$uH=E1HL[]A\A]A^A_fDMHHEtvH #.H=FIHDH5 LIHVI>GH$HML $LT$Kf.HxGfDoFH)$DH~HL#LLH5# ny-HFHH$CIwH5w LHVFHtHD$IGLmD@AWfAVAUATIUSHH(L5LnHD$Lt$)$HrIXIHV0HK(HC LKHI$It$HH=AHEAVj5y QHj5}} Pj5d}  IHEHPMHHEH FHBIH5} HIHVnEH$H<LkDIH!H !HMHAUH"L@HLH5mL %H81@Xi-ZH \!H=E1H(L[]A\A]A^A_LI^HH&JcHHF0HD$HC(HD$HC HHD$HCH$AIItlI~.IML $HD$HL$HT$-MIH5{ HHVDHD$H1IH5w HHVCHD$HIzfDHHEH  p-H=H5t HHVCHtHD$IM#HLN LHH5 U-CHCqHFHH$P@IHH5H VH5OjL h#AHH81_>Y^K-HHAH5 jL $#H H8H1>_E-AXff.fAWAVAUATUHSHH(L5LnH$HD$Lt$HrIHIHV(HE LMHHHsHH=AHEHAVj5{ 56p j5y Pj5uy ~ IHEHPMHHEH(L[]A\A]A^A_H>H51y LIHVI~AH$HLmDIH HAHMEIHHcH!H5}AUL !H81LHHAII[qMIHFHH$=IH5Dx LHV@HD$H3IML $HD$HT$8f.IHF(HHD$HF HD$HFH$T=IM~HLLLH5 蠵y,fHHEt&H =,H=@H(@fDH AXfDH5o LHV?H`HD$IMfDHF HHD$HFH$ A1A%XZH H=E1HL[]A\A]A^ILHHsHHW H=0AI$Hb LUjRPjRLPj5?Z e II$HPMHI$xL)HL[]A\A]A^LV LNfIIIIt]MH%IHL $LT$fHI$H :H=UoFH)$%H~HLgLLH5ޝyvHFHH$P%IMdH5uX LHV(HtHD$IF뒐LX(QH5X LHV'HtH$Iff.AWAVAUATUHSHH(L5LnH$HD$Lt$HrIHIHV(HE LMHHHsHH=tAHEHAVj5` 5fU j5_ Pj5` Oc IHEHPMHHEH(L[]A\A]A^A_H#H5Q` LIHVI&H$HLmDIH HAHMEIHHHH5AUL H81!X8ZH H=E1>LHx&II[qMIHFHH$"IH5^ LHV%HD$H3IML $HD$HT$8f.IHF(HHD$HF HD$HFH$"IM~HLeLLH5Кy&fHHEt&H maH=٫@HX%fDH !AXfDH5)U LHV$H`HD$IMfDHF HHD$HFH$!IHHH H5jL AHkH81Y^H {AAVAUATUHHHCLfHD$HMpILfHEH5U HHH4HH6&IH5H5T LHS&+HEL5` LMH=)LLHAIH!M'HmImL#HL]A\A]A^fMHNH OHILH8H?L XMLIL@HHATHH5H816XZH H=dE1tHL]A\A]A^DIH"/AHmuH"ImD溽E1H H=HL]A\A]A^IMIHFHHD$HLd$fDLLH!IHwfDAKDL"S[HA5DHmAHA! HT$LLLbH5ΖFm@HZHH5H8d"?HIHH55Q LHV HlHD$IFf.AVAUATUSHHL%LnLd$HMIMHH HILHH?L MLIL@HHMAUHKH5eH81X8ZH [H=E1ɦHL[]A\A]A^fHVHMH=Hs HEH\ IHEMtmHHEuHHL[]A\A]A^LHMIHFHHD$yHNHT$nf.HHEt&H }YH=@HhHT$LHLH5xy*f.HIHhH5 O HHVHtHD$IF=fDAWIAVAUATUSH(HH5O HGHHHHH{H;=/!HkHLCHEIH+-IxH5H9tL$L$ IPBHZ M`H=L$L$ HLIML$ HmI(IFH;0zIVHIF(MfMn H$I$IEHI.JHHHN HHHN HCHH=O !IHH+I@L5pN LD$LHLHLD$HHH}I(mIEH;uH;5H@hHrH@He1LHHIHHXLI$H5R L` IGHHHHIHHsH5Q HtHCH5X HHH=bHL$Ht$ Ht$HL$LHHD$oLD$MH+I/LD$LD$HIfHnfInHEflL@(@I,$IEMHIELZDHL$DL$LHL$ɗHmL$IMH$1E1E1E11E1A+BfDE1LMMt Im`H<$tH $HHD$HHHt HmH(L[]A\A]A^A_H5H9t HSBHj LcH=,T1LIMII($LfHQHQL/HCL%}V HHMH=1LHIMjfDIHWE1E1A+I1E1E1I(LH$LLT$T$D\$(D\$T$LT${DHL$L$MfHLD$LD$fLLD$LD$fHHD$LD$NfLIEH;-IEHHHLD$SLD$BfI]HHHmH$1E1E1AE1HtH+t_MtI*t4H DH=>MI,$E1fL׉T$D\$T$D\$@HLT$T$D\$zLT$T$D\$xH; 3LHHI.HEHHIH%HIHHH$HHHcHHm+HfAE1I.tTE1Mt I(\MI/LLT$T$D\$qLT$T$D\$dfDLLD$T$D\$BD\$T$LD$fHHx.HHH HEH|H5H81H$1E1E1A41E1E11AqfDE1`IE111AvfDE1?HQLD$H8LD$LúAy?@LD$H fDIVHIFL LhH@H$>f.kLD$AH HSH54LD$T$H8D\$D\$T$LD$HٺLH5(H81LD$AyLhD19IHHLI.HzL-mLHIH8AGfAA+E1E1E1i@AE1E1Af.HE1ADHmLA+E1E1E1fDL$HIHmtlE1E1LA+HH5UH81fDAE1#D1A fDHLD$1E1E1LD$E1ҺA+1E1H$fDLp61LHII91ۺAAH$1E1E1E1E1ANfHH5zH8E1A+E1E1Hmtob HIHLH H5mHEH*H81 MMH$1E1E11A`AHpHH5H8XHmL$E1LA+A?H(tsHH5lH81 L4$IMLAXHCH8S t H&H86 Hf.AVAUATIUHHGH5@ HHHHH}H;=LuMLmIIEHmtWLLLiI.ItPMImI,$H·HH]A\A]A^f.HfDLfDH5H9t / HUBLj LuH=l u@LLAII MD HuHҵH5H8;HmH H=耕H1]A\A]A^fLLLHIrIf.f H1E1Ht$ t$ SL8AVAUATUHHGH5zD HHHHlIH{HLH5B HeHEL5J LMH=;uxLLHAI^ MtQHmt2ImtHL]A\A]A^@L HL]A\A]A^ÐH ImuHAHmuHU ImtnDE1H *H=m蠓HL]A\A]A^ÐALLH IHtHm-Zf.L fDA|wDHnHmA~QHA~ >H H5H8v AUH5C ATUHGHHHIM ID$H5? LHH0II$HI$M&HH}H5C HGHH.HH0HEH5e? HHHrIHEHM<HEHH=D LP HHDI,$H5D H* IH.HmLLHHI,$ImHEIHPHUHEHt.L]A\A]DH TL H L]A\A]L BH X-I,$uL LE1H ^H=ԐIEHPLH -L8 IMH H=orL]A\A]fKIHH E1H=$'f HA"f.DLE1H iH=ߏIEH[fA$HmuHK fI'fDA*L ,AWAVIAUATUSHHGH5$8 HHHHH}H;=CH]HLeHI$HmI|$L=L9tL?IT$B/Hj Ml$H=2LHIMH+I,$H=@ H5v> HGHHRIMTHC H{L9tLDHSB Hj H[H= LHI MpI|$H;=I\$HM|$HII,$8LHLjH+H-I.HI/CHLIIEMHIEHmHL[]A\A]A^A_f.HLH5IHfImAu LkMt I,${H BDE1H=譌lHLłH+IMLHm|HAf.L=L9tLLHUBLb ALmH=u1LAIMII,$bLfUL9t;Lu/LLMI.H!@L(IT$BtHZ Il$H=L?HLHoHIMHItI/LHEL-]A LMH=޼y|1LHAIMHRHتH5H8A7@H0MfHGL,LHLx1,fHHABDIIEAHIEE1H+NHLKAfDE1HIE:AHmH @{HrHsH5TH8WKHuHGH5(H8I.t"IEAHIE'DM1NA1ffAH0HHH5H8H+ DMVL1HIpILff.AUATUHSHHHHH/>HkHEHH59 HH,IM/H5DLGH5-LhIHAoEHC AoMK0IE HC@HC HCHHCPHHH(tdHEH55 HHHHHHH/tSHHHI,$tH[]A\A]땐LH$H$H[]A\A]D{fkfDIMH 6H=袆H1[]A\A]DH-; L-- HEHHH=G1HLHhHHtHmtnH H=h1@H>#fD6@HrA@Ht$ $9$t$ wH1LHHE@ K|HuHxH5YH8 fAWAVAUATIUHH H|HEHEHIT$H5%. LHHHyIMKIHzH5K. HHIFL=: LMH=n LLLAIMdI.I,$wHmH L]A\A]A^A_@HIMI|$H;=M|$MMt$III,$!I~H5H9xLLyI/IM}I.ELHm=f.H&H5H9[ID$L59 LMH=uG1LLAI$MHuHH5H8hA(I,$DH .E1H=衂T@IT$BELj TMt$H=Pu1LAIwMVMfDIVByLb MnH=LLAIM#I/RLq<@L`LPaL@HIHpHD$L|$Lt$HD$1LLLHD$H@uHHtH5E1H81A(NLT$ T$ 1A(I.L[LLLIHBfDA(A(A([IKI9I.LA(tfDE1I/tpMBfE1 HRHH5H8l7HuHןH5H8@I/uLMA()L1LMIHL$HEHHH5 E1H81A(xfATUHH5 H k! H9HHR! HHH->! HHEH50 HHHH}HWHUHt4HtH]A\HHD$+HD$H]A\fDSHH H=b~H1]A\H=) H H5 }H8fDS말 IfDH=a4 L%B) HGLHHtIHHHbH5LH81fSB@H@S,Hff.AUATIUSHH3 H  H9HH HHH- HHH9HELHIHH(t9HIHmtHL[]A\A]DHHL[]A\A]fHhfDH=!3 L-( HGLHH]HHVHLH5H81"H 7TE1H=b|HL[]A\A]@H=' H H5 {HfDHi1 HH}H9HXHHqH~1H;THH9uHHJH5HWH81\Hmt]H ,9TE1H=R{HL[]A\A]H DTH= c{LfDH9HHRH5uH8jHHH9HuH;fAWAVAUATUHH HIMVH1 HEH # H9HIH  HHL5 MMIFH5R% LHHgIIHIM'HL^IHIELMH=LHLAIM}ImtfI.tPHmt)I,$t1H L]A\A]A^A_ÐLHmH8I,$uL)LfDLI.u{HI.TImH H=E1-yHLHL"IHtImzf.ILt$Ll$HD$HD$1LLLtH5LH81BTPT3fDLt$ t$ #H|H5]H8LT)IkAWAVAUATUHH H"IMVH_- HEH s H9HIHZ HHL5F MMIFH5! LHHgIIHIM'HLIHIELMH=7LHLAIVM}ImtfI.tPHmt)I,$t1H L]A\A]A^A_ÐLmHxI,$uLiLXfDLHI.uHI.TImH H=E1muHLHLbIHtImzf.ILt$Ll$HD$HD$1LLLPtHHtAHmhHHD$HD$H []A\A]A^LxfDH RAHD$H=hHD$IIMnLH5 LIHVIHD$HMH|$HT$HI$t.H AH=N!h1lf.LfDoFL)D$>H~HT$LLLaH5 ViiAfHFLHD$I4H5 LHVHtHD$IF|LmZH /H=f1H []A\A]A^ÐHV H}LHHs E1H AI$H % jQPjQLPj5j WH=҇ HHPI$HHI$t>HHtAHmhHHD$HD$H []A\A]A^LhfDH B>HD$H= eHD$IIMnLH5 LIHVIHD$HMH|$HT$HI$t.H u>H=ne1lf.LfDoFL)D$.H~HT$LLL̾H5{Si<>fHFLHD$I4H5 LHVHtHD$IF|LmHHtAHmhHHD$yHD$H []A\A]A^LXfDH 2=HD$H=+bHD$IIMnLH5 LIHVIqHD$HMH|$HT$HI$t.H =H=b1lf.LxfDoFL)D$H~HT$LLLKH5kPi=fHFLHD$I4H5 LHV}HtHD$IF|LmH <H=f[1fH AfDLfDH5 LHVeHYHD$ IFfDHF LHD$HFHD$^IfDLHH51 LIHVIHD$HsLmHH{H %H5jL 7AH9H81.Y^v<H AAWAVAUATIUSH(H{LnH\$HMIkLvIl$H HEH wH9HH^HHL-JMLHNAHEAHHEWIm<E I|$H5 HGHHHHH}H;=z<LEM/L}IIHm LLLLD$QLD$II(cMI/ImID$LPHH>H(HH(H[]A\A]A^A_fDMH&H 'HILHH?L 0MLIL@HHxAUHhH5H81XZH ޭH=dOXH(1[]A\A]A^A_fDIH5yH9t$/uLHI=NIDHUBtLj L}H=YLLIAIMAAH~HwH5H8cLHLFL+JfDHLD$LD$HHEImAADDH \H=VH(1[]A\A]A^A_ÐLHHM IHFHHD$HsLt$GfDLAAlH= HH5UIHHH=OHLLIoMI/HmIEH5> LHHCHH%ImzH}HsH9VL}MILmIIEHmH LLII/I"M ImID$H5 LHHHI$HI$HHH}H9L}MLmIIEHmI}H5rH9t IUBHZ ImH=RzLHIMI/ImID$HXpHjH{_LHHgH=qHHHmI[M>LLSI.II$HPMI$HHL[]A\A]A^A_@DvLMfHLIII8DvFII DvFII IIuAIHH $E1H=nO[fDvAMc@LLuEI/IhMf.HPH]oH5H81I$HI$A&FH ^DE1H= NHAFf.L(iH5ApL%H9t tdHUBtYLj L}H=2u+LLAIIUM7HALHI2DIXf.LA&FmLX LHH5aoH9t tKHUBt@HZ $LeH=Yuv1LIMtTI@HEL% HH%H=u+1LHI3MufDHA#Hm8FE1E1fDH`H3I L@L0HmdHWLyLAEi;IHrLAEE1I/t|Hm)MtI,$t%MImLi@LXfDI/uLBEAfL(LwLE1AEKDHLLZIHfH LAE1EDH=qHBH5C.JIfDLXE1AFDL0oH= L- HGLHHIMwHkLFH5gE1AH81ܾ%IYAFL1HIIHxFA軾HAE1EE1f.諾IHE1AEj@cHE1E1I/t@LHf1[f;HuH7iH5{H8I/uLLf.H@`H[HHKHIH:H@H;izMuIFHHJHcHEuIm}L@ImALE1KEuAMcImt밐EuAEII I@EuAEII xLH5-IH0H@ L8IL覼I@HgAEH5yH8v蓼HHgH5\H8HpgH5QyH8&HUgH56yH8/H@`H HHLHHH,hH9GuzHGLwIFHHHcHDwAMcH/y2oDwGII IDwGII DwH5,HHnImLٿ@H|$覺H|$IxH|$/@H|$IaHuHIfH5H8AWAVAUATUHSHHHFH uu IL=eL9H;zfH@hH' H@H 1HIM L5M9HfI9D$I9FyqA|$ A~ b IT$I;VI,$uL茾H--L%HEHHm H=vNr 1HLHԻHWHs2HmsXuHT$4$4$T$H H=΋E1fDHL[]A\A]A^A_@H5 H kH-dL%HEHH H=u腸 1HLH H2H1HmmI.L-eM9uM9uLLoIH1H;LdL;5*dM9LI.AEAIrE11I,$aE1E1LL$裼ML$7Im,LL$L$fDHEL I$hH5HY9 xHEE1HEH5`HHH5 IM ID$L=CLMH=4t϶1LLAITM+ I,$HbI9ELf.Y ImCXHEH5HHH IM IEL=LM H={s1LLAI蛸MR ImID$ID$HHO H#HcHfID$INH9@H@t HA|$ Av 8@ } I|$H@  IvHȃ  DA9HE1HAI,$EEsH-L%JHEHH7 H=r躴1HLH@HtH-Hmq5~cDAEDLeI$kfI,$^лIHHEL9OH;`H@hHMH@H@1HIMeH5&LL;[ImSIHHEL9*H;`H@hHHH@H;HIMH5LLL$跺L$cI(HEL9H;_GH@hHH@HHIMH5(LLL$AL$I(3H5,LLIm~HHHLE1\I,$AnuE11E1ImMtI(t\H ;DE1H==HtHmt%M.I.$LDHfDLfDHEL(IELȶL踶L$裶$fHEL@ILmIEmf.LX\LHHEL@I(1LL蛵IHTA|81ɶIHuIH ǑrGE1H=0<LE I.L蠵HHImILwrfE1DI,$H{DkPHH5HGHH7ЅH\HIf.El$AAuuH AyEl$fAD$AT$HH HHcAH9tHz\H5nH8蛵fAD$AT$HH HcAH9pfDI.DBLRtDLc讯IHR H q1E1H=ׁr:H1LpHHH ˏsTE1H=4:H1L+HH-H mE1H=T9f.Lh~H =H=E19L0u1LL蛲IH }E1ADLE(IHEL9#H;ZtH@hHH@HHIM*H5wLL贯I.HEL9H;YH@hHH@HHIM6H5yLLFI.I$ML@|;I1該IHt,HHFI.IL衱@E11AluL[HcAH9lHݬHUl-f$蠬$H|E1AA|vI,$E11nE11AxvIU}Cf A|HHVH5hH8a@.IHHHǪI.ILH$L$DAvfDAzvLXAuE1D+HH#VH5hH8茰VIHeHHI.IBLH$FL$-DA~v;LwE1lf@IL$0MD$HIDHoIN0IH@IDH_fHELpI H1LHH_LxHELp I7LWNLu0IDLu8IDIHHH菨ImIRLE_IHHTTH55fH8轮sH TH=u{4E1AzvLAMxE1Lx0IHHHɧImIL#LAMyE1EE1A~vLyA}H@`H@HH0LIHH@L-?TL9fDCIGHHHΎHcHEoDI/FL>9I/AL!EoAAGAWHH HHcAH9tHSH5`eH8I/LΫAGAWHH HcAH9[fLH5IHH@LTHcAH9He֦HWh軦IHHQH5cH8mH H=xl1E1Lx1AIrHQQH52cH8身qH 1H=rx 1E1Ly-H{H&QH5DŽH8_`H AujH=x0BH@`HOHH?LIH.L9hI@<I@HHHVHcHEhI(oL跩bE1EhAA@APHH HHcAH9tHPQL$H5bAH8iL$A@APHH HcAH9tH5HAIH%@LL$L$HcAH98HprL$H^AWHHOH5 H8踩H@`HtzHHtnLL$L$HHtYL9hu6LD$H<$H<$LD$AH/L$VL$H5̆QL$HHuKL$解L$H5HNH5lAH8L$.L1E111E1E1uAnLff.AWAVAUATUSHHhL-`OLfLl$XH MILfHD$0HD$8HD$@HD$HHD$PHD$XM9HH 2H9Hh HHP HH-Hl$@H HEH5HHHY IL|$8MV H|$@H/oHD$@HD$@HHk I$L`HD$0IHl H5H ^H9Ht HEHHH-1H HEH5HHHJIM HmH5H|$0L5 I.Hl$8Lt$0Ll$@HELM H=]蔠 HLLAHHH|$8H/ HD$8H|$@H/X HD$@H|$0H/0 Hl$0H} HEuHD$0H} HD$HHD$(HIHC E1HD$ LL-IVLHHT$>IH H@HT$HHYLLIHLL-gIRLLT$HHT$ޥIH HxHT$LT$HH4LLIHHxHD$@H;=KIFHD$@HMnHIEI. Lt$@I}MH5KH9FLLS H|$@HD$0HtH/7HD$0HD$@H ImH|$0H/iHD$0H|$ INHL$LJmIGL5jLM H=k[ 1LLAI苠M Ll$HI/SL|$HME I/KHD$HIL9d$(HEIHPHUMHf~H g~HILHP~H?L pMLIL@HHIATHrH5(`H81NXZH ~lH=$pE1(HhL[]A\A]A^A_f.HLIUBHJHL$ M]H=ZL\$蛜L\$VHD$LLIMH|$@Lt$0HH/^fI;fDMH5NIH9蘞IEH5LMH=>YHt$ԛ1Ht$LAIWM>fD3HbH+GH5 YH8蔡H|$@HD$0HDMAf.IUBAHJHL$ x MuH=X HD$1LfIfDLkL؟HD$0HD$8HD$@HD$HHD$PHD$XLL=TMt$LLHH H@HHHLLHH]LL=Mt$LL藠HH H@HHgLLHD$8HH H@HD$@H;bFtHGHD$@HbHWHHH|$8HT$8H/Ld$8Lt$@I|$M/H5FH9t.u%LL H|$@HD$0HuerfIT$BtLz AMd$H=|V<LLAI螛MH|$@Ld$0Ht!H/HD$0HD$@H H|$8H/FHD$8H|$0H/HT$XHt$PHD$0H|$H豝H{ 舖H0HD$0IHlHT$XHt$PHD$0H|$HL|$XHEL%zHH H={U1LHH蜚H H\$PHmsHl$PHe HmjLd$XHM I"HFHHD$XiHLd$X'fDL舜LItHD$HL9d$(pAAE11H|$0Ht H/H|$8Ht H/H|$@Ht H/Mt I/Mt I.DDE1H wH=i:"HHEHHEHH褛蓛bfD胛ffDsjfDL`fLPeHD$0H|$@H@H Hl$0fD fDfDvfDHBLE1AAH8ϖzf.H=HH5V HfDHiBLAH8耖I/AE1E1E1AAfHLL袙HHfDAE1E11AE1HT$XLHLzH5i)H=yL-HGLHHzHH1Hl$@ӔIH|$01E1AAH,@E1FE1lE1E1AAf1E1AAf.H=HʻH5˻HfDHpIHH5HHV!HHD$XIFf.E1E1AAfIA1Af.H=L%HGLHH>HHH?LE1E1H5nsAAH81#ILAE1觗bfHEQfDE1AA<@1LLIqLd$8I|$H5w?H9ID$L=LMH=fOu*1LLAI芔MlH3HD$0H|$@HAHm AE1E11hHH|$8f苖fD{fDIT$B'Lr tMd$H=N3X1LAI軓M,DlfD胑IHHx<H5YNH:@ӕfDE1;HHH0<H5NE1E1AAH8臖BfHxHhHA=LE1E1AAH8LH=LAH8(HD$8D1H|$8Ht H/HD$8H|$@Ht H/H=bHD$@H pHT$@Ht$8H|$0HL$@HT$81Ht$0IHHELM.H=L91LHAI辑MHm I. ML;%@;L;%;M9LڍI,$A8HT$XHt$PE11H|$HAP[HAE1E1臓1@E1E1AkA)1LHܒHQHE;LE1E1H5nAAH817HD$@tH~臎HHH|9H5]KH8E1A+(HH|$0Ht H/@HD$0H|$8Ht H/HD$8H|$@Ht H/HT$XHt$PMHD$@H|$HI,$wL/]1L蠑h" 5LHHT$@Ht$8A@؎HD$0HD$8HD$@E1H18H5JH8蚒A/腑{ 1LL_ՌIHuH7H5IH87 A4h1LH蚐IAE1AAA"E1E1AAAWAVAUATUHSHHL58LfH<$HD$ Lt$(Lt$0HIIIMHDlH ElHOAL -lHQpELOODHH6H\pATH5NH81-X*)ZH kH=3^E1kHHL[]A\A]A^A_fLt$MLmIEHI$M9H H9H HH9 HH-H HEH5HHH HHEHH HEHdH=H5HGHH" HH$ 菑IH3 H5DLHHD$1HLIMHmtI.t)I/dLDžWfH踅fDL訅fD`A)H!E1E1`A)^f.H=!L5HGLHHHHH-LH5`H81"`A)f[HE1E1bA)f.E1bA)DLLL$諄LL$GHLL$蓄LL$%fZAl)E1RDH 9`AL Vd 1HL軃IHIE1L`A*fLMIDLcLDHLL$˃LL$ LLL$賃LL$fA<*DH舃LxeA**}~I H=yHH5HfD1A,*I/eu@LH}IHZAm)YA[)E1E1YA])f. ~H[HXH5ٶLHHVHHD$ HLeHF HHD$(HFHD$ HHH5+LHV跁HHD$0HH;)"HH)LKMLsIIH+LL$LL$HHLM\f.bA)E1bD|HfE1E1ZAh)FfA1*%DHHHE`HHQ1LH莀IHMHݺYLA)E1fL $W|L $bA)HH@'H5!9T$DD$H8蠁L $DD$T$sfDE1ɺZAj)ZDH5ILHVHHD$(HfDA9*DTfDL`E1E1A*HHLyIHA=*0f[{HH$HO&H508LH8赀L $`A*HUHJH]HHHmmLL$HT$HT$LL$HIIW+ZAm)E1 L $zL $ZAm)H@HݺYLE1A)E1L $ZzL $HHN%H5/7HLE1H8L $YA)zA=*HH%H56D$H8lD${CHH H{CSHH H)aHLL$&~LL$OwyHHLL$HT$}HT$LL$rIyHɅXCxHIHݺYA)L`A*ff.AWAVAUIATUSH(H@%LfH\$HfMEILfI9.H5'HFFID$LHHIMI.nI$LIELHHH H/AH+H|MH~XH XHILHhXH?L \MLIL@HH(#ATH\H5@:H81fwX ZH 6XH=JAH(D[]A\A]A^A_HH jH9HHQHHH-=HH}H;=Y#KH5#H96yHEL%[HHC H=3wv1LHHxH@vA AH.H!H53H80|HUBkHZ -LeH=R3u1LHvxHtHmIELHHHH VcH=IALzzHDH H9HHHHH-H_H}H;=!~HO"H$H9tHw>HUB/Lz H]H=82t1HAH[wHBHmHCH5tHHHHHH}H;=!L}MLuIIHmLLLI/IMI.I,$'Ly y8fDHMdI:HFHHD$uHLd$wfDHxH$H9t'HvuLHI-IJDHUBtLz uLuH=0MsLLIAIuMsA@AHfDHmH SDDH=YFdHJAE1HEH}LMH=/r1HHAHuHDrA%AHNA A15HT$LHLwWH5K_W LAA@fDHwHv4HH5jEAH8wH R H=Eyf.LvkHHsIHH5HHVuHHD$IFyfHHvH=qHH5H9fDH=ٲL%BHGLHHHH HLH5\QH81pf.H Q AH=D gpIrLeML}I$IHmhIH5xH9t rtZIWBtOHZ >IoH=p- phLHHrHlI,$tcL LL}I,$HtCHuA AI/LH PDDH=C LtfDLxt81zfH=HH5HfDH=L=bHGLHH[HHHLH5|OH81nf.H OAH="B-LUMuL}IIHmeHIH$H9HLT$pLT$uxLLLT$LT$HI*tHLmL sfD1A%AD[nH_HrIWByHj bI_H=+LT$mLT$}LHH)pHLT$EL$nL$HeI*LA%A?1LHIqHeA@AHL$rL$mmHH1HIuqH1fDH M2AH="@-mHI,$t_LA A1v mHt"I,$t8LA AQ1HH5)H8LrI,$uLAA 1qjI*t:L1A%AHH5r)H8qL$I*LA%ApH LcAH=?"1I.H L@AH=>A%ADAWAVAUATUSHHLfH\$HMILvHH ĘH9HHHHL%MID$H5LHHHI$HHI$HHmH 6H9H HH|HL% MLH=nAŃHmI,${HH@EH9 HH HH-HFHEH5kHHH0 IHEHHEMHI}H;=2 H5H9t lcIUBTHZ `ImH=&PiPLHIkM? ImL,nMHIH IHILHIH?L MMLIL@HHATH0NH5+H81hX2SZH IH=ME1HL[]A\A]A^A_fIfH9;HHHL%ЕMID$H5kLHH@HI$HHI$HHEHL}HmIH5H9t ajiIWBZHj MgH=%gVLLI&jMIEH5LHH_IIEHIEMHHH H9H4HHHL-M8IEH5-LHHHHImiHEH5.LHHHЅI,$[Hm@HIܐI/L~kfLLIHfH 7GSE1H=zKfDLLLIIMImH FzSE1H='KM4LjLjxHj]HM4IjHFHHD$9gHLt$DfDH= HH5IOfDH=L-ڛHGLHH]IMHLH5|EH81df.H ETSH= J3fDE1AVSI,$uLiIMtImt I/I*H[HHmH-EHc{P`IH H5HL&^~ I.CX[IH H5HL]d I.L9H;-D$ H;-OЈD$L9LsH4IH 9H9H H Ho HL= MC LL_IHImI/S+ |$>L$,H5HLI9D$p IMO H5HqLI9D$d ^IM? H5HDLI9Ew 2IIEHM8 IEH H I9D$LT$LH5~= LT$IM9 HI9FLT$LH5- LT$IIHM. IHHI9D$LT$LH5 kLT$IM HSI9D$LT$LH5ǔ@4LT$HH LT$^LT$HH fInfInHX8LflfIn@fInfl@(I$IHMH9H 9HILH9H?L =MLIL@HHhATH>H5H81XXxZH v9H=<,E1H8L[]A\A]A^A_fLIL=M9uL9uLrZHHvL9H;=1fL9]H|$VH|$AH/OE0I.VEjIFHNH9@H@t H3EN ~ Dȉ8A _MVH@ 9HvHȃA >9HLE1H[A4@HU&AL[,L[HU;HUD$2A3fDHIE1E11ۺ-I/Mt I*Mt ImHtH+tMH 7H=)I$LE1HHHtHmHZx@ZݐH߉t$T$ZT$t$H 6H=z)%LLT$t$T$ZLT$t$T$2fLt$T$pZt$T$+L׉t$T$PZt$T$L8ZI. LZfHMI HFHHD$(VHHl$(fDLYLHt$3[Ht$6fD&I._1E1E1E1LLT$t$T$lYMT$t$LT$HHYH-i@L0YLLT$YLT$%AH 4DH='_M;TI&AKSII$LHPI$@H/DX'E1E11fHHt$YHt$LLT$3XLT$hRLT$HHT$(LHLc8H5&)jX1LVWIH|f.E1'1K@LWLW+As+HTIHlH5HHVVHHD$(IF9fH=IHyH5yIfD,A'AH=L=HGLHHIMHLH52AH81Q'mfQI',LLT$3VLT$f-A QLT$HI:H{H5\1H8VLT$'mfDL-iL5*IEHHVH= P[1LLISM]LA!I..ULT$NUT$@DH=qH xH5 xIfDHN0HH@HE@IN0MVHA@LE -CfH=L-HGLHHIMrHLH50H81O -LhT3A>KNIE1Һ3@1NIA >M1E1B3I.tD{NInNLT$IE13E1\HNLT$IA >l1۾G3덺4R1 NLT$IT4[^3LL1LRIH.A"NItNA.HHH5c T$H8ST$9NI+-1E1E19f.AWAVAUATIUHSH(HFHTHzH9X3HzH#HL5zMwIFH5LHHHIHHtIH@HH{HD$H9/H5H9t fOHSBLj  LsH= LSLHAI*OMH+ID$H5LHH8IMJI}H;|$kMeM^I]I$HImLLHXI,$HHH+HmIHPIIH^H(L[]A\A]A^A_@H)H9%HXHHqH~1DH;THH9uHEHxH9X HxH HL5qxM IFH5ňLHHJ IIHM IHAHIHD$H9 H5 H9t WM'IWBLj MwH=JLHAIMM I/IFH5LHH:IMLH5|1LLIH?ImL;=]L;=;L;={LHÅ^I/IFH5TLHHIMH5T|L9IEH;8-I}uA}xHE1HHHH9uAImNH+THH@EH99vH$vHpHL-vMIEH5LHHHIEHHIEH7HD$H9CMvAQIH$1E1H+OMt I+ Ht Hm`H 7)DH=/Mt I.E1MIIm>L M1HH9eHuH;bNfHHIbIMLfDHL=LIMlIfDLhL?LXL-LHL9H=)HjtH5ktITfA.E@H9AHHH5H9t OItsIUBthHZ  ImH=Fu*HLLHIHFHLA3Q1E1lE1LLLH8fL߉T$DD$/KT$DD$H߉T$DD$L\$ KL\$DD$T$HT$DD$JT$DD$LJLJLJ+HJ E1LMJUDLpJH`JoHPJTL@J\H91r HrHTHL-rM IEH5LHHIIEHM IEH8IFL\$LH5yHHYL\$IM`IEH;5H; H@hHH@HL\$1LL\$HHUImW L\$IL\$HII]L\$KL\$HIHHpH9XHpHHH-pHHEL\$HH5"~HHL\$HHzHm H5eHLL\$JL\$"H+u ICHH H={L\$CL\$LLLHEHL\$I+Q Im9 I/" ID$H5wLHHtHHSH}H;|$ LmM| L}IEIHm HLLImI MI/ I,$ IFHPpHbHRHUHLIMI. MIfLFHFE1 E1H=yHnH5nVIELL\$FL\$LoFIEHHIغ1IMtHEIlI@HHH=rL\$LD$ALD$L\$ 1LLICMLD$L\$ I+I(L;-L;-tL;-mgL?Ņ} Im^cHCH|mH9X HcmH HL-OmM IEH5}LHH IIEHM IEHrI{H;|$6H5H9#L\$TBL\$ LLL\$WL\$HLIHTI+I.ID$H5tLHH[ IM: I}H;|$(MuMM}IIImmHLL^I.IM I/ I,$HEIHPHUE1AQkuDE1ATQo]DMkE1A QE1I1HHIHtE1HfLT$DD$L\$ZCT$DD$L\$f.H=L-"vHGLHH; IMHLH5E1H81=AQk3>IH=L-uHGLHH IMUHLH5 E1H81=ATQo=H'LE1E1oAVQMoM;IOIEHI/HHLHL$9ImHL$IOLHL$AHL$8LkML{IEIH+lHLLֿImILA(LLD$>LD$LL\$>L\$IUBtHZ @MeH=9u*LHMILH50H9t,~;u#HHI茺II]HHUBtLj "LeH=8u+LHIAI(;MA 9H IAoRHE11}HL\$Y=L\$LHL$B=HL$,H0=1LLL\$LD$1AKR|,1LH8IHAQq;4I.4I(HPhHHJHHMH9ScHSHrHHDcHH;H;HhhHHMHsMLLIIA Rxm3IImL|IAMRP?3L\$HJH=jL\$AL\$HE1H=jH_H5_RICAQw*H=cjIARR|1tIAaR}2HLwAQ|2IL\$2L\$ARR|HH rH5ST$DD$H97L\$DD$T$MkMI[IEHI+LLHƴImHLp6E1HH5H8M7oAeQAQE1qHPHH5H811IA|R~HLD$5LD$_1AQqH{H LH5-T$DD$H96DD$T$NE1HH5H861AAR|DcCII IIIFH;H;MIVLI;VsOMdI$L5jMLL\$LI4L\$|AMR:IL5HHHLHF(Lv HD$@H//vLX-f.6D$ L<-f.6 \D$HkH TD$ H9H: HSHJHL=SMIGH5cLHHXHIHH3IH D$ \)IHH{H;=wH5@H9t +HSBLj 3LsH=0(LLAIR+M I/H+ L;-L;-u L;-N Im H\$D$LIo(IHD$ HD$S(LL$HI2HSfHL5jH APH=H5ZjPAUjPHT$XLL$`hHPLL$HHI/ I)p ImE E1E1I,$HmIH~, fHh,EA;D$MDHHD$LHͨHII/Ht.MfE1E1A&(E1DL+fD1LHT$+HT$IF@L+dL+EHaLv HD$gH߉T$LD$DT$r+T$LD$DT$LljT$DT$G+T$DT$fLT$LD$DT$+T$LD$DT$QL*H AHHH9 H5ATL H81%X&ZH H=E1AHI# M IHFHHD$0'IM~TH5*`HHV)HtHD$8IM~.H5ZHHV)HSHD$@IMAHD$@Ll$0Lt$8HD$MH HMHII?IA@IHF(HHD$@HF HD$8HFHD$0K&IfDIٺ1HEIlI$MdIALM~H=LL$-$LL$1LLAI&MLL$I/I)HeIL0IHKH8HeH MH9HHMHHHMHHCH5jaHHHgIMAH+H(eH qMH9Hg HXMH7 HL DMM IALL$LH5]HH LL$HH] I)SH{H;=` H5H9t A%HSBHJHL$ LCH=LD$ t" LD$ LHD$LHD$$LL$M H+IH;= H5XH9tLL$$LL$tIWBeHZ IMGH=>LL$LD$ !LL$bLD$ LLHM$HLL$I)]I/H;H;u H;DfDH+E? HL$HMAH=HHq HHh_H5jP5SjPAUjPHT$XaHHPH H+RHHD$%HT$;f.LHeHIHMaAq'SLLLL$0LL$LHI)I0HE1A' @LLL$K%LL$$LL$HIL$LL$LxÅMA&1E1E1E1DI/ LT$E1LL$DT$-$LL$I؋T$DT$L@@HADžA'E1f.UHE1E1E1Ay&1D$ D$ HE1E1E1A&1DH=YVHHH5HIfDL@#HLL$+#LL$H#nL#hH=_L-UHGLHH IMHLH5DH81fE1E11ۺA'H=yUHGH5G6IfDL`"HP"A 'E1E1I/tE1E1RfDLT$E1DT$ "DT$T$E1'SHLkMLKIEIH+LL$"LL$HI) MoE11ۺAL'A4'1E1E1LHT$E1E1M!HT$LHT$3!HT$yfLHD$!LL$HT$J@LL$vLL$HI1E1A4'H=SHEH5EvHfDE1A['DH=Q]L=rSHGLHHHHHNLH5E1H81OA['XHغ1A]'I9kIE1ckHuHgH5HH8 I/E1E1E1A&DLL$LL$Hu$HH5H8~ LL$fI)A'1E1H=\L-2RHGLHHIMHLH5H81fE1E11ۺA&H=QHCH5C膤IfDE1E1HXIHH5RHHV HHD$0I fHF HHD$8HFHD$0IfD1A`'fDA&kHH=ZHPHGHHHIMHHLL$1H5=H81LL$A`'FfDLAb'MϺH/>T$DT$gDT$T$"fLL$HHfDE1E1A&KfLCMAHKIHH+LHLLD$ HL$%LD$ HL$II(LHL$ LL$HL$ LL$DLsM|HKIHH+bHLLHL$(貚I.HL$(I9LHL$(SHL$("fMGMIOIHI/LLHLL$(LD$ HL$@LD$ HL$LL$(HI(_LHL$ LL$HL$ LL$>Aq'E1]DL-)WL5RHIEHHH=jL1LLHHH菏H+FE1E1E1A&1Sf.L=VH5GIGHHcH=GHt$1Ht$LHaHHH+A'E11LL$LL$Aq'H-HH5T$LL$H8DT$YDT$LL$T$fD1E1E1A&E1A'6DHLL$LL$-1E1A&LLME1E1A&=HT$0LHLH5sΎ&L1A'E1HHL$ LD$]LD$HL$ QLHL$(LD$ LL$7LL$LD$ HL$(HHL$(HL$(gI)HHE11A'HVH571H8LL$A4'XH51LHHE1A'1LLHHE1E1E1A&iI IPE1E1E1A&14A'E11qIH`HfH5GH8A'3IHNH(H5 E1E1H8A&LMﺋA)'(IE1LAWfAVAUATUHSHXHLnH<$HD$@HD$H)D$0HIFIHF0HD$H](L} LuHSH(hE111HALIHe H8#H|SH(hE111HALHHY H8H8SL(hE111HAHAIH\ H8EA;D$HRH ^<H9HHE<HDHL51<MIFH5}NLHHjIMDI.HcRH ;H9H2H;HHL5;MFIFH5JLHHIIHIMHHQH Q;H9H H8;HHL$;MsI@L\$LLD$H5MHHLD$L\$IMI(HI9AL\$LL$LL$L\$HI E1Һs'@E1I+1E1Mt I/hMt I.yMt I+Mt I(Mt I)Mt I*H H=BMtI,$E1Ht HmMt ImHH+HL0HH5FLHHVHD$0HLm@IH H HMHAUHL@HĹH5L H81X>ZH H="E1BHXL[]A\A]A^A_HD$fDL LL\$L\$jfHxHhA;EJBIF"MFI@HHHcHEFAFII fIG$MOIAHH~HcHDEOAGII fHC"LsIFH]HBHcHDDsCII f.KL9;H$LLL$LIL$L$HILL$L\$H$LL$L\$HILL\$L$L\$HIHH E1LjHH=A5CPj57DAWj5CARHT$XLT$`L\$PVLHPL$LT$HHI+I*I/I.H=#IHI,$tMHEMHHEkDHTHLDE1LMDL׉t$$t$$DLL\$(t$ T$LT$LL$L$L\$(t$ T$LT$LL$L$DDL߉t$ T$LT$LL$L$Bt$ T$LT$LL$L$Lljt$T$LT$L $t$T$LT$L $Lωt$T$L$t$T$L$LL\$(t$ T$LT$LL$L$L\$(t$ T$LT$LL$L$UDL߉t$ 1ۉT$LT$LL$L$`L$T$E1LL$LT$t$ fIIHJcHHF0HD$HHE(HD$@HE LHD$8HEHD$0 HItsI~5ItHJHD$HLt$0L|$8H\$@HD$@M:I6H5.ALHV HD$8H9HH5Y@LHV HD$@HHsfD1E11E1E1E1`.?fDE1E1E1f.1E1E1E1E1a=?fDLL\$LL$ L\$LL$61E1E1E1bL?뉐EFEFAMcIH=IrfEFAFII I@EOEOAMcILD$HLD$I`EOAGII I@DsDsAMcItLL$LD$EHZLL$LD$IFfDDsCII IDH5y;LHV HtHD$HHHHT$0LLLH5pf>AfDH=>HR1L\$H5N1ɐL\$I@HFLHD$0HMκ1I$MdHEIlIFL\$HH H=LD$LD$L\$NL\$ 1LLLD$HD$LT$LD$L\$ MI(tbI.t|HI9CL\$LT$w LT$L\$HIE1ɾV@sI*tVE1E1LL\$LT$ L\$LT$fLL\$LT$ L\$LT$cLL\$t$T$L $ L $T$E1ҋt$L\$E1Lۺ1MTIEMlHCHH+ H=|HL$HL$1LHHD$LD$MI. H+IH;=\H5H9LD$4LD$LLLD$L4LD$II(IPM5I/L;L;@L;3LL\$6L\$I+eID$H5@LHHIMHDH .H9HH-H&HL5-MIFH51=LHH&IMI.IH;=]MwMPMGIII/ LLLL\$ LD$jI.LD$HD$L\$ ! I+GH|$I( HD$H;I,$ HEH5Q?HHHIMHqCH ,H9HIH,H/HL5,M IFH5;LHHIIHMIH IH;=EMgM8MOI$II/X LLLLD$(LL$ I,$LL$ HD$LD$(_I(H|$ I)> HD$H;bxHm1 IEH5>LHHIMH&BH _+H9HHF+HHL52+MIFH5:LHHIMI. IH;=h IoH MGHEII/ LLHL\$(LD$ ނHmLD$ L\$(I+I+ MI(k L;5*ImF H$HH="LHp ILL 7j57AVj57t$HjLD$PHT$HL\$@?H@L$HHI+X HMHl$Ld$IHE1E1~AIWBHZ MwH=ʻLD$`LD$LD$LLHD$L\$LD$MI(LL\$"L\$HLD$LD$ fLLD$LD$fLL\$L\$=fH=a?L=5HGLHHIMH^LH5H81bf1E1E1s?fH=5Hz(H5{(ևI5fDH5)H9L\$nL\$LLL\$q~L\$MHD$LLVL<L"LL$lL$1E1ۺs?$@IIWBHHZ U MwH=bL\$L\$ L\$LLHHD$qHL\$ IMHILLD$LD$f1LLLD$LD$L\$HI0MME1s2@H=93H&H5&ILIfD1E1ۺs@LLD$LD$If.H=<H2HGHHH IMOHH1H5"H81E1ۺs@VfDI0HLL$1E1NL$E1E1E1s@fHLL$LD$LD$LL$I.@1LH{IHME1Iۺsa@H5H9'LD$=LD$LLLD$ @{LD$ MHD$fDLL\${L\$E1E1ɺs@ H=;L5:1L\$HGLHH L\$IM8H LH5L\$L$H81L$E1E1L\$s@fE1Һs@mfDL\$LD$IIWBHZ * MgH=ʵLD$`LD$ LD$ LLHHD$HLD$ > IMHIxLLL$ LL$ ULLIYH[MqHII)VL\$L\$HIGIXLD$LD$HuHH5H8pLD$I($1ۺss@E1E1LL\$LD$.LD$L\$L57L=(IFHH H=?1LLI`MLoI/1ۺk?E1E1E1sE1MKMXI[IHI+0LT$LL$LL$LT$HIMNMMs2@E1E1E1E1m?LL\$ LD$L\$ LD$H5H9L\$ 5L\$ LLL\$ 8wL\$ MILLD$sLD$Ld$E1E1Mw@ IH=:'nILLT$LL$+LL$LT$L=0Mm HpI9kHXH HJH~1@L;|AHH9uH}HV1E1IOH5E1H81Ld$L\$w@cIHl$Ld$1E1E1x@2LLD$MLD$VHHAH5 jL $H H8H1_>AXLD$sLD$HuHjH5KH8LD$I([Ld$1ۺwE1۾@E1H=%HH5NxI\Hl$Ld$1E1ۺx@@IE1H=<%lIHl$Ld$1ۺx@IL1LIH71E1E1k?1E1E1MǺv@L*IL=$.M IvI9:HXH HJH~1L;|HH9uHHVH51IOE1H81OMHl$Ld$x@E1MHl$Ld$E1E1yAyL\$ HuHpH5QH8L\$ I+Hl$Ld$1E1ۺx@E1L\$Is1Ld$1E1E1MϺw@IL1LIH1E1E1t@1ۺfn?E1E1E1H@`HeHHULIHDH H@HL$ H9fpMFI@HHHcHEFI.LLD$LD$I.rLjE1bEFAMcI.EFAFII IEFAFII LH5([IHH@3LIH@`HHHLL$HLD$LD$LL$HIH UH@HL$ H9MwIFHHjHcHEwI/LLL$LD$ILL$LD$\I/LLL$E1LD$LD$LL$EwAMcI/EwAGII IEwAGII iLH5JYLD$LL$HIH@LLL$LD$nLD$LL$Iz1ۺgx?E1E1E1:1ۺh?E1E1E1 H@`HHHLD$LLD$HIH H@HL$ H9MOIAHHHcHEOI/LLL$LD$LL$LD$TI/LLD${LD$E1EOAMcI/EOAGII IEOAGII sLH51XLD$HIH@LLD$LD$IHD$M1ۺk?E1E1E1@IHH5H51H8E1ۺk?bHl$Ld$1E1MǺx@AHD$M0IHHȑH51H8/E1ۺt@1ۺt@E1E1E1ME1uHyH51E1H8L\$v@HHI9HuL;=H H5C1E1H8Ld$L\$w@CHHI9HuL;=Du1E1ss@LD$LD$HHƐH5gH8LD$LL$LD$fLD$LL$HHH5%H8LD$LL$)HHQH5H8H@`HHHuLL$LLD$LD$LL$HHPHD$ H9CHCLsIFHH*HcHDsAMcH+HLL$LD$LL$LD$DsCII IDsCII DsHH5TLD$LL$HHJI/LLL$LD$bLD$LL$H@`HHHLHHyHD$ H9GHGLGI@HH HcHDGAMcH/wLD$LD$cDGGII IDGGII DGH5SHHdI.LhH|$9H|$IsH|$gH|$I\HLL$LD$LD$LL$IHLL$LD$gLL$LD$IH@`HHHLD$LLD$HIHD$ I9FIF>MNIAHHHcHENAMcI.LLL$LD$MLL$LD$ENAFII IENAFII ENLH5RLD$HIMI/LLD$LD$HHI9cHuL;=QDH)H5L1H8MHl$Ld$E1x@ILLD$DLD$ILLD$eLD$II۾V@s7LD$LD$H HċH5eH8LD$LL$LD$dLD$LL$HSHH5#H8LD$LL$.'H$HOH5H8 E1E1E1E11ۺs@,E1MIغs'@ff.AWAVAUATUSHHL-LfH|$HDŽ$L$HII+HF H$HkHD$hHHD$pHD$xHDŽ$HDŽ$HDŽ$HD$ HHb  H(hE1ɹAHƺHHD$hIHHD$pHH|$hH/ HLd$pHD$hHD$pH w H9HH^ H@HL=J L|$hMIGH5LHHHHl$xH|$hHH/H]H HD$hH9HH H&HH- H`HEH5HHHmIMyHm L5M9p"LD$XE1E1LD$HH H|$hE1HD$AJAHt H/uH|$pHt H/1H|$xHt H/Mt I)Mt I(Mt I*pH GDDH=İgMtI,$'E1Ht HmMt I/HT$HHH$HHHfDHIInM MHH HOHL nLOL@HH<ATHH5TH81zXIZH J)H=ȯE1fHĨL[]A\A]A^A_L,$DfDH|$ HD$pHH1KHD$xIHAHD$xHD$pHD$pIGHH H9H4HHHLLD$hM]I@H5k LHH2HHl$pH|$hHH/ HH5`HD$hH9ptH>HHL*LD$hMI@H5iLHHIH|$hM_H/ HD$hHD$pL9pwE1L$L$HIHD$hHt IAHD$hIO|MTHl$pHELMhH=L $L $L $1LHAHHL $Hl$xI){H|$pH/bHD$pHD$xHHl$xHD$HEuHD$xH} HD$HH5^HD$(HD$H4$LIUHHT$0H4$HI5H@HT$0HHcLLH$IHHD$H5LH4$IUHHT$0sH4$HI3HxHT$0HHLLHD$pIHHxL9; MhM. I@IEHH|$pHD$pH/ Lt$pH5I~H9 z LLXILt$xImLl$xMjH|$pH/HD$pH|$xH/HD$xH|$(HD$@HL$ HT$L|$HE1Hl$PHHLd$XHHD$ HHIHHt$8HHD$0@HHD$fE1LhHBL$}$CIXL9u =^HHD$0J LfHfY@H9uHD$8H9tHL$LHYILd$ L9t$(BL|$HHl$PLd$XH|$@L$MtwIEL5HHH=1LLH$ L $MAH$H/HDŽ$I)HD$HI,$ HmLd$HLbfDLLL$L$\LLD$L$LD$L$+fLD$LT$L $LD$LT$L $f.LD$LT$L $]LD$LT$L $f.LD$LT$L $-H|$pLD$LT$L $eD1LIfDHLD$LD$Lú1I$LdH}HHDHCLMUH= 1HHAIM L|$hHmhH+nHl$xH}L9H5mL|$hH9LHTHH\$pH|$hH/(HD$hL|$pMH|$xH/HD$xH|$pH;= H;=~L9Å H|$pH/HD$p I\$L9,$aH$H$H$tHH5H9pq H}H HH=iH|$pHE HGH5 HH HD$hLD$pHi I("HD$pHl$hH}L9>H5~H9G):H4$H5SDH|$pHD$xHtH/HD$xHD$pH|$hIHI(HD$hH|$ HD$hHHD$pIH~D$xH$HD$xHD$pD$hHD$h@Ht H/H$HDŽ$Ht H/H$HDŽ$Ht H/uHDŽ$f.HUB7HZ HmH=5u*LHHH HH1mHHfDfD%fDH|$hAIE11AE1E1E1HD$E1H|$pE1E1E1AIAHD$1E11HHHD$hIHH!|LH5H81%HD$h@H|$pE1E1AIAH=HBH5CYIH~H5 HIHVI4H$HMH$H$H$lHUBLj HmH=,HH4$AHH H|$pHl$xHLD$hH}yH5^H8;H"E1E1E1E1HD$AIAH=nL5HGLHH IMJL|$h1`d!H=HHH5XHH=L5HGLHH HHHyLH5H81H|$hE1E1E1E1AIHD$AILIH|$hE1E11AIHD$A1(H5yH9L$L$I@L5LMH=L$L$1LLAIMLl$xL$L$,MHMIXIHI(( LL$LL$HHLML4IPBLj x MpH=Gu'1LAIM9H HD$xAKH$H/ HDŽ$H|$hE1E1E1A"IVBwHR  MvH=H$u.LH$LI/MDIH$ E1+xLkx1E1E1A#JAL}MHEIHH|$xHD$xH/4HT$hH|$xLNMHD$pI/LH|$pIIE1HD$hAJAdI;H HD$hH|$pIIAJA'DA&JAH|$hE1E1E1HD$E11o^H)$HH$LHLH5qLC}If.HFHH$IQIL$H-t L-HEHH H= 1HLHH Hl$pHAH|$pH/HD$pA5JAG5ILD$pw/H\$ETH5HHVHH$IF1LHL $DL $HHD$xHHD$H|$hE1E1A8KAf.Lx1H=\HH5RIE1E1E1AKHD$AjHH|$pE1AKAH=L-HGLHHIMyLD$hMH=dHH5qQHH=HH5OQIL5JHD$hMt I(nHD$pH|$xHt H/^H=HD$xH QHT$xHt$hH|$p9dH$H FrH9HHH|$ IHHD$#LT$HIH,$LPHHD$HjLL$HI>HEHD$HHEI)H|$pHt H/HD$pH|$hHt H/UHD$hH|$xHt H/BH$H$HD$xH$0L3H|$hJHtH/LD$p;LD$p,H=bL-3HGLHHHHdH|$pQE1E11A KHD$A~IdH|$pE1A KAH=L-HGLHHIM;LD$h1LLRH$IH/u L $L $HDŽ$MPH|$hE1E1ALAHEHD$pHHUHHH|$hHT$hH/9Ht$pHl$hHH$HoFHPHT$hHwH@HHH|$pHD$pH/AOLnH<$LL$LL$LH|$hE11A-KHD$ALLL$LL$LD$pJJH|$hJH|$hA8KE11HD$xAE1HD$gH$AoKAHnH8H|$hE1E1HDŽ$L $(]L $HHD$xfHnAqKH8HD$pH$AJE1E1AH$LT$E11H$H$LL$L$H|$hHD$LL$LT$r1HuHlH5~H8 H$H/uHDŽ$AL^H|$hE1E1E1E1A[JHD$AH|$hE1E1E11A]JHD$AjZHIHFE1H$AJAHuHkH5}H8HD$xH|$pH}H|$hJ+E1E1:IOE1AJA;1oAJA#H1LHD$pHH H|$hE1E1E1E1A1JHD$AD賿H8E1AJAL$A#L$HjH5m|H8>Lc'bHHD$pCHD$pA1JA2HkLL$H5!A KAH81蓾H|$pL$E1HD$hľIjHMkLL$H5ϞAKAH81AH|$pL$E1HD$hrI,HjL꽇JH5|H81H|$hHD$p2HSAKH@iH5!{H8L $1LLIH iH5zAKH8oHD$xDHhH5zH8KH}AqKH|$hE1E1AoKARE1I*H$HhH5dzH8AWAVAUATUHSHHiLnH|$HD$pHD$xH$H IIIH ZHKAHMEIHHhHڡH5*AUL CH81GXHFZH H=ŐE1GHĘL[]A\A]A^A_HF(H$oMH})L$pHG HGH$HH1 HHcHGWHH H$HiL|$xHD$p HD$HH$HD$PL(HD$XHD$`HD$hhE1ɹ1HALAIH HD$HH8HD$HAt$ HI|$ I\$H5LIƺH\$ MtIvHf/ H;-NgHT$pHt$hH|$`茿H5H nH9H\HUHdHH=AH|$HH{HGH5HHHD$XH\$HHH+HD$HL|$XIH;=`fH5fH9=HLJ;f.H|$HHD$PHHxLd$HnfDGWHH HH$H<$ĹHeH$fDGHH$GH$|@H E11A1AFE1fLD$HMt I(E1H|$PHt H/Mt I)H|$XHt H/Ht H+H jDDL$H=CML$tI,$E1Ht HmMt I(Mt ImHHmHIWB4Lj 7 MH="u轷]HLAIDMH|$HLl$PHt%@H/ HD$PHD$HH'H|$XH/ HD$XLHD$XH蟼IH~D$PH|$`HD$PHD$HD$XHD$X@Ht H/ HD$`H|$hHt H/x HD$hH|$pHt H/k HD$pOLXHD$HIHGHD$PIHDHD$PHD$HHD$HIEH*H SH9HH:HHH=&H|$PH@HGH5 HHUILL$XMH|$PH/KHD$P0HD$PHIEHD$PLh/HD$HIHH?bH5Hؼ L|$XHT$HH\$PIGHH=H=rHT$<HT$HLH迷H&H|$XH/ HD$XH|$PH/o HD$PH|$HH/G HEuHD$HH} HEHD$(HfH=TH*$HHD$Mt HHIHD$HD$L=5HLHSHHT$ºIHH@HT$HH LHHD$pIHHD$HLHIWHHT$]HHaH@HT$HH LHD$HHH?H@HD$PH;&` HGHD$PHv HWHHH|$HHT$HH/^H\$HLL$PH{MD H5p`H9HSBHJIϨ H[H=fpLL$ LL$LHAH~HR H|$PHtH/!HD$PH|$HH/i HD$HH+f 1H|$HD$0~{HD$H\$(Ld$8E1Ld$ Hl$(LH`H H,$L,$HD$JILHD$H|$LIMLHILl$L;|$uHl$(L,$Ld$8H|$0?Lt$pMtuIFL=2HHH=3oαF1LLH$SL $MH|$pH/HD$pI)uL萶HEI,$P HEIHHE@E1HL$TL$ L@ L0 LL $L $E1GLD$L $LD$L $?@LL$L$5L$ϵL$4fDHL$贵L$(1LL$I藵L$?iH$,IIMIHFHHD$p IH5'LHVôHD$xHIH|$pM-H5LHV萴HH$IMH|$pIf HF(HH$HF HD$xHFHD$p`I蓴fDHIGH9tIL9DHEH$HHH΄HcHDEH$HmH6HmHH$EHH$HmEUHH HH$f.EUHH H$qHH56 HHVH@HnH$1LHHD$HIHE1AAGfDHD$PE111AAFJLAE1AFnL $H-H=vI1E11E1AAF)H'E1AHD$HAGHHQFH r{H5k]jL AHH81{Y^,F/ZКL$H HD$HAAGv$蚚HuHEH5wWH8HD$HE1E1AAG,E11A1A#GE1HFLH5;zAAWHH81譙H\$HHD$PHHSBLz 7LsH=VBLLAIɛMLl$PLl$XMH|$HHFLH5yAGH81H|$PCCH%9E1AA2G IH\$HAA4G藝u H;H5DH5wH8n LHTnJ@7E11A1A6GE1@LHCH5kUH8iH@`HHH HHHL9`HGHGH$HHH HcHGHH$H/ee[GWHH HH$֋GWHH H$ËGH$H5z2HH`HmHH|$ԖH|$H$rH|$\H|$H$ZIHGBH5(TH8谜H1!HE1A AEGHuHAH5SH8eHD$PLl$XMH\$HA ASGi蓖I5E1HAH5SH8 L$E11A 1AXGE1 E1A AVG"A ASG?HHgAH5uH8蠛1MAAG:H ^vAAH1A"AHuIzAAGff.AWfAVAUATUHSHXHALnH|$HD$@HD$H)D$0H?IIHF0HD$H](L} LuH H(hE111HALIHH8RH[ H(hE111HALHHH8H L(hE111HAHAH$HR H8EA;D$HH H9HRHHHL-MVIEH5[LHHHHIm'H@H )H9HHHHL-MSIEH58LHHeIIEHM7IEHH{?I9F ?IH[E1 A6I.Ht H+tMtIm@Mt I/H psDH=PgMtI,$ E1Ht HmH<$H $HHD$HHH$LГHH5LHHV膖HD$0HLm@IHrH rHMHAUHLwL@Hd=H5TL vH81褑X5ZH tr H=RfE1HXL[]A\A]A^A_HD$fDLHL89H(HH $;ALSf.f(HL$ 3f.ӝL$ D$!LL$ f.L$ f(f/Af/T$-f.L$HD$f(T$ LIrIHT$ f(WHH#D$@IHHt$HMLH65<AH=<jH PAUjPSjPHT$`HD$XHPHI. I/ H+ Imm I,$t?HmLd$9fDHpH`cHD$LGLd$DH߉T$DD$'T$DD$mfLT$DD$DD$T$YLT$DD$ߓT$DD$HLT$DD$迓T$DD$IIH4wJcHHF0HD$HHE(HD$@HE LHD$8HEHD$0$HItsI~5IHHD$HLt$0L|$8H\$@HD$^@M IH5LHV芒HD$8HjHH5LHVeHD$@HHsfD1E1 A5H$H$E1 A5DM1I$MdHEIlI@LMvH=JLD$%LD$ 1LLAI襏MLD$g I/-I(H{H;=p9 H5:H9t QyHSBjLz n LsH=I莌FLLAHD$LT$MImbH+L;8L;w8TL;9GLLT$*LT${I*)9 HHH9X HH1 HH~H HCH5zHHH IMq H+H`H)H9X HH&HHH HCH5XHHH]IHHM HHlH7I9Ga Ab7HH>M1HEHlH$HHDIALMa H=GLL$蔊LL$1LHAIMLL$$H+<I)RI}H;=6]H5r7H9輌LLL II.IM4ImL;;6L;6L;6LLT$̈LT$XI*HtH-H9X@HHHL-M>IEH5LHHHHIm( HH H9HHHjHL=}MIGH5LHH+IM I/ HG5I9Ey IH E1ɺ A7Im7LT$MDD$LL$mLL$DD$T$MLHI IImLt-ME1E1 A7[fDE1DLLT$LT$@L،LLD$ÌLD$fE1 A5DHLT$苌LT$!H5qLHVHtHD$HHHHT$0LLLlH5^h5fDIVH*MFHII.w LD$ HT$IHT$LD$ HIIIWRHFLHD$0oHL A7E1MI*L׉T$DD$nT$DD$IUBHZ r M}H=C%LLHD$諈LT$MOI.LLT$LT$@L1LLLD$>LD$HIMƺ A7f.H舊H=iH:H5;&IfDE1 A6DH=L5"HGLHH6IM_H1LH5eE1H81 A6GE1 A6DH(H訉H=HJH5KFI:fDHuH/H5AH8XIm5 A7E1E1tL LaLGHKE1 A6DE1ɺ A6,DIH=iL5HGLHHIM:Hf0LH5cE1H81g A6E1HLL$#LL$fLLLT$LT$fL؇ D$?L$H A6E1DL$ L$ H A6E1DMƺ A7E1f.LsMQL{IIH+ LLLOI.ILLT$LT$OL$ D$(QL$ T$(H) A&6E1fLD$LD$ MA7HH,H5>T$DD$H8[DD$T$DL5L=IFHHN H=o>  1LLH萃H H/H+ A>6E1:fL5qL=rIFHH H==蒀 1LLHH HH+ A)7E1f.L5L=IFHH H=w= 1LLH蘂H H7H+c A^6E1Bf.YL5kL=\IFHHM H=<6 1LLHH. HH+H A~6E11LHLL$ƒLL$HIM IAm71fLH=HH5 HfDE1 A;7-DI H=qL5HGLHH HHHn+LH5^E1H81o~ A;7E11ۺ A6nE1 A=7bDLEE1 A6I/ E1#H=ɵHjH5kHZ9~HuH5)H5;H8螃I.  A7E1E1 A@7 A6dI޺ 1AB7H= L5+HGLHH^ HHH*LH5]E1H81} A@7<;}IE1M1I$MdH$HIDIGLMH=9z|1LLAI~MI.I/H{H;=(H5b)H9~LHIIImLDM6 H+L;+(L; (6L;()LLT$zLT$I*sHD$HH=c(MLHp IL LjAQt$jAQUjHT$PLT$XHD$HH@HLT$I*L'fHSBLz sLsH=J8z LLAHD$j}LT$MImLLT$LT$f.L9L"MwMfMOIII/KLL$LL$HH LsH A6HLT$ LT$L~QM}MI]IHImLLHI/ILLT$~LT${M IAm71LLD$ HT$t~HT$LD$ h A7E1LL$yLL$H H$H56MI1H8LL$ Am7vL5L=IFHHH=.6x@1LLHO{H8HH+ A7E11LL}IH A7nHX}6H=<H͢H5΢IE1 A7HE1} A>6kHH#H XH5:jL \AHi]H81wY^}51H=2I2xHQE1 A7HE1| A^6HE1d| A)7E1 A7H=*HH5IHE1| A~6lwHuH{"H5\4H8|Im A7E1E1 A7wIH=IE1E1HH/"AH5J9jL c[H :WH8H[1Yv_w5AXMMMzM}IIIm;LL$i{LL$HIHMN$@vI1LLKzHHE1 A:6@ A7E1LsMGL{IIH+LLLI.IDLLT$JzLT$-1LLyHH(E1 AZ6|@LLL$zLL$1LLlyHHaE1 A%75D A7E1#0u A7HH H51T$DD$H8}zDD$T$utI1LLxHHE1 Az6L5ɴL=IFHHH=O1s1LLHpvHHH+ A7E1LxE1 A8HE1x A7 A:6E1sE1 A:6HHH50H82y A:6E1Y AZ6E1o A%7E1\OsHabsE1 AZ6H3HLH5-0H8x AZ6 DsE1 A%7HHH5/H8nx A%7 Az6E1rE1 Az6HHH5/H8x Az6q1LL|vHH:E1 A7ELLL$vLL$!rHHv:HE1v A7 A7E1qHuHH5.E1H8[w A71LLuHH[E1 A7 A7E1tqHuHH5d.E1H8v A7AE1 A7E1E1 A7M A7 MϺ Ab7MIE1 A6wLT$E1DD$]uDD$T$MM@AWAVAUATUHSHHLnH<$HD$pHD$xH$HRIXILn(Lu LeH HD$pHD$HHD$PH(HD$XHD$`HD$hhE111HALHHH8H9D}H(hE111HALHHH8Eu sH5HwH5HͯeL;-uHH H9HHH'HL=MIGH5hLHHHD$HHI/H HD$HHL$H9HsIHD$P1M[E1Aa9 H|$HHtH/LD$PMt I( H|$XHt H/ H NDH=BMtI,$ E1Mt I/| Ht HmLHH+HarHoH5LIHVIqHD$pH LmIH MHMAHMEIHHHRH5/AUL QH81lX8ZH M H=AE1 HĘL[]A\A]A^A_IHq#Lof.xyD$*ID$V$Mt$IFH=HTHcHEt$f.H=D$SfH=I* L;-)HH5H9pHHHL=sL|$PMXIGH5rLHHIL|$HLD$PMxI(H HD$HHD$PHL$H9HHD$E1spIHPHD$PHt IGHD$PIEHOlHt$Ld$HHIDID$LMH=(jn1LLAHD$mLL$MLL$XI/H|$HH/zHD$XHD$HHH|$XH/SLd$HHHD$XHD$HAt$I|$ H5إHD$ID$Ht$ HD$0H$LIUHHT$(PpHt$ HI{H@HT$(HH LLIHZ%H$H5wLHt$ IQLL$8HHT$(oHt$ HIKHxHT$(LL$8HH@LLHD$XIH%HxH;|$MEMIEIHH|$XHD$XH/[Ll$XH5 I}H9WLD$NkLD$@LLLD$QLD$ILl$HI(zLl$HM!H|$XH/HD$XH|$HH/UHD$HE1iHD$H$Hp`H H|$H~]Hl$ Hl$0L<$IH\$(H\$Ld$MIfD$LLLLmJDIL9uL<$Hl$ H\$(Ld$H|$iIGL-LMH=$g1LLAIjMLl$pI/L|$pMI/=HD$pI<$uLGlHmDH(lHlLPMH@IHH|$HHD$HH/MLT$7lLT$HHD$PIi#MWIELd$HMlHEIDHID$LM H=#cf1LLAIhMH|$PH/HD$PH|$HH/Ld$HI<$HD$HE1At$I|$ LL$H֧HHLLHD$1ALL$HIH;}HD$PEMt I) IGH5КLHD$PHHHH|$PHL8HD$HHH|$PH/ HD$PH|$HH/ H$H5HD$HLHt$ MqLL$(LbkHt$ HHD$H@LL$(HHE H|$LLHD$HC!H$H5LHt$ IVHHT$(jHt$ HHH@HT$(HH LHD$PHH H@HD$XH;D$HGHD$XHHWHHH|$PHT$PH/n LD$PLt$XIxMH5H9t6LD$LfLD$u#LLXH|$XHD$HHukxIPBtLj M@H= LD$dcLD$LLAIeMH|$XLt$HHt!H/c HD$HHD$XHH|$PH/W HD$PH|$HH/5 HD$HE1udHD$H$Hp`H H|$H,$Ll$LH\$HLd$ MIfH0HLL0H@H0H8H0H6g1IMHE %pH(H0H0H@(;u}MHcHLH0H@H0Pt8H(HR8HcR H0;u|IM94IH\$H,$Ld$ H|$bHD$L5H@LMH=}a1H|$LAIdMkHL$Ll$pHH$HHH|$pHH/+HD$pI<$I/Lfwu;HP0H;0HHP0H0H0H0y:HG(H0J(H)0SH0LcJ[HmHD$HH|$XH`A:HL$ HH$HHHω$o_$A+9E1E1 H5rH91\$IEH5HH7H=bHL$ Ht$YJ1LHt$HL$ Ir\MLl$H:HH|$P^^L1L ^IHEE1Al9 Ly^'MIPBBHR  MpH=HT$,Y1LHT$I[MOfIUBHR ( MmH=:HT$Xu+1LHT$IX[M:YHHD$HA;I/ E1 ]IUBHR  MmH=LD$HT$ HD$HHD$XHSH|$PH/HD$PH|$HH/kHT$`Ht$hHD$HH|$pZH$D$LHP`Hx ZHVZHD$HIH HT$`Ht$hHD$HH|$p(WIGLd$`L%LMgH=H5A;H8IHD$HA;H H5H8rILD$A: Aa9 HT$XHt$PA";3EHD$HHD$PHD$X$HH@`HHHLIHL9pu5ID$5Mt$IFHH}+HcHH5&H衳IHuI/LwGEt$AMcI,$LTGHHI9&HuL;=(BHuHH5m!H8Hw{BHEHH5D!H8G*Et$fEt$AD$II IMEt$AD$II 7LvAI'H@`HtEHHt9LHHt,L9pu9H|$H|$IH/EFAHtIH5$1HHuHH5i H8Gff.@AWAVAUATUHSHXHHisH|$LnHD$0H\$8H\$@HD$HD$HH2 III H\$Le LmIEI$I9L5߁I~H;=ttzH5kBujHt$LHHH !E1H==pImLDI,$LDDIVBtHj MvH=?uHt$LH#BHHEH5yHHHL IMN L;=L;=uunI9tiL9>Aƅy^E1I/$LT$M DT$I HF0HD$HE(HD$fDI/REHH@H9og9HZgHyHL=FgMUIGH5qLHHIIHMIHIFH5kpLM{H=tHt$ >1Ht$ LAI@MI.* I/ HEH5=vHHHIMtI~H;=13M^M&MNIII. IyH;=t(H5LL$(L\$ ?L\$ LL$(IQBLr MyH=uLL$(L\$ =L\$ LL$(TLLAI?ML\$ LL$(KI+V I), Hm LH~H `eH9HHGeHHL=3eMIGH5wLHHIIHMIHLH>IHI.L;=VL;=4u I9$DI/nE]H}H wdH9HMH^dHHL5JdMiIFH5.vLHH[IIHMIHLH=IH&I/L;5}L;5[u I9 DI.EH|H cH9HzHucHHL=acMGIGH5euLHH}IIHMYIHLH9AƅI/E8HD$HE1LLLLH IAWHT$ {IXZMH\$INIEH+H-blHEImMI1HD$HE1LLLLH IAWHT$ z_AXIHI/H9\$IHEMHHENI.ImI,$HXL[]A\A]A^A_fDL7Aƅb E1LMI=?DH=L=5Lp=L`=UHD$HLE1LLLH IAWHT$ ]yY^IHM f.E1I. MH H= E1HHEHHEuH HL5-SMIFH5)eLHHIIHM IHLH,IH I/L;5L;5^~I9uL)ADž I.EHD$HE1LLLLH IAWHT$ jAZA[IHsML.o \H=iaH RH5 R&IfDEE1 +)IH=jL=aHGLHHIMDHLH5d H81(  I+]Mκ3@L-E1E1<HX-<H=9`HZPH5[PH&fD fL- K(H|$I fDH=iL=_HGLHHHHHLH5H81' M IfLX,*MYMIIIHI)LHLL\$HL$UL\$HL$II+LHL$+HL$3( $'I +pD- pLL$L\$'L\$LL$HHH5H8],L\$LL$I+LLL$E134+LL$M%L+L|$E1HgH [NH9HHBNHlHL=.NMIIGH5ZYLHHIIHMIH,LH'IHMI."L;=L;=oI9L+$AƅI/EHD$HE1LLLLH IAWHT$ :fAXAYIHHL$HHD$HH MMH=c\HLH5L IH=A\̣Ik$I{HXHHLHHIHHD$HHLLHLL$$H+LL$IHLL$(LL$|L(A $IL(>E1NLk(aLHL$L\$T(L\$HL$#IBDL|$^1L'MIHdH NKH9HQH5KH-HL5!KM IFH5eVLHHIIHMIHLH$IHI/9L;5L;5rYI9PL.!ADžoI.ESHD$LLE1LLH IQAWHT$ 8c^_IHMFH=YHJH5JWI!IH=kYI_!I\LJ&L.&HLL$&LL$L &E1N!H8DH^H5?H8&LL$vHbH HH9HHHHgHL=HMH52TLRIH)I/LH"IHI.L;=oL;=MAI98L AƅI/REHD$E1LLLLH IASAWHT$  aIXZMfqx iL$IrH=NWٞIH=:WHGH5GIiL'$ L$ E1HuH5VH8$ DL#gI2H=VIH=|VHFH5F9I4bH+`H FH9H~H{FHZHL5gFM7H5QLIH I.@LHA IHI/1L;5L;5I9LADž|I.eEHD$LLLLH IAPE1AWHT$ ^AYAZIH M=_ H=U袜IH=UHTEH5UEIn]H^H EH9HoHDHKHL=DM(H5YL~IHI/LHIHI.L;=L;=yI9wL5AƅRI/;EzHD$LLE1LLH IQAWHT$ ']^_IHDL E19L L R7TH|$LL$Mg LL$MXdE1b4L, E1M H=RTIH=RHBH5BrI{LH=XH9t`HEt H;uJHIMtwH=[LܜIHtQI.tAL膓I/t+HYILL뵺LE1DLNL `H=_QIH=KQH|AH5}AIAWAVAUATIUSHxH<$HHD$8HD$@HD$HHD$PHD$XHD$`HD$hDHzHKZI|$IH9t:HXHHqH#1DHH9 H;TuHH5TLH HH|$8H HGH5dKHHa!HD$@LT$8HV!I*eH|$@H-YHD$8H1H9H9H\$u H;=DH/iHD$@EI|$H3YH9,HH5,PLHHH|$@H%H5.JH9HGH;'Hu HD$HHD$8H/ HD$@H|$8H9H;|$g H;=Z AƅKH|$8D Aƅ H|$@#f.GpjHEHfHDHH9 HuH;HXH HqH 1DHH9 H9TuHH56KLH.IL|$HMf.H5HL9IGH;IHH-%HHD$HD$8I/'$HD$HH|$8H9H;|$H;=jAƅ/H|$8H/$HD$8E@ID$H5MLHH4HH|$8Hh4H5GH9A-HGH;n6Hu -HD$HHD$HH/M-HD$8H|$HH9H;|$l%H;=_%Aƅ 4H|$HH/-HD$HEjID$H5yPLHH;IL|$HM;IGH5HLHH;HD$8H[;H|$HH/o3HUH@HD$HH9XT9H@H;HL=@L|$HM:IGH5yKLHHv:IL|$PML:H|$HH/_4L|$8Lt$PHD$HI/54HD$8H|$PH/ 4HD$PHUHPM9&H9?:H?H9HL=?L|$PMN9IGH5AFLHH&9HD$8H^:H|$PH/3HD$PAHD$PIH$9HYEH5GH1HT$PH5CH|$8謁HD$HIH:H|$8H/C4HD$8H|$PH/7HD$PH|$HH/ 7HD$HHSH>H9X3H>H8HL=>L|$PM3IGH5MLHHO8HD$8HI<H|$PH/0ID$L=BHD$PHPpH0HRH0LLIL|$PMF9H|$8HT$PHD$@H˽H9G8H豓H|$@HD$HHt H/1HD$@H|$PH//L|$HLT$8HD$PM6I*/H$L5LHD$8L|$HHD$HLLIRLT$HHT$HH6H@HT$LT$HH0HLHH[?H$L5/LLLIPLD$HHT$HH#8H@HT$LD$HH.0LHD$8HHx?H@HD$PH;c/HGHD$PH/HWHHH|$8HT$8H/4Ht$PLT$8H/LH|$PHD$HHtH/m4HD$HHD$PH5H|$8H//HD$8H|$HH//HT$XHt$`IHD$HH|$h0MH$Hl$H$H HD$H|$LDIL9dID$H|H9~*H; -H@hHY3H@HL3LLHHT$HH3H5<L]0H|$HH/*HD$HID$H9+H;/H@hH7H@H7LLHHl$HH7ID$H9+H@hH32H@(H&2HLLЉŅc6H|$HH/<+HD$HID$H9*H@hH2H@(H2LLLЉÅW3IyH$H|$hHt H/p8HD$hH|$`Ht H/]8HD$`H|$XHt H/8H5I?1HHD$X.|HD$XH+U8H\$XH28H+8HD$XHHD$DHDHH9THuH;BfDHQNH 9H9XH8H(HL=8L|$PMLLÃH|$PH/bHD$P0HMH8H9X#Hj8H^$HL=V8L|$PM#IGH5?LHH$HD$8Hi$H|$PH/HD$PHD$PIHw$Hh>HHD$PHX>H5CHPI|$HGHHIL|$HM$IGH;F$IHD$HHD$@H$H|$HH/HD$@HL$PHD$HHxBHXHX @&݃#$HA HD$@HLt$PHBIF(IH4AG IGHHD$1HcLl$ 1Ld$(II܉(LRH9t1LHIIKTz LBMtLHJ  tHz0LRH@LELDH/DHD$8I|$HKEH9/fDkfDLT$8zL1E1AMI*Lt$@tMt I.H|$HHt H/Ht H+H|$PHt H/H DH=>I1MtI/t,Hx[]A\A]A^A_DL Lt$@ofDLH$ H$Hx[]A\A]A^A_ÐLx Kk ufD[ BfDHH >H-HL%z9HEHHPH=g 1HLH Hx Hl$@H"H|$@H/ LT$8LHD$@AE11MjHH5<LH?HH|$8HH9H;|$ H;= KÅ/ H|$8H/HD$8HHD$I9IT$ID$H53;LHD$HH\HHl$8H4HEH;u H;*H@hHH@H1HHD$@HH|$8H/ HD$8H|$@HG."HoHEH H8HcHH|$PLl$ L|$@Ld$(H/u HD$Pa HD$PIH HD$@IGH`FHHT$PHD$@HB E HD$@HH H]8H5: 9Hl$8L|$@Lt$PHEHH{"H=Z%LLHHH%H\$HH|$8H/HD$8H|$PH/HD$PH|$@H/]HD$@H|$HH/;HD$HH$H@HHLuL0 IH, H@HHLLHIH0H$H^@HHLuL HHH@HHLHHD$@HH/H@HD$PH;HGHD$PHHWHHH|$@HT$@H/Lt$@Hl$PI~HH5H97HLDH|$PHD$HHtH/%HD$HHD$PHH|$@H/HD$@H|$HH/HT$hHt$`IHD$HH|$XLMiL4$HȮI LLiHID$H9a H;AH@hH5H@H(HLHD$HHID$H9 H;H@hHH@HLLHD$@H ID$HT$HH9H@hH]H@(HPLLЅ[H|$HH/>ID$HT$@HD$HH9H@hH H@(HHLЉŅlH|$@H/IHD$@IH|$XHt H/X'HD$XH|$`Ht H/E'HD$`H|$hHt H/+H531LHD$hpHD$hI/+L|$hM-I/+HHOoHcH H|$@H/ID$H5<LHD$@HH HH|$@H HGH5:HH4 HD$8Lt$@H I.HD$@H|$8HGLgID$HY HRHcHoGHH HoGHH oDgGII II H|$8H/ HAH,HD$8H9X] H,H HH=|,H|$8H HGH5{;HH HD$@LT$8H I*HD$8LHD$8H HD$HIH[ HD$8HD$8IGHD$8H H@H+H9X'H+HHH+H HCH5>9HHHIL|$PMH+HT$PH5r:H|$8AH|$PH/Lt$@HT$8HD$PL|$HIFHH H=vHT$ HT$LLIMLt$PH|$@H/HD$@H|$HH/HD$HH|$8H/HD$8L|$PHD$PL;|$H$H5M9MwLHt$IRLT$(HHT$ Ht$HH.H@HT$ LT$(HH HLHH)H$H58LHt$IRLT$(HHT$ _Ht$HHH@HT$ LT$(HH LHD$8HH(H@HD$HH; HGHD$HH HWHHH|$8HT$8H/ LT$8L\$HIzM H5hH9LT$ L\$L\$LT$ LL}H|$HHD$PHtH/V HD$PHD$HHH|$8H/ HD$8H|$PH/ HT$hHt$`HD$PH|$XIMIH~H$L|$ ILH\$Ll$HH H$H<$HHL$LLHL<L1LLL#LLLI)HuH\$L|$ HD$H8HD$PH HD$PH|$XHt H/ HD$XH|$`Ht H/[ HD$`H|$hHt H/H HCL%,HD$hHHH=T1LHHHHl$hH+ H\$hH0H+ HD$hHD$H,DgDgGII DgAMc!H&LT$8L1E1AJIYHiH1۽OA-H8yLT$8H;hLHD$8Hp(Lt$@MA E11fDHEHHH|$8HD$@H/}sfDLHfDILHD$IH<0LHu@HEHvLpFHUYHLt$@LAIRB HJHL$( MRH=WLT$ L\$L\$LT$  HD$(LLHD$_HD$H{ H|$HHD$PHHLt$@LAgHL=8MIT$I9UHXHHqH~1L;|.HH9uIOH٢1E1HRH5aLAH81 LT$8fI>LAD91IH6HH(I.LHD$HD$ffA.GtnH-H{HEHH\$eLT$8L1E1ALT$8M1E1AHuHH5UH8hHH|$@H)HLT$8LALT$8LAID$HHHD$H{LAFHHt$HT$HT$Ht$>I/LT$8OHD$@A&VID$JHHD$@=IVBHZ 7MvH=HLH(HH|$PH\$HHLt$@MABH=K,H,!H5-!HHuHH5H8 HH|$8IHʠH5{H8ID$JH9cVIFH-0HHH=.u)1HLHSH+5HHD$HH|$PHOI/ LT$8A-1E1&lH/I|$QG=jH=,mIHLT$8WO1E1A#LH$H$HHLH$L$I(Llj$$oIVBHZ InH=Ԫo1HHeFLT$8O1E1A+HuHH5H8&HD$PH|$HH?MH+8LT$8A1HH$H$HeHLH$L$I(LHH/ HLH$L$I(LH$oH$aLOA&ILT$8HD$@|OH|$PHt H/HD$PH|$8Ht H/HD$8H|$@Ht H/bHD$@H|$HHt H/O0H=HD$HH wHT$PHt$HH|$@_ HL$PHT$H1Ht$@HD$8HH IGHHY H=+1HLHH I/HH|$8H//HD$8H( H;-&H;-H;-HHmy5'PHT$hHt$`1E1H|$XA-7LT$82PH|$@Ht H/ HD$@H|$HHt H/ HD$HH|$PHt H/Q HT$hHt$`HD$PH|$XHĖH'HmoH)PLH@HLH$>L$I("LH$H$ d|rhcHuHڔH5H8CLT$8M1E1HD$PAH/MAH8LT$8Hm*HIwH9OHXHHJH~1H;\)HH9uHaHK$M1HVH5AH81LT$8.fP5O+Lt$@LA'HH=cHH5rI:LT$81۽cOA%f.GHEHH1MH8HD$8LT$81۽MA WI^H=fIHuHH5ϤH8WLT$8LHD$@ALt$@eOA%jE1LT$8L1E1ALT$81۽pOA&J1LwULT$8M1E1A "LT$8LHD$@ALT$81۽xOA&LT$8zOA&iH;Ē- LHA HD$@HLt$PHIF(mIHo@HA1LT$8cHH޽OH8HD$@E1HH5%1۽LAH8^LT$8 HT$PHt$H/P@HD$@HD$HHD$PzLT$81۽OA%LT$8OA+:LT$8MA1LHRHHHI9HuL;=tE1(HHHH5H8q+LA-1E1YLT$8JID$JHHT$H1H@`H HHIHH@L5L9 IoHEHHHcHAoI/LI/H|$@L1AoHcI/dAoAGHH HAoAGHH LH5GTIH>H@:LHLT$8L1E1ALLHmHD$HIHLT$81۽OA%zDID$J,HEHl$HID$JHHLI.LHD$@O9H$Hl$ANAVL(HHR HLHmHHHD$ HT$ {KLIHLHLI.DL7eH=HSH5TOeIHhLNAH8zLT$81۽sNAI?NH+t,LT$8A1H$Hl$ANA  HA1LT$8H;H5}L=#LT$81۽1NA\H(H5 H8LT$81۽>NA*H=TXIgLT$81۽NA IH=\ XIH=H HH5cILT$81۽NA H=ZXILt$@NA xLT$81۽NA QI&1LuLt$@3NAKHGHD$@H HWHHH|$8HT$8H/3Ht$@HT$PH|$8H\[LT$81۽dNAHLNH8HD$8HIIID$H;oXH;M:LMt$M;t$KDHHD$P LT$81۽INAH$Hl$ANARH@`HHHLHHL9puxHCHkHEHH:HcHkHcH+\HOkCHH H׋kCHH ȋkH5HGHHmI/LsHFHH[HtH$Hl$ANA=LHHLHD$ LD$ HI(LHHH9THuH;ЁBHT$PHt$8OHD$HHD$8HD$PHtLz2pLt$@aNA6LNHʀH5$MAH8(LT$8H|$8OzO61HLAHLT$84OAOHkHH&H5dzH8_H$Hl$HD$HANA= OLT$81۽IPA-LT$8/MALT$81۽OA-1MILIHtAHLRI.IJL=HPHUH5H81LT$8dN1E1HD$PALMy`LMt$M;t$iID$JHHHu9ID$H;~t%IH;acLIHH0"NIGH5zHPH~H81$H;Y~tCH;~t3HzIItnIuXEwAWII 511 LHHHH+I_H,RL_IBEwAGII I+H@`HIHH9LHH(L9pu~HCLcID$HHHcHDcAMcH+H|DcCII IDcCII DcH5ʹHOBHHgI/ L!HIHVIqLT$8NAH$Hl$HD$HANAH|L|$HH-|HD$Ӻ%HiHM|H5H8NH|$@2N8IHfLHxITH}H8c3ff.AWfAVAUATUHSHH|LnH|$H$H!)D$pHD$0H$HHD$8H$HJIIIJHE(HD$Lu L}II#IHGHpLL$HI@HZHH=1DL$HH I(WHEL%$HHLH H$H<$ H$H8gHmLHHH9X HH<HHHxHCH5- HHHIMH+IzH;=zH5r{H9tLT$LT$&IRBHj 3 MbH=XLT$LT$LLIqMLT$I*I/HHH9X'HxHwHHdHHCH5 HHH IMH+IxH;=yH5]zH9txLD$LD$ueLLLD$NLD$IMMGMIHF8HD$8HE0HD$09LIPBtHj qMhH=LD$|LD$LLIMLD$I(I.HD$H; y[H@HH#Ht$H\$HH@H0ID$H53LHHpHHH*IHHm IXIEH5LHHBHHDHHD$@HHmH|$@H- L5HEHH/H=ux1LHHH5H5DH+*M]CAoHD$1E11HD$(E1E1HD$ HD$HawHD$HHpHmH H9&HXHHqH1HH9H;TuLLLT$KLT$ILM5II*LI/LHH9vHuH;udfDHD$H{f.H=1 L%RHGLHHHHH.vLH5H812fM1E1E1HD$E1yBAcHD$(HD$ HD$HD$DHLT$LT$WfHLD$LD$LfL6LxIIHбJcHHF8H$HE0H$HE(H$HE LHD$xHEHD$pIIHJcHDH AHHsHH5AUL ìH81X0BZH H=UE1SHĨL[]A\A]A^A_L8IH5LIHVHD$pHtLmIH !HAHMEOD@A@HD$M11HD$(E1E1E1HD$ cBA`HD$HD$H$Mt I)Mt I+Ht H*DH E1H=GQH<$tH<$HHD$0HHHL$HtHH$HH Ht H+ Mt Im HL$HtHH$HHtsH\$ HtHH$HHHT$(HtHH$HHtKHL$HtHH$HHtAI,$tII.LHfDHfDHxI,$uLiHX^KfDH8H(LLωt$@HT$8L\$0t$@HT$8L\$0&L߉t$8HT$0t$8HT$0H׉t$0t$0HHD$M11HD$(E1E1E1H$E1hBA`HD$ HD$HD$Mt I(HtHmtJMQI*GLLL$Ht$@HT$8L\$0LL$Ht$@HT$8L\$0@HLL$PLT$Ht$@HT$8L\$0LL$PLT$Ht$@HT$8L\$0yLLL$PLT$Ht$@HT$8L\$0LL$PLT$Ht$@HT$8L\$0-H$ZfH1pH8!M1E1E1HD$E11E1HD$(kBA`HD$ HD$HD$fE1H=HH5vMH fDHAIEH5LHH&HH%HEH;n H;nFH@hH!H@H!1HIM!HmfIELD$LH5CHH!LD$HH!HEH;mH;nH@hH#H@H#LD$HLD$IMl$HmLLǺLT$ LD$LD$LT$ HHD#I(I*H;-rmH;-PmH;-mHAƅ]$HmEID$H5LHHHHdHEH;alH;l>H@hHAH@H41HIMHmIELT$LH5HHKLT$HHPHEH;kH;hl"H@hHH@HLT$1HLT$IMHmLL׺LD$ LT$LT$LD$ HHI*I(H;-kH;-k H;-1l H[D$HsHmL$HH|$HWHBpHrH@HeH5fHH(HhHHHmYID$H5LHH8IM7I@H;Yj#H;jNH@hHkH@H^LD$ L1LD$ HD$H|$?I(HS HCHt$HHH9kH9bHHSH4HHCHt$HHD$ HH=H|$H5HGHHZ IM IzH;=LjMrMIjIHEI*HLH@I.IM2Hm;I@LD$LH50HHM!LD$IML!I(ID$LT$LH5WHHD"LT$IM!I@H;hH;,iH@hH"H@Hu"LT$L1LD$LD$LT$IM#I(DH iI9B!MBMIjIHEI*LL$ LD$LD$LL$ HHD$ML@HHHHDLLHELMzH=9yԻ.