PyInns - Home

More Unpacking in Loops in Python for Data Science – Best Practices 2026

Advanced unpacking inside loops is a powerful Pythonic skill that makes data science code dramatically cleaner and more readable. Once you master enumerate(), zip(), *rest, and nested unpacking, your feature engineering, result processing, and configuration handling become much more elegant.

TL;DR — Advanced Unpacking Patterns

for idx, (a, b) in enumerate(zip(...)) → index + paired values
for key, (val1, val2) in data.items() → nested unpacking
for first, *rest in records → head/tail splitting
for **kwargs in config_list → dictionary unpacking in loops

1. Enumerate + zip() – The Most Common Power Combo

features = ["amount", "quantity", "profit", "region"]
importance = [0.42, 0.31, 0.18, 0.09]

for rank, (feature, score) in enumerate(zip(features, importance), start=1):
    print(f"#{rank:2d} {feature:12} importance: {score:.4f}")

2. Nested Unpacking in Loops

results = [
    ("random_forest", (200, 10, 42)),   # model, (n_estimators, max_depth, random_state)
    ("xgboost", (300, 8, 123))
]

for model_name, (n_est, depth, seed) in results:
    print(f"{model_name:15} → n_estimators={n_est}, max_depth={depth}, seed={seed}")

3. Head/Tail Unpacking with *rest

feature_list = ["customer_id", "order_date", "amount", "quantity", "profit", "region", "category"]

primary, *metrics, geo = feature_list

print(f"Primary: {primary}")
print(f"Metrics: {metrics}")
print(f"Geo: {geo}")

4. Real-World Data Science Examples

import pandas as pd

df = pd.read_csv("sales_data.csv")

# Example 1: Unpacking column groups in a loop
numeric_cols = [col for col in df.columns if df[col].dtype in ["int64", "float64"]]
cat_cols = [col for col in df.columns if df[col].dtype == "object"]

for group_name, cols in [("Numeric", numeric_cols), ("Categorical", cat_cols)]:
    print(f"{group_name} features ({len(cols)}): {cols[:5]}...")

# Example 2: Processing model hyperparameter grids
hyperparams = [
    ("rf", {"n_estimators": 200, "max_depth": 10}),
    ("xgb", {"n_estimators": 300, "learning_rate": 0.1})
]

for model_type, params in hyperparams:
    print(f"Training {model_type} with {params}")

5. Best Practices in 2026

Always unpack directly in the for statement for maximum readability
Use enumerate(zip(...)) whenever you need both rank and paired values
Use *rest to cleanly separate primary columns from the rest
Keep unpacking simple — avoid more than two levels of nesting
Combine with zip() and dict.items() for the cleanest code

Conclusion

Advanced unpacking in loops is one of the hallmarks of professional Python data science code in 2026. It eliminates manual indexing, reduces boilerplate, and makes your feature processing, model configuration, and result handling far more elegant. Master enumerate(zip()), nested unpacking, and *rest, and your code will feel significantly more Pythonic and maintainable.

Next steps:

Review your current loops and refactor any that use manual counters or index access into clean unpacking patterns