scikit-learn

What it is

The most-used ML library on Earth. Linear and tree models, clustering, dimensionality reduction, model selection, preprocessing — all behind one consistent fit/predict API.

How Vaaani uses it

• Tabular baselines before reaching for boosting or deep learning
• Building reproducible Pipeline objects (preprocess + model)
• Cross-validation and hyperparameter search with GridSearchCV
• Quick MVPs that ship to prod and stay there for years

Why it makes the cut

Most business problems don't need deep learning. scikit-learn ships in days and is good enough — and when it isn't, it's a clean baseline to beat.

Sample code

from sklearn.ensemble import RandomForestClassifier
from sklearn.pipeline import Pipeline

pipe = Pipeline([
    ("scaler", StandardScaler()),
    ("clf", RandomForestClassifier(n_estimators=200))
])
pipe.fit(X_train, y_train)