Question

我是python和机器学习的新手

我的代码是

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
from sklearn import cross_validation, svm,cross_validation,neighbors

df=pd.read_csv('train_loan_prediction.csv')
df.drop(['Loan_ID'],1,inplace=True)

df.replace('Graduate',1,inplace=True)
df.replace('Not Graduate',0,inplace=True)
df.replace('Urban',2,inplace=True)
df.replace('Semiurban',1,inplace=True)
df.replace('Rural',1,inplace=True)
df.replace('Yes',1,inplace=True)
df.replace('No',0,inplace=True)
df.replace('Male',1,inplace=True)
df.replace('Female',0,inplace=True)
df.replace('Y',1,inplace=True)
df.replace('N',0,inplace=True)
df.replace('3+',4,inplace=True)

x=np.array(df.drop(['Loan_Status'],1))
y=np.array(df['Loan_Status'])

df.fillna(0,inplace=True)

df.head(20)
    x_train,x_test,y_train,y_test=cross_validation.train_test_split(x,y,test_size=0.2
)

clf=neighbors.KNeighborsClassifier()
clf.fit(x_train, y_train)

Answer 1

在设置x和y之前，您需要替换中的数据框中的NaN：

# replace NaNs first df.fillna(0,inplace=True) # then set features and labels x = np.array(df.drop(['Loan_Status'],1)) y = np.array(df['Loan_Status'])

错误获取输入内容包含NaN，无穷大或dtype（'float64'）太大的值

1 个答案: