fitctree

다중클래스분류를위한이진결정결정트리

페이지내내모두

구문

树= fitctree（tbl，responsevarname）

树= FITCTREE（TBL，公式）

树= fitctree(资源描述,Y)

树= fitctree (X, Y)

树= fitctree (＿＿＿、名称、值)

설명

树= fitctree（资源描述，ResponseVarName）은 테이블资源描述에포함된입력변수（예측변수，특징또는특성이라고도함）와Tbl.ResponseVarName에 포함된 출력 변수(응답 변수 또는 레이블)를 기반으로 하여 피팅된 이진 분류 결정 트리를 반환합니다. 반환된 이진 트리는资源描述의의열값에에따라노드를분할분할

树= fitctree（资源描述，公式）는테이블资源描述에포함된입력변수를으로하여피팅된이진분류결정트리를반환합니다。公式는树를피팅하는데사용된资源描述에 포함된 응답 변수와 예측 변수의 부분 집합에 대한 설명 모델입니다.

树= fitctree（资源描述，Y）는테이블资源描述에포함된입력변수와벡터Y에포함된출력변수를기반으로하여피팅된이진분류결정트리를반환합니다。

예제

树= fitctree（X，Y）는행렬X에 포함된 입력 변수와 출력 변수Y를기반으로하여피팅된이진분류결정트리를반환합니다。반환된이진트리는X의의열값에에따라노드를분할분할

예제

树= fitctree（＿＿＿，名称，价值）는 위에 열거된 구문 중 하나를 사용하여 하나 이상의 이름-값 쌍의 인수로 지정된 추가 옵션으로 트리를 피팅합니다. 예를 들어, 범주형 예측 변수에 대한 최상의 분할을 찾거나 교차 검증된 트리를 성장시키거나 검증을 위해 입력 데이터의 일부를 홀드아웃하는 데 사용되는 알고리즘을 지정할 수 있습니다.

예제

모두축소

분류트리성장시키기

라이브스크립트열기

电离层데이터세트를사용하여분류트리를성장시킵니다。

加载电离层tc = fitctree（x，y）

TC = ClassificationTree Recordename：'Y'类别预防icon：[] ClassNames：{'B'G'} ScorEtransform：'无'NumObServations：351属性，方法

트리트리이제어제어

라이브스크립트열기

maxnumsplits.，小叶大小또는MinParentSize이름——값쌍의모수를사용하여트리의깊이를제어할수있습니다。fitctree는는기본적으로으로깊은트리를성장성장모델복잡성또는계산시간을이기위해더얕은트리를시킬수있습니다。

电离层데이터세트를불러옵니다。

加载电离层

분류트리를성장시키는데사용할수있는트리깊이제어값의디폴트는다음과같습니다。

maxnumsplits.에대해n - 1．n은훈련표본크기입니다。
小叶大小에대해1．
MinParentSize에대해10.．

이러한디폴트값은훈련크기크기가클경우깊은트리를시키는경향이있습니다시키는경향경향이있습니다。

디폴트 트리 깊이 제어 값을 사용하여 분류 트리를 훈련시킵니다. 10겹 교차 검증을 사용하여 모델을 교차 검증합니다.

rng (1);%的再现性MdlDefault = fitctree（X，Y，'横向'，“上”）;

트리에적용된분할수를나타내는히스토그램을그립니다。또한，트리중하나를표시합니다。

numBranches = @ (x)和(x.IsBranch);mdldefaultnumpartitions = cellfun(numBranches, mdldefault . training);图;直方图(mdlDefaultNumSplits)

视图(MdlDefault。训练有素的{1},“模式”，“图形”）

평균분할수는약15입니다。

디폴트수의분할을사용하여훈련된분류트리만큼복잡하지(깊)지않은분류트리를원한다고가정하겠습니다。또다른분류트리를훈련시키되,최대분할수를7로설정합니다。이는디폴트분류트리에서계산한평균분할수의약절반에해당합니다。10겹교차검증을사용하여모델을교차검증합니다。

Mdl7=fitctree（X，Y，“MaxNumSplits”7'横向'，“上”); 视图（Mdl7.1}，“模式”，“图形”）

모델의교차검증분류오차를비교합니다。

classErrorDefault = kfoldLoss (MdlDefault)

classErrorDefault = 0.1168

classError7 = kfoldLoss (Mdl7)

ClassError7 = 0.1311.

Mdl7은mdldefault.보다훨씬덜복잡하지만성능은약간더떨어질뿐입니다。

분류트리최적화하기

라이브스크립트열기

이예제에서는fitctree를사용하여자동으로하이퍼파라미터파라미터를최적화하는방법을보여이예제에서는피셔의붓꽃데이터사용용。

피셔의붓꽃데이터를불러옵니다。

加载渔民

분류기의 교차 검증 손실을 최적화하되量의데이터를사용하여物种의응답변수를예측합니다。

X =量;Y =物种;Mdl = fitctree (X, Y,'OptimizeHyperparameters'，“汽车”）

| ====================================================================================== ||磨练|eval |目标|目标|Bestsofar |Bestsofar |minleafsize || | result | | runtime | (observed) | (estim.) | | |======================================================================================| | 1 | Best | 0.066667 | 1.3662 | 0.066667 | 0.066667 | 31 |

| 2 |接受| 0.066667 | 0.3454 | 0.066667 | 0.066667 | 12 |

|3 |最佳| 0.04 | 0.14567 | 0.04 | 0.040003 | 2|

|4 |接受|0.66667 |0.12831 |0.04 |0.15796 |73 |

| 5 |接受| 0.04 | 0.26487 | 0.04 | 0.040009 | 2 |

| 6 |接受| 0.66667 | 0.148 | 0.04 | 0.040012 | 75 |

| 7 |接受| 0.066667 | 0.090839 | 0.04 | 0.040012 | 20 |

|8 |接受|0.04 |0.095666 |0.04 |0.040009 |4 |

|9 |最佳| 0.033333 | 0.09839 | 0.033333 | 0.033351 | 1|

|10 |接受| 0.066667 | 0.075485 | 0.033333 | 0.03335 | 7|

| 11 |接受| 0.04 | 0.10088 | 0.033333 | 0.033349 | 3 |

|12 |接受|0.066667 |0.08178 |0.033333 |0.033348 |26 |

| 13 |接受| 0.046667 | 0.088121 | 0.033333 | 0.033347 | 5 |

|14 |接受| 0.066667 | 0.30009 | 0.033333 | 0.033454 | 15|

|15 |接受| 0.033333 | 0.1827 | 0.033333 | 0.033339 | 1|

| 16 |接受| 0.033333 | 0.24765 | 0.033333 | 0.033337 | 1 |

|17 |接受| 0.033333 | 0.091024 | 0.033333 | 0.033336 | 1|

|18 |接受| 0.33333 | 0.085329 | 0.033333 | 0.033336 | 43|

| 19 |接受| 0.066667 | 0.17471 | 0.033333 | 0.033336 | 9 |

|20 |接受|0.066667 |0.20179 |0.033333 |0.033336 |6 |

|======================================================================================| | Iter | Eval客观客观| | | BestSoFar | BestSoFar | MinLeafSize | | |结果| |运行时| | (estim(观察) .) | | |======================================================================================| | 21日|接受| 0.066667 | 0.092659 |0.033333 | 0.033336 | 17 |

|22 |接受|0.066667 |0.15854 |0.033333 |0.033336 |10 |

|23 |接受| 0.066667 | 0.073741 | 0.033333 | 0.033336 | 36|

| 24 |接受| 0.33333 | 0.16776 | 0.033333 | 0.034142 | 55 |

| 25 |接受| 0.04 | 0.24091 | 0.033333 | 0.034121 | 2 |

|26 |接受|0.04 |0.097961 |0.033333 |0.034087 |3 |

|27 |接受|0.04 |0.074085 |0.033333 |0.034062 |4 |

|28 |接受| 0.066667 | 0.14195 | 0.033333 | 0.034035 | 23|

|29 |接受|0.066667 |0.17074 |0.033333 |0.034005 |8 |

|30 |接受|0.066667 |0.11 |0.033333 |0.03398 |13 |

__________________________________________________________优化完成。30 MaxObjectiveEvaluations达到。总功能评价：30总的经过时间：49.2264秒总目标函数评估时间：5.6413最佳观察到的可行点：MinLeafSize ___________ 1个中观察到的目标函数值= 0.033333估计目标函数值= 0.03398功能评估时间= 0.09839最佳估计可行点（根据模型）：MinLeafSize ___________ 1个估计值的目标函数值= 0.03398估计函数评估时间= 0.14353

Mdl=ClassificationTree ResponseName:'Y'分类预测值：[]类名：{'setosa''versicolor''virginica'}ScoreTransform:'none'NumObservations:150超参数优化结果：[1x1 BayesianOptimization]属性、方法

무편향예측변수중요도추정값

라이브스크립트열기

census1994데이터세트를불러옵니다。연령,노동자계층,교육수준,결혼여부,인종,성별,자본금손익,주당근무시간이주어진경우개인의급여범주를예측하는모델이있다고가정하겠습니다。

加载census1994X=成人数据（：{'年龄'，'工作组'，'education_num'，“婚姻状况”，'种族'，...“性”，“capital_gain”，'capital_loss'，“hours_per_week”，“工资”});

概括를사용하여범주형변수로표현된범주의개수를표시합니다。

摘要（十）

变量：年龄：32561x1双倍值：最低17中位数37最高90工人阶级：32561x1分类值：联邦政府960地方政府2093从未工作过7私人22696自我emp公司1116自我emp不公司2541州政府1298无薪14数字1836教育人数：32561x1双倍值：最低1中位数10最高16婚姻状况：32561x1分类价值观：离婚4443已婚AF配偶23已婚civ配偶14976已婚配偶缺席418未婚10683分居1025丧偶993种族：32561x1分类价值观：美国印第安爱斯基摩人311亚裔太平洋岛民1039黑人3124其他271白人27816性别：32561x1分类价值观：女性10771男性21790资本收益：32561x1双重价值观：最低0中位数0最高99999资本损失：32561x1双倍值：最低0中位数0最高每周4356小时：32561x1双倍值：最低1中位数40最高99工资：32561x1分类值：<=50K 24720>50K 7841

연속형 변수의 수준에 비해 범주형 변수로 표현된 범주가 적기 때문에 표준 运货马车예측 변수 분할 알고리즘이 범주형 변수를 사용하는 대신 연속형 예측 변수를 분할하려 합니다.

전체데이터세트를사용하여분류트리를훈련시킵니다。무편향트리를성장시키려면예측변수분할에곡률검정을사용하도록지정하십시오。데이터에누락된관측값이있기때문에대리분할을사용하도록지정합니다。

Mdl = fitctree (X,“工资”，'PredictorSelection'，“弯曲”，...“代理”，“上”）;

모든예측변수에대한분할인한위험의에대한합구하고이합을가지노드노드개수로나눠서변수중요도값을추정추정예측변수도값값을추정추정막대그래프를사용하여추정값을비교。

imp=预测重要性（Mdl）；图形bar（imp）；头衔(的预测估计的重要性）;ylabel（“估计”）;包含('预测器'); h=gca；h、 XTickLabel=Mdl.predictor名称；h、 XTickLabelRotation=45；h、 滴答计=“没有”；

이경우,资本收益이가장중요중요한예측이고，education_num이 그 다음으로 중요한 예측 변수입니다.

高的형 배열에 대한 분류 트리 최적화하기

라이브스크립트열기

이 예제에서는 高的형 배열을 사용하여 분류 트리의 하이퍼파라미터를 자동으로 최적화하는 방법을 보여줍니다. 표본 데이터 세트airlinesmall.csv는 항공편 데이터로 구성된 테이블 형식 파일을 포함하는 대규모 데이터 세트입니다. 이 예제에서는 데이터를 포함하는 高的형 테이블을 생성하고 이를 사용하여 최적화 절차를 실행합니다.

