采样调度与引导
引导生成结果符合提示词
采样调度与引导 是 CoddyKit 上的免费 Deep Learning Academy 课时。 这是第 3 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 Deep Learning Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Deep Learning 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。
Sampling Is the Payoff
Training gives you a denoiser. Sampling is where you cash it in, running the reverse steps to turn noise into a finished image.
Steps Cost Time
Each reverse step calls the U-Net once. More steps usually means crisper results but a slower generation. ⏳
The DDPM Sampler
The original DDPM sampler is faithful but slow, often needing hundreds or even a thousand tiny denoising steps.
Faster with DDIM
DDIM skips many steps while keeping quality high, so you can sample in twenty or fifty steps instead of a thousand.
Choosing a Schedule
A sampling schedule decides which timesteps you actually visit. Fewer, well-chosen steps trade a little detail for big speed.
scheduler.set_timesteps(num_inference_steps=30)Plain Generation Wanders
Left alone, the model produces a valid image but you cannot steer it. You get something plausible, not necessarily what you asked for.
Conditioning on a Prompt
To get control, you condition the U-Net on a prompt, often a text embedding, so it aims toward your description.
Classifier-Free Guidance
Classifier-free guidance runs the model twice, with and without the prompt, then pushes the result toward the prompted version.
The Guidance Scale
A guidance scale sets how hard to push. Higher values follow the prompt more closely but can hurt realism if overdone.
image = pipe(prompt, guidance_scale=7.5)Finding the Sweet Spot
Too low and the image ignores you; too high and it looks harsh. A scale around seven is a common sweet spot to start from.
Two Knobs, Big Impact
Steps control speed and detail; guidance controls how closely you obey the prompt. Tuning both is how you dial in your output. 🎛️
Quick Check
What does raising the guidance scale do?
Recap
Schedules like DDIM cut steps for speed, and guidance steers toward your prompt. Tune steps and guidance scale to balance quality and control. 🎉
常见问题解答
「采样调度与引导」课时是免费的吗？
是的 — 「采样调度与引导」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 Deep Learning Academy 课程的其余内容，请升级到 CoddyKit PRO。 Deep Learning Academy 课程共包含 4 节课。
「采样调度与引导」这节课中我会学到什么？
引导生成结果符合提示词 你通过在浏览器中直接运行的动手代码来练习 Deep Learning 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 Deep Learning Academy 需要有经验吗？
无需任何先前经验。CoddyKit 上的 Deep Learning 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 3 节课，共 4 节。
「采样调度与引导」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 Deep Learning Academy 课中编写并运行代码吗？
能。每节 Deep Learning 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。