Scikit-learn: Standardizing Features
Intermediate
The problem, from first principles
Many models, such as linear models and neural networks, train better when features are on similar scales. Standardization centers each feature and divides by its spread, so no feature dominates just because of its units.
From theory to code
Implement standard_scale(X), which centers each column and scales it to unit standard deviation.
Constraints
- A constant column has standard deviation 0; return zeros for it.
Hints
Hint 1
Compute the column means and standard deviations, guard zero deviations, then transform.
Keywords
research-papersclassical-mltoolingpreprocessing
Python
Run your code to see results
No results yet
Run your code to check it against the examples, or submit to run every test.