高的형 배열에 대한 계산을 수행할 때 MATLAB®은 병렬 풀(并行计算工具箱™를 사용할 경우 디폴트 값) 또는 로컬 MATLAB세션을 사용합니다. 并行计算工具箱가 있는 상태에서 로컬 MATLAB세션을 사용하여 예제를 실행하려는 경우Mapreducer.함수를사용하여하여전역실행실행환경을하면하면하면

데이터와 함께 폴더 위치를 참조하는 데이터저장소를 생성합니다. 사용할 변수의 부분 집합을 선택하고,数据存储에서“NA”값을 누락된 데이터로 처리하여 이를南값으로 바꿀 수 있도록 합니다. 데이터저장소에 있는 데이터를 포함하는 高的형 테이블을 생성합니다.

ds=数据存储(“airlinesmall.csv”）;ds。年代electedVariableNames = {“月”，'DayofMonth'，“星期几”，...“DepTime”，“ArrDelay”，“距离”，“DepDelay”};ds.TreatAsMissing =“NA”；tt =高（DS）%高表

使用“本地”配置文件启动并行池（Parpool）连接到并行池（工人数：6）。TT = M×7高大表月DAYOFMONTH DAYOFWEEK DepTime ArrDelay距离DepDelay _____ __________ _________ _______ ________ ________ ________ 10 21 3 642 8 308 12 10 26 1 1021 8 296 1 10 23 5 2055 21 480 20 10 23 5 1332 13 296 12 1022 4 629 4 373 10 -1 28 3 1446 59 308 63 10 8 4 928 3 447 10 -2 10 6 859 11 954 -1：：：：：：：：：：：：：：

늦은항공편에대해真实인논리형변수를정의하여10분이상늦은항공편을확인합니다。이변수는클래스레이블을포함포함。이변수의미리보기에는처음몇행만행만됩니다。

Y = tt。DepDelay > 10%的类标签

Y=M×1高逻辑阵列1 0 1 0 0:：

예측변수데이터에대한高형배열을생성합니다。

X = tt {: 1: end-1}%的预测数据

X = M×6高，双矩阵10 21 3 642 8 308 10 2 23 5 2055 2 23 10 23 5 133213 23 10 23 5 13329 4 3230 23 5 1329 4 3231 10 23 3 1446 59 3030 8 4 928 3 44710 10 10 6 859 11 954 ::::::::::::::::::::::::::::::::::::::::::

X와Y에서누락된데이터를포함하는행을제거합니다。

R = rmmissing([X Y]);%删除缺少条目的数据X = R (:, 1: end-1);Y = R(:,结束);

예측변수를표준화합니다。

Z=zscore（X）；

'OptimizeHyperparameters'이름 - 값값의인수를사용하여하이퍼파라미터파라미터를자동으로최적화화홀드홀드아웃교차검증검증손실을최소화하는최적'minleafsize'값을 구합니다. (“汽车”를 지정하면'minleafsize'가사용됩니다。）재현이가능하도록“expected-improvement-plus”수집함수를사용하고RNG.및塔林를사용하여난수생성기의시드값을설정합니다。결과는워커의개수및高형배열의실행환경에따라다를수있습니다。자세한내용은控制代码运行的位置항목을참조하십시오。

rng (“默认”) tallrng (“默认”) [Mdl,FitInfo,HyperparameterOptimizationResults] = fitctree(Z,Y, x)...'OptimizeHyperparameters'，“汽车”，...'hyperparameteroptimizationOptions'结构('坚持',0.3,...“AcquisitionFunctionName”，“expected-improvement-plus”）））

使用并行池“本地”评估高表达： - 通过3的3号：5.6秒 - 通行证3：在2.1秒内完成 -  3.4秒的第3条：3.4秒评估完成13秒评估高层表达并行池“本地”： - 通过1：在0.73秒的评估中完成，在0.9秒的评估中完成，使用并行池“本地”评估高表达： - 通过1：在1.2秒评估中完成的1.5秒评估表达式使用并行池“本地”： - 通过4的第1条：0.64秒内完成 -  PASS 2的第2条：在1.1秒内完成 -  PASS 3为4：在0.72秒内完成 -  PASS 4为4：在1.1秒评估中完成在5.9秒内完成使用并行池“本地”评估高表达： -  PASS 1，共4个：在0.53秒内完成 -  PASS 2：在0.73秒内完成 -  PASS 3，共4条：0.58秒内完成 - 通过4：在0.8秒的评估中完成3.5秒的评估，使用并行池“本地”评估高表达式： - 通过1中的第1条：在0.54秒内完成 - 通过4个：0.72秒内完成 -  PASS 3，共4分：0.54秒 - 通行证4：在0.79秒内完成，在0.79秒的评估中完成3.5秒评估使用并行游泳池的高表达式。'当地'： - 通过第1条：1/4：0.51秒 - 通行证的第2条：在0.72秒内完成 - 通过4个：0.59秒内完成 - 通过4分：在0.87秒评估中完成3.3秒评估使用并行池“本地”的高表达： - 通过1的4：0.56秒 - 通行证的第2条：0.8秒内完成 -  PASS 3的第3条：在0.54秒内完成 -  PASS 4为4：在0.89秒内完成评估在3.4秒中完成使用并行池“本地”评估高表达式： - 通过4的4：在0.57秒内完成 -  PASS 2，共4个：在0.74秒内完成 -  PASS 3，共4个：在0.58秒内完成4：在1秒评估中完成3.5秒评估，使用并行池“本地”评估高表达： - 通过4个：在0.54秒内完成 - 通过4：在1.3秒内完成 - 第3条：在4：0.68秒内完成 - 通过4项，共有4个：在1.3秒的评估中完成4.3秒的评估，使用并行池“本地”评估高表达式： -  PASS 1，共4个：在0.65秒内完成 - Pass 2 of 4: Completed in 0.73 sec - Pass 3 of 4: Completed in 0.65 sec - Pass 4 of 4: Completed in 1.7 sec Evaluation completed in 4.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.63 sec - Pass 2 of 4: Completed in 0.85 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 2.2 sec Evaluation completed in 4.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.2 sec - Pass 2 of 4: Completed in 0.88 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 3 sec Evaluation completed in 6.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.72 sec - Pass 2 of 4: Completed in 0.96 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 4.2 sec Evaluation completed in 7.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.77 sec - Pass 2 of 4: Completed in 0.95 sec - Pass 3 of 4: Completed in 0.65 sec - Pass 4 of 4: Completed in 4.8 sec Evaluation completed in 7.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.79 sec - Pass 2 of 4: Completed in 1 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 5.1 sec Evaluation completed in 8.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.89 sec - Pass 2 of 4: Completed in 1.1 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 5.8 sec Evaluation completed in 9.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.63 sec - Pass 4 of 4: Completed in 5.2 sec Evaluation completed in 8.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.6 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.74 sec - Pass 4 of 4: Completed in 4.8 sec Evaluation completed in 9.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.68 sec - Pass 4 of 4: Completed in 3.9 sec Evaluation completed in 7.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.6 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.7 sec - Pass 4 of 4: Completed in 3 sec Evaluation completed in 7.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.66 sec - Pass 4 of 4: Completed in 2.5 sec Evaluation completed in 6.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.66 sec - Pass 4 of 4: Completed in 2.2 sec Evaluation completed in 5.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.69 sec - Pass 4 of 4: Completed in 1.9 sec Evaluation completed in 5.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.2 sec - Pass 2 of 4: Completed in 1.4 sec - Pass 3 of 4: Completed in 0.67 sec - Pass 4 of 4: Completed in 1.6 sec Evaluation completed in 5.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.3 sec - Pass 2 of 4: Completed in 1.4 sec - Pass 3 of 4: Completed in 0.65 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 5.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.67 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.2 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.73 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 5.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.65 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 5.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 2.4 sec Evaluation completed in 2.6 sec |======================================================================================| | Iter | Eval | Objective | Objective | BestSoFar | BestSoFar | MinLeafSize | | | result | | runtime | (observed) | (estim.) | | |======================================================================================| | 1 | Best | 0.11572 | 197.12 | 0.11572 | 0.11572 | 10 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.4 sec Evaluation completed in 0.56 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.93 sec Evaluation completed in 1.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 0.84 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 1.4 sec Evaluation completed in 1.6 sec | 2 | Accept | 0.19635 | 10.496 | 0.11572 | 0.12008 | 48298 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.33 sec Evaluation completed in 0.47 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.83 sec Evaluation completed in 0.99 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.68 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 0.74 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.48 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 0.73 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.68 sec - Pass 4 of 4: Completed in 0.77 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.48 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.86 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.5 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.75 sec Evaluation completed in 0.87 sec | 3 | Best | 0.1048 | 44.614 | 0.1048 | 0.11431 | 3166 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.3 sec Evaluation completed in 0.45 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.83 sec Evaluation completed in 0.97 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.99 sec - Pass 2 of 4: Completed in 0.68 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 0.73 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.47 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.82 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 0.89 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.6 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.5 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.63 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.8 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.66 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 4.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.62 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 1.6 sec Evaluation completed in 4.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.6 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 1 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 4.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.8 sec Evaluation completed in 0.94 sec | 4 | Best | 0.10094 | 91.723 | 0.10094 | 0.10574 | 180 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.3 sec Evaluation completed in 0.42 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.93 sec Evaluation completed in 1.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.66 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.83 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 0.88 sec Evaluation completed in 4.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.5 sec - Pass 4 of 4: Completed in 0.98 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.5 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.7 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.73 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.8 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.64 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.57 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.97 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.89 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.6 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 0.85 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.6 sec - Pass 2 of 4: Completed in 0.82 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.79 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 1.3 sec Evaluation completed in 1.4 sec | 5 | Best | 0.10087 | 82.84 | 0.10087 | 0.10085 | 219 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.32 sec Evaluation completed in 0.45 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.87 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.79 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.66 sec - Pass 3 of 4: Completed in 0.5 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.68 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.68 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.86 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 1 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.85 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.6 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.6 sec - Pass 4 of 4: Completed in 0.84 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 0.87 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 0.92 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.77 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.68 sec - Pass 4 of 4: Completed in 0.86 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.77 sec Evaluation completed in 0.93 sec | 6 | Accept | 0.10155 | 61.043 | 0.10087 | 0.10089 | 1089 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.33 sec Evaluation completed in 0.46 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.89 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.8 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.85 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.83 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.87 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 0.98 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.62 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 4.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 2 sec Evaluation completed in 4.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 2.7 sec Evaluation completed in 5.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.64 sec - Pass 2 of 4: Completed in 0.87 sec - Pass 3 of 4: Completed in 1.2 sec - Pass 4 of 4: Completed in 3.7 sec Evaluation completed in 7.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.73 sec - Pass 2 of 4: Completed in 0.92 sec - Pass 3 of 4: Completed in 0.6 sec - Pass 4 of 4: Completed in 4.4 sec Evaluation completed in 7.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.86 sec - Pass 2 of 4: Completed in 1.5 sec - Pass 3 of 4: Completed in 0.64 sec - Pass 4 of 4: Completed in 4.8 sec Evaluation completed in 8.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.9 sec - Pass 2 of 4: Completed in 1.1 sec - Pass 3 of 4: Completed in 0.65 sec - Pass 4 of 4: Completed in 5.2 sec Evaluation completed in 8.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.73 sec - Pass 4 of 4: Completed in 5.6 sec Evaluation completed in 9.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.5 sec - Pass 2 of 4: Completed in 1.6 sec - Pass 3 of 4: Completed in 0.75 sec - Pass 4 of 4: Completed in 5.8 sec Evaluation completed in 10 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.3 sec - Pass 2 of 4: Completed in 1.4 sec - Pass 3 of 4: Completed in 1.2 sec - Pass 4 of 4: Completed in 5.1 sec Evaluation completed in 9.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.4 sec - Pass 2 of 4: Completed in 1.5 sec - Pass 3 of 4: Completed in 0.7 sec - Pass 4 of 4: Completed in 4.1 sec Evaluation completed in 8.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.4 sec - Pass 2 of 4: Completed in 1.6 sec - Pass 3 of 4: Completed in 0.71 sec - Pass 4 of 4: Completed in 3.6 sec Evaluation completed in 7.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.5 sec - Pass 2 of 4: Completed in 1.8 sec - Pass 3 of 4: Completed in 0.74 sec - Pass 4 of 4: Completed in 3.2 sec Evaluation completed in 7.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.4 sec - Pass 2 of 4: Completed in 1.7 sec - Pass 3 of 4: Completed in 0.73 sec - Pass 4 of 4: Completed in 2.8 sec Evaluation completed in 7.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.5 sec - Pass 2 of 4: Completed in 1.7 sec - Pass 3 of 4: Completed in 0.82 sec - Pass 4 of 4: Completed in 2.4 sec Evaluation completed in 7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 2 sec - Pass 2 of 4: Completed in 1.9 sec - Pass 3 of 4: Completed in 0.79 sec - Pass 4 of 4: Completed in 2.3 sec Evaluation completed in 7.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.6 sec - Pass 2 of 4: Completed in 1.8 sec - Pass 3 of 4: Completed in 0.73 sec - Pass 4 of 4: Completed in 2.2 sec Evaluation completed in 6.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.6 sec - Pass 2 of 4: Completed in 1.7 sec - Pass 3 of 4: Completed in 0.79 sec - Pass 4 of 4: Completed in 2.3 sec Evaluation completed in 7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.7 sec - Pass 2 of 4: Completed in 1.9 sec - Pass 3 of 4: Completed in 0.8 sec - Pass 4 of 4: Completed in 1.8 sec Evaluation completed in 6.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.7 sec - Pass 2 of 4: Completed in 1.8 sec - Pass 3 of 4: Completed in 0.77 sec - Pass 4 of 4: Completed in 1.8 sec Evaluation completed in 6.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.4 sec - Pass 2 of 4: Completed in 1.6 sec - Pass 3 of 4: Completed in 0.73 sec - Pass 4 of 4: Completed in 1.8 sec Evaluation completed in 6.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.5 sec - Pass 2 of 4: Completed in 1.7 sec - Pass 3 of 4: Completed in 1.3 sec - Pass 4 of 4: Completed in 1.7 sec Evaluation completed in 6.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.5 sec - Pass 2 of 4: Completed in 1.7 sec - Pass 3 of 4: Completed in 0.73 sec - Pass 4 of 4: Completed in 1.8 sec Evaluation completed in 6.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 1.3 sec Evaluation completed in 1.5 sec | 7 | Accept | 0.13495 | 241.76 | 0.10087 | 0.10089 | 1 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.31 sec Evaluation completed in 0.44 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.87 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.47 sec - Pass 2 of 4: Completed in 0.67 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.74 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.85 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 0.89 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.73 sec - Pass 3 of 4: Completed in 0.6 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.62 sec - Pass 2 of 4: Completed in 0.8 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 1.6 sec Evaluation completed in 4.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.2 sec - Pass 2 of 4: Completed in 1.5 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 1.9 sec Evaluation completed in 6.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.69 sec - Pass 2 of 4: Completed in 0.88 sec - Pass 3 of 4: Completed in 0.75 sec - Pass 4 of 4: Completed in 2.1 sec Evaluation completed in 5.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 2.2 sec Evaluation completed in 4.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.84 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 2.2 sec Evaluation completed in 4.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.68 sec - Pass 2 of 4: Completed in 0.85 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 2.2 sec Evaluation completed in 4.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.68 sec - Pass 2 of 4: Completed in 0.91 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 2.4 sec Evaluation completed in 5.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.92 sec - Pass 2 of 4: Completed in 0.86 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 1.6 sec Evaluation completed in 4.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.69 sec - Pass 2 of 4: Completed in 0.91 sec - Pass 3 of 4: Completed in 0.63 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.67 sec - Pass 2 of 4: Completed in 0.86 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.99 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.2 sec - Pass 2 of 4: Completed in 0.9 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.95 sec Evaluation completed in 4.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.73 sec - Pass 2 of 4: Completed in 0.91 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.91 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.76 sec - Pass 2 of 4: Completed in 0.93 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.91 sec Evaluation completed in 1.1 sec | 8 | Accept | 0.10246 | 115.31 | 0.10087 | 0.10089 | 58 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.34 sec Evaluation completed in 0.49 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.87 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.8 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.48 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.79 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.6 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.73 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.6 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.95 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.94 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.83 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.83 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.6 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.77 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.72 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.6 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 1.3 sec Evaluation completed in 1.4 sec | 9 | Accept | 0.10173 | 77.229 | 0.10087 | 0.10086 | 418 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.32 sec Evaluation completed in 0.46 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.84 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.75 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.68 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.91 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 0.82 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 0.82 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.95 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.73 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 1.7 sec Evaluation completed in 4.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1.7 sec Evaluation completed in 4.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.71 sec - Pass 2 of 4: Completed in 1.7 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 1.7 sec Evaluation completed in 5.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.65 sec - Pass 2 of 4: Completed in 0.83 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.67 sec - Pass 2 of 4: Completed in 0.87 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.82 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.65 sec - Pass 2 of 4: Completed in 0.84 sec - Pass 3 of 4: Completed in 0.62 sec - Pass 4 of 4: Completed in 0.89 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.64 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.88 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.62 sec - Pass 2 of 4: Completed in 0.9 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.86 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.65 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.8 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.77 sec Evaluation completed in 0.89 sec | 10 | Accept | 0.10114 | 94.532 | 0.10087 | 0.10091 | 123 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.86 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.83 sec Evaluation completed in 0.99 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.48 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.8 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.79 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.73 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.85 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.82 sec - Pass 3 of 4: Completed in 0.64 sec - Pass 4 of 4: Completed in 0.94 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.97 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.8 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 0.8 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 1.1 sec Evaluation completed in 4.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.65 sec - Pass 2 of 4: Completed in 0.84 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 1 sec Evaluation completed in 4.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.63 sec - Pass 2 of 4: Completed in 0.84 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.2 sec - Pass 2 of 4: Completed in 0.83 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.81 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.65 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.8 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.77 sec Evaluation completed in 0.89 sec | 11 | Best | 0.1008 | 90.637 | 0.1008 | 0.10088 | 178 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.38 sec Evaluation completed in 0.52 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.88 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.59 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.79 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.93 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.57 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 0.83 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.91 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.85 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 1 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 1.1 sec - Pass 2 of 4: Completed in 0.81 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 3.9 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 1.5 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.82 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 1.4 sec Evaluation completed in 4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.66 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.79 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1.2 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.62 sec - Pass 2 of 4: Completed in 0.85 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 1 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.61 sec - Pass 2 of 4: Completed in 0.86 sec - Pass 3 of 4: Completed in 1.1 sec - Pass 4 of 4: Completed in 0.96 sec Evaluation completed in 4.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.65 sec - Pass 2 of 4: Completed in 0.8 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.86 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.69 sec - Pass 2 of 4: Completed in 0.84 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.83 sec Evaluation completed in 3.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.76 sec Evaluation completed in 0.89 sec | 12 | Accept | 0.1008 | 90.267 | 0.1008 | 0.10086 | 179 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.32 sec Evaluation completed in 0.45 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.9 sec Evaluation completed in 1.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.77 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.49 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 0.77 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.72 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.74 sec - Pass 3 of 4: Completed in 0.51 sec - Pass 4 of 4: Completed in 1.3 sec Evaluation completed in 3.6 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.74 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.83 sec Evaluation completed in 0.97 sec | 13 | Accept | 0.11126 | 32.134 | 0.1008 | 0.10084 | 10251 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.32 sec Evaluation completed in 0.45 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.85 sec Evaluation completed in 0.99 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.75 sec - Pass 3 of 4: Completed in 0.55 sec - Pass 4 of 4: Completed in 0.74 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.7 sec - Pass 3 of 4: Completed in 0.57 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.68 sec - Pass 3 of 4: Completed in 0.53 sec - Pass 4 of 4: Completed in 0.79 sec Evaluation completed in 3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.91 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.86 sec Evaluation completed in 3.7 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 1 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.71 sec - Pass 3 of 4: Completed in 0.64 sec - Pass 4 of 4: Completed in 0.99 sec Evaluation completed in 3.4 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.54 sec - Pass 2 of 4: Completed in 1.2 sec - Pass 3 of 4: Completed in 0.58 sec - Pass 4 of 4: Completed in 0.94 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.51 sec - Pass 2 of 4: Completed in 0.77 sec - Pass 3 of 4: Completed in 0.59 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 0.78 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.9 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.55 sec - Pass 2 of 4: Completed in 0.72 sec - Pass 3 of 4: Completed in 0.52 sec - Pass 4 of 4: Completed in 0.89 sec Evaluation completed in 3.2 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.58 sec - Pass 2 of 4: Completed in 0.76 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.8 sec Evaluation completed in 3.3 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.56 sec - Pass 2 of 4: Completed in 1.3 sec - Pass 3 of 4: Completed in 0.61 sec - Pass 4 of 4: Completed in 0.76 sec Evaluation completed in 3.8 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.83 sec Evaluation completed in 0.97 sec | 14 | Accept | 0.10154 | 66.262 | 0.1008 | 0.10085 | 736 | Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.36 sec Evaluation completed in 0.5 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 1: Completed in 0.89 sec Evaluation completed in 1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.53 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.74 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.5 sec - Pass 2 of 4: Completed in 0.67 sec - Pass 3 of 4: Completed in 0.56 sec - Pass 4 of 4: Completed in 0.78 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': - Pass 1 of 4: Completed in 0.52 sec - Pass 2 of 4: Completed in 0.69 sec - Pass 3 of 4: Completed in 0.54 sec - Pass 4 of 4: Completed in 0.84 sec Evaluation completed in 3.1 sec Evaluating tall expression using the Parallel Pool 'local': Evaluation 0% ...

