MLOps Academy · 课时

为提示词建立版本并评估输出

跟踪提示词变化并为响应评分

为提示词建立版本并评估输出 是 CoddyKit 上的免费 MLOps Academy 课时。这是第 2 节课，共 4 节。你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。这是 MLOps Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 MLOps Academy 课程共包含 4 节课。

本课时的部分内容尚未翻译，以英文显示。

Prompts Are Code

A reworded prompt can change behavior overnight. So you treat each prompt as a versioned artifact, with history, review, and the ability to roll back. 📝

Store Prompts, Not Strings

Hard-coded strings scattered in code are impossible to track. Keep prompts in a registry or files so every change is visible and diffable.

prompts/summarize_ticket.v2.txt
prompts/summarize_ticket.v3.txt

Tag Each Version

Give every prompt a clear version id you can pin in code. Then you always know exactly which wording produced a given output.

PROMPT_VERSION = "summarize-v3"
template = load_prompt(PROMPT_VERSION)

Build an Eval Dataset

Collect real inputs with the answers you want. This eval set is your yardstick for judging whether a prompt change actually helped.

cases = [
  {"input": "ticket text...", "expected": "Refund requested"},
]

Score Against References

For tasks with a known answer, compare output to the expected text. A simple metric like exact match or F1 gives you a fast pass-fail signal.

Use an LLM as Judge

For open-ended text, ask another model to grade the answer against a rubric. This LLM-as-judge pattern scales scoring beyond hand-written rules.

judge_prompt = "Rate 1-5 how well the summary captures the ticket:\n{output}"

Watch the Judge Itself

Judges can be biased or inconsistent. You keep them honest by spot-checking scores against a few human ratings on the same cases.

Run Evals on Every Change

Before shipping a new prompt, run it across the whole eval set. You only promote it if the score holds or improves, never on a hunch.

old = run_eval("summarize-v2")
new = run_eval("summarize-v3")
assert new.score >= old.score

Track Regressions

A prompt that fixes one case can break another. Comparing per-case results catches these regressions before users ever see them.

Frameworks Help

Tools like promptfoo or OpenAI Evals run datasets, call models, and report scores for you. A framework turns ad-hoc testing into a repeatable suite.

Pin the Winner

Once a version wins on your evals, pin it in production and archive the rest. You now have a clear, auditable trail from prompt to result.

Quick Check

You need to score open-ended summaries with no single correct answer. What fits best?

Recap

You learned to version prompts, build an eval set, score with metrics or an LLM judge, and gate releases on results. No more guessing! 🎉

常见问题解答

「为提示词建立版本并评估输出」课时是免费的吗？

是的 — 「为提示词建立版本并评估输出」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 MLOps Academy 课程的其余内容，请升级到 CoddyKit PRO。 MLOps Academy 课程共包含 4 节课。

「为提示词建立版本并评估输出」这节课中我会学到什么？

跟踪提示词变化并为响应评分你通过在浏览器中直接运行的动手代码来练习 MLOps Academy，全天候 AI 导师会在你学习这节课的过程中回答你的问题。

学习 MLOps Academy 需要有经验吗？

无需任何先前经验。CoddyKit 上的 MLOps Academy 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。这是第 2 节课，共 4 节。

「为提示词建立版本并评估输出」课时需要多长时间？

大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。

我能在这节 MLOps Academy 课中编写并运行代码吗？

能。每节 MLOps Academy 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。

此课程中的所有课时

← 返回 MLOps Academy