python
23 lines · 1 tab
Dr. Elena Vasquez
Apr 2026
1 tab
from sklearn.compose import ColumnTransformer
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.impute import SimpleImputer
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn.ensemble import RandomForestClassifier
preprocessor = ColumnTransformer([
('num', Pipeline([
('imputer', SimpleImputer(strategy='median')),
('scaler', StandardScaler()),
]), ['age', 'income', 'days_since_last_login']),
('cat', Pipeline([
('imputer', SimpleImputer(strategy='most_frequent')),
('encoder', OneHotEncoder(handle_unknown='ignore')),
]), ['country', 'plan_tier']),
('text', TfidfVectorizer(max_features=5000, ngram_range=(1, 2)), 'support_ticket_text'),
])
model = Pipeline([
('preprocessor', preprocessor),
('classifier', RandomForestClassifier(n_estimators=300, random_state=42, n_jobs=-1)),
])
1 file · python
Explain with highlit
I push nearly all preprocessing into a Pipeline so training and inference paths share exactly the same logic. ColumnTransformer is the workhorse here because real-world tables mix numeric, categorical, boolean, and text fields. It gives you reproducibility without having to manage fragile pre-fit artifacts by hand.
Share this code
Here's the card — post it anywhere.