#Ht$1HAHD$ULL$M#Ht$HHD$HHHmIELL$LH5HH#LL$IMl"H%H5H9p#HH4$HLM#ICLL$(LLT$ H5HL\$HQ$L\$LT$ HD$LL$(H|$#I+IzH;=Dg!MBMIjIHEI*HT$LHLL$ LD$=LD$LL$ II(OHt$HHD$HHM4HmImH$HHxH;=f2%H50gH9LL$uLL$L,$LLL$Lt;LL$MHH6I*HEH;e[%HUHW&HEHD$HE HD$ HE(HD$(HD$HHD$ HHD$(HHm-H-H9l$0Ht$0HheH9FH9E~ 1} 0HD$0HPH9UHHHEH9@H@t HHD$0DE p D8^@ 3Ht$0H~HA 3HuHȃ[:F:DA9HtLL$H,LL$HD$HH5H9p-HH.HH-H,HELL$0HH5HH.LL$0IMR.Hm HH5+H9p.HH/HLM.ICLL$HLLT$8H5MHL\$0H.L\$0LT$8LL$HHH0I+C!HcH9B^/Ht$ HLL$HLT$8HT$0|9HT$0LT$8LL$HHIH /I(!HEL)HPpH*HRHs*LL$8LHLT$0LT$0LL$8IM4Hm'Ht$(LLL$HLT$8LD$0耷LD$0LT$8HLL$HH,1I('HbI9B/LL$8LT$0vLT$0LL$8H'M1E1E1ɾEA1艼HD$HM1E1E1HD$1E1BHD$(AfHD$ HD$۵IIM1E1HD$E111HD$(E1{BAcHD$ HD$HD$fDHY`LH5H81 HFLHD$p蟶IH5ELHVYHD$xHIMRH$L|$pLt$xHD$H$HD$0H$HD$8@HELIGH5QLHVݸHtH$IM~H5LHV贸HtH$IMXH5LHV臸HtH$IM+HT$pLLLH5-BHZMYCD$H5@MjM"IjIEHEI*LLH6ImIRL7EfE1H=L-HGLHHHHjH_LH5dH81fHD$1E1E1HD$(E1BAdHD$ HD$HD$H=iHJH5K&=HfD賲II1E1E1HD$E11۾BHD$(AdHD$ HD$HD$MBAc1HD$E1E11HD$(E1HD$ HD$HD$HELI'HLT$胶LT$fIhH'MxHEII(LHL~4HmI.H(!蛱1E1E1HHD$LT$sME11۾BHD$AcHD$(HD$ HD$HD$M1E1E1HD$1E1CHD$(AlHD$ HD$@˰HHELIuHLD$ LT$>LD$ LT$_MCAlHD$1E11HD$(E1E1E1HD$ HD$HD$1E11HD$(BE11HD$ AdE1E1HD$HD$LL舴LLD$sLD$fLUIH-L5HEHHzH=wlY1LHH蘱HH7(H+=M#CAmf.KLD$HHD$1E1E1E11HD$1۾BAdHD$(HD$ HD$HD$'@H`DR裮HM1E1E1HD$1E15CHD$(AnHD$ HD$$@LEIM7CAnH5ZH9tILT$7LT$u6LHLT$>/LT$ILfDH耲IRBtLr IjH=jLT$M1M;DHD$E1AHD$(HD$ HD$1/HLL$8LT$0LL$8LT$0GH=LL$LT$jLT$LL$IM1M;DHD$E1AHD$(HD$ xLL$(LT$ HD$L\$M1E1E1HD$E1vCApHD$(HD$ HD$ M1E1E1HD$1E1CHD$(AwHD$ HD$.MCAy1E1PLLL$HLT$8HT$0ϚLL$HLT$8HT$01LH.HHM1E1E1HD$E1CAqHD$(HD$ HD$ەHD$HH@H5RMAoE1H8+1E1E1HD$YCHD$(HD$ .LLL$8LT$0LL$8LT$0 LhHMLPIEIHHD$HH LLLLL$LT$ImLT$LL$HLLL$LT$iLL$LT$1E1E1IJH;? HLL$蠖LL$HIHmI@LL$LLD$HLD$LL$HIqLL$(LHD$LD$LT$HHD$ LL$(QLL$PLLD$LT$HHD$(LL$PLվH_LD$LT$LL$PI(LT$LLL$LLL$fH Hx8HH(xH sLL$HEH>H5ZH81LL$M;DDA1HD$E1E1E1HD$(HD$ HD$IJHIjHHEI*mLL$`LD$XHL$PL\$H՗L\$HHL$PHHD$LD$XLL$`HD$D$HIHHIHcD$HHT$(LLLL$XMD@HHIT1L\$PLT$H3LT$HL\$PHLL$XHIIEHD$HHIE6 I* LL$HL\$L\$LL$HHIHhILp LL$PL\$HHD$LT$L\$HHLL$PH$Ll$8H5HLT$HL\$L葘L\$LT$HLL$PtH5LHLL$HLT$8^L\$LT$8LL$HJLLHLL$PLT$HL\$8 L\$8LT$HHHD$LL$PZI+I*HmHt$H;5<H;5< H;5>= H|$LL$8aLL$8H|$0LL$8H5n LL$8HH5H9pwHH'HHHH5HLL$8HT$0HT$0LL$8HHH*H51HLL$0LL$0HILHm I(7LLL$0`LL$0 яLT$LL$ Hu H:H5LH8,LT$LL$ Ht$HHD$HHM!DMAHD$1E1E1HD$(HD$ HD$1HLL$HLD$8LT$0訓LL$HLD$8LT$0LLL$8LT$0肓LL$8LT$08L$M;;DAeHD$@MIE1HD$@IOLIlIELL$8HH>H=gKL\$0L\$0LL$8 LL$@1LLL\$8HD$0mLD$0L\$8LL$@M I+ ImI)LLLD$0LD$0HII(HHH IEHT$0LH5HH HT$0 H*>IEMHt$@HHD$PHHH|$LL$PLL$PLLD$0ڑLD$0*LLL$8LD$0辑LL$8LD$0LLL$8LD$0蝑LL$8LD$0L膑M1E1E1HD$E1E1ACHD$(AnHD$ HD$HPhH Hz H58I9p@ MhIUHMtAPII H;7H;'8EH@hHD$0H Hx MHD$0LL$HLHLT$8PLT$8LL$HIM1E1E1E11E1CHD$AxHD$(HD$M;DA1E1E1HD$OM1E1E1ɾfEA4H=H~LL$0H5zLL$0H1E1E1IMCAqؿ0HD$HH#6H5HMAqE1H8耐1E1E1HD$CHD$(HD$ M1E1E1HD$E1E1FCHD$(AnHD$ HD$HLL$PL\$LL$PL\$yH1LLL\$0NL\$0LL$8HIM1M;EE1AsHLL$`LT$XL\$P膎LL$`LT$XL\$PM1E1E1ɾhEA!HUH HEH0Ht$HpH@Ht$ HD$(聉LL$0IGH=LL$0LL$0HM1E1ɾkEAH=HHLL$8H5DLT$0zLL$8LT$0I"LL$HLT$8L\$0HPH=uLL$8LT$0LT$0LL$8IM1E1E11E1CAyHD$HD$(HD$ EAE1E1CHLL$膎LL$M;DA1HD$E1E1M1E1E1ɾ|EAuLjMLBIEIH* HT$ LLLL$HLT$8LD$0 ImLD$0LT$8LL$HHsLLL$HLD$8LT$0CLL$HLD$8LT$0HM;mEE1AH|$0LL$芍LL$H|$LL$hLD$`L\$PߋLL$hLD$`LT$XL\$PHLL$LT$贋LT$LL$H2LL$H5HH81蓆LL$MBMBMjIIEI*wLL$8LD$0衋LD$0LL$8HI MCHLL$MALT$0LL$1E1۾!DME1HD$LT$0HD$(HD$ HD$hLLL$HL\$誊LL$HL\$H|$LT$PL\$H臊LL$XLT$PL\$HM1ME1HD$IE1ɾ!DHD$(AHD$ HD$1E1E1E11MnM1E1E1ɾEA1E1E1MžEAA@HuHOHu0FH|$0HH@@%Ht$0H~0H|$LL$XL\$PIM1E1ɾDA駽HLL$HD$eLL$LD$BM1LL$DHD$AcLL$8L\$0褄L\$0LL$8HYH/H5sAAH8LL$8M1L\$0EME1鬿1LL$ILL$LT$LLL$LT$LL$LT$M;pDLL$ 1HD$AE1HD$(HD$ HD$sM;EA1E1E1鲳M; EA1E1E1闳H5ȫH=1LL$0}LL$0HHLL$8HD$0mLD$0LL$8I(tVM;IEA1E1E12H&LL$HLT$8H8YzLT$8LL$HtCLL$HLT$8LLL$0~LL$0M1E1E1ɾEEAȲM;EA1E1E1GA M1M;EE1AHLL$8HD$0E~LL$8LD$0M1E1E1E1ɾ.EAHLL$0 ~LL$0XME1E1ɾ#EAME1E1ɾ!EAH=LL$0>LL$0HLL$HLT$8LD$0}LL$HLT$8LD$0"M1E1E1ɾDA鞱I)IApAPHH IXIM11IɾDALLL$XLD$PHL$HL\$ }L\$HL$HLD$PLL$X^I%H$LL$@LT$8H0LD$0 wLT$8LL$@ |LD$0H5W:AI@HPH#H81wLL$@1E1LT$8EME13LLL$HLT$8LD$0yLD$0LT$8LL$HIApA@HH HIAWAVAUATUHSHH#LnH|$H$H-#HDŽ$H$H#H$H )I II0*L5"IHE HmH$HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$HDŽ$HEH$HHIEHH9P+HH+HL=L$MM*IGH5jLHH+IH$M+H/HDŽ$zH$IH,HEHh|H$IH{,Hl!H5H|ID$H$L$HH0*H=Q2H$tH$5LLIlwM;6L$I,$&H$H/&H$HDŽ$H/&H$HDŽ$HmHD$ht&H|$hH5HDŽ$HGHH@4IL$M:4H5ۦL9tIGH; 9IJH; HH$I/(H$H;= HDŽ$H;=H;=v rH$Ņ8H/'HDŽ$H$H$H$vxHHHH9P>H/HAHH=H$H>HGH5HHAH$H@H$H/*4H|$hH5HDŽ$HGHHAIM@HIxHD$ H9?*MxM2*IXIHI(8H{H-yH980Ht(0LHIL$I/8L$[*I&Ln0Lu(^vOfDL5IBffA.Gz fDHHD$XE1E11D$dE1E1H$HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$Ht H/Mt I,$H$Ht H/H$Ht H/Mt I)Mt I(T$dt$XE1H QH=|FMt I/Ht H+HL$HtHHD$XHHH\$PHtHHD$HHH $HtHHD$HHMt I.HT$HtHH$HHH$Ht H/Ht$(HtHH$HHH\$HHtHH$HH}HL$@HtHH$HHoHT$ HtHH$HHaHt$0HtHH$HHSH\$8HtHH$HHEHL$HtHH$HH7Hm?H$Ht H/5Mt Im5HL[]A\A]A^A_LsHsHxs)Hhs8HXsFLHsEH8sS+s]fDHshHsvHrHrHrHrHrHmHrrfDLrLL$pLD$hqrLL$pLD$hafLLL$pLD$hNrLL$pLD$hNLL$pLD$h)rLL$pLD$h@f.LL$pLD$hrLL$pLD$h/fLLD$hqLD$h%fLqLu(I!@H|$hH5,HGHH(8IL$Mz7H5#L9+IGH;:IuA+HyHH$I/k-H$H;=THDŽ$H;=&u H;=(H/0HDŽ$;H|$hH5HGHHw=IL$M<IGH;8"2H;U*H@hHPH@HPL1IL$MPH$H/1L$L;=םHDŽ$HDŽ$-L;-*LLlH$H;SHHH9P\THHUHHsH$HSHCH5wHHHUHH$HUH$H/L8HHHDŽ$H9PTVHHWHL%͓MRVID$H5LHHfWIMyWI,$2:HHvH9PYH]H\HL%IMXID$LL$LH5HH\LL$HH \I,$MMHIyHD$ H9^H5^H92LL$kLL$2LHLL$LL$IL$MH+LH$H^I,$YNH$H5HGHH_HH _H$H/NH$HD$ HDŽ$H9G.bHH$H$Ht H/H+= L$IH;|$ aH$H9HIHLIL$H$H/ H$H$HDŽ$aH/ H$L$HDŽ$HDŽ$Ht H/9H$HDŽ$Ht H/&H$HDŽ$Ht H/H5DyLHDŽ$HH$HHz*H;H;v+H;H.EH$Aą+H+HDŽ$EHŇHoH9P/HoH1HHoH$H/HCH5|HHHo1HH$H$H0H/H$L$HDŽ$H{H;|$ H2H9HGHLHH$H$HD$Ht!HT$HH$HH(H$HDŽ$H$5H$H/>H5owH$HDŽ$H9BHGH;7A9H&HHH$H/EH$H;=HDŽ$H;=dH;=tCH$Ņ$6H/'HDŽ$H5vH=1 H$HH`HH$H/;BD$XE1E11D$dE1E1H$HD$Hl$hHDŽ$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$fDBH$Ņ D$XE1E11D$dE1E1H$HD$Hl$hE1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$L@IPB1Lb 'MxH=L$BL$ 1LA@HSBHRH$ .(LcH=6Au.LH$LIYDM;BIH_eE1}HFLH$LCI1@H5qvLHVEHtH$IMH5,wLHVEHH$IfDIGHH$HH9\HuH;EMfA.GfA/GHH{HH9HuH;fIWB Lb MH=E@u+HLAIBM@IH}UE1fDD1)[@IHD$Hl$hE1E1HD$8E1E11HD$0H$HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$dNfDD$XE1E11D$dE1E1H$HDŽ$HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$LCCfD?H,HD$E1E11HDŽ$E1H$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$d@HHgH9P,HgH-HHgH$H+HCH5tHHH-HH$HQ.H$H/H$L$HDŽ$HH9Gx1L豿HH$H$HtH/#H$HDŽ$H$H0H/!H5oH$HDŽ$H9HGH;Y;5HHHH$H/$"H$H;=HDŽ$H;=H;=A;H$Ņg3H/"HDŽ$H-|L% nHEHHAH=8;[1LHHY>HXH$HH$H/8D$X=Hl$hHDŽ$D$dE1E11HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$@!fDD$XE1E11D$dE1E1H$HD$Hl$hE1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$SHX?K? fDH;IcLvffDLh>IGHH8K>fDL8>HQHH^H9?҉T$dkHH$HDbHD$PHzH9P'1HbH3HL bM|0I@LD$LH5oHH\3LD$H$H$H2I(=IH4HD$PLD$HI@?LD$HH$H$26HyH\aH9P97HCaH7HL /aM?7IALD$LLL$H5VrHHn9LL$LD$H$H$H8I)p$H$H5sL$H$>L$6H$H/L'H$H$HDŽ$HELM)2H=vHT$L$7L$HT$BLD$LHAHH$}9HLD$wKH$H$H$H/E*HDŽ$I("*H$H/*H$H$HDŽ$H$H/)HHDŽ$HD$fDH@;*IQBLb ^$IiH=bLL$5LL$u5HHAI~8MLL$[6LL$HIbE1f{:fD:fD{:ff.Gf/G@HyH ff.Gf/G~fDH9HsHD$Hl$hE1E1HD$8E1E11HD$0H$HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$d^fD4IH=kHB^H5C^HfDD$XH$HtH/u9HDŽ$H$Ht H/H$HDŽ$Ht H/H=ctHDŽ$"4D$dt$XH= H yH$H$H$YH-tL%seHEHH$H=#3$1HLH5H3HHHmD$XqD$dDH$E1E11H$H$E1p4Hl$hE1E1HD$H$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$6fD;7IfD+7\fDH=i蔱H@7fDD$XC2fDE1o6TfDD$XaD$dD$X H$s1I{6H;L3H$IHcHD$Hl$hE1E1HD$8E1E11HD$0H$HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$d馿H-WqL%bHEHH<,H=x091LHI2M;L$L蕩H$H/D D$XE1E11D$dE1E1H$HDŽ$Hl$hE1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$钾LLL$4LL$HSBNHj .H[H=P/,LHH1H],H$H$HD$HL4HDŽ$H$HtH/zHDŽ$I(D$X.L3HD$Hl$hE1E1HD$8E1E11HD$0H$HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$d".IMgMIGI$HH$H$H/"H$H$LH$I,$wL2jD$X>2(IHHH5tVH'H$H$L謏H$IHH$H/H$HDŽ$H/ـHDŽ$I,$m~H$HGH;^}HWHbkHG LgH$I$H$HH$H/%kHDŽ$HT$0HtHH$HHF{HT$8H$HtHH$HHRjH5`THHDŽ$#4*LD$Xr"E1E1H$D$dHl$hE1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$鍆יHjHD$H:Ld$HLII$H$HI$zH|$IkHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$XD$d魅D$XE1E11D$dE1E1H$HDŽ$Hl$hE1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$ LD$LD$BH=-HH5讀HD$逎HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$dGD$X"E1E1H$D$dE1E1Hl$hHD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$ɃTHS>HD$Hl$hE1E1HD$8E1E11HDŽ$H$HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$dD$X!Hl$hE1E1H;E1HDŽ$D$dH$HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$銂H=;4H5%1bH$HH?;lH$H/W!D$Xw IHl$hE1HDŽ$E1D$dHD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$ՁH|$IqH;H|$H$H+?IE1Hl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$Xe D$dH,$E1Hl$hE1HD$E1H$HD$8HD$0HD$ HD$@HD$HHD$(HD$D$X!D$dxHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$D$X!D$dH=(pHD$雉H|$UI齉H5#H|$\sH$類H鑟Hl$hHt$HwHD$IHD$X%H$E1E1D$dhHD$Hl$hHD$HITHt$LLD$rLD$H$Mߓ#HD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X"D$d~H5 'H|$w_H$H$H*H$HH9PHhHH@HEHH$H$H/%H$LHfrH$Hmm%H$H$H+H$H/g%H$E1E11H$1HDŽ$1fH$H$Hy*H$H/ %H$H5#HDŽ$HDŽ$HT$PHGH|$HH6ЅZD$X$E1E1H$D$dE1E1Hl$hHD$HD$8HD$0HD$ HD$@HD$HHD$(H$|1HɍIE1Hl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$dP|H5H=-1\\H$H$H%H<$NfH$H/!D$X"E1E1H$D$dE1E1Hl$hHDŽ$HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$b{HD$E1E1Hl$hHD$8HD$0HD$ HD$@HD$HHD$(HD$D$X "D$d{HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$XD$dvzLL$HD$阄mLD$H4HvE1E1HDŽ$HD$Hl$hE1HD$8H$HD$0HD$ HD$@HD$HHD$(HD$D$X!D$dHD$yLLD$LD$JH*H5 H87IE1Hl$hE1HD$E1E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$X D$dxE鍇IE1E1D$X D$dE1E1H$HD$Hl$hHD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$`xLLD$cLD$錄HD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X"D$dwD$X"E1E1H$D$dE1E1Hl$hHD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$Uw`]LD$QLD$gL?5HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(D$X"D$dvHD$E1E1Hl$hHD$8HD$0HD$ HD$@HD$HHD$(D$X"D$dvvH<$HD$HD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$duIAH$HMaHI$I)nMAD鋋Hu阔H59H= '1UH$H$H.1H<$}_H$H/-D$X"E1E1H$D$dE1E1Hl$hHDŽ$HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$tLL$LD$LL$LD$黁D$XHl$hHDŽ$D$d-LUFKIE1Hl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$dsH=HH5doH雅H5H=6%1SHD$H$H(H|$]H$H/'ID$X H$E1D$dE1E1Hl$hHDŽ$E1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$rHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$drH/HDŽ$PHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$D$Xh"D$dq6fڸHD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X"D$dq LL$L$xL$LL$H[<HDŽ$ D$XE1E1Hl$hD$dE1E1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$=pLETD$X;"E1E1H$D$dHl$hHD$HD$8HD$0HD$ HD$@HD$HHD$(oLD$LD$n|HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$D$Xf"D$d&oH|$ILD$X9Hl$hHDŽ$tH龍銍HH$}HD$E1E1Hl$hHD$8HD$0HD$ HD$@HD$HHD$(HD$D$Xp"D$dxnd鹶LWM鄵HD$E1E1Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$D$X"D$dmH=HH5iH$H韲H<$L+aH$߂IE1HD$Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$X D$d'mIE1Hl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$dlI~L{-H=_\Ha~]ѳSڈLLD$ALD$yLL$LD$(LL$LD$xyLD$LD$yHD$E1E1Hl$hHD$8HD$0HD$ HD$@HD$HHD$(D$X "D$dkD$X!%Hl$hE1E1D$d H$E1HD$ZkHbvzLUmyHчH5H8: jHD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HHD$(D$X1"D$djH|$($IۅHD$E1E1Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$D$X"D$d`jHD$ff.@HpHۊHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$D$Xs"D$diH5 H=N1IHD$H$H<H|$SH$H/)ID$X H$E1D$dE1E1Hl$hHDŽ$E1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$hLUwHD$E1E1Hl$hHD$8E1E1HD$0HD$ HD$@HD$HHD$(HD$D$X"D$dChH<$H$[H=*XH$HHoHHGHEHH$H$H/{H$H$H\H$HmkH^IE1E1D$X D$dE1E1H$HD$Hl$hHD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$gH"uvuMhMyM`IEI$I(H$LL[H$ImpyLcyH=HH5WbHD$(xHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HD$X/"D$dfH=VHD$(HD$E1E1Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$D$X"D$deHD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HD$X4"D$d=e(L=P{H$/L$zIMHl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$X D$ddHl$hE1E1E1D$X%H$D$d ldH|$hHD$ȃc@YHD$HTJ鐭@h6@Ll$XHl$hE1E1D$XH#E1E1H$D$dHD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$cHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X!#D$d(cH5H=1SCH$H$H! H<$EMH$H/:D$X#E1E1H$D$dE1E1Hl$hHDŽ$HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$YbD$XK"E1E1H$D$dHl$hE1HD$HD$8HD$0HD$ HD$@HD$HHD$(aHl$hE1E1E1D$X$H$D$daHU~H56H8LL$iHD$Hl$hE1HD$8HD$0HD$ HD$@HD$HD$X<"D$dpaHD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HD$X9"D$d`TLD$HtIE1HD$H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$dG`H=@ PIsH=, HH5[IsHD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$d_[HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$XD$d^HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$XD$du^H=nLD$NLD$HD$jH=NHLD$H5ZLD$HD$jLL$IEjHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$XD$d]ԧIE1HD$Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$X D$d!]IE1HD$Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$d\IE1Hl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$d[`IoH;yH|$,HD$H$H'Hl$hE1E1E1D$X%H$D$d [HD$E1E1Hl$hHD$8E1HD$0HD$ HD$@HD$HHD$(HD$D$X/#D$d?[LϽ#MAoeHl$hE1E1E1D$X%H$D$dZH=YKIfH=HKH5LwVIfD$X"E1E1H$D$dE1E1Hl$hHD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$!Z,+ͦIMHl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$X!D$dYHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$XD$d YHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X"D$dXHVv8zlyH_uUuKuA|uH=%HH5SHD$yH=HH5SH$Ll$XHl$hE1E1HD$E1E1H$HD$8HD$0HD$ HD$@HD$HHD$(HD$D$X<#D$djWH=cGH$xIE1D$X-!H$D$dE1E1Hl$hHD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$VHl$hE1E1E1D$XH%H$D$d VD$X IE1H$D$dHl$hE1HD$HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$ VHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X"D$dUϡHT$H|$HE11A1=IHH&H|$XH:I,$LFL9/2LHLHHD$@H S靠IVIFHlHE邢 L鱟IE1HD$H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X+!D$d#TIE1E1HD$HD$8E1E1H$HD$0Hl$hHD$ HD$@HD$HHD$(HD$H$HD$PD$X)!D$dSHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X$D$d#SHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X$D$dRHl$hE1E1E1D$XZ%H$D$dRH={CHD$sHD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X$D$dQHl$hE1E1E1D$X\%H$D$dQHD$Hl$hE1E1D$Xa%H$D$dQHl$hE1E1E1D$X_%H$D$doQLwatmtcsLVsD$Xc%Hl$hE1E1D$dH$HD$QL,$H\$xIE1L|$ Ll$XE1Hl$@L|$pHD$H$HD$Hl$hD$Xt#D$dPؚLl$XHl$hE1E1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$D$XF#D$d[PLl$XHl$hE1E1HD$E1E1H$HD$8HD$0HD$ HD$@HD$HHD$(HD$D$XA#D$dOHD$IE1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$d4O?Ll$XHl$hE1E1HD$E1E1HD$8HD$0HD$ HD$@HD$H$D$Xf#D$dN鱚H5H=a1.HD$H1#Lt$L8IH$HIv"D$X#D$djIu{HD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HD$XI"D$dHHD$E1E1Hl$hHD$8HD$0HD$ HD$@HD$HD$XD"D$dH贾HD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PD$X D$dHHD$E1E1H$HD$8Hl$hHD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$X D$dGHD$E1E1Hl$hHD$8E1HD$0HD$ HD$@HD$HD$XA"D$d;GFH|$WIE1Hl$hE1HD$E1HD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$D$Xs D$dFHHH9T$_VHuHt$H;5|cHVHbIE1E1H5H8ZH$E1H$HD$Hl$hE1HD$8HD$0HD$ HD$@HD$HHD$(HD$HD$PD$XY D$dEHD$E1E1H$HD$8Hl$hE1HD$0HD$ HD$@HD$HD$XL"D$dhEHaH5sH8bHaH5sH8GLD$LgM?RHGI$HH$H$H/H$H$LLD$,9I,$LD$IQLLL$LD$ǺLL$LD$QIE1E1D$X!D$dE1E1H$HD$Hl$hHD$8HD$0HD$ HD$@HD$HHD$(HD$H$HD$PHD$DH顕HD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X"D$dC蝹єL|$pH\$xL,$HH-HHnH <HEH_E1E1H5{H81SHl$ E1E1Ll$XHl$hD$X$D$dH$HD$HD$BL$L$HD$E1E1H$HD$8E1E1Hl$hHD$0HD$ HD$@HD$HHD$(HD$D$X#D$dgBH^H5pH8aDH$HD$RHl$ Ll$XE1E1D$X$E1E1H$D$dHl$hHD$HD$AHq^E1E1H5?uH81Hl$ E1E1Ll$XHl$hD$X$D$dH$HD$HD$AH=HH5?=IkL,$MLl$XE1L|$ H\$xE1E1D$X#L|$pD$dH$HD$Hl$hHD$AL,$MLl$XE1L|$ H\$xE1E1D$X#L|$pD$dH$HD$Hl$hHD$@L,$H\$xIE1L|$ Ll$XE1HD$L|$pHD$H$D$X#Hl$hD$db@L,$H\$xIE1L|$ Ll$XE1HD$L|$pHD$H$D$X#Hl$hD$d @锍H]H8ttk+HXL,$MLl$XE1L|$ H\$xE1E1Hl$@L|$pD$Xu#Hl$hD$dH$HD$HD$?H,$E1Hl$hE1D$X!E1E1H$HDŽ$D$dHD$HD$8HD$0HD$ HD$@HD$HHD$(HD$>ML,$IHl$hL|$ H\$xE1E1Ll$8L|$pLD$0Ll$XE1D$X$H$HD$HD$D$d>ML,$IHl$hL|$ H\$xE1E1Ll$8L|$pLD$0Ll$XE1D$X$H$HD$HD$D$d5>@^L,$IHl$hE1L|$ H\$xE1E1Ll$8L|$pLd$0Ll$XE1D$X$H$D$dHD$HD$=LγґML,$IHl$hL|$ H\$xE1E1Ll$8L|$pLD$0Ll$XE1D$X$H$HD$HD$D$dT=LH NH5kLd$HHCL,$IHDHPL|$ E1HYH\$xE1L|$pH81LD$E1Ll$8D$dLl$XLD$0Hl$hE1D$X$H$HD$HD$<輲QML,$IHl$hL|$ H\$xE1E1Ll$8L|$pLD$0Ll$XE1D$X$H$HD$HD$D$dBL$HcT$\$L$%D$Yf(Y=~T$L$f($YD$Xf/wAI>AVfII LH!H*AYLtfW I94?H(f([]A\A]A^A_@IFI>fWύ~ I>Y $AVfWu~ $fWf(XYf/vX AzfW im@H~CAUIATIUHS1HfDL~ADHH9uH[]A\A]DAWAVIAUATL%UH-SHHDIFI> HcL$ \D$%fAnfZAYAf(Yb|L$ f(fA*YwYD$XZf/wDI>AVfAA A*AYfA~t W_fA~D9l/HfAn[]A\A]A^A_IFI>f*YW} I>YL$AVf*YW}L$W(XY/vX AfA~RW fA~AH~CAUIATIUHS1HfDLADHH9uH[]A\A]DUHH0f. D$d|$ff.8=f/|$@H}UHD$} ˉt$T$\f/r5 D$f(^z~\$f/rH0]D$rL$ \^D$譃|$L$ D$f(Y\f( 3^ ~T$\$\f/(H0]f.|$\=BBY|$(ff.2Q5ƈ^t$@H`zff(D$YXf/sf(L$H}YYD$UL$Èf(YYYK\f/wbL$ 膂D$D$u%\d$f(L$ f(XeYD$(YYXf/D$'D$(YD$H0]DH0f]H0]{kfDUHH .qD$ud$f.R=J/|$%-d$H}UfH*YT$T$  t$T$ \/r- D$ (^}\$ /rH ]fD$L$\^D$|$L$D$ (Y\( u^|}T$ \$\/#H ]Dt$\5>f6Yt$.;Q=d$^|$H}f(D$YX/s(L$H}YYD$ UL$f(*YYD$YY~\/w_L$D$D$ =Q\|$ (L$(XCYD$YYX/D$D$YD$ H ]@H f]H ]~%ff.HHH?PHHff.HHH?PHff.HHH?PHHff.HGH?f.XE„f.f(D„USH(-]f/Kf1X A\$^T$YYX )Y\ %YX !Y\ YX Y\ YX Y\ Yf(L$}L$X T$f(5\$^f(\f/YX X\vHH~C\L$Hf(T$?}L$H9T$\}H(f([]f(f\H,H*Dff(ff.@H$L$,tYD$X$Hff.HD$AvYD$HfDHHH?$L$PYD$X$HHL$1sYD$HfDHL$ zYD$ HfDUHH -D$f/L$rf/sBD$HrH$D$r $H ]X^f(ff/wvH}UH}$U$ F^L$D$f(vT$ #^L$$f(uX$=f/rf. $H ]^f(ff.fHY,qHXfDUHH $D$v $Hf(Yf($v\$$H]Y^f(ff.fUHHqHD$qL$H]^f(ÐHD$t^D$Hxff(ff.z uHT$s T$H^tff.HD$s~ ifWPx~ XfW |^L$HMtff.fSHH$L$ ff/wFH;Sf/r \\f(eyL$Y$H[\@X?yYD$X$H[ff.AVfI~SHHL$H;S~f(~\f/vxfWWxL$H[YfInA^\SHH$L$DH;Sff/v 0~\^sxYD$X$H[fHtHnff.HD$qfXf.wQYD$Hwff.fUHHD$MoL$HY }$f(L$nL$ff(f.w!QY $f.w1Q^H]f(D$f(wT$ff(f( $v $f(ff.AWf(AVATUSHH@f/D$f.}zuE1H@L[]A\A^A_f.D$fWr}E1Jm |D$f.IL$H;SL$Yf/L$wH@L[]A\A^A_fDff.-Qf(D$L$qvL$Y X D$0Yf(L$ \ \-|f(\%^ f(d$8XfI~X|fI~w^\fH~f.H;SH;f(\{T$ST$ {D$f(fTr{\fIn^L$XD$ YXD$XtL$f/ L,\$rfHnf/M_5f/v f/Gf(L$(uD$fIntL$(t$8D$D$ Y^Xt|$ID$X|$f(f(\fI*YL$0\L$D$fH*L$3lL$\f/L$H@L[]A\A^A_D$sf(fyUH\^f(rH]kf.AWfAVH*AUIATIUSHH$t H9rm |yMe$AEf(\A}|$Pf/t$H L$PT$Yf(fD$HAEX\$AM(f(L$ur\$Y\$HH,fL$T$f.Im0f(\$hz Qf(Y\$HY%x->y\f(fTf.(5xd$PXf(t$(fD(D$AE8ffA(H*XXYXf(Am@$A\f(\$xA]P-Y|$8^f(A}H\XfD(\$@A]Xf(\^f(YXYf(Yl$H\\$`^A]`Yf(XYfA(AXXL$pAMhAYf(D$AEpfA(^D^Xf(D$XAExfA(XD$ AMI)IEH$H;Sd$ H;Yd$Sd$f/d$f(Nf/d$"t$l$@fH*v\YT$0f(^XXD$8\X\$(fTv^\f/L$[oL$T$0L,MI)LH?HL1H)H~#D$hYD$(fH*\f/ID$fd$P^d$HH*YI9/f(ef/$M)f/L$(MGH[]LA\A]A^A_f.uADEf/d$Xwcd$0D$o^D$`XD$8nL,MSEJd$0\d$L$@uYYL$`d$0D$oo\$x^D$p\f(vnL,M9EL$d$0\d$XtYYL$pqfDHEf(I9ff(H*H^\YI9}IFH9ff(H*H^\^H9~dff(^tt$hLIXOf(XYHX@^^XT$(YfH*^f(T$0\$$n\$T$0f(f(\f/Xd$0f/IFfEfEL*HEfEL*ID$fL*$L)H*D$fE(EYfA(fA(D$A^fE(D$AYfD(l$EYDYD$$D$D$0mD$$D$fA(^D$mT$HD$$t$Yt$PAYf(^l=$5d$0f(-D$^f(D$D$DY$D$D$D$D$\f(^\9fD(^D\D^ &f(A\fEM*DX\$(DY$A^DEXfEM*AYDXf(A^A^AXfD(D\f(E^A\fD(A^D\f(E^A\fD(A^A^Xf(A^A^D\f(E^\A^A\fD(A^\A^D\f(E^\A^A\A^\^L$A^A^XXf/f($YT$\XiL,fDH,ffUH*f(fT\fVf(f(t$P\l$Hf.Bf/B}J8rz Hj0L$J@Yt$P59p$JH|$HL$8JPYt$(L$xJXL$@J`L$`Jh\$hL$pJpL$JxL$XL$ -D$hT$L$ iT$L$ff.AUIATIUSHH8D$t H9r_d$-nfMeI*AE\AeAm f($l$i$YR_d$$L$f(AEfYYAeXX mnf.Q-f(YXf/aH,Im0$H;S$1f/f(vt@HBH9}%$H;S$f/vQ1f(HBH9|Lf\H)H*YT$YfH*YT$^f/vH1H8[]A\A]f.Bf/Bz ZHj0|$1f(d$(\$ T$ $gd$(T$-ն $\$ YXf/f(l$ d$$f$d$l$ f(fYXH,ff.Hf.lEurmfUHH*f/rYlf/rM]f`l\lYf/f(r&qfIH]L)fD1D]bf bIH]L)UHH$f(L$D`L$ff.k$f/v]f(HL$\f(aH$N]L$ff(f.~QX$H]YXY kHf(]fHHH*X$H]fafDkH]fB$HH]8aD$f(dT$f(hf.UHH $f(D$dd$$Hf(Yf($`\$$H]Y^f(ff(f(SHXf(H ^L$\$l$[\$L$Y jf(YYYf(YXff.w\Q\YT$\$H;XT$S\$T$f(f(X^f/s Y^f(H f([f(\$T$rc\$T$f(UHH@D$8f(L$]icif/D$if/D$s|$f/ i-ift$Yl$YXf.Qf(XL$f(Xf.JQ\T$f(f(X^f(YXXL$^L$0&f(^bXD$L$ \f/D$sH}UY iZ\$0H}f(YXXL$f(L$^\d$(YL$ UL$ f(h\Y\f/D$\H}UD$D$($_f(0hf/D$vfWhXT$8 rhf(fTgT$X@hZt$T$\'hf/vfW9hH@]H}UX\MgYgH@]f,g^D$f.QD$HXYD$ tXD$8f/vXgf/gq\gdD=ff(|$|$^Xf|$D$0(^`D$qf(L$ D`L$ f(f(,`l$f(hff.fSHH0D$ fWfWD$(H;Sf/D$ D$H;SYD$(^T$f(fWfYf/~f(T$\$`\$D$f(_L$^eX_L,MaT$ff.E„EH0L[f/ArA@H4eHH?D$\$PL$$f(f/vf(fDYHXf/wHÐHD$X~eT$$fWf(QV~ye $f(fW4ef(^df(fTf.v3H,ff(%adfUH*fTXfVf(f/HsH,HDf/hdrU]ff.AVSHH(\cf(D$dXD$H;SH;D$S%c\d$fI~ c^L$f(tX=,df(fTf.v;H,f=ecH*f(fT\ ϭfUf(fVf/]5%cf/Kf(L$T$^XWT$l$f(fInYf(\b^Yf(\b^f/H(H,[A^f.f(f(H8H\\T$H?L$D$(f(d$^l$ 4$P4$d$L$T$f/r;l$ \$(Yf(Yff.wPQXH8f(f(\ a\fYYf.w*Q\H8f(f($[$f(f($r[$f(1HATIIUHI SHLHI LHL IIL III LH I ĸH9wfDH;SD!H9r[]A\DH;SL!H9sH;SL!H9rf.ff.@AWAVAUATIUSHHttHHH?IIH9wqHEAEDjAME9v,D1AAA9sH}UAME9wI IHL[]A\A]A^A_@HEHEuQLbIHHI9v+L1HIIH9sfDH}UIHI9wHL$DHEH}L!I9rL$oDHEH}D!D9wL$ODЉI>@I0fDAWAVAUATAUSHteHGIH?AEu`JL$ Dl$ AME9v%1AAA9sI>AVAME9wI AHD[]A\A]A^A_@IFI>D!9rD$f.AfAWAVAUATUSHt$H\$PfA։IAMfAAEEF-DD$ H?AUDD$ A$3EAfA9v^AAƙAAf9r&ID3A$xA<$3AfA9v%uI}AUA$3AfA9wD$H[]A\A]A^A_+A$PA$D!f9s!uI}AUA$D!f9rfD$H[]A\A]A^A_fuDH?AUA$D$fH[]A\A]A^A_@+AxA9+A)ff.@AWAVAUATUSHt$ H\$PA։IAMAAEE~H?AUA$3D@A8v^AEAƙAA8r#Ff3A$xDA<$3@A8v%uI}AUDA$3@A8wD$ fH[]A\A]A^A_f+A$PA$D!@8s!uI}AUA$D!@8rD$ H[]A\A]A^A_uDH?AUA$D$ H[]A\A]A^A_D+AxA9+A)ff.@USHHl$ tAHL˅tmA)EH[]fDH?QEAWAVAUATIULSH(HL$HujH~OHHHfInLHHLHH9uHt$HHtLdH([]A\A]A^A_IHH9EH|$~DrE1Dt$D$fI}AUt$I9s$D$19s@I}AUI9wH LJLIL9|$uLf.HEE1HLrH|$H\$fI}AUIHHL9s-HD$1IHH9sI}AUIHH9wHLJDIL9|$ufH|$1fI}AULHDHH9\$ufDH|$n1f.I}AULHDHH9\$uBIII LHI LHI LHI LHI LH I H|$E1fDI}AUL!H9rLJDIL9|$uIII LHI LHI LHI LHI H|$E1I}AUD!9wLJDIL9|$u[1OfAWAVAUATMUHSHt$H~pHAHHfnLHfpHLHH9uHH@t,t$HHHA4H9~HAtH9~AtH[]A\A]A^A_@IӃEDrE1Dt$\$ H~f.I}AUt$I9v$D$ 19s@I}AUI9wH L$C IL9uH[]A\A]A^A_fIII LHI LHI LHI LHA HE1I}AUD!9rD$CIL9uH[]A\A]A^A_ÐH1DI}AUD$AHH9uH[]A\A]A^A_1zAWAVAUATUSLHfHHAHCHfnLHfaHfpLʐHH9uHHt\Hxf4CHH9~KHxftH9~=HxftH9~/HxftH9~!HxftH9~Hft H9~ft H[]A\A]A^A_DIAfHEH~DrII1H$A1D$ ufI?AWfD9s@D$ Af9s1t1fA9vI?AWȉfA9wۿHDfKH;$2t1뒐AHHH HHH HHH HH HM4I11t@1!fA9sI?AW!fA9rDHfSL9uH[]A\A]A^A_@HI,IHDfCH9mI?AWATfHSH9uP1DAWAVIAUATUSH@t$ uVH~3#33>5(5 54R5???@@)EEELE 1. The Zipf distribution (also known as the zeta distribution) is a discrete probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. .. note:: New code should use the `~numpy.random.Generator.zipf` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Distribution parameter. Must be greater than 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Zipf distribution. See Also -------- scipy.stats.zipf : probability density function, distribution, or cumulative density function, etc. random.Generator.zipf: which should be used for new code. Notes ----- The probability density for the Zipf distribution is .. math:: p(k) = \frac{k^{-a}}{\zeta(a)}, for integers :math:`k \geq 1`, where :math:`\zeta` is the Riemann Zeta function. It is named for the American linguist George Kingsley Zipf, who noted that the frequency of any word in a sample of a language is inversely proportional to its rank in the frequency table. References ---------- .. [1] Zipf, G. K., "Selected Studies of the Principle of Relative Frequency in Language," Cambridge, MA: Harvard Univ. Press, 1932. Examples -------- Draw samples from the distribution: >>> a = 4.0 >>> n = 20000 >>> s = np.random.zipf(a, n) Display the histogram of the samples, along with the expected histogram based on the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import zeta # doctest: +SKIP `bincount` provides a fast histogram for small integers. >>> count = np.bincount(s) >>> k = np.arange(1, s.max() + 1) >>> plt.bar(k, count[1:], alpha=0.5, label='sample count') >>> plt.plot(k, n*(k**-a)/zeta(a), 'k.-', alpha=0.5, ... label='expected count') # doctest: +SKIP >>> plt.semilogy() >>> plt.grid(alpha=0.4) >>> plt.legend() >>> plt.title(f'Zipf sample, a={a}, size={n}') >>> plt.show() weibull(a, size=None) Draw samples from a Weibull distribution. Draw samples from a 1-parameter Weibull distribution with the given shape parameter `a`. .. math:: X = (-ln(U))^{1/a} Here, U is drawn from the uniform distribution over (0,1]. The more common 2-parameter Weibull, including a scale parameter :math:`\lambda` is just :math:`X = \lambda(-ln(U))^{1/a}`. .. note:: New code should use the `~numpy.random.Generator.weibull` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape parameter of the distribution. Must be nonnegative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Weibull distribution. See Also -------- scipy.stats.weibull_max scipy.stats.weibull_min scipy.stats.genextreme gumbel random.Generator.weibull: which should be used for new code. Notes ----- The Weibull (or Type III asymptotic extreme value distribution for smallest values, SEV Type III, or Rosin-Rammler distribution) is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. This class includes the Gumbel and Frechet distributions. The probability density for the Weibull distribution is .. math:: p(x) = \frac{a} {\lambda}(\frac{x}{\lambda})^{a-1}e^{-(x/\lambda)^a}, where :math:`a` is the shape and :math:`\lambda` the scale. The function has its peak (the mode) at :math:`\lambda(\frac{a-1}{a})^{1/a}`. When ``a = 1``, the Weibull distribution reduces to the exponential distribution. References ---------- .. [1] Waloddi Weibull, Royal Technical University, Stockholm, 1939 "A Statistical Theory Of The Strength Of Materials", Ingeniorsvetenskapsakademiens Handlingar Nr 151, 1939, Generalstabens Litografiska Anstalts Forlag, Stockholm. .. [2] Waloddi Weibull, "A Statistical Distribution Function of Wide Applicability", Journal Of Applied Mechanics ASME Paper 1951. .. [3] Wikipedia, "Weibull distribution", https://en.wikipedia.org/wiki/Weibull_distribution Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> s = np.random.weibull(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> x = np.arange(1,100.)/50. >>> def weib(x,n,a): ... return (a / n) * (x / n)**(a - 1) * np.exp(-(x / n)**a) >>> count, bins, ignored = plt.hist(np.random.weibull(5.,1000)) >>> x = np.arange(1,100.)/50. >>> scale = count.max()/weib(x, 1., 5.).max() >>> plt.plot(x, weib(x, 1., 5.)*scale) >>> plt.show() vonmises(mu, kappa, size=None) Draw samples from a von Mises distribution. Samples are drawn from a von Mises distribution with specified mode (mu) and dispersion (kappa), on the interval [-pi, pi]. The von Mises distribution (also known as the circular normal distribution) is a continuous probability distribution on the unit circle. It may be thought of as the circular analogue of the normal distribution. .. note:: New code should use the `~numpy.random.Generator.vonmises` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mu : float or array_like of floats Mode ("center") of the distribution. kappa : float or array_like of floats Dispersion of the distribution, has to be >=0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mu`` and ``kappa`` are both scalars. Otherwise, ``np.broadcast(mu, kappa).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized von Mises distribution. See Also -------- scipy.stats.vonmises : probability density function, distribution, or cumulative density function, etc. random.Generator.vonmises: which should be used for new code. Notes ----- The probability density for the von Mises distribution is .. math:: p(x) = \frac{e^{\kappa cos(x-\mu)}}{2\pi I_0(\kappa)}, where :math:`\mu` is the mode and :math:`\kappa` the dispersion, and :math:`I_0(\kappa)` is the modified Bessel function of order 0. The von Mises is named for Richard Edler von Mises, who was born in Austria-Hungary, in what is now the Ukraine. He fled to the United States in 1939 and became a professor at Harvard. He worked in probability theory, aerodynamics, fluid mechanics, and philosophy of science. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] von Mises, R., "Mathematical Theory of Probability and Statistics", New York: Academic Press, 1964. Examples -------- Draw samples from the distribution: >>> mu, kappa = 0.0, 4.0 # mean and dispersion >>> s = np.random.vonmises(mu, kappa, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import i0 # doctest: +SKIP >>> plt.hist(s, 50, density=True) >>> x = np.linspace(-np.pi, np.pi, num=51) >>> y = np.exp(kappa*np.cos(x-mu))/(2*np.pi*i0(kappa)) # doctest: +SKIP >>> plt.plot(x, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() uniform(low=0.0, high=1.0, size=None) Draw samples from a uniform distribution. Samples are uniformly distributed over the half-open interval ``[low, high)`` (includes low, but excludes high). In other words, any value within the given interval is equally likely to be drawn by `uniform`. .. note:: New code should use the `~numpy.random.Generator.uniform` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- low : float or array_like of floats, optional Lower boundary of the output interval. All values generated will be greater than or equal to low. The default value is 0. high : float or array_like of floats Upper boundary of the output interval. All values generated will be less than or equal to high. The high limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. The default value is 1.0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``low`` and ``high`` are both scalars. Otherwise, ``np.broadcast(low, high).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized uniform distribution. See Also -------- randint : Discrete uniform distribution, yielding integers. random_integers : Discrete uniform distribution over the closed interval ``[low, high]``. random_sample : Floats uniformly distributed over ``[0, 1)``. random : Alias for `random_sample`. rand : Convenience function that accepts dimensions as input, e.g., ``rand(2,2)`` would generate a 2-by-2 array of floats, uniformly distributed over ``[0, 1)``. random.Generator.uniform: which should be used for new code. Notes ----- The probability density function of the uniform distribution is .. math:: p(x) = \frac{1}{b - a} anywhere within the interval ``[a, b)``, and zero elsewhere. When ``high`` == ``low``, values of ``low`` will be returned. If ``high`` < ``low``, the results are officially undefined and may eventually raise an error, i.e. do not rely on this function to behave when passed arguments satisfying that inequality condition. The ``high`` limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. For example: >>> x = np.float32(5*0.99999999) >>> x 5.0 Examples -------- Draw samples from the distribution: >>> s = np.random.uniform(-1,0,1000) All values are within the given interval: >>> np.all(s >= -1) True >>> np.all(s < 0) True Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 15, density=True) >>> plt.plot(bins, np.ones_like(bins), linewidth=2, color='r') >>> plt.show() triangular(left, mode, right, size=None) Draw samples from the triangular distribution over the interval ``[left, right]``. The triangular distribution is a continuous probability distribution with lower limit left, peak at mode, and upper limit right. Unlike the other distributions, these parameters directly define the shape of the pdf. .. note:: New code should use the `~numpy.random.Generator.triangular` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- left : float or array_like of floats Lower limit. mode : float or array_like of floats The value where the peak of the distribution occurs. The value must fulfill the condition ``left <= mode <= right``. right : float or array_like of floats Upper limit, must be larger than `left`. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``left``, ``mode``, and ``right`` are all scalars. Otherwise, ``np.broadcast(left, mode, right).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized triangular distribution. See Also -------- random.Generator.triangular: which should be used for new code. Notes ----- The probability density function for the triangular distribution is .. math:: P(x;l, m, r) = \begin{cases} \frac{2(x-l)}{(r-l)(m-l)}& \text{for $l \leq x \leq m$},\\ \frac{2(r-x)}{(r-l)(r-m)}& \text{for $m \leq x \leq r$},\\ 0& \text{otherwise}. \end{cases} The triangular distribution is often used in ill-defined problems where the underlying distribution is not known, but some knowledge of the limits and mode exists. Often it is used in simulations. References ---------- .. [1] Wikipedia, "Triangular distribution" https://en.wikipedia.org/wiki/Triangular_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.triangular(-3, 0, 8, 100000), bins=200, ... density=True) >>> plt.show() tomaxint(size=None) Return a sample of uniformly distributed random integers in the interval [0, ``np.iinfo(np.int_).max``]. The `np.int_` type translates to the C long integer type and its precision is platform dependent. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray Drawn samples, with shape `size`. See Also -------- randint : Uniform sampling over a given half-open interval of integers. random_integers : Uniform sampling over a given closed interval of integers. Examples -------- >>> rs = np.random.RandomState() # need a RandomState object >>> rs.tomaxint((2,2,2)) array([[[1170048599, 1600360186], # random [ 739731006, 1947757578]], [[1871712945, 752307660], [1601631370, 1479324245]]]) >>> rs.tomaxint((2,2,2)) < np.iinfo(np.int_).max array([[[ True, True], [ True, True]], [[ True, True], [ True, True]]]) sum(pvals[:-1].astype(np.float64)) > 1.0. The pvals array is cast to 64-bit floating point prior to checking the sum. Precision changes when casting may cause problems even if the sum of the original pvals is valid.state must be a dict or a tuple. standard_t(df, size=None) Draw samples from a standard Student's t distribution with `df` degrees of freedom. A special case of the hyperbolic distribution. As `df` gets large, the result resembles that of the standard normal distribution (`standard_normal`). .. note:: New code should use the `~numpy.random.Generator.standard_t` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard Student's t distribution. See Also -------- random.Generator.standard_t: which should be used for new code. Notes ----- The probability density function for the t distribution is .. math:: P(x, df) = \frac{\Gamma(\frac{df+1}{2})}{\sqrt{\pi df} \Gamma(\frac{df}{2})}\Bigl( 1+\frac{x^2}{df} \Bigr)^{-(df+1)/2} The t test is based on an assumption that the data come from a Normal distribution. The t test provides a way to test whether the sample mean (that is the mean calculated from the data) is a good estimate of the true mean. The derivation of the t-distribution was first published in 1908 by William Gosset while working for the Guinness Brewery in Dublin. Due to proprietary issues, he had to publish under a pseudonym, and so he used the name Student. