批归一化与层归一化
稳定激活值，以便训练更深的网络
批归一化与层归一化 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。
Why Normalize Activations
As signals flow through layers, their scale drifts and training slows. Normalization keeps activations in a stable, well-behaved range. ⚖️
Batch Norm in One Line
Batch normalization rescales each feature using the mean and variance computed across the current mini-batch.
Add Batch Norm
You drop in a batch norm layer after a linear or conv layer. Use nn.BatchNorm1d for dense features and BatchNorm2d for images.
self.bn = nn.BatchNorm1d(128)
x = torch.relu(self.bn(self.fc(x)))Learnable Scale and Shift
Batch norm adds two trainable parameters, gamma and beta, so the network can rescale and shift the normalized output as it needs.
It Lets You Train Deeper
By steadying activations, batch norm allows higher learning rates and deeper networks that would otherwise be hard to train.
Train vs Eval Stats
During training, batch norm uses batch statistics; at eval it switches to running averages, so call model.eval() before testing.
Batch Size Matters
Batch norm depends on the batch. With very small batches the statistics get noisy and the benefit can disappear.
Enter Layer Norm
Layer normalization normalizes across the features of a single sample, so it does not depend on the batch at all.
Add Layer Norm
Layer norm shines in transformers and RNNs. You add it with nn.LayerNorm, passing the size of the features to normalize.
self.ln = nn.LayerNorm(256)
x = self.ln(x)Same Behavior Both Modes
Because layer norm uses per-sample statistics, it behaves the same in training and eval, which suits variable batch sizes.
Which One to Pick
Reach for batch norm in conv vision models and layer norm in sequence and transformer models, where batches and lengths vary.
Quick Check
Recall the key difference between the two normalization styles.
Recap
You compared normalization: batch norm uses batch statistics and helps CNNs, while layer norm works per-sample and powers transformers. 🚀
常见问题解答
「批归一化与层归一化」课时是免费的吗？
是的 — 「批归一化与层归一化」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。
「批归一化与层归一化」这节课中我会学到什么？
稳定激活值，以便训练更深的网络 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。
「批归一化与层归一化」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。