我的标准缩放器输出 ValueError: could not convert string to float答案

【问题标题】：my standardscaler outputs ValueError: could not convert string to float我的标准缩放器输出 ValueError: could not convert string to float
【发布时间】：2021-07-21 21:43:32
【问题描述】：

我作为 Python 和机器学习初学者正在做一个项目，并且遇到了 Titanic 数据集。将我的数据集拆分为训练和测试后，我想使用 StandardScaler 规范化 x_train，但这不断出现：

ValueError：无法将字符串转换为浮点数：'PassengerId'

这是我的代码

    feature =df[['PassengerId', 'PClass', 'Age', 'SibSp', 'Parch']].values
    target = df[['Survived']].values
    x_train, x_test, y_train, y_test = train_test_split(feature, target, test_size=0.2)
    from sklearn.preprocessing import StandardScaler, OneHotEncoder
    scaler = StandardScaler().fit(x_train)
    x_train = scaler.transform(x_train)
    x_test = scaler.transform(x_test)

我该如何解决这个问题？

【问题讨论】：

标签： python pandas machine-learning scikit-learn

【解决方案1】：

# it's a good idea to exclude PassengerId, since it's probabaly not a predictive feature
# also convert all the values to float
feature =df[['PClass', 'Age', 'SibSp', 'Parch']].astype('float').values
target = df[['Survived']].values
    
x_train, x_test, y_train, y_test = train_test_split(feature, target, test_size=0.2)
from sklearn.preprocessing import StandardScaler, OneHotEncoder
scaler = StandardScaler().fit(x_train)
x_train = scaler.transform(x_train)
x_test = scaler.transform(x_test)

【讨论】：

现在它给了我 ValueError: could not convert string to float: 'Pclass'