20 items
Mean Imputation With Sentinel Values
Preventing Data Leakage in Imputation
Scaling Train and Test Data
Encoding High-Cardinality Categorical Features
Feature Engineering & Encoding Essentials
Dropping Missing Rows in Pandas
Handling Missing Data in Pandas
Filtering a DataFrame with two conditions
Filtering Rows in Pandas
Handling Missing Data & Duplicates
Pandas: Load, Filter & Select Data
Pandas: Loading, Filtering & Selecting Data
Encoding, Scaling & Transformations
Handling Missing Data & Outliers
Detecting Outliers with IQR
One-Hot Encoding a Category Column
Learn why imputation statistics must be computed from training data only to avoid leakage.
You are filling missing values in a numeric feature using the mean. To avoid data leakage in a train/test split, how should the mean be computed and applied?