MDL = CompactClassificationTree ReplactEname：'Y'类分类：[]类法：[0 1] ScorEtransform：'无'属性，方法

FitInfo=结构没有字段。

HyperparameterOptimizationResults = BayesianOptimization属性:ObjectiveFcn: @createObjFcn / tallObjFcn VariableDescriptions:[4×1 optimizableVariable]选项:[1×1 struct] MinObjective: 0.1004 XAtMinObjective:[1×1表]MinEstimatedObjective: 0.1008 XAtMinEstimatedObjective:[1×1表]NumObjectiveEvaluations: 30 TotalElapsedTime:3.0367e+03 NextPoint: [1×1 table] XTrace: [30×1 table] ObjectiveTrace: [30×1 double] ConstraintsTrace: [] UserDataTrace: {30×1 cell} ObjectiveEvaluationTimeTrace: [30×1 double] IterationTimeTrace: [30×1 double] ErrorTrace: [30×1 double]可行性跟踪:[30×1 logical]可行性probabilitytrace: [30×1 double] IndexOfMinimumTrace:[30×1 double] ObjectiveMinimumTrace: [30×1 double] EstimatedObjectiveMinimumTrace: [30×1 double]

입력인수

모두축소

`资源描述`- - - - - -표본데이터
表형

모델을훈련시키는데사용되는표본데이터로,테이블로지정됩니다。资源描述의각행은하나의관측값에대응되고,각열은하나의예측변수에대응됩니다。선택적으로,资源描述은응답변수에대해하나의추가열을포함할수있습니다。문자형벡터로구성된셀형배열이외의셀형배열과다중열변수는허용되지않습니다。

资源描述이응답변수를포함하며资源描述의나머지모든변수를예측변수로사용하려는경우ResponseVarName을사용하여응답변수를지정하십시오。

资源描述이응답변수를포함하며资源描述의의나머지변수중중일부만예측변수사용용하려는公式를사용하여하여공식공식을지정

资源描述이응답변수를포함하지않는경우Y를사용하여응답변수를지정하십시오。응답변수의길이와资源描述의행개수는동일합니다합니다。

데이터형:桌子

`ResponseVarName`- - - - - -응답 변수 이름
`资源描述`에포함된변수이름

응답변수이름으로，资源描述의변수이름으로지정됩니다。

ResponseVarName은문자형벡터나弦나형스칼라로지정합니다。예예를，응답응답Y가待定로 저장된 경우 이를“是的”로지정하십시오。이렇게하지하지않으면모델모델을훈련시킬Y를포함한资源描述의모든열이예측예측로처리。

응답응답는分类형배열，문자형문자형，字符串형배열，논리형벡터또는벡터，문자형문자형구성된배열이어야어야。Y가문자형배열인경우,응답변수의각요소는배열의각행에대응되어야합니다。

一会이름——값쌍의인수를사용하여클래스의순서를지정하는것이좋습니다。

데이터형:字符|一串

`公式`- - - - - -응답변수와예측변수의부분집합에대한설명모델
문자형 벡터|字符串형스칼라

응답변수，그리고예측변수의부분에대한설명모델로，“Y ~ X1 + X2 + X3”형식의 문자형 벡터나 一串형 스칼라로 지정됩니다. 이 형식에서Y는 응답 변수를 나타내고,X1，X2，X3은예측변수를나타냅니다。

资源描述의일부변수를모델훈련에사용할예측변수로지정하려면식을사용하십시오。사용자가식을지정하면资源描述의 변수 중 해당公式에 표시되지 않은 변수는 사용되지 않습니다.

식에 포함되는 변수 이름은资源描述에포함된변수이름（Tbl.Properties.VariableNames）이면서동시동시에유효유효matlab^®식별자여야합니다。

伊斯瓦名称함수를사용용资源描述에포함된변수이름을확인할수있습니다。다음코드는유효한변수이름을갖는각변수에대해논리값1（真的）을반환합니다。

cellfun（@ isvarname，tbl.properties.variablenames）

资源描述에 포함된 변수 이름이 유효하지 않으면matlab.lang.makeValidName함수를사용하여변수이름을변환하십시오。

tbl.properties.variablenames = matlab.lang.makevallname（tbl.properties.variablenames）;

데이터형:字符|一串

`Y`- - - - - -클래스레이블
숫자형 벡터|绝对的형 벡터|논리형벡터|문자형 배열|字符串형배열|문자형벡터로구성된셀형배열

클래스레이블로，숫자형벡터，分类형벡터，논리형벡터，문자형배열，串형배열또는문자형벡터로구성된셀형배열로지정됩니다。Y의각행은이에대응되는X행의분류를나타냅니다。

트리를피팅할때fitctree는Y의南，''(빈문자형벡터),＂＂(빈 一串형),< >失踪，<未定义>값을결측값으로간주합니다。fitctree는피팅할때Y에대해결측값이있는관측값을사용하지않습니다。

숫자형Y의경우，이대신菲特里를 사용하여 회귀 트리를 피팅하는 방안을 고려해 보십시오.

