Filter Pandas DataFrame Based on Index

Pandas DataFrame filtering based on index is a fundamental operation for data analysis. The filter() method and boolean indexing provide flexible ways to select specific rows and columns based on their index labels.

Syntax

df.filter(items=None, like=None, regex=None, axis=None)

Parameters

items: List of labels to keep. Returns only rows/columns with matching names.

like: String pattern. Keeps labels containing this substring.

regex: Regular expression pattern for matching labels.

axis: 0 for rows, 1 for columns. Default is None (columns).

Filtering by Numeric Index Positions

Use iloc[] to filter rows by their numeric positions ?

import pandas as pd

# Create a sample DataFrame
data = {'A': [1, 2, 3, 4, 5],
        'B': [6, 7, 8, 9, 10],
        'C': [11, 12, 13, 14, 15]}
df = pd.DataFrame(data)

print("Original DataFrame:")
print(df)
print("\nFiltered rows at positions 1 and 3:")

# Filter rows based on numeric positions
filtered_df = df.iloc[[1, 3]]
print(filtered_df)

Original DataFrame:
   A  B   C
0  1  6  11
1  2  7  12
2  3  8  13
3  4  9  14
4  5 10  15

Filtered rows at positions 1 and 3:
   A  B   C
1  2  7  12
3  4  9  14

Filtering by Custom Index Labels

Filter DataFrames with custom string indices using boolean conditions ?

import pandas as pd

# Create DataFrame with custom index labels
data = {'A': [1, 2, 3, 4, 5],
        'B': [6, 7, 8, 9, 10],
        'C': [11, 12, 13, 14, 15]}
df = pd.DataFrame(data, index=['apple', 'banana', 'orange', 'grape', 'kiwi'])

print("DataFrame with custom indices:")
print(df)
print("\nFiltered rows containing letter 'a':")

# Filter rows where index contains 'a'
filtered_df = df[df.index.str.contains('a')]
print(filtered_df)

DataFrame with custom indices:
        A  B   C
apple   1  6  11
banana  2  7  12
orange  3  8  13
grape   4  9  14
kiwi    5 10  15

Filtered rows containing letter 'a':
        A  B   C
apple   1  6  11
banana  2  7  12
orange  3  8  13
grape   4  9  14

Using filter() Method

The filter() method provides pattern-based filtering for index labels ?

import pandas as pd

# Create DataFrame with descriptive indices
data = {'Sales': [100, 200, 150, 300],
        'Profit': [20, 40, 30, 60]}
df = pd.DataFrame(data, index=['Jan_2023', 'Feb_2023', 'Jan_2024', 'Feb_2024'])

print("Original DataFrame:")
print(df)

# Filter rows containing 'Jan'
jan_data = df.filter(like='Jan', axis=0)
print("\nRows containing 'Jan':")
print(jan_data)

# Filter using regex for 2024 data
data_2024 = df.filter(regex='.*2024$', axis=0)
print("\nRows ending with '2024':")
print(data_2024)

Original DataFrame:
          Sales  Profit
Jan_2023    100      20
Feb_2023    200      40
Jan_2024    150      30
Feb_2024    300      60

Rows containing 'Jan':
          Sales  Profit
Jan_2023    100      20
Jan_2024    150      30

Rows ending with '2024':
          Sales  Profit
Jan_2024    150      30
Feb_2024    300      60

Comparison of Methods

Conclusion

DataFrame index filtering enables precise data selection using iloc[] for positions, filter() for patterns, and boolean indexing for complex conditions. Choose the method that best fits your filtering requirements and index structure.