How to Fill Missing Data with sklearn’s SimpleImputer and KNNImputer

scikit-learn provides several estimators for missing‑value imputation, such as SimpleImputer, IterativeImputer, and KNNImputer.

Univariate Imputation with SimpleImputer

It performs univariate imputation using simple strategies, replacing missing values with column‑wise descriptive statistics (e.g., mean, median, most frequent) or a constant.

Example

<code>import numpy as np
from sklearn.impute import SimpleImputer
imp_mean = SimpleImputer(missing_values=np.nan, strategy='mean')
imp_mean.fit([[7, 2, 3], [4, np.nan, 6], [10, 5, 9]])

X = [[np.nan, 2, 3], [4, np.nan, 6], [10, np.nan, 9]]
print(imp_mean.transform(X))
</code>

The result is

<code>[[ 7.   2.   3.]
 [ 4.   3.5  6.]
 [10.   3.5  9.]]
</code>

The available imputation strategies include

mean

median

most_frequent

constant

K‑Nearest Neighbors Imputation with KNNImputer

It uses a K‑nearest neighbors model for imputation.

<code>import numpy as np
from sklearn.impute import KNNImputer
X = [[1, 2, np.nan], [3, 4, 3], [np.nan, 6, 5], [8, 8, 7]]
imputer = KNNImputer(n_neighbors=2)
imputer.fit_transform(X)
</code>

The result is:

<code>array([[1. , 2. , 4. ],
       [3. , 4. , 3. ],
       [5.5, 6. , 5. ],
       [8. , 8. , 7. ]])
</code>

References

https://scikit-learn.org/stable/modules/generated/sklearn.impute.SimpleImputer.html#sklearn.impute.SimpleImputer