Isolate Numerical and Categorical Columns with select_dtypes in Pandas

Question

Given a mixed DataFrame of numerical, object, and datetime columns, select all numerical columns (int, float) for statistical scaling.

Accepted Answer

Optimal Solution:
import pandas as pd

# Select all integer and floating point columns
numeric_df = df.select_dtypes(include=['number'])

# Exclude object/string columns
# non_text_df = df.select_dtypes(exclude=['object'])

Technical Explanation:
`df.select_dtypes(include=['number'])` dynamically inspects column dtypes and returns a subset matching numeric numpy types.