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics With R", Springer, 2002. .. [2] Wikipedia, "Student's t-distribution" https://en.wikipedia.org/wiki/Student's_t-distribution Examples -------- From Dalgaard page 83 [1]_, suppose the daily energy intake for 11 women in kilojoules (kJ) is: >>> intake = np.array([5260., 5470, 5640, 6180, 6390, 6515, 6805, 7515, \ ... 7515, 8230, 8770]) Does their energy intake deviate systematically from the recommended value of 7725 kJ? Our null hypothesis will be the absence of deviation, and the alternate hypothesis will be the presence of an effect that could be either positive or negative, hence making our test 2-tailed. Because we are estimating the mean and we have N=11 values in our sample, we have N-1=10 degrees of freedom. We set our significance level to 95% and compute the t statistic using the empirical mean and empirical standard deviation of our intake. We use a ddof of 1 to base the computation of our empirical standard deviation on an unbiased estimate of the variance (note: the final estimate is not unbiased due to the concave nature of the square root). >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> t -2.8207540608310198 We draw 1000000 samples from Student's t distribution with the adequate degrees of freedom. >>> import matplotlib.pyplot as plt >>> s = np.random.standard_t(10, size=1000000) >>> h = plt.hist(s, bins=100, density=True) Does our t statistic land in one of the two critical regions found at both tails of the distribution? >>> np.sum(np.abs(t) < np.abs(s)) / float(len(s)) 0.018318 #random < 0.05, statistic is in critical region The probability value for this 2-tailed test is about 1.83%, which is lower than the 5% pre-determined significance threshold. Therefore, the probability of observing values as extreme as our intake conditionally on the null hypothesis being true is too low, and we reject the null hypothesis of no deviation. set_state can only be used with legacy MT19937state instances. rayleigh(scale=1.0, size=None) Draw samples from a Rayleigh distribution. The :math:`\chi` and Weibull distributions are generalizations of the Rayleigh. .. note:: New code should use the `~numpy.random.Generator.rayleigh` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. Must be non-negative. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Rayleigh distribution. See Also -------- random.Generator.rayleigh: which should be used for new code. Notes ----- The probability density function for the Rayleigh distribution is .. math:: P(x;scale) = \frac{x}{scale^2}e^{\frac{-x^2}{2 \cdotp scale^2}} The Rayleigh distribution would arise, for example, if the East and North components of the wind velocity had identical zero-mean Gaussian distributions. Then the wind speed would have a Rayleigh distribution. References ---------- .. [1] Brighton Webs Ltd., "Rayleigh Distribution," https://web.archive.org/web/20090514091424/http://brighton-webs.co.uk:80/distributions/rayleigh.asp .. [2] Wikipedia, "Rayleigh distribution" https://en.wikipedia.org/wiki/Rayleigh_distribution Examples -------- Draw values from the distribution and plot the histogram >>> from matplotlib.pyplot import hist >>> values = hist(np.random.rayleigh(3, 100000), bins=200, density=True) Wave heights tend to follow a Rayleigh distribution. If the mean wave height is 1 meter, what fraction of waves are likely to be larger than 3 meters? >>> meanvalue = 1 >>> modevalue = np.sqrt(2 / np.pi) * meanvalue >>> s = np.random.rayleigh(modevalue, 1000000) The percentage of waves larger than 3 meters is: >>> 100.*sum(s>3)/1000000. 0.087300000000000003 # random random_integers(low, high=None, size=None) Random integers of type `np.int_` between `low` and `high`, inclusive. Return random integers of type `np.int_` from the "discrete uniform" distribution in the closed interval [`low`, `high`]. If `high` is None (the default), then results are from [1, `low`]. The `np.int_` type translates to the C long integer type and its precision is platform dependent. This function has been deprecated. Use randint instead. .. deprecated:: 1.11.0 Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution (unless ``high=None``, in which case this parameter is the *highest* such integer). high : int, optional If provided, the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- randint : Similar to `random_integers`, only for the half-open interval [`low`, `high`), and 0 is the lowest value if `high` is omitted. Notes ----- To sample from N evenly spaced floating-point numbers between a and b, use:: a + (b - a) * (np.random.random_integers(N) - 1) / (N - 1.) Examples -------- >>> np.random.random_integers(5) 4 # random >>> type(np.random.random_integers(5)) >>> np.random.random_integers(5, size=(3,2)) array([[5, 4], # random [3, 3], [4, 5]]) Choose five random numbers from the set of five evenly-spaced numbers between 0 and 2.5, inclusive (*i.e.*, from the set :math:`{0, 5/8, 10/8, 15/8, 20/8}`): >>> 2.5 * (np.random.random_integers(5, size=(5,)) - 1) / 4. array([ 0.625, 1.25 , 0.625, 0.625, 2.5 ]) # random Roll two six sided dice 1000 times and sum the results: >>> d1 = np.random.random_integers(1, 6, 1000) >>> d2 = np.random.random_integers(1, 6, 1000) >>> dsums = d1 + d2 Display results as a histogram: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(dsums, 11, density=True) >>> plt.show() randint(low, high=None, size=None, dtype=int) Return random integers from `low` (inclusive) to `high` (exclusive). Return random integers from the "discrete uniform" distribution of the specified dtype in the "half-open" interval [`low`, `high`). If `high` is None (the default), then results are from [0, `low`). .. note:: New code should use the `~numpy.random.Generator.integers` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- low : int or array-like of ints Lowest (signed) integers to be drawn from the distribution (unless ``high=None``, in which case this parameter is one above the *highest* such integer). high : int or array-like of ints, optional If provided, one above the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). If array-like, must contain integer values size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result. Byteorder must be native. The default value is int. .. versionadded:: 1.11.0 Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- random_integers : similar to `randint`, only for the closed interval [`low`, `high`], and 1 is the lowest value if `high` is omitted. random.Generator.integers: which should be used for new code. Examples -------- >>> np.random.randint(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) # random >>> np.random.randint(1, size=10) array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0]) Generate a 2 x 4 array of ints between 0 and 4, inclusive: >>> np.random.randint(5, size=(2, 4)) array([[4, 0, 2, 1], # random [3, 2, 2, 0]]) Generate a 1 x 3 array with 3 different upper bounds >>> np.random.randint(1, [3, 5, 10]) array([2, 2, 9]) # random Generate a 1 by 3 array with 3 different lower bounds >>> np.random.randint([1, 5, 7], 10) array([9, 8, 7]) # random Generate a 2 by 4 array using broadcasting with dtype of uint8 >>> np.random.randint([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7], # random [ 1, 16, 9, 12]], dtype=uint8) power(a, size=None) Draws samples in [0, 1] from a power distribution with positive exponent a - 1. Also known as the power function distribution. .. note:: New code should use the `~numpy.random.Generator.power` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Parameter of the distribution. Must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized power distribution. Raises ------ ValueError If a <= 0. See Also -------- random.Generator.power: which should be used for new code. Notes ----- The probability density function is .. math:: P(x; a) = ax^{a-1}, 0 \le x \le 1, a>0. The power function distribution is just the inverse of the Pareto distribution. It may also be seen as a special case of the Beta distribution. It is used, for example, in modeling the over-reporting of insurance claims. References ---------- .. [1] Christian Kleiber, Samuel Kotz, "Statistical size distributions in economics and actuarial sciences", Wiley, 2003. .. [2] Heckert, N. A. and Filliben, James J. "NIST Handbook 148: Dataplot Reference Manual, Volume 2: Let Subcommands and Library Functions", National Institute of Standards and Technology Handbook Series, June 2003. https://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> samples = 1000 >>> s = np.random.power(a, samples) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=30) >>> x = np.linspace(0, 1, 100) >>> y = a*x**(a-1.) >>> normed_y = samples*np.diff(bins)[0]*y >>> plt.plot(x, normed_y) >>> plt.show() Compare the power function distribution to the inverse of the Pareto. >>> from scipy import stats # doctest: +SKIP >>> rvs = np.random.power(5, 1000000) >>> rvsp = np.random.pareto(5, 1000000) >>> xx = np.linspace(0,1,100) >>> powpdf = stats.powerlaw.pdf(xx,5) # doctest: +SKIP >>> plt.figure() >>> plt.hist(rvs, bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('np.random.power(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('inverse of 1 + np.random.pareto(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('inverse of stats.pareto(5)') pareto(a, size=None) Draw samples from a Pareto II or Lomax distribution with specified shape. The Lomax or Pareto II distribution is a shifted Pareto distribution. The classical Pareto distribution can be obtained from the Lomax distribution by adding 1 and multiplying by the scale parameter ``m`` (see Notes). The smallest value of the Lomax distribution is zero while for the classical Pareto distribution it is ``mu``, where the standard Pareto distribution has location ``mu = 1``. Lomax can also be considered as a simplified version of the Generalized Pareto distribution (available in SciPy), with the scale set to one and the location set to zero. The Pareto distribution must be greater than zero, and is unbounded above. It is also known as the "80-20 rule". In this distribution, 80 percent of the weights are in the lowest 20 percent of the range, while the other 20 percent fill the remaining 80 percent of the range. .. note:: New code should use the `~numpy.random.Generator.pareto` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape of the distribution. Must be positive. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Pareto distribution. See Also -------- scipy.stats.lomax : probability density function, distribution or cumulative density function, etc. scipy.stats.genpareto : probability density function, distribution or cumulative density function, etc. random.Generator.pareto: which should be used for new code. Notes ----- The probability density for the Pareto distribution is .. math:: p(x) = \frac{am^a}{x^{a+1}} where :math:`a` is the shape and :math:`m` the scale. The Pareto distribution, named after the Italian economist Vilfredo Pareto, is a power law probability distribution useful in many real world problems. Outside the field of economics it is generally referred to as the Bradford distribution. Pareto developed the distribution to describe the distribution of wealth in an economy. It has also found use in insurance, web page access statistics, oil field sizes, and many other problems, including the download frequency for projects in Sourceforge [1]_. It is one of the so-called "fat-tailed" distributions. References ---------- .. [1] Francis Hunt and Paul Johnson, On the Pareto Distribution of Sourceforge projects. .. [2] Pareto, V. (1896). Course of Political Economy. Lausanne. .. [3] Reiss, R.D., Thomas, M.(2001), Statistical Analysis of Extreme Values, Birkhauser Verlag, Basel, pp 23-30. .. [4] Wikipedia, "Pareto distribution", https://en.wikipedia.org/wiki/Pareto_distribution Examples -------- Draw samples from the distribution: >>> a, m = 3., 2. # shape and mode >>> s = (np.random.pareto(a, 1000) + 1) * m Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, _ = plt.hist(s, 100, density=True) >>> fit = a*m**a / bins**(a+1) >>> plt.plot(bins, max(count)*fit/max(fit), linewidth=2, color='r') >>> plt.show() numpy.core.multiarray failed to import normal(loc=0.0, scale=1.0, size=None) Draw random samples from a normal (Gaussian) distribution. The probability density function of the normal distribution, first derived by De Moivre and 200 years later by both Gauss and Laplace independently [2]_, is often called the bell curve because of its characteristic shape (see the example below). The normal distributions occurs often in nature. For example, it describes the commonly occurring distribution of samples influenced by a large number of tiny, random disturbances, each with its own unique distribution [2]_. .. note:: New code should use the `~numpy.random.Generator.normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats Mean ("centre") of the distribution. scale : float or array_like of floats Standard deviation (spread or "width") of the distribution. Must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized normal distribution. See Also -------- scipy.stats.norm : probability density function, distribution or cumulative density function, etc. random.Generator.normal: which should be used for new code. Notes ----- The probability density for the Gaussian distribution is .. math:: p(x) = \frac{1}{\sqrt{ 2 \pi \sigma^2 }} e^{ - \frac{ (x - \mu)^2 } {2 \sigma^2} }, where :math:`\mu` is the mean and :math:`\sigma` the standard deviation. The square of the standard deviation, :math:`\sigma^2`, is called the variance. The function has its peak at the mean, and its "spread" increases with the standard deviation (the function reaches 0.607 times its maximum at :math:`x + \sigma` and :math:`x - \sigma` [2]_). This implies that normal is more likely to return samples lying close to the mean, rather than those far away. References ---------- .. [1] Wikipedia, "Normal distribution", https://en.wikipedia.org/wiki/Normal_distribution .. [2] P. R. Peebles Jr., "Central Limit Theorem" in "Probability, Random Variables and Random Signal Principles", 4th ed., 2001, pp. 51, 51, 125. Examples -------- Draw samples from the distribution: >>> mu, sigma = 0, 0.1 # mean and standard deviation >>> s = np.random.normal(mu, sigma, 1000) Verify the mean and the variance: >>> abs(mu - np.mean(s)) 0.0 # may vary >>> abs(sigma - np.std(s, ddof=1)) 0.1 # may vary Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=True) >>> plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) * ... np.exp( - (bins - mu)**2 / (2 * sigma**2) ), ... linewidth=2, color='r') >>> plt.show() Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> np.random.normal(3, 2.5, size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random noncentral_f(dfnum, dfden, nonc, size=None) Draw samples from the noncentral F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters > 1. `nonc` is the non-centrality parameter. .. note:: New code should use the `~numpy.random.Generator.noncentral_f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Numerator degrees of freedom, must be > 0. .. versionchanged:: 1.14.0 Earlier NumPy versions required dfnum > 1. dfden : float or array_like of floats Denominator degrees of freedom, must be > 0. nonc : float or array_like of floats Non-centrality parameter, the sum of the squares of the numerator means, must be >= 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum``, ``dfden``, and ``nonc`` are all scalars. Otherwise, ``np.broadcast(dfnum, dfden, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral Fisher distribution. See Also -------- random.Generator.noncentral_f: which should be used for new code. Notes ----- When calculating the power of an experiment (power = probability of rejecting the null hypothesis when a specific alternative is true) the non-central F statistic becomes important. When the null hypothesis is true, the F statistic follows a central F distribution. When the null hypothesis is not true, then it follows a non-central F statistic. References ---------- .. [1] Weisstein, Eric W. "Noncentral F-Distribution." From MathWorld--A Wolfram Web Resource. 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. >>> dfnum = 3 # between group deg of freedom >>> dfden = 20 # within groups degrees of freedom >>> nonc = 3.0 >>> nc_vals = np.random.noncentral_f(dfnum, dfden, nonc, 1000000) >>> NF = np.histogram(nc_vals, bins=50, density=True) >>> c_vals = np.random.f(dfnum, dfden, 1000000) >>> F = np.histogram(c_vals, bins=50, density=True) >>> import matplotlib.pyplot as plt >>> plt.plot(F[1][1:], F[0]) >>> plt.plot(NF[1][1:], NF[0]) >>> plt.show() noncentral_chisquare(df, nonc, size=None) Draw samples from a noncentral chi-square distribution. The noncentral :math:`\chi^2` distribution is a generalization of the :math:`\chi^2` distribution. .. note:: New code should use the `~numpy.random.Generator.noncentral_chisquare` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must be > 0. .. versionchanged:: 1.10.0 Earlier NumPy versions required dfnum > 1. nonc : float or array_like of floats Non-centrality, must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` and ``nonc`` are both scalars. Otherwise, ``np.broadcast(df, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral chi-square distribution. See Also -------- random.Generator.noncentral_chisquare: which should be used for new code. Notes ----- The probability density function for the noncentral Chi-square distribution is .. math:: P(x;df,nonc) = \sum^{\infty}_{i=0} \frac{e^{-nonc/2}(nonc/2)^{i}}{i!} P_{Y_{df+2i}}(x), where :math:`Y_{q}` is the Chi-square with q degrees of freedom. References ---------- .. [1] Wikipedia, "Noncentral chi-squared distribution" https://en.wikipedia.org/wiki/Noncentral_chi-squared_distribution Examples -------- Draw values from the distribution and plot the histogram >>> import matplotlib.pyplot as plt >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, density=True) >>> plt.show() Draw values from a noncentral chisquare with very small noncentrality, and compare to a chisquare. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, .0000001, 100000), ... bins=np.arange(0., 25, .1), density=True) >>> values2 = plt.hist(np.random.chisquare(3, 100000), ... bins=np.arange(0., 25, .1), density=True) >>> plt.plot(values[1][0:-1], values[0]-values2[0], 'ob') >>> plt.show() Demonstrate how large values of non-centrality lead to a more symmetric distribution. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, density=True) >>> plt.show() negative_binomial(n, p, size=None) Draw samples from a negative binomial distribution. Samples are drawn from a negative binomial distribution with specified parameters, `n` successes and `p` probability of success where `n` is > 0 and `p` is in the interval [0, 1]. .. note:: New code should use the `~numpy.random.Generator.negative_binomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : float or array_like of floats Parameter of the distribution, > 0. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized negative binomial distribution, where each sample is equal to N, the number of failures that occurred before a total of n successes was reached. See Also -------- random.Generator.negative_binomial: which should be used for new code. Notes ----- The probability mass function of the negative binomial distribution is .. math:: P(N;n,p) = \frac{\Gamma(N+n)}{N!\Gamma(n)}p^{n}(1-p)^{N}, where :math:`n` is the number of successes, :math:`p` is the probability of success, :math:`N+n` is the number of trials, and :math:`\Gamma` is the gamma function. When :math:`n` is an integer, :math:`\frac{\Gamma(N+n)}{N!\Gamma(n)} = \binom{N+n-1}{N}`, which is the more common form of this term in the pmf. The negative binomial distribution gives the probability of N failures given n successes, with a success on the last trial. If one throws a die repeatedly until the third time a "1" appears, then the probability distribution of the number of non-"1"s that appear before the third "1" is a negative binomial distribution. References ---------- .. [1] Weisstein, Eric W. "Negative Binomial Distribution." From MathWorld--A Wolfram Web Resource. 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.negative_binomial(1, 0.1, 100000) >>> for i in range(1, 11): # doctest: +SKIP ... probability = sum(s>> mean = [0, 0] >>> cov = [[1, 0], [0, 100]] # diagonal covariance Diagonal covariance means that points are oriented along x or y-axis: >>> import matplotlib.pyplot as plt >>> x, y = np.random.multivariate_normal(mean, cov, 5000).T >>> plt.plot(x, y, 'x') >>> plt.axis('equal') >>> plt.show() Note that the covariance matrix must be positive semidefinite (a.k.a. nonnegative-definite). Otherwise, the behavior of this method is undefined and backwards compatibility is not guaranteed. References ---------- .. [1] Papoulis, A., "Probability, Random Variables, and Stochastic Processes," 3rd ed., New York: McGraw-Hill, 1991. .. [2] Duda, R. O., Hart, P. E., and Stork, D. G., "Pattern Classification," 2nd ed., New York: Wiley, 2001. Examples -------- >>> mean = (1, 2) >>> cov = [[1, 0], [0, 1]] >>> x = np.random.multivariate_normal(mean, cov, (3, 3)) >>> x.shape (3, 3, 2) Here we generate 800 samples from the bivariate normal distribution with mean [0, 0] and covariance matrix [[6, -3], [-3, 3.5]]. The expected variances of the first and second components of the sample are 6 and 3.5, respectively, and the expected correlation coefficient is -3/sqrt(6*3.5) ≈ -0.65465. >>> cov = np.array([[6, -3], [-3, 3.5]]) >>> pts = np.random.multivariate_normal([0, 0], cov, size=800) Check that the mean, covariance, and correlation coefficient of the sample are close to the expected values: >>> pts.mean(axis=0) array([ 0.0326911 , -0.01280782]) # may vary >>> np.cov(pts.T) array([[ 5.96202397, -2.85602287], [-2.85602287, 3.47613949]]) # may vary >>> np.corrcoef(pts.T)[0, 1] -0.6273591314603949 # may vary We can visualize this data with a scatter plot. The orientation of the point cloud illustrates the negative correlation of the components of this sample. >>> import matplotlib.pyplot as plt >>> plt.plot(pts[:, 0], pts[:, 1], '.', alpha=0.5) >>> plt.axis('equal') >>> plt.grid() >>> plt.show() multinomial(n, pvals, size=None) Draw samples from a multinomial distribution. The multinomial distribution is a multivariate generalization of the binomial distribution. Take an experiment with one of ``p`` possible outcomes. An example of such an experiment is throwing a dice, where the outcome can be 1 through 6. Each sample drawn from the distribution represents `n` such experiments. Its values, ``X_i = [X_0, X_1, ..., X_p]``, represent the number of times the outcome was ``i``. .. note:: New code should use the `~numpy.random.Generator.multinomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : int Number of experiments. pvals : sequence of floats, length p Probabilities of each of the ``p`` different outcomes. These must sum to 1 (however, the last element is always assumed to account for the remaining probability, as long as ``sum(pvals[:-1]) <= 1)``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. See Also -------- random.Generator.multinomial: which should be used for new code. Examples -------- Throw a dice 20 times: >>> np.random.multinomial(20, [1/6.]*6, size=1) array([[4, 1, 7, 5, 2, 1]]) # random It landed 4 times on 1, once on 2, etc. Now, throw the dice 20 times, and 20 times again: >>> np.random.multinomial(20, [1/6.]*6, size=2) array([[3, 4, 3, 3, 4, 3], # random [2, 4, 3, 4, 0, 7]]) For the first run, we threw 3 times 1, 4 times 2, etc. For the second, we threw 2 times 1, 4 times 2, etc. A loaded die is more likely to land on number 6: >>> np.random.multinomial(100, [1/7.]*5 + [2/7.]) array([11, 16, 14, 17, 16, 26]) # random The probability inputs should be normalized. As an implementation detail, the value of the last entry is ignored and assumed to take up any leftover probability mass, but this should not be relied on. A biased coin which has twice as much weight on one side as on the other should be sampled like so: >>> np.random.multinomial(100, [1.0 / 3, 2.0 / 3]) # RIGHT array([38, 62]) # random not like: >>> np.random.multinomial(100, [1.0, 2.0]) # WRONG Traceback (most recent call last): ValueError: pvals < 0, pvals > 1 or pvals contains NaNs logseries(p, size=None) Draw samples from a logarithmic series distribution. Samples are drawn from a log series distribution with specified shape parameter, 0 <= ``p`` < 1. .. note:: New code should use the `~numpy.random.Generator.logseries` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- p : float or array_like of floats Shape parameter for the distribution. Must be in the range [0, 1). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logarithmic series distribution. See Also -------- scipy.stats.logser : probability density function, distribution or cumulative density function, etc. random.Generator.logseries: which should be used for new code. Notes ----- The probability density for the Log Series distribution is .. math:: P(k) = \frac{-p^k}{k \ln(1-p)}, where p = probability. The log series distribution is frequently used to represent species richness and occurrence, first proposed by Fisher, Corbet, and Williams in 1943 [2]. It may also be used to model the numbers of occupants seen in cars [3]. References ---------- .. [1] Buzas, Martin A.; Culver, Stephen J., Understanding regional species diversity through the log series distribution of occurrences: BIODIVERSITY RESEARCH Diversity & Distributions, Volume 5, Number 5, September 1999 , pp. 187-195(9). .. [2] Fisher, R.A,, A.S. Corbet, and C.B. Williams. 1943. The relation between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology, 12:42-58. .. [3] D. J. Hand, F. Daly, D. Lunn, E. Ostrowski, A Handbook of Small Data Sets, CRC Press, 1994. .. [4] Wikipedia, "Logarithmic distribution", https://en.wikipedia.org/wiki/Logarithmic_distribution Examples -------- Draw samples from the distribution: >>> a = .6 >>> s = np.random.logseries(a, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s) # plot against distribution >>> def logseries(k, p): ... return -p**k/(k*np.log(1-p)) >>> plt.plot(bins, logseries(bins, a)*count.max()/ ... logseries(bins, a).max(), 'r') >>> plt.show() lognormal(mean=0.0, sigma=1.0, size=None) Draw samples from a log-normal distribution. Draw samples from a log-normal distribution with specified mean, standard deviation, and array shape. Note that the mean and standard deviation are not the values for the distribution itself, but of the underlying normal distribution it is derived from. .. note:: New code should use the `~numpy.random.Generator.lognormal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : float or array_like of floats, optional Mean value of the underlying normal distribution. Default is 0. sigma : float or array_like of floats, optional Standard deviation of the underlying normal distribution. Must be non-negative. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``sigma`` are both scalars. Otherwise, ``np.broadcast(mean, sigma).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized log-normal distribution. See Also -------- scipy.stats.lognorm : probability density function, distribution, cumulative density function, etc. random.Generator.lognormal: which should be used for new code. Notes ----- A variable `x` has a log-normal distribution if `log(x)` is normally distributed. The probability density function for the log-normal distribution is: .. math:: p(x) = \frac{1}{\sigma x \sqrt{2\pi}} e^{(-\frac{(ln(x)-\mu)^2}{2\sigma^2})} where :math:`\mu` is the mean and :math:`\sigma` is the standard deviation of the normally distributed logarithm of the variable. A log-normal distribution results if a random variable is the *product* of a large number of independent, identically-distributed variables in the same way that a normal distribution results if the variable is the *sum* of a large number of independent, identically-distributed variables. References ---------- .. [1] Limpert, E., Stahel, W. A., and Abbt, M., "Log-normal Distributions across the Sciences: Keys and Clues," BioScience, Vol. 51, No. 5, May, 2001. https://stat.ethz.ch/~stahel/lognormal/bioscience.pdf .. [2] Reiss, R.D. and Thomas, M., "Statistical Analysis of Extreme Values," Basel: Birkhauser Verlag, 2001, pp. 31-32. Examples -------- Draw samples from the distribution: >>> mu, sigma = 3., 1. # mean and standard deviation >>> s = np.random.lognormal(mu, sigma, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 100, density=True, align='mid') >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, linewidth=2, color='r') >>> plt.axis('tight') >>> plt.show() Demonstrate that taking the products of random samples from a uniform distribution can be fit well by a log-normal probability density function. >>> # Generate a thousand samples: each is the product of 100 random >>> # values, drawn from a normal distribution. >>> b = [] >>> for i in range(1000): ... a = 10. + np.random.standard_normal(100) ... b.append(np.prod(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, density=True, align='mid') >>> sigma = np.std(np.log(b)) >>> mu = np.mean(np.log(b)) >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, color='r', linewidth=2) >>> plt.show() logistic(loc=0.0, scale=1.0, size=None) Draw samples from a logistic distribution. Samples are drawn from a logistic distribution with specified parameters, loc (location or mean, also median), and scale (>0). .. note:: New code should use the `~numpy.random.Generator.logistic` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats, optional Parameter of the distribution. Default is 0. scale : float or array_like of floats, optional Parameter of the distribution. Must be non-negative. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logistic distribution. See Also -------- scipy.stats.logistic : probability density function, distribution or cumulative density function, etc. random.Generator.logistic: which should be used for new code. Notes ----- The probability density for the Logistic distribution is .. math:: P(x) = P(x) = \frac{e^{-(x-\mu)/s}}{s(1+e^{-(x-\mu)/s})^2}, where :math:`\mu` = location and :math:`s` = scale. The Logistic distribution is used in Extreme Value problems where it can act as a mixture of Gumbel distributions, in Epidemiology, and by the World Chess Federation (FIDE) where it is used in the Elo ranking system, assuming the performance of each player is a logistically distributed random variable. References ---------- .. [1] Reiss, R.-D. and Thomas M. (2001), "Statistical Analysis of Extreme Values, from Insurance, Finance, Hydrology and Other Fields," Birkhauser Verlag, Basel, pp 132-133. .. [2] Weisstein, Eric W. "Logistic Distribution." From MathWorld--A Wolfram Web Resource. 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.logistic(loc, scale, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=50) # plot against distribution >>> def logist(x, loc, scale): ... return np.exp((loc-x)/scale)/(scale*(1+np.exp((loc-x)/scale))**2) >>> lgst_val = logist(bins, loc, scale) >>> plt.plot(bins, lgst_val * count.max() / lgst_val.max()) >>> plt.show() hypergeometric(ngood, nbad, nsample, size=None) Draw samples from a Hypergeometric distribution. Samples are drawn from a hypergeometric distribution with specified parameters, `ngood` (ways to make a good selection), `nbad` (ways to make a bad selection), and `nsample` (number of items sampled, which is less than or equal to the sum ``ngood + nbad``). .. note:: New code should use the `~numpy.random.Generator.hypergeometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- ngood : int or array_like of ints Number of ways to make a good selection. Must be nonnegative. nbad : int or array_like of ints Number of ways to make a bad selection. Must be nonnegative. nsample : int or array_like of ints Number of items sampled. Must be at least 1 and at most ``ngood + nbad``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if `ngood`, `nbad`, and `nsample` are all scalars. Otherwise, ``np.broadcast(ngood, nbad, nsample).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized hypergeometric distribution. Each sample is the number of good items within a randomly selected subset of size `nsample` taken from a set of `ngood` good items and `nbad` bad items. See Also -------- scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. random.Generator.hypergeometric: which should be used for new code. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{g}{x}\binom{b}{n-x}}{\binom{g+b}{n}}, where :math:`0 \le x \le n` and :math:`n-b \le x \le g` for P(x) the probability of ``x`` good results in the drawn sample, g = `ngood`, b = `nbad`, and n = `nsample`. Consider an urn with black and white marbles in it, `ngood` of them are black and `nbad` are white. If you draw `nsample` balls without replacement, then the hypergeometric distribution describes the distribution of black balls in the drawn sample. Note that this distribution is very similar to the binomial distribution, except that in this case, samples are drawn without replacement, whereas in the Binomial case samples are drawn with replacement (or the sample space is infinite). As the sample space becomes large, this distribution approaches the binomial. References ---------- .. [1] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [2] Weisstein, Eric W. "Hypergeometric Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", https://en.wikipedia.org/wiki/Hypergeometric_distribution Examples -------- Draw samples from the distribution: >>> ngood, nbad, nsamp = 100, 2, 10 # number of good, number of bad, and number of samples >>> s = np.random.hypergeometric(ngood, nbad, nsamp, 1000) >>> from matplotlib.pyplot import hist >>> hist(s) # note that it is very unlikely to grab both bad items Suppose you have an urn with 15 white and 15 black marbles. If you pull 15 marbles at random, how likely is it that 12 or more of them are one color? >>> s = np.random.hypergeometric(15, 15, 15, 100000) >>> sum(s>=12)/100000. + sum(s<=3)/100000. # answer = 0.003 ... pretty unlikely! gumbel(loc=0.0, scale=1.0, size=None) Draw samples from a Gumbel distribution. Draw samples from a Gumbel distribution with specified location and scale. For more information on the Gumbel distribution, see Notes and References below. .. note:: New code should use the `~numpy.random.Generator.gumbel` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats, optional The location of the mode of the distribution. Default is 0. scale : float or array_like of floats, optional The scale parameter of the distribution. Default is 1. Must be non- negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Gumbel distribution. See Also -------- scipy.stats.gumbel_l scipy.stats.gumbel_r scipy.stats.genextreme weibull random.Generator.gumbel: which should be used for new code. Notes ----- The Gumbel (or Smallest Extreme Value (SEV) or the Smallest Extreme Value Type I) distribution is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. The Gumbel is a special case of the Extreme Value Type I distribution for maximums from distributions with "exponential-like" tails. The probability density for the Gumbel distribution is .. math:: p(x) = \frac{e^{-(x - \mu)/ \beta}}{\beta} e^{ -e^{-(x - \mu)/ \beta}}, where :math:`\mu` is the mode, a location parameter, and :math:`\beta` is the scale parameter. The Gumbel (named for German mathematician Emil Julius Gumbel) was used very early in the hydrology literature, for modeling the occurrence of flood events. It is also used for modeling maximum wind speed and rainfall rates. It is a "fat-tailed" distribution - the probability of an event in the tail of the distribution is larger than if one used a Gaussian, hence the surprisingly frequent occurrence of 100-year floods. Floods were initially modeled as a Gaussian process, which underestimated the frequency of extreme events. It is one of a class of extreme value distributions, the Generalized Extreme Value (GEV) distributions, which also includes the Weibull and Frechet. The function has a mean of :math:`\mu + 0.57721\beta` and a variance of :math:`\frac{\pi^2}{6}\beta^2`. References ---------- .. [1] Gumbel, E. J., "Statistics of Extremes," New York: Columbia University Press, 1958. .. [2] Reiss, R.