How to Split a Cell into Multiple Rows in a Pandas Dataframe Based on Comma-Separated Values?

Split Cell into Multiple Rows in Pandas Dataframe

Problem:

You have a Pandas dataframe with orders data where each order contains multiple packages stored as comma-separated strings in the 'package' and 'package_code' columns. You aim to split the packages data and create a new row for each package with its corresponding order details.

Solution:

For Pandas versions >= 0.25:

<code class="python">df.set_index(['order_id', 'order_date']) \
    .apply(lambda x: x.str.split(',').explode()) \
    .reset_index()</code>

For Pandas versions <= 0.24:

<code class="python">df.set_index(['order_date', 'order_id']) \
    .stack() \
    .str.split(',', expand=True) \
    .stack() \
    .unstack(-2) \
    .reset_index(-1, drop=True) \
    .reset_index()

Explanation:

1. Set Index: Set the columns that should remain intact ('order_id' and 'order_date') as the index of the dataframe.

2. Split and Stack:

• For Pandas >= 0.25: Use apply(lambda x: x.str.split(',').explode()) to split the comma-separated 'package' and 'package_code' columns and expand them into multiple rows.
• For Pandas <= 0.24: Use stack(), str.split(), and stack() again to split and stack the data.

3. Unstack and Reset Index:

• Unstack on the second last level (-2) to create new columns for the split package and package code data.
• Reset the index to revert the index change and obtain the final dataframe.

The above is the detailed content of How to Split a Cell into Multiple Rows in a Pandas Dataframe Based on Comma-Separated Values?. For more information, please follow other related articles on the PHP Chinese website!