
    {Kgq                       U d Z ddlZddlZddlZddlZddlZddlmZmZ ddl	Z	ddl
ZddlZddlmZ ddlmZ ddlmZmZmZ ddlmZ ddlmZ dd	lmZ dd
lmZmZmZ ddlm Z  ddl!m"Z" ddl#m$Z$ ddl%m&Z&m'Z'm(Z(m)Z) ddl*m+Z+m,Z,m-Z-m.Z. ddl/m0Z0m1Z1m2Z2m3Z3m4Z4m5Z5 ddl/m6Z7 ddl8m9Z9 ddl:m;Z;m<Z<m=Z=m>Z>m?Z?m@Z@ ddlAmBZB ddlCmDZDmEZEmFZFmGZG ddlHmIZI dZJdZKe&e(dZLe'e)dZM eN       ZOeNePd<   eOj                  eL       eOj                  eM       g dZR ej                  g dg dg dg dg d g d!g d"g d#g d$g d%g d&g d'g d(g d)g d*g d+g d,g d-g d.g d/g d0g d1g d2g      ZTg d3ZUg d4ZVd5d6gd6d6gd6d5gd7d7gd7d8gd8d7ggZWg d9ZXd6d6gd8d8gd:d8ggZYg d;ZZ ej                         Z\ej                  j                  d7      Z_e_j                  e\j                  j                        Zce\j                  ec   e\_d        e\j                  ec   e\_a         ej                         Zfe_j                  efj                  j                        Zcefj                  ec   ef_d        efj                  ec   ef_a         ej                         Zhe_j                  ehj                  j                        Zcehj                  ec   eh_d        ehj                  ec   eh_a         eId      Zi ej                  dd<d=>      \  ZkZleij                  d?@      ZndAenendBk  <   eij                  ddCdD@      Zp e$dEd=dFdG      j                         Zre\j                  e\j                  dHefj                  efj                  dHehj                  ehj                  dHeWeXdHeTeUdHeTeVdHekeldHenepdHen epdHerepdH ej                  dI      epdHdJZtdK ZudL ZvdM Zwej                  j                  dNeMj                               ej                  j                  dOeK      dP               Z{dQ Z|dR Z}ej                  j                  dSeMj                               ej                  j                  dOeK      dT               Ze@ej                  j                  dSeMj                               ej                  j                  dUdVdWedXfdYdEedXfdZdWedXfd[dWed<fg      d\                      Zd] Zd^ Zd_ Zd` Zda Zdb Zdc Zdd Zde Zdf Zdg ZddhZej                  j                  dieO      dj        Zej                  j                  dieR      ej                  j                  dkeF      dl               Z	 ddmZej                  j                  dieO      dn        Zej                  j                  dieR      ej                  j                  dkeF      do               Zdp Zdq Zdr Zds Zdt Zdu Zdv Zdw Zej                  j                  dieL      dx        Zej                  j                  dieL      dy        Zdz Zd{ Zd| Zd} Zd~ Zd Zd Zd Zd ZddZej                  j                  deR      ej                  j                  dd      d               Zej                  j                  d e eeR      jS                  eM                  ej                  j                  dddg      d               Zej                  j                  deR      ej                  j                  dg d      ej                  j                  dkeF      d                      Zej                  j                  d e eeRD  cg c]	  } | eMv s|  c} eK             e eeRD  cg c]	  } | eLv s|  c} eJ            z         ej                  j                  dg d      ej                  j                  dkeF      d                      Zej                  j                  deR      ej                  j                  d eeFeG            d               Ze?d        Zej                  j                  dieO      d        Zej                  j                  dieO      ej                  j                  ddgeFz         d               Zej                  j                  dieO      d        Zej                  j                  dieR      ej                  j                  deG      d               Zd Zej                  j                  dieO      d        Zej                  j                  dieO      ej                  j                  deG      d               Zd Zd Zej                  j                  ddgeFz         d        Zej                  j                  d e eetjw                               ddhz
              ej                  j                  de&e(g      d               Zej                  j                  detjw                               ej                  j                  de'e)g      d               Zd Zd Zd Zej                  j                  dieO      ej                  j                  dddg      ej                  j                  ddgeFz   eGz         d                      Zej                  j                  dOg d      ej                  j                  dNeMj                               d               Zej                  j                  d ed:            d        Zd Zej                  j                  dOeK      d        Zej                  j                  dNe&e(g      ej                  j                  dd8dCg      d               Zd Zd Zd Zd Zd Zd Zd Zd Zd Zd Zej                  j                  d e e-j                          e.j                                     d        Zd Zej                  j                  dNeOj                               d        Zej                  j                  dOdVdZg      d        Zej                  j                  dOddg      d        Zej                  j                  dOddg      d        Zej                  j                  dOddg      dÄ        Zej                  j                  dOddg      dĄ        Zej                  j                  ddgeGz         ej                  j                  d e&dƫ       e'dYǫ      g      dȄ               ZdɄ Zdʄ Zej                  j                  dej                  e'fee&fg      ej                  j                  dddg      d΄               Zdτ Zej                  j                  dej                  e'fej                  e&fg      dф        Zd҄ Zej                  j                  d ej                  ej                  d8ej                  dCddg       ej                  ej                  ej                  d:dCddg       ej                  d7d8d:dCej                  ej                  g       ej                  d7d8d:ej                  dej                  g      g      ej                  j                  dOdVdZg      dք               Zdׄ Zyc c} w c c} w )z-
Testing for the tree module (sklearn.tree).
    N)chainproduct)NumpyPickler)assert_allclose)clonedatasetstree)DummyRegressor)NotFittedError)SimpleImputer)accuracy_scoremean_poisson_deviancemean_squared_error)train_test_split)make_pipeline)_sparse_random_matrix)DecisionTreeClassifierDecisionTreeRegressorExtraTreeClassifierExtraTreeRegressor)CRITERIA_CLFCRITERIA_REGDENSE_SPLITTERSSPARSE_SPLITTERS)
NODE_DTYPE	TREE_LEAFTREE_UNDEFINED_check_n_classes_check_node_ndarray_check_value_ndarray)Tree)compute_sample_weight)assert_almost_equalassert_array_almost_equalassert_array_equalcreate_memmap_backed_dataignore_warningsskip_if_32bit)check_sample_weights_invariance)	_IS_32BITCOO_CONTAINERSCSC_CONTAINERSCSR_CONTAINERS)check_random_state)ginilog_loss)squared_errorabsolute_errorfriedman_msepoisson)r   r   )r   r   	ALL_TREES)r   r      r   r   r      ir   r   r   r   r   )r   r         r   r8   r   r   r7   皙?r   r6   r7   )r=   r   r         r   r    @r7   r   r   r>   r   r7   )r=   r=   r   g333333r   r   r   r   r   r   r<   r   r   r7   )r=   r=   r   r   r   r   r   r:   r   r   r   r   r   r7   )r=   r   r6   
   r6   r   皙	r   r6   r:   r8   r7   )zG @r         r      r   r   rC            ?r   rA   r7   )rD   r   rE   rF   r   rG   r   r   rC   rH   r   r   r@   r7   )rD      rE   rF   r   rG   r   r   rC   rH   r   r   r@   r7   )rD   rJ   rE   rF   r   rG   r   r   rC   rH   rI   r   r=   r   )   rJ   r9   r7   rI   r8   rB   r   r7   r;   r:   r   rK   r   )rK   r   r7   r7   r7   r=   r7   r   r   r@   r:   r   r7   r   )rK   r   r7   rK   r:   r=   rB   rK   r   r=   r7   rK   rK   r   )r7   r7   r   rK   rK   r=   r7   rK   r   r;   r7   rK   r:   r   )r:   r7   r   r:   r   r8   rB   r   r7   r;   r:   r   r:   r7   )rD   rJ   rE   rF   r   r7   r   r   rC   rH   rI   r   rA   r7   )rD   rJ   rE   rF   r   r7   r   r   rC   rH         ?r7   r=   r=   )rD   rJ   rE   rF   r   rB   r   r   rC   rH   rI   r   r=   r=   )rK   r   r9   r7   rI   r@   rB   r   r7   r;   r:   r7   r   r=   )rK   r   r7   r7   r7   r@   r7   r   r   r@   r   r   r   r7   )rK   r7   r7   r7   rK   r=   rB   rK   r   r=   r   rK   r7   r7   )r7   r7   r   r   r7   rA   r7   rK   r   r;   r7   rK   r7   r7   )r:   r7   r   r7   r   r8   r7   r   r7   r@   r   r   r7   r   )r7   r7   r   r   r   r   r7   r7   r7   r7   r7   r7   r   r   r   r7   r   r   r7   r   r   r   r   )      ?r?   333333?皙?rB   g333333@@g)\(?{Gz?gףp=
@rP   g?        rN   rK   rG   r   r         @g|?5^?g(\??r   r@   r=   r7   rK   )r=   r=   r=   r7   r7   r7   r:   )r=   r7   r7      rB   )random_state	n_samples
n_features)   r9   sizerR   g?r6   )rY   rY   g      ?)densityrV   Xy)rY   r:   )irisdiabetesdigitstoy	clf_small	reg_small
multilabel
sparse-pos
sparse-neg
sparse-mixzerosc                 ~   |j                   | j                   k(  s,J dj                  ||j                   | j                                t        | j                  |j                  |dz          t        | j                  |j                  |dz          | j                  t
        k(  }t        j                  |      }t        | j                  |   |j                  |   |dz          t        | j                  |   |j                  |   |dz          t        | j                  j                         |j                  j                         |dz          t        | j                  |j                  |dz          t        | j                  |j                  |dz   	       t        | j                  |   |j                  |   |d
z   	       y )Nz({0}: inequal number of node ({1} != {2})z: inequal children_rightz: inequal children_leftz: inequal featuresz: inequal thresholdz: inequal sum(n_node_samples)z: inequal n_node_samplesz: inequal impurityerr_msgz: inequal value)
node_countformatr%   children_rightchildren_leftr   nplogical_notfeature	thresholdn_node_samplessumr#   impurityr$   value)dsmessageexternalinternals        `/home/alanp/www/video.onchill/myenv/lib/python3.12/site-packages/sklearn/tree/tests/test_tree.pyassert_tree_equalr      s   	$188q||$
 	!**G6P,P 	'4M*M 9,H~~h'H			(QYYx0'<P2P 	Hq{{84g@U6U 		11
 	!**G6P,P 

AJJBV8VW	1778,g@Q6Q    c                     t         j                         D ]  \  } } |d      }|j                  t        t               t        |j                  t              t        dj                  |               |dd      }|j                  t        t               t        |j                  t              t        dj                  |               y )Nr   rV   Failed with {0}r7   )max_featuresrV   )
	CLF_TREESitemsfitr^   r_   r%   predictTtrue_resultro   namer!   clfs      r   test_classification_toyr      s    oo'
d"13;;q>;8I8P8PQU8VW213;;q>;8I8P8PQU8VW (r   c            
         t         j                         D ]  \  } } |d      }|j                  t        t        t        j                  t        t                           t        |j                  t              t        dj                  |              |j                  t        t        t        j                  t        t              d             t        |j                  t              t        dj                  |               y )Nr   r   sample_weightr   rI   )r   r   r   r^   r_   rr   oneslenr%   r   r   r   ro   fullr   s      r    test_weighted_classification_toyr      s    oo'
d"1BGGCFO43;;q>;8I8P8PQU8VW1BGGCFC$893;;q>;8I8P8PQU8VW (r   r!   	criterionc                    |dk(  rht        j                  t        j                  t                    dz   }t        j                  t              |z   }t        j                  t
              |z   }nt        }t
        } | |d      }|j                  t        |       t        |j                  t              |        | |dd      }|j                  t        |       t        |j                  t              |       y )Nr4   r7   r   rV   r   r   rV   )rr   absminr_   arrayr   r   r^   r   r   r   )r!   r   ay_trainy_testregr   s          r   test_regression_toyr     s     I FF266!9!((1+/+&*

3CGGAwCKKNF+

CCGGAwCKKNF+r   c                     t        j                  d      } d| d dd df<   d| dd dd f<   t        j                  | j                        \  }}t        j                  |j                         |j                         g      j                  }| j                         } t        j                         D ]  \  }} |d      }|j                  ||        |j                  ||       dk(  sJ dj                  |              |dd      }|j                  ||        |j                  ||       dk(  r~J dj                  |              y )	N)rB   rB   r7   r9   r   r   rM   r   rV   r   )rr   rj   indicesshapevstackravelr   r   r   r   scorero   )r_   gridxgridyr^   r   r!   r   s          r   test_xorr     s   
AAbqb"1"fIAab!"fI::agg&LE5
		5;;=%++-0133A		Aoo'
d"1yyA#%E'8'?'?'EE%21yyA#%E'8'?'?'EE% (r   c                     t        t        j                         t              D ]"  \  \  } }} ||d      }|j	                  t
        j                  t
        j                         t        |j                  t
        j                        t
        j                        }|dkD  sJ dj                  | ||              ||dd      }|j	                  t
        j                  t
        j                         t        |j                  t
        j                        t
        j                        }|dkD  rJ dj                  | ||              y )Nr   r   rT   z0Failed with {0}, criterion = {1} and score = {2}rK   r   rI   )r   r   r   CLF_CRITERIONSr   r`   datatargetr   r   ro   )r   r!   r   r   r   s        r   	test_irisr   1  s    #*9??+<n#MtiYQ7		4;;'s{{4995t{{Cs{ 	
NUU)U
 	
{ YQQG		4;;'s{{4995t{{Cs{ 	
NUU)U
 	
{ $Nr   z
name, Treec                 2    ||d      }|j                  t        j                  t        j                         t	        t        j                  |j                  t        j                              }|t        j                  d      k(  sJ d|  d| d|        y )Nr   r   zFailed with z, criterion = z and score = )r   ra   r   r   r   r   pytestapprox)r   r!   r   r   r   s        r   test_diabetes_overfitr   C  s    
 
3CGGHMM8??+xHMM0JKEFMM	  J	dV>)M%IJ r   z&criterion, max_depth, metric, max_lossr1      <   r2   r3   r4   c                      |||dd      }|j                  t        j                  t        j                          |t        j                  |j	                  t        j                              }d|cxk  r|k  sJ  J y )NrH   r   )r   	max_depthr   rV   )r   ra   r   r   r   )r   r!   r   r   metricmax_lossr   losss           r   test_diabetes_underfitr   P  sa     iaVW
XCGGHMM8??+(//3;;x}}#=>Dthr   c            	          t         j                         D ]v  \  } } |ddd      }|j                  t        j                  t        j
                         |j                  t        j                        }t        t        j                  |d      t        j                  t        j                  j                  d         dj                  |              t        t        j                  |d      |j                  t        j                        dj                  |              t!        |j                  t        j                        t        j"                  |j%                  t        j                              ddj                  |              y y )Nr7   *   r   r   rV   r   r   rl   rJ   )r   r   r   r`   r   r   predict_probar$   rr   rw   r   r   ro   r%   argmaxr   r#   exppredict_log_proba)r   r!   r   prob_predicts       r   test_probabilityr   e  s     oo'
dQQR@		4;;'((3!FF<#GGDIIOOA&'%,,T2	

 	IIlA&KK		"%,,T2	

 	dii(FF3((34%,,T2		
 (r   c                      t        j                  d      d d t         j                  f   } t        j                  d      }t        j	                         D ]!  \  }} |d d      }|j                  | |       # y )Ni'  r   r   rV   )rr   arangenewaxis	REG_TREESr   r   r^   r_   r   r!   r   s        r   test_arrayreprr     s[     			%BJJ'A
		%Aoo'
dT21 (r   c                     ddgddgddgddgddgddgg} g d}t         j                         D ]L  \  }} |d      }|j                  | |       t        |j	                  |       |dj                  |      	       N t        j                         D ]L  \  }} |d      }|j                  | |       t        |j	                  |       |dj                  |      	       N y )
Nr@   r=   r7   rK   )r7   r7   r7   r7   r7   r7   r   r   r   rl   )r   r   r   r%   r   ro   r   r#   )r^   r_   r   TreeClassifierr   TreeRegressorr   s          r   test_pure_setr     s    
bB8b"X1v1v1v>AA ) 1n!,13;;q>16G6N6Nt6TU !2
  )0m+1CKKNA7H7O7OPT7UV  1r   c            
         t        j                  g dg dg dg dg dg dg dg      } t        j                  g d      }t        j                  d	
      5  t        j	                         D ]Z  \  }} |d      }|j                  | |       |j                  | |        |j                  |  |       |j                  |  |        \ 	 d d d        y # 1 sw Y   y xY w)N)gs_c@d	a@籛 `8`@?c@)g_9a@g 8`@g-Vu]@g    @Xd@)gSW j_@r   r   r   )g ً`@4Ta@	lKa@{c@)g|@Y@g~G`a@gwI?lKa@g/"c@)g_@r   r   r   )g:^@r   r   r   )rM   gAw?gtQ?5??rR   g7G?gۺ?gb'?raise)allr   r   )rr   r   errstater   r   r   r   s        r   test_numerical_stabilityr     s    
DDDDDDD	

	A 	WXA		!#//+JD$A&CGGAqMGGArNGGQBNGGQBO , 
"	!	!s   A2CCc            	         t        j                  ddddddd      \  } }t        j                         D ]  \  }} |d      }|j	                  | |       |j
                  }t        j                  |dkD        }|j                  d   dk(  sJ d	j                  |             |dk(  rsJ d	j                  |              t        d      }|j	                  t        j                  t        j                         t        dt        t        j                        
      }|j	                  t        j                  t        j                         t        |j
                  |j
                         y )N  rB   r:   r   FrW   rX   n_informativen_redundant
n_repeatedshufflerV   r   皙?r   rV   max_leaf_nodes)r   make_classificationr   r   r   feature_importances_rr   rw   r   ro   r   r`   r   r   r   r%   )r^   r_   r   r!   r   importancesn_importantclf2s           r   test_importancesr     s"   ''DAq  oo'
d"1..ff[3./  #r)I+<+C+CD+II)a?!2!9!9$!?? ( !a
0CGGDIIt{{#!qTYYPDHHTYY$s//1J1JKr   c                      t               } t        j                  t              5  t	        | d       d d d        y # 1 sw Y   y xY w)Nr   )r   r   raises
ValueErrorgetattr)r   s    r   test_importances_raisesr     s-    
 
"C	z	"+, 
#	"	"s	   :Ac            	         t        j                  ddddddd      \  } }t        ddd	      j                  | |      }t	        d
dd	      j                  | |      }t        |j                  |j                         t        |j                  j                  |j                  j                         t        |j                  j                  |j                  j                         t        |j                  j                  |j                  j                         t        |j                  j                  |j                  j                         y )Ni  rB   r:   r   Fr   r/   r9   )r   r   rV   r1   )r   r   r   r   r   r#   r   r%   tree_rt   rq   rp   rv   )r^   r_   r   r   s       r   )test_importances_gini_equal_squared_errorr     s     ''DAq !6QQ
O
S
S	1C  !QQ	c!Qi  00#2J2JKsyy((#))*;*;<syy..		0G0GHsyy//1I1IJsyy//1I1IJr   c                  V   t         j                         D ]  \  } } |d      }|j                  t        j                  t        j
                         |j                  t        t        j                  t        j                  j                  d               k(  sJ  |d      }|j                  t        j                  t        j
                         |j                  t        t        j                  t        j                  j                  d               k(  sJ  |d      }|j                  t        j                  t        j
                         |j                  dk(  sJ  |d      }|j                  t        j                  t        j
                         |j                  dk(  sJ  |d      }|j                  t        j                  t        j
                         |j                  dk(  sJ  |d      }|j                  t        j                  t        j
                         |j                  t        dt        j                  j                  d   z        k(  sJ  |d      }|j                  t        j                  t        j
                         |j                  t        j                  j                  d   k(  sJ  |d       }|j                  t        j                  t        j
                         |j                  t        j                  j                  d   k(  rJ  y )	Nsqrt)r   r7   log2r:   rQ   rI   rM   )r5   r   r   r`   r   r   max_features_intrr   r   r   r   )r   TreeEstimatorests      r   test_max_featuresr     s   (0m0		4;;'  C		0B(C$DDDD0		4;;'  C		0B(C$DDDD+		4;;'  A%%%+		4;;'  A%%%.		4;;'  A%%%-		4;;'  Cdiiooa.@(@$AAAA-		4;;'  DIIOOA$6666.		4;;'  DIIOOA$6666?  1r   c                  J	   t         j                         D ]o  \  } } |       }t        j                  t              5  |j                  t               d d d        |j                  t        t               g dg}t        j                  t              5  |j                  |       d d d         |       }t        d d }t        j                  t              5  |j                  t        |       d d d        t        j                  t              } |       }|j                  |t               t        |j                  t              t                |       }t        j                  t              5  |j                  t               d d d        |j                  t        t               t        j                   t              }t        j                  t              5  |j                  |d d dd f          d d d        t        j"                  t              j                  } |       }|j                  t        j$                  t        |      t               t        j                  t              5  |j                  t               d d d        t        j                  t              5  |j'                  t               d d d         |       }|j                  t        t               t        j                  t              5  |j                  |       d d d        t        j                  t              5  |j'                  |       d d d         |       }t        j                  t              5  |j'                  t               d d d        r t)        d      }t        j                  t        d      5  |j                  g dgg d	       d d d        t        j                  t        d
      5  |j                  g dgg d       d d d        y # 1 sw Y   xY w# 1 sw Y   vxY w# 1 sw Y   ;xY w# 1 sw Y   xY w# 1 sw Y   OxY w# 1 sw Y   xY w# 1 sw Y   xY w# 1 sw Y   \xY w# 1 sw Y   6xY w# 1 sw Y   uxY w# 1 sw Y   xY w# 1 sw Y   y xY w)N)r@   r=   r7   r=   r7   r4   r   zy is not positive.*Poissonmatchr   r7   rK   )r   r   r   zSome.*y are negative.*Poisson)r9   grK   )r   r   r   r   r   r   r^   r   r_   r   rr   asfortranarrayr#   r   r   r   asarrayr   dotapplyr   )	r   r   r   X2y2XftXtr   s	            r   
test_errorr    s   (0mo]]>*a  + 	1]]]:&b! ' osV]]:&GGArN ' q!oACKKNK8 o]]>*KKN + 	1JJqM]]:&KK!QR%! ' XXa[]]oq"q!]]:&KKN ']]:&IIaL ' o1]]:&KKO ']]:&IIbM ' o]]>*IIaL +*k  1r  )
4C	z)E	FY' 
G	z)H	I\* 
J	Is +*
 '& '& +* '& '&&&
 '&&&
 +*
 