-D. and Thomas, M., "Statistical Analysis of Extreme Values from Insurance, Finance, Hydrology and Other Fields," Basel: Birkhauser Verlag, 2001. Examples -------- Draw samples from the distribution: >>> mu, beta = 0, 0.1 # location and scale >>> s = np.random.gumbel(mu, beta, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=True) >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp( -np.exp( -(bins - mu) /beta) ), ... linewidth=2, color='r') >>> plt.show() Show how an extreme value distribution can arise from a Gaussian process and compare to a Gaussian: >>> means = [] >>> maxima = [] >>> for i in range(0,1000) : ... a = np.random.normal(mu, beta, 1000) ... means.append(a.mean()) ... maxima.append(a.max()) >>> count, bins, ignored = plt.hist(maxima, 30, density=True) >>> beta = np.std(maxima) * np.sqrt(6) / np.pi >>> mu = np.mean(maxima) - 0.57721*beta >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp(-np.exp(-(bins - mu)/beta)), ... linewidth=2, color='r') >>> plt.plot(bins, 1/(beta * np.sqrt(2 * np.pi)) ... * np.exp(-(bins - mu)**2 / (2 * beta**2)), ... linewidth=2, color='g') >>> plt.show() geometric(p, size=None) Draw samples from the geometric distribution. Bernoulli trials are experiments with one of two outcomes: success or failure (an example of such an experiment is flipping a coin). The geometric distribution models the number of trials that must be run in order to achieve success. It is therefore supported on the positive integers, ``k = 1, 2, ...``. The probability mass function of the geometric distribution is .. math:: f(k) = (1 - p)^{k - 1} p where `p` is the probability of success of an individual trial. .. note:: New code should use the `~numpy.random.Generator.geometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- p : float or array_like of floats The probability of success of an individual trial. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized geometric distribution. See Also -------- random.Generator.geometric: which should be used for new code. Examples -------- Draw ten thousand values from the geometric distribution, with the probability of an individual success equal to 0.35: >>> z = np.random.geometric(p=0.35, size=10000) How many trials succeeded after a single run? >>> (z == 1).sum() / 10000. 0.34889999999999999 #random gamma(shape, scale=1.0, size=None) Draw samples from a Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, `shape` (sometimes designated "k") and `scale` (sometimes designated "theta"), where both parameters are > 0. .. note:: New code should use the `~numpy.random.Generator.gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. Must be non-negative. scale : float or array_like of floats, optional The scale of the gamma distribution. Must be non-negative. Default is equal to 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(shape, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. random.Generator.gamma: which should be used for new code. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. 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.gamma(shape, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1)*(np.exp(-bins/scale) / # doctest: +SKIP ... (sps.gamma(shape)*scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() f(dfnum, dfden, size=None) Draw samples from an F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters must be greater than zero. The random variate of the F distribution (also known as the Fisher distribution) is a continuous probability distribution that arises in ANOVA tests, and is the ratio of two chi-square variates. .. note:: New code should use the `~numpy.random.Generator.f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Degrees of freedom in numerator, must be > 0. dfden : float or array_like of float Degrees of freedom in denominator, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum`` and ``dfden`` are both scalars. Otherwise, ``np.broadcast(dfnum, dfden).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Fisher distribution. See Also -------- scipy.stats.f : probability density function, distribution or cumulative density function, etc. random.Generator.f: which should be used for new code. Notes ----- The F statistic is used to compare in-group variances to between-group variances. Calculating the distribution depends on the sampling, and so it is a function of the respective degrees of freedom in the problem. The variable `dfnum` is the number of samples minus one, the between-groups degrees of freedom, while `dfden` is the within-groups degrees of freedom, the sum of the number of samples in each group minus the number of groups. References ---------- .. [1] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [2] Wikipedia, "F-distribution", https://en.wikipedia.org/wiki/F-distribution Examples -------- An example from Glantz[1], pp 47-40: Two groups, children of diabetics (25 people) and children from people without diabetes (25 controls). Fasting blood glucose was measured, case group had a mean value of 86.1, controls had a mean value of 82.2. Standard deviations were 2.09 and 2.49 respectively. Are these data consistent with the null hypothesis that the parents diabetic status does not affect their children's blood glucose levels? Calculating the F statistic from the data gives a value of 36.01. Draw samples from the distribution: >>> dfnum = 1. # between group degrees of freedom >>> dfden = 48. # within groups degrees of freedom >>> s = np.random.f(dfnum, dfden, 1000) The lower bound for the top 1% of the samples is : >>> np.sort(s)[-10] 7.61988120985 # random So there is about a 1% chance that the F statistic will exceed 7.62, the measured value is 36, so the null hypothesis is rejected at the 1% level. exponential(scale=1.0, size=None) Draw samples from an exponential distribution. Its probability density function is .. math:: f(x; \frac{1}{\beta}) = \frac{1}{\beta} \exp(-\frac{x}{\beta}), for ``x > 0`` and 0 elsewhere. :math:`\beta` is the scale parameter, which is the inverse of the rate parameter :math:`\lambda = 1/\beta`. The rate parameter is an alternative, widely used parameterization of the exponential distribution [3]_. The exponential distribution is a continuous analogue of the geometric distribution. It describes many common situations, such as the size of raindrops measured over many rainstorms [1]_, or the time between page requests to Wikipedia [2]_. .. note:: New code should use the `~numpy.random.Generator.exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats The scale parameter, :math:`\beta = 1/\lambda`. Must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized exponential distribution. Examples -------- A real world example: Assume a company has 10000 customer support agents and the average time between customer calls is 4 minutes. >>> n = 10000 >>> time_between_calls = np.random.default_rng().exponential(scale=4, size=n) What is the probability that a customer will call in the next 4 to 5 minutes? >>> x = ((time_between_calls < 5).sum())/n >>> y = ((time_between_calls < 4).sum())/n >>> x-y 0.08 # may vary See Also -------- random.Generator.exponential: which should be used for new code. References ---------- .. [1] Peyton Z. Peebles Jr., "Probability, Random Variables and Random Signal Principles", 4th ed, 2001, p. 57. .. [2] Wikipedia, "Poisson process", https://en.wikipedia.org/wiki/Poisson_process .. [3] Wikipedia, "Exponential distribution", https://en.wikipedia.org/wiki/Exponential_distribution choice(a, size=None, replace=True, p=None) Generates a random sample from a given 1-D array .. versionadded:: 1.7.0 .. note:: New code should use the `~numpy.random.Generator.choice` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : 1-D array-like or int If an ndarray, a random sample is generated from its elements. If an int, the random sample is generated as if it were ``np.arange(a)`` size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. replace : boolean, optional Whether the sample is with or without replacement. Default is True, meaning that a value of ``a`` can be selected multiple times. p : 1-D array-like, optional The probabilities associated with each entry in a. If not given, the sample assumes a uniform distribution over all entries in ``a``. Returns ------- samples : single item or ndarray The generated random samples Raises ------ ValueError If a is an int and less than zero, if a or p are not 1-dimensional, if a is an array-like of size 0, if p is not a vector of probabilities, if a and p have different lengths, or if replace=False and the sample size is greater than the population size See Also -------- randint, shuffle, permutation random.Generator.choice: which should be used in new code Notes ----- Setting user-specified probabilities through ``p`` uses a more general but less efficient sampler than the default. The general sampler produces a different sample than the optimized sampler even if each element of ``p`` is 1 / len(a). Sampling random rows from a 2-D array is not possible with this function, but is possible with `Generator.choice` through its ``axis`` keyword. Examples -------- Generate a uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3) array([0, 3, 4]) # random >>> #This is equivalent to np.random.randint(0,5,3) Generate a non-uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0]) array([3, 3, 0]) # random Generate a uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False) array([3,1,0]) # random >>> #This is equivalent to np.random.permutation(np.arange(5))[:3] Generate a non-uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0]) array([2, 3, 0]) # random Any of the above can be repeated with an arbitrary array-like instead of just integers. For instance: >>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher'] >>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3]) array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'], # random dtype=' 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized chi-square distribution. Raises ------ ValueError When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``) is given. See Also -------- random.Generator.chisquare: which should be used for new code. Notes ----- The variable obtained by summing the squares of `df` independent, standard normally distributed random variables: .. math:: Q = \sum_{i=0}^{\mathtt{df}} X^2_i is chi-square distributed, denoted .. math:: Q \sim \chi^2_k. The probability density function of the chi-squared distribution is .. math:: p(x) = \frac{(1/2)^{k/2}}{\Gamma(k/2)} x^{k/2 - 1} e^{-x/2}, where :math:`\Gamma` is the gamma function, .. math:: \Gamma(x) = \int_0^{-\infty} t^{x - 1} e^{-t} dt. References ---------- .. [1] NIST "Engineering Statistics Handbook" https://www.itl.nist.gov/div898/handbook/eda/section3/eda3666.htm Examples -------- >>> np.random.chisquare(2,4) array([ 1.89920014, 9.00867716, 3.13710533, 5.62318272]) # random bytes(length) Return random bytes. .. note:: New code should use the `~numpy.random.Generator.bytes` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- length : int Number of random bytes. Returns ------- out : bytes String of length `length`. See Also -------- random.Generator.bytes: which should be used for new code. Examples -------- >>> np.random.bytes(10) b' eh\x85\x022SZ\xbf\xa4' #random binomial(n, p, size=None) Draw samples from a binomial distribution. Samples are drawn from a binomial distribution with specified parameters, n trials and p probability of success where n an integer >= 0 and p is in the interval [0,1]. (n may be input as a float, but it is truncated to an integer in use) .. note:: New code should use the `~numpy.random.Generator.binomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : int or array_like of ints Parameter of the distribution, >= 0. Floats are also accepted, but they will be truncated to integers. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized binomial distribution, where each sample is equal to the number of successes over the n trials. See Also -------- scipy.stats.binom : probability density function, distribution or cumulative density function, etc. random.Generator.binomial: which should be used for new code. Notes ----- The probability density for the binomial distribution is .. math:: P(N) = \binom{n}{N}p^N(1-p)^{n-N}, where :math:`n` is the number of trials, :math:`p` is the probability of success, and :math:`N` is the number of successes. When estimating the standard error of a proportion in a population by using a random sample, the normal distribution works well unless the product p*n <=5, where p = population proportion estimate, and n = number of samples, in which case the binomial distribution is used instead. For example, a sample of 15 people shows 4 who are left handed, and 11 who are right handed. Then p = 4/15 = 27%. 0.27*15 = 4, so the binomial distribution should be used in this case. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics with R", Springer-Verlag, 2002. .. [2] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [3] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [4] Weisstein, Eric W. "Binomial Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/BinomialDistribution.html .. [5] Wikipedia, "Binomial distribution", https://en.wikipedia.org/wiki/Binomial_distribution Examples -------- Draw samples from the distribution: >>> n, p = 10, .5 # number of trials, probability of each trial >>> s = np.random.binomial(n, p, 1000) # result of flipping a coin 10 times, tested 1000 times. A real world example. A company drills 9 wild-cat oil exploration wells, each with an estimated probability of success of 0.1. All nine wells fail. What is the probability of that happening? Let's do 20,000 trials of the model, and count the number that generate zero positive results. >>> sum(np.random.binomial(9, 0.1, 20000) == 0)/20000. # answer = 0.38885, or 38%. a must be greater than 0 unless no samples are taken'a' cannot be empty unless no samples are takenUnsupported dtype %r for randintRandomState.standard_gamma (line 1562)RandomState.multivariate_normal (line 4057)RandomState.logseries (line 3968)RandomState.lognormal (line 2973)RandomState.hypergeometric (line 3833)RandomState.geometric (line 3771)RandomState.dirichlet (line 4393)RandomState.chisquare (line 1909) wald(mean, scale, size=None) Draw samples from a Wald, or inverse Gaussian, distribution. As the scale approaches infinity, the distribution becomes more like a Gaussian. Some references claim that the Wald is an inverse Gaussian with mean equal to 1, but this is by no means universal. The inverse Gaussian distribution was first studied in relationship to Brownian motion. In 1956 M.C.K. Tweedie used the name inverse Gaussian because there is an inverse relationship between the time to cover a unit distance and distance covered in unit time. .. note:: New code should use the `~numpy.random.Generator.wald` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : float or array_like of floats Distribution mean, must be > 0. scale : float or array_like of floats Scale parameter, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(mean, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Wald distribution. See Also -------- random.Generator.wald: which should be used for new code. Notes ----- The probability density function for the Wald distribution is .. math:: P(x;mean,scale) = \sqrt{\frac{scale}{2\pi x^3}}e^ \frac{-scale(x-mean)^2}{2\cdotp mean^2x} As noted above the inverse Gaussian distribution first arise from attempts to model Brownian motion. It is also a competitor to the Weibull for use in reliability modeling and modeling stock returns and interest rate processes. References ---------- .. [1] Brighton Webs Ltd., Wald Distribution, https://web.archive.org/web/20090423014010/http://www.brighton-webs.co.uk:80/distributions/wald.asp .. [2] Chhikara, Raj S., and Folks, J. Leroy, "The Inverse Gaussian Distribution: Theory : Methodology, and Applications", CRC Press, 1988. .. [3] Wikipedia, "Inverse Gaussian distribution" https://en.wikipedia.org/wiki/Inverse_Gaussian_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.wald(3, 2, 100000), bins=200, density=True) >>> plt.show() standard_normal(size=None) Draw samples from a standard Normal distribution (mean=0, stdev=1). .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray A floating-point array of shape ``size`` of drawn samples, or a single sample if ``size`` was not specified. See Also -------- normal : Equivalent function with additional ``loc`` and ``scale`` arguments for setting the mean and standard deviation. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use one of:: mu + sigma * np.random.standard_normal(size=...) np.random.normal(mu, sigma, size=...) Examples -------- >>> np.random.standard_normal() 2.1923875335537315 #random >>> s = np.random.standard_normal(8000) >>> s array([ 0.6888893 , 0.78096262, -0.89086505, ..., 0.49876311, # random -0.38672696, -0.4685006 ]) # random >>> s.shape (8000,) >>> s = np.random.standard_normal(size=(3, 4, 2)) >>> s.shape (3, 4, 2) Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.standard_normal(size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random standard_gamma(shape, size=None) Draw samples from a standard Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, shape (sometimes designated "k") and scale=1. .. note:: New code should use the `~numpy.random.Generator.standard_gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats Parameter, must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` is a scalar. Otherwise, ``np.array(shape).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. random.Generator.standard_gamma: which should be used for new code. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. 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.standard_gamma(shape, 1000000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ # doctest: +SKIP ... (sps.gamma(shape) * scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() standard_exponential(size=None) Draw samples from the standard exponential distribution. `standard_exponential` is identical to the exponential distribution with a scale parameter of 1. .. note:: New code should use the `~numpy.random.Generator.standard_exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray Drawn samples. See Also -------- random.Generator.standard_exponential: which should be used for new code. Examples -------- Output a 3x8000 array: >>> n = np.random.standard_exponential((3, 8000)) standard_cauchy(size=None) Draw samples from a standard Cauchy distribution with mode = 0. Also known as the Lorentz distribution. .. note:: New code should use the `~numpy.random.Generator.standard_cauchy` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- samples : ndarray or scalar The drawn samples. See Also -------- random.Generator.standard_cauchy: which should be used for new code. Notes ----- The probability density function for the full Cauchy distribution is .. math:: P(x; x_0, \gamma) = \frac{1}{\pi \gamma \bigl[ 1+ (\frac{x-x_0}{\gamma})^2 \bigr] } and the Standard Cauchy distribution just sets :math:`x_0=0` and :math:`\gamma=1` The Cauchy distribution arises in the solution to the driven harmonic oscillator problem, and also describes spectral line broadening. It also describes the distribution of values at which a line tilted at a random angle will cut the x axis. When studying hypothesis tests that assume normality, seeing how the tests perform on data from a Cauchy distribution is a good indicator of their sensitivity to a heavy-tailed distribution, since the Cauchy looks very much like a Gaussian distribution, but with heavier tails. References ---------- .. [1] NIST/SEMATECH e-Handbook of Statistical Methods, "Cauchy Distribution", https://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm .. [2] Weisstein, Eric W. "Cauchy Distribution." From MathWorld--A Wolfram Web Resource. 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.standard_cauchy(1000000) >>> s = s[(s>-25) & (s<25)] # truncate distribution so it plots well >>> plt.hist(s, bins=100) >>> plt.show() shuffle(x) Modify a sequence in-place by shuffling its contents. This function only shuffles the array along the first axis of a multi-dimensional array. The order of sub-arrays is changed but their contents remains the same. .. note:: New code should use the `~numpy.random.Generator.shuffle` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- x : ndarray or MutableSequence The array, list or mutable sequence to be shuffled. Returns ------- None See Also -------- random.Generator.shuffle: which should be used for new code. Examples -------- >>> arr = np.arange(10) >>> np.random.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] # random Multi-dimensional arrays are only shuffled along the first axis: >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.shuffle(arr) >>> arr array([[3, 4, 5], # random [6, 7, 8], [0, 1, 2]]) seed(seed=None) Reseed a legacy MT19937 BitGenerator Notes ----- This is a convenience, legacy function. The best practice is to **not** reseed a BitGenerator, rather to recreate a new one. This method is here for legacy reasons. This example demonstrates best practice. >>> from numpy.random import MT19937 >>> from numpy.random import RandomState, SeedSequence >>> rs = RandomState(MT19937(SeedSequence(123456789))) # Later, you want to restart the stream >>> rs = RandomState(MT19937(SeedSequence(987654321))) random_sample(size=None) Return random floats in the half-open interval [0.0, 1.0). Results are from the "continuous uniform" distribution over the stated interval. To sample :math:`Unif[a, b), b > a` multiply the output of `random_sample` by `(b-a)` and add `a`:: (b - a) * random_sample() + a .. note:: New code should use the `~numpy.random.Generator.random` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray of floats Array of random floats of shape `size` (unless ``size=None``, in which case a single float is returned). See Also -------- random.Generator.random: which should be used for new code. Examples -------- >>> np.random.random_sample() 0.47108547995356098 # random >>> type(np.random.random_sample()) >>> np.random.random_sample((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) # random Three-by-two array of random numbers from [-5, 0): >>> 5 * np.random.random_sample((3, 2)) - 5 array([[-3.99149989, -0.52338984], # random [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) randn(d0, d1, ..., dn) Return a sample (or samples) from the "standard normal" distribution. .. note:: This is a convenience function for users porting code from Matlab, and wraps `standard_normal`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. If positive int_like arguments are provided, `randn` generates an array of shape ``(d0, d1, ..., dn)``, filled with random floats sampled from a univariate "normal" (Gaussian) distribution of mean 0 and variance 1. A single float randomly sampled from the distribution is returned if no argument is provided. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, must be non-negative. If no argument is given a single Python float is returned. Returns ------- Z : ndarray or float A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from the standard normal distribution, or a single such float if no parameters were supplied. See Also -------- standard_normal : Similar, but takes a tuple as its argument. normal : Also accepts mu and sigma arguments. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use:: sigma * np.random.randn(...) + mu Examples -------- >>> np.random.randn() 2.1923875335537315 # random Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.randn(2, 4) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random rand(d0, d1, ..., dn) Random values in a given shape. .. note:: This is a convenience function for users porting code from Matlab, and wraps `random_sample`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. Create an array of the given shape and populate it with random samples from a uniform distribution over ``[0, 1)``. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, must be non-negative. If no argument is given a single Python float is returned. Returns ------- out : ndarray, shape ``(d0, d1, ..., dn)`` Random values. See Also -------- random Examples -------- >>> np.random.rand(3,2) array([[ 0.14022471, 0.96360618], #random [ 0.37601032, 0.25528411], #random [ 0.49313049, 0.94909878]]) #random poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. .. note:: New code should use the `~numpy.random.Generator.poisson` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- lam : float or array_like of floats Expected number of events occurring in a fixed-time interval, must be >= 0. A sequence must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. See Also -------- random.Generator.poisson: which should be used for new code. Notes ----- The Poisson distribution .. math:: f(k; \lambda)=\frac{\lambda^k e^{-\lambda}}{k!} For events with an expected separation :math:`\lambda` the Poisson distribution :math:`f(k; \lambda)` describes the probability of :math:`k` events occurring within the observed interval :math:`\lambda`. Because the output is limited to the range of the C int64 type, a ValueError is raised when `lam` is within 10 sigma of the maximum representable value. References ---------- .. [1] Weisstein, Eric W. "Poisson Distribution." From MathWorld--A Wolfram Web Resource. 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 >>> s = np.random.poisson(5, 10000) Display histogram of the sample: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 14, density=True) >>> plt.show() Draw each 100 values for lambda 100 and 500: >>> s = np.random.poisson(lam=(100., 500.), size=(100, 2)) permutation(x) Randomly permute a sequence, or return a permuted range. If `x` is a multi-dimensional array, it is only shuffled along its first index. .. note:: New code should use the `~numpy.random.Generator.permutation` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- x : int or array_like If `x` is an integer, randomly permute ``np.arange(x)``. If `x` is an array, make a copy and shuffle the elements randomly. Returns ------- out : ndarray Permuted sequence or array range. See Also -------- random.Generator.permutation: which should be used for new code. Examples -------- >>> np.random.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) # random >>> np.random.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) # random >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.permutation(arr) array([[6, 7, 8], # random [0, 1, 2], [3, 4, 5]]) ' object which is not a subclass of 'Sequence'; `shuffle` is not guaranteed to behave correctly. E.g., non-numpy array/tensor objects with view semantics may contain duplicates after shuffling. laplace(loc=0.0, scale=1.0, size=None) Draw samples from the Laplace or double exponential distribution with specified location (or mean) and scale (decay). The Laplace distribution is similar to the Gaussian/normal distribution, but is sharper at the peak and has fatter tails. It represents the difference between two independent, identically distributed exponential random variables. .. note:: New code should use the `~numpy.random.Generator.laplace` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats, optional The position, :math:`\mu`, of the distribution peak. Default is 0. scale : float or array_like of floats, optional :math:`\lambda`, the exponential decay. Default is 1. Must be non- negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Laplace distribution. See Also -------- random.Generator.laplace: which should be used for new code. Notes ----- It has the probability density function .. math:: f(x; \mu, \lambda) = \frac{1}{2\lambda} \exp\left(-\frac{|x - \mu|}{\lambda}\right). The first law of Laplace, from 1774, states that the frequency of an error can be expressed as an exponential function of the absolute magnitude of the error, which leads to the Laplace distribution. For many problems in economics and health sciences, this distribution seems to model the data better than the standard Gaussian distribution. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] Kotz, Samuel, et. al. "The Laplace Distribution and Generalizations, " Birkhauser, 2001. .. [3] Weisstein, Eric W. "Laplace Distribution." From MathWorld--A Wolfram Web Resource. 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.laplace(loc, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=True) >>> x = np.arange(-8., 8., .01) >>> pdf = np.exp(-abs(x-loc)/scale)/(2.*scale) >>> plt.plot(x, pdf) Plot Gaussian for comparison: >>> g = (1/(scale * np.sqrt(2 * np.pi)) * ... np.exp(-(x - loc)**2 / (2 * scale**2))) >>> plt.plot(x,g) dirichlet(alpha, size=None) Draw samples from the Dirichlet distribution. Draw `size` samples of dimension k from a Dirichlet distribution. A Dirichlet-distributed random variable can be seen as a multivariate generalization of a Beta distribution. The Dirichlet distribution is a conjugate prior of a multinomial distribution in Bayesian inference. .. note:: New code should use the `~numpy.random.Generator.dirichlet` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- alpha : sequence of floats, length k Parameter of the distribution (length ``k`` for sample of length ``k``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n)``, then ``m * n * k`` samples are drawn. Default is None, in which case a vector of length ``k`` is returned. Returns ------- samples : ndarray, The drawn samples, of shape ``(size, k)``. Raises ------ ValueError If any value in ``alpha`` is less than or equal to zero See Also -------- random.Generator.dirichlet: which should be used for new code. Notes ----- The Dirichlet distribution is a distribution over vectors :math:`x` that fulfil the conditions :math:`x_i>0` and :math:`\sum_{i=1}^k x_i = 1`. The probability density function :math:`p` of a Dirichlet-distributed random vector :math:`X` is proportional to .. math:: p(x) \propto \prod_{i=1}^{k}{x^{\alpha_i-1}_i}, where :math:`\alpha` is a vector containing the positive concentration parameters. The method uses the following property for computation: let :math:`Y` be a random vector which has components that follow a standard gamma distribution, then :math:`X = \frac{1}{\sum_{i=1}^k{Y_i}} Y` is Dirichlet-distributed References ---------- .. [1] David McKay, "Information Theory, Inference and Learning Algorithms," chapter 23, 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.dirichlet((10, 5, 3), 20).transpose() >>> import matplotlib.pyplot as plt >>> plt.barh(range(20), s[0]) >>> plt.barh(range(20), s[1], left=s[0], color='g') >>> plt.barh(range(20), s[2], left=s[0]+s[1], color='r') >>> plt.title("Lengths of Strings") RandomState.vonmises (line 2264)RandomState.rayleigh (line 3089)RandomState.logistic (line 2887)RandomState.binomial (line 3352)state dictionary is not valid.probabilities do not sum to 1RandomState.weibull (line 2456)RandomState.uniform (line 1049)RandomState.tomaxint (line 620)RandomState.shuffle (line 4542)RandomState.poisson (line 3592)RandomState.laplace (line 2669)pvals must be a 1-d sequenceRandomState.randint (line 678)RandomState.pareto (line 2353)RandomState.normal (line 1453)RandomState.gumbel (line 2763)'a' and 'p' must have same sizeRandomState.randn (line 1220)RandomState.power (line 2560)RandomState.gamma (line 1644)RandomState.choice (line 841)mean must be 1 dimensionalRange exceeds valid boundsRandomState.zipf (line 3675)RandomState.wald (line 3166)RandomState.rand (line 1176)RandomState.bytes (line 804)probabilities contain NaNRandomState.seed (line 228)'p' must be 1-dimensionala must be 1-dimensionalRandomState.f (line 1728)standard_exponentialnoncentral_chisquareyou are shuffling a 'numpy.random.mtrandmultivariate_normalngood + nbad < nsamplecline_in_tracebackarray is read-onlyDeprecationWarningset_bit_generatornegative_binomialget_bit_generator__randomstate_ctormay_share_memorysum(pvals[:-1]) > 1.0standard_normalstandard_cauchyrandom_integers_poisson_lam_maxcollections.abc_bit_generatorstandard_gamma_legacy_seedinghypergeometricRuntimeWarningrandom_samplecount_nonzerobit_generatorOverflowErrorsearchsortedreturn_indexnumpy.linalgnoncentral_fnewbyteorderpermutationmultinomialexponentialcheck_validUserWarningRandomStateImportErrortriangularstandard_tstacklevel__pyx_vtable__mtrand.pyxmode > rightlogical_orless_equalleft == rightissubdtypeempty_likeValueErrorIndexErrorwriteablesingletonset_statelogserieslognormalleft > modehas_gaussget_stategeometricdirichletchisquareTypeErrorMT19937warningsvonmisessubtractreversedreducerayleighoperatorlogisticitemsizeisscalarisnativeisfinitebinomialallcloseSequenceweibulluniformtobytesstridesshufflereshapereplacerandintpoissonnsample_mt19937laplacegreaterfloat64castingcapsulebg_type at 0x{:X}asarrayalpha <= 0_MT19937unsafeuniqueuint64uint32uint16sample__reduce__random_rand_pickleparetoobject_normallegacykwargs__import__ignoregumbelformatdoublecumsumchoicebitgenastypearangezerosuint8statesigmashapescalerightravelrangerandnraisepvalspowernumpyngoodkappaisnanint64int32int16indexgaussgammaflagsfinfoequal__enter__emptydtypedfnumdfden__class__bytesarrayalpha__all__zipfwarnwaldtype__test__takesqrtsortsizesideseedrtolranfrandprodnoncndimnbad__name__modemean__main__locklessleftitemintpint8high__exit__copybool_betaatolargstolsvdsum__str__poslowloclamkeygetepsdotcovanyalladdqh??UUUUUU?"@m{??@>@3?r?q?0@0C@-DT! @h㈵>@-DT!@?/*p?3f?(_?xY?յS?N?J?F?DB?Qt??u+ ?A?Į?"?ʝ?G??i>l>7>>*J>>>^>>F>>7P>>K{>>>u>;->>>|>eO>4(>8>L>N>ȿ>>>>·>ε>߳>>>0>U>~>|>ަ>Y>IP>w>ҟ>B>e>)>~>_>>C>{>>J>﷍>X(>'>N>Í>x >b>x>>!>^}>;z>Хw>@t>wr>byl>i>g>Sd>3a>^>]\>&Y>z)W>T>P R>O>L>5~J>3H>E>nC>@>VK>>;>9>07>4>2>>0>p->+>s)>7'>%>"> >s>L>*>= >T>T>4>>y >ϣ > >>L>>l>=+==0 =C==8==hp==0== =n==|===;=ں=Z=o,=ް=ߗ=.W===%==r=W= C=4=M,=4*=D.=y8=H=~=x=3E3lM3FT3/[3b3i34p3fw3&~3[3B3ψ3g37!3>3T3d3n3r3Fq3j3_31P3r<3$3k 33ȸ3q3|{3P3#3C333dY3"3+3®3r35333x37333p3-33731b33l44(44h4C44 4` 4M 47 44?4nB44L4 i4a4T044542`44p. 4!4"4i$4%4@'4t(4>*4+4,4j.4/4'P1424):4454&)7484c:4;4$=4+>4@4A4KC4vD4B(F4G4:I4J4rTL4M4GuO4Q4R44T4U4EiW4Y4 Z4G\4]4_4:a4b4d4Bf4\g4ji4bk4m4n4p4\r4}"t4Yu4Hw4[y4X{46.}4 4q4a4]S4F4N<434,4+(4{%4$4o&4,*4'04m84 C4P4_4q474{4w4>ԕ44s4<4d444$4 (4a44lߢ4$4l44x 4_444{4 4EP4±4{:4귴4);4nķ4S444<*4տ44A4.44ע44f4RW4R4*Y4Fk44δ444444g44k4<444y44u4_45555@5ó 5 5]5^555q5v 5!5%5V*5s/5;S55:<5D5NO5^5Nv5QHqoMֻanjDotTrotou$w'xx,jyy7\z׻z{W{S{{.|3|]|ȃ|||I||}C0}F}Z}m}}S}(}}-}}"}}|}M}~i ~~~B(~o0~C8~?~F~M~T~Z~a~f~l~r~]w~v|~`~ ~~$~m~~~w~:~ަ~f~ѭ~#~Z~y~~q~K~~~^~~a~~~`~~~~~~~~f~*~~~-~~J~~=~~~\~~~$~U~}~~~~~pH`  i   6  H  A!B+m 5XttW3 `wK\ L   s   G {V~~~d~~x~~K~~~~~~~~~)~~~a~~~{~;~~A~~~m~~z~~~"~k~]~~~ԃ~|~s~j~Ua~W~K~?~2~$~~~ }}} }}i}A}}|Q|D|{3N{zeyww7ms?7E?P?'{{?*!?bv?mU?9U1T?/v?x]?&1$-?~ n?cK[!?I?\Omg?f?uLi=?sڂl?x?Qf?ij?%ᨯC?+?Dܻ?z?cE#;?^E#?$O?2m?P"K?>?{s?%;?omo?3;?J9?++?*T[?};1s?HeC?$`?vE!=?ſ-r?MBц?K=?Q}6Ei?7u? !?z}k? ~?@?`x?*?8? Qi?oTC?_(4?ָ?@je?!u v?7Zi?{ ?I?]T?9]??}?8aD:?Yζi?Ɲҷ?r^sSw?ꍰ07?d>[?%۹? Ə{?'HB>?vX#?l1&?:l?磽!O?ލ?&?ڋ?ﾀ+ j?A1?N0Z?0H?}G?(V?5$1!?pB9 ?b"FS?)vEW(?vG}rO?~ /? {^?Z? ?ބS?i"?lR?3Sn?>N?Ґ]b?,|y2?jG>?TLҫ?~>\O??@YH?/֎@?9O"H?>?1 7?8?Ox?]4?5D9g?r|?>ܸ$8? [B/?I䠟?O?y%d?bPޱ?c?PR?j?F}?9(Q1?c?(ڦ^w?0U^Q?1j?T ξ?x.BTv?Imb.??6YJ?)ِ?\C}?%d?w?SN?эv?pa?,Q&?@oű?SuFe?PV?;?I?viׯ?4D?.g?X1Iα?Jy?!dJ?پz?j»?8G;?L|{ʎ?mwn?k9:9??Ry?A&E?U?Ŗ<?k&_?G??~#? V#?_?S?Q| z? Y&?$?htQz? 3Tݜ?pXP?N梚?H*g?gS(u?1c?w@rT?Q=I?QA?]1%? RD?lj?W'n?-BU؊?h?t4? n?boQ?qvi?_)N?]tQW}?6H#z? 67w?"Ηs?C@Wi=q?ḰXl?f?$ka?%> T+Y? O?K 2=?]d<A]X`<+M[Ij<[5q&<.8eG< h#ឪV <;LC%K<ꆭh NVeΙVn6nvK zicp%E tQ)2U1WQ9Lin?23F:L"3\LQ V f[_rWDdx h+*k2=Ko:qr Mu\x?A{FS~8;b=ZV`bBtu9=JE>XدGwdO 8cx AFẙi&zqVYםΡag6 X83:뇡koɣj_ۤ| Mg^ݧt|Ψ_ΓXp2X^ttH蟿W;ޭl~$\z[߁İPp:J+N!X ɦ֬ ᆴX7(. Ɉ?5}h.G{tr&oya=cA/˺DH0⤮<)9O@ᣩTrVj֋@?˷dsI^i@(0߾ta&⊂l1EA1T[n&mi#d)B}QJwt}B < EOvpc/F<Ң"Ae އ0~ Rfq(*QtH3D@M`P}hwx%ƿ8*JG+[EliPIw+ E>ҙ02yΩ4A (Nt.