Scikit学习中的OneHotEncoder和KNNImpute之间的循环循环 [英] Cyclical Loop Between OneHotEncoder and KNNImpute in Scikit-learn

问题描述

我正在使用一个非常简单的数据集.它在分类和数字功能中都缺少一些值.因此，我正在尝试使用sklearn.preprocessing.KNNImpute获得我能得到的最准确的插补.但是，当我运行以下代码时:

I'm working with a really simple dataset. It has some missing values, both in categorical and numeric features. Because of this, I'm trying to use sklearn.preprocessing.KNNImpute to get the most accurate imputation I can. However, when I run the following code:

imputer = KNNImputer(n_neighbors=120)

imputer.fit_transform(x_train)

我收到错误: ValueError:无法将字符串转换为float:'Private'

这很有意义，显然不能处理分类数据.但是当我尝试使用以下命令运行OneHotEncoder时:

That makes sense, it obviously can't handle categorical data. But when I try to run OneHotEncoder with:

encoder = OneHotEncoder(drop="first")

encoder.fit_transform(x_train[categorical_features])

它引发错误: ValueError:输入包含NaN

即使是分类数据，我还是更愿意使用 KNNImpute ，因为如果我只使用 ColumnTransform 并用数字进行插补，我会失去一些准确性.和分类数据分开.有什么方法可以使 OneHotEncoder 忽略这些缺失的值?如果不是，使用 ColumnTransform 或更简单的方法是解决此问题的更好方法吗?

I'd prefer to use KNNImpute even with the categorical data as I feel like I'd be losing some accuracy if I just use a ColumnTransform and impute with numeric and categorical data seperately. Is there any way to get OneHotEncoder to ignore these missing values? If not, is using ColumnTransform or a simpler imputer a better way of tackling this problem?

预先感谢

推荐答案

在处理 OneHotEncoder 上的缺失值方面存在未解决的问题/PR，但目前尚不清楚这些选项是什么.在此期间，这是一种手动方法.

There are open issues/PRs to handle missing values on OneHotEncoder, but it's not clear yet what the options would be. In the interim, here's a manual approach.

• 用熊猫或 SimpleImputer 用字符串"missing"填充分类缺失.
• 然后使用 OneHotEncoder .
• 使用一次性编码器的 get_feature_names 来标识与每个原始功能相对应的列，尤其是缺少"的列.指示器.
对于每行和每个原始分类特征，当1为缺失"时，则为1.列，用 np.nan 替换 0；然后删除缺少的指标列.
• 现在应将所有内容设置为运行 KNNImputer .
• 最后，如果需要，可以对插补的分类编码列进行后处理.(简单地四舍五入可能会为您提供分类特征的全零行，但我不认为使用 KNNImputer 可以连续获得多个1.您可以使用argmax代替，以便准确返回一个1.)

• Fill categorical missings with pandas or SimpleImputer with the string "missing".
• Use OneHotEncoder then.
• Use the one-hot encoder's get_feature_names to identify the columns corresponding to each original feature, and in particular the "missing" indicator.
• For each row and each original categorical feature, when the 1 is in the "missing" column, replace the 0's with np.nan; then delete the missing indicator column.
• Now everything should be set up to run KNNImputer.
• Finally, if desired, postprocess the imputed categorical-encoding columns. (Simply rounding might get you an all-zeros row for a categorical feature, but I don't think with KNNImputer you could get more than one 1 in a row. You could argmax instead to get back exactly one 1.)

这篇关于Scikit学习中的OneHotEncoder和KNNImpute之间的循环循环的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！