<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Recommended Python UDFs for On-Demand Feature Computation in Databricks in Machine Learning</title>
    <link>https://community.databricks.com/t5/machine-learning/recommended-python-udfs-for-on-demand-feature-computation-in/m-p/154226#M4606</link>
    <description>&lt;P&gt;&lt;SPAN&gt;The Databricks documentation page on on-demand feature computation (&lt;/SPAN&gt;&lt;A class="" title="/aws/en/machine-learning/feature-store/on-demand-features#what-are-on-demand-features" href="https://docs.databricks.com/aws/en/machine-learning/feature-store/on-demand-features#what-are-on-demand-features" target="_blank" rel="noopener"&gt;https://docs.databricks.com/aws/en/machine-learning/feature-store/on-demand-features#what-are-on-demand-features&lt;/A&gt;&lt;SPAN&gt;) mentions using Python UDFs for computing on-demand features. What types of Python UDFs are supported for this purpose? Are only scalar UDFs allowed, as shown in the examples on that page, or can other types of UDFs (such as pandas UDFs) also be used? What is the recommended approach for implementing on-demand feature computation in Databricks Feature Store?&lt;/SPAN&gt;&lt;/P&gt;</description>
    <pubDate>Sun, 12 Apr 2026 20:44:37 GMT</pubDate>
    <dc:creator>nj21</dc:creator>
    <dc:date>2026-04-12T20:44:37Z</dc:date>
    <item>
      <title>Recommended Python UDFs for On-Demand Feature Computation in Databricks</title>
      <link>https://community.databricks.com/t5/machine-learning/recommended-python-udfs-for-on-demand-feature-computation-in/m-p/154226#M4606</link>
      <description>&lt;P&gt;&lt;SPAN&gt;The Databricks documentation page on on-demand feature computation (&lt;/SPAN&gt;&lt;A class="" title="/aws/en/machine-learning/feature-store/on-demand-features#what-are-on-demand-features" href="https://docs.databricks.com/aws/en/machine-learning/feature-store/on-demand-features#what-are-on-demand-features" target="_blank" rel="noopener"&gt;https://docs.databricks.com/aws/en/machine-learning/feature-store/on-demand-features#what-are-on-demand-features&lt;/A&gt;&lt;SPAN&gt;) mentions using Python UDFs for computing on-demand features. What types of Python UDFs are supported for this purpose? Are only scalar UDFs allowed, as shown in the examples on that page, or can other types of UDFs (such as pandas UDFs) also be used? What is the recommended approach for implementing on-demand feature computation in Databricks Feature Store?&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Sun, 12 Apr 2026 20:44:37 GMT</pubDate>
      <guid>https://community.databricks.com/t5/machine-learning/recommended-python-udfs-for-on-demand-feature-computation-in/m-p/154226#M4606</guid>
      <dc:creator>nj21</dc:creator>
      <dc:date>2026-04-12T20:44:37Z</dc:date>
    </item>
  </channel>
</rss>

