我正在尝试在此处提供的数据集中使用R执行逻辑回归：http://archive.ics.uci.edu/ml/machine-learning-databases/00451/ 这与乳腺癌有关。该数据集包含 Classification 列，其中仅包含 1 （如果患者没有癌症）或 2 （如果患者患有癌症）
library(ISLR)
dataCancer <- read.csv("~/Desktop/Isep/Machine
Leaning/TD/Project_Cancer/dataR2.csv")
attach(dataCancer)
names(dataCancer)
summary(dataCancer)
cor(dataCancer[,-11])
pairs(dataCancer[,-11])
#Step : Split data into training and testing data
training = (BMI>25)
testing = !training
training_data = dataCancer[training,]
testing_data = dataCancer[testing,]
Classification_testing = Classification[testing]
#Step : Fit a logistic regression model using training data
as.factor(dataCancer$Classification)
classification_model = glm(Classification ~ ., data =
training_data,family = binomial )
summary(classification_model)
运行脚本时，我得到：
> classification_model = glm(Classification ~ ., data = training_data,family = binomial )
Error in eval(family$initialize) : y values must be 0 <= y <= 1
> summary(classification_model)
Error in summary(classification_model) : object 'classification_model' not found .
我在其他帖子中添加了 as.factor（dataCancer $ Classification），但它没有解决我的问题。 如果这是该预测变量的内容，您能否建议我将分类的值设置为0到1之间的方法？ 感谢您的帮助。
答案 0 :(得分：2)
您在脚本中添加了import UIKit
import CoreData
class ViewController: UIViewController {
let fetchResultController = CoreDataStack.shared.fetchedResultsController
@IBOutlet weak var tableView: UITableView!
override func viewDidLoad() {
super.viewDidLoad()
print(NSTemporaryDirectory())
title = "List"
tableView.dataSource = self
}
override func viewWillAppear(_ animated: Bool) {
super.viewWillAppear(animated)
CoreDataStack.shared.fetchedResultsController.delegate = self
try! CoreDataStack.shared.fetchedResultsController.performFetch()
}
override func didReceiveMemoryWarning() {
super.didReceiveMemoryWarning()
// Dispose of any resources that can be recreated.
}
@IBAction func addItems(_ sender: UIBarButtonItem) {
let alert = UIAlertController(title: "Enter The Item Name", message: nil, preferredStyle: .alert)
alert.addTextField(configurationHandler: nil)
alert .addTextField { (passwordTextField) in
let placeholder = NSAttributedString(string: "Password", attributes: nil)
passwordTextField.attributedPlaceholder = placeholder
}
let cancel = UIAlertAction(title: "Cancel", style: .cancel, handler: nil)
let save = UIAlertAction(title: "Save", style:.default) { [unowned self] (action) in
guard let nameTextfield = alert.textFields?.first, let name = nameTextfield.text else {
return
}
guard let passwordTextfield = alert.textFields?[1], let password = passwordTextfield.text else {
return
}
let entity = NSEntityDescription.entity(forEntityName: "UserEntity", in: CoreDataStack.shared.context)
let userEntity = UserEntity(entity: entity!, insertInto: CoreDataStack.shared.context)
userEntity.name = name
userEntity.password = password
try? CoreDataStack.shared.context.save()
self.tableView.reloadData()
}
alert.addAction(cancel)
alert.addAction(save)
self.present(alert, animated: true, completion: nil)
}
}
extension ViewController: UITableViewDataSource {
func numberOfSections(in tableView: UITableView) -> Int {
guard let sectionCount = CoreDataStack.shared.fetchedResultsController.sections?.count else {
return 0
}
return sectionCount
}
func tableView(_ tableView: UITableView, numberOfRowsInSection section: Int) -> Int {
guard let sections = CoreDataStack.shared.fetchedResultsController.sections else {
return 0
}
return sections[section].numberOfObjects
}
func tableView(_ tableView: UITableView, cellForRowAt indexPath: IndexPath) -> UITableViewCell {
let cell = tableView.dequeueReusableCell(withIdentifier: "DataCell", for: indexPath)
let userEntity = CoreDataStack.shared.fetchedResultsController.object(at: indexPath)
cell.textLabel?.text = userEntity.name
cell.detailTextLabel?.text = userEntity.password
return cell
}
}
extension ViewController: NSFetchedResultsControllerDelegate {
func controllerWillChangeContent(_ controller: NSFetchedResultsController<NSFetchRequestResult>) {
tableView.beginUpdates()
}
func controller(_ controller: NSFetchedResultsController<NSFetchRequestResult>, didChange anObject: Any, at indexPath: IndexPath?, for type: NSFetchedResultsChangeType, newIndexPath: IndexPath?) {
switch(type){
case .insert:
if let indexPath = newIndexPath {
tableView.insertRows(at: [indexPath], with: .fade)
}
case .delete:
if let indexPath = indexPath {
tableView.deleteRows(at: [indexPath], with: .fade)
}
case .move:
if let indexPath = indexPath, let newIndexPath = newIndexPath {
tableView.moveRow(at: indexPath, to: newIndexPath)
}
case .update:
if let indexPath = indexPath, let cell = tableView.cellForRow(at: indexPath){
cell.textLabel?.text = "Hello"
}
}
}
func controllerDidChangeContent(_ controller: NSFetchedResultsController<NSFetchRequestResult>) {
tableView.endUpdates()
}
}
，但是即使附加了数据集 dataCancer ，上面的命令也不会转换数据集变量 Classification 成为一个因素。它只会在控制台上返回一个因子。
由于您希望模型适合训练数据集，因此可以指定
as.factor(dataCancer$Classification)或在 glm 行代码中使用 as.factor 函数
training_data$Classification <- as.factor(training_data$Classification)
classification_model <- glm(Classification ~ ., data =
training_data, family = binomial)
答案 1 :(得分：1)
您需要将因变量重新编码为0,1，因此请使用以下代码。
library(car)
dataCancer$Classification <- recode(dataCancer$Classification, "1=0; 2=1")
答案 2 :(得分：1)
classification_model = glm（分类〜。，data = training_data，family =二项式） eval（family $ initialize）中的错误：y值必须为0 <= y <= 1
这是因为您的数据包含数字值，而不是因子值。希望你做到了
dataCancer $ Classification <-as.factor（dataCancer $ Classification）
理想情况下，只要是1,0或1,2就没有关系。但是，如果执行上述操作也无济于事，则可以尝试将1,2转换为1,0，然后尝试相同的代码。
第二个错误当然是因为根本没有创建逻辑回归变量。