我试过这个，但无法让它适用于我的数据： Use Scikit Learn to do linear regression on a time series pandas data frame
我的数据包含2个DataFrame。 DataFrame_1.shape = (40,5000)和DataFrame_2.shape = (40,74)。我正在尝试进行某种类型的线性回归，但DataFrame_2包含NaN个缺失的数据值。当我DataFrame_2.dropna(how="any")时，形状会降至(2,74)。
sklearn中是否有可以处理NaN值的线性回归算法？
我正在load_boston之后sklearn.datasets建模X,y = boston.data, boston.target = (506,13),(506,) X = DataFrame_1
for col in DataFrame_2.columns:
y = DataFrame_2[col]
model = LinearRegression()
model.fit(X,y)
#ValueError: Input contains NaN, infinity or a value too large for dtype('float64').
这是我的简化代码：
DataFrame_2我使用上面的格式来获得与矩阵相匹配的形状
如果发布#include <iostream>
#include <cmath>
#include <cstdlib>
#include <string>
#include <algorithm>
#include <cctype>
using namespace std;
int main() {
string s;
char selection;
string w;
cout << "Enter a paragraph or a sentence : ";
getline(cin, s);
int sizeOfString = s.length();
//cout << "The paragraph has " << sizeOfString << " characters. " << endl; ***Dummy call to see if size works.
//cout << "You entered " << s << endl; *** Dummy function !!
cout << "" << endl;
cout << " Menu " << endl;
cout << " ------------------------" << endl;
cout << "" << endl;
cout << "A -- Convert paragraph to all caps " << endl;
cout << "B -- Convert paragraph to all lowercase " << endl;
cout << "C -- Delete whitespaces " << endl;
cout << "D -- Split words & remove duplicates " << endl;
cout << "E -- Search a certain word " << endl;
cout << "" << endl;
cout << "Please select one of the above: ";
cin >> selection;
cout << "" << endl;
switch (selection) //Switch statement
{
case 'a':
case 'A':
cout << "You chose to convert the paragraph to all uppercase" << endl;
cout << "" << endl;
for (int i = 0; s[i] != '\0'; i++) {
s[i] = toupper(s[i]);
}
cout << "This is it: " << s << endl;
break;
case 'b':
case 'B':
cout << "You chose to convert the paragragh to all lowercase" << endl;
cout << "" << endl;
for (int i = 0; s[i] != '\0'; i++) {
s[i] = tolower(s[i]);
}
cout << "This is it: " << s << endl;
break;
case 'c':
case 'C':
cout << "You chose to delete the whitespaces in the paragraph" << endl;
cout << "" << endl;
for (int i = 0; i < s.length(); i++) {
if (s[i] == ' ')
s.erase(i, 1);
}
cout << "This is it: " << s << endl;
break;
case 'd':
case 'D':
cout << "You chose to split the words & remove the duplicates in the paragraph" << endl;
cout << "" << endl;
case 'e':
case 'E':
cout << "You chose to search for a certain word in the paragraph. " << endl;
cout << "" << endl;
cout << "Enter the word you want to search for: ";
cin >> w;
s.find(w);
if (s.find(w) != std::string::npos) {
cout << w << " was found in the paragraph. " << endl;
} else {
cout << w << " was not found in the paragraph. " << endl;
}
}
return 0;
}
会有所帮助，请在下方发表评论，我会添加它。
答案 0 :(得分：3)
您可以使用插补填写y中的空值。在scikit-learn中，使用以下代码段完成此操作：
from sklearn.preprocessing import Imputer
imputer = Imputer()
y_imputed = imputer.fit_transform(y)
否则，您可能希望使用74列的子集作为预测变量来构建模型，也许您的某些列包含的空值较少？
答案 1 :(得分：0)
如果变量是DataFrame，则可以使用fillna。在这里，我用该列的平均值替换了丢失的数据。
df.fillna(df.mean(), inplace=True)