G	F	I	Is   PPP%-P2P?+Q"Q:Q&-Q3'R /R*RP	P"	%P/	2P<	?Q		Q	Q#	&Q0	3Q=	 R
	RR"c                     t        j                  t        j                  t        j
                  j                        } t        j                  }t        dt        j                               D ]  \  }}t        |   } |d|d      }|j                  | |       |j                  j                  |j                  j                  dk7     }t        j                  |      dkD  sJ dj!                  |              |d	|d      }|j                  | |       |j                  j                  |j                  j                  dk7     }t        j                  |      dkD  rJ dj!                  |              y
)z Test min_samples_split parameterdtypeN  rB   r   )min_samples_splitr   rV   r=   	   r   r<   N)rr   r  r`   r   r	   _treeDTYPEr   r   r5   keysr   r   rv   rq   r   ro   )r^   r_   r   r   r   r   node_sampless          r   test_min_samples_splitr  ]  s.   
$))4::+;+;<AA !(inn6F G!$  a
 	1yy//		0G0G20MNvvl#a'G):)A)A$)GG' !.q
 	1yy//		0G0G20MNvvl#a'G):)A)A$)GG'+ !Hr   c                     t        j                  t        j                  t        j
                  j                        } t        j                  }t        dt        j                               D ]  \  }}t        |   } |d|d      }|j                  | |       |j                  j                  |       }t        j                  |      }||dk7     }t        j                  |      dkD  sJ dj!                  |              |d|d      }|j                  | |       |j                  j                  |       }t        j                  |      }||dk7     }t        j                  |      dkD  rJ dj!                  |              y )	Nr  r  r9   r   )min_samples_leafr   rV   r6   r   r   )rr   r  r`   r   r	   r  r  r   r   r5   r  r   r   r  bincountr   ro   )	r^   r_   r   r   r   r   outnode_counts
leaf_counts	            r   test_min_samples_leafr  |  sF   
$))4::+;+;<AA !(inn6F G!$ ~A
 	1iiooa kk#& !12
vvj!A%E'8'?'?'EE%  a
 	1iiooa kk#& !12
vvj!A%E'8'?'?'EE%/ !Hr   c                    t         |   d   j                  t        j                        }| ||      }t         |   d   }t        j                  |j                  d         }t        j                  |      }t        |    }t        dt        j                  ddd            D ]  \  }}	 ||	|d      }
|
j                  |||	       |*|
j                  j                  |j                               }n|
j                  j                  |      }t        j                  ||
      }||dk7     }t        j                   |      ||
j"                  z  k\  rJ dj%                  | |
j"                                |j                  d   }t        dt        j                  ddd            D ]  \  }}	 ||	|d      }
|
j                  ||       |*|
j                  j                  |j                               }n|
j                  j                  |      }t        j                  |      }||dk7     }t        j                   |      ||
j"                  z  k\  rJ dj%                  | |
j"                                y)zPTest if leaves contain at least min_weight_fraction_leaf of the
    training setr^   Nr_   r   r  rI   rH   )min_weight_fraction_leafr   rV   r   )weightsz,Failed with {0} min_weight_fraction_leaf={1})DATASETSastyperr   float32rngrandr   rw   r5   r   linspacer   r   r  tocsrr  r   r   ro   )r   r   sparse_containerr^   r_   r!  total_weightr   r   fracr   r  node_weightsleaf_weightss                 r   check_min_weight_fraction_leafr.    s-    	3&&rzz2A#Q3Ahhqwwqz"G66'?LdOM !(bkk!S!6L M%).WX
 	1G,'))//!''),C))//!$C{{38#LA$56FF< L33O3O$OO	
9@@#..
	
O !N* 771:L 'bkk!S!6L M%).WX
 	1'))//!''),C))//!$C{{3'#LA$56FF< L33O3O$OO	
9@@#..
	
O !Nr   r   c                     t        | d       y Nr`   r.  r   s    r   ,test_min_weight_fraction_leaf_on_dense_inputr3    s    "40r   csc_containerc                      t        | d|       y Nrf   )r)  r1  r   r4  s     r   -test_min_weight_fraction_leaf_on_sparse_inputr8    s     #4Vr   c                    t         |   d   j                  t        j                        }| ||      }t         |   d   }|j                  d   }t
        |    }t        dt        j                  ddd            D ]  \  }} |||dd	      }	|	j                  ||       |*|	j                  j                  |j                               }
n|	j                  j                  |      }
t        j                  |
      }||dk7     }t        j                  |      t        ||	j                  z  d      k\  rJ d
j!                  | |	j                  |	j"                                t        dt        j                  ddd            D ]  \  }} |||dd	      }	|	j                  ||       |*|	j                  j                  |j                               }
n|	j                  j                  |      }
t        j                  |
      }||dk7     }t        j                  |      t        ||	j                  z  ||	j"                  z        k\  rJ d
j!                  | |	j                  |	j"                                y)zzTest the interaction between min_weight_fraction_leaf and
    min_samples_leaf when sample_weights is not provided in fit.r^   Nr_   r   r  rI   r:   r9   )r   r   r  rV   zBFailed with {0} min_weight_fraction_leaf={1}, min_samples_leaf={2}r   )r"  r#  rr   r$  r   r5   r   r'  r   r   r  r(  r  r   maxr   ro   r  )r   r   r)  r^   r_   r*  r   r   r+  r   r  r,  r-  s                r   4check_min_weight_fraction_leaf_with_min_samples_leafr;    sD   
 	3&&rzz2A#Q3A771:LdOM 'bkk!S!6L M%))	
 	1'))//!''),C))//!$C{{3'#LA$56vvl#sC8881(
 
 	
OVV#..0D0D
	
 
% !N. !(bkk!S!6L M%)) 	
 	1'))//!''),C))//!$C{{3'#LA$56vvl#sC888C000(
 
 	
 PVV#..0D0D
	
 
% !Nr   c                     t        | d       y r0  r;  r2  s    r   Btest_min_weight_fraction_leaf_with_min_samples_leaf_on_dense_inputr>     s    8vFr   c                      t        | d|       y r6  r=  r7  s     r   Ctest_min_weight_fraction_leaf_with_min_samples_leaf_on_sparse_inputr@  %  s    
 9l]r   c                    t        j                  d|       \  }}t        dt        j	                               D ]  \  }}t        |   } ||d      } ||dd      } ||dd      } ||d	d      }	|d
f|df|df|	d	ffD ]  \  }
}|
j
                  |k  s!J dj                  |
j
                  |             |
j                  ||       t        |
j                  j                        D ]M  }|
j                  j                  |   t        k7  s%|
j                  j                  |   }|
j                  j                  |   }|
j                  j                  |   }|
j                  j                  |   }|
j                  j                  |   }||z  }|
j                  j                  |   }|
j                  j                  |   }|
j                  j                  |   }||z  }||z   }||z  }|
j                  j                  |   |j                   d   z  }|||z
  z  }||k\  r9J dj                  ||                y )Nd   rW   rV   r  r   r   rV   rO   )r   min_impurity_decreaserV   g-C6?r   gHz>z)Failed, min_impurity_decrease = {0} > {1}z2Failed with {0} expected min_impurity_decrease={1})r   r   r   r5   r  rE  ro   r   ranger   rn   rq   r   rx   weighted_n_node_samplesrp   r   )global_random_seedr^   r_   r   r   r   est1est2est3est4r   expected_decreasenode
imp_parent
wtd_n_nodeleft
wtd_n_leftimp_leftwtd_imp_leftrightwtd_n_right	imp_rightwtd_imp_rightwtd_avg_left_right_impfractional_node_weightactual_decreases                             r   test_min_impurity_decreaser\  /  sf    ''#DVWDAq !(inn6F G!$ NK)TU
 )VW
 )ST

 4L4L6N3K	'
"C" ))->>:AA))+<> GGAqMcii223 99**40I=!$!3!3D!9J!$!B!B4!HJ992248D!$!B!B4!HJ"yy11$7H#-#8LII44T:E"%))"C"CE"JK #		 2 25 9I$/)$;M-:\-I**j8* 		99$?!''!*L + '="%;;'O
 (+<<KRR'):<; 4'
% !Hr   c            
         t         j                         D ]C  \  } }d| v r!t        j                  t        j                  }}n t
        j                  t
        j                  }} |d      }|j                  ||       |j                  ||      }g d}|D ci c]  }|t        |j                  |       }}t        j                  |      }	t        j                  |	      }
t        |
      |j                  k(  sJ |
j                  ||      }||k(  sJ dj                  |              |D ]-  }t!        t        |
j                  |      ||   d| d|         / F y	c c}w )
z8Test pickling preserves Tree properties and performance.
Classifierr   r   )r   rn   capacity	n_classesrq   rp   n_leavesrt   ru   rx   rv   rG  ry   z6Failed to generate same score  after pickling with {0}z"Failed to generate same attribute z after pickling with rl   N)r5   r   r`   r   r   ra   r   r   r   r   pickledumpsloadstype	__class__ro   r%   )r   r   r^   r_   r   r   
attributes	attributefitted_attributeserialized_objectrJ  score2s               r   test_picklerl  w  sN   (0m499dkkqA==(//qA+1		!Q

  GQ
FPIwsyy)44j 	 
 #LL-||-.DzS]]***Aq!VO	QCJJ4P	Q)I

I. +8 Dv	 *M  14
s   Ec                     ddgddgddgddgddgddgddgddgddgddgddgddgg} ddgddgddgddgddgddgddgddgddgddgddgddgg}ddgddgddgddgg}ddgddgddgddgg}t         j                         D ]  \  }} |d      }|j                  | |      j                  |      }t	        ||       |j
                  dk(  sJ |j                  |      }t        |      dk(  sJ |d   j
                  dk(  sJ |d   j
                  d	k(  sJ |j                  |      }	t        |	      dk(  sJ |	d   j
                  dk(  sJ |	d   j
                  d	k(  rJ  t        j                         D ]L  \  }}
 |
d      }|j                  | |      j                  |      }t        ||       |j
                  dk(  rLJ  y )
Nr@   r=   r7   rK   r   r:   r   r6   rK   )r6   r6   )r   r   r   r   r%   r   r   r   r   r   r#   )r^   r_   r   y_truer   r   r   y_hatproba	log_probar   r   s               r   test_multioutputrs    sH    
R	R	R	
A	
A	
A	Q	Q	Q	
B	
B	
B	A  
Q	Q	Q	
A	
A	
A	Q	Q	Q	
A	
A	
A	A bAq6B7QG,A1g1vAwA/F !* 1n!,1%%a(5&){{f$$$!!!$5zQQx~~'''Qx~~'''))!,	9~"""|!!V+++|!!V+++ !2"  )0m+1%%a(E6*{{f$$$	  1r   c                  d   t         j                         D ]  \  } } |d      }|j                  t        t               |j
                  dk(  sJ t        |j                  ddg       t        j                  t        t        j                  t              dz  f      j                  } |d      }|j                  t        |       t        |j
                        dk(  sJ t        |j                        dk(  sJ t        |j
                  ddg       t        |j                  ddgddgg        y )Nr   r   rK   r=   r7   r@   )r   r   r   r^   r_   
n_classes_r%   classes_rr   r   r   r   r   )r   r   r   _ys       r   test_classes_shaperx    s     ) 1n!,1~~"""3<<"a1 YY288A;?+,..!,23>>"a'''3<< A%%%3>>Aq623<<2q'B7);< !2r   c                     t         j                  d d } t         j                  d d }t        d|      }t        j                         D ]=  \  }} |d      }|j                  | ||       t        |j                  |       |       ? y )N}   balancedr   r   r   )	r`   r   r   r"   r   r   r   r#   r   )unbalanced_Xunbalanced_yr   r   r   r   s         r   test_unbalanced_irisr~    sr    99Tc?L;;t$L)*lCM ) 1n!,l-HCKK5|D !2r   c                     t        t        j                         t        j                  t        j
                  g      D ]  \  \  } }} |d      }t        j                  t        j                  |      }t        j                  }t        |j                  ||      j                  |      |       t        j                  t        j                  d|      }t        j                  }t        |j                  ||      j                  |      |       t        j                  t        j                  d|      }t        j                  }t        |j                  ||      j                  |      |       t        j                  t        j                  |      }t        j                  }t        |j                  ||      j                  |      |       t        D ]U  } |t        j                  |      }t        j                  }t        |j                  ||      j                  |      |       W t        D ]U  } |t        j                  |      }t        j                  }t        |j                  ||      j                  |      |       W t        j                  t        j                  d d d   |      }t        j                  d d d   }t        |j                  ||      j                  |      |        y )Nr   r   r  C)orderr  Fr:   )r   r5   r   rr   float64r$  r  r`   r   r   r%   r   r   ascontiguousarrayr-   r,   )r   r   r  r   r^   r_   csr_containerr4  s           r   test_memory_layoutr    s   (/BJJ

