Handling Missing Values

Posted 2020-12-29 hotsnow

1) A Simple Option: Drop Columns with Missing Values

如果这些列具有有用信息（在未丢失的位置），则在删除列时，模型将失去对此信息的访问权限。 此外，如果您的测试数据在您的训练数据没有的地方缺少值，则会导致错误。

data_without_missing_values = original_data.dropna(axis=1)

#同时操作tran和test部分
cols_with_missing = [col for col in original_data.columns 
                                 if original_data[col].isnull().any()]
redued_original_data = original_data.drop(cols_with_missing, axis=1)
reduced_test_data = test_data.drop(cols_with_missing, axis=1)

 

2) A Better Option: Imputation

默认行为填写了插补的平均值。 统计学家已经研究了更复杂的策略，但是一旦将结果插入复杂的机器学习模型，那些复杂的策略通常没有任何好处。

关于Imputation的一个（很多）好处是它可以包含在scikit-learn Pipeline中。 管道简化了模型构建，模型验证和模型部署。

from sklearn.impute import SimpleImputer
my_imputer = SimpleImputer()
data_with_imputed_values = my_imputer.fit_transform(original_data)

 

3) An Extension To Imputation

估算是标准方法，通常效果很好。 但是，估算值可能系统地高于或低于其实际值（未在数据集中收集）。 或者具有缺失值的行可能以某种其他方式看来是唯一的。 在这种情况下，您的模型会通过考虑最初缺少哪些值来做出更好的预测。

# make copy to avoid changing original data (when Imputing)
new_data = original_data.copy()

# make new columns indicating what will be imputed
cols_with_missing = (col for col in new_data.columns 
                                 if new_data[col].isnull().any())
for col in cols_with_missing:
    new_data[col + ‘_was_missing‘] = new_data[col].isnull()

# Imputation
my_imputer = SimpleImputer()
new_data = pd.DataFrame(my_imputer.fit_transform(new_data))
new_data.columns = original_data.columns

 

 

Example (Comparing All Solutions)

import pandas as pd

# Load data
melb_data = pd.read_csv(‘../input/melbourne-housing-snapshot/melb_data.csv‘)

from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_absolute_error
from sklearn.model_selection import train_test_split

melb_target = melb_data.Price
melb_predictors = melb_data.drop([‘Price‘], axis=1)

# For the sake of keeping the example simple, we‘ll use only numeric predictors. 
melb_numeric_predictors = melb_predictors.select_dtypes(exclude=[‘object‘])

from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_absolute_error
from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(melb_numeric_predictors, 
                                                    melb_target,
                                                    train_size=0.7, 
                                                    test_size=0.3, 
                                                    random_state=0)

def score_dataset(X_train, X_test, y_train, y_test):
    model = RandomForestRegressor()
    model.fit(X_train, y_train)
    preds = model.predict(X_test)
    return mean_absolute_error(y_test, preds)


# Get Model Score from Dropping Columns with Missing Values
cols_with_missing = [col for col in X_train.columns 
                                 if X_train[col].isnull().any()]
reduced_X_train = X_train.drop(cols_with_missing, axis=1)
reduced_X_test  = X_test.drop(cols_with_missing, axis=1)
print("Mean Absolute Error from dropping columns with Missing Values:")
print(score_dataset(reduced_X_train, reduced_X_test, y_train, y_test))


# Get Model Score from Imputation
from sklearn.impute import SimpleImputer

my_imputer = SimpleImputer()
imputed_X_train = my_imputer.fit_transform(X_train)
imputed_X_test = my_imputer.transform(X_test)
print("Mean Absolute Error from Imputation:")
print(score_dataset(imputed_X_train, imputed_X_test, y_train, y_test))


# Get Score from Imputation with Extra Columns Showing What Was Imputed
imputed_X_train_plus = X_train.copy()
imputed_X_test_plus = X_test.copy()

cols_with_missing = (col for col in X_train.columns 
                                 if X_train[col].isnull().any())
for col in cols_with_missing:
    imputed_X_train_plus[col + ‘_was_missing‘] = imputed_X_train_plus[col].isnull()
    imputed_X_test_plus[col + ‘_was_missing‘] = imputed_X_test_plus[col].isnull()

# Imputation
my_imputer = SimpleImputer()
imputed_X_train_plus = my_imputer.fit_transform(imputed_X_train_plus)
imputed_X_test_plus = my_imputer.transform(imputed_X_test_plus)

print("Mean Absolute Error from Imputation while Track What Was Imputed:")
print(score_dataset(imputed_X_train_plus, imputed_X_test_plus, y_train, y_test))