叠加线性层
从输入经过隐藏层到达输出
叠加线性层 是 CoddyKit 上的免费 Deep Learning 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 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。
One Layer, One Transform
A single nn.Linear layer maps inputs to outputs with a weighted sum plus a bias. It reshapes data from one size to another.
layer = nn.Linear(in_features=4, out_features=8)Input to Hidden
Your first layer takes the raw input features and projects them into a hidden space, often a larger or smaller width.
self.fc1 = nn.Linear(4, 16) # 4 inputs -> 16 hiddenHidden to Output
A later layer maps the hidden features down to your final output size, like the number of classes you want to predict.
self.fc2 = nn.Linear(16, 3) # 16 hidden -> 3 classesChaining Sizes
Each layer's out_features must equal the next layer's in_features. These matching dimensions form a clean pipeline.
Add a Nonlinearity Between
Stacking bare linear layers stays linear, so place an activation like ReLU between them to unlock real depth.
x = F.relu(self.fc1(x))
x = self.fc2(x)Width Versus Depth
More neurons per layer means more width; more layers means more depth. Both add capacity in different ways.
The Hidden Dimension Choice
Picking the hidden size is a design call. Bigger learns more patterns but costs memory and risks overfitting.
Data Flows Top to Bottom
In forward, the input passes through layer one, then the activation, then layer two. Each step transforms the tensor in turn.
def forward(self, x):
x = F.relu(self.fc1(x))
return self.fc2(x)A Two-Layer MLP
Input, hidden, output with a nonlinearity is the classic multilayer perceptron. It is the workhorse of feedforward nets.
Parameters Grow With Layers
Every linear layer adds its own weights and bias. Stacking more layers steadily grows the model's parameter count.
Start Small, Then Grow
Begin with one hidden layer and expand only if needed. A lean stack trains fast and is easy to debug. 🌱
Quick Check
Recall what must sit between two linear layers to make stacking worthwhile.
Recap
Stack nn.Linear layers so each output size feeds the next input size, with an activation between them. That builds a real MLP. 🎯
常见问题解答
「叠加线性层」课时是免费的吗？
是的 — 「叠加线性层」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。
「叠加线性层」这节课中我会学到什么？
从输入经过隐藏层到达输出 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 2 节课，共 4 节。
「叠加线性层」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。