데이터형:单身的|双重的|分类|逻辑|字符|一串|细胞

`X`- - - - - -예측변수데이터
숫자형 행렬

예측 변수 데이터로, 숫자형 행렬로 지정됩니다.X의각행은하나의관측값에대응되고,각열은하나의예측변수에대응됩니다。

fitctree는X의南값을결측값으로간주합니다。fitctree는피팅할때전부결측값인X의관측값을사용하지않습니다。fitctree는일부가결측값인X의관측값을사용하여이러한관측값이유효한값을가지는변수에서분할을。

데이터형:单身的|双重的

이름-값 쌍의 인수

선택적으로名称，价值인수가쉼표로구분되어지정됩니다。여기서姓名은인수이름이고价值는 대응값입니다.姓名은 따옴표 안에 표시해야 합니다.Name1, Value1,…,的家과과이여러개의이름 - 값쌍의인수를어떤순서로든할수있습니다。

예:‘CrossVal’,‘上’,‘MinLeafSize’,40岁40은리프당최소개의관측값을가지는교차검증된분류트리를지정합니다。

참고

교차검증이름 - 값쌍의인수는'OptimizeHyperparameters'이름——값쌍의인수와함께사용할수없습니다。'OptimizeHyperparameters'에에대한교차검증검증을수정'hyperparameteroptimizationOptions'이름 - 값쌍의인수사용해야만해야만。

모델의 파라미터

모두축소

`“分类算法”`- - - - - -최상의 범주형 예측 변수 분할을 위한 알고리즘
`“准确”`|`'PullLeft'`|`主成分分析的`|`'ovabyclass'`

데이터에대한C개의범주와K≥3개클래스를사용하여범주형예측변수에대한최상의분할을구하기위한알고리즘으로,“分类算法”과함께다음값중하나가쉼표로구분되어지정됩니다。

값	설명
`“准确”`	2^C-1- 1개조합모두를고려。
`'PullLeft'`	오른쪽 가지에 있는 모든 C개의 범주로 시작합니다. 각각의 범주를 왼쪽 가지로 이동시키면 나머지 범주 중에서 K개의 클래스에 대해 최소 불순도를 얻을 수 있으므로 이를 고려해 보십시오. 이 시퀀스에서 불순도가 가장 낮은 분할을 선택합니다.
`主成分分析的`	(중심화된 클래스 확률 행렬의) 가중 공분산 행렬에 대한 첫 번째 주성분과 해당 범주에 대한 클래스 확률 벡터 간의 내적을 사용하여 각 범주의 점수를 계산합니다. 점수를 오름차순으로 정렬하고 C-1개 분할을 모두 고려합니다.
`'ovabyclass'`	오른쪽가지에있는모든C개의범주로시작합니다。각클래스마다해당클래스의확률을기준으로범주를정렬합니다。첫번째클래스에대해각범주를순서대로왼쪽가지로이동하고이동할때마다불순도기준을기록해봅니다。나머지클래스에대해이과정을반복합니다。이시퀀스에서최소불순도를갖는분할을선택합니다。

fitctree는알려진개수의클래스와범주형예측변수수준을사용하여각분할에대한최적의알고리즘조합을자동으로선택합니다。K = 2개클래스인경우，fitctree는 항상 완전 탐색을 수행합니다. 특정 알고리즘을 지정하려면“分类算法”이름——값쌍의인수를사용하십시오。

자세한내용은分类树分类预测器的分裂항목을참조하십시오。

예:'AlgorithmForCategorical'， 'PCA'

`“CategoricalPredictors”`- - - - - -범주형예측변수목록
양의 정수로 구성된 벡터|논리형벡터|문자형행렬|字符串형배열|문자형벡터로구성된셀형배열|`'全部'`

범주형예측변수목록으로,“CategoricalPredictors”와함께다음표에나와있는값중하나가쉼표로구분되어지정됩니다。

값	설명
양의 정수로 구성된 벡터	벡터벡터의각요소는범주형변수를포함하는예측변수변수터의（`X`또는`资源描述`)에 대응되는 인덱스 값입니다.
논리형벡터	요소의값이`真的`이면대응대응되는예측변수변수이터의의（`X`또는`资源描述`）이범주형변수임을의미합니다。
문자형행렬	행렬의각행은예측변수의이름입니다。이름은`PredictorNames`의요소와일치해야합니다。이이같게되도록이름뒤에가로추추채웁니다채웁니다채웁니다채웁니다채웁니다채웁니다채웁니다채웁니다채웁니다。
字符串형배열또는문자형벡터로구성된셀형배열	배열의각요소는예측변수의이름입니다。이름은`PredictorNames`의요소와일치해야합니다。
`'全部'`	모든 예측 변수가 범주형 변수입니다.

기본적으로예측변수데이터가테이블（资源描述) 내에 있는 경우, 변수가 논리형 벡터, 순서가 없는 绝对的형 벡터, 문자형 배열, 一串형 배열 또는 문자형 벡터로 구성된 셀형 배열이면fitctree함수는그변수를범주형변수라고가정합니다。예측변수데이터가행렬（X)이면fitctree함수는 모든 예측 변수를 연속형 변수라고 가정합니다. 다른 모든 예측 변수를 범주형 예측 변수로 식별하려면 이러한 변수를“CategoricalPredictors”이름 - 값쌍의인수사용하여지정하십시오。

예:“分类预测因子”，“全部”

데이터형:单身的|双重的|逻辑|字符|一串|细胞

`“类名”`- - - - - -훈련에 사용할 클래스의 이름
分类형형.|문자형 배열|字符串형배열|논리형벡터|숫자형 벡터|문자형벡터로구성된셀형배열

훈련에사용할클래스의이름으로，“类名”와함께直言형배열,문자형배열,字符串형배열,논리형벡터또는숫자형벡터,문자형벡터로구성된셀형배열이쉼표로구분되어지정됩니다。一会는Y와같은데이터형이어야합니다。

一会가문자형배열인경우，각요소는배열의각에대응합니다。

“类名”를 사용하여 다음을 수행할 수 있습니다.

훈련중에클래스를정렬합니다。
입력인수차원또는출력인수차원의순서를지정합니다。이순서는클래스순서와일치합니다。예를들어,“类名”를사용하여成本차원의순서나预测로반환되는분류점수의열순서를지정할수있습니다。
훈련에 사용할 클래스의 일부를 선택합니다. 예를 들어,Y에포함된모든고유한이름의집합이{' a ', ' b ', ' c '}라고가정해해。클래스'一种'와'C'의 관측값만 사용하여 모델을 훈련시키려면'classnames'，{'a'，'c'}를지정하십시오。

一会의디폴트값은Y에포함된모든고유한이름의집합입니다。

예:“类名”,{' b ', ' g '}

데이터형:分类|字符|一串|逻辑|单身的|双重的|细胞

`'成本'`- - - - - -오분류비용
정사각 행렬|구조체

점의오분류비용으로，'成本'와 함께 다음 중 하나가 쉼표로 구분되어 지정됩니다.

정사각각，여기서成本（i，j）가는특정점에대해실제클래스가我인경우이점을j클래스로분류하는비용입니다(즉,행은실제클래스에대응되고,열은예측클래스에대응됨)。成本의의대응행과대응대응열에클래스클래스순서를지정一会이름——값쌍의인수도지정하십시오。
다음두개의필드를갖는구조체年代：Y와같은데이터형의변수로그룹이름을포함하는美国类名와비용행렬을포함하는S.Classificycosts.．

디폴트값은我~ = j인 경우成本(i, j) = 1이고,我= J.인 경우成本(i, j) = 0입니다。

데이터형:单身的|双重的|结构体

`'maxdepth'`- - - - - -최대트리깊이
양의정수

최대트리깊이로，'maxdepth'와함께양의정수가쉼표로구분되어지정됩니다。수준의개수가더적고계산을위해高형배열을통과해야하는횟수가더적은트리를반환하려면이인수의값을지정하십시오。일반적으로fitctree알고리즘은이터전체에대해한번통과，각트리수준마다추가적적한번씩더통과합니다。기본적으로함수는최대트리트리이를설정하지않습니다。

참고

이옵션은高형배열에fitctree를사용하는경우에만적용됩니다。자세한내용은高형배열을참조하십시오。

`“最大分类”`- - - - - -최대범주수준
`10.`(디폴트값)|음이 아닌 스칼라 값

최대범주수준으로，“最大分类”와함께음이아닌스칼라값이쉼표로구분되어지정됩니다。fitctree는예측변수가분할노드에최대maxnum类别개의수준을갖는경우완전탐색알고리즘을사용하여범주형예측변수를분할합니다。그렇지않은경우，fitctree는완전하지않은탐색알고리즘중를사용하여하여최상의범주분할을

무료와이발생할수있으며큰값전달하면하면시간과메모리가증가할수있습니다。

예:'MaxNumCategories'，8

`“MaxNumSplits”`- - - - - -결정분할의최대개수
`大小(X, 1) - 1`(디폴트값)|양의정수

결정분할(또는가지노드)의최대개수로,“MaxNumSplits”와함께양의정수가쉼표로구분되어지정됩니다。fitctree는maxnumsplits.개이하의가지노드노드를분할。분분할동작에에대한내용내용알고리즘을참조하십시오。

예:'maxnumsplits'，5

데이터형:单身的|双重的

`“MergeLeaves”`- - - - - -리프 병합 플래그
`“上”`(디폴트값)|`“关闭”`

리프 병합 플래그로,“MergeLeaves”와 함께“上”또는“关闭”가쉼표로쉼표로구분되어지정지정。

MergeLeaves가“上”인 경우fitctree는다음을수행합니다。

동일한부모노드에서발생하고부모노드와연관된위험보다크거나같은위험값의합을생성하는리프를병합합니다。
가지치기된 하위 트리의 최적의 시퀀스를 추정하되, 분류 트리를 가지치기하지는 않습니다.

그렇지않은경우，fitctree는 리프를 병합하지 않습니다.

예:'mergeleaves'，'关闭'

`'minleafsize'`- - - - - -리프리프노드관측값값의최소
`1`(디폴트값)|양의정수값

리프노드관측값의최소개수로,'minleafsize'와함께양의정수값이쉼표로구분되어지정됩니다。각리프는트리리프당최소小叶大小개의관측값을가집니다。MinParentSize와小叶大小를 모두 지정하는 경우,fitctree는더큰리프를생성하는설정을사용합니다(MinParentSize = MAX（MinParentSize，2 * MinLeafSize））。

예:“MinLeafSize”,3

데이터형:单身的|双重的

`'迷人'`- - - - - -가지노드관측값의최소개수
`10.`(디폴트값)|양의정수값

가지 노드 관측값의 최소 개수로,'迷人'와함께양의정수값이쉼표로구분되어지정됩니다。트리의각가지노드는최소MinParentSize개의관측값을가집니다。MinParentSize와小叶大小를 모두 지정하는 경우,fitctree는더큰리프를생성하는설정을사용합니다(MinParentSize = MAX（MinParentSize，2 * MinLeafSize））。

예:“MinParentSize”,8

데이터형:单身的|双重的

`“麻木”`- - - - - -숫자형예측변수의箱개수
`[]`(비어있음)(디폴트값)|양의정수스칼라

숫자형예측변수의箱，“麻木”와함께양의정수스칼라가쉼표로구분되어지정됩니다。

“麻木”값이비어있으면(디폴트값)예측변수를비닝하지않습니다。
“麻木”값을양의정수스칼라로지정하면모든숫자형예측변수를지정된개수의등확률本으로비닝한다음원래데이터가아닌本인덱스에서트리를성장시킵니다。
- “麻木”값이 예측 변수의 고유한 값의 개수(（u）를 초과하는 경우fitctree함수는예측변수를u개의宾으로비닝합니다。
- fitctree함수는 범주형 예측 변수를 비닝하지 않습니다.

큰훈련데이터세트를사용할경우이비닝옵션은훈련속도를높이지만정확도가떨어질가능성이있습니다。먼저“麻木人”，50岁을사용해본후에정확도와훈련속도에따라“麻木”를변경해볼수있습니다。

훈련된 모델은 箱子경계값을BinEdges속성에 저장합니다.

예:“麻木人”，50岁

데이터형:单身的|双重的

`“NumVariablesToSample”`- - - - - -각분할에대해임의로선택할예측변수의개수
`'全部'`(디폴트값)|양의정수값

각분할에대해임의로선택할예측변수의개수로,“NumVariablesToSample”과함께양의정수값이쉼표로구분되어지정됩니다。또는'全部'을을하여사용가능한모든예측변수사용할수도있습니다。

훈련데이터가다수의예측변수를포함하며예측변수의중요도를분석하려는경우에는“NumVariablesToSample”을'全部'로지정하십시오。이렇게하지않으면일부예측변수가선택되지않아서이러한변수의중요도가과소평가될수있습니다。

임의선택을재현하려면RNG.를사용하여하여난수난수생성기의시드값을설정'可重复'，真实를 지정해야 합니다.

