如何修复：值错误：标签数=21638与样本数=13不匹配

Question

我对Python2.7非常陌生，我尝试在我的数据集上运行决策树分类器，但是在一个教程之后，我遇到了这个问题，我首先将我的features列矢量化并将其保存到数组中，然后使用labelencoder将目标列保存在数组中。请你给我解释一下如何修正这个错误？在

数据：

代码：

import pandas as pd

dataset = "C:/Users/ashik swaroop/Desktop/anaconda/Gene Dataset/Final.csv"
datacan = pd.read_csv(dataset)
datacan = datacan.fillna('')
features = datacan[[ 
"Tumour_Types_Somatic","Tumour_Types_Germline",
"Cancer_Syndrome","Tissue_Type", 
"Role_in_Cancer","Mutation_Types","Translocation_Partner",
"Other_Syndrome","Tier","Somatic","Germline",
"Molecular_Genetics","Other_Germline_Mut"]]

 from sklearn.feature_extraction import DictVectorizer
from sklearn.preprocessing import LabelEncoder

X_dict = features.to_dict().values()
vect = DictVectorizer(sparse=False)
X_vector = vect.fit_transform(X_dict)

le = LabelEncoder()
y_train = le.fit_transform(datacan['Gene_Symbol'][:-1])


X_Train = X_vector[:-1]

X_Test = X_vector[-1:] 

from sklearn import tree

clf = tree.DecisionTreeClassifier(criterion='entropy')
clf = clf.fit(X_Train,y_train)    `

我得到了这个错误：

^{pr2}$

Answer 1

首先，要理解错误： 似乎您的训练样本数（即np.shape(X_train)[0]）与标签数目（即np.shape(y_train)[0]）不匹配。在

在查看代码时，我注意到一些不一致之处。请参考下面的在线评论。在

import pandas as pd
from apyori import apriori
dataset = "C:/Users/ashik swaroop/Desktop/anaconda/Gene Dataset/Final.csv"
datacan = pd.read_csv(dataset)
datacan = datacan.fillna('')
features = datacan[[ 
"Tumour_Types_Somatic","Tumour_Types_Germline",
"Cancer_Syndrome","Tissue_Type", 
"Role_in_Cancer","Mutation_Types","Translocation_Partner",
"Other_Syndrome","Tier","Somatic","Germline",
"Molecular_Genetics","Other_Germline_Mut"]]
# EDIT replace by features = [ 
#"Tumour_Types_Somatic","Tumour_Types_Germline",
#"Cancer_Syndrome","Tissue_Type", 
#"Role_in_Cancer","Mutation_Types","Translocation_Partner",
#"Other_Syndrome","Tier","Somatic","Germline",
#"Molecular_Genetics","Other_Germline_Mut"]

orders = datacan[features].to_dict( orient = 'records' ) # this variable is not used

from sklearn.feature_extraction import DictVectorizer
from sklearn.preprocessing import LabelEncoder

X_dict = features.to_dict().values() # try replacing this line with X_dict = orders
vect = DictVectorizer(sparse=False)
X_vector = vect.fit_transform(X_dict)

le = LabelEncoder()
y_train = le.fit_transform(datacan['Gene_Symbol'][:-1])


X_Train = X_vector[:-1]

X_Test = X_vector[-1:] 

from sklearn import tree

clf = tree.DecisionTreeClassifier(criterion='entropy')
clf = clf.fit(X_Train,y_train)