Ȱ--̕^&܌z#;ޖu~g6X .pmF 3n bH޵LaEZvpR(-x_b˿ӰdyQӶVg<7܆ut7$MH𯋉ld"rqտH)݄ /0 wپ}2}K D5z&R cM,}uc?Ѡp5.bJ3ʸT[vv+\[U@ضBi"7oLeiFγ>SR(D2Z> B0$y1gWr-ެ @樫(afoeW-|&aY +M?V#z?u?q?}n?k?Lh?e?Rc?`?Zw^?*+\?Y?RW?U?_S?XQ?߱O?M?3K?J?GH?F?jD?`C?(`A?j??>?x,>N>>q>>>j>>k>>Π>>F>>>'>\>#>u>J>*>_F>d>+>$>w>>>JK>y>|>iݿ>>I>;>ʾ>t>5<> ~>>>O>>>~3>T>ե>(>g~>ՠ>G/>>>F>J> >:n>bԓ>Q<>>x>~>>>^>Ј>D>l>1>>%>\D>@|>?y>Bv>Hs>Qp>#^m>mj>|g>md>a>^>$[> Y>=3V>[S>P>M>J>~H>UE>B>?>=>S:>7>"4>=22>T/>d,>m+*>m'>c$>N?">,>>m>t>F>>1*> > >Y>>ʗ>>I=_={==^==&=_=g=='0===P6=˙=\= s==d= =yo=/=6=.=fЍ=x=i'=܀=a1y=p=xIh=_==W=TO=G=>=N6=.=&===-H==<א<̀<<<.4V?4=3@4A4A4qB4C4D4udE4-CF4K"G4H4H41I4J4vK4\fL4HM4+N4aO4O4bP4ٽQ4R4ԊS4crT4ZU4CV4-W4ZX4Y4UY4Z4[4(\4_]4^4_4C`4va4alb40cc47[d4~Te4Of4Jg42Hh4Fi4Fj4Hk4Kl4MPm4Vn4^o48hp4sq4r4s4 t4u4v4Cw4x4 z42{40S|4u}4~44v4@ 4L4>4ق4v444lV44R4F44p4 I44"4_44Ќ4l4L4`4ԏ4坐4y4ݖ4%44r&4k44(4444.4Q4N4t44\۶4H94̻4p44~X4w4p_4~444wE`mru\zw8xky5zz/ {ԃ{{7|3}|&|H|}C}g}ۇ}}a}g}]}~~4%~5~C~Q~g^~ij~u~>~2~~r~դ~Ƭ~N~u~C~~~~k~~~~~~t~~~6 < :#%](*.-z/13579;=?EABD:FGNIJ8LMNLPQR T=UdVWXYZ[\]^~__`;abbcod.eefLggh~~7~~/~7~~ ~ ~w~G]~>~Y~,}6}b}|O|06{x?yjD?l[T?w'??o?Wp?xI?-3?x^j??N?R:e?4:>?l?*?%z?PՋt?4?e;?$"?zaWF}?Gz‘B?Oq1? OU?ߺH?7a?nV,? K?Xhw?հ<?Vp\?m?)?zP?ZcX?*;Q^?#*'g? U7?e&$ ?jJo?\Ȭ)?L&?FS?leZ&?g ?NIO??xRr!?P_hy?y6IJO?_5%?[X~?1>?bU?+À?PX?5:pɗ0?8d?;U?J?͓?)m?ېZ]G?/|!? ?iT??Wq?PF9 ?ߓ^??ۮY?3???i?Z8o? O5?ٸ?P?R9?igP?La;?L?!ވ?%o?{7=8?Ҁt?DvC?6?=p\?;So&?mj?W?j?$O?z5Ѽ?Ҏ?C|P?yh|?%H?/ZM?f!w;??>ǭ?MAz?G?y?.?P9կ?TT}?g4K?#$O? Y?BM?6C;?B"_U?~t$?œ߉?52?Ҙl'?DɤT?<(i?qE8 ? Uī?OQM?o^?Sq͒?Gط5?zx?1zd}?:R!?Wg?~& ~k?=~-2?ZҿҶ?'|j_]?it?[?8R?uqb?#h?z|J?G~`?\!>?GF?vJ?l󈬚?5hȩmE?㭍?-l ?uG?1i%?調?M?e*|?zè?^V?4<%F?B}u?c-@c?n? R=?Kr?*}T#?,"k>?R) ?K{o?vaӽ?命8? t;I_? h?3xk?3Ӻ?b3?vZ9S?LJisk?M$a.?ftW?+ ?"@|?&#?p>_?1fҲ? DE?} ?/?%,?0?5nl+,&?QG?b. ?,*(>?p_8?cU)?h*?'wާ?dИۦ?ԭ<ڥ?]']ۤ?ݣ?=|?j?.?ĥׁ?u? ̓0?"NR? y? ڥ?d֔?^8 ?0`4I?IrO*?O'?x A?B?/)?7h`|?] ٨v?p?gC_e?T?yx;I< <[,L< Ŀk<4xV<=A[<'?}y<NG<~;[xo6xu{fUY>9>{ppCBwS(:5^dܓAN}8) YfHqն&|s f2,2Ztզޗ .n ZR'ӯB)[l@u Pҍ'TȈt(5wI'L/$;nXMØT`OArW,+jtȳRfARnqӊ<KZW$eKs) 4<=>)G'QA@Y.(5bX jz>lq{2Xx{~JH҄Cc`Qz%~ )Q\HsrUb'Bkq-hnק Ψ;3Kd)P^٨Tv$Hx"$ 5..&$ŗ: Aޓ=?~)@ lѿ3 ; @@5gG8?SˆB?AAz?<ٰj_?$+K?88C?J?llf?UUUUUU?dg?$@= ףp=@n?[ m?h|?5?333333 @r?$~?B>٬ @r鷯?Q?Q?9v?(\@ffffff@.@4@x&??UUUUUU?a@X@`@|@@MA.A-DT! C;,H0d - * 0   0-p@P`L`0 Dph@ Pl  !\ " %d `' ( ( )@ *d + + ,T.p/@33T:>L@Dh@HLO$pR U$W`[4^``<cfLj nxq@txx Pz`P`hp<8$ h ` !d!!!"##$$D%%%<&/& =&0JH'Z'{'0(@))PT*@8,+g+@,`-`1t-P -p.,.@x.`... .@. /D/d//@/`//@/p0h0p00 001@,1P@1X1l1p!1"1"1"1#02p#l2$2$2&L3`&3&3P'4(P40)4*4@+5-`505050515162<63T603l6`3636364646 57P587p5P75h7676777`777 87(8p8L8;8;8PF(9Hd9`I9J9J9K:0OL:0Pp:P:PQ:pQ:R:S;T8;U;V;`XL<Z<PZ<p]@=`_=a@>@d>d>zRx $@ FJ w?;*3$"DX4XdEBDD d GBI AAB| xXD{ A \BED D(D0| (D ABBH Q (D DBBC D (A ABBE   D4BFB B(A0A8DP8A0A(B BBB4|QADD ^ CAE M CAG A@ G h H <AD A AI G AH V AI G AC tAT K H`<GBBB B(D0D8G` 8C0A(B BBBC K 8F0A(B BBBF 4wAAG O CAD I CAC \4BBE A(A0R (D BBBI I (D BBBE {(A EBBD8ĦBEL E(A0A8E@8D0A(B BBBDyBEL E(A0A8E@8D0A(B BBB8.~BEH A(A0b(D BBB8p`BBE A(A0J(D BBB(@BAG U DBK Plh BHE A(D0JHfA\0D(A BBBlBBB A(D0DPj 0D(A BBBL ` 0D(A BBBE ` 0D(D BBBB P0\BBA s EBI A BBC D BBH H IBM d`BBB B(D0D8Hh 8A0A(B BBBF  8F0A(B BBBA HBEE E(D0A8G8D0A(B BBBL8LBEA A(D@( (A ABBE m (A ABBE (BD V EG k EH 0tBEG e EBF lEB(dxBED ^ EBH yD F AA D8BIA [ ABF L ABA \ AEF DLBKD R ABJ A GBF Q AEI \(BEI A(G0u (D ABBJ [ (D ABBD \ (D DBBH @( BIA Y BBG Q EBH qBB(l`kBD d EA I EB P BBD G0Z  ABBB   DBBC `  ABBA 8 $BEA G(G0s (D ABBF d( $BBE E(D0A8F 8A0A(B BBBH d 8A0A(B BBBJ *BBB B(A0D8G`Uh[pBxFFBFABFQ`[ 8D0A(B BBBD ^hWpRhF`hWp_hA` -BBB A(A0G@dHMP`HF@_ 0D(A BBBH NHhPBXA`AhBpAxDBFQ@c 0D(A BBBH /BEA A(GPuXW`BhFpAxEFAEFQPW (D ABBC l (D ABBK AXW`RXFP_ (D ABBH < 3BEA A(GPuXW`BhFpAxEFAEFQPW (D ABBC l (D ABBK AXW`RXFP_ (D ABBH <6BEA A(GPuXW`BhFpAxEFAEFQPW (D ABBC l (D ABBK AXW`RXFP_ (D ABBH d h9BEA A(GPuXW`BhFpAxEFAEFQPW (D ABBC l (D ABBK AXW`RXFP_ (D ABBH | <BBB B(A0D8GPHXi`BhApAxBADBFQP|XI`YXFP_ 8D0A(B BBBG |x >BBB B(A0D8GPHXi`BhApAxBADBFQP|XI`YXFP_ 8D0A(B BBBG | @BBB B(A0D8GPHXi`BhApAxBADBFQP|XI`YXFP_ 8D0A(B BBBG x$CBBB B(A0D8G`Uh[pBxFFBFABFQ`[ 8D0A(B BBBD ^hWpRhF`hWp_hA`|$FBBB B(A0D8GPHXi`BhApAxBADBFQP|XI`YXFP_ 8D0A(B BBBG THJBBB A(A0G@eHIPYHF@_ 0D(A BBBJ OHfPBXA`BhBpAxBBAQ@[ 0D(A BBBG JBBB B(A0D8G`Uh[pBxFFBFABFQ`[ 8D0A(B BBBD ^hWpRhF`hWp_hA`|MBBB B(A0D8GPHXi`BhApAxBADBFQP|XI`YXFP_ 8D0A(B BBBG LOBFB B(D0A8G`NhapBxFAEFABFQ`hEpkhF`_ 8D0A(B BBBI hWp_hA`NhVp`hG`RBBB B(A0D8G`Uh[pBxFFBFABFQ`[ 8D0A(B BBBD ^hWpRhF`hWp_hA`|LUjBBB B(A0D8G`hPpRhF`_ 8D0A(B BBBE Rh[pBxFFBFABFQ`|XBBB B(A0D8GPHXi`BhApAxBADBFQP|XI`YXFP_ 8D0A(B BBBG LZBEA A(GPuXX`BhBpAxEFAEBWPW (D ABBD l (D ABBK AXW`RXFP_ (D ABBH |]ABBA A(G@eHIPYHF@_ (D ABBF OHXPBXB`BhBpBxBBBQ@[ (D ABBD @`_BBB B(D0GP 0D(B BBBE |eABBA A(G@eHIPYHF@_ (D ABBF OHXPBXB`BhBpBxBBBQ@[ (D ABBD $XgBBB A(A0G@dHMP`HF@_ 0D(A BBBH NHhPBXA`AhBpAxDBFQ@c 0D(A BBBH iBBB B(A0D8G`Uh[pBxFFBFABFQ`[ 8D0A(B BBBD ^hWpRhF`hWp_hA`h<lvBBB A(G@ (D BBBC yHIPYHF@_ (D BBBF b (D BBBD doBBB A(A0G@eHIPYHF@_ 0D(A BBBJ I 0D(A BBBD HLqFBEB B(A0A8D`Q 8D0A(B BBBA L\P0BBB D(D@ (A BBBK  (C BBBC \0BBB A(D0 (D BBBE L (D BBBB d (D BBBB D ЂmBIA j BBF l BBH  BBJ HT%BBE B(A0A8D@' 8D0A(B BBBK `܌BBA D(G@  (A ABBH \ (A ABBF R (C ABBF \BBB A(J DEFFBBAIbDEFFBBBIEFHFFBBAI{MHFFBBAI{MHFFBBAIM(A BBB@BBB B(D0GP 0D(B BBBE @<PBAD0v ABH V ABG j CBI pBBD A(D0| (D ABBF L (D ABBC  (D ABBE  (D ABBH @ؗBBB B(A0GP 0D(B BBBB @8TBBB B(A0GP 0D(B BBBB L|ОPBB E(A0  (E BBBH r (E BBBK BBB D(D0GPkXI`YXFP^ 0A(A BBBB SXb`AhApBxADBFATPE 0A(A BBBH XBBB D(D0GPkXI`YXFP^ 0A(A BBBB SXb`AhApBxADBFATPE 0A(A BBBH BBB D(D0GPkXI`YXFP^ 0A(A BBBB SXb`AhApBxADBFATPE 0A(A BBBH |p BBA D(G@dHMP`HF@_ (D ABBA SHcPBXA`BhBpGxEFAQ@z (D ABBD |BBB D(D0G`ehWpRhF`^ 0A(A BBBA Wh[pFxFBFAEFAQ`A 0A(A BBBC  hWp_hA`BBB B(D0A8D` 8D0A(B BBBG yhIpYhF`\ 8C0A(B BBBG j 8C0A(B BBBB HBBB B(A0D8DP 8D0A(B BBBE HdhBBB B(A0D8GPz 8D0A(B BBBE \BBB B(A0A8G@IYF_ 8D0A(B BBBK X l2BBB B(A0D8DPRF_ 8D0A(B BBBJ Tl P + BBB E(A0A8D`hIpYhF`b 8D0A(B BBBH T ( BBB B(A0A8DPkXI`YXFP_ 8D0A(B BBBC X!"BBB B(A0A8GpxIYxFp_ 8D0A(B BBBC Hx!$3 !BBB B(D0D8D` 8D0A(B BBBE !SIBBB B(A0A8DG 8D0A(B BBBF !IBKPBABBATWRFZoBAFBABBAR"|m3BFB B(A0D8DEkF_ 8D0A(B BBBA LVABFBBFBY9 bFBBFDBY<W_AFV`G\X#Y"BBB B(A0A8JIYFb 8D0A(B BBBD l#'BBB B(A0D8GmWRFb 8D0A(B BBBA `!W_A($$BFB B(A0D8DXEkF_ 8D0A(B BBBA USOEBBAABAT5eBDBBABYHW_AJV`Gl% A/BBB B(A0D8G2WRFb 8D0A(B BBBD !V`Gp%;'BBB B(A0D8D_OAi_LBf 8D0A(B BBBG _LAPRFi _MB_MBGLABOAFEMBbGLAdp&bIBBB B(D0A8Dg 8A0A(B BBBF l 8A0A(B BBBB l&TXBFB B(A0D8G)WRFb 8D0A(B BBBA +W_F\H'BBB B(A0D8GN 8D0A(B BBBA {"PRF'' ('` AG0_ AH  AA 'T/DjH(lAG@ AF t AK 4 AK J EA J AE L(D Ud('D b|(B\ ](?D n(@DQ(HHD | A 4(|AO ~ EE L AK J AE )dWAG AA 8)IO0 EA \)$D _t)DI )AG _ AE ),&IX)@/AG ]AL)PBAD D@  AAFG j  AAFN  AAF<*SAG }A\*D U(t*Ma J n J L D AH*LBBB E(A0D8G[ 8D0A(B BBBA *+ +n\ L,+`4@+\AGP AG ^ AA L AC x+AG@ DD +%G]+ 8+IGED D(F0a(A ABBG8,YGED D(F0t(A ABBDH@,BEH H(KP (E ABBK [(A AFB8,IGED D(F0b(A ABBFT,-BBE H(H0K@ 0D(A BBBE m0A(A FBB8 -IGED D(F0b(A ABBF8\-aGED D(F0v(A ABBJ<-yGED D(G0O(A ABBHH-DBBE I(H0K8K` 8E0A(B BBBE 8$.IGED D(F0b(A ABBFH`.BBE B(H0H8KP 8F0A(B BBBD 8.IGED D(F0b(A ABBFH.$AG@ AI t AK 1 AF J EI J AE H4/xAG0 AC q AF B AE J EA J AE /GJ/GJ/GJ/ 0/vAD@O EAH `0$D _(0D U@0(D cX0D Up0D U(0AG0J AM A0DQ0SAG }A0/AG ]A 1D U$1B\ ]<1PCD z(T1AG V AI XA$1iBFG0IAK14NAG DA1dDI1l3D g A 1AG _ AE l 2BFB A(A0Gpg 0D(A BBBK V 0D(A BBBG / 0D(A BBBA 2&IXL2 BFG E(D0A8J 8A0A(E BBBA 82 RBED A(G`k (A ABBD 083Dka L nJHA G L<l3&AG  AL l AK L AC O DE 3WAG AA 3IO0 EA ,3CAGP# AD _ AH 4AG@ DD D4_D Z\4D t4(4VBAG@CFB 4 L@ M r E 04MGG T ABF hH 5PZBBB B(D0A8D@} 8D0A(B BBBE HX5dBBB B(D0A8DPm 8D0A(B BBBE x5BDB B(A0A8DP 8A0A(B BBBD C 8A0A(B BBBC a 8A0A(B BBBE x 6 BDB B(A0A8DP 8A0A(B BBBC B 8A0A(B BBBD ` 8A0A(B BBBF (60IACD g AAG H6TBBB B(D0D8D`b 8A0A(B BBBH 7( BBB B(D0D8DP 8A0A(B BBBE  8A0A(B BBBC d 8A0A(B BBBB m 8A0A(B BBBA d7!{BBB B(A0A8GP 8A0A(B BBBF I 8A0A(B BBBE x8#`BBE B(A0A8DPR 8K0A(B BBBI _ 8A0A(B BBBG L 8A0A(B BBBJ ,8|%wKHE }ABH\8%BIB B(A0D8D`n 8A0A(B BBBA S8A0A(B BBB00k& kko` G pHЦ83s o2oo0ok6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FV This is an alias of `random_sample`. See `random_sample` for the complete documentation. This is an alias of `random_sample`. See `random_sample` for the complete documentation. Sets the singleton RandomState's bit generator Parameters ---------- bitgen A bit generator instance Notes ----- The singleton RandomState provides the random variate generators in the ``numpy.random``namespace. This function, and its counterpart get method, provides a path to hot-swap the default MT19937 bit generator with a user provided alternative. These function are intended to provide a continuous path where a single underlying bit generator can be used both with an instance of ``Generator`` and with the singleton instance of RandomState. See Also -------- get_bit_generator numpy.random.Generator Returns the singleton RandomState's bit generator Returns ------- BitGenerator The bit generator that underlies the singleton RandomState instance Notes ----- The singleton RandomState provides the random variate generators in the ``numpy.random`` namespace. This function, and its counterpart set method, provides a path to hot-swap the default MT19937 bit generator with a user provided alternative. These function are intended to provide a continuous path where a single underlying bit generator can be used both with an instance of ``Generator`` and with the singleton instance of RandomState. See Also -------- set_bit_generator numpy.random.Generator seed(seed=None) Reseed the singleton RandomState instance. Notes ----- This is a convenience, legacy function that exists to support older code that uses the singleton RandomState. Best practice is to use a dedicated ``Generator`` instance rather than the random variate generation methods exposed directly in the random module. See Also -------- numpy.random.Generator permutation(x) Randomly permute a sequence, or return a permuted range. If `x` is a multi-dimensional array, it is only shuffled along its first index. .. note:: New code should use the `~numpy.random.Generator.permutation` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- x : int or array_like If `x` is an integer, randomly permute ``np.arange(x)``. If `x` is an array, make a copy and shuffle the elements randomly. Returns ------- out : ndarray Permuted sequence or array range. See Also -------- random.Generator.permutation: which should be used for new code. Examples -------- >>> np.random.permutation(10) array([1, 7, 4, 3, 0, 9, 2, 5, 8, 6]) # random >>> np.random.permutation([1, 4, 9, 12, 15]) array([15, 1, 9, 4, 12]) # random >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.permutation(arr) array([[6, 7, 8], # random [0, 1, 2], [3, 4, 5]]) shuffle(x) Modify a sequence in-place by shuffling its contents. This function only shuffles the array along the first axis of a multi-dimensional array. The order of sub-arrays is changed but their contents remains the same. .. note:: New code should use the `~numpy.random.Generator.shuffle` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- x : ndarray or MutableSequence The array, list or mutable sequence to be shuffled. Returns ------- None See Also -------- random.Generator.shuffle: which should be used for new code. Examples -------- >>> arr = np.arange(10) >>> np.random.shuffle(arr) >>> arr [1 7 5 2 9 4 3 6 0 8] # random Multi-dimensional arrays are only shuffled along the first axis: >>> arr = np.arange(9).reshape((3, 3)) >>> np.random.shuffle(arr) >>> arr array([[3, 4, 5], # random [6, 7, 8], [0, 1, 2]]) dirichlet(alpha, size=None) Draw samples from the Dirichlet distribution. Draw `size` samples of dimension k from a Dirichlet distribution. A Dirichlet-distributed random variable can be seen as a multivariate generalization of a Beta distribution. The Dirichlet distribution is a conjugate prior of a multinomial distribution in Bayesian inference. .. note:: New code should use the `~numpy.random.Generator.dirichlet` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- alpha : sequence of floats, length k Parameter of the distribution (length ``k`` for sample of length ``k``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n)``, then ``m * n * k`` samples are drawn. Default is None, in which case a vector of length ``k`` is returned. Returns ------- samples : ndarray, The drawn samples, of shape ``(size, k)``. Raises ------ ValueError If any value in ``alpha`` is less than or equal to zero See Also -------- random.Generator.dirichlet: which should be used for new code. Notes ----- The Dirichlet distribution is a distribution over vectors :math:`x` that fulfil the conditions :math:`x_i>0` and :math:`\sum_{i=1}^k x_i = 1`. The probability density function :math:`p` of a Dirichlet-distributed random vector :math:`X` is proportional to .. math:: p(x) \propto \prod_{i=1}^{k}{x^{\alpha_i-1}_i}, where :math:`\alpha` is a vector containing the positive concentration parameters. The method uses the following property for computation: let :math:`Y` be a random vector which has components that follow a standard gamma distribution, then :math:`X = \frac{1}{\sum_{i=1}^k{Y_i}} Y` is Dirichlet-distributed References ---------- .. [1] David McKay, "Information Theory, Inference and Learning Algorithms," chapter 23, 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.dirichlet((10, 5, 3), 20).transpose() >>> import matplotlib.pyplot as plt >>> plt.barh(range(20), s[0]) >>> plt.barh(range(20), s[1], left=s[0], color='g') >>> plt.barh(range(20), s[2], left=s[0]+s[1], color='r') >>> plt.title("Lengths of Strings") multinomial(n, pvals, size=None) Draw samples from a multinomial distribution. The multinomial distribution is a multivariate generalization of the binomial distribution. Take an experiment with one of ``p`` possible outcomes. An example of such an experiment is throwing a dice, where the outcome can be 1 through 6. Each sample drawn from the distribution represents `n` such experiments. Its values, ``X_i = [X_0, X_1, ..., X_p]``, represent the number of times the outcome was ``i``. .. note:: New code should use the `~numpy.random.Generator.multinomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : int Number of experiments. pvals : sequence of floats, length p Probabilities of each of the ``p`` different outcomes. These must sum to 1 (however, the last element is always assumed to account for the remaining probability, as long as ``sum(pvals[:-1]) <= 1)``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. See Also -------- random.Generator.multinomial: which should be used for new code. Examples -------- Throw a dice 20 times: >>> np.random.multinomial(20, [1/6.]*6, size=1) array([[4, 1, 7, 5, 2, 1]]) # random It landed 4 times on 1, once on 2, etc. Now, throw the dice 20 times, and 20 times again: >>> np.random.multinomial(20, [1/6.]*6, size=2) array([[3, 4, 3, 3, 4, 3], # random [2, 4, 3, 4, 0, 7]]) For the first run, we threw 3 times 1, 4 times 2, etc. For the second, we threw 2 times 1, 4 times 2, etc. A loaded die is more likely to land on number 6: >>> np.random.multinomial(100, [1/7.]*5 + [2/7.]) array([11, 16, 14, 17, 16, 26]) # random The probability inputs should be normalized. As an implementation detail, the value of the last entry is ignored and assumed to take up any leftover probability mass, but this should not be relied on. A biased coin which has twice as much weight on one side as on the other should be sampled like so: >>> np.random.multinomial(100, [1.0 / 3, 2.0 / 3]) # RIGHT array([38, 62]) # random not like: >>> np.random.multinomial(100, [1.0, 2.0]) # WRONG Traceback (most recent call last): ValueError: pvals < 0, pvals > 1 or pvals contains NaNs multivariate_normal(mean, cov, size=None, check_valid='warn', tol=1e-8) Draw random samples from a multivariate normal distribution. The multivariate normal, multinormal or Gaussian distribution is a generalization of the one-dimensional normal distribution to higher dimensions. Such a distribution is specified by its mean and covariance matrix. These parameters are analogous to the mean (average or "center") and variance (standard deviation, or "width," squared) of the one-dimensional normal distribution. .. note:: New code should use the `~numpy.random.Generator.multivariate_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : 1-D array_like, of length N Mean of the N-dimensional distribution. cov : 2-D array_like, of shape (N, N) Covariance matrix of the distribution. It must be symmetric and positive-semidefinite for proper sampling. size : int or tuple of ints, optional Given a shape of, for example, ``(m,n,k)``, ``m*n*k`` samples are generated, and packed in an `m`-by-`n`-by-`k` arrangement. Because each sample is `N`-dimensional, the output shape is ``(m,n,k,N)``. If no shape is specified, a single (`N`-D) sample is returned. check_valid : { 'warn', 'raise', 'ignore' }, optional Behavior when the covariance matrix is not positive semidefinite. tol : float, optional Tolerance when checking the singular values in covariance matrix. cov is cast to double before the check. Returns ------- out : ndarray The drawn samples, of shape *size*, if that was provided. If not, the shape is ``(N,)``. In other words, each entry ``out[i,j,...,:]`` is an N-dimensional value drawn from the distribution. See Also -------- random.Generator.multivariate_normal: which should be used for new code. Notes ----- The mean is a coordinate in N-dimensional space, which represents the location where samples are most likely to be generated. This is analogous to the peak of the bell curve for the one-dimensional or univariate normal distribution. Covariance indicates the level to which two variables vary together. From the multivariate normal distribution, we draw N-dimensional samples, :math:`X = [x_1, x_2, ... x_N]`. The covariance matrix element :math:`C_{ij}` is the covariance of :math:`x_i` and :math:`x_j`. The element :math:`C_{ii}` is the variance of :math:`x_i` (i.e. its "spread"). Instead of specifying the full covariance matrix, popular approximations include: - Spherical covariance (`cov` is a multiple of the identity matrix) - Diagonal covariance (`cov` has non-negative elements, and only on the diagonal) This geometrical property can be seen in two dimensions by plotting generated data-points: >>> mean = [0, 0] >>> cov = [[1, 0], [0, 100]] # diagonal covariance Diagonal covariance means that points are oriented along x or y-axis: >>> import matplotlib.pyplot as plt >>> x, y = np.random.multivariate_normal(mean, cov, 5000).T >>> plt.plot(x, y, 'x') >>> plt.axis('equal') >>> plt.show() Note that the covariance matrix must be positive semidefinite (a.k.a. nonnegative-definite). Otherwise, the behavior of this method is undefined and backwards compatibility is not guaranteed. References ---------- .. [1] Papoulis, A., "Probability, Random Variables, and Stochastic Processes," 3rd ed., New York: McGraw-Hill, 1991. .. [2] Duda, R. O., Hart, P. E., and Stork, D. G., "Pattern Classification," 2nd ed., New York: Wiley, 2001. Examples -------- >>> mean = (1, 2) >>> cov = [[1, 0], [0, 1]] >>> x = np.random.multivariate_normal(mean, cov, (3, 3)) >>> x.shape (3, 3, 2) Here we generate 800 samples from the bivariate normal distribution with mean [0, 0] and covariance matrix [[6, -3], [-3, 3.5]]. The expected variances of the first and second components of the sample are 6 and 3.5, respectively, and the expected correlation coefficient is -3/sqrt(6*3.5) ≈ -0.65465. >>> cov = np.array([[6, -3], [-3, 3.5]]) >>> pts = np.random.multivariate_normal([0, 0], cov, size=800) Check that the mean, covariance, and correlation coefficient of the sample are close to the expected values: >>> pts.mean(axis=0) array([ 0.0326911 , -0.01280782]) # may vary >>> np.cov(pts.T) array([[ 5.96202397, -2.85602287], [-2.85602287, 3.47613949]]) # may vary >>> np.corrcoef(pts.T)[0, 1] -0.6273591314603949 # may vary We can visualize this data with a scatter plot. The orientation of the point cloud illustrates the negative correlation of the components of this sample. >>> import matplotlib.pyplot as plt >>> plt.plot(pts[:, 0], pts[:, 1], '.', alpha=0.5) >>> plt.axis('equal') >>> plt.grid() >>> plt.show() logseries(p, size=None) Draw samples from a logarithmic series distribution. Samples are drawn from a log series distribution with specified shape parameter, 0 <= ``p`` < 1. .. note:: New code should use the `~numpy.random.Generator.logseries` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- p : float or array_like of floats Shape parameter for the distribution. Must be in the range [0, 1). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logarithmic series distribution. See Also -------- scipy.stats.logser : probability density function, distribution or cumulative density function, etc. random.Generator.logseries: which should be used for new code. Notes ----- The probability density for the Log Series distribution is .. math:: P(k) = \frac{-p^k}{k \ln(1-p)}, where p = probability. The log series distribution is frequently used to represent species richness and occurrence, first proposed by Fisher, Corbet, and Williams in 1943 [2]. It may also be used to model the numbers of occupants seen in cars [3]. References ---------- .. [1] Buzas, Martin A.; Culver, Stephen J., Understanding regional species diversity through the log series distribution of occurrences: BIODIVERSITY RESEARCH Diversity & Distributions, Volume 5, Number 5, September 1999 , pp. 187-195(9). .. [2] Fisher, R.A,, A.S. Corbet, and C.B. Williams. 1943. The relation between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology, 12:42-58. .. [3] D. J. Hand, F. Daly, D. Lunn, E. Ostrowski, A Handbook of Small Data Sets, CRC Press, 1994. .. [4] Wikipedia, "Logarithmic distribution", https://en.wikipedia.org/wiki/Logarithmic_distribution Examples -------- Draw samples from the distribution: >>> a = .6 >>> s = np.random.logseries(a, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s) # plot against distribution >>> def logseries(k, p): ... return -p**k/(k*np.log(1-p)) >>> plt.plot(bins, logseries(bins, a)*count.max()/ ... logseries(bins, a).max(), 'r') >>> plt.show() hypergeometric(ngood, nbad, nsample, size=None) Draw samples from a Hypergeometric distribution. Samples are drawn from a hypergeometric distribution with specified parameters, `ngood` (ways to make a good selection), `nbad` (ways to make a bad selection), and `nsample` (number of items sampled, which is less than or equal to the sum ``ngood + nbad``). .. note:: New code should use the `~numpy.random.Generator.hypergeometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- ngood : int or array_like of ints Number of ways to make a good selection. Must be nonnegative. nbad : int or array_like of ints Number of ways to make a bad selection. Must be nonnegative. nsample : int or array_like of ints Number of items sampled. Must be at least 1 and at most ``ngood + nbad``. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if `ngood`, `nbad`, and `nsample` are all scalars. Otherwise, ``np.broadcast(ngood, nbad, nsample).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized hypergeometric distribution. Each sample is the number of good items within a randomly selected subset of size `nsample` taken from a set of `ngood` good items and `nbad` bad items. See Also -------- scipy.stats.hypergeom : probability density function, distribution or cumulative density function, etc. random.Generator.hypergeometric: which should be used for new code. Notes ----- The probability density for the Hypergeometric distribution is .. math:: P(x) = \frac{\binom{g}{x}\binom{b}{n-x}}{\binom{g+b}{n}}, where :math:`0 \le x \le n` and :math:`n-b \le x \le g` for P(x) the probability of ``x`` good results in the drawn sample, g = `ngood`, b = `nbad`, and n = `nsample`. Consider an urn with black and white marbles in it, `ngood` of them are black and `nbad` are white. If you draw `nsample` balls without replacement, then the hypergeometric distribution describes the distribution of black balls in the drawn sample. Note that this distribution is very similar to the binomial distribution, except that in this case, samples are drawn without replacement, whereas in the Binomial case samples are drawn with replacement (or the sample space is infinite). As the sample space becomes large, this distribution approaches the binomial. References ---------- .. [1] Lentner, Marvin, "Elementary Applied Statistics", Bogden and Quigley, 1972. .. [2] Weisstein, Eric W. "Hypergeometric Distribution." From MathWorld--A Wolfram Web Resource. http://mathworld.wolfram.com/HypergeometricDistribution.html .. [3] Wikipedia, "Hypergeometric distribution", https://en.wikipedia.org/wiki/Hypergeometric_distribution Examples -------- Draw samples from the distribution: >>> ngood, nbad, nsamp = 100, 2, 10 # number of good, number of bad, and number of samples >>> s = np.random.hypergeometric(ngood, nbad, nsamp, 1000) >>> from matplotlib.pyplot import hist >>> hist(s) # note that it is very unlikely to grab both bad items Suppose you have an urn with 15 white and 15 black marbles. If you pull 15 marbles at random, how likely is it that 12 or more of them are one color? >>> s = np.random.hypergeometric(15, 15, 15, 100000) >>> sum(s>=12)/100000. + sum(s<=3)/100000. # answer = 0.003 ... pretty unlikely! geometric(p, size=None) Draw samples from the geometric distribution. Bernoulli trials are experiments with one of two outcomes: success or failure (an example of such an experiment is flipping a coin). The geometric distribution models the number of trials that must be run in order to achieve success. It is therefore supported on the positive integers, ``k = 1, 2, ...``. The probability mass function of the geometric distribution is .. math:: f(k) = (1 - p)^{k - 1} p where `p` is the probability of success of an individual trial. .. note:: New code should use the `~numpy.random.Generator.geometric` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- p : float or array_like of floats The probability of success of an individual trial. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``p`` is a scalar. Otherwise, ``np.array(p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized geometric distribution. See Also -------- random.Generator.geometric: which should be used for new code. Examples -------- Draw ten thousand values from the geometric distribution, with the probability of an individual success equal to 0.35: >>> z = np.random.geometric(p=0.35, size=10000) How many trials succeeded after a single run? >>> (z == 1).sum() / 10000. 0.34889999999999999 #random zipf(a, size=None) Draw samples from a Zipf distribution. Samples are drawn from a Zipf distribution with specified parameter `a` > 1. The Zipf distribution (also known as the zeta distribution) is a discrete probability distribution that satisfies Zipf's law: the frequency of an item is inversely proportional to its rank in a frequency table. .. note:: New code should use the `~numpy.random.Generator.zipf` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Distribution parameter. Must be greater than 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Zipf distribution. See Also -------- scipy.stats.zipf : probability density function, distribution, or cumulative density function, etc. random.Generator.zipf: which should be used for new code. Notes ----- The probability density for the Zipf distribution is .. math:: p(k) = \frac{k^{-a}}{\zeta(a)}, for integers :math:`k \geq 1`, where :math:`\zeta` is the Riemann Zeta function. It is named for the American linguist George Kingsley Zipf, who noted that the frequency of any word in a sample of a language is inversely proportional to its rank in the frequency table. References ---------- .. [1] Zipf, G. K., "Selected Studies of the Principle of Relative Frequency in Language," Cambridge, MA: Harvard Univ. Press, 1932. Examples -------- Draw samples from the distribution: >>> a = 4.0 >>> n = 20000 >>> s = np.random.zipf(a, n) Display the histogram of the samples, along with the expected histogram based on the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import zeta # doctest: +SKIP `bincount` provides a fast histogram for small integers. >>> count = np.bincount(s) >>> k = np.arange(1, s.max() + 1) >>> plt.bar(k, count[1:], alpha=0.5, label='sample count') >>> plt.plot(k, n*(k**-a)/zeta(a), 'k.-', alpha=0.5, ... label='expected count') # doctest: +SKIP >>> plt.semilogy() >>> plt.grid(alpha=0.4) >>> plt.legend() >>> plt.title(f'Zipf sample, a={a}, size={n}') >>> plt.show() poisson(lam=1.0, size=None) Draw samples from a Poisson distribution. The Poisson distribution is the limit of the binomial distribution for large N. .. note:: New code should use the `~numpy.random.Generator.poisson` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- lam : float or array_like of floats Expected number of events occurring in a fixed-time interval, must be >= 0. A sequence must be broadcastable over the requested size. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``lam`` is a scalar. Otherwise, ``np.array(lam).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Poisson distribution. See Also -------- random.Generator.poisson: which should be used for new code. Notes ----- The Poisson distribution .. math:: f(k; \lambda)=\frac{\lambda^k e^{-\lambda}}{k!} For events with an expected separation :math:`\lambda` the Poisson distribution :math:`f(k; \lambda)` describes the probability of :math:`k` events occurring within the observed interval :math:`\lambda`. Because the output is limited to the range of the C int64 type, a ValueError is raised when `lam` is within 10 sigma of the maximum representable value. References ---------- .. [1] Weisstein, Eric W. "Poisson Distribution." From MathWorld--A Wolfram Web Resource. 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 >>> s = np.random.poisson(5, 10000) Display histogram of the sample: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 14, density=True) >>> plt.show() Draw each 100 values for lambda 100 and 500: >>> s = np.random.poisson(lam=(100., 500.), size=(100, 2)) negative_binomial(n, p, size=None) Draw samples from a negative binomial distribution. Samples are drawn from a negative binomial distribution with specified parameters, `n` successes and `p` probability of success where `n` is > 0 and `p` is in the interval [0, 1]. .. note:: New code should use the `~numpy.random.Generator.negative_binomial` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- n : float or array_like of floats Parameter of the distribution, > 0. p : float or array_like of floats Parameter of the distribution, >= 0 and <=1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``n`` and ``p`` are both scalars. Otherwise, ``np.broadcast(n, p).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized negative binomial distribution, where each sample is equal to N, the number of failures that occurred before a total of n successes was reached. See Also -------- random.Generator.negative_binomial: which should be used for new code. Notes ----- The probability mass function of the negative binomial distribution is .. math:: P(N;n,p) = \frac{\Gamma(N+n)}{N!\Gamma(n)}p^{n}(1-p)^{N}, where :math:`n` is the number of successes, :math:`p` is the probability of success, :math:`N+n` is the number of trials, and :math:`\Gamma` is the gamma function. When :math:`n` is an integer, :math:`\frac{\Gamma(N+n)}{N!\Gamma(n)} = \binom{N+n-1}{N}`, which is the more common form of this term in the pmf. The negative binomial distribution gives the probability of N failures given n successes, with a success on the last trial. If one throws a die repeatedly until the third time a "1" appears, then the probability distribution of the number of non-"1"s that appear before the third "1" is a negative binomial distribution. References ---------- .. [1] Weisstein, Eric W. "Negative Binomial Distribution." From MathWorld--A Wolfram Web Resource. 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.negative_binomial(1, 0.1, 100000) >>> for i in range(1, 11): # doctest: +SKIP ... probability = sum(s>> n, p = 10, .5 # number of trials, probability of each trial >>> s = np.random.binomial(n, p, 1000) # result of flipping a coin 10 times, tested 1000 times. A real world example. A company drills 9 wild-cat oil exploration wells, each with an estimated probability of success of 0.1. All nine wells fail. What is the probability of that happening? Let's do 20,000 trials of the model, and count the number that generate zero positive results. >>> sum(np.random.binomial(9, 0.1, 20000) == 0)/20000. # answer = 0.38885, or 38%. triangular(left, mode, right, size=None) Draw samples from the triangular distribution over the interval ``[left, right]``. The triangular distribution is a continuous probability distribution with lower limit left, peak at mode, and upper limit right. Unlike the other distributions, these parameters directly define the shape of the pdf. .. note:: New code should use the `~numpy.random.Generator.triangular` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- left : float or array_like of floats Lower limit. mode : float or array_like of floats The value where the peak of the distribution occurs. The value must fulfill the condition ``left <= mode <= right``. right : float or array_like of floats Upper limit, must be larger than `left`. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``left``, ``mode``, and ``right`` are all scalars. Otherwise, ``np.broadcast(left, mode, right).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized triangular distribution. See Also -------- random.Generator.triangular: which should be used for new code. Notes ----- The probability density function for the triangular distribution is .. math:: P(x;l, m, r) = \begin{cases} \frac{2(x-l)}{(r-l)(m-l)}& \text{for $l \leq x \leq m$},\\ \frac{2(r-x)}{(r-l)(r-m)}& \text{for $m \leq x \leq r$},\\ 0& \text{otherwise}. \end{cases} The triangular distribution is often used in ill-defined problems where the underlying distribution is not known, but some knowledge of the limits and mode exists. Often it is used in simulations. References ---------- .. [1] Wikipedia, "Triangular distribution" https://en.wikipedia.org/wiki/Triangular_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.triangular(-3, 0, 8, 100000), bins=200, ... density=True) >>> plt.show() wald(mean, scale, size=None) Draw samples from a Wald, or inverse Gaussian, distribution. As the scale approaches infinity, the distribution becomes more like a Gaussian. Some references claim that the Wald is an inverse Gaussian with mean equal to 1, but this is by no means universal. The inverse Gaussian distribution was first studied in relationship to Brownian motion. In 1956 M.C.K. Tweedie used the name inverse Gaussian because there is an inverse relationship between the time to cover a unit distance and distance covered in unit time. .. note:: New code should use the `~numpy.random.Generator.wald` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : float or array_like of floats Distribution mean, must be > 0. scale : float or array_like of floats Scale parameter, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(mean, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Wald distribution. See Also -------- random.Generator.wald: which should be used for new code. Notes ----- The probability density function for the Wald distribution is .. math:: P(x;mean,scale) = \sqrt{\frac{scale}{2\pi x^3}}e^ \frac{-scale(x-mean)^2}{2\cdotp mean^2x} As noted above the inverse Gaussian distribution first arise from attempts to model Brownian motion. It is also a competitor to the Weibull for use in reliability modeling and modeling stock returns and interest rate processes. References ---------- .. [1] Brighton Webs Ltd., Wald Distribution, https://web.archive.org/web/20090423014010/http://www.brighton-webs.co.uk:80/distributions/wald.asp .. [2] Chhikara, Raj S., and Folks, J. Leroy, "The Inverse Gaussian Distribution: Theory : Methodology, and Applications", CRC Press, 1988. .. [3] Wikipedia, "Inverse Gaussian distribution" https://en.wikipedia.org/wiki/Inverse_Gaussian_distribution Examples -------- Draw values from the distribution and plot the histogram: >>> import matplotlib.pyplot as plt >>> h = plt.hist(np.random.wald(3, 2, 100000), bins=200, density=True) >>> plt.show() rayleigh(scale=1.0, size=None) Draw samples from a Rayleigh distribution. The :math:`\chi` and Weibull distributions are generalizations of the Rayleigh. .. note:: New code should use the `~numpy.random.Generator.rayleigh` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats, optional Scale, also equals the mode. Must be non-negative. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Rayleigh distribution. See Also -------- random.Generator.rayleigh: which should be used for new code. Notes ----- The probability density function for the Rayleigh distribution is .. math:: P(x;scale) = \frac{x}{scale^2}e^{\frac{-x^2}{2 \cdotp scale^2}} The Rayleigh distribution would arise, for example, if the East and North components of the wind velocity had identical zero-mean Gaussian distributions. Then the wind speed would have a Rayleigh distribution. References ---------- .. [1] Brighton Webs Ltd., "Rayleigh Distribution," https://web.archive.org/web/20090514091424/http://brighton-webs.co.uk:80/distributions/rayleigh.asp .. [2] Wikipedia, "Rayleigh distribution" https://en.wikipedia.org/wiki/Rayleigh_distribution Examples -------- Draw values from the distribution and plot the histogram >>> from matplotlib.pyplot import hist >>> values = hist(np.random.rayleigh(3, 100000), bins=200, density=True) Wave heights tend to follow a Rayleigh distribution. If the mean wave height is 1 meter, what fraction of waves are likely to be larger than 3 meters? >>> meanvalue = 1 >>> modevalue = np.sqrt(2 / np.pi) * meanvalue >>> s = np.random.rayleigh(modevalue, 1000000) The percentage of waves larger than 3 meters is: >>> 100.*sum(s>3)/1000000. 0.087300000000000003 # random lognormal(mean=0.0, sigma=1.0, size=None) Draw samples from a log-normal distribution. Draw samples from a log-normal distribution with specified mean, standard deviation, and array shape. Note that the mean and standard deviation are not the values for the distribution itself, but of the underlying normal distribution it is derived from. .. note:: New code should use the `~numpy.random.Generator.lognormal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mean : float or array_like of floats, optional Mean value of the underlying normal distribution. Default is 0. sigma : float or array_like of floats, optional Standard deviation of the underlying normal distribution. Must be non-negative. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mean`` and ``sigma`` are both scalars. Otherwise, ``np.broadcast(mean, sigma).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized log-normal distribution. See Also -------- scipy.stats.lognorm : probability density function, distribution, cumulative density function, etc. random.Generator.lognormal: which should be used for new code. Notes ----- A variable `x` has a log-normal distribution if `log(x)` is normally distributed. The probability density function for the log-normal distribution is: .. math:: p(x) = \frac{1}{\sigma x \sqrt{2\pi}} e^{(-\frac{(ln(x)-\mu)^2}{2\sigma^2})} where :math:`\mu` is the mean and :math:`\sigma` is the standard deviation of the normally distributed logarithm of the variable. A log-normal distribution results if a random variable is the *product* of a large number of independent, identically-distributed variables in the same way that a normal distribution results if the variable is the *sum* of a large number of independent, identically-distributed variables. References ---------- .. [1] Limpert, E., Stahel, W. A., and Abbt, M., "Log-normal Distributions across the Sciences: Keys and Clues," BioScience, Vol. 51, No. 5, May, 2001. https://stat.ethz.ch/~stahel/lognormal/bioscience.pdf .. [2] Reiss, R.D. and Thomas, M., "Statistical Analysis of Extreme Values," Basel: Birkhauser Verlag, 2001, pp. 31-32. Examples -------- Draw samples from the distribution: >>> mu, sigma = 3., 1. # mean and standard deviation >>> s = np.random.lognormal(mu, sigma, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 100, density=True, align='mid') >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, linewidth=2, color='r') >>> plt.axis('tight') >>> plt.show() Demonstrate that taking the products of random samples from a uniform distribution can be fit well by a log-normal probability density function. >>> # Generate a thousand samples: each is the product of 100 random >>> # values, drawn from a normal distribution. >>> b = [] >>> for i in range(1000): ... a = 10. + np.random.standard_normal(100) ... b.append(np.prod(a)) >>> b = np.array(b) / np.min(b) # scale values to be positive >>> count, bins, ignored = plt.hist(b, 100, density=True, align='mid') >>> sigma = np.std(np.log(b)) >>> mu = np.mean(np.log(b)) >>> x = np.linspace(min(bins), max(bins), 10000) >>> pdf = (np.exp(-(np.log(x) - mu)**2 / (2 * sigma**2)) ... / (x * sigma * np.sqrt(2 * np.pi))) >>> plt.plot(x, pdf, color='r', linewidth=2) >>> plt.show() logistic(loc=0.0, scale=1.0, size=None) Draw samples from a logistic distribution. Samples are drawn from a logistic distribution with specified parameters, loc (location or mean, also median), and scale (>0). .. note:: New code should use the `~numpy.random.Generator.logistic` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats, optional Parameter of the distribution. Default is 0. scale : float or array_like of floats, optional Parameter of the distribution. Must be non-negative. Default is 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized logistic distribution. See Also -------- scipy.stats.logistic : probability density function, distribution or cumulative density function, etc. random.Generator.logistic: which should be used for new code. Notes ----- The probability density for the Logistic distribution is .. math:: P(x) = P(x) = \frac{e^{-(x-\mu)/s}}{s(1+e^{-(x-\mu)/s})^2}, where :math:`\mu` = location and :math:`s` = scale. The Logistic distribution is used in Extreme Value problems where it can act as a mixture of Gumbel distributions, in Epidemiology, and by the World Chess Federation (FIDE) where it is used in the Elo ranking system, assuming the performance of each player is a logistically distributed random variable. References ---------- .. [1] Reiss, R.-D. and Thomas M. (2001), "Statistical Analysis of Extreme Values, from Insurance, Finance, Hydrology and Other Fields," Birkhauser Verlag, Basel, pp 132-133. .. [2] Weisstein, Eric W. "Logistic Distribution." From MathWorld--A Wolfram Web Resource. 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.logistic(loc, scale, 10000) >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=50) # plot against distribution >>> def logist(x, loc, scale): ... return np.exp((loc-x)/scale)/(scale*(1+np.exp((loc-x)/scale))**2) >>> lgst_val = logist(bins, loc, scale) >>> plt.plot(bins, lgst_val * count.max() / lgst_val.max()) >>> plt.show() gumbel(loc=0.0, scale=1.0, size=None) Draw samples from a Gumbel distribution. Draw samples from a Gumbel distribution with specified location and scale. For more information on the Gumbel distribution, see Notes and References below. .. note:: New code should use the `~numpy.random.Generator.gumbel` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats, optional The location of the mode of the distribution. Default is 0. scale : float or array_like of floats, optional The scale parameter of the distribution. Default is 1. Must be non- negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Gumbel distribution. See Also -------- scipy.stats.gumbel_l scipy.stats.gumbel_r scipy.stats.genextreme weibull random.Generator.gumbel: which should be used for new code. Notes ----- The Gumbel (or Smallest Extreme Value (SEV) or the Smallest Extreme Value Type I) distribution is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. The Gumbel is a special case of the Extreme Value Type I distribution for maximums from distributions with "exponential-like" tails. The probability density for the Gumbel distribution is .. math:: p(x) = \frac{e^{-(x - \mu)/ \beta}}{\beta} e^{ -e^{-(x - \mu)/ \beta}}, where :math:`\mu` is the mode, a location parameter, and :math:`\beta` is the scale parameter. The Gumbel (named for German mathematician Emil Julius Gumbel) was used very early in the hydrology literature, for modeling the occurrence of flood events. It is also used for modeling maximum wind speed and rainfall rates. It is a "fat-tailed" distribution - the probability of an event in the tail of the distribution is larger than if one used a Gaussian, hence the surprisingly frequent occurrence of 100-year floods. Floods were initially modeled as a Gaussian process, which underestimated the frequency of extreme events. It is one of a class of extreme value distributions, the Generalized Extreme Value (GEV) distributions, which also includes the Weibull and Frechet. The function has a mean of :math:`\mu + 0.57721\beta` and a variance of :math:`\frac{\pi^2}{6}\beta^2`. References ---------- .. [1] Gumbel, E. J., "Statistics of Extremes," New York: Columbia University Press, 1958. .. [2] Reiss, R.-D. and Thomas, M., "Statistical Analysis of Extreme Values from Insurance, Finance, Hydrology and Other Fields," Basel: Birkhauser Verlag, 2001. Examples -------- Draw samples from the distribution: >>> mu, beta = 0, 0.1 # location and scale >>> s = np.random.gumbel(mu, beta, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=True) >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp( -np.exp( -(bins - mu) /beta) ), ... linewidth=2, color='r') >>> plt.show() Show how an extreme value distribution can arise from a Gaussian process and compare to a Gaussian: >>> means = [] >>> maxima = [] >>> for i in range(0,1000) : ... a = np.random.normal(mu, beta, 1000) ... means.append(a.mean()) ... maxima.append(a.max()) >>> count, bins, ignored = plt.hist(maxima, 30, density=True) >>> beta = np.std(maxima) * np.sqrt(6) / np.pi >>> mu = np.mean(maxima) - 0.57721*beta >>> plt.plot(bins, (1/beta)*np.exp(-(bins - mu)/beta) ... * np.exp(-np.exp(-(bins - mu)/beta)), ... linewidth=2, color='r') >>> plt.plot(bins, 1/(beta * np.sqrt(2 * np.pi)) ... * np.exp(-(bins - mu)**2 / (2 * beta**2)), ... linewidth=2, color='g') >>> plt.show() laplace(loc=0.0, scale=1.0, size=None) Draw samples from the Laplace or double exponential distribution with specified location (or mean) and scale (decay). The Laplace distribution is similar to the Gaussian/normal distribution, but is sharper at the peak and has fatter tails. It represents the difference between two independent, identically distributed exponential random variables. .. note:: New code should use the `~numpy.random.Generator.laplace` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats, optional The position, :math:`\mu`, of the distribution peak. Default is 0. scale : float or array_like of floats, optional :math:`\lambda`, the exponential decay. Default is 1. Must be non- negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Laplace distribution. See Also -------- random.Generator.laplace: which should be used for new code. Notes ----- It has the probability density function .. math:: f(x; \mu, \lambda) = \frac{1}{2\lambda} \exp\left(-\frac{|x - \mu|}{\lambda}\right). The first law of Laplace, from 1774, states that the frequency of an error can be expressed as an exponential function of the absolute magnitude of the error, which leads to the Laplace distribution. For many problems in economics and health sciences, this distribution seems to model the data better than the standard Gaussian distribution. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] Kotz, Samuel, et. al. "The Laplace Distribution and Generalizations, " Birkhauser, 2001. .. [3] Weisstein, Eric W. "Laplace Distribution." From MathWorld--A Wolfram Web Resource. 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.laplace(loc, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=True) >>> x = np.arange(-8., 8., .01) >>> pdf = np.exp(-abs(x-loc)/scale)/(2.*scale) >>> plt.plot(x, pdf) Plot Gaussian for comparison: >>> g = (1/(scale * np.sqrt(2 * np.pi)) * ... np.exp(-(x - loc)**2 / (2 * scale**2))) >>> plt.plot(x,g) power(a, size=None) Draws samples in [0, 1] from a power distribution with positive exponent a - 1. Also known as the power function distribution. .. note:: New code should use the `~numpy.random.Generator.power` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Parameter of the distribution. Must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized power distribution. Raises ------ ValueError If a <= 0. See Also -------- random.Generator.power: which should be used for new code. Notes ----- The probability density function is .. math:: P(x; a) = ax^{a-1}, 0 \le x \le 1, a>0. The power function distribution is just the inverse of the Pareto distribution. It may also be seen as a special case of the Beta distribution. It is used, for example, in modeling the over-reporting of insurance claims. References ---------- .. [1] Christian Kleiber, Samuel Kotz, "Statistical size distributions in economics and actuarial sciences", Wiley, 2003. .. [2] Heckert, N. A. and Filliben, James J. "NIST Handbook 148: Dataplot Reference Manual, Volume 2: Let Subcommands and Library Functions", National Institute of Standards and Technology Handbook Series, June 2003. https://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/powpdf.pdf Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> samples = 1000 >>> s = np.random.power(a, samples) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, bins=30) >>> x = np.linspace(0, 1, 100) >>> y = a*x**(a-1.) >>> normed_y = samples*np.diff(bins)[0]*y >>> plt.plot(x, normed_y) >>> plt.show() Compare the power function distribution to the inverse of the Pareto. >>> from scipy import stats # doctest: +SKIP >>> rvs = np.random.power(5, 1000000) >>> rvsp = np.random.pareto(5, 1000000) >>> xx = np.linspace(0,1,100) >>> powpdf = stats.powerlaw.pdf(xx,5) # doctest: +SKIP >>> plt.figure() >>> plt.hist(rvs, bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('np.random.power(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('inverse of 1 + np.random.pareto(5)') >>> plt.figure() >>> plt.hist(1./(1.+rvsp), bins=50, density=True) >>> plt.plot(xx,powpdf,'r-') # doctest: +SKIP >>> plt.title('inverse of stats.pareto(5)') weibull(a, size=None) Draw samples from a Weibull distribution. Draw samples from a 1-parameter Weibull distribution with the given shape parameter `a`. .. math:: X = (-ln(U))^{1/a} Here, U is drawn from the uniform distribution over (0,1]. The more common 2-parameter Weibull, including a scale parameter :math:`\lambda` is just :math:`X = \lambda(-ln(U))^{1/a}`. .. note:: New code should use the `~numpy.random.Generator.weibull` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape parameter of the distribution. Must be nonnegative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Weibull distribution. See Also -------- scipy.stats.weibull_max scipy.stats.weibull_min scipy.stats.genextreme gumbel random.Generator.weibull: which should be used for new code. Notes ----- The Weibull (or Type III asymptotic extreme value distribution for smallest values, SEV Type III, or Rosin-Rammler distribution) is one of a class of Generalized Extreme Value (GEV) distributions used in modeling extreme value problems. This class includes the Gumbel and Frechet distributions. The probability density for the Weibull distribution is .. math:: p(x) = \frac{a} {\lambda}(\frac{x}{\lambda})^{a-1}e^{-(x/\lambda)^a}, where :math:`a` is the shape and :math:`\lambda` the scale. The function has its peak (the mode) at :math:`\lambda(\frac{a-1}{a})^{1/a}`. When ``a = 1``, the Weibull distribution reduces to the exponential distribution. References ---------- .. [1] Waloddi Weibull, Royal Technical University, Stockholm, 1939 "A Statistical Theory Of The Strength Of Materials", Ingeniorsvetenskapsakademiens Handlingar Nr 151, 1939, Generalstabens Litografiska Anstalts Forlag, Stockholm. .. [2] Waloddi Weibull, "A Statistical Distribution Function of Wide Applicability", Journal Of Applied Mechanics ASME Paper 1951. .. [3] Wikipedia, "Weibull distribution", https://en.wikipedia.org/wiki/Weibull_distribution Examples -------- Draw samples from the distribution: >>> a = 5. # shape >>> s = np.random.weibull(a, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> x = np.arange(1,100.)/50. >>> def weib(x,n,a): ... return (a / n) * (x / n)**(a - 1) * np.exp(-(x / n)**a) >>> count, bins, ignored = plt.hist(np.random.weibull(5.,1000)) >>> x = np.arange(1,100.)/50. >>> scale = count.max()/weib(x, 1., 5.).max() >>> plt.plot(x, weib(x, 1., 5.)*scale) >>> plt.show() pareto(a, size=None) Draw samples from a Pareto II or Lomax distribution with specified shape. The Lomax or Pareto II distribution is a shifted Pareto distribution. The classical Pareto distribution can be obtained from the Lomax distribution by adding 1 and multiplying by the scale parameter ``m`` (see Notes). The smallest value of the Lomax distribution is zero while for the classical Pareto distribution it is ``mu``, where the standard Pareto distribution has location ``mu = 1``. Lomax can also be considered as a simplified version of the Generalized Pareto distribution (available in SciPy), with the scale set to one and the location set to zero. The Pareto distribution must be greater than zero, and is unbounded above. It is also known as the "80-20 rule". In this distribution, 80 percent of the weights are in the lowest 20 percent of the range, while the other 20 percent fill the remaining 80 percent of the range. .. note:: New code should use the `~numpy.random.Generator.pareto` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Shape of the distribution. Must be positive. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` is a scalar. Otherwise, ``np.array(a).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Pareto distribution. See Also -------- scipy.stats.lomax : probability density function, distribution or cumulative density function, etc. scipy.stats.genpareto : probability density function, distribution or cumulative density function, etc. random.Generator.pareto: which should be used for new code. Notes ----- The probability density for the Pareto distribution is .. math:: p(x) = \frac{am^a}{x^{a+1}} where :math:`a` is the shape and :math:`m` the scale. The Pareto distribution, named after the Italian economist Vilfredo Pareto, is a power law probability distribution useful in many real world problems. Outside the field of economics it is generally referred to as the Bradford distribution. Pareto developed the distribution to describe the distribution of wealth in an economy. It has also found use in insurance, web page access statistics, oil field sizes, and many other problems, including the download frequency for projects in Sourceforge [1]_. It is one of the so-called "fat-tailed" distributions. References ---------- .. [1] Francis Hunt and Paul Johnson, On the Pareto Distribution of Sourceforge projects. .. [2] Pareto, V. (1896). Course of Political Economy. Lausanne. .. [3] Reiss, R.D., Thomas, M.(2001), Statistical Analysis of Extreme Values, Birkhauser Verlag, Basel, pp 23-30. .. [4] Wikipedia, "Pareto distribution", https://en.wikipedia.org/wiki/Pareto_distribution Examples -------- Draw samples from the distribution: >>> a, m = 3., 2. # shape and mode >>> s = (np.random.pareto(a, 1000) + 1) * m Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, _ = plt.hist(s, 100, density=True) >>> fit = a*m**a / bins**(a+1) >>> plt.plot(bins, max(count)*fit/max(fit), linewidth=2, color='r') >>> plt.show() vonmises(mu, kappa, size=None) Draw samples from a von Mises distribution. Samples are drawn from a von Mises distribution with specified mode (mu) and dispersion (kappa), on the interval [-pi, pi]. The von Mises distribution (also known as the circular normal distribution) is a continuous probability distribution on the unit circle. It may be thought of as the circular analogue of the normal distribution. .. note:: New code should use the `~numpy.random.Generator.vonmises` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- mu : float or array_like of floats Mode ("center") of the distribution. kappa : float or array_like of floats Dispersion of the distribution, has to be >=0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``mu`` and ``kappa`` are both scalars. Otherwise, ``np.broadcast(mu, kappa).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized von Mises distribution. See Also -------- scipy.stats.vonmises : probability density function, distribution, or cumulative density function, etc. random.Generator.vonmises: which should be used for new code. Notes ----- The probability density for the von Mises distribution is .. math:: p(x) = \frac{e^{\kappa cos(x-\mu)}}{2\pi I_0(\kappa)}, where :math:`\mu` is the mode and :math:`\kappa` the dispersion, and :math:`I_0(\kappa)` is the modified Bessel function of order 0. The von Mises is named for Richard Edler von Mises, who was born in Austria-Hungary, in what is now the Ukraine. He fled to the United States in 1939 and became a professor at Harvard. He worked in probability theory, aerodynamics, fluid mechanics, and philosophy of science. References ---------- .. [1] Abramowitz, M. and Stegun, I. A. (Eds.). "Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing," New York: Dover, 1972. .. [2] von Mises, R., "Mathematical Theory of Probability and Statistics", New York: Academic Press, 1964. Examples -------- Draw samples from the distribution: >>> mu, kappa = 0.0, 4.0 # mean and dispersion >>> s = np.random.vonmises(mu, kappa, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> from scipy.special import i0 # doctest: +SKIP >>> plt.hist(s, 50, density=True) >>> x = np.linspace(-np.pi, np.pi, num=51) >>> y = np.exp(kappa*np.cos(x-mu))/(2*np.pi*i0(kappa)) # doctest: +SKIP >>> plt.plot(x, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() standard_t(df, size=None) Draw samples from a standard Student's t distribution with `df` degrees of freedom. A special case of the hyperbolic distribution. As `df` gets large, the result resembles that of the standard normal distribution (`standard_normal`). .. note:: New code should use the `~numpy.random.Generator.standard_t` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard Student's t distribution. See Also -------- random.Generator.standard_t: which should be used for new code. Notes ----- The probability density function for the t distribution is .. math:: P(x, df) = \frac{\Gamma(\frac{df+1}{2})}{\sqrt{\pi df} \Gamma(\frac{df}{2})}\Bigl( 1+\frac{x^2}{df} \Bigr)^{-(df+1)/2} The t test is based on an assumption that the data come from a Normal distribution. The t test provides a way to test whether the sample mean (that is the mean calculated from the data) is a good estimate of the true mean. The derivation of the t-distribution was first published in 1908 by William Gosset while working for the Guinness Brewery in Dublin. Due to proprietary issues, he had to publish under a pseudonym, and so he used the name Student. References ---------- .. [1] Dalgaard, Peter, "Introductory Statistics With R", Springer, 2002. .. [2] Wikipedia, "Student's t-distribution" https://en.wikipedia.org/wiki/Student's_t-distribution Examples -------- From Dalgaard page 83 [1]_, suppose the daily energy intake for 11 women in kilojoules (kJ) is: >>> intake = np.array([5260., 5470, 5640, 6180, 6390, 6515, 6805, 7515, \ ... 7515, 8230, 8770]) Does their energy intake deviate systematically from the recommended value of 7725 kJ? Our null hypothesis will be the absence of deviation, and the alternate hypothesis will be the presence of an effect that could be either positive or negative, hence making our test 2-tailed. Because we are estimating the mean and we have N=11 values in our sample, we have N-1=10 degrees of freedom. We set our significance level to 95% and compute the t statistic using the empirical mean and empirical standard deviation of our intake. We use a ddof of 1 to base the computation of our empirical standard deviation on an unbiased estimate of the variance (note: the final estimate is not unbiased due to the concave nature of the square root). >>> np.mean(intake) 6753.636363636364 >>> intake.std(ddof=1) 1142.1232221373727 >>> t = (np.mean(intake)-7725)/(intake.std(ddof=1)/np.sqrt(len(intake))) >>> t -2.8207540608310198 We draw 1000000 samples from Student's t distribution with the adequate degrees of freedom. >>> import matplotlib.pyplot as plt >>> s = np.random.standard_t(10, size=1000000) >>> h = plt.hist(s, bins=100, density=True) Does our t statistic land in one of the two critical regions found at both tails of the distribution? >>> np.sum(np.abs(t) < np.abs(s)) / float(len(s)) 0.018318 #random < 0.05, statistic is in critical region The probability value for this 2-tailed test is about 1.83%, which is lower than the 5% pre-determined significance threshold. Therefore, the probability of observing values as extreme as our intake conditionally on the null hypothesis being true is too low, and we reject the null hypothesis of no deviation. standard_cauchy(size=None) Draw samples from a standard Cauchy distribution with mode = 0. Also known as the Lorentz distribution. .. note:: New code should use the `~numpy.random.Generator.standard_cauchy` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- samples : ndarray or scalar The drawn samples. See Also -------- random.Generator.standard_cauchy: which should be used for new code. Notes ----- The probability density function for the full Cauchy distribution is .. math:: P(x; x_0, \gamma) = \frac{1}{\pi \gamma \bigl[ 1+ (\frac{x-x_0}{\gamma})^2 \bigr] } and the Standard Cauchy distribution just sets :math:`x_0=0` and :math:`\gamma=1` The Cauchy distribution arises in the solution to the driven harmonic oscillator problem, and also describes spectral line broadening. It also describes the distribution of values at which a line tilted at a random angle will cut the x axis. When studying hypothesis tests that assume normality, seeing how the tests perform on data from a Cauchy distribution is a good indicator of their sensitivity to a heavy-tailed distribution, since the Cauchy looks very much like a Gaussian distribution, but with heavier tails. References ---------- .. [1] NIST/SEMATECH e-Handbook of Statistical Methods, "Cauchy Distribution", https://www.itl.nist.gov/div898/handbook/eda/section3/eda3663.htm .. [2] Weisstein, Eric W. "Cauchy Distribution." From MathWorld--A Wolfram Web Resource. 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.standard_cauchy(1000000) >>> s = s[(s>-25) & (s<25)] # truncate distribution so it plots well >>> plt.hist(s, bins=100) >>> plt.show() noncentral_chisquare(df, nonc, size=None) Draw samples from a noncentral chi-square distribution. The noncentral :math:`\chi^2` distribution is a generalization of the :math:`\chi^2` distribution. .. note:: New code should use the `~numpy.random.Generator.noncentral_chisquare` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Degrees of freedom, must be > 0. .. versionchanged:: 1.10.0 Earlier NumPy versions required dfnum > 1. nonc : float or array_like of floats Non-centrality, must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` and ``nonc`` are both scalars. Otherwise, ``np.broadcast(df, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral chi-square distribution. See Also -------- random.Generator.noncentral_chisquare: which should be used for new code. Notes ----- The probability density function for the noncentral Chi-square distribution is .. math:: P(x;df,nonc) = \sum^{\infty}_{i=0} \frac{e^{-nonc/2}(nonc/2)^{i}}{i!} P_{Y_{df+2i}}(x), where :math:`Y_{q}` is the Chi-square with q degrees of freedom. References ---------- .. [1] Wikipedia, "Noncentral chi-squared distribution" https://en.wikipedia.org/wiki/Noncentral_chi-squared_distribution Examples -------- Draw values from the distribution and plot the histogram >>> import matplotlib.pyplot as plt >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, density=True) >>> plt.show() Draw values from a noncentral chisquare with very small noncentrality, and compare to a chisquare. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, .0000001, 100000), ... bins=np.arange(0., 25, .1), density=True) >>> values2 = plt.hist(np.random.chisquare(3, 100000), ... bins=np.arange(0., 25, .1), density=True) >>> plt.plot(values[1][0:-1], values[0]-values2[0], 'ob') >>> plt.show() Demonstrate how large values of non-centrality lead to a more symmetric distribution. >>> plt.figure() >>> values = plt.hist(np.random.noncentral_chisquare(3, 20, 100000), ... bins=200, density=True) >>> plt.show() chisquare(df, size=None) Draw samples from a chi-square distribution. When `df` independent random variables, each with standard normal distributions (mean 0, variance 1), are squared and summed, the resulting distribution is chi-square (see Notes). This distribution is often used in hypothesis testing. .. note:: New code should use the `~numpy.random.Generator.chisquare` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- df : float or array_like of floats Number of degrees of freedom, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``df`` is a scalar. Otherwise, ``np.array(df).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized chi-square distribution. Raises ------ ValueError When `df` <= 0 or when an inappropriate `size` (e.g. ``size=-1``) is given. See Also -------- random.Generator.chisquare: which should be used for new code. Notes ----- The variable obtained by summing the squares of `df` independent, standard normally distributed random variables: .. math:: Q = \sum_{i=0}^{\mathtt{df}} X^2_i is chi-square distributed, denoted .. math:: Q \sim \chi^2_k. The probability density function of the chi-squared distribution is .. math:: p(x) = \frac{(1/2)^{k/2}}{\Gamma(k/2)} x^{k/2 - 1} e^{-x/2}, where :math:`\Gamma` is the gamma function, .. math:: \Gamma(x) = \int_0^{-\infty} t^{x - 1} e^{-t} dt. References ---------- .. [1] NIST "Engineering Statistics Handbook" https://www.itl.nist.gov/div898/handbook/eda/section3/eda3666.htm Examples -------- >>> np.random.chisquare(2,4) array([ 1.89920014, 9.00867716, 3.13710533, 5.62318272]) # random noncentral_f(dfnum, dfden, nonc, size=None) Draw samples from the noncentral F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters > 1. `nonc` is the non-centrality parameter. .. note:: New code should use the `~numpy.random.Generator.noncentral_f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Numerator degrees of freedom, must be > 0. .. versionchanged:: 1.14.0 Earlier NumPy versions required dfnum > 1. dfden : float or array_like of floats Denominator degrees of freedom, must be > 0. nonc : float or array_like of floats Non-centrality parameter, the sum of the squares of the numerator means, must be >= 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum``, ``dfden``, and ``nonc`` are all scalars. Otherwise, ``np.broadcast(dfnum, dfden, nonc).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized noncentral Fisher distribution. See Also -------- random.Generator.noncentral_f: which should be used for new code. Notes ----- When calculating the power of an experiment (power = probability of rejecting the null hypothesis when a specific alternative is true) the non-central F statistic becomes important. When the null hypothesis is true, the F statistic follows a central F distribution. When the null hypothesis is not true, then it follows a non-central F statistic. References ---------- .. [1] Weisstein, Eric W. "Noncentral F-Distribution." From MathWorld--A Wolfram Web Resource. 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. >>> dfnum = 3 # between group deg of freedom >>> dfden = 20 # within groups degrees of freedom >>> nonc = 3.0 >>> nc_vals = np.random.noncentral_f(dfnum, dfden, nonc, 1000000) >>> NF = np.histogram(nc_vals, bins=50, density=True) >>> c_vals = np.random.f(dfnum, dfden, 1000000) >>> F = np.histogram(c_vals, bins=50, density=True) >>> import matplotlib.pyplot as plt >>> plt.plot(F[1][1:], F[0]) >>> plt.plot(NF[1][1:], NF[0]) >>> plt.show() f(dfnum, dfden, size=None) Draw samples from an F distribution. Samples are drawn from an F distribution with specified parameters, `dfnum` (degrees of freedom in numerator) and `dfden` (degrees of freedom in denominator), where both parameters must be greater than zero. The random variate of the F distribution (also known as the Fisher distribution) is a continuous probability distribution that arises in ANOVA tests, and is the ratio of two chi-square variates. .. note:: New code should use the `~numpy.random.Generator.f` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- dfnum : float or array_like of floats Degrees of freedom in numerator, must be > 0. dfden : float or array_like of float Degrees of freedom in denominator, must be > 0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``dfnum`` and ``dfden`` are both scalars. Otherwise, ``np.broadcast(dfnum, dfden).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized Fisher distribution. See Also -------- scipy.stats.f : probability density function, distribution or cumulative density function, etc. random.Generator.f: which should be used for new code. Notes ----- The F statistic is used to compare in-group variances to between-group variances. Calculating the distribution depends on the sampling, and so it is a function of the respective degrees of freedom in the problem. The variable `dfnum` is the number of samples minus one, the between-groups degrees of freedom, while `dfden` is the within-groups degrees of freedom, the sum of the number of samples in each group minus the number of groups. References ---------- .. [1] Glantz, Stanton A. "Primer of Biostatistics.", McGraw-Hill, Fifth Edition, 2002. .. [2] Wikipedia, "F-distribution", https://en.wikipedia.org/wiki/F-distribution Examples -------- An example from Glantz[1], pp 47-40: Two groups, children of diabetics (25 people) and children from people without diabetes (25 controls). Fasting blood glucose was measured, case group had a mean value of 86.1, controls had a mean value of 82.2. Standard deviations were 2.09 and 2.49 respectively. Are these data consistent with the null hypothesis that the parents diabetic status does not affect their children's blood glucose levels? Calculating the F statistic from the data gives a value of 36.01. Draw samples from the distribution: >>> dfnum = 1. # between group degrees of freedom >>> dfden = 48. # within groups degrees of freedom >>> s = np.random.f(dfnum, dfden, 1000) The lower bound for the top 1% of the samples is : >>> np.sort(s)[-10] 7.61988120985 # random So there is about a 1% chance that the F statistic will exceed 7.62, the measured value is 36, so the null hypothesis is rejected at the 1% level. gamma(shape, scale=1.0, size=None) Draw samples from a Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, `shape` (sometimes designated "k") and `scale` (sometimes designated "theta"), where both parameters are > 0. .. note:: New code should use the `~numpy.random.Generator.gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats The shape of the gamma distribution. Must be non-negative. scale : float or array_like of floats, optional The scale of the gamma distribution. Must be non-negative. Default is equal to 1. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(shape, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. random.Generator.gamma: which should be used for new code. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. 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.gamma(shape, scale, 1000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1)*(np.exp(-bins/scale) / # doctest: +SKIP ... (sps.gamma(shape)*scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() standard_gamma(shape, size=None) Draw samples from a standard Gamma distribution. Samples are drawn from a Gamma distribution with specified parameters, shape (sometimes designated "k") and scale=1. .. note:: New code should use the `~numpy.random.Generator.standard_gamma` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- shape : float or array_like of floats Parameter, must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``shape`` is a scalar. Otherwise, ``np.array(shape).