예:“NumVariablesToSample”,3

데이터형:字符|一串|单身的|双重的

`“PredictorNames”`- - - - - -예측 변수 이름
고유한이름으로으로으로구성구성된형배열|고유한 문자형 벡터로 구성된 셀형 배열

예측 변수 이름으로,“PredictorNames”와함께고유한이름으로구성된字符串형배열또는고유문자형문자형구성셀형배열이쉼표로구분배열배열됩니다됩니다됩니다지정됩니다。“PredictorNames”의기능은훈련데이터를어떤방식으로제공하느냐에따라달라집니다。

X와Y를제공하는경우，“PredictorNames”를사용하여X의예측변수에이름을할당할수。
- PredictorNames의이름의순서는X의 열 순서와 일치해야 합니다. 즉,PredictorNames {1}은X (: 1)의이름이고,PredictorNames {2}는X (:, 2)의 이름이 되는 식입니다. 또한,尺寸（X，2）와numel（预测名称）는같아야합니다。
- 기본적으로PredictorNames는{'x1'，'x2'，...}입니다。
资源描述을제공하는경우，“PredictorNames”를사용하여훈련에사용할예측변수를선택할할수있습니다。즉，fitctree는PredictorNames의예측변수와이에대한응답변수만을훈련중에사용합니다。
- PredictorNames는Tbl.Properties.VariableNames의의집합이어야하므로응답변수의이름은은포함할수수름름포함할수
- 기본적으로，PredictorNames는 모든 예측 변수의 이름을 포함합니다.
- “PredictorNames”와公式중 하나만 사용하여 훈련에 사용할 예측 변수를 지정하는 것이 좋습니다.

예:PredictorNames,{‘SepalLength’,‘SepalWidth’,‘PetalLength’,‘PetalWidth}

데이터형:一串|细胞

`'PredictorSelection'`- - - - - -최상의최상의분할예측예측변수를선택하는사용용되는
`“allsplits”`(디폴트값)|`“弯曲”`|`“interaction-curvature”`

각 노드에서 최상의 분할 예측 변수를 선택하는 데 사용되는 알고리즘으로,'PredictorSelection'과함께다음표에나와있는값이쉼표로구분되어지정됩니다。

값	설명
`“allsplits”`	표준车 - 모든예측변수의모든가능한분할에대해분할조건이득을최대화하는분할예측변수를선택합니다［1］．
`“弯曲”`	곡률 검정- 각예측변수와응답변수간독립성에대한카이제곱검정의P-값을최소화하는분할예측변수를선택합니다［4］．훈련속도는표준购物车와비슷합니다。
`“interaction-curvature”`	상호작용검정— 각 예측 변수와 응답 변수 간 독립성에 대한 카이제곱 검정의 p-값을 최소화하고 각 예측 변수 쌍과 응답 변수 간의 독립성에 대한 카이제곱 검정의 p-값을 최소화하는 분할 예측 변수를 선택합니다［3］. 훈련 속도는 표준 运货马车보다 느릴 수 있습니다.

“弯曲”및“interaction-curvature”의경우，모든검정에서0.05보다큰p-값이생성생성fitctree가 노드 분할을 중지합니다.

팁

표준购物车는는적은고유값을포함하는분할예측（예：범주형변수）보다많은고유값포함포함할할변수（예：연속형연속형）를를하는경향이있습니다［4］. 다음 중 하나에 해당되는 경우 곡률 검정이나 상호 작용 검정을 지정해 보십시오.
- 다른예측변수보다상대적으로더적은수의고유값을갖는예측변수가있는경우（예를들어，예측변수데이터세트가이종인경우）。
- 예측 변수 중요도 분석이 목표인 경우. 예측 변수 중요도 추정에 대한 자세한 내용은predictorImportance및特征选择简介항목을참조하십시오。
표준购物车를사용하여하여성장된트리는예측변수상호에민감하지하지또한，이러한트리는관련없는예측가많이있을경우상호작용검정을때때중요한변수를할가능성이더더。따라서예측변수상호작용을을고려，관련관련변수가많이있을있을경우중요한변수를식별상호작용작용검정을지정해야해야［3］．
예측 속도는'PredictorSelection'값의 영향을 받지 않습니다.

fitctree가분할예측변수를선택하는방법에대한자세한내용은노드분할규칙및选择分割预测器选择技术항목을참조하십시오。

예:'预测互联'，'曲率'

`“之前”`- - - - - -사전확률
`“经验的”`(디폴트값)|`“制服”`|스칼라 값으로 구성된 벡터|구조체

각클래스에대한사전확률로,“之前”와 함께 다음 중 하나가 쉼표로 구분되어 지정됩니다.

문자형벡터또는字符串형스칼라。
- “经验的”은Y또는资源描述에 포함된 응답 변수의 클래스 빈도에서 클래스 확률을 확인합니다. 관측값 가중치를 전달하면fitctree는이가중치를사용하여클래스확률을계산합니다。
- “制服”은클래스확률을모두같은값으로설정합니다。
벡터(각 클래스마다 하나의 스칼라 값 구성).“之前”의의대응요소에에대한클래스순서를하려면一会이름——값쌍의인수도지정하십시오。
다음과다음과같은두개개의필드갖는갖는年代입니다。
- 美国类名는 클래스 이름을 변수로 포함하며Y또는资源描述응답 변수와 동일한 데이터형을 가집니다.
- S.ClassProbs는는대응되는확률확률로구성벡터를포함포함된벡터를포함

“重量”및“之前”값을둘다설정하면fitctree함수는각클래스의가중치의총합이각클래스의사전확률값이되도록가중치를정규화합니다。

예:“优先”，“制服”

데이터형:字符|一串|单身的|双重的|结构体

`'修剪'`- - - - - -가지치기된하위트리에대한최적의시퀀스를추정하는플래그
`“上”`(디폴트값)|`“关闭”`

가지치기된하위트리에대한최적의시퀀스를추정하는플래그로，'修剪'과함께“上”또는“关闭”가쉼표로쉼표로구분되어지정지정。

修剪이“上”인 경우fitctree는분류트리를가지치기하지않고성장，가지지된된하위트리에에최적의시퀀스추정추정추정그렇지않은경우，fitctree는 가지치기된 하위 트리에 대한 최적의 시퀀스를 추정하지 않고 분류 트리를 성장시킵니다.

훈련된ClassificationTree모델을가지치기하려면이를修剪에에하십시오。

예:“删除”,“关闭”

`“PruneCriterion”`- - - - - -가지치기치기.
`“错误”`(디폴트값)|`“杂质”`

가지치기 기준으로,“PruneCriterion”과함께“错误”또는“杂质”가쉼표로쉼표로구분되어지정지정。

“杂质”를지정할경우fitctree함수는“拆分标准”이름-값 쌍의 인수로 지정된 불순도 측정값을 사용합니다.

자세한내용은불순도및노드오차항목을참조하십시오。

예:'Prunecrerion'，'杂质'

`“可复制”`- - - - - -재현성을적용하기위한플래그
`错误的`(논리값`0`）(디폴트값)|`真的`(논리값`1`）

，“可复制”과함께错误的또는真的가쉼표로쉼표로구분되어지정지정。

“NumVariablesToSample”이'全部'이아니면각각분할에에대해임의로예측변수를선택임의선택을재현하려면'可重复'，真实를지정하고RNG.를 사용하여 난수 생성기의 시드값을 설정해야 합니다.“可复制”을真的로설정하면훈련속도가느려질수있습니다。

예:'可重复'，真实

데이터형:逻辑

`'responsebame'`- - - - - -응답 변수 이름
`“是的”`(디폴트값)|문자형 벡터|字符串형스칼라

응답변수이름으로，'responsebame'과과응답변수이름을나타내는문자형벡터또는형스칼라가쉼표로쉼표로되어지정됩니다。

이이름——값쌍은ResponseVarName또는公式입력인수를사용하는경우유효하지않습니다。

예:'responsebame'，'iristype'

데이터형:字符|一串

`“分数变换”`- - - - - -점수변환방식
`“没有”`(디폴트값)|`'doublelogit'`|`“因弗罗吉特”`|`'ismax'`|`分对数的`|함수핸들|...

점수변환방식으로，“分数变换”과함께문자형벡터,弦형스칼라또는함수핸들이쉼표로구분되어지정됩니다。

다음 표에는 사용 가능한 문자형 벡터와 一串형 스칼라가 요약되어 있습니다.

값	설명
`'doublelogit'`	1 /（1 + e^–2倍）
`“因弗罗吉特”`	对数（x/（1-x））
`'ismax'`	최대최대점수를갖는클래스의를를를를설정하고를모든모든클래스의점수0으로으로설정합니다
`分对数的`	1 /（1 + e^–x）
`“没有”`또는`'身份'`	x(변환없음)
`“标志”`	x <0의경우-1 x = 0의경우0 x > 0의경우1
`“对称”`	2x - 1
`“symmetricismax”`	최대최대점수를갖는클래스클래스점수를를를를설정，다른모든클래스의를를-1로설정합니다。
`“symmetriclogit”`	2 /（1 + e^–x) - 1

MATLAB함수나 사용자가 직접 정의하는 함수의 경우, 이에 대한 함수 핸들을 점수 변환에 사용하십시오. 함수 핸들은 행렬(원래 점수)을 받아 동일한 크기의 행렬(변환된 점수)을 반환합니다.

예:'ScoreTransform'， '分对数'

데이터형:字符|一串|function_handle

`“拆分标准”`- - - - - -분할기준
`'GDI'`(디폴트값)|`“两个”`|`“越轨”`

분할기준으로,“拆分标准”과함께'GDI'(지니다양성지수)“两个”（투잉규칙）또는“越轨”(최대 이탈도 감소, 교차 엔트로피라고도 함)가 쉼표로 구분되어 지정됩니다.

자세한내용은불순도및노드오차항목을참조하십시오。

예:“分裂标准”，“偏差”

`“代理”`- - - - - -대리결정분할플래그
`“关闭”`(디폴트값)|`“上”`|`'全部'`|양의정수값

대리결정분할플래그로,“代理”와 함께“上”，“关闭”，'全部'또는양의정수값이쉼표로구분되어지정됩니다。

“上”으로 설정된 경우,fitctree는각가지노드노드에서최대최대최대최대최대의을을을
'全部'로설정된경우，fitctree는각가지노드에서모든대리분할을찾습니다。'全部'설정을사용하면시간과메모리가상당히소모될수있습니다。
양의정수값으로설정된된경우，fitctree는각가지노드노드에서최대지정된정수값만큼대리분할을을

대리분할을사용하면결측값이있는데이터에대한예측의정확도를높일수있습니다。이설정을사용하면예측변수간예측연관성측도도계산할수있습니다。자세한내용은노드분할규칙항목을참조하십시오。

예:“代孕”，“上”

데이터형:单身的|双重的|字符|一串

`“重量”`- - - - - -관측값 가중치
`（大小（x，1），1）`(디폴트값)|스칼라 값으로 구성된 벡터|`资源描述`에포함된변수이름

관측값가중치로，“重量”와와함께스칼라값값으로구성벡터벡터资源描述에포함된변수의이름이쉼표로구분되어지정됩니다。소프트웨어는X또는资源描述의각행에있는관측값에이에대응하는权重의의을함께사용가중치를적용합니다。权重의 크기는X또는资源描述의행개수와일치합니다합니다。

입력 데이터를 테이블资源描述로지정하는경우，权重는资源描述에서숫자형벡터를포함하는변수의이름일수있습니다。이경우,权重를 문자형 벡터 또는 一串형 스칼라로 지정해야 합니다. 예를 들어, 가중 벡터W가待定로 저장된 경우, 이를'W'로지정하십시오。이렇게하지하지않으면모델모델을훈련시킬W를포함한资源描述의모든열이예측예측로처리。

fitctree는각클래스의가중치의총합이각클래스의사전확률값이되도록가중치를정규화합니다。

데이터형:单身的|双重的|字符|一串

교차검증옵션

모두축소

`'横向'`- - - - - -교차교차검증된결정결정트리를시키는시키는
`“关闭”`(디폴트값)|`“上”`

교차검증된결정트리를성장시키는플래그로,'横向'과함께“上”또는“关闭”가쉼표로쉼표로구분되어지정지정。

“上”인 경우,fitctree는10겹교차검증된결정트리를성장시킵니다。“KFold”，'坚持'，“Leaveout”또는“CVPartition”이름——값쌍의인수중하나를사용하여이교차검증설정을재정의할수있습니다。교차검증된트리를생성할때한번에이네개인수중하나씩만사용할수있습니다。

또는,crossval메서드를 사용하여 나중에树를교차검증하십시오。

예:'横向'，'开'

`“CVPartition”`- - - - - -교차 검증된 트리에 사용할 분할
`cvpartition`객체

교차 검증된 트리에 사용할 분할로,“CVPartition”과함께cvpartition을사용하여생성된객체객체가쉼표로구분되어지정됩니다。

“CVPartition”을사용하는하는“KFold”，'坚持'또는“Leaveout”이름-값 쌍의 인수를 사용할 수 없습니다.

`'坚持'`- - - - - -홀드아웃검증에사용할데이터의비율
`0`(디폴트값)|`[0,1]`범위의스칼라값

홀드아웃검증에사용할데이터의비율로,'坚持'과함께범위[0,1]내스칼라값이쉼표로구분되어지정됩니다。홀드아웃검증은데이터에서지정된비율만큼을검증하고나머지데이터는훈련에사용합니다。

'坚持'을사용하는하는“CVPartition”，“KFold”또는“Leaveout”이름-값 쌍의 인수를 사용할 수 없습니다.