3)$}u + JJtyy.KK3771a=003Q7 JJtyy59KK3771a=003Q7 JJtyy59KK3771a=003Q7   %8KK3771a=003Q7 ,Mdiiu5AAswwq!}44Q7; , ,Mdiiu5AAswwq!}44Q7; , JJtyy1~U3KK!3771a=003Q7Q)r   c                  z   t        j                  d      d d t         j                  f   } t        j                  d      }d|d d t        j                  d      }d||dk(  <   t	        d      }|j                  | ||       t        |j                  |       t        j                  d             t        j                  d      d d t         j                  f   } t        j                  d      }d|dd d	|dd d| dddf<   t        j                  d      }d
||d	k(  <   t	        dd      }|j                  | ||       |j                  j                  d   dk(  sJ d||d	k(  <   t	        dd      }|j                  | ||       |j                  j                  d   dk(  sJ t        j                  } t        j                  }t        j                  d| j                   d   d      }t	        d      }|j                  | |   ||          t        j"                  || j                   d         }t	        d      }|j                  | ||       |j                  j$                  t&        j(                  j*                  k7  }t-        |j                  j                  |   |j                  j                  |          y )NrB  rR   2   r   r   r      r7   rK   gRQ?r   g     b@rI   g     H@)	minlength)rr   r   r   r   r   r   r%   r   rj   r   ru   r`   r   r   r%  randintr   r  rq   r	   r  r   r$   )r^   r_   r   r   
duplicatesr   r~   s          r   test_sample_weightr  3  sA    			#q"**}%A
AAcrFGGCLMM!q&
 a
0CGGAqG.s{{1~rwws|4 			#q"**}%A
AAbIAc#JAc#gqjMGGCLM M!q&
 11
=CGGAqG.99q!U***M!q&
 11
=CGGAqG.99q!T))) 			AAQ
C0J
 a
0CGGAjM1Z=)KK
aggajAM!q1DHHQH/yy&&$***>*>>H		H%tzz';';H'Er   c                     t        j                  d      d d t         j                  f   } t        j                  d      }d|d d t	        d      }t         j
                  j                  dd      }t        j                  t              5  |j                  | ||       d d d        t        j                  d      }d}t        j                  t        |	      5  |j                  | ||       d d d        y # 1 sw Y   YxY w# 1 sw Y   y xY w)
NrB  rR   r  r   r   r7   r   z3Singleton.* cannot be considered a valid collectionr   )rr   r   r   r   r   randomr&  r   r   r   r   r   	TypeError)r^   r_   r   r   expected_errs        r   test_sample_weight_invalidr  g  s    
		#q"**}%A
AAcrF
 a
0CIINN3*M	z	"1M2 
# HHQKMIL	y	51M2 
6	5 
#	"
 
6	5s   
C7D7D Dc                    t         |    } |d      }|j                  t        j                  t        j                          |dd      }|j                  t        j                  t        j                         t        |j                  |j                         t        j                  t        j                  t        j                  t        j                  f      j                  } |ddddddddddddgd      }|j                  t        j                  |       t        |j                  |j                          |dd      }|j                  t        j                  |       t        |j                  |j                         t        j                  t        j                  j                        }|t        j                  dk(  xx   d	z  cc<   dd
dd} |d      }|j                  t        j                  t        j                  |        ||d      }|j                  t        j                  t        j                         t        |j                  |j                          |d      }|j                  t        j                  t        j                  |dz          ||d      }|j                  t        j                  t        j                  |       t        |j                  |j                         y )Nr   r   r{  class_weightrV   g       @rM   r  r7   rB  g      Y@rK   )r   r   r`   r   r   r#   r   rr   r   r   r   r   )	r   r   clf1r   
iris_multiclf3clf4r   r  s	            r   test_class_weightsr  y  s    t_N q)DHHTYY$zBDHHTYY$1143L3LM DKKdkkBCEEJ$$$

 D 	HHTYY
#1143L3LMzBDHHTYY
#1143L3LM GGDKK--.M$++"#s*#u-Lq)DHHTYY]3|!DDHHTYY$1143L3LM q)DHHTYY]A%56|!DDHHTYY]31143L3LMr   c                 @   t         |    }t        j                  t        t        j                  t              dz  f      j
                  } |dddgd      }d}t        j                  t        |      5  |j                  t        |       d d d        y # 1 sw Y   y xY w)	NrK   rI   rM   r=   r7   r   r  zBnumber of elements in class_weight should match number of outputs.r   )r   rr   r   r_   r   r   r   r   r   r   r^   )r   r   rw  r   rm   s        r   test_class_weight_errorsr    st     t_N	Arxx{Q'	(	*	*B CC'8&9
JCRG	z	12 
2	1	1s   4BBc                      t        j                  dd      \  } }d}t        j                         D ]:  \  }} |d |dz         j	                  | |      }|j                         |dz   k(  r:J  y NrB  r7   rC  r6   )r   r   )r   make_hastie_10_2r5   r   r   get_n_leavesr^   r_   kr   r   r   s         r   test_max_leaf_nodesr    sk    $$sCDAq	A(0md1q5AEEaK!QU***  1r   c                      t        j                  dd      \  } }d}t        j                         D ]4  \  }} |d|      j	                  | |      }|j                         dk(  r4J  y r  )r   r  r5   r   r   	get_depthr  s         r   test_max_leaf_nodes_max_depthr    s`    $$sCDAq	A(0ma:>>q!D}}!###  1r   c                      dD ]]  } t        t               j                  dgdggddg      j                  |       }d|j                  d   cxk  rdk  rPJ d        J d        y )N)r`  ry   rq   rp   ru   rx   rt   rv   r   r7   rA   r:   z Array points to arbitrary memory)r   r   r   r   flat)attrry   s     r   test_arrays_persistr    sm    	 .044qcA3Z!QHNNPTUUZZ]&Q&J(JJ&J(JJ&	r   c                     t        d      } t        j                  d      }| j                  ddd      }t        j                         D ];  \  }} |d      }|j                  ||       |j                  j                  dk(  r;J  y )Nr   )rB   rY   rK   )rB   r   )	r.   rr   rj   r  r5   r   r   r   r   )rV   r^   r_   r   r   r   s         r   test_only_constant_featuresr    ss    %a(L
AQ5)A(0m+1yy""a'''  1r   c                  r   t        j                  t        j                  g dgt        j                  d      f            } g d}t        j                         D ]\  \  }}d|vs |dd      }|j                  | |       |j                  j                  dk(  sJ |j                  j                  d	k(  r\J  y )
N)r   r   r   r   r   r7   rK   r6   r9   rH      )r6   rG   )r   r   r   r7   r7   rK   rK   rK   r:   r:   r:   	ExtraTreer   r7   r   rK   r9   )
rr   	transposer   rj   r5   r   r   r   r   rn   r^   r_   r   r   r   s        r   ,test_behaviour_constant_feature_after_splitsr    s    

		568IJK	A 	*A(0md"QQ?CGGAqM99&&!+++99''1,,,  1r   c                     t        j                  t        j                  dgdgdgdgg      t        j                  d      g      } t        j                  g d      }t        j                         D ]k  \  }} |dd      }|j                  | |       |j                  j                  dk(  sJ t        |j                  |       t        j                  dd	             m t        j                         D ]k  \  }} |dd      }|j                  | |       |j                  j                  dk(  sJ t        |j                  |       t        j                  d
d	             m y )NrM   rR   )r6   r  )rR   rM   rR   rM   r   r7   r   rn  rI   )r6   )rr   hstackr   rj   r   r   r   r   r   r%   r   r   r   r   r  s        r   (test_with_only_one_non_constant_featuresr    s   
		288cUSEC53%89288I;NOPA
%&A(0m;1yy""a'''3,,Q/1EF	  1  )0m;1yy""a'''3;;q>2774+=>	  1r   c                  &   t        j                  dd      j                  t         j                        j	                  dd      } t               }t        j                  t        d      5  |j                  | g d       d d d        y # 1 sw Y   y xY w)Ng\)c=Hr6   r=   r7   r$  r   )r   r7   r   r7   )
rr   repeatr#  r  reshaper   r   r   r   r   )r^   r   s     r   test_big_inputr    s^    
		(A%%bjj199"a@A
 
"C	z	3<  
4	3	3s   )BBc                  z    ddl m}  t        j                  t              5   |         d d d        y # 1 sw Y   y xY w)Nr   _realloc_test)sklearn.tree._utilsr  r   r   MemoryErrorr  s    r   test_reallocr    s"    1	{	# 
$	#	#s   1:c                     dt        j                  d      z  } t        j                  j	                  dd      }t        j                  j                  ddd      }d| dz   z  }t        d|      }t        j                  t              5  |j                  ||       d d d        d| dz
  z  dz
  }t        d|      }t        j                  t              5  |j                  ||       d d d        y # 1 sw Y   VxY w# 1 sw Y   y xY w)	NrJ   PrB   rK   r   r7   best)splitterr   )structcalcsizerr   r  randnr  r   r   r   	Exceptionr   r  )n_bitsr^   r_   huger   s        r   test_huge_allocationsr    s    %%F
		AA
		!Q#A !D
 &
FC	y	!1 
"
 !q D
 &
FC	{	#1 
$	# 
"	! 
$	#s   C0C<0C9<Dc                     t         |    }t        |   d   }t        |   d   }|dv r|j                  d   dz  }|d | }|d | }t        t        z   t
        z   D ]5  } ||      } |d|      j                  ||      }	 |d|      j                  ||      }
t        |	j                  |
j                  dj                  |              |	j                  |      }| t        v r"|	j                  |      }|	j                  |      }t        t
        z   t        z   D ]t  } ||t        j                        }t!        |
j                  |      |       | t        v s?t!        |
j                  |             t!        |
j                  |             v 8 y )	Nr^   r_   )rb   ra   r   r9   rV   r   5{0} with dense and sparse format gave different treesr  )r5   r"  r   r+   r,   r-   r   r   r   ro   r   r   r   r   rr   r$  r$   )r	   datasetr   r   r^   r_   rW   r)  X_sparserz   r{   y_predy_probay_log_probasparse_container_testX_sparse_tests                   r   check_sparse_inputr  '  st   dOM#A#A ((GGAJ!O	jyMjyM*^;nL#A& qI>BB1aHqI>BB8QOGGGGCJJ4P	
 19ooa(G--a0K%3n%D~%U!1("**MM%aii&>Gy )!//-*H'R)''6 &V% Mr   	tree_typer  )rd   rc   rb   rf   rg   rh   ri   rj   c                 0    |dk(  rdnd }t        | ||       y )Nrb   r:   r  )r  r  r   s      r   test_sparse_inputr  P  s     (dIy'95r   ra   re   c                     t        | |d       y )NrK   r  )r  r  s     r   test_sparse_input_reg_treesr  c  s    
 y'1-r   )rg   rh   ri   rj   c                    t         |    }t        |   d   } ||      }t        |   d   } |ddd      j                  ||      } |ddd      j                  ||      }t        |j                  |j                  dj                  |              t        |j                  |      |j                  |              |ddd	      j                  ||      } |ddd	      j                  ||      }t        |j                  |j                  dj                  |              t        |j                  |      |j                  |              |d|j                  d   dz  
      j                  ||      } |d|j                  d   dz  
      j                  ||      }t        |j                  |j                  dj                  |              t        |j                  |      |j                  |              |dd      j                  ||      } |dd      j                  ||      }t        |j                  |j                  dj                  |              t        |j                  |      |j                  |             y )Nr^   r_   r   r7   rK   )rV   r   r   r  rB   )rV   r   r  )rV   r  r:   r   )	r5   r"  r   r   r   ro   r$   r   r   )	r  r  r4  r   r^   r  r_   rz   r{   s	            r   test_sparse_parametersr  k  s9    i(M#AQH#A 	11BFFq!LA11BFFxQRSA		?FFyQ
 aiilAIIaL9 	11KOOPQSTUA11KOO!	A 		?FFyQ
 aiilAIIaL9 	1x~~a7HA7MNRRSTVWXA1x~~a7HA7MNRR!	A 		?FFyQ
 aiilAIIaL9 	1Q7;;AqAA1Q7;;HaHA		?FFyQ
 aiilAIIaL9r   ztree_type, criterionc                 v   t         |    }t        |   d   } ||      }t        |   d   } |dd|      j                  ||      } |dd|      j                  ||      }	t        |j                  |	j                  dj                  |              t        |	j                  |      |j                  |             y )Nr^   r_   r   r:   rV   r   r   r  )r5   r"  r   r   r   ro   r$   r   )
r  r  r4  r   r   r^   r  r_   rz   r{   s
             r   test_sparse_criteriar    s     i(M#AQH#A1YGKKAqQA1YGKKHVWXA		?FFyQ
 aiilAIIaL9r   zcsc_container,csr_containerc                    t         |    }d}d}|}t        j                  |      }t        d      }g }	g }
d}|g}t	        |      D ]x  }|j                  |d      }|j                  |      d | }|	j                  |       |j                  dd|f      dz
  }|
j                  |       ||z  }|j                  |       z t        j                  |	      j                  t        j                        }	t        j                  |t        j                        }t        j                  t        j                  |
      t        j                        }
 ||
|	|f||f      }|j                         } ||
|	|f||f      }|j                         }|j                  dd|f      }|j                         }|j                   d	k(  j#                         dkD  sJ |j                   d	k(  j#                         dkD  sJ  |d|
      j%                  ||      } |d|
      j%                  ||      }t'        |j(                  |j(                  dj+                  t,                     ||f}t/        ||      D ]  \  }}t1        |j(                  j3                  |      |j(                  j3                  |             t1        |j3                  |      |j3                  |             t1        |j3                  |      |j(                  j3                  |             t1        |j(                  j5                  |      j                         |j(                  j5                  |      j                                t1        |j5                  |      j                         |j5                  |      j                                t1        |j5                  |      j                         |j(                  j5                  |      j                                t1        |j7                  |      |j7                  |             t,        t8        v st1        |j;                  |      |j;                  |              y )Nr:   rB   r   rI   rZ   r7   r  r   rR   r  r  )r5   rr   r   r.   rF  binomialpermutationappendconcatenater#  int32r   r$  toarrayr  copyr   rw   r   r   r   ro   r	   r   r$   r  decision_pathr   r   r   )r  r4  r  r   r   rX   rW   samplesrV   r   r   offsetindptrin_nonzero_i	indices_idata_ir  r^   r  X_testr_   rz   r{   XsX1r  s                              r   test_explicit_sparse_zerosr    s   
 i(MIJ Iii	"G &a(LGDFXF:"++Is; ,,W5l{C	y!&&q#[N&CaGF+f  nnW%,,RXX6GXXfBHH-F88BNN4(

