Select a Subset of Data

When working with data, you may not always need all columns or want to anallyze a specific portion of your dataset. In such situations, you can use the loc[] or iloc[] functions to extract the desired subset of data.

This tutorial uses classic Iris dataset, which can be downloaded here Iris dataset.

1. Select a Subset of Data via iloc[]

The df.iloc[] function is used for integer based indexing for selection by position, allowing access to a group of rows and columns by integer indices.

Usage

.iloc[row_index, column_index]takes two arguments:

• row_index: This argument specifies the rows you want to select based on their integer positions. You can provide a single integer, a slice (e.g., 0:5), or a list of integers to select specific rows.
• column_index: This argument specifies the columns you want to select based on their integer positions. Similar to the rows, you can provide a single integer, a slice, or a list of integers to select specific columns.

Select One Column by Column Index

In this example, we've selected all rows using : and specifically chosen the first column with [0].

Select First 3 Rows and One column

In this example. we've selected the first 3 rows of data with :3 and the second column with [1].

2. Select a Subset of Data via loc[]

Different from iloc[] using positions, loc[] accesses specific rows and columns by labels or a filter rows based on a condition.

Select one column

In this example, we've selected the first 5 rows and the SepalLengthCm column.

It's essential to understand that in Python, data slicing with [:] is typically left-inclusive and right-exclusive. However, with loc[:], it becomes inclusive on both sides. Therefore, to select the first 5 rows, we used loc[:4] or iloc[:5].

Select multiple columns

Select first 10 rows and two columns out from the data using loc[]

3. Filter Data Based on Conditions with loc

Single condition Example

select records where its SepalLengthCm column values are larger than 7.

Multiple conditions example

To select records where its SepalLengthCm are larger than 7 and PetalWidthCm are smaller than 2, we can use the & operator for 'and' to combine the two conditions. It's important to remember that when using and & operator and or | opeartor to combine multiple conditions, we need to use parenthesis () to wrap each condition to ensure the filtering is working properly.

Awesome! In this lesson, you've learned the skills to select specific data subsets using iloc and loc. In our next chapter, we'll delve into data sorting. Stay tuned!