예:“坚持”,0.1

데이터형:单身的|双重的

`“KFold”`- - - - - -겹의개수
`10.`(디폴트값)|1보다큰양의정수값

교차검증된분류기에사용할겹의개수로,“KFold”와 함께 1.보다 큰 양의 정수 값이 쉼표로 구분되어 지정됩니다. 지정할 경우(즉,KFold, k），소프트웨어가다음과같이동작동작。

데이터를 K개 세트로 임의로 분할합니다.
각세트마다해당세트를검증데이터로남겨두고나머지k - 1개의세트를사용하여모델을。
교차검증된모델의训练有素的속성에k×1셀형벡터의셀로k개의 훈련된 간소 모델을 저장합니다.

교차교차검증된모델을을하려면다음다음다음다음의의옵션하나만사용용할수CVPartition.，坚持，kfold.또는忽略．

예:'kfold'，8

데이터형:单身的|双重的

`“Leaveout”`- - - - - -리브-원아웃(分析)교차검증플래그
`“关闭”`(디폴트값)|`“上”`

리브-원아웃교차검증플래그로,“Leaveout”과함께“上”이나“关闭”가 쉼표로 구분되어 지정됩니다. 리브-원-아웃 교차 검증을 사용하려면“上”을지정하십시오。

“Leaveout”을사용하는하는“CVPartition”，'坚持'또는“KFold”이름-값 쌍의 인수를 사용할 수 없습니다.

예:'留下'，'开'

하이퍼파라미터파라미터최적화

모두축소

`'OptimizeHyperparameters'`- - - - - -최적화할모수
`“没有”`(디폴트값)|`“汽车”`|`'全部'`|적합한모수이름으로으로구성된된형형배열또는셀형배열|`optimizableVariable`객체로 구성된 벡터

최적화할모수로,'OptimizeHyperparameters'와함께다음값중하나가쉼표로구분되어지정됩니다。

“没有”- 최적화하지않습니다。
“汽车”- - - - - -{'MinLeafSize'}를 사용합니다.
'全部'- 모든적합한모수를최적화합니다。
적합한모수이름으로으로구성된된형형배열또는셀형배열
optimizableVariable객체로 구성된 벡터. 일반적으로hyperparameters의출력값임

최적화는모수를변경하여fitctree에에대한교차검증（오차）을을화하려고합니다。이와는는다른맥락의의교차검증손실에대한자세한내용내용分类损失항목을참조하십시오。교차검증유형과최적화의기타측면을제어하려면HyperParameterOptimizationOptions.이름——값쌍을사용하십시오。

참고

'OptimizeHyperparameters'값은 다른 이름-값 쌍의 인수를 사용하여 설정하는 모든 값을 재정의합니다. 예를 들어,'OptimizeHyperparameters'를“汽车”로설정하면“汽车”값이적용됩니다。

fitctree에에대한적합한한모수는다음과다음과

maxnumsplits.- - - - - -fitctree가기본적으로범위[1，max（2，numobervations-1）]에서로그스케일링된정수중에서탐색을수행합니다。
小叶大小- - - - - -fitctree가기본적으로범위[1,马克斯(2楼(NumObservations / 2)))에서로그스케일링된정수중에서탐색을수행합니다。
SplitCriterion——두클래스에대해fitctree가'GDI'및“越轨”중에서탐색을수행합니다。세개이상의의의에에대해서대해서fitctree가“两个”에대해서도탐색을수행합니다。
NumVariablesToSample- - - - - -fitctree는이하이퍼파라미터에대해최적화수행하지않습니다。NumVariablesToSample을 모수 이름으로 전달하면fitctree는 전체 개수의 예측 변수를 사용합니다. 그러나,fitcensemble.은 이 하이퍼파라미터에 대해 최적화를 수행합니다.

디폴트가아닌값을가지는optimizableVariable객체로 구성된 벡터를 전달하여 디폴트가 아닌 모수를 설정합니다. 예를 들면 다음과 같습니다.

加载渔民params = hyperparameters ('fitctree'，MEAS，物种）;PARAMS（1）.Range = [1,30];

参数를优化超参数의 값으로 전달합니다.

이파라미터에따라플롯플롯플롯하하하하되고하플롯플롯되고플롯플롯플롯되고플롯플롯하플롯하플롯플롯플롯플롯플롯플롯에에최적화최적화와플롯에대해대해목적함수는회귀경우日志（1 +交叉验证丢失）분이고,류의경우오분류율입니다。반복표시를제어하려면'hyperparameteroptimizationOptions'이름 - 값값쌍의인수에에verb필드를 설정하십시오. 플롯을 제어하려면'hyperparameteroptimizationOptions'이름 - 값값쌍의인수에에ShowPlots필드를설정하십시오。

예제는분류트리최적화하기항목을참조하십시오。

예:“汽车”

`'hyperparameteroptimizationOptions'`- - - - - -최적화에 사용할 옵션
구조체

최적화에사용할옵션으로，'hyperparameteroptimizationOptions'와함께가쉼표로구분되어지정됩니다。이인수는优化超参数이름-값 쌍의 인수의 효과를 수정합니다. 이 구조체에 포함된 모든 필드는 선택 사항입니다.

필드 이름	값	디폴트값
`优化器`	`“bayesopt”`- 베이즈최적최적화사용합니다。내부적으로이설정은`Bayesopt.`를를합니다。 `'gridsearch'`— 차원당`NumGridDivisions`개값으로그리드탐색을수행합니다。 `“随机搜索”`- - - - - -`MaxObjectiveEvaluations`개점중에서무작위로탐색합니다。 `'gridsearch'`는 그리드에서 균등한 비복원추출을 사용하여 무작위 순서로 탐색을 수행합니다. 최적화를 수행한 후, 명령`sortrows（Mdl.HyperparameterOptimizationResults）`를사용의그리드순그리드순으로된된이블테얻을수있습니다。	`“bayesopt”`
`收集功能名称`	`“expected-improvement-per-second-plus”` `“预期改善”` `“expected-improvement-plus”` `'预期 - 每秒改善'` `“置信下限”` `“probability-of-improvement”` 최적화는목적함수의런타임에종속적이기때문에이름에`每秒`가포함된수집함수는재현가능한결과를산출하지않습니다。이름에`加`가의된된수집수집함수특정영역을과도하게수정수정하게될경우동작을수정수정하게될될경우경우수정수정자세한내용은采集函数类型항목을참조하십시오。	`“expected-improvement-per-second-plus”`
`MaxObjectiveEvaluations`	목적함수실행의최대입니다입니다。	`“bayesopt”`또는`“随机搜索”`의경우`30.`이고,`'gridsearch'`의 경우 그리드 전체입니다.
`MaxTime`	시간제한으로,양의실수로지정됩니다。시간제한은초단위이며,`抽搐`과`TOC.`으로 측정됩니다.`MaxTime`은은함수계산을을중단시키지않으므로실행은`MaxTime`을초과할수있습니다。	`正`
`NumGridDivisions`	`'gridsearch'`의경우,각차원의값개수입니다。이값은각차원에대한값의개수를제공하는양의정수로구성된벡터또는모든차원에적용되는스칼라일수있습니다。이필드는범주형변수의경우무시됩니다。	`10.`
`ShowPlots`	플롯표시여부를나타내는논리값입니다。`真的`인경우,이필드는반복횟수에대해가장적합한목적함수값을플로팅합니다。하나또는두개의최적화모수가있고`优化器`가`“bayesopt”`인 경우,`ShowPlots`는이모수에대해서도목적의모델을플로팅합니다。	`真的`
`SaveIntermediateResults`	`优化器`가`“bayesopt”`인경우결과를저장할지여부를나타내는논리값입니다。`真的`인경우,이필드는각반복마다`“BayeSotResults”`라는이름의작업공간변수덮어씁니다덮어씁니다덮어씁니다。변수는`贝叶斯偏见`객체입니다。	`错误的`
`verb`	명령줄에대한표시입니다。 `0`— 반복 표시 안 함 `1`— 반복 표시 `2`— 추가 정보와 함께 반복 표시 자세한내용은`Bayesopt.`의`verb`이름-값 쌍의 인수를 참조하십시오.	`1`
`使用指α.`	아이즈최적최적화를병렬로실행할지여부를나타내는값값，并行计算工具箱™가필요합니다。병렬시간재현이불가능하기때문에，병렬베이즈최적화에서반드시재현가능한결과를산출하지는않습니다。자세한내용은并行贝叶斯优化항목을참조하십시오。	`错误的`
`再分配`	매반복시교차검증을다시분할할지여부를나타내는논리값입니다。`错误的`인경우,최적화함수는최적화에단일분할을사용합니다。 `真的`인경우,분할잡음이고려되므로일반적으로가장견고한결과가제공됩니다。그러나,`真的`에서좋은결과를하려면적어도배배더횟수함수실행실행이필요실행이이이이이이이이	`错误的`
다음과 같은 3.개 필드 이름 중 하나만 사용합니다.
`CVPartition.`	`cvpartition`으로생성되는`cvpartition`객체입니다。	교차 검증 필드를 지정하지 않을 경우`“Kfold”，5`
`坚持`	홀드아웃비율을나타내는범위`(0,1)`내스칼라입니다。
`肯福尔德`	1보다큰정수입니다。

예:“HyperparameterOptimizationOptions”、结构(MaxObjectiveEvaluations, 60)

데이터형:结构体

출력인수

모두축소

`树`- 분류트리
분류트리객체

분류트리로,분류트리객체로반환됩니다。

'横向'，“KFold”，'坚持'，“Leaveout”또는“CVPartition”옵션을사용하면ClassificationededModel.클래스의트리가생성됩니다。예측에는분할된트리를사용할수없으므로이러한유형의트리에는预测메서드가없습니다。대신,kfoldPredict를사용하여훈련에사용되지않는관측값에대한응답변수를예측할수있습니다。

그렇지않은경우，树는ClassificationTree클래스이며,预测메서드를 사용하여 예측을 수행할 수 있습니다.

세부정보

모두축소

곡률 검정

곡률 검정은'두변수간에이없다'는가설을가하는평검정입니다。

예측 변수 x와 Y간의 곡률 검정은 다음 과정을 통해 수행됩니다.

x가연속형변수이면이변수를사분위수단위로분할합니다。관측값이어느사분위수로분할되었는지에따라관측값을本으로분류하는명목형변수를생성합니다。결측값이있는경우결측값에대한本을따로생성합니다。
분할된예측j = 1 ... j의각수준과응답k = 1，...，k의클래스에k클래스클래스포함된값의가중비율을합니다의。

${\overset{＾}{π}}_{j k} ＝ \sum_{我＝ 1}^{n} 我｛ y_{我} ＝ k ｝ w_{我} ．$

w_我는관측값我의가중치이고( $\sum w_{我} ＝ 1$ )，我는 표시 함수이며, N은 표본 크기입니다. 모든 관측값의 가중치가 동일한 경우 ${\overset{＾}{π}}_{j k} ＝ \frac{n_{j k}}{n}$ 입니다。여기서n_jk는k클래스에속하는예측변수의수준j에있는관측값개수입니다。
검정통계량을계산합니다。

$t ＝ n \sum_{k ＝ 1}^{K} \sum_{j ＝ 1}^{J} \frac{{（ {\overset{＾}{π}}_{j k} - {\overset{＾}{π}}_{j +} {\overset{＾}{π}}_{+ k} ）}^{2}}{{\overset{＾}{π}}_{j +} {\overset{＾}{π}}_{+ k}}$

${\overset{＾}{π}}_{j +} ＝ \sum_{k} {\overset{＾}{π}}_{j k}$ ，이는수준j에서예측변수가관측될주변확률입니다。 ${\overset{＾}{π}}_{+ k} ＝ \sum_{j} {\overset{＾}{π}}_{j k}$ , 이는 K클래스가 관측될 주변 확률입니다. N이 충분히 클 경우 T는 자유도가 (K-1）（J-1）인 χ²로분산됩니다。
검정에 대한 p-값이 0.05보다 작을 경우 'x와 Y간에 연관성이 없다'는 귀무가설을 기각합니다.

각 노드에서 최상의 분할 예측 변수를 결정할 때 표준 运货马车알고리즘은 많은 수준을 갖는 연속형 예측 변수를 선택하려 합니다. 하지만 가끔은 이는 잘못된 선택이 될 수 있으며 더 적은 개수의 수준을 갖지만 더 중요한 예측 변수(예: 범주형 예측 변수)를 감추는 결과를 낳을 수도 있습니다.

표준购物车대신곡률검정을적용하여각노드에서최상의분할예측변수를결정할수있습니다。이경우,최상의분할예측변수는각예측변수와응답변수간곡률검정의유의미한p -값(0.05보다작은값)을최소화하는예측변수입니다。이러한선택은개별예측변수의수준개수에대해견고합니다。

참고

예측 변수 수준이 특정 클래스에 대해 순수하면(즉, 모든 수준이 한 특정 클래스에 속하면),fitctree는 이러한 수준을 병합합니다. 따라서, 이 알고리즘의 3.단계에서 J는 예측 변수의 실제 수준 개수보다 작을 수 있습니다. 예를 들어, x에 4.개의 수준이 있고 垃圾箱1및 垃圾箱2에 있는 모든 관측값이 클래스 1.에 속하는 경우 이 수준은 클래스 1.에 대해 순수합니다. 따라서fitctree中用가宾1및bin 2에있는관측값을병합j가3가3으로줄어듭니다。

분류 트리를 성장시킬 때 곡률 검정이 어떻게 적용되는지에 대한 자세한 내용은노드분할규칙항목및［4］항목을참조하십시오。

불순도및노드오차

ClassificationTree는불순도또는노드 오차를기반으로하여노드를분할합니다。

불순도는사용자가선택하는SplitCriterion이름-값 쌍의 인수에 따라 다음 여러 항목 중 하나를 의미합니다.

