<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Baselines on Programmer.ie</title>
    <link>http://programmer.ie/tags/baselines/</link>
    <description>Recent content in Baselines on Programmer.ie</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Fri, 28 Aug 2026 10:35:00 +0100</lastBuildDate>
    <atom:link href="http://programmer.ie/tags/baselines/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>DSPy From First Principles 08: You Cannot Optimize What You Cannot Measure</title>
      <link>http://programmer.ie/books/dspy-from-first-principles/08-chapter/</link>
      <pubDate>Fri, 28 Aug 2026 10:35:00 +0100</pubDate>
      <guid>http://programmer.ie/books/dspy-from-first-principles/08-chapter/</guid>
      <description>&lt;h1 id=&#34;you-cannot-optimize-what-you-cannot-measure&#34;&gt;You Cannot Optimize What You Cannot Measure&lt;/h1&gt;&#xA;&lt;p&gt;Chapter 7 turned examples into data. We now have rows, labels, provenance, and splits.&lt;/p&gt;&#xA;&lt;p&gt;That does not yet give us evidence. It gives us the material from which evidence can be produced.&lt;/p&gt;&#xA;&lt;p&gt;The next step is a baseline evaluation:&lt;/p&gt;&#xA;&lt;div class=&#34;highlight&#34;&gt;&lt;pre tabindex=&#34;0&#34; style=&#34;color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;&#34;&gt;&lt;code class=&#34;language-text&#34; data-lang=&#34;text&#34;&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;frozen program&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;      +&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;frozen cases&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;      +&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;metric&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;      ↓&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;per-case results&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;      ↓&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;aggregate result&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;      ↓&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;failure inspection&#xA;&lt;/span&gt;&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;p&gt;No optimization happens in this chapter. Evaluation answers one question:&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