;DdGV4Y
<STHA!	w	:'>M ""$FQ5A "&&(M MMS %%'!+++#%**,q000 	1	:>>q!DA1	:>>xKA		?FFtL -	 B"b/B!!''--"3QWW]]25FG!!''"+qwwr{;!!''"+qww}}R/@A!GG!!"%--/1F1Fr1J1R1R1T	
 	"OOB'')1??2+>+F+F+H	
 	"OOB'')177+@+@+D+L+L+N	
 	"!))B-2?9%aoob&91??2;NO% "r   c                    t         |    }t        j                  d d df   j                         }t        j                  d d df   j	                  d      }t        j
                  }t        j                  t              5   |d      j                  ||       d d d         |d      }|j                  ||       t        j                  t              5  |j                  |g       d d d        y # 1 sw Y   YxY w# 1 sw Y   y xY w)Nr   r  r   )r5   r`   r   r   r  r   r   r   r   r   r   )r   r   r^   X_2dr_   r   s         r   check_raise_error_on_1d_inputr  
  s    dOM		!Q$A99QT?""7+DA	z	"1%))!Q/ 
# Q
'CGGD!	z	"QC 
#	" 
#	"
 
#	"s   >C0C<0C9<Dc                 X    t               5  t        |        d d d        y # 1 sw Y   y xY wN)r'   r  r2  s    r   test_1d_inputr    s    		%d+ 
		s    )r)  c                 Z   t         |    }t        j                  dgdgdgdgdgg      }g d}g d}| ||      } |d      }|j                  |||       |j                  j
                  dk(  sJ  |dd      }|j                  |||       |j                  j
                  dk(  sJ y )	Nr   r7   )r   r   r   r   r7   )r<   r<   r<   r<   r<   r   r   g?)rV   r   )r5   rr   r   r   r   r   )r   r)  r   r^   r_   r   r   s          r    test_min_weight_leaf_split_levelr  !  s     dOM
1#sQC!qc*+AA-M#Q
Q
'CGGAqG.99!###
Q
ECGGAqG.99!###r   c                     t         j                  t        j                  j                  d      }t        |           }|j                  t         t               t        |j                  t               |j                  j                  |             y NFr  X_smallr#  r	   r  r  r5   r   y_smallr%   r  r   )r   	X_small32r   s      r   test_public_apply_all_treesr  5  sX    tzz//e<I
D/
CGGGWsyy)399??9+EFr   r  c                 ,    |t         j                  t        j                  j                  d            }t        |           }|j                  t         t               t        |j                  t               |j                  j                  |             y r   r  )r   r  r  r   s       r   test_public_apply_sparse_treesr  >  s_     gnnTZZ-=-=EnJKI
D/
CGGGWsyy)399??9+EFr   c                      t         j                  } t         j                  }t        dd      j	                  | |      }|j                  | d d       j                         }t        |g dg dg       y )Nr   r7   r  rK   )r7   r7   r   r7   r   r7   )r`   r   r   r   r   r  r  r%   )r^   r_   r   node_indicators       r   test_decision_path_hardcodedr  H  s[    		AA
 a1
=
A
A!Q
GC&&q!u-557N~	9'=>r   c                    t         j                  }t         j                  }|j                  d   }t        |    } |dd      }|j                  ||       |j                  |      }|j                         }|j                  ||j                  j                  fk(  sJ |j                  |      }t        |      D 	
cg c]  \  }	}
||	|
f    }}	}
t        |t        j                  |             |j                  j                  t         k(  }t        t        j"                  ||      t        j                  |             |j%                  d      j'                         }|j                  j(                  |k  sJ y c c}
}	w )Nr   rK   r  r  r7   axis)r`   r   r   r   r5   r   r  r  r   rn   r  	enumerater$   rr   r   rq   r   r  rw   r:  r   )r   r^   r_   rW   r   r   node_indicator_csrr  leavesr  jleave_indicator
all_leavesr   s                 r   test_decision_pathr  P  sB   		AA
IdOM
Q!
4CGGAqM**1-'//1NIsyy/C/C#DDDD YYq\F8A&8IJ8I1~ad+8IOJorwwY/GH ((I5J
~z*BGG),D
 """*..0I99)+++ Ks   <E=c                     t          |t              }}t        |    }t        j                  t
              5   |d      j                  ||       d d d        y # 1 sw Y   y xY wNr   r   )X_multilabely_multilabelr5   r   r   r  r   )r   r  r^   r_   r   s        r   test_no_sparse_y_supportr  n  sH     |4qAdOM	y	!1%))!Q/ 
"	!	!s   AA!c                     t        ddd      } | j                  dgdgdgdgdggg dg d	
       t        | j                  j                  g d       t        | j                  j                  j                  g d       | j                  dgdgdgdgdggg dt        j                  d      
       t        | j                  j                  g d       t        | j                  j                  j                  g d       | j                  dgdgdgdgdggg d       t        | j                  j                  g d       t        | j                  j                  j                  g d       y)aQ	  Check MAE criterion produces correct results on small toy dataset:

    ------------------
    | X | y | weight |
    ------------------
    | 3 | 3 |  0.1   |
    | 5 | 3 |  0.3   |
    | 8 | 4 |  1.0   |
    | 3 | 6 |  0.6   |
    | 5 | 7 |  0.3   |
    ------------------
    |sum wt:|  2.3   |
    ------------------

    Because we are dealing with sample weights, we cannot find the median by
    simply choosing/averaging the centre value(s), instead we consider the
    median where 50% of the cumulative weight is found (in a y sorted data set)
    . Therefore with regards to this test data, the cumulative weight is >= 50%
    when y = 4.  Therefore:
    Median = 4

    For all the samples, we can get the total error by summing:
    Absolute(Median - y) * weight

    I.e., total error = (Absolute(4 - 3) * 0.1)
                      + (Absolute(4 - 3) * 0.3)
                      + (Absolute(4 - 4) * 1.0)
                      + (Absolute(4 - 6) * 0.6)
                      + (Absolute(4 - 7) * 0.3)
                      = 2.5

    Impurity = Total error / total weight
             = 2.5 / 2.3
             = 1.08695652173913
             ------------------

    From this root node, the next best split is between X values of 3 and 5.
    Thus, we have left and right child nodes:

    LEFT                    RIGHT
    ------------------      ------------------
    | X | y | weight |      | X | y | weight |
    ------------------      ------------------
    | 3 | 3 |  0.1   |      | 5 | 3 |  0.3   |
    | 3 | 6 |  0.6   |      | 8 | 4 |  1.0   |
    ------------------      | 5 | 7 |  0.3   |
    |sum wt:|  0.7   |      ------------------
    ------------------      |sum wt:|  1.6   |
                            ------------------

    Impurity is found in the same way:
    Left node Median = 6
    Total error = (Absolute(6 - 3) * 0.1)
                + (Absolute(6 - 6) * 0.6)
                = 0.3

    Left Impurity = Total error / total weight
            = 0.3 / 0.7
            = 0.428571428571429
            -------------------

    Likewise for Right node:
    Right node Median = 4
    Total error = (Absolute(4 - 3) * 0.3)
                + (Absolute(4 - 4) * 1.0)
                + (Absolute(4 - 7) * 0.3)
                = 1.2

    Right Impurity = Total error / total weight
            = 1.2 / 1.6
            = 0.75
            ------
    r   r2   rK   )rV   r   r   r:   r9   rJ   )rH   r  r:   r6   r:   )333333?333333?r   rM   r  )r^   r_   r   )g,d?gܶm۶m?g?)      @g      @r  )ffffff?rL   gUUUUUU?)r6   rS   r  r]   N)
r   r   r   r   rx   r%   ry   r  rr   r   )dt_maes    r   test_maer"  x  s0   T #"21F
 JJ3aS1#s
#
/  
 FLL))+LMv||))..@ JJ1#sQC!qc*oRWWUVZJXv||,,.CDv||))..>
 JJ1#sQC!qc*oJ>v||,,.CDv||))..>r   c                     d} t        j                  dt         j                        }d}d }t        j                  t        j                  |fD ]  }t        j                         D ]G  \  }} || |      } ||      j                         }|\  }	\  }
}}||	k(  sJ | |
k(  sJ t        ||       I t        j                         D ]B  \  }} || |      } ||      j                         }|\  }	\  }
}}||	k(  sJ | |
k(  sJ ||k(  rBJ   y )Nr:   r  rB  c                 R    t        j                  t        j                  |             S r  )rb  rd  rc  )objs    r   _pickle_copyz)test_criterion_copy.<locals>._pickle_copy  s    ||FLL-..r   )
rr   r   intpr  deepcopyr   r   
__reduce__r%   r   )	n_outputsr`  rW   r&  	copy_func_typenamecriteriaresult	typename_
n_outputs_ru  
n_samples_s                r   test_criterion_copyr3    s"    I		!277+II/ ii=	'--/KAx	95Hx(335F5;2I/
Jy(((
***y*5 0 (--/KAx	95Hx(335F5;2I/
Jy(((
***
*** 0 >r   c                    t         j                  j                  d      j                  dd      dz  }t        j                  |j                  d            }|d d d df   }|  | |      }|d d df   }t        d      j                  ||      } |j                  |      }t        t        j                  |j                  j                  t        k(        d         }|j                  |      }t        j                  t        j                  |j                  j                                d   }t#        |      dk(  sJ t#        |      dk(  sJ y )Nr   rB  rG   g*Gr$  r=   r   )rr   r  RandomStater  
nan_to_numr#  r   r   r  setwherer   rq   r   
differenceisfiniteru   r   )	r)  r   r^   r_   r	   terminal_regions	left_leaf
empty_leafinfinite_thresholds	            r   "test_empty_leaf_infinite_thresholdr?    s    99  #))#r2T9D==Y/0DQVA#QQUA a044Q:D!tzz!}BHHTZZ55BCAFGI%%&67J2;;tzz/C/C#D"DEaH!"a'''z?ar   tree_clsc                 b   t         |    } | d   | d   }} |dd      }|j                  ||      }|j                  }|j                  }t	        j
                  t	        j                  |      dk\        sJ t	        j
                  t	        j                  |      dk\        sJ t        ||||       y Nr^   r_   rY   r   rD  r"  cost_complexity_pruning_path
ccp_alphas
impuritiesrr   r   diffassert_pruning_creates_subtreer  r@  r^   r_   r   infopruning_pathrF  s           r   'test_prune_tree_classifier_are_subtreesrL    s    
 wG3<qA
"1
5C++Aq1D??LJ66"'','1,---66"''*%*+++"8Q<@r   c                 b   t         |    } | d   | d   }} |dd      }|j                  ||      }|j                  }|j                  }t	        j
                  t	        j                  |      dk\        sJ t	        j
                  t	        j                  |      dk\        sJ t        ||||       y rB  rC  rI  s           r   'test_prune_tree_regression_are_subtreesrN    s     wG3<qA
"1
5C++Aq1D??LJ66"'','1,---66"''*%*+++"8Q<@r   c                      t        d      } | j                  dgdggddg       t        dd      }|j                  dgdggddg       t        | j                  |j                         y )Nr   r   r7   rB   )rV   	ccp_alpha)r   r   assert_is_subtreer   )r  r   s     r   test_prune_single_node_treerR  .  s`    !q1DHHqcA3Z!Q  "qB?DHHqcA3Z!Q djj$**-r   c                     g }|D ].  } | d|d      j                  ||      }|j                  |       0 t        ||dd        D ]%  \  }}t        |j                  |j                         ' y )NrY   r   )r   rP  rV   r7   )r   r  ziprQ  r   )	estimator_clsr^   r_   rK  
estimatorsrP  r   prev_estnext_ests	            r   rH  rH  :  sp    J!	2QRSWWq
 	#	 " "*jn=((..(..9 >r   c                 >   | j                   |j                   k\  sJ | j                  |j                  k\  sJ | j                  }| j                  }|j                  }|j                  }dg}|r1|j	                         \  }}t        | j                  |   |j                  |          t        | j                  |   |j                  |          t        | j                  |   |j                  |          t        | j                  |   |j                  |          ||   ||   k(  rt        t        |j                  |          nXt        | j                  |   |j                  |          |j                  ||   ||   f       |j                  ||   ||   f       |r0y y )N)r   r   )rn   r   rq   rp   popr$   ry   r#   rx   rv   rG  r   ru   r  )	r	   subtreetree_c_lefttree_c_rightsubtree_c_leftsubtree_c_rightstacktree_node_idxsubtree_node_idxs	            r   rQ  rQ  I  s   ??g00000>>W.....$$K&&L**N,,OHE
*/))+''!JJ}%w}}5E'F	
 	MM-('*:*:;K*L	
 	.0F0FGW0X	
 	((7++,<=	