지니다양성지수(gdi） - 노드의지니지수는다음과같습니다。

$1 - \sum_{我} p^{2} （我），$

여기서합계는노드에있는클래스我에대한것이며p (i)는클래스중노드에도달하는클래스我에대해관측된비율입니다。클래스를하나만갖는노드(순수노드）의지니지수는0이고,그렇지않은노드의경우지니지수는양수입니다。따라서지니지수는노드불순도를측정한값입니다。
이탈도(“越轨”)-p（i）가 지니 지수에서와 동일하게 정의되었을 때 노드의 이탈도는 다음과 같습니다.

$- \sum_{我} p （我） {日志}_{2} p （我）．$

순수노드의이탈도는0이고，그렇지그렇지노드의이탈도는는양수。
투잉규칙(“两个”)——투잉은노드의순수도를측정한값이아니며노드분할방법결정에사용하는다른측정값입니다。L(我)는분할후왼쪽자식노드에서클래我에스속하는비율을나타내고,R (i)는분할후오른쪽자식노드에서클래我에스속하는비율을나타낸다고하겠습니다。다음을최대화할분할기준을선택합니다。

$P （ l ） P （ R ） {（ \sum_{我} | l （我） - R （我） | ）}^{2} ，$

여기서P (L)및P (R)은각각왼쪽과오른쪽으로분할되는관측값의비율입니다。표현식값이클경우분할로인해각자식노드가더순수해졌다는의미입니다。마찬가지로,표현식값이작을경우분할로인해각자식노드가서로비슷해졌고,이는즉부모노드와비슷해졌다는것입니다。분할로인해노드의순수도가높아지지않았습니다。
노드오차——노드오차는노드에서오분류된클래스의비율입니다。j가노드에서최대개수의훈련표본을갖는클래스인경우노드오차는다음과같습니다。

1 - p（j）。

상호작용검정

상호작용검정은'예측변수쌍과응답변수간에상호작용이없다’는귀무가설을평가하는통계검정입니다。

y에대한예측변수x₁과x₂간의 연관성을 평가하는 상호 작용 검정은 다음 과정을 통해 수행됩니다.

x₁또는X₂가연속형변수이면이변수를사분위수단위로분할합니다。관측값이어느사분위수로분할되었는지에따라관측값을本으로분류하는명목형변수를생성합니다。결측값이있는경우결측값에대한本을따로생성합니다。
当J = J₁J₂개수준을갖는명목형변z수를생성합니다。이변수는x₁과x₂중어떤수준에속하느냐에따라인덱스를관측값我에할당합니다。관측값에대응되지않는ž의수준을제거합니다。
z와y간간곡률 검정을 수행합니다.

결정트리를성장시킬때예측변수쌍간에중요한상호작용이있지만해당데이터에덜중요한다른예측변수도많은경우표준购物车는중요한상호작용을놓치는경향이있습니다。그러나,예측변수선택에곡률검정과상호작용검정을대신수행하면중요한상호작용을더잘감지할수있으며,따라서더정확한결정트리를생성할수있습니다。

결정 트리를 성장시킬 때 상호 작용 검정이 어떻게 적용되는지에 대한 자세한 내용은곡률 검정，노드분할규칙，［3］항목을참조하십시오。

예측연관성측도

예측연관성측도는 관측값을 분할하는 결정 규칙 간 유사성을 나타내는 값입니다. (트리를 성장시켜서 확인할 수 있는) 최적의 분할과 비교되는 모든 가능한 결정 분할 중에 최상의대리결정분할은최대의예측연관성측도를생성합니다。두번째로최상인대리분할은두번째로큰예측연관성측도를생성합니다。

x_j및 x_k는각각예측변수j및k이고，j≠k라고가정하겠습니다。노드t에서최적의분할x_jk

$λ_{j k} ＝ \frac{闵（ P_{l} ， P_{R} ） - （ 1 - P_{l_{j} l_{k}} - P_{R_{j} R_{k}} ）}{闵（ P_{l} ， P_{R} ）} ．$

P_l은 노드 T에서 x_j
P_R은 노드 T에서 x_j< u를충족하는관측값의비율입니다。아래첨자R은노드t의오른쪽자식을나타냅니다。
$P_{l_{j} l_{k}}$ 는 노드 T에서 x_j< u및x_k
$P_{R_{j} R_{k}}$ 는 노드 T에서 x_j≥u및x_k≥v를충족하는관측값의비율입니다。
x_j또는X_k에대한결측값이있는관측값은비율계산에포함되지않습니다。

λ_jk는（-∞，1]범위의값입니다。λ_jk> 0이면x_kj

대리결정분할

대리결정분할은 결정 트리의 주어진 노드에서의 최적의 결정 분할에 대한 대안입니다. 최적의 분할은 트리를 성장시켜서 확인할 수 있으며, 대리 분할은 이와 유사하거나 상관관계가 있는 예측 변수 및 분할 기준을 사용합니다.

관측값에대한최적의분할예측변수값이누락된경우,관측값은최상의대리예측변수를통해왼쪽또는오른쪽자식노드로보내집니다。관측값에대한최상의대리분할예측변수값도누락된경우,관측값은두번째로최상인대리예측변수을통해왼쪽또는오른쪽자식노드로보내지며,이렇게값이누락된경우그다음으로최상인대리예측변수를통해보내지는식입니다。후보분할은해당예측연관성측도를기준으로내림차순으로정렬됩니다。

팁

기본적으로修剪은“上”입니다。그러나，이렇게지정되도분류트리가가지치기되지않습니다。훈련된분류트리를가지치기하려면분류트리를修剪에전달해야합니다。
모델을훈련시킨후에는새데이터에대한레이블을예측하는C / c++코드를생성할수있습니다。C / c++코드를생성하려면MATLAB编码器™가필요합니다。자세한내용은代码生成简介항목을참조하십시오。

알고리즘

모두축소

노드분할규칙

fitctree는

표준购物车（즉，PredictorSelection이'allpaes'인경우）및모든예측변수X_我， I = 1，…p의경우:
1. fitctree가노드t의가중불순도(我_t)를 계산합니다. 지원되는 불순도 측정법은SplitCriterion을참조하십시오。
2. fitctree가다음을사용하여관측값이노드吨에있을확률을추정합니다。
  
  $P （ T ）＝ \sum_{j \in T} w_{j} ．$
  
  w_j는 관측값 J의 가중치이고 T는 노드 T에 있는 모든 관측값 인덱스의 집합입니다.事先的또는权重를지정하지않을경우w_j=1/n입니다. 여기서 N은 표본 크기입니다.
3. fitctree가 x_我를 오름차순으로 정렬합니다. 정렬된 예측 변수의 각 요소는 분할 후보, 즉 절단점입니다.fitctree는분할되지않은집합인T_U의결측값에대응되는인덱스를저장합니다。
4. fitctree가 모든 분할 후보에 대한 불순도 이득(ΔI）을 최대화함으로써 x_我를사용하여하여노드노드분할하는하는최상의방법결정결정결정결정즉，X._我의모든분할후보에대해다음을수행합니다。
  1. fitctree가노드t에있는관측값을왼쪽및오른쪽자식노드(각각t_l및 T_R)로 분할합니다.
  2. fitctree使用权限ΔI를를계산。특정분할후보에대해t_l및 T_R이각각집합t_l및 T_R의관측값인덱스를포함한다고가정합니다。
    
    x_我에아무런결측값이없을경우현재분할후보의불순도이득은다음과같습니다。
    
    $Δ 我＝ P （ T ）我_{t} - P （ T_{l} ）我_{t_{l}} - P （ T_{R} ）我_{t_{R}} ．$
    
    x_我에결측값이있는경우관측값이임의로누락되었다는가정하에불순도이득은다음과같습니다。
    
    $Δ 我_{U} ＝ P （ T - T_{U} ）我_{t} - P （ T_{l} ）我_{t_{l}} - P （ T_{R} ）我_{t_{R}} ．$
    
    T - T_U는 노드 T의 누락되지 않은 모든 관측값 인덱스의 집합입니다.
    
    대리결정분할을 사용하는 경우 다음과 같습니다.
    
    fitctree가결정분할x_jk< v, j≠k간의예측연관성측도를계산합니다。
    
    fitctree가의의분할에대한연관성측도를기준으로가능한한대체결정분할을차순으로정렬정렬대리분할은최대측도값을생성하는결정분할입니다。
    
    fitctree가대리대리분할사용하여x_我의결측값이있는관측값에대한자식노드할당을결정합니다。대리예측변수에도결측값이있는경우fitctree는두번째로큰측도를갖는결정분할을사용하며，이런식으로다른대리분할이없을때까지그다음으로큰측도를갖는결정분할을계속사용합니다。fitctree가두개의서로다른대리분할을사용하여노드t에서두개의각기다른관측값을분할하는것도가능합니다。예를들어,예측변수x₁및 x₂가 각각 노드 T에서 예측 변수 x_我，我∉{1,2}에대한최상의대리분할과번째최상인대리분할이라가라가정정하겠습니다。예측변수X._我의 관측값 M이 누락(즉, x_惯性矩가누락됨)되었지만x_m1이누락되지않은경우x₁이 관측값 x_惯性矩에대한대리예측변수입니다。관측값x_{（m + 1），我}및 x（m+1），1이 누락되었지만 x_{（m + 1），2}가누락되지않은경우x₂가관측관측M+ 1에대한대리예측변수。
    
    fitctree가적합한불순도이득식을사용합니다。즉，fitctree가 대리 분할을 사용하여 노드 T에서 누락된 모든 관측값을 자식 노드에 할당하지 못하면 불순도 이득이 ΔI_U입니다。그렇지않은경우fitctree가불순도이득에ΔI를사용합니다。
  3. fitctree가최대불순도이득을생성하는후보를선택합니다。
fitctree가 불순도 이득을 최대화하는 절단점에서 예측 변수를 분할합니다.
곡률검정의경우(즉PredictorSelection이“弯曲”인 경우):
1. fitctree가 노드 T의 관측값에 대한 각 예측 변수와 응답 변수 간에곡률 검정을 수행합니다.
  - 모든对 - 값이최소0.05이면fitctree가노드吨를분할하지않습니다。
  - 최소 p-값이 있으면fitctree가 이에 대응하는 예측 변수를 선택하여 노드 T를 분할합니다.
  - 언더플로로인해둘이상의p-값이0인경우fitctree가표준购物车를를대응하는예측에적용하여분할예측를선택선택선택를를선택
2. fitctree가의할예측변수를선택표준표준购物车를사용하여절단점을선택（표준购物车과정의4단계참조）。
상호작용검정의경우（즉，PredictorSelection이“interaction-curvature”인 경우):
1. t의노드관측값에대해fitctree가 각 예측 변수와 응답 변수 간에곡률 검정을 수행하고 각 예측 변수 쌍과 응답 변수 간에상호작용검정을 수행합니다.
  - 모든对 - 값이최소0.05이면fitctree가노드吨를분할하지않습니다。
  - 최소p -값이있고이값이곡률검정의결과이면fitctree가 이에 대응하는 예측 변수를 선택하여 노드 T를 분할합니다.
  - 최소p -값이있고이값이상호작용검정의결과이면fitctree가이이에대응대응하는예측변수쌍에대해购物车를사용하여하여분할예측를선택선택선택를선택합니다
  - 언더플로로인해둘이상의p-값이0인경우fitctree가표준购物车를를대응하는예측에적용하여분할예측를선택선택선택를를선택
2. fitctree가의할예측변수를선택표준표준购物车를사용하여절단점을선택（표준购物车과정의4단계참조）。

트리깊이제어

MergeLeaves가“上”이고普吕尼特里翁이“错误”인경우（이이름 - 값쌍의인수에대한디폴트값임），소프트웨어가분류오차를사용하여리프에만가지치기를적용합니다。이렇게지정하면리프당가장많이사용되는클래스를공유하는리프를병합하는것과같은결과를낳습니다。
maxnumsplits.를수용할수있도록fitctree는현재계층의모든노드를분할한가지노드노드의의개수계산계산계층은루트노드에서등거리에있는노드의집합입니다。가지노드의가maxnumsplits.를초과할경우fitctree는다음절차를따릅니다。
1. 가지 노드 개수가 최대maxnumsplits.개가되려면현재계층에서분할되지않아야하는가지노드의개수가몇개인지판별합니다。
2. 불순도이득을기준으로가지노드를정렬。
3. 성공도가낮은가지의개수만큼분할을취소합니다。
4. 지금까지 성장한 결정 트리를 반환합니다.
이절차를통해최대로균형잡힌가생성생성。
다음중최소이상의일이생길때까지소프트웨어는각가지노드를분할합니다。
- maxnumsplits.개의가지노드가있습니다。
- 제안된분할로인해하나이상의가지노드에있는관측값개수가MinParentSize보다작아집니다。
- 제안된분할로인해하나이상의리프노드에있는관측값개수가小叶大小보다작아집니다。
- 알고리즘이계층내에서알맞은분할을찾을수없습니다(즉,계층내제안된모든분할에대해가지치기기준(普吕尼特里翁참조）이개선되지않음）。모든노드가순수한경우는특수한사례입니다（즉，노드에포함된모든관측값이동일한클래스를가짐）。
- PredictorSelection의 값이“弯曲”또는“interaction-curvature”인경우,모든검정에서0.05보다큰p -값이생성됩니다。
maxnumsplits.및小叶大小의의디폴트값은은분할영향을미치지미치지따라서，“MaxNumSplits”를설정하면maxnumsplits.개의분할이발생하기전에MinParentSize값으로 인해 분할이 중지될 수 있습니다.

