如何使用 cross_val_score 解决连续错误？

Question

我对机器学习很陌生，在过去的两天里，我一直在努力摆脱 Unknown label type: 'continuous' 错误。

我的代码：将 numpy 导入为 np

import pandas as pd
from sklearn.model_selection import train_test_split  
from sklearn.preprocessing import StandardScaler  
from sklearn.ensemble import RandomForestClassifier  
from sklearn.model_selection import cross_val_score  

dataset = pd.read_csv(r'allData.csv', sep=',')  
X = dataset.iloc[:, 1:3].values  
y = dataset.iloc[:, 4].values  

train_features, test_features, train_lables, test_lables = train_test_split(X, y, test_size=10, random_state=10)  

feature_scaler = StandardScaler()  
train_features = feature_scaler.fit_transform(train_features)  
test_features = feature_scaler.transform(test_features)  

classifier = RandomForestClassifier(n_estimators=300, random_state=10)  
all_accuracies = cross_val_score(estimator=classifier, X=train_features, y=train_lables, cv="warn")  
#all_accuracies = cross_val_score(estimator=classifier, X=train_features, y=train_lables, cv=3)
#print(all_accuracies)  

错误出现在cross_val_score 部分，我不明白为什么会收到Unknown label type: 'continuous' 错误。

任何帮助将不胜感激。

---

如果有帮助，我的数据都是数字，有 4 列 300 行。

Answer 1

您正在使用RandomForestClassifier，同时具有连续输出。如果您要解决的问题是回归，那么您应该使用RandomForestRegressor。