 *+?O/PP0A0ABR0ST  }-w/@/@AQ/R LL+m4nEU6VWXLLm,o>N.OP3 r   r  r  r  c                 8   t         d   }|d   j                  t        j                  j                  d      }|t        |      }n ||d         }t        j                  |j                  t        j                  j                        |_        t        |j                  |j                  |j                  f      \  |_        |_	        |_
        t        t        j                  t        t        j                  j                              }t        |    |      }|j                  ||       t        |j                  |      |j                  |             t        |j!                  |      j#                         |j!                  |      j#                                y )Nrd   r^   Fr  r  r  )r"  r#  r	   r  r  r&   rr   r   r   r   r  r  r5   r   r%   r   r  todense)r   r  r)  r  r  
X_readonly
y_readonlyr   s           r   "test_apply_path_readonly_all_treesrh  q  s7    {#Gcl!!$**"2"2!?G.w7
%gcl3
((:??$**:J:JK

 &__j00*2C2CD
		
O
 +288G4::CSCS+TUJ
D/8
,CGGJ
#s{{:.G0DE*%--/1B1B71K1S1S1Ur   )r1   r3   r4   c                    t         j                  t         j                  }} ||       }|j                  ||       t	        j
                  |j                  |            t        j                  t	        j
                  |            k(  sJ y )Nr   )	ra   r   r   r   rr   rw   r   r   r   )r   r!   r^   r_   r   s        r   test_balance_propertyrj    s\     ==(//qA

#CGGAqM66#++a.!V]]266!9%====r   seedc           	         ddgddgddgddgddgddgddgddgg}g d}t        d|       }|j                  ||       t        j                  |j	                  |            dk(  sJ t        d|       }|j                  ||       t        j
                  |j	                  |      dkD        sJ d	}t        j                  |dz  dz  d
d||dz  dz  |       \  }}d|d|k  |dk  z  <   t        j                  |      }t        d|       }|j                  ||       t        j
                  |j	                  |      dkD        sJ y )Nr   r7   rK   r:   )r   r   r   r   r7   rK   r:   r6   r1   r   r4   rB   r  r  )effective_ranktail_strengthrW   rX   r   rV   r=   )	r   r   rr   aminr   r   r   make_regressionr   )rk  r^   r_   r   rX   s        r   test_poisson_zero_nodesrq    sN    Q!Q!Q!Q!Q!Q!Q!QHA A  /
MCGGAqM773;;q>"a'''
)$
GCGGAqM66#++a.1$%%% J##!A~* 1n)DAq ArAv!a%
q	A
)$
GCGGAqM66#++a.1$%%%r   c            	      0   t         j                  j                  d      } d\  }}}t        j                  ||z   ||       }| j                  dd|      t        j                  |d      z  }| j                  t        j                  ||z        	      }t        |||| 
      \  }}}	}
t        dd|       }t        dd|       }|j                  ||	       |j                  ||	       t        d      j                  ||	      }||	df||
dffD ]  \  }}}t        ||j                  |            }t        |t        j                  |j                  |      dd             }t        ||j                  |            }|dk(  r
|d|z  k  sJ |d|z  k  rJ  y )Nr   )  rs  rB   rW   rX   rV   r@   rK   )lowhighr[   r   r  )lam)	test_sizerV   r4   rB   )r   r  rV   r1   mean)strategytraintestgV瞯<rI   g      ?)rr   r  r5  r   make_low_rank_matrixuniformr:  r4   r   r   r   r   r
   r   r   clip)r%  n_trainn_testrX   r^   coefr_   X_trainr  r   r   tree_poitree_msedummyval
metric_poi
metric_msemetric_dummys                     r   test_poisson_vs_mser    s   
 ))


#C".GVZ%%F"z	A
 ;;2AJ;7"&&:KKDq4x()A'7	1S($GVWf %rH %!RcH LL'"LL'"F+//AE1FFF3KL	1c*1h.>.>q.AB
*1bggh6F6Fq6I5RV.WX
,Qa0@A &=j 0000D<//// Mr   c                 H   t        |       }t        di |ddi}dD ]  }t        d| z   |d        t        j                  j                  d      }d	\  }}|j                  ||      }t        j                  |d
      |j                  |      z   }|t        j                  |      dz   z  }t        j                  ||d|dz   gd      }	t        j                  ||d|dz   g      }
t        j                  t        |            }d|d|dz   t        di |j                  |||      }t        di |j                  |	|
d      }|j                  j                  |j                  j                  k(  sJ t        |j!                  |      |j!                  |             y)z4Test that the impact of sample_weight is consistent.r   rV   r   )rj   r   DecisionTreeRegressor_rj   )kindr   )rB   r9   r7   r  r   NrK   r    )dictr   r)   rr   r  r5  r&  ry  r   r  r   r   r   r   rn   r   r   )r   tree_paramsr	   r  r%  rW   rX   r^   r_   r  r  sample_weight_1tree1tree2s                 r   6test_decision_tree_regressor_sample_weight_consistencyr    s    +K @;@R@D!'$y0$W	
 "
 ))


"C!IzJ'A
SXXi00ASA 
A.	Q/0q	9B	A.	Q/0	1Bggc!foO()O$i1n%!0K044	1O 5 E "0K044R44PE;;!!U[[%;%;;;; EMM!$emmA&67r   r`  c                 N   d\  }}t        j                  ||||dd      \  }} | dd      j                  ||      } | dd      j                  ||      }t        |j                  |j                  | d	       t        |j                  |      |j                  |             y
)z3Test that criterion=entropy gives same as log_loss.)r  r9   r   r   )r`  rW   rX   r   r   rV   r0   +   r   entropyz> with criterion 'entropy' and 'log_loss' gave different trees.N)r   r   r   r   r   r   r   )r!   r`  rW   rX   r^   r_   tree_log_losstree_entropys           r   'test_criterion_entropy_same_as_log_lossr    s     "Iz'' DAq :B?CCAqIM)"=AA!QGL(PQ
 M))!,l.B.B1.EFr   c                  6   t        j                  d      \  } }t        dd      j                  | |       j	                  | |      }d fd}t        j                   |             }|j	                  | |      }t        j                  ||      sJ y )Nr   r   r:   r  c                     | j                         j                  | j                  j                               j	                         S r  )byteswapviewr  newbyteorderr)  )arrs    r   reduce_ndarrayz8test_different_endianness_pickle.<locals>.reduce_ndarray-  s/    ||~""399#9#9#;<GGIIr   c                     t        j                         } t        j                  |       }t        j
                  j                         |_        |j
                  t        j                  <   |j                         | j                  d       | S Nr   )ioBytesIOrb  Picklercopyregdispatch_tabler  rr   ndarraydumpseek)fpr   r  s     r    get_pickle_non_native_endiannesszJtest_different_endianness_pickle.<locals>.get_pickle_non_native_endianness0  sb    JJLNN1"11668'5$	s	q	r   )	r   r   r   r   r   rb  loadrr   isclose)r^   r_   r   r  new_clf	new_scorer   r  s         @@r    test_different_endianness_pickler  &  s    ''Q7DAq
 a1
=CGGAqMIIaOEJ kk:<=Ga#I::eY'''r   c                  N   t        j                  d      \  } }t        dd      j                  | |       j	                  | |      } G d dt
              fd}t        j                   |             }|j	                  | |      }t        j                  ||      sJ y )Nr   r   r:   r  c                        e Zd Z fdZ xZS )Ptest_different_endianness_joblib_pickle.<locals>.NonNativeEndiannessNumpyPicklerc                     t        |t        j                        r7|j                         j	                  |j
                  j                               }t        | !  |       y r  )	
isinstancerr   r  r  r  r  r  supersave)selfr%  rf  s     r   r  zUtest_different_endianness_joblib_pickle.<locals>.NonNativeEndiannessNumpyPickler.saveG  s@    #rzz*lln))#))*@*@*BCGLr   )__name__
__module____qualname__r  __classcell__)rf  s   @r   NonNativeEndiannessNumpyPicklerr  F  s    	 	r   r  c                      t        j                         }  |       }|j                         | j                  d       | S r  )r  r  r  r  )r  r  r  r   s     r   'get_joblib_pickle_non_native_endiannesszXtest_different_endianness_joblib_pickle.<locals>.get_joblib_pickle_non_native_endiannessL  s3    JJL+A.	s	q	r   )
r   r   r   r   r   r   joblibr  rr   r  )r^   r_   r   r  r  r  r  r   s         @@r   'test_different_endianness_joblib_pickler  ?  s    ''Q7DAq
 a1
=CGGAqMIIaOE,  kkACDGa#I::eY'''r   c                    t         rt        j                  nt        j                  }g d}| j                  j
                  j                         D ci c]  \  }\  }}|| }}}}|D ]  }|||<   	 t        j                  t        |j                               t        |j                               d      }| j                  |d      S c c}}}w )N)
left_childright_childrt   rv   namesformats	same_kindcasting)r*   rr   int64r  r  fieldsr   listr  valuesr#  )node_ndarraynew_dtype_for_indexing_fieldsindexing_field_namesr   r  r,  new_dtype_dict	new_dtypes           r   "get_different_bitness_node_ndarrayr  Y  s    09BHHrxx! V -9,>,>,E,E,K,K,M,M(juae,M   %<t % ~**,-$~?T?T?V:WXI y+>>s   Cc                    | j                   j                  j                         D ci c]  \  }\  }}|| }}}}| j                   j                  j                         D cg c]  \  }}|	 }}}|D cg c]  }d|z   	 }}t	        j                   t        |j                               t        |j                               |d      }| j                  |d      S c c}}}w c c}}w c c}w )NrJ   )r  r  offsetsr  r  )r  r  r   r  rr   r  r  r#  )	r  r   r  r,  r  r  r  shifted_offsetsr  s	            r   $get_different_alignment_node_ndarrayr  k  s    ,8,>,>,E,E,K,K,M,M(juae,M   ,8+=+=+D+D+K+K+MN+M-%v+MGN078fq6zO8.--/0N1134&	
I y+>> O8s   C$C&7C,c                     t         rt        j                  nt        j                  } | j                         \  }\  }}}}|j                  |d      }|j                         }t        |d         |d<   ||||f|fS )Nr  r  nodes)r*   rr   r  r  r)  r#  r  r  )	r	   r  r@  rX   r`  r*  statenew_n_classes	new_states	            r   "reduce_tree_with_different_bitnessr  |  sw    %288I:I$//:K7H0z9i%$$Y$DM