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized standard gamma distribution. See Also -------- scipy.stats.gamma : probability density function, distribution or cumulative density function, etc. random.Generator.standard_gamma: which should be used for new code. Notes ----- The probability density for the Gamma distribution is .. math:: p(x) = x^{k-1}\frac{e^{-x/\theta}}{\theta^k\Gamma(k)}, where :math:`k` is the shape and :math:`\theta` the scale, and :math:`\Gamma` is the Gamma function. The Gamma distribution is often used to model the times to failure of electronic components, and arises naturally in processes for which the waiting times between Poisson distributed events are relevant. References ---------- .. [1] Weisstein, Eric W. "Gamma Distribution." From MathWorld--A Wolfram Web Resource. 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.standard_gamma(shape, 1000000) Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> import scipy.special as sps # doctest: +SKIP >>> count, bins, ignored = plt.hist(s, 50, density=True) >>> y = bins**(shape-1) * ((np.exp(-bins/scale))/ # doctest: +SKIP ... (sps.gamma(shape) * scale**shape)) >>> plt.plot(bins, y, linewidth=2, color='r') # doctest: +SKIP >>> plt.show() normal(loc=0.0, scale=1.0, size=None) Draw random samples from a normal (Gaussian) distribution. The probability density function of the normal distribution, first derived by De Moivre and 200 years later by both Gauss and Laplace independently [2]_, is often called the bell curve because of its characteristic shape (see the example below). The normal distributions occurs often in nature. For example, it describes the commonly occurring distribution of samples influenced by a large number of tiny, random disturbances, each with its own unique distribution [2]_. .. note:: New code should use the `~numpy.random.Generator.normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- loc : float or array_like of floats Mean ("centre") of the distribution. scale : float or array_like of floats Standard deviation (spread or "width") of the distribution. Must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``loc`` and ``scale`` are both scalars. Otherwise, ``np.broadcast(loc, scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized normal distribution. See Also -------- scipy.stats.norm : probability density function, distribution or cumulative density function, etc. random.Generator.normal: which should be used for new code. Notes ----- The probability density for the Gaussian distribution is .. math:: p(x) = \frac{1}{\sqrt{ 2 \pi \sigma^2 }} e^{ - \frac{ (x - \mu)^2 } {2 \sigma^2} }, where :math:`\mu` is the mean and :math:`\sigma` the standard deviation. The square of the standard deviation, :math:`\sigma^2`, is called the variance. The function has its peak at the mean, and its "spread" increases with the standard deviation (the function reaches 0.607 times its maximum at :math:`x + \sigma` and :math:`x - \sigma` [2]_). This implies that normal is more likely to return samples lying close to the mean, rather than those far away. References ---------- .. [1] Wikipedia, "Normal distribution", https://en.wikipedia.org/wiki/Normal_distribution .. [2] P. R. Peebles Jr., "Central Limit Theorem" in "Probability, Random Variables and Random Signal Principles", 4th ed., 2001, pp. 51, 51, 125. Examples -------- Draw samples from the distribution: >>> mu, sigma = 0, 0.1 # mean and standard deviation >>> s = np.random.normal(mu, sigma, 1000) Verify the mean and the variance: >>> abs(mu - np.mean(s)) 0.0 # may vary >>> abs(sigma - np.std(s, ddof=1)) 0.1 # may vary Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 30, density=True) >>> plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) * ... np.exp( - (bins - mu)**2 / (2 * sigma**2) ), ... linewidth=2, color='r') >>> plt.show() Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> np.random.normal(3, 2.5, size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random standard_normal(size=None) Draw samples from a standard Normal distribution (mean=0, stdev=1). .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray A floating-point array of shape ``size`` of drawn samples, or a single sample if ``size`` was not specified. See Also -------- normal : Equivalent function with additional ``loc`` and ``scale`` arguments for setting the mean and standard deviation. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use one of:: mu + sigma * np.random.standard_normal(size=...) np.random.normal(mu, sigma, size=...) Examples -------- >>> np.random.standard_normal() 2.1923875335537315 #random >>> s = np.random.standard_normal(8000) >>> s array([ 0.6888893 , 0.78096262, -0.89086505, ..., 0.49876311, # random -0.38672696, -0.4685006 ]) # random >>> s.shape (8000,) >>> s = np.random.standard_normal(size=(3, 4, 2)) >>> s.shape (3, 4, 2) Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.standard_normal(size=(2, 4)) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random random_integers(low, high=None, size=None) Random integers of type `np.int_` between `low` and `high`, inclusive. Return random integers of type `np.int_` from the "discrete uniform" distribution in the closed interval [`low`, `high`]. If `high` is None (the default), then results are from [1, `low`]. The `np.int_` type translates to the C long integer type and its precision is platform dependent. This function has been deprecated. Use randint instead. .. deprecated:: 1.11.0 Parameters ---------- low : int Lowest (signed) integer to be drawn from the distribution (unless ``high=None``, in which case this parameter is the *highest* such integer). high : int, optional If provided, the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- randint : Similar to `random_integers`, only for the half-open interval [`low`, `high`), and 0 is the lowest value if `high` is omitted. Notes ----- To sample from N evenly spaced floating-point numbers between a and b, use:: a + (b - a) * (np.random.random_integers(N) - 1) / (N - 1.) Examples -------- >>> np.random.random_integers(5) 4 # random >>> type(np.random.random_integers(5)) >>> np.random.random_integers(5, size=(3,2)) array([[5, 4], # random [3, 3], [4, 5]]) Choose five random numbers from the set of five evenly-spaced numbers between 0 and 2.5, inclusive (*i.e.*, from the set :math:`{0, 5/8, 10/8, 15/8, 20/8}`): >>> 2.5 * (np.random.random_integers(5, size=(5,)) - 1) / 4. array([ 0.625, 1.25 , 0.625, 0.625, 2.5 ]) # random Roll two six sided dice 1000 times and sum the results: >>> d1 = np.random.random_integers(1, 6, 1000) >>> d2 = np.random.random_integers(1, 6, 1000) >>> dsums = d1 + d2 Display results as a histogram: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(dsums, 11, density=True) >>> plt.show() randn(d0, d1, ..., dn) Return a sample (or samples) from the "standard normal" distribution. .. note:: This is a convenience function for users porting code from Matlab, and wraps `standard_normal`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. .. note:: New code should use the `~numpy.random.Generator.standard_normal` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. If positive int_like arguments are provided, `randn` generates an array of shape ``(d0, d1, ..., dn)``, filled with random floats sampled from a univariate "normal" (Gaussian) distribution of mean 0 and variance 1. A single float randomly sampled from the distribution is returned if no argument is provided. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, must be non-negative. If no argument is given a single Python float is returned. Returns ------- Z : ndarray or float A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from the standard normal distribution, or a single such float if no parameters were supplied. See Also -------- standard_normal : Similar, but takes a tuple as its argument. normal : Also accepts mu and sigma arguments. random.Generator.standard_normal: which should be used for new code. Notes ----- For random samples from the normal distribution with mean ``mu`` and standard deviation ``sigma``, use:: sigma * np.random.randn(...) + mu Examples -------- >>> np.random.randn() 2.1923875335537315 # random Two-by-four array of samples from the normal distribution with mean 3 and standard deviation 2.5: >>> 3 + 2.5 * np.random.randn(2, 4) array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677], # random [ 0.39924804, 4.68456316, 4.99394529, 4.84057254]]) # random rand(d0, d1, ..., dn) Random values in a given shape. .. note:: This is a convenience function for users porting code from Matlab, and wraps `random_sample`. That function takes a tuple to specify the size of the output, which is consistent with other NumPy functions like `numpy.zeros` and `numpy.ones`. Create an array of the given shape and populate it with random samples from a uniform distribution over ``[0, 1)``. Parameters ---------- d0, d1, ..., dn : int, optional The dimensions of the returned array, must be non-negative. If no argument is given a single Python float is returned. Returns ------- out : ndarray, shape ``(d0, d1, ..., dn)`` Random values. See Also -------- random Examples -------- >>> np.random.rand(3,2) array([[ 0.14022471, 0.96360618], #random [ 0.37601032, 0.25528411], #random [ 0.49313049, 0.94909878]]) #random uniform(low=0.0, high=1.0, size=None) Draw samples from a uniform distribution. Samples are uniformly distributed over the half-open interval ``[low, high)`` (includes low, but excludes high). In other words, any value within the given interval is equally likely to be drawn by `uniform`. .. note:: New code should use the `~numpy.random.Generator.uniform` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- low : float or array_like of floats, optional Lower boundary of the output interval. All values generated will be greater than or equal to low. The default value is 0. high : float or array_like of floats Upper boundary of the output interval. All values generated will be less than or equal to high. The high limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. The default value is 1.0. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``low`` and ``high`` are both scalars. Otherwise, ``np.broadcast(low, high).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized uniform distribution. See Also -------- randint : Discrete uniform distribution, yielding integers. random_integers : Discrete uniform distribution over the closed interval ``[low, high]``. random_sample : Floats uniformly distributed over ``[0, 1)``. random : Alias for `random_sample`. rand : Convenience function that accepts dimensions as input, e.g., ``rand(2,2)`` would generate a 2-by-2 array of floats, uniformly distributed over ``[0, 1)``. random.Generator.uniform: which should be used for new code. Notes ----- The probability density function of the uniform distribution is .. math:: p(x) = \frac{1}{b - a} anywhere within the interval ``[a, b)``, and zero elsewhere. When ``high`` == ``low``, values of ``low`` will be returned. If ``high`` < ``low``, the results are officially undefined and may eventually raise an error, i.e. do not rely on this function to behave when passed arguments satisfying that inequality condition. The ``high`` limit may be included in the returned array of floats due to floating-point rounding in the equation ``low + (high-low) * random_sample()``. For example: >>> x = np.float32(5*0.99999999) >>> x 5.0 Examples -------- Draw samples from the distribution: >>> s = np.random.uniform(-1,0,1000) All values are within the given interval: >>> np.all(s >= -1) True >>> np.all(s < 0) True Display the histogram of the samples, along with the probability density function: >>> import matplotlib.pyplot as plt >>> count, bins, ignored = plt.hist(s, 15, density=True) >>> plt.plot(bins, np.ones_like(bins), linewidth=2, color='r') >>> plt.show() choice(a, size=None, replace=True, p=None) Generates a random sample from a given 1-D array .. versionadded:: 1.7.0 .. note:: New code should use the `~numpy.random.Generator.choice` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : 1-D array-like or int If an ndarray, a random sample is generated from its elements. If an int, the random sample is generated as if it were ``np.arange(a)`` size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. replace : boolean, optional Whether the sample is with or without replacement. Default is True, meaning that a value of ``a`` can be selected multiple times. p : 1-D array-like, optional The probabilities associated with each entry in a. If not given, the sample assumes a uniform distribution over all entries in ``a``. Returns ------- samples : single item or ndarray The generated random samples Raises ------ ValueError If a is an int and less than zero, if a or p are not 1-dimensional, if a is an array-like of size 0, if p is not a vector of probabilities, if a and p have different lengths, or if replace=False and the sample size is greater than the population size See Also -------- randint, shuffle, permutation random.Generator.choice: which should be used in new code Notes ----- Setting user-specified probabilities through ``p`` uses a more general but less efficient sampler than the default. The general sampler produces a different sample than the optimized sampler even if each element of ``p`` is 1 / len(a). Sampling random rows from a 2-D array is not possible with this function, but is possible with `Generator.choice` through its ``axis`` keyword. Examples -------- Generate a uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3) array([0, 3, 4]) # random >>> #This is equivalent to np.random.randint(0,5,3) Generate a non-uniform random sample from np.arange(5) of size 3: >>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0]) array([3, 3, 0]) # random Generate a uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False) array([3,1,0]) # random >>> #This is equivalent to np.random.permutation(np.arange(5))[:3] Generate a non-uniform random sample from np.arange(5) of size 3 without replacement: >>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0]) array([2, 3, 0]) # random Any of the above can be repeated with an arbitrary array-like instead of just integers. For instance: >>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher'] >>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3]) array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'], # random dtype='>> np.random.bytes(10) b' eh\x85\x022SZ\xbf\xa4' #random randint(low, high=None, size=None, dtype=int) Return random integers from `low` (inclusive) to `high` (exclusive). Return random integers from the "discrete uniform" distribution of the specified dtype in the "half-open" interval [`low`, `high`). If `high` is None (the default), then results are from [0, `low`). .. note:: New code should use the `~numpy.random.Generator.integers` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- low : int or array-like of ints Lowest (signed) integers to be drawn from the distribution (unless ``high=None``, in which case this parameter is one above the *highest* such integer). high : int or array-like of ints, optional If provided, one above the largest (signed) integer to be drawn from the distribution (see above for behavior if ``high=None``). If array-like, must contain integer values size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. dtype : dtype, optional Desired dtype of the result. Byteorder must be native. The default value is int. .. versionadded:: 1.11.0 Returns ------- out : int or ndarray of ints `size`-shaped array of random integers from the appropriate distribution, or a single such random int if `size` not provided. See Also -------- random_integers : similar to `randint`, only for the closed interval [`low`, `high`], and 1 is the lowest value if `high` is omitted. random.Generator.integers: which should be used for new code. Examples -------- >>> np.random.randint(2, size=10) array([1, 0, 0, 0, 1, 1, 0, 0, 1, 0]) # random >>> np.random.randint(1, size=10) array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0]) Generate a 2 x 4 array of ints between 0 and 4, inclusive: >>> np.random.randint(5, size=(2, 4)) array([[4, 0, 2, 1], # random [3, 2, 2, 0]]) Generate a 1 x 3 array with 3 different upper bounds >>> np.random.randint(1, [3, 5, 10]) array([2, 2, 9]) # random Generate a 1 by 3 array with 3 different lower bounds >>> np.random.randint([1, 5, 7], 10) array([9, 8, 7]) # random Generate a 2 by 4 array using broadcasting with dtype of uint8 >>> np.random.randint([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7], # random [ 1, 16, 9, 12]], dtype=uint8) tomaxint(size=None) Return a sample of uniformly distributed random integers in the interval [0, ``np.iinfo(np.int_).max``]. The `np.int_` type translates to the C long integer type and its precision is platform dependent. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : ndarray Drawn samples, with shape `size`. See Also -------- randint : Uniform sampling over a given half-open interval of integers. random_integers : Uniform sampling over a given closed interval of integers. Examples -------- >>> rs = np.random.RandomState() # need a RandomState object >>> rs.tomaxint((2,2,2)) array([[[1170048599, 1600360186], # random [ 739731006, 1947757578]], [[1871712945, 752307660], [1601631370, 1479324245]]]) >>> rs.tomaxint((2,2,2)) < np.iinfo(np.int_).max array([[[ True, True], [ True, True]], [[ True, True], [ True, True]]]) standard_exponential(size=None) Draw samples from the standard exponential distribution. `standard_exponential` is identical to the exponential distribution with a scale parameter of 1. .. note:: New code should use the `~numpy.random.Generator.standard_exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray Drawn samples. See Also -------- random.Generator.standard_exponential: which should be used for new code. Examples -------- Output a 3x8000 array: >>> n = np.random.standard_exponential((3, 8000)) exponential(scale=1.0, size=None) Draw samples from an exponential distribution. Its probability density function is .. math:: f(x; \frac{1}{\beta}) = \frac{1}{\beta} \exp(-\frac{x}{\beta}), for ``x > 0`` and 0 elsewhere. :math:`\beta` is the scale parameter, which is the inverse of the rate parameter :math:`\lambda = 1/\beta`. The rate parameter is an alternative, widely used parameterization of the exponential distribution [3]_. The exponential distribution is a continuous analogue of the geometric distribution. It describes many common situations, such as the size of raindrops measured over many rainstorms [1]_, or the time between page requests to Wikipedia [2]_. .. note:: New code should use the `~numpy.random.Generator.exponential` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- scale : float or array_like of floats The scale parameter, :math:`\beta = 1/\lambda`. Must be non-negative. size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``scale`` is a scalar. Otherwise, ``np.array(scale).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized exponential distribution. Examples -------- A real world example: Assume a company has 10000 customer support agents and the average time between customer calls is 4 minutes. >>> n = 10000 >>> time_between_calls = np.random.default_rng().exponential(scale=4, size=n) What is the probability that a customer will call in the next 4 to 5 minutes? >>> x = ((time_between_calls < 5).sum())/n >>> y = ((time_between_calls < 4).sum())/n >>> x-y 0.08 # may vary See Also -------- random.Generator.exponential: which should be used for new code. References ---------- .. [1] Peyton Z. Peebles Jr., "Probability, Random Variables and Random Signal Principles", 4th ed, 2001, p. 57. .. [2] Wikipedia, "Poisson process", https://en.wikipedia.org/wiki/Poisson_process .. [3] Wikipedia, "Exponential distribution", https://en.wikipedia.org/wiki/Exponential_distribution beta(a, b, size=None) Draw samples from a Beta distribution. The Beta distribution is a special case of the Dirichlet distribution, and is related to the Gamma distribution. It has the probability distribution function .. math:: f(x; a,b) = \frac{1}{B(\alpha, \beta)} x^{\alpha - 1} (1 - x)^{\beta - 1}, where the normalization, B, is the beta function, .. math:: B(\alpha, \beta) = \int_0^1 t^{\alpha - 1} (1 - t)^{\beta - 1} dt. It is often seen in Bayesian inference and order statistics. .. note:: New code should use the `~numpy.random.Generator.beta` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- a : float or array_like of floats Alpha, positive (>0). b : float or array_like of floats Beta, positive (>0). size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. If size is ``None`` (default), a single value is returned if ``a`` and ``b`` are both scalars. Otherwise, ``np.broadcast(a, b).size`` samples are drawn. Returns ------- out : ndarray or scalar Drawn samples from the parameterized beta distribution. See Also -------- random.Generator.beta: which should be used for new code. random(size=None) Return random floats in the half-open interval [0.0, 1.0). Alias for `random_sample` to ease forward-porting to the new random API. random_sample(size=None) Return random floats in the half-open interval [0.0, 1.0). Results are from the "continuous uniform" distribution over the stated interval. To sample :math:`Unif[a, b), b > a` multiply the output of `random_sample` by `(b-a)` and add `a`:: (b - a) * random_sample() + a .. note:: New code should use the `~numpy.random.Generator.random` method of a `~numpy.random.Generator` instance instead; please see the :ref:`random-quick-start`. Parameters ---------- size : int or tuple of ints, optional Output shape. If the given shape is, e.g., ``(m, n, k)``, then ``m * n * k`` samples are drawn. Default is None, in which case a single value is returned. Returns ------- out : float or ndarray of floats Array of random floats of shape `size` (unless ``size=None``, in which case a single float is returned). See Also -------- random.Generator.random: which should be used for new code. Examples -------- >>> np.random.random_sample() 0.47108547995356098 # random >>> type(np.random.random_sample()) >>> np.random.random_sample((5,)) array([ 0.30220482, 0.86820401, 0.1654503 , 0.11659149, 0.54323428]) # random Three-by-two array of random numbers from [-5, 0): >>> 5 * np.random.random_sample((3, 2)) - 5 array([[-3.99149989, -0.52338984], # random [-2.99091858, -0.79479508], [-1.23204345, -1.75224494]]) set_state(state) Set the internal state of the generator from a tuple. For use if one has reason to manually (re-)set the internal state of the bit generator used by the RandomState instance. By default, RandomState uses the "Mersenne Twister"[1]_ pseudo-random number generating algorithm. Parameters ---------- state : {tuple(str, ndarray of 624 uints, int, int, float), dict} The `state` tuple has the following items: 1. the string 'MT19937', specifying the Mersenne Twister algorithm. 2. a 1-D array of 624 unsigned integers ``keys``. 3. an integer ``pos``. 4. an integer ``has_gauss``. 5. a float ``cached_gaussian``. If state is a dictionary, it is directly set using the BitGenerators `state` property. Returns ------- out : None Returns 'None' on success. See Also -------- get_state Notes ----- `set_state` and `get_state` are not needed to work with any of the random distributions in NumPy. If the internal state is manually altered, the user should know exactly what he/she is doing. For backwards compatibility, the form (str, array of 624 uints, int) is also accepted although it is missing some information about the cached Gaussian value: ``state = ('MT19937', keys, pos)``. References ---------- .. [1] M. Matsumoto and T. Nishimura, "Mersenne Twister: A 623-dimensionally equidistributed uniform pseudorandom number generator," *ACM Trans. on Modeling and Computer Simulation*, Vol. 8, No. 1, pp. 3-30, Jan. 1998. get_state(legacy=True) Return a tuple representing the internal state of the generator. For more details, see `set_state`. Parameters ---------- legacy : bool, optional Flag indicating to return a legacy tuple state when the BitGenerator is MT19937, instead of a dict. Raises ValueError if the underlying bit generator is not an instance of MT19937. Returns ------- out : {tuple(str, ndarray of 624 uints, int, int, float), dict} If legacy is True, the returned tuple has the following items: 1. the string 'MT19937'. 2. a 1-D array of 624 unsigned integer keys. 3. an integer ``pos``. 4. an integer ``has_gauss``. 5. a float ``cached_gaussian``. If `legacy` is False, or the BitGenerator is not MT19937, then state is returned as a dictionary. See Also -------- set_state Notes ----- `set_state` and `get_state` are not needed to work with any of the random distributions in NumPy. If the internal state is manually altered, the user should know exactly what he/she is doing. seed(seed=None) Reseed a legacy MT19937 BitGenerator Notes ----- This is a convenience, legacy function. The best practice is to **not** reseed a BitGenerator, rather to recreate a new one. This method is here for legacy reasons. This example demonstrates best practice. >>> from numpy.random import MT19937 >>> from numpy.random import RandomState, SeedSequence >>> rs = RandomState(MT19937(SeedSequence(123456789))) # Later, you want to restart the stream >>> rs = RandomState(MT19937(SeedSequence(987654321))) 0 H H ` H H ` H ` H ` H H H p ` H ` H H  H H ` x 0 0  A  @& `  0  >     $   P  P  !  @M" ` M" #x p @h L"` X L'P  H !@ @L"8 L"0 `$( K,  * - %   `$     `  ( % ! `   ( `, K' ( #x  p #h  ` @!X P  H @ 8 `0  ( `>   G L   @K! @    p       K0  & J5 l h P  x Fp  h d` cX &P @H @ 8 \0 x ( !     p     ;   U  : : 9 ( h h ` 0  @4x  p  h /6 ` NX NP ! H 0 @ 8 0 (  ` % 3 G ~ ~ $ $       a  \ @       X x  p  h  ` @ X P H @ 8 0 X( 9          { T @ @     @u P 2 2 @!    P Px np {h +`  X P H @ 8 0 (           x    P   L L H H        Tx @p h  ` HX HP H  @ h 8 h 0 @p (     ` p  p  `  D      #     8    x x     Q x `p `h ` X P  H @ 8 0 (  w           `k      ]p    N u 0 0  M'x `"p  h @` X P X H @ 8 0 (  =       ( (    @   /  # @ P    @ x p h ` X qP H @ @$v 8 0 ( @(        p p  `     H  H      A    x 8 p h ` X P H @ 8  0 (  w   `  `  ?     @{  | | P    ~    x pA p h ` l6X 0P 0H a @ 8 0 Yz(      v v 0 ! 8  (  3   / y p   + S p  p   x gp qh ` X P pH @ 8 0 (        b b Mk  ] ]       @  /  j X X  U 5@D 17 J1p1`3@̨٨  `d 0 o h @ cy W o Nj ӧ e Z 0`W KI `; W6 \- ܧ6" >0 Gp / 8P - "Ѐ@  ~ z @x u r Po ڣlz ԣil ̣0f^ ţpbK ^ ? Z@- @X # T E- ʧpR``% `9 /ppu}`u0u`v y PW |GCC: (GNU) 10.2.1 20210130 (Red Hat 10.2.1-11) 1; 1Ep p1  1XP h % 19 2S 2 `3Q 3 P4  5> 6GQ 7w p8  ` ] Z~ `@u 9) 8 N :r ; <` D @?L A 0Btx   Bx1 0CE PD` 0Ez E( G  0Hk I pM  Np 0p Eh _@ |@  T  @   % 4H D@ Y h @X   Z    5 ^nX |`   pb  0f# 8 in z   l`  Po @ 4  rm  }         u h       2  @xJr    z       ~9  N  Ѐ         ` (   @ '  Pj]  m         p  2  0Ar        A - i }     v ) g   F    3 D 0  @ @mH P & 4 B %z 0 ( `   P %X >8 M \e` wX @ Cx p ` X P p   % 9P Q8 jH   @  8   0 9x L a( z    @   0 * V( e       = M ] m        ) T d    8   , < a q      " N ^ x p @ h  `  X + AP Q oH  @  8  0 - = i( y       p   0H N ^(       !x M _0 n      X X  ! Q   h 8 0 ( *  E a    +  c      ! ?! v! ! ! " K" u" " "x #p 8#P K# ^#0 q#8 # # # # # # # $H $` #$ :$ V$ e$ |$x $ $p $h $ $` %X !%x 5%P L%H h%` y%@ %8 % %0 %( %p  &  & <&X H& _& {& & & & & & ' ' 8' I' `' |' ' ' ' ' '  (@  ( 7( S(( g( ~( ( ( (x ( (p )h *)x H)` _)X {)0 )P )H ) )@ *8 * 3*0 J*( f* w* * *0 * * * + + 7+ I+ `+ |+ + + +p + + ,P , *, F,8 V, m, , , , , , , - 1-x M- `-p w-h -` -X - -P -H  . %.@ <.8 X.p w.0 .( .X . . .@ / '/ C/( X/ o/ / / / /H / / 0 +0 B0 ^0 m0 0 0 0 0 0 0h 1 :1 o1 1 1 1 2 2 (2p <2X M2 a2h w2P 2 2 2 2 2 2 3 38 &3 :3 N3 d3p 3( 3 3 3 3( 3  4 4 .4h =4H O40 _4 p4 4 4 4 4P 4( 4 5 5 &5 @5h _5P x58 5 5 5@ 5 5 5 6 6 '6 76 a6 6 6 6` 7 37` ]7 7 7x 7@ 8 08 [8 8P 8 88 8h '9X Q9 w9( 9 9 9 :0 I: u: : :H : ; H;p o;H ;x ; ; <h ?< j< < <x < = B= n= =` = = > A>P j> > > >P ? :?` b?0 ?H ? ?@  @ 8@ b@ @X @ @p  A 4A ^A A  A A B 3BH YB Bh Bx BX C 0C8 [Ch C( C C0 D 2D ]D D D D E E KE ~Ep Eh E E E Fx 1F 0YF pF F F F F G )G EG XG8 iG( zG G G@ GH Gp G -H BH wH H H H H H 7IX CI XI` dI I Ix Ix I` I 0*J @Jx [J` lJ |Jp J J 0J J K K  K .K i  j j $Dj Zj j P j !j j @M" k `-k M"Uk #~k k @k L"k l L'7l  ]l !l @L"l L"l `$l K,%m *Nm -wm %m m m `$n  9n ]n n `n n (n %o !Bo `do  o (o `,o K'p (-p #Vp  |p #p  p @!p q  ;q ^q q `q  q `>q q Gr L0r  Br @K!jr @ ~r  r pr r r r  r K0 s )s &Rs J5zs ls hs Ps  s Fs  s ds cs &s @t  t 0t \?t x Ot !\t ft st pt  t ;t t t Ut u :u 9|  Q| h b| @p |  | `| p | ` | D }  } 0} =} #f} } }  } 8 }  } x}  } Q ~ `/~ W~ a~  n~ {~ ~ ~ w ~  ~ ~   2 `k Z  o ]p   N u 0  M' `"H  ] @y   X  ʀ   =!  5 \ k ({    @Ɓ ԁ /  #% @G Pm z  @  ΂ ۂ  q ! @$v I W @( ~    ̓ p   `# 0 > L H ]     A  ń  ڄ 8     #  8 E wl  `  ? ΅ ޅ @{  |  P 2 ? L ~Y  l  pA  Ȇ l6 0 a - E Yzl    v 0ۇ ! 8  ( 0 3< T | / y p   + S p   ! g. q9 H W f pt    ʉ  ۉ   b Mk 7 ]D  U e   @  /Ȋ  j X  ' 5 0K | P w  Y @ Ќy u o Cj 6e SÍZ v `W 3I k`; 6 qڎ- ( "  R z ˏ    B u@ p  6   4 A u    #z \l ^ ̒K ! ? p >@- y #  k  %_` ` ۔ {Ln Ǖ  Q Ka }`uh&uhO`v y |ݗ @0ߗ p0 0@ k; 0Gkf p z        Ƙ Ϙ ؘ|Y0p@ k(k8`0283Ц   `  0` kkkkmp@u@ 1C\q` ә p '  ) OB8DW Kk @. pEƚך ;' y{B Qd p/u e A `j_ ś Jۛ (/ Q3' 0K 3H @KVf CI R 0ɜ՜  |`#2@X OCez P3ʝ؝ p8#0 L_ tzў M- @M@ PO/Wg p+t  AyßП = @a)"EVfv͠ <% `i .=J\n ,?{ fCǡء(: 2M^o @2Sx nҢ p0/ r ,9J EIh w U  -Wǣ < ߣ k () U&B QS Ndp Jʤ +'ؤ p~w4 M$B @0&[ 8u `M( nZ @QNʥݥ O#7E kVV m h @=YȦ+٦ `8  `C7 QIY <Iv b 8n Piѧ / M*; >I\h p,Hx P, 2b ~˨٨ @P H* M7J[k 30 Ω +B NS  0tI(7H c&d j pȪ ?-  `R9H 9Xix @@I ëΫ dW @/$__pyx_f_5numpy_6random_6mtrand_11RandomState__reset_gauss__pyx_tp_traverse_5numpy_6random_6mtrand_RandomState__pyx_getprop_5numpy_6random_6mtrand_11RandomState__bit_generator__pyx_tp_new_5numpy_6random_6mtrand_RandomState__pyx_vtabptr_5numpy_6random_6mtrand_RandomState__pyx_empty_tuple__Pyx_PyObject_Call__Pyx_PyObject_GetAttrStr__pyx_f_5numpy_6random_6mtrand_11RandomState__shuffle_raw__pyx_setprop_5numpy_6random_6mtrand_11RandomState__bit_generator__Pyx_PyNumber_IntOrLongWrongResultType__Pyx_PyInt_As_int__pyx_tp_dealloc_5numpy_6random_6mtrand_RandomState__Pyx_GetException__pyx_tp_clear_5numpy_6random_6mtrand_RandomState__Pyx_Import__pyx_m__Pyx_ImportVoidPtr_0_29_36__Pyx_ImportFunction_0_29_36__Pyx_copy_spec_to_module__pyx_pymod_createmain_interpreter_id.0__Pyx_PyDict_GetItem__Pyx_ImportType_0_29_36.constprop.0__Pyx_PyObject_GetSlice.constprop.0__Pyx_Raise.constprop.0__Pyx_ParseOptionalKeywords.constprop.0__Pyx_PyCode_New.constprop.0__Pyx_PyUnicode_Equals__Pyx_IternextUnpackEndCheck__Pyx_GetBuiltinName__pyx_b__Pyx_ImportFrom__Pyx_PyInt_As_long__Pyx__PyObject_CallOneArg__Pyx_PyObject_CallOneArg__Pyx_PyObject_Call2Args__Pyx_PyObject_CallNoArg__Pyx_PyInt_As_Py_intptr_t.part.0__Pyx_PyObject_GetItem__Pyx__GetModuleGlobalName__pyx_d__Pyx_AddTraceback__pyx_cython_runtime__pyx_dict_version.2__pyx_dict_cached_value.1__pyx_n_s_cline_in_traceback__pyx_code_cache__pyx_pw_5numpy_6random_6mtrand_11RandomState_83wald__pyx_kp_u__13__pyx_float_0_0__pyx_n_u_scale__pyx_n_u_mean__pyx_f_5numpy_6random_7_common_cont__pyx_n_s_mean__pyx_n_s_scale__pyx_pyargnames.120__pyx_n_s_size__pyx_pw_5numpy_6random_6mtrand_11RandomState_81rayleigh__pyx_float_1_0__pyx_pyargnames.119__pyx_pw_5numpy_6random_6mtrand_11RandomState_79lognormal__pyx_n_u_sigma__pyx_n_s_sigma__pyx_pyargnames.118__pyx_pw_5numpy_6random_6mtrand_11RandomState_77logistic__pyx_n_u_loc__pyx_n_s_loc__pyx_pyargnames.117__pyx_pw_5numpy_6random_6mtrand_11RandomState_75gumbel__pyx_pyargnames.116__pyx_pw_5numpy_6random_6mtrand_11RandomState_73laplace__pyx_pyargnames.115__pyx_pw_5numpy_6random_6mtrand_11RandomState_71power__pyx_n_u_a__pyx_n_s_a__pyx_pyargnames.114__pyx_pw_5numpy_6random_6mtrand_11RandomState_69weibull__pyx_pyargnames.113__pyx_pw_5numpy_6random_6mtrand_11RandomState_67pareto__pyx_pyargnames.112__pyx_pw_5numpy_6random_6mtrand_11RandomState_65vonmises__pyx_n_u_kappa__pyx_n_u_mu__pyx_n_s_mu__pyx_n_s_kappa__pyx_pyargnames.111__pyx_pw_5numpy_6random_6mtrand_11RandomState_63standard_t__pyx_int_0__pyx_n_u_df__pyx_n_s_df__pyx_pyargnames.110__pyx_pw_5numpy_6random_6mtrand_11RandomState_61standard_cauchy__pyx_pyargnames.109__pyx_pw_5numpy_6random_6mtrand_11RandomState_59noncentral_chisquare__pyx_n_u_nonc__pyx_n_s_nonc__pyx_pyargnames.108__pyx_pw_5numpy_6random_6mtrand_11RandomState_57chisquare__pyx_pyargnames.107__pyx_pw_5numpy_6random_6mtrand_11RandomState_55noncentral_f__pyx_n_u_dfden__pyx_n_u_dfnum__pyx_n_s_dfnum__pyx_n_s_dfden__pyx_pyargnames.106__pyx_pw_5numpy_6random_6mtrand_11RandomState_53f__pyx_pyargnames.105__pyx_pw_5numpy_6random_6mtrand_11RandomState_51gamma__pyx_n_u_shape__pyx_pyargnames.104__pyx_n_s_shape__pyx_pw_5numpy_6random_6mtrand_11RandomState_49standard_gamma__pyx_pyargnames.103__pyx_pw_5numpy_6random_6mtrand_11RandomState_47normal__pyx_pyargnames.102__pyx_pw_5numpy_6random_6mtrand_11RandomState_45standard_normal__pyx_pyargnames.101__pyx_pw_5numpy_6random_6mtrand_11RandomState_39rand__pyx_n_s_random_sample__pyx_pw_5numpy_6random_6mtrand_11RandomState_27standard_exponential__pyx_pyargnames.15__pyx_pw_5numpy_6random_6mtrand_11RandomState_25exponential__pyx_pyargnames.14__pyx_pw_5numpy_6random_6mtrand_11RandomState_23beta__pyx_n_u_b__pyx_n_s_b__pyx_pyargnames.13__pyx_pw_5numpy_6random_6mtrand_11RandomState_21random__pyx_pyargnames.12__pyx_pw_5numpy_6random_6mtrand_11RandomState_19random_sample__pyx_f_5numpy_6random_7_common_double_fill__pyx_pyargnames.11__pyx_pw_5numpy_6random_6mtrand_11RandomState_11__reduce____pyx_n_s_reduce__pyx_n_s_randomstate_ctor__pyx_n_s_pickle__pyx_n_s_get_state__pyx_n_s_legacy__pyx_pw_5numpy_6random_6mtrand_11RandomState_9__setstate____pyx_n_s_set_state__pyx_pw_5numpy_6random_6mtrand_11RandomState_7__getstate____pyx_pw_5numpy_6random_6mtrand_11RandomState_5__str____pyx_n_s_class__pyx_n_s_name__pyx_kp_u__2__pyx_kp_u__3__pyx_pw_5numpy_6random_6mtrand_11RandomState_3__repr____pyx_n_s_str__pyx_kp_u_at_0x_X__pyx_n_s_format__pyx_builtin_id__pyx_f_5numpy_6random_6mtrand_11RandomState__initialize_bit_generator__pyx_n_s_capsule__pyx_n_s_lock__pyx_builtin_ValueError__pyx_tuple__4__pyx_pymod_exec_mtrand__pyx_empty_bytes__pyx_empty_unicode__pyx_string_tab__pyx_float_1eneg_8__pyx_float_1_0001__pyx_int_1__pyx_int_4294967296__pyx_int_neg_1__pyx_n_s_id__pyx_n_s_main__pyx_n_s_ValueError__pyx_n_s_TypeError__pyx_builtin_TypeError__pyx_n_s_RuntimeWarning__pyx_builtin_RuntimeWarning__pyx_n_s_range__pyx_n_s_DeprecationWarning__pyx_builtin_DeprecationWarning__pyx_n_s_OverflowError__pyx_builtin_OverflowError__pyx_n_s_UserWarning__pyx_builtin_UserWarning__pyx_n_s_reversed__pyx_n_s_IndexError__pyx_builtin_IndexError__pyx_n_s_ImportError__pyx_builtin_ImportError__pyx_n_u_l__pyx_tuple___pyx_kp_u_Invalid_bit_generator_The_bit_ge__pyx_kp_u_can_only_re_seed_a_MT19937_BitGe__pyx_tuple__5__pyx_kp_u_get_state_and_legacy_can_only_be__pyx_tuple__6__pyx_kp_u_legacy_can_only_be_True_when_the__pyx_tuple__7__pyx_kp_u_state_dictionary_is_not_valid__pyx_tuple__8__pyx_kp_u_state_must_be_a_dict_or_a_tuple__pyx_tuple__9__pyx_kp_u_set_state_can_only_be_used_with__pyx_tuple__10__pyx_n_u_gauss__pyx_tuple__11__pyx_n_u_has_gauss__pyx_tuple__12__pyx_tuple__14__pyx_kp_u_Providing_a_dtype_with_a_non_nat__pyx_tuple__16__pyx_tuple__17__pyx_kp_u_a_must_be_1_dimensional_or_an_in__pyx_tuple__18__pyx_kp_u_a_must_be_greater_than_0_unless__pyx_tuple__19__pyx_kp_u_a_must_be_1_dimensional__pyx_tuple__20__pyx_kp_u_a_cannot_be_empty_unless_no_sam__pyx_tuple__21__pyx_kp_u_p_must_be_1_dimensional__pyx_tuple__22__pyx_kp_u_a_and_p_must_have_same_size__pyx_tuple__23__pyx_kp_u_probabilities_contain_NaN__pyx_tuple__24__pyx_kp_u_probabilities_are_not_non_negati__pyx_tuple__25__pyx_kp_u_probabilities_do_not_sum_to_1__pyx_tuple__26__pyx_kp_u_Cannot_take_a_larger_sample_than__pyx_tuple__27__pyx_kp_u_Negative_dimensions_are_not_allo__pyx_tuple__28__pyx_kp_u_Fewer_non_zero_entries_in_p_than__pyx_tuple__29__pyx_tuple__30__pyx_kp_u_Range_exceeds_valid_bounds__pyx_tuple__31__pyx_kp_u_left_mode__pyx_tuple__32__pyx_kp_u_mode_right__pyx_tuple__33__pyx_kp_u_left_right__pyx_tuple__34__pyx_kp_u_ngood_nbad_nsample__pyx_tuple__35__pyx_kp_u_mean_must_be_1_dimensional__pyx_tuple__36__pyx_kp_u_cov_must_be_2_dimensional_and_sq__pyx_tuple__37__pyx_kp_u_mean_and_cov_must_have_same_leng__pyx_tuple__38__pyx_slice__39__pyx_kp_u_check_valid_must_equal_warn_rais__pyx_tuple__40__pyx_kp_u_covariance_is_not_symmetric_posi__pyx_tuple__41__pyx_tuple__42__pyx_tuple__43__pyx_kp_u_pvals_must_be_a_1_d_sequence__pyx_tuple__44__pyx_kp_u_alpha_0__pyx_tuple__45__pyx_kp_u_array_is_read_only__pyx_tuple__46__pyx_kp_u_Shuffling_a_one_dimensional_arra__pyx_tuple__47__pyx_tuple__48__pyx_kp_u_x_must_be_an_integer_or_at_least__pyx_tuple__49__pyx_kp_u_numpy_core_multiarray_failed_to__pyx_tuple__50__pyx_kp_u_numpy_core_umath_failed_to_impor__pyx_n_s_bg_type__pyx_n_s_seed__pyx_kp_s_mtrand_pyx__pyx_n_s_get_bit_generator__pyx_n_s_singleton__pyx_n_s_bitgen__pyx_n_s_set_bit_generator__pyx_n_s_kwargs__pyx_n_s_args__pyx_n_s_sample__pyx_n_s_ranf__pyx_vtable_5numpy_6random_6mtrand_RandomState__pyx_type_5numpy_6random_6mtrand_RandomState__pyx_n_s_pyx_vtable__pyx_n_s_RandomState__pyx_ptype_5numpy_6random_6mtrand_RandomState__pyx_ptype_5numpy_dtype__pyx_ptype_5numpy_broadcast__pyx_ptype_5numpy_ndarray__pyx_ptype_5numpy_integer__pyx_ptype_5numpy_floating__pyx_vp_5numpy_6random_7_common_POISSON_LAM_MAX__pyx_vp_5numpy_6random_7_common_LEGACY_POISSON_LAM_MAX__pyx_vp_5numpy_6random_7_common_MAXSIZE__pyx_f_5numpy_6random_17_bounded_integers__rand_uint64__pyx_f_5numpy_6random_17_bounded_integers__rand_uint32__pyx_f_5numpy_6random_17_bounded_integers__rand_uint16__pyx_f_5numpy_6random_17_bounded_integers__rand_uint8__pyx_f_5numpy_6random_17_bounded_integers__rand_bool__pyx_f_5numpy_6random_17_bounded_integers__rand_int64__pyx_f_5numpy_6random_17_bounded_integers__rand_int32__pyx_f_5numpy_6random_17_bounded_integers__rand_int16__pyx_f_5numpy_6random_17_bounded_integers__rand_int8__pyx_f_5numpy_6random_7_common_check_constraint__pyx_f_5numpy_6random_7_common_check_array_constraint__pyx_f_5numpy_6random_7_common_kahan_sum__pyx_f_5numpy_6random_7_common_validate_output_shape__pyx_f_5numpy_6random_7_common_disc__pyx_f_5numpy_6random_7_common_cont_broadcast_3__pyx_f_5numpy_6random_7_common_discrete_broadcast_iii__pyx_n_s_operator__pyx_n_s_warnings__pyx_n_s_Sequence__pyx_n_s_collections_abc__pyx_n_s_numpy__pyx_n_s_np__pyx_n_s_MT19937_2__pyx_n_s_mt19937__pyx_n_s_MT19937PyArray_API__pyx_n_s_poisson_lam_max__pyx_k__15__pyx_n_s_rand_2__pyx_dict_version.339__pyx_dict_cached_value.338__pyx_n_s_beta__pyx_dict_version.337__pyx_dict_cached_value.336__pyx_n_s_binomial__pyx_dict_version.335__pyx_dict_cached_value.334__pyx_n_s_bytes__pyx_dict_version.333__pyx_dict_cached_value.332__pyx_n_s_chisquare__pyx_dict_version.331__pyx_dict_cached_value.330__pyx_n_s_choice__pyx_dict_version.329__pyx_dict_cached_value.328__pyx_n_s_dirichlet__pyx_dict_version.327__pyx_dict_cached_value.326__pyx_n_s_exponential__pyx_dict_version.325__pyx_dict_cached_value.324__pyx_n_s_f__pyx_dict_version.323__pyx_dict_cached_value.322__pyx_n_s_gamma__pyx_dict_version.321__pyx_dict_cached_value.320__pyx_dict_version.319__pyx_dict_cached_value.318__pyx_n_s_geometric__pyx_dict_version.317__pyx_dict_cached_value.316__pyx_n_s_gumbel__pyx_dict_version.315__pyx_dict_cached_value.314__pyx_n_s_hypergeometric__pyx_dict_version.313__pyx_dict_cached_value.312__pyx_n_s_laplace__pyx_dict_version.311__pyx_dict_cached_value.310__pyx_n_s_logistic__pyx_dict_version.309__pyx_dict_cached_value.308__pyx_n_s_lognormal__pyx_dict_version.307__pyx_dict_cached_value.306__pyx_n_s_logseries__pyx_dict_version.305__pyx_dict_cached_value.304__pyx_n_s_multinomial__pyx_dict_version.303__pyx_dict_cached_value.302__pyx_n_s_multivariate_normal__pyx_dict_version.301__pyx_dict_cached_value.300__pyx_n_s_negative_binomial__pyx_dict_version.299__pyx_dict_cached_value.298__pyx_n_s_noncentral_chisquare__pyx_dict_version.297__pyx_dict_cached_value.296__pyx_n_s_noncentral_f__pyx_dict_version.295__pyx_dict_cached_value.294__pyx_n_s_normal__pyx_dict_version.293__pyx_dict_cached_value.292__pyx_n_s_pareto__pyx_dict_version.291__pyx_dict_cached_value.290__pyx_n_s_permutation__pyx_dict_version.289__pyx_dict_cached_value.288__pyx_n_s_poisson__pyx_dict_version.287__pyx_dict_cached_value.286__pyx_n_s_power__pyx_dict_version.285__pyx_dict_cached_value.284__pyx_n_s_rand__pyx_dict_version.283__pyx_dict_cached_value.282__pyx_n_s_randint__pyx_dict_version.281__pyx_dict_cached_value.280__pyx_n_s_randn__pyx_dict_version.279__pyx_dict_cached_value.278__pyx_n_s_random__pyx_dict_version.277__pyx_dict_cached_value.276__pyx_n_s_random_integers__pyx_dict_version.275__pyx_dict_cached_value.274__pyx_dict_version.273__pyx_dict_cached_value.272__pyx_n_s_rayleigh__pyx_dict_version.271__pyx_dict_cached_value.270__pyx_dict_version.269__pyx_dict_cached_value.268__pyx_n_s_shuffle__pyx_dict_version.267__pyx_dict_cached_value.266__pyx_n_s_standard_cauchy__pyx_dict_version.265__pyx_dict_cached_value.264__pyx_n_s_standard_exponential__pyx_dict_version.263__pyx_dict_cached_value.262__pyx_n_s_standard_gamma__pyx_dict_version.261__pyx_dict_cached_value.260__pyx_n_s_standard_normal__pyx_dict_version.259__pyx_dict_cached_value.258__pyx_n_s_standard_t__pyx_dict_version.257__pyx_dict_cached_value.256__pyx_n_s_triangular__pyx_dict_version.255__pyx_dict_cached_value.254__pyx_n_s_uniform__pyx_dict_version.253__pyx_dict_cached_value.252__pyx_n_s_vonmises__pyx_dict_version.251__pyx_dict_cached_value.250__pyx_n_s_wald__pyx_dict_version.249__pyx_dict_cached_value.248__pyx_n_s_weibull__pyx_dict_version.247__pyx_dict_cached_value.246__pyx_n_s_zipf__pyx_n_s_numpy_random_mtrand__pyx_mdef_5numpy_6random_6mtrand_1seed__pyx_mdef_5numpy_6random_6mtrand_3get_bit_generator__pyx_mdef_5numpy_6random_6mtrand_5set_bit_generator__pyx_mdef_5numpy_6random_6mtrand_7sample__pyx_mdef_5numpy_6random_6mtrand_9ranf__pyx_n_u_beta__pyx_n_u_binomial__pyx_n_u_bytes__pyx_n_u_chisquare__pyx_n_u_choice__pyx_n_u_dirichlet__pyx_n_u_exponential__pyx_n_u_f__pyx_n_u_gamma__pyx_n_u_geometric__pyx_n_u_get_bit_generator__pyx_n_u_get_state__pyx_n_u_gumbel__pyx_n_u_hypergeometric__pyx_n_u_laplace__pyx_n_u_logistic__pyx_n_u_lognormal__pyx_n_u_logseries__pyx_n_u_multinomial__pyx_n_u_multivariate_normal__pyx_n_u_negative_binomial__pyx_n_u_noncentral_chisquare__pyx_n_u_noncentral_f__pyx_n_u_normal__pyx_n_u_pareto__pyx_n_u_permutation__pyx_n_u_poisson__pyx_n_u_power__pyx_n_u_rand__pyx_n_u_randint__pyx_n_u_randn__pyx_n_u_random__pyx_n_u_random_integers__pyx_n_u_random_sample__pyx_n_u_ranf__pyx_n_u_rayleigh__pyx_n_u_sample__pyx_n_u_seed__pyx_n_u_set_bit_generator__pyx_n_u_set_state__pyx_n_u_shuffle__pyx_n_u_standard_cauchy__pyx_n_u_standard_exponential__pyx_n_u_standard_gamma__pyx_n_u_standard_normal__pyx_n_u_standard_t__pyx_n_u_triangular__pyx_n_u_uniform__pyx_n_u_vonmises__pyx_n_u_wald__pyx_n_u_weibull__pyx_n_u_zipf__pyx_n_u_RandomState__pyx_n_s_all_2__pyx_kp_u_seed_seed_None_Reseed_a_legacy__pyx_kp_u_RandomState_seed_line_228__pyx_kp_u_random_sample_size_None_Return__pyx_kp_u_RandomState_random_sample_line_3__pyx_kp_u_exponential_scale_1_0_size_None__pyx_kp_u_RandomState_exponential_line_499__pyx_kp_u_standard_exponential_size_None__pyx_kp_u_RandomState_standard_exponential__pyx_kp_u_tomaxint_size_None_Return_a_sam__pyx_kp_u_RandomState_tomaxint_line_620__pyx_kp_u_randint_low_high_None_size_None__pyx_kp_u_RandomState_randint_line_678__pyx_kp_u_bytes_length_Return_random_byte__pyx_kp_u_RandomState_bytes_line_804__pyx_kp_u_choice_a_size_None_replace_True__pyx_kp_u_RandomState_choice_line_841__pyx_kp_u_uniform_low_0_0_high_1_0_size_N__pyx_kp_u_RandomState_uniform_line_1049__pyx_kp_u_rand_d0_d1_dn_Random_values_in__pyx_kp_u_RandomState_rand_line_1176__pyx_kp_u_randn_d0_d1_dn_Return_a_sample__pyx_kp_u_RandomState_randn_line_1220__pyx_kp_u_random_integers_low_high_None_s__pyx_kp_u_RandomState_random_integers_line__pyx_kp_u_standard_normal_size_None_Draw__pyx_kp_u_RandomState_standard_normal_line__pyx_kp_u_normal_loc_0_0_scale_1_0_size_N__pyx_kp_u_RandomState_normal_line_1453__pyx_kp_u_standard_gamma_shape_size_None__pyx_kp_u_RandomState_standard_gamma_line__pyx_kp_u_gamma_shape_scale_1_0_size_None__pyx_kp_u_RandomState_gamma_line_1644__pyx_kp_u_f_dfnum_dfden_size_None_Draw_sa__pyx_kp_u_RandomState_f_line_1728__pyx_kp_u_noncentral_f_dfnum_dfden_nonc_s__pyx_kp_u_RandomState_noncentral_f_line_18__pyx_kp_u_chisquare_df_size_None_Draw_sam__pyx_kp_u_RandomState_chisquare_line_1909__pyx_kp_u_noncentral_chisquare_df_nonc_si__pyx_kp_u_RandomState_noncentral_chisquare__pyx_kp_u_standard_cauchy_size_None_Draw__pyx_kp_u_RandomState_standard_cauchy_line__pyx_kp_u_standard_t_df_size_None_Draw_sa__pyx_kp_u_RandomState_standard_t_line_2149__pyx_kp_u_vonmises_mu_kappa_size_None_Dra__pyx_kp_u_RandomState_vonmises_line_2264__pyx_kp_u_pareto_a_size_None_Draw_samples__pyx_kp_u_RandomState_pareto_line_2353__pyx_kp_u_weibull_a_size_None_Draw_sample__pyx_kp_u_RandomState_weibull_line_2456__pyx_kp_u_power_a_size_None_Draws_samples__pyx_kp_u_RandomState_power_line_2560__pyx_kp_u_laplace_loc_0_0_scale_1_0_size__pyx_kp_u_RandomState_laplace_line_2669__pyx_kp_u_gumbel_loc_0_0_scale_1_0_size_N__pyx_kp_u_RandomState_gumbel_line_2763__pyx_kp_u_logistic_loc_0_0_scale_1_0_size__pyx_kp_u_RandomState_logistic_line_2887__pyx_kp_u_lognormal_mean_0_0_sigma_1_0_si__pyx_kp_u_RandomState_lognormal_line_2973__pyx_kp_u_rayleigh_scale_1_0_size_None_Dr__pyx_kp_u_RandomState_rayleigh_line_3089__pyx_kp_u_wald_mean_scale_size_None_Draw__pyx_kp_u_RandomState_wald_line_3166__pyx_kp_u_triangular_left_mode_right_size__pyx_kp_u_RandomState_triangular_line_3243__pyx_kp_u_binomial_n_p_size_None_Draw_sam__pyx_kp_u_RandomState_binomial_line_3352__pyx_kp_u_negative_binomial_n_p_size_None__pyx_kp_u_RandomState_negative_binomial_li__pyx_kp_u_poisson_lam_1_0_size_None_Draw__pyx_kp_u_RandomState_poisson_line_3592__pyx_kp_u_zipf_a_size_None_Draw_samples_f__pyx_kp_u_RandomState_zipf_line_3675__pyx_kp_u_geometric_p_size_None_Draw_samp__pyx_kp_u_RandomState_geometric_line_3771__pyx_kp_u_hypergeometric_ngood_nbad_nsamp__pyx_kp_u_RandomState_hypergeometric_line__pyx_kp_u_logseries_p_size_None_Draw_samp__pyx_kp_u_RandomState_logseries_line_3968__pyx_kp_u_multivariate_normal_mean_cov_si__pyx_kp_u_RandomState_multivariate_normal__pyx_kp_u_multinomial_n_pvals_size_None_D__pyx_kp_u_RandomState_multinomial_line_425__pyx_kp_u_dirichlet_alpha_size_None_Draw__pyx_kp_u_RandomState_dirichlet_line_4393__pyx_kp_u_shuffle_x_Modify_a_sequence_in__pyx_kp_u_RandomState_shuffle_line_4542__pyx_kp_u_permutation_x_Randomly_permute__pyx_kp_u_RandomState_permutation_line_466__pyx_n_s_test__pyx_pw_5numpy_6random_6mtrand_11RandomState_41randn__pyx_pw_5numpy_6random_6mtrand_3get_bit_generator__pyx_dict_version.239__pyx_dict_cached_value.238__pyx_n_s_bit_generator_2__pyx_pw_5numpy_6random_6mtrand_5set_bit_generator__pyx_dict_version.241__pyx_dict_cached_value.240__pyx_pw_5numpy_6random_6mtrand_7sample__pyx_dict_version.243__pyx_dict_cached_value.242__pyx_pw_5numpy_6random_6mtrand_9ranf__pyx_dict_version.245__pyx_dict_cached_value.244__pyx_f_5numpy_6random_6mtrand_int64_to_long__pyx_dict_version.159__pyx_dict_cached_value.158__pyx_n_s_isscalar__pyx_n_s_astype__pyx_n_u_unsafe__pyx_n_s_casting__pyx_pw_5numpy_6random_6mtrand_11RandomState_99logseries__pyx_n_u_p__pyx_n_s_p__pyx_pyargnames.160__pyx_pw_5numpy_6random_6mtrand_11RandomState_95geometric__pyx_pyargnames.144__pyx_pw_5numpy_6random_6mtrand_11RandomState_93zipf__pyx_pyargnames.143__pyx_pw_5numpy_6random_6mtrand_11RandomState_91poisson__pyx_n_u_lam__pyx_pyargnames.142__pyx_n_s_lam__pyx_pw_5numpy_6random_6mtrand_11RandomState_89negative_binomial__pyx_n_u_n__pyx_pyargnames.141__pyx_n_s_n__pyx_pw_5numpy_6random_6mtrand_11RandomState_13seed__pyx_dict_version.4__pyx_dict_cached_value.3__pyx_n_s_legacy_seeding__pyx_pyargnames.5__pyx_pw_5numpy_6random_6mtrand_11RandomState_33bytes__pyx_dict_version.45__pyx_dict_cached_value.44__pyx_n_s_uint32__pyx_n_s_dtype__pyx_kp_u_u4__pyx_n_s_tobytes__pyx_pw_5numpy_6random_6mtrand_11RandomState_17set_state__pyx_n_u_MT19937_2__pyx_n_u_bit_generator__pyx_n_u_state__pyx_n_s_get__pyx_n_u_key__pyx_n_u_pos__pyx_n_s_state__pyx_pw_5numpy_6random_6mtrand_11RandomState_29tomaxint__pyx_dict_version.19__pyx_dict_cached_value.18__pyx_n_s_empty__pyx_dict_version.17__pyx_dict_cached_value.16__pyx_n_s_int64__pyx_n_s_exit__pyx_n_s_enter__pyx_pyargnames.20__pyx_pw_5numpy_6random_6mtrand_11RandomState_43random_integers__pyx_dict_version.97__pyx_dict_cached_value.96__pyx_n_s_warn__pyx_kp_u_This_function_is_deprecated_Plea_2__pyx_n_s_low__pyx_n_s_high__pyx_dict_version.99__pyx_dict_cached_value.98__pyx_kp_u_This_function_is_deprecated_Plea__pyx_pyargnames.100__pyx_pw_5numpy_6random_6mtrand_11RandomState_1__init____pyx_n_u_capsule__pyx_dict_version.225__pyx_dict_cached_value.224__pyx_dict_version.223__pyx_dict_cached_value.222__pyx_pyargnames.226__pyx_pw_5numpy_6random_6mtrand_1seed__pyx_dict_version.236__pyx_dict_cached_value.235__pyx_dict_version.234__pyx_dict_cached_value.233__pyx_dict_version.232__pyx_dict_cached_value.231__pyx_dict_version.230__pyx_dict_cached_value.229__pyx_dict_version.228__pyx_dict_cached_value.227__pyx_pyargnames.237__pyx_pw_5numpy_6random_6mtrand_11RandomState_15get_state__pyx_dict_version.9__pyx_dict_cached_value.8__pyx_dict_version.7__pyx_dict_cached_value.6__pyx_pyargnames.10__pyx_pw_5numpy_6random_6mtrand_11RandomState_109permutation__pyx_dict_version.221__pyx_dict_cached_value.220__pyx_n_s_arange__pyx_dict_version.219__pyx_dict_cached_value.218__pyx_n_s_asarray__pyx_n_s_ndim__pyx_dict_version.217__pyx_dict_cached_value.216__pyx_n_s_may_share_memory__pyx_dict_version.213__pyx_dict_cached_value.212__pyx_dict_version.211__pyx_dict_cached_value.210__pyx_n_s_intp__pyx_dict_version.215__pyx_dict_cached_value.214__pyx_n_s_array__pyx_pw_5numpy_6random_6mtrand_11RandomState_37uniform__pyx_dict_version.92__pyx_dict_cached_value.91__pyx_n_s_subtract__pyx_dict_version.94__pyx_dict_cached_value.93__pyx_n_s_isfinite__pyx_dict_version.90__pyx_dict_cached_value.89__pyx_n_s_all__pyx_dict_version.88__pyx_dict_cached_value.87__pyx_pyargnames.95__pyx_pw_5numpy_6random_6mtrand_11RandomState_97hypergeometric__pyx_dict_version.156__pyx_dict_cached_value.155__pyx_n_s_any__pyx_dict_version.154__pyx_dict_cached_value.153__pyx_n_s_less__pyx_dict_version.152__pyx_dict_cached_value.151__pyx_n_s_add__pyx_n_s_ngood__pyx_n_u_nsample__pyx_n_u_nbad__pyx_n_u_ngood__pyx_n_s_nbad__pyx_n_s_nsample__pyx_pyargnames.157__pyx_dict_version.150__pyx_dict_cached_value.149__pyx_dict_version.148__pyx_dict_cached_value.147__pyx_dict_version.146__pyx_dict_cached_value.145__pyx_pw_5numpy_6random_6mtrand_11RandomState_105dirichlet__pyx_dict_version.194__pyx_dict_cached_value.193__pyx_dict_version.192__pyx_dict_cached_value.191__pyx_n_s_less_equal__pyx_dict_version.188__pyx_dict_cached_value.187__pyx_n_s_zeros__pyx_dict_version.186__pyx_dict_cached_value.185__pyx_n_s_float64__pyx_dict_version.190__pyx_dict_cached_value.189__pyx_n_s_index__pyx_n_s_alpha__pyx_pyargnames.195__pyx_pw_5numpy_6random_6mtrand_11RandomState_103multinomial__pyx_n_u_pvals__pyx_dict_version.181__pyx_dict_cached_value.180__pyx_dict_version.179__pyx_dict_cached_value.178__pyx_n_s_pvals__pyx_dict_version.183__pyx_dict_cached_value.182__pyx_n_s_issubdtype__pyx_kp_u_sum_pvals_1_1_0__pyx_n_s_sum__pyx_kp_u_sum_pvals_1_astype_np_float64_1__pyx_pyargnames.184__pyx_pw_5numpy_6random_6mtrand_11RandomState_85triangular__pyx_dict_version.132__pyx_dict_cached_value.131__pyx_dict_version.130__pyx_dict_cached_value.129__pyx_n_s_greater__pyx_n_s_left__pyx_n_s_mode__pyx_n_s_right__pyx_dict_version.128__pyx_dict_cached_value.127__pyx_dict_version.126__pyx_dict_cached_value.125__pyx_dict_version.124__pyx_dict_cached_value.123__pyx_dict_version.122__pyx_dict_cached_value.121__pyx_n_s_equal__pyx_pyargnames.133__pyx_pw_5numpy_6random_6mtrand_11RandomState_87binomial__pyx_dict_version.139__pyx_dict_cached_value.138__pyx_dict_version.135__pyx_dict_cached_value.134__pyx_pyargnames.140__pyx_dict_version.137__pyx_dict_cached_value.136__pyx_pw_5numpy_6random_6mtrand_11RandomState_31randint__pyx_n_s_isnative__pyx_dict_version.42__pyx_dict_cached_value.41__pyx_n_s_newbyteorder__pyx_dict_version.40__pyx_dict_cached_value.39__pyx_n_s_int32__pyx_dict_version.38__pyx_dict_cached_value.37__pyx_dict_version.36__pyx_dict_cached_value.35__pyx_n_s_int16__pyx_pyargnames.43__pyx_dict_version.22__pyx_dict_cached_value.21__pyx_dict_version.34__pyx_dict_cached_value.33__pyx_n_s_int8__pyx_dict_version.32__pyx_dict_cached_value.31__pyx_n_s_uint64__pyx_dict_version.30__pyx_dict_cached_value.29__pyx_dict_version.28__pyx_dict_cached_value.27__pyx_n_s_uint16__pyx_dict_version.26__pyx_dict_cached_value.25__pyx_n_s_uint8__pyx_dict_version.24__pyx_dict_cached_value.23__pyx_n_s_bool__pyx_kp_u_Unsupported_dtype_r_for_randint__pyx_pw_5numpy_6random_6mtrand_11RandomState_107shuffle__pyx_n_s_flags__pyx_n_s_writeable__pyx_n_s_type__pyx_dict_version.205__pyx_dict_cached_value.204__pyx_n_s_object__pyx_dict_version.203__pyx_dict_cached_value.202__pyx_n_s_stacklevel__pyx_dict_version.201__pyx_dict_cached_value.200__pyx_n_s_empty_like__pyx_dict_version.199__pyx_dict_cached_value.198__pyx_dict_version.197__pyx_dict_cached_value.196__pyx_kp_u_you_are_shuffling_a__pyx_kp_u_object_which_is_not_a_subclass__pyx_n_s_strides__pyx_n_s_itemsize__pyx_dict_version.209__pyx_dict_cached_value.208__pyx_dict_version.207__pyx_dict_cached_value.206__pyx_pw_5numpy_6random_6mtrand_11RandomState_101multivariate_normal__pyx_n_u_warn__pyx_n_s_svd__pyx_n_s_numpy_linalg__pyx_dict_version.176__pyx_dict_cached_value.175__pyx_dict_version.174__pyx_dict_cached_value.173__pyx_n_s_reshape__pyx_dict_version.172__pyx_dict_cached_value.171__pyx_n_s_double__pyx_n_u_ignore__pyx_dict_version.164__pyx_dict_cached_value.163__pyx_n_s_dot__pyx_dict_version.162__pyx_dict_cached_value.161__pyx_n_s_sqrt__pyx_n_s_cov__pyx_n_s_check_valid__pyx_n_s_tol__pyx_pyargnames.177__pyx_n_u_raise__pyx_dict_version.170__pyx_dict_cached_value.169__pyx_n_s_allclose__pyx_dict_version.168__pyx_dict_cached_value.167__pyx_n_s_T__pyx_n_s_rtol__pyx_n_s_atol__pyx_dict_version.166__pyx_dict_cached_value.165__pyx_pw_5numpy_6random_6mtrand_11RandomState_35choice__pyx_dict_version.85__pyx_dict_cached_value.84__pyx_n_s_copy__pyx_dict_version.83__pyx_dict_cached_value.82__pyx_n_s_item__pyx_dict_version.77__pyx_dict_cached_value.76__pyx_dict_version.75__pyx_dict_cached_value.74__pyx_n_s_finfo__pyx_dict_version.73__pyx_dict_cached_value.72__pyx_n_s_eps__pyx_dict_version.71__pyx_dict_cached_value.70__pyx_dict_version.69__pyx_dict_cached_value.68__pyx_dict_version.67__pyx_dict_cached_value.66__pyx_dict_version.65__pyx_dict_cached_value.64__pyx_n_s_isnan__pyx_dict_version.63__pyx_dict_cached_value.62__pyx_n_s_logical_or__pyx_n_s_reduce_2__pyx_n_s_cumsum__pyx_n_s_searchsorted__pyx_n_u_right__pyx_n_s_side__pyx_dict_version.57__pyx_dict_cached_value.56__pyx_dict_version.47__pyx_dict_cached_value.46__pyx_pyargnames.86__pyx_dict_version.81__pyx_dict_cached_value.80__pyx_n_s_prod__pyx_n_s_replace__pyx_dict_version.79__pyx_dict_cached_value.78__pyx_dict_version.61__pyx_dict_cached_value.60__pyx_dict_version.59__pyx_dict_cached_value.58__pyx_dict_version.55__pyx_dict_cached_value.54__pyx_n_s_count_nonzero__pyx_dict_version.53__pyx_dict_cached_value.52__pyx_n_s_ravel__pyx_dict_version.51__pyx_dict_cached_value.50__pyx_dict_version.49__pyx_dict_cached_value.48__pyx_n_s_unique__pyx_n_s_return_index__pyx_n_s_sort__pyx_n_s_take__pyx_moduledef__pyx_k_Cannot_take_a_larger_sample_than__pyx_k_DeprecationWarning__pyx_k_Fewer_non_zero_entries_in_p_than__pyx_k_ImportError__pyx_k_IndexError__pyx_k_Invalid_bit_generator_The_bit_ge__pyx_k_MT19937__pyx_k_MT19937_2__pyx_k_Negative_dimensions_are_not_allo__pyx_k_OverflowError__pyx_k_Providing_a_dtype_with_a_non_nat__pyx_k_RandomState__pyx_k_RandomState_binomial_line_3352__pyx_k_RandomState_bytes_line_804__pyx_k_RandomState_chisquare_line_1909__pyx_k_RandomState_choice_line_841__pyx_k_RandomState_dirichlet_line_4393__pyx_k_RandomState_exponential_line_499__pyx_k_RandomState_f_line_1728__pyx_k_RandomState_gamma_line_1644__pyx_k_RandomState_geometric_line_3771__pyx_k_RandomState_gumbel_line_2763__pyx_k_RandomState_hypergeometric_line__pyx_k_RandomState_laplace_line_2669__pyx_k_RandomState_logistic_line_2887__pyx_k_RandomState_lognormal_line_2973__pyx_k_RandomState_logseries_line_3968__pyx_k_RandomState_multinomial_line_425__pyx_k_RandomState_multivariate_normal__pyx_k_RandomState_negative_binomial_li__pyx_k_RandomState_noncentral_chisquare__pyx_k_RandomState_noncentral_f_line_18__pyx_k_RandomState_normal_line_1453__pyx_k_RandomState_pareto_line_2353__pyx_k_RandomState_permutation_line_466__pyx_k_RandomState_poisson_line_3592__pyx_k_RandomState_power_line_2560__pyx_k_RandomState_rand_line_1176__pyx_k_RandomState_randint_line_678__pyx_k_RandomState_randn_line_1220__pyx_k_RandomState_random_integers_line__pyx_k_RandomState_random_sample_line_3__pyx_k_RandomState_rayleigh_line_3089__pyx_k_RandomState_seed_line_228__pyx_k_RandomState_shuffle_line_4542__pyx_k_RandomState_standard_cauchy_line__pyx_k_RandomState_standard_exponential__pyx_k_RandomState_standard_gamma_line__pyx_k_RandomState_standard_normal_line__pyx_k_RandomState_standard_t_line_2149__pyx_k_RandomState_tomaxint_line_620__pyx_k_RandomState_triangular_line_3243__pyx_k_RandomState_uniform_line_1049__pyx_k_RandomState_vonmises_line_2264__pyx_k_RandomState_wald_line_3166__pyx_k_RandomState_weibull_line_2456__pyx_k_RandomState_zipf_line_3675__pyx_k_Range_exceeds_valid_bounds__pyx_k_RuntimeWarning__pyx_k_Sequence__pyx_k_Shuffling_a_one_dimensional_arra__pyx_k_T__pyx_k_This_function_is_deprecated_Plea__pyx_k_This_function_is_deprecated_Plea_2__pyx_k_TypeError__pyx_k_Unsupported_dtype_r_for_randint__pyx_k_UserWarning__pyx_k_ValueError__pyx_k__13__pyx_k__2__pyx_k__3__pyx_k_a__pyx_k_a_and_p_must_have_same_size__pyx_k_a_cannot_be_empty_unless_no_sam__pyx_k_a_must_be_1_dimensional__pyx_k_a_must_be_1_dimensional_or_an_in__pyx_k_a_must_be_greater_than_0_unless__pyx_k_add__pyx_k_all__pyx_k_all_2__pyx_k_allclose__pyx_k_alpha__pyx_k_alpha_0__pyx_k_any__pyx_k_arange__pyx_k_args__pyx_k_array__pyx_k_array_is_read_only__pyx_k_asarray__pyx_k_astype__pyx_k_at_0x_X__pyx_k_atol__pyx_k_b__pyx_k_beta__pyx_k_bg_type__pyx_k_binomial__pyx_k_binomial_n_p_size_None_Draw_sam__pyx_k_bit_generator__pyx_k_bit_generator_2__pyx_k_bitgen__pyx_k_bool__pyx_k_bytes__pyx_k_bytes_length_Return_random_byte__pyx_k_can_only_re_seed_a_MT19937_BitGe__pyx_k_capsule__pyx_k_casting__pyx_k_check_valid__pyx_k_check_valid_must_equal_warn_rais__pyx_k_chisquare__pyx_k_chisquare_df_size_None_Draw_sam__pyx_k_choice__pyx_k_choice_a_size_None_replace_True__pyx_k_class__pyx_k_cline_in_traceback__pyx_k_collections_abc__pyx_k_copy__pyx_k_count_nonzero__pyx_k_cov__pyx_k_cov_must_be_2_dimensional_and_sq__pyx_k_covariance_is_not_symmetric_posi__pyx_k_cumsum__pyx_k_df__pyx_k_dfden__pyx_k_dfnum__pyx_k_dirichlet__pyx_k_dirichlet_alpha_size_None_Draw__pyx_k_dot__pyx_k_double__pyx_k_dtype__pyx_k_empty__pyx_k_empty_like__pyx_k_enter__pyx_k_eps__pyx_k_equal__pyx_k_exit__pyx_k_exponential__pyx_k_exponential_scale_1_0_size_None__pyx_k_f__pyx_k_f_dfnum_dfden_size_None_Draw_sa__pyx_k_finfo__pyx_k_flags__pyx_k_float64__pyx_k_format__pyx_k_gamma__pyx_k_gamma_shape_scale_1_0_size_None__pyx_k_gauss__pyx_k_geometric__pyx_k_geometric_p_size_None_Draw_samp__pyx_k_get__pyx_k_get_bit_generator__pyx_k_get_state__pyx_k_get_state_and_legacy_can_only_be__pyx_k_greater__pyx_k_gumbel__pyx_k_gumbel_loc_0_0_scale_1_0_size_N__pyx_k_has_gauss__pyx_k_high__pyx_k_hypergeometric__pyx_k_hypergeometric_ngood_nbad_nsamp__pyx_k_id__pyx_k_ignore__pyx_n_s_import__pyx_k_import__pyx_k_index__pyx_k_int16__pyx_k_int32__pyx_k_int64__pyx_k_int8__pyx_k_intp__pyx_k_isfinite__pyx_k_isnan__pyx_k_isnative__pyx_k_isscalar__pyx_k_issubdtype__pyx_k_item__pyx_k_itemsize__pyx_k_kappa__pyx_k_key__pyx_k_kwargs__pyx_k_l__pyx_k_lam__pyx_k_laplace__pyx_k_laplace_loc_0_0_scale_1_0_size__pyx_k_left__pyx_k_left_mode__pyx_k_left_right__pyx_k_legacy__pyx_k_legacy_can_only_be_True_when_the__pyx_k_legacy_seeding__pyx_k_less__pyx_k_less_equal__pyx_k_loc__pyx_k_lock__pyx_k_logical_or__pyx_k_logistic__pyx_k_logistic_loc_0_0_scale_1_0_size__pyx_k_lognormal__pyx_k_lognormal_mean_0_0_sigma_1_0_si__pyx_k_logseries__pyx_k_logseries_p_size_None_Draw_samp__pyx_k_low__pyx_k_main__pyx_k_may_share_memory__pyx_k_mean__pyx_k_mean_and_cov_must_have_same_leng__pyx_k_mean_must_be_1_dimensional__pyx_k_mode__pyx_k_mode_right__pyx_k_mt19937__pyx_k_mtrand_pyx__pyx_k_mu__pyx_k_multinomial__pyx_k_multinomial_n_pvals_size_None_D__pyx_k_multivariate_normal__pyx_k_multivariate_normal_mean_cov_si__pyx_k_n__pyx_k_name__pyx_k_nbad__pyx_k_ndim__pyx_k_negative_binomial__pyx_k_negative_binomial_n_p_size_None__pyx_k_newbyteorder__pyx_k_ngood__pyx_k_ngood_nbad_nsample__pyx_k_nonc__pyx_k_noncentral_chisquare__pyx_k_noncentral_chisquare_df_nonc_si__pyx_k_noncentral_f__pyx_k_noncentral_f_dfnum_dfden_nonc_s__pyx_k_normal__pyx_k_normal_loc_0_0_scale_1_0_size_N__pyx_k_np__pyx_k_nsample__pyx_k_numpy__pyx_k_numpy_core_multiarray_failed_to__pyx_k_numpy_core_umath_failed_to_impor__pyx_k_numpy_linalg__pyx_k_numpy_random_mtrand__pyx_k_object__pyx_k_object_which_is_not_a_subclass__pyx_k_operator__pyx_k_p__pyx_k_p_must_be_1_dimensional__pyx_k_pareto__pyx_k_pareto_a_size_None_Draw_samples__pyx_k_permutation__pyx_k_permutation_x_Randomly_permute__pyx_k_pickle__pyx_k_poisson__pyx_k_poisson_lam_1_0_size_None_Draw__pyx_k_poisson_lam_max__pyx_k_pos__pyx_k_power__pyx_k_power_a_size_None_Draws_samples__pyx_k_probabilities_are_not_non_negati__pyx_k_probabilities_contain_NaN__pyx_k_probabilities_do_not_sum_to_1__pyx_k_prod__pyx_k_pvals__pyx_k_pvals_must_be_a_1_d_sequence__pyx_k_pyx_vtable__pyx_k_raise__pyx_k_rand__pyx_k_rand_2__pyx_k_rand_d0_d1_dn_Random_values_in__pyx_k_randint__pyx_k_randint_low_high_None_size_None__pyx_k_randn__pyx_k_randn_d0_d1_dn_Return_a_sample__pyx_k_random__pyx_k_random_integers__pyx_k_random_integers_low_high_None_s__pyx_k_random_sample__pyx_k_random_sample_size_None_Return__pyx_k_randomstate_ctor__pyx_k_ranf__pyx_k_range__pyx_k_ravel__pyx_k_rayleigh__pyx_k_rayleigh_scale_1_0_size_None_Dr__pyx_k_reduce__pyx_k_reduce_2__pyx_k_replace__pyx_k_reshape__pyx_k_return_index__pyx_k_reversed__pyx_k_right__pyx_k_rtol__pyx_k_sample__pyx_k_scale__pyx_k_searchsorted__pyx_k_seed__pyx_k_seed_seed_None_Reseed_a_legacy__pyx_k_set_bit_generator__pyx_k_set_state__pyx_k_set_state_can_only_be_used_with__pyx_k_shape__pyx_k_shuffle__pyx_k_shuffle_x_Modify_a_sequence_in__pyx_k_side__pyx_k_sigma__pyx_k_singleton__pyx_k_size__pyx_k_sort__pyx_k_sqrt__pyx_k_stacklevel__pyx_k_standard_cauchy__pyx_k_standard_cauchy_size_None_Draw__pyx_k_standard_exponential__pyx_k_standard_exponential_size_None__pyx_k_standard_gamma__pyx_k_standard_gamma_shape_size_None__pyx_k_standard_normal__pyx_k_standard_normal_size_None_Draw__pyx_k_standard_t__pyx_k_standard_t_df_size_None_Draw_sa__pyx_k_state__pyx_k_state_dictionary_is_not_valid__pyx_k_state_must_be_a_dict_or_a_tuple__pyx_k_str__pyx_k_strides__pyx_k_subtract__pyx_k_sum__pyx_k_sum_pvals_1_1_0__pyx_k_sum_pvals_1_astype_np_float64_1__pyx_k_svd__pyx_k_take__pyx_k_test__pyx_k_tobytes__pyx_k_tol__pyx_k_tomaxint_size_None_Return_a_sam__pyx_k_triangular__pyx_k_triangular_left_mode_right_size__pyx_k_type__pyx_k_u4__pyx_k_uint16__pyx_k_uint32__pyx_k_uint64__pyx_k_uint8__pyx_k_uniform__pyx_k_uniform_low_0_0_high_1_0_size_N__pyx_k_unique__pyx_k_unsafe__pyx_k_vonmises__pyx_k_vonmises_mu_kappa_size_None_Dra__pyx_k_wald__pyx_k_wald_mean_scale_size_None_Draw__pyx_k_warn__pyx_k_warnings__pyx_k_weibull__pyx_k_weibull_a_size_None_Draw_sample__pyx_k_writeable__pyx_k_x_must_be_an_integer_or_at_least__pyx_k_you_are_shuffling_a__pyx_k_zeros__pyx_k_zipf__pyx_k_zipf_a_size_None_Draw_samples_f__pyx_methods__pyx_moduledef_slots__pyx_methods_5numpy_6random_6mtrand_RandomState__pyx_getsets_5numpy_6random_6mtrand_RandomState__pyx_doc_5numpy_6random_6mtrand_11RandomState_12seed__pyx_doc_5numpy_6random_6mtrand_11RandomState_14get_state__pyx_doc_5numpy_6random_6mtrand_11RandomState_16set_state__pyx_doc_5numpy_6random_6mtrand_11RandomState_18random_sample__pyx_doc_5numpy_6random_6mtrand_11RandomState_20random__pyx_doc_5numpy_6random_6mtrand_11RandomState_22beta__pyx_doc_5numpy_6random_6mtrand_11RandomState_24exponential__pyx_doc_5numpy_6random_6mtrand_11RandomState_26standard_exponential__pyx_doc_5numpy_6random_6mtrand_11RandomState_28tomaxint__pyx_doc_5numpy_6random_6mtrand_11RandomState_30randint__pyx_doc_5numpy_6random_6mtrand_11RandomState_32bytes__pyx_doc_5numpy_6random_6mtrand_11RandomState_34choice__pyx_doc_5numpy_6random_6mtrand_11RandomState_36uniform__pyx_doc_5numpy_6random_6mtrand_11RandomState_38rand__pyx_doc_5numpy_6random_6mtrand_11RandomState_40randn__pyx_doc_5numpy_6random_6mtrand_11RandomState_42random_integers__pyx_doc_5numpy_6random_6mtrand_11RandomState_44standard_normal__pyx_doc_5numpy_6random_6mtrand_11RandomState_46normal__pyx_doc_5numpy_6random_6mtrand_11RandomState_48standard_gamma__pyx_doc_5numpy_6random_6mtrand_11RandomState_50gamma__pyx_doc_5numpy_6random_6mtrand_11RandomState_52f__pyx_doc_5numpy_6random_6mtrand_11RandomState_54noncentral_f__pyx_doc_5numpy_6random_6mtrand_11RandomState_56chisquare__pyx_doc_5numpy_6random_6mtrand_11RandomState_58noncentral_chisquare__pyx_doc_5numpy_6random_6mtrand_11RandomState_60standard_cauchy__pyx_doc_5numpy_6random_6mtrand_11RandomState_62standard_t__pyx_doc_5numpy_6random_6mtrand_11RandomState_64vonmises__pyx_doc_5numpy_6random_6mtrand_11RandomState_66pareto__pyx_doc_5numpy_6random_6mtrand_11RandomState_68weibull__pyx_doc_5numpy_6random_6mtrand_11RandomState_70power__pyx_doc_5numpy_6random_6mtrand_11RandomState_72laplace__pyx_doc_5numpy_6random_6mtrand_11RandomState_74gumbel__pyx_doc_5numpy_6random_6mtrand_11RandomState_76logistic__pyx_doc_5numpy_6random_6mtrand_11RandomState_78lognormal__pyx_doc_5numpy_6random_6mtrand_11RandomState_80rayleigh__pyx_doc_5numpy_6random_6mtrand_11RandomState_82wald__pyx_doc_5numpy_6random_6mtrand_11RandomState_84triangular__pyx_doc_5numpy_6random_6mtrand_11RandomState_86binomial__pyx_doc_5numpy_6random_6mtrand_11RandomState_88negative_binomial__pyx_doc_5numpy_6random_6mtrand_11RandomState_90poisson__pyx_doc_5numpy_6random_6mtrand_11RandomState_92zipf__pyx_doc_5numpy_6random_6mtrand_11RandomState_94geometric__pyx_doc_5numpy_6random_6mtrand_11RandomState_96hypergeometric__pyx_doc_5numpy_6random_6mtrand_11RandomState_98logseries__pyx_doc_5numpy_6random_6mtrand_11RandomState_100multivariate_normal__pyx_doc_5numpy_6random_6mtrand_11RandomState_102multinomial__pyx_doc_5numpy_6random_6mtrand_11RandomState_104dirichlet__pyx_doc_5numpy_6random_6mtrand_11RandomState_106shuffle__pyx_doc_5numpy_6random_6mtrand_11RandomState_108permutation__pyx_doc_5numpy_6random_6mtrand_8ranf__pyx_doc_5numpy_6random_6mtrand_6sample__pyx_doc_5numpy_6random_6mtrand_4set_bit_generator__pyx_doc_5numpy_6random_6mtrand_2get_bit_generator__pyx_doc_5numpy_6random_6mtrand_seedderegister_tm_clones__do_global_dtors_auxcompleted.0__do_global_dtors_aux_fini_array_entryframe_dummy__frame_dummy_init_array_entryfe_doublewe_doubleke_doublefe_floatwe_floatke_floatwi_doubleki_doublefi_doublewi_floatki_floatfi_float__FRAME_END____GNU_EH_FRAME_HDR_GLOBAL_OFFSET_TABLE___TMC_END____dso_handle_DYNAMICPyExc_ImportErrorPyInterpreterState_GetIDPyDict_SetItemStringexp@@GLIBC_2.2.5PyUnicode_Compare__pyx_module_is_main_numpy__random__mtrandPyExc_StopIterationrandom_buffered_bounded_uint32legacy_gaussPyFrame_New_PyUnicode_FastCopyCharactersrandom_weibullPyBool_TypePyModule_NewObjectrandom_positive_intmemset@@GLIBC_2.2.5legacy_waldrandom_standard_gammaPyNumber_Remainderceil@@GLIBC_2.2.5PyExc_ValueErrorPyType_ReadyPyUnicode_Formatlegacy_logseriesPyList_AsTuplePyExc_SystemErrorrandom_bounded_uint16_fill__gmon_start__log1p@@GLIBC_2.2.5legacy_lognormalrandom_waldrandom_standard_normalrandom_geometric_searchPyCapsule_Type_finirandom_positive_int64legacy_standard_exponentialPyList_TypePyObject_GC_UnTrackrandom_rayleighrandom_uint_PyObject_GetDictPtrrandom_loggamPyObject_IsTruerandom_standard_normal_fillPyExc_TypeErrorPyMem_ReallocPyExc_Exceptionlegacy_betaPyObject_NotPyDict_TypePyErr_GivenExceptionMatchesPyDict_GetItemWithErrorrandom_bounded_uint8_fillPyExc_KeyErrorPyOS_snprintfPyUnicode_AsUTF8AndSizerandom_powerPyUnicode_FromFormatlegacy_random_hypergeometriclog1pf@@GLIBC_2.2.5PyDict_NextPyImport_AddModulePyLong_AsLongPy_EnterRecursiveCallPyObject_CallFinalizerFromDealloclegacy_random_zipfPyErr_Format_ITM_deregisterTMCloneTablePyFloat_FromDoublerandom_bounded_uint64_fillPyBaseObject_Typefmod@@GLIBC_2.2.5PyNumber_LongPyNumber_InPlaceAddPyObject_GetAttrPyErr_ExceptionMatchesPyErr_OccurredPyDict_CopyPyNumber_InPlaceTrueDividerandom_gamma_frandom_exponentialrandom_standard_cauchyPyType_Modifiedlegacy_gamma_ITM_registerTMCloneTablerandom_standard_exponential_inv_fill_fPyCapsule_NewPyFloat_Typerandom_standard_exponentialrandom_standard_exponential_inv_fillPyObject_GC_IsFinalized__cxa_finalize@@GLIBC_2.2.5PyExc_IndexError_Py_FalseStructPyErr_SetObjectPyNumber_MultiplyPyEval_RestoreThread__isnan@@GLIBC_2.2.5PyUnicode_InternFromStringrandom_standard_uniform_fPyUnicode_TypePyExc_DeprecationWarningrandom_logseriesPyObject_Size_Py_TrueStructPyTuple_PackPyLong_FromStringPyEval_SaveThreadlegacy_powerPyThreadState_GetPyTuple_TypePyErr_SetExcInforandom_vonmisesPyDict_Sizepow@@GLIBC_2.2.5PyModule_GetDictPyCode_NewEmptyPyErr_RestorePyImport_ImportModuleLevelObjectPyCapsule_GetNamelegacy_exponentialPyObject_SetItemPyLong_AsSsize_tlegacy_fPyObject_RichComparerandom_intervalPyObject_GetIterPyExc_RuntimeErrorPyType_IsSubtypePy_LeaveRecursiveCalllegacy_standard_cauchyrandom_buffered_bounded_uint8legacy_noncentral_chisquarePyUnicode_Decoderandom_standard_normal_fill_frandom_bounded_uint32_fillPy_GetVersionPyNumber_Addrandom_binomial_btpelegacy_noncentral_frandom_standard_uniformPyErr_WarnExPyErr_NormalizeExceptionPyMethod_Typelegacy_standard_gammarandom_negative_binomialrandom_lognormalrandom_chisquarePyErr_FetchPyModule_GetNameexpf@@GLIBC_2.2.5random_positive_int32PyFloat_AsDoublePyObject_Formatlegacy_paretorandom_bounded_bool_fillPyObject_HashPyNumber_SubtractPySlice_NewPyObject_GetAttrStringrandom_normallegacy_negative_binomiallegacy_random_multinomialPyObject_IsInstancePyLong_Typerandom_uniformrandom_bounded_uint64random_logistic_Py_EllipsisObjectrandom_paretopowf@@GLIBC_2.2.5PyObject_SetAttrString_Py_NoneStructPyExc_OverflowErrorPyList_AppendPyObject_SetAttrrandom_triangularrandom_standard_uniform_fill_facos@@GLIBC_2.2.5PyBytes_FromStringAndSizePyDict_GetItemStringPySequence_Tuplelegacy_random_poisson_PyDict_GetItem_KnownHashPyExc_AttributeErrorrandom_standard_normal_frandom_standard_tPyLong_FromLongrandom_standard_uniform_fillPyTraceBack_HerePyExc_NameErrorlegacy_random_geometricrandom_gumbelmemcmp@@GLIBC_2.2.5PyException_SetTraceback_Py_Dealloclegacy_standard_tPyImport_ImportModulerandom_betaPyErr_GetExcInforandom_standard_exponential_fillPyErr_Clearlegacy_rayleighlegacy_chisquarelegacy_random_binomialPyTuple_NewPyCFunction_TypePyCMethod_NewPyUnicode_NewPyImport_GetModuleDictrandom_laplaceexpm1@@GLIBC_2.2.5random_standard_gamma_frandom_gammaPyLong_FromSsize_tPyErr_WarnFormatPyErr_SetStringPySequence_ContainsPyModuleDef_InitPyInit_mtrandPyUnicode_FromStringAndSize_PyDict_NewPresizedPyCapsule_IsValidfloor@@GLIBC_2.2.5legacy_weibullrandom_frandom_buffered_bounded_bool_PyType_Lookup_PyUnicode_Readyrandom_noncentral_chisquarerandom_geometric_inversionrandom_buffered_bounded_uint16PyList_Newsqrt@@GLIBC_2.2.5PyMem_Mallocrandom_standard_exponential_fPyUnicode_AsUTF8memcpy@@GLIBC_2.14PyCapsule_GetPointerrandom_binomial_inversionPyNumber_Indexlegacy_vonmiseslog@@GLIBC_2.2.5PyDict_SetItemlogf@@GLIBC_2.2.5PySequence_Listrandom_standard_exponential_fill_f_initPyDict_Newsqrtf@@GLIBC_2.2.5random_noncentral_flegacy_normal.symtab.strtab.shstrtab.note.gnu.build-id.gnu.hash.dynsym.dynstr.gnu.version.gnu.version_r.rela.dyn.rela.plt.init.text.fini.rodata.eh_frame_hdr.eh_frame.init_array.fini_array.data.rel.ro.dynamic.got.got.plt.data.bss.comment88$.o``8 @GHo00 Uo22Pd8383snBЦЦHxs @ ~``? 0 00,` `  9k[k[k[k[m]p`0@u@eu @ @ X 0@ /p 8 z &