When to use .loc and when to use .iloc

 Absolutely! Let’s break it down clearly with a comparison between .loc and .iloc, and then explain why they are often used when filling empty rows in a DataFrame.

---

1. .loc vs .iloc in Pandas

Feature	.loc	.iloc
Selection type	Label-based (row/column names)	Integer position-based (row/column indices)
Includes the endpoint in slices?	Yes	No
Accepts boolean masks?	Yes	Yes
Examples	df.loc[0, 'Age'] → value in row label 0, column ‘Age’df.loc[:, ['Name','City']] → select columns by names	df.iloc[0, 1] → value in row 0, column 1df.iloc[:, 0:2] → select first 2 columns by position
When used	When you know labels	When you know positions

---

2. Why they are used in filling empty rows (NaN)

When a DataFrame has missing values (NaN), you often want to fill them with specific values, either for a whole column or a specific row. Both .loc and .iloc are helpful:

• .loc → Use it if you want to fill based on row or column labels.

• .iloc → Use it if you want to fill based on row or column positions.

Example 1: Using .loc

import pandas as pd
import numpy as np

df = pd.DataFrame({
    'Name': ['Alice', 'Bob', None],
    'Age': [25, np.nan, 35]
})

# Fill missing name for row label 2
df.loc[2, 'Name'] = 'Charlie'

# Fill missing Age for all rows where Age is NaN
df.loc[df['Age'].isna(), 'Age'] = 30

print(df)

Output:

      Name   Age
0    Alice  25.0
1      Bob  30.0
2  Charlie  35.0

---

Example 2: Using .iloc

# Fill the missing Age using row/column position
df.iloc[1, 1] = 30  # Row 1, Column 1 → Age

• Handy when labels are unknown or you just want quick positional indexing.

• Especially useful in loops, numeric operations, or filling missing values systematically.

---

✅ Summary:

• Use .loc for clarity and readability when you know labels.

• Use .iloc when working with positions, often in automated or iterative filling of missing data.

• Both allow direct assignment of missing values, making them essential tools for data cleaning.