I;Ig<NOIgz=)<iHHr   c                  0   t        j                  d      \  } }t        dd      j                  | |       j	                  | |      }fd}t        j                   |             }|j	                  | |      }|t        j                  |      k(  sJ y )Nr   r   r:   r  c                     t        j                         } t        j                  |       }t        j
                  j                         |_        t        |j
                  t        <   |j                         | j                  d       | S r  )r  r  rb  r  r  r  r  r  
CythonTreer  r  r  r  r   s     r   "pickle_dump_with_different_bitnesszItest_different_bitness_pickle.<locals>.pickle_dump_with_different_bitness  s^    JJLNN1"11668'I$	s	q	r   )	r   r   r   r   r   rb  r  r   r   )r^   r_   r   r  r  r  r   s         @r   test_different_bitness_pickler    s    ''Q7DAq
 a1
=CGGAqMIIaOE kk<>?Ga#IFMM),,,,r   c                  0   t        j                  d      \  } }t        dd      j                  | |       j	                  | |      }fd}t        j                   |             }|j	                  | |      }|t        j                  |      k(  sJ y )Nr   r   r:   r  c                      t        j                         } t        |       }t        j                  j                         |_        t        |j                  t        <   |j                         | j                  d       | S r  )
r  r  r   r  r  r  r  r  r  r  r  s     r   "joblib_dump_with_different_bitnesszPtest_different_bitness_joblib_pickle.<locals>.joblib_dump_with_different_bitness  sY    JJLO"11668'I$	s	q	r   )	r   r   r   r   r   r  r  r   r   )r^   r_   r   r  r  r  r   s         @r   $test_different_bitness_joblib_pickler    s     ''Q7DAq
 a1
=CGGAqMIIaOE kk<>?Ga#IFMM),,,,r   c                  L   t         r#t        j                  t        j                        n"t        j                  t        j                        } t        j                  t        j                        t        j                  t        j                        g}||D cg c]  }|j                          c}z  }t        j                  ddg|       }|D ]  }t        |j                  |      |         t        j                  t        d      5  t        j                  ddgg|       }t        ||        d d d        t        j                  t        d      5  |j                  t        j                        }t        ||        d d d        y c c}w # 1 sw Y   ^xY w# 1 sw Y   y xY w)Nr   r7   r  zWrong dimensions.+n_classesr   zn_classes.+incompatible dtype)r*   rr   r  r  r  r  r   r   r#  r   r   r   r  )expected_dtypeallowed_dtypesdtr`  wrong_dim_n_classeswrong_dtype_n_classess         r   test_check_n_classesr    s)   +4RXXbhh'"((288:LNhhrxx("((288*<=N>B>Rr(>BBN!Q~6I))"-~>  
z)F	G hhAx~F,n= 
H 
z)H	I ) 0 0 <.? 
J	I C 
H	G 
J	Is   F	
'F,FFF#c                     t        j                  t         j                        } d}t        j                  ||       }| | j	                         g}|D ]  }t        |||        t        j                  t        d      5  t        || d       d d d        |d d d d d df   t        j                  |      fD ]>  }t        j                  t        d      5  t        || |j                         d d d        @ t        j                  t        d	      5  t        |j                  t         j                        | |       d d d        y # 1 sw Y   xY w# 1 sw Y   xY w# 1 sw Y   y xY w)
N)r9   r7   rK   r  )r  expected_shapezWrong shape.+value arrayr   )r7   rK   r7   zvalue array.+C-contiguouszvalue array.+incompatible dtype)rr   r  r  rj   r  r    r   r   r   r  r   r#  r$  )r  r  value_ndarrayr  r  problematic_arrs         r   test_check_value_ndarrayr    s.   XXbjj)NNHH^>BM$n&A&A&CDN"^	
 
 
z)C	D.	
 
E
 *!Q(3R5F5F}5UV]]:-HI -.44 JI W 
z)J	K  ,))	
 
L	K 
E	D JI 
L	Ks$   ?E
E,E"
EE	"E+c                     t         } t        j                  d|       }|t        |      t	        |      g}||D cg c]+  }|j                  |j                  j                               - c}z  }|D ]  }t        ||         t        j                  t        d      5  t        j                  d|       }t        ||        d d d        t        j                  t        d      5  |d d d   }t        ||        d d d        |j                  j                  j                         D ci c]  \  }\  }}|| }}}}|j                         }	t        j                  |	d	<   t        j                  t!        |	j#                               t!        |	j%                               d
      }
|j                  |
      }t        j                  t        d      5  t        ||        d d d        |j                         }	t        j&                  |	d<   t        j                  t!        |	j#                               t!        |	j%                               d
      }
|j                  |
      }t        j                  t        d      5  t        ||        d d d        y c c}w # 1 sw Y   xY w# 1 sw Y   xY wc c}}}w # 1 sw Y   xY w# 1 sw Y   y xY w)N)r9   r  )r  zWrong dimensions.+node arrayr   )r9   rK   znode array.+C-contiguousrK   ru   r  znode array.+incompatible dtyper  )r   rr   rj   r  r  r#  r  r  r   r   r   r   r  r   r  r  r  r  r  r  )r  r  valid_node_ndarraysr  problematic_node_ndarrayr   r  r,  
dtype_dictr  r  s              r   test_check_node_ndarrayr    s\   N88D7L 	*<8,\:
 8K8K

399))+,8K  #LH # 
z)G	H#%88F.#I 4^T 
I 
z)C	D#/!#4 4^T 
E 7C6H6H6O6O6U6U6WX6W"2$
$+6WJX  __&N"$((N;~**,-$~?T?T?V:WXI  ,229=	z)I	J4^T 
K  __&N#%::N< ~**,-$~?T?T?V:WXI  ,229=	z)I	J4^T 
K	JM 
I	H 
E	D Y 
K	J 
K	Js;   0J%J&J),J6J= K	J&)J3=K	KSplitterc                 d   t         j                  j                  d      }d}dt        j                  ddgt         j                        }}t        d   ||      } | ||dd|d	
      }t        j                  |      }t        j                  |      }|j                  |k(  sJ t        ||       sJ y	)z&Check that splitters are serializable.r   rB   rK   r:   r  r/   r9   rI   N)monotonic_cst)rr   r  r5  r   r'  r   rb  rc  rd  r   r  )	r  r%  r   r*  r`  r   r  splitter_serializesplitter_backs	            r   test_splitter_serializabler  	  s    
 ))


#CLbhh1vRWW=yIV$Y	:I	<CDQHh/LL!34M%%555mX...r   c                     t        | j                  d            }t        d      }|j                  t        t
               t        j                  ||       t        j                  |d      }t        |j                  |j                  d       y)zhCheck that Trees can be deserialized with read only buffers.

    Non-regression test for gh-25584.
    z
clf.joblibr   r   r)	mmap_modez?The trees of the original and loaded classifiers are not equal.N)strjoinr   r   r  r  r  r  r  r   r   )tmpdirpickle_pathr   
loaded_clfs       r   /test_tree_deserialization_from_read_only_bufferr  1	  sh    
 fkk,/0K
 a
0CGGGW
KK[![C8J		Ir   c                 6   t        j                  ddgddgg      }t        j                  ddg      } | d      j                  ||        | d      }d}t        j                  t
        |      5   |j                  ||       ddd       y# 1 sw Y   yxY w)zhCheck that an error is raised when min_sample_split=1.

    non-regression test for issue gh-25481.
    r   r7   rM   )r  zb'min_samples_split' .* must be an int in the range \[2, inf\) or a float in the range \(0.0, 1.0\]r   N)rr   r   r   r   r   r   )r!   r^   r_   r	   msgs        r   test_min_sample_split_1_errorr  D	  s     	1a&1a&!"A
!QA 	3##Aq) !$D	0  
z	-A 
.	-	-s   2BBc                    t        j                  g dg      j                  }t        j                  g d      }t        dd|       }|j	                  ||       |j                  t         j                  gg      }t        |t        j                  |dd       g       |dd }|dd }t        dd|       }|j	                  ||       |j                  t         j                  gg      }t        |t        j                  |d	d       g       y)
z<Check missing values goes to correct node during predictions)	r   r7   rK   r:   rJ   r  rG      r   )	r   r<   r  r<   r   r   rL   g?g@r   r7   r  r;   Nr=   r8   )	rr   r   r   r   r   r   nanr   ry  )r   r^   r_   dtcr  X_equaly_equals          r   -test_missing_values_on_equal_nodes_no_missingr  Z	  s     	01244A
>?A
R1	
RCGGAqM [[266($FFRWWQrsV_-. fGfG
R1	
RCGGGW [[266($FFRWWWRS\234r   r  r/   c                    d}t        j                  t         j                  gdz  g dz   g      j                  }t        j                  |gdz  dgdz  z   dgdz  z         }t	        dd|       }|j                  ||       t        j                  t         j                  dd	gg      j                  }|j                  |      }t        ||ddg       y
)zITest when missing values are uniquely present in a class among 3 classes.r   r6   )r   r7   rK   r:   rJ   r  rG   r  r7   rK   r   r  r:   r  Nrr   r   r  r   r   r   r   r%   )r   missing_values_classr^   r_   r  r  
y_nan_preds          r   /test_missing_values_best_splitter_three_classesr  t	  s     
266(Q,!;;<=??A
&'!+qcAg5a?@A
 bA
SCGGAqMXX2'(**FV$Jz$8!Q#?@r   c                    t        j                  t         j                  gdz  g dz   g      j                  }t        j                  dgdz  dgdz  z         }t	        dd|       }|j                  ||       t        j                  t         j                  d	t         j                  gg      j                  }|j                  |      }t        |g d
       y)zMissing values spanning only one class at fit-time must make missing
    values at predict-time be classified has belonging to this class.r6   r   r7   rK   r:   r6   r9   r   r7   rH   r   rK   r  r9   )r   r7   r   Nr  r   r^   r_   r  r  r  s         r   )test_missing_values_best_splitter_to_leftr  	  s     	266(Q,!334577A
!qA37"#A
 bA
SCGGAqMXX266*+,..F[[ Fvy)r   c                    t        j                  t         j                  gdz  g dz   g      j                  }t        j                  dgdz  dgdz  z   dgdz  z         }t	        dd|       }|j                  ||       t        j                  t         j                  dd	gg      j                  }|j                  |      }t        |g d
       y)zMissing values and non-missing values sharing one class at fit-time
    must make missing values at predict-time be classified has belonging
    to this class.r6   r  r7   r   rK   r   r  rN   g333333@r
  Nr  r  s         r   *test_missing_values_best_splitter_to_rightr!  	  s    
 	266(Q,!334577A
!qA37"aS1W,-A
 bA
SCGGAqMXXS)*+--F[[ Fvy)r   c                    t        j                  ddddt         j                  ddddt         j                  g
g      j                  }t        j                  d	gdz  dgdz  z         }t	        d
d|       }|j                  ||       t        j                  t         j                  ddgg      j                  }|j                  |      }t        |g d       y)zNCheck behavior of missing value when there is one missing value in each class.r7   rK   r:   r9   rB   rY   rU   r   r   r   r  gffffff@gA@r
  Nr  r  s         r   0test_missing_values_missing_both_classes_has_nanr#  	  s     	1aArvvr2r2rvv>?@BBA
!qA37"#A
 bA
SCGGAqMXXT*+,..F[[ F vy)r   r	   rd  r   c                 j   t        j                  ddddt         j                  ddddt         j                  g
g      j                  }t        j                  d	gdz  dgdz  z         }|  | |      }t	        j
                  t        d      5   |j                  ||       d
d
d
       y
# 1 sw Y   y
xY w)z4Check unsupported configurations for missing values.r7   rK   r:   r9   rB   rY   rU   r   r   NzInput X contains NaNr   )rr   r   r  r   r   r   r   r   )r)  r	   r^   r_   s       r   test_missing_value_errorsr%  	  s     	1aArvvr2r2rvv>?@BBA
!qA37"#A#Q	z)?	@A 
A	@	@s   B))B2c                  J   t         j                  j                         t         j                  }} t        j
                  | ddddf<   t        j
                  | ddddf<   t        dd      }|j                  | |       |j                  |       }|d	k\  j                         sJ y)
z5Smoke test for poisson regression and missing values.Nr9   r   rH   r=   r4   r   r   rR   )
ra   r   r  r   rr   r  r   r   r   r   )r^   r_   r   r  s       r   test_missing_values_poissonr'  	  s    ==qA Acc1fIAcc2gJ
)"
ECGGAqM[[^FcM   r   c                  D    t        j                  | i |\  }}|dkD  }||fS )N   )r   make_friedman1)argskwargsr^   r_   s       r   make_friedman1_classificationr-  	  s-    ""D3F3DAq	BAa4Kr   zmake_data,Treesample_weight_trainr   c                 l   d\  }} | |||      \  }}|j                         }t        j                  j                  |      }	t        j                  ||	j                  ddg|j                  ddg      <   t        |||      \  }
}}}|d	k(  r#t        j                  |
j                  d
         }nd} |d|      }|j                  |
||       |j                  ||      }t        t                |d|            }|j                  |
|       |j                  ||      }||kD  sJ d|d|        y)zFCheck that trees can deal with missing values have decent performance.)r   rB   rt  FTrT   r   r[   r  r   r   r   NrB   r   r   zscore_native_tree=z! should be strictly greater than )r  rr   r  r5  r  choicer   r   r   r   r   r   r   )	make_datar!   r.  rH  rW   rX   r^   r_   	X_missingr%  X_missing_trainX_missing_testr   r   r   native_treescore_native_treetree_with_imputerscore_tree_with_imputers                      r   !test_missing_values_is_resiliencer:  	  sR    &Iz
ASDAq I
))

 2
3CGIvvIcjj%QWWc
jCD7G1#584O^Wf f$ 5 5a 892DEKOOOWMOJ#)).&A%9KL /73/55nfM 	33Y

	>?V>WXY3r   c                  8   t         j                  j                  d      } d}| j                  |df      }| j	                  dd|      }| j                  ddg|d	d
g      }|j                         j                  t              }||    ||<   | j                  |      }t         j                  ||<   ||dddf<   t        |||       \  }}}	}
t        |       j                  ||	      } |j                  ||	      dk\  sJ  |j                  ||
      dk\  sJ y)z@Check the tree learns when only the missing value is predictive.r   r  rB   rZ   rK   )rv  r[   FTgffffff?rO   r0  Nr9   r   g333333?)rr   r  r5  standard_normalr  r1  r  r#  boolr  r   r   r   r   )r%  rW   r^   r_   X_random_masky_maskX_predictiver  r  r   r   r	   s               r    test_missing_value_is_predictiverA  	
  s   
))


"CI)R1AAAI.A JJt}9tJMMVVX__T"F#M22F=&&I&6L66LAadG'713'O$GVWf!s377ID4::gw'4///4::ff%---r   zmake_data, Treec                    t         j                  j                  d      }d\  }} | |||      \  }}t         j                  ||j	                  ddg|j
                  ddg      <   t        j                  |j
                  d         }d	|d
d
d<    |d      }|j                  |||        |d      }	|	j                  |dd
dd
d
f   |dd
d          t        |	j                  |      |j                  |             y
)z=Check sample weight is correctly handled with missing values.r   )r  rB   rt  FTrT   r   r0  rR   NrK   r   r   r7   )
rr   r  r5  r  r1  r   r   r   r   r   )
r2  r!   r%  rW   rX   r^   r_   r   tree_with_swtree_samples_removeds
             r   test_sample_weight_non_uniformrE  "
  s     ))


"C$IzyZcRDAq @BvvAcjj%QWWc
j;< GGAGGAJ'MM#A#Q'LQ7Q/Qqt!tQwZ14a41(003\5I5I!5LMr   c                  F   t        d      j                  t        j                  t        j                        } t        d      j                  t        j                  t        j                        }t        j                  |       }t        j                  |      }||k(  sJ y r  )r   r   r`   r   r   rb  rc  )r  r  pickle1pickle2s       r   test_deterministic_picklerI  ?
  sl     #266tyy$++NE"266tyy$++NEll5!Gll5!Ggr   r^   r9   rH   c                 p   | j                  dd      } t        j                  d      }t        |d      j	                  | |      }t        |      j	                  |j                  dd      |      }t        |j                  j                  dk\        sJ t        |j                  j                  dd |j                  j                  dd        t        j                  |j                  j                  dk(  |j                  j                  dk(  z        }t        |j                  j                  |   d       y)	a'  Check that we properly handle missing values in regression trees using a toy
    dataset.

    The regression targeted by this test was that we were not reinitializing the
    criterion when it comes to the number of missing values. Therefore, the value
    of the critetion (i.e. MSE) was completely wrong.

    This test check that the MSE is null when there is a single sample in the leaf.

    Non-regression test for:
    https://github.com/scikit-learn/scikit-learn/issues/28254
    https://github.com/scikit-learn/scikit-learn/issues/28316
    r=   r7   rH   r   r   NrK   rR   )r  rr   r   r   r   r   r   r   rx   r   flatnonzerorq   rv   )r^   r   r_   r	   tree_ref
leaves_idxs         r   'test_regression_tree_missing_values_toyrN  L
  s    4 	
		"aA
		!A 91EII!QODT{qyyQ/3Htzz""a'(((DJJ''+X^^-D-DRa-HI 		!	!R	'DJJ,E,E,JKJ DJJ''
3S9r   c                  >   t        j                  d      \  } }t        j                  j	                  d      }| j                         }|j                  t        j                  dt        j                        | dddgf   dz  	      j                  t              }t        j                  ||<   t        ||d
      \  }}}}t        j                  g dt        j                        }t        ddd      }	 |	j                  ||   ||          t!        |	j"                  j$                  dk\        sJ t        j&                  |	j"                  j(                  dk(  |	j"                  j*                  dk(  z        }
t-        |	j"                  j$                  |
   d       y)a  Check that we properly handle missing values in clasification trees using a toy
    dataset.

    The test is more involved because we use a case where we detected a regression
    in a random forest. We therefore define the seed and bootstrap indices to detect
    one of the non-frequent regression.

    Here, we check that the impurity is null or positive in the leaves.

    Non-regression test for:
    https://github.com/scikit-learn/scikit-learn/issues/28254
    T)
return_X_yr   )r7   r6   )r   r  NrK   rJ   )nr     r   )prK   Q   '   a   [   &   .      e   rR  Y   R   rB  r   E      rS     I   J   3   /   k      K   n   rY   r   h   9      r   rf  O   #   M   Z   rb  rX  rR  ^   rV     rJ   ]   rq  r`  rm  r  rR  ra  m   rg     rB   rp  rh  r^  \   4   rY   rr  rJ   rJ      r^  rl  r  r  r  r  r   rU   rY  N   r  rs  i   r  r   r`  r  f   rx  rR  rY  r7   r]  rG       rf  rn  j   ro  r   8   rl  re  >   U   rS  rT  P   r_  ?   rH   r  T   r:   r:   L   rw  r  r:   r   iHnr   r   r=   r7   rR   )r   	load_irisrr   r  r5  r  r  r   r  r#  r=  r  r   r   r   r   r   r   rx   rK  rq   rv   r   )r^   r_   r%  r3  maskr  r,  r   r   r	   rM  s              r   +test_classification_tree_missing_values_toyr  v
  sW    .DAq
))


#CI<<
''bhh
/1QV9q=  fTl 	 ffIdO-iLGQ hh  XXG "&zD DHHWWww/0tzz""a'(((		!	!R	'DJJ,E,E,JKJ DJJ''
3S9r   r  )__doc__r  r  r  rb  r  	itertoolsr   r   r  numpyrr   r   joblib.numpy_pickler   numpy.testingr   sklearnr   r   r	   sklearn.dummyr
   sklearn.exceptionsr   sklearn.imputer   sklearn.metricsr   r   r   sklearn.model_selectionr   sklearn.pipeliner   sklearn.random_projectionr   sklearn.treer   r   r   r   sklearn.tree._classesr   r   r   r   sklearn.tree._treer   r   r   r   r   r    r!   r  sklearn.utilsr"   sklearn.utils._testingr#   r$   r%   r&   r'   r(   sklearn.utils.estimator_checksr)   sklearn.utils.fixesr*   r+   r,   r-   sklearn.utils.validationr.   r   REG_CRITERIONSr   r   r  r5   __annotations__updateSPARSE_TREESr   r  r  y_small_regr^   r_   r   r   r  r`   r  r5  r%  r  r   r[   permr   load_diabetesra   load_digitsrb   rV   make_multilabel_classificationr  r  r~  X_sparse_posr  y_randomr  X_sparse_mixrj   r"  r   r   r   markparametrizer  r   r   r   r   r   r   r   r   r   r   r   r   r   r   r  r  r  r.  r3  r8  r;  r>  r@  r\  rl  rs  rx  r~  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  sortedr7  intersectionr  r  r  r  rT  r  r  r  r  r  r  r  r  r  r"  r3  r?  r  rL  rN  rR  rH  rQ  rh  rj  rF  rq  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  r!  r#  r%  r'  r-  r*  r:  rA  rp  r   rE  rI  r  rN  r  )r	   s   0r   <module>r     s     	   $    , ) ) ) ( - ( U U 4 * ;    2 /  K  8%O 5.	 3,	
 &	4  	    	    "((56<94:@>>@74545?@A84544/8 P6 	"XBx"bAq6Aq6Aq6:"X1v1v xiiA
t{{''(IIdO	kk$ "8!!#
x++,d#//$'				
v}}))*kk$d#!!$DXDDbR l
 ###1$'\S  !151$RTJRRT ))$++.mm(//:KKfmm4W-[1$<8$84%H5$84288G$84$N	X	X !1!1!34n5, 6 5,*F*
$ y'89n5J 6 :J y'89,	"0"5	2126	/4	B-r2	 : 
4W 2L>-K<!7H?+DH>FB8
v +1 ,1 ..9W : /W
 &*:
z +G ,G ..9 : /EP0f9%x=(	E*8Z1h3$ +,N ,,N^ +	 ,	+$K$(-?"!*&R l3	6 46
 fS->-K-KI-V&WXZ$=>. ? Y. l3$WX.90: : Y 40:f <E<4493D$<E~	VW
,D,$$)2C,DnU $WX.9: : Y:" l3!3~~#FHP 4HPV    +, ,,
 ++dVn-DE$ F ,$$ +G ,G ..9G : /G? +, ,,: +.90 : ,0a?H+8 +dVn-DE  F $ vc(--/*k:-FFG &<>Q%RSA TA HMMO4&;=O%PQA R 5A	.:%P +fh%78+dVn-D~-UV W 9 ,4 &RS!1!1!34	> 5 T	> q*& +&B'0T n5"8 6"8J "8:M!NOq!f-G . PG,(2(4?$?"I-,-6@$
B1Uh ,o,,.0G0@0G0G0IJ//& !1!1!34 5* &GH5 I52 y&&9:A ;A y&&9:* ;* y&&9:* ;*  y&&9:* ;* +dVn-DE
1(89
 F
! 		 	 "78	&(>? .v? Y @ YF.2 		!	!#89		%	%'=>NN,
  	"&&!RVVQ1-."&&"&&!Q1-.!Q1bffbff-.!Q2661bff-.
 &GH: I
:<,:g& FDs   	{{,	{6{