병렬화

듀얼 코어 이상의 시스템인 경우,fitctree는英特尔^®TBB（穿线构建块）를사용하여하여훈련결정트리를병렬화英特尔TBB에에대한자세한내용내용https://software.intel.com/en-us/intel-tbb를참조하십시오。

참고 문헌

[1] Breiman, L.， J. Friedman, R. Olshen, C. Stone。分类和回归树。佛罗里达州博卡拉顿:CRC出版社，1984。

Coppersmith, D.， S. J. Hong, J. R. M. Hosking。在决策树中划分标称属性。数据挖掘与知识发现，1999年第3卷，第197-217页。

[3] Loh, W.Y.， <具有无偏变量选择和交互检测的回归树>《统计》，2002年第12卷，第361-386页。

Loh w.y y and Y.S. Shih分类树的分裂选择方法中国统计，1997年第7卷，第815-840页。

확장기능

高형배열
메모리에 담을 수 없을 정도로 많은 행을 가진 배열을 계산할 수 있습니다.

用户使用사용법：

자세한내용은高형배열항목을참조하십시오。

자동병렬지원
并行计算工具箱™를사용해해자동병렬을실행하여코드실행속도높일수수수높일수

병렬하이퍼파라미터최적화를수행하려면이함수에대한호출에'hyperparameteroptimizationOptions'，struct（'使用反应'，true）이름——값쌍의인수를사용하십시오。

병렬하이퍼파라미터최적화에대한자세한내용은并行贝叶斯优化항목을참조하십시오。

병렬연산에대한일반적인내용은자동병렬연산을지원하는MATLAB함수실행하기（并行计算工具箱）항목을참조하십시오。

참고항목

ClassificationededModel.|ClassificationTree|kfoldPredict|预测|修剪

도움말항목

分类树分类预测器的分裂

R2014A에개발됨

fitctree

구문

설명

예제

분류트리성장시키기

트리트리이제어제어

분류트리최적화하기

무편향예측변수중요도추정값

高的형 배열에 대한 분류 트리 최적화하기

입력인수

资源描述- - - - - -표본데이터表형

ResponseVarName- - - - - -응답 변수 이름资源描述에포함된변수이름

公式- - - - - -응답변수와예측변수의부분집합에대한설명모델문자형 벡터|字符串형스칼라

Y- - - - - -클래스레이블숫자형 벡터|绝对的형 벡터|논리형벡터|문자형 배열|字符串형배열|문자형벡터로구성된셀형배열

X- - - - - -예측변수데이터숫자형 행렬

이름-값 쌍의 인수

“分类算法”- - - - - -최상의 범주형 예측 변수 분할을 위한 알고리즘“准确”|'PullLeft'|主成分分析的|'ovabyclass'

“CategoricalPredictors”- - - - - -범주형예측변수목록양의 정수로 구성된 벡터|논리형벡터|문자형행렬|字符串형배열|문자형벡터로구성된셀형배열|'全部'

“类名”- - - - - -훈련에 사용할 클래스의 이름分类형형.|문자형 배열|字符串형배열|논리형벡터|숫자형 벡터|문자형벡터로구성된셀형배열

'成本'- - - - - -오분류비용정사각 행렬|구조체

'maxdepth'- - - - - -최대트리깊이양의정수

“最大分类”- - - - - -최대범주수준10.(디폴트값)|음이 아닌 스칼라 값

“MaxNumSplits”- - - - - -결정분할의최대개수大小(X, 1) - 1(디폴트값)|양의정수

“MergeLeaves”- - - - - -리프 병합 플래그“上”(디폴트값)|“关闭”

'minleafsize'- - - - - -리프리프노드관측값값의최소1(디폴트값)|양의정수값

'迷人'- - - - - -가지노드관측값의최소개수10.(디폴트값)|양의정수값

“麻木”- - - - - -숫자형예측변수의箱개수[](비어있음)(디폴트값)|양의정수스칼라

“NumVariablesToSample”- - - - - -각분할에대해임의로선택할예측변수의개수'全部'(디폴트값)|양의정수값

“PredictorNames”- - - - - -예측 변수 이름고유한이름으로으로으로구성구성된형배열|고유한 문자형 벡터로 구성된 셀형 배열

'PredictorSelection'- - - - - -최상의최상의분할예측예측변수를선택하는사용용되는“allsplits”(디폴트값)|“弯曲”|“interaction-curvature”

“之前”- - - - - -사전확률“经验的”(디폴트값)|“制服”|스칼라 값으로 구성된 벡터|구조체

'修剪'- - - - - -가지치기된하위트리에대한최적의시퀀스를추정하는플래그“上”(디폴트값)|“关闭”

“PruneCriterion”- - - - - -가지치기치기.“错误”(디폴트값)|“杂质”

“可复制”- - - - - -재현성을적용하기위한플래그错误的(논리값0）(디폴트값)|真的(논리값1）

'responsebame'- - - - - -응답 변수 이름“是的”(디폴트값)|문자형 벡터|字符串형스칼라

“分数变换”- - - - - -점수변환방식“没有”(디폴트값)|'doublelogit'|“因弗罗吉特”|'ismax'|分对数的|함수핸들|...

“拆分标准”- - - - - -분할기준'GDI'(디폴트값)|“两个”|“越轨”

“代理”- - - - - -대리결정분할플래그“关闭”(디폴트값)|“上”|'全部'|양의정수값

“重量”- - - - - -관측값 가중치（大小（x，1），1）(디폴트값)|스칼라 값으로 구성된 벡터|资源描述에포함된변수이름

'横向'- - - - - -교차교차검증된결정결정트리를시키는시키는“关闭”(디폴트값)|“上”

“CVPartition”- - - - - -교차 검증된 트리에 사용할 분할cvpartition객체

'坚持'- - - - - -홀드아웃검증에사용할데이터의비율0(디폴트값)|[0,1]범위의스칼라값

“KFold”- - - - - -겹의개수10.(디폴트값)|1보다큰양의정수값

“Leaveout”- - - - - -리브-원아웃(分析)교차검증플래그“关闭”(디폴트값)|“上”

'OptimizeHyperparameters'- - - - - -최적화할모수“没有”(디폴트값)|“汽车”|'全部'|적합한모수이름으로으로구성된된형형배열또는셀형배열|optimizableVariable객체로 구성된 벡터

'hyperparameteroptimizationOptions'- - - - - -최적화에 사용할 옵션구조체

출력인수

树- 분류트리분류트리객체

세부정보

곡률 검정

불순도및노드오차

상호작용검정

예측연관성측도

대리결정분할

팁

알고리즘

노드분할규칙

트리깊이제어

병렬화

참고 문헌

확장기능

高형배열메모리에 담을 수 없을 정도로 많은 행을 가진 배열을 계산할 수 있습니다.

자동병렬지원并行计算工具箱™를사용해해자동병렬을실행하여코드실행속도높일수수수높일수

참고항목

도움말항목

统计和机器学习工具箱문서

지원

掌握机器学习：使用MATLAB逐步指南

`资源描述`- - - - - -표본데이터
表형

`ResponseVarName`- - - - - -응답 변수 이름
`资源描述`에포함된변수이름

`公式`- - - - - -응답변수와예측변수의부분집합에대한설명모델
문자형 벡터|字符串형스칼라

`Y`- - - - - -클래스레이블
숫자형 벡터|绝对的형 벡터|논리형벡터|문자형 배열|字符串형배열|문자형벡터로구성된셀형배열

`X`- - - - - -예측변수데이터
숫자형 행렬

`“分类算法”`- - - - - -최상의 범주형 예측 변수 분할을 위한 알고리즘
`“准确”`|`'PullLeft'`|`主成分分析的`|`'ovabyclass'`

`“CategoricalPredictors”`- - - - - -범주형예측변수목록
양의 정수로 구성된 벡터|논리형벡터|문자형행렬|字符串형배열|문자형벡터로구성된셀형배열|`'全部'`

`“类名”`- - - - - -훈련에 사용할 클래스의 이름
分类형형.|문자형 배열|字符串형배열|논리형벡터|숫자형 벡터|문자형벡터로구성된셀형배열

`'成本'`- - - - - -오분류비용
정사각 행렬|구조체

`'maxdepth'`- - - - - -최대트리깊이
양의정수

`“最大分类”`- - - - - -최대범주수준
`10.`(디폴트값)|음이 아닌 스칼라 값

`“MaxNumSplits”`- - - - - -결정분할의최대개수
`大小(X, 1) - 1`(디폴트값)|양의정수

`“MergeLeaves”`- - - - - -리프 병합 플래그
`“上”`(디폴트값)|`“关闭”`

`'minleafsize'`- - - - - -리프리프노드관측값값의최소
`1`(디폴트값)|양의정수값

`'迷人'`- - - - - -가지노드관측값의최소개수
`10.`(디폴트값)|양의정수값

`“麻木”`- - - - - -숫자형예측변수의箱개수
`[]`(비어있음)(디폴트값)|양의정수스칼라

`“NumVariablesToSample”`- - - - - -각분할에대해임의로선택할예측변수의개수
`'全部'`(디폴트값)|양의정수값

`“PredictorNames”`- - - - - -예측 변수 이름
고유한이름으로으로으로구성구성된형배열|고유한 문자형 벡터로 구성된 셀형 배열

`'PredictorSelection'`- - - - - -최상의최상의분할예측예측변수를선택하는사용용되는
`“allsplits”`(디폴트값)|`“弯曲”`|`“interaction-curvature”`

`“之前”`- - - - - -사전확률
`“经验的”`(디폴트값)|`“制服”`|스칼라 값으로 구성된 벡터|구조체

`'修剪'`- - - - - -가지치기된하위트리에대한최적의시퀀스를추정하는플래그
`“上”`(디폴트값)|`“关闭”`

`“PruneCriterion”`- - - - - -가지치기치기.
`“错误”`(디폴트값)|`“杂质”`

`“可复制”`- - - - - -재현성을적용하기위한플래그
`错误的`(논리값`0`）(디폴트값)|`真的`(논리값`1`）

`'responsebame'`- - - - - -응답 변수 이름
`“是的”`(디폴트값)|문자형 벡터|字符串형스칼라

`“分数变换”`- - - - - -점수변환방식
`“没有”`(디폴트값)|`'doublelogit'`|`“因弗罗吉特”`|`'ismax'`|`分对数的`|함수핸들|...

`“拆分标准”`- - - - - -분할기준
`'GDI'`(디폴트값)|`“两个”`|`“越轨”`

`“代理”`- - - - - -대리결정분할플래그
`“关闭”`(디폴트값)|`“上”`|`'全部'`|양의정수값

`“重量”`- - - - - -관측값 가중치
`（大小（x，1），1）`(디폴트값)|스칼라 값으로 구성된 벡터|`资源描述`에포함된변수이름

`'横向'`- - - - - -교차교차검증된결정결정트리를시키는시키는
`“关闭”`(디폴트값)|`“上”`

`“CVPartition”`- - - - - -교차 검증된 트리에 사용할 분할
`cvpartition`객체

`'坚持'`- - - - - -홀드아웃검증에사용할데이터의비율
`0`(디폴트값)|`[0,1]`범위의스칼라값

`“KFold”`- - - - - -겹의개수
`10.`(디폴트값)|1보다큰양의정수값

`“Leaveout”`- - - - - -리브-원아웃(分析)교차검증플래그
`“关闭”`(디폴트값)|`“上”`

`'OptimizeHyperparameters'`- - - - - -최적화할모수
`“没有”`(디폴트값)|`“汽车”`|`'全部'`|적합한모수이름으로으로구성된된형형배열또는셀형배열|`optimizableVariable`객체로 구성된 벡터

`'hyperparameteroptimizationOptions'`- - - - - -최적화에 사용할 옵션
구조체

`树`- 분류트리
분류트리객체

高형배열
메모리에 담을 수 없을 정도로 많은 행을 가진 배열을 계산할 수 있습니다.

자동병렬지원
并行计算工具箱™를사용해해자동병렬을실행하여코드실행속도높일수수수높일수