LSTM 与 GRU 门控
记住长距离依赖关系
LSTM 与 GRU 门控 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。
The Long-Memory Problem
Vanilla RNNs forget early clues over long sequences. Gated cells fix this by deciding what to keep, update, or throw away.
Meet the LSTM
The LSTM adds a separate cell state, a memory highway that runs straight across time with only small, controlled edits.
Gates Are Soft Switches
A gate is a sigmoid layer outputting values from 0 to 1. Zero blocks information, one lets it pass, and in between mixes the two.
gate = torch.sigmoid(W @ x + U @ h)The Forget Gate
The forget gate looks at the input and memory and chooses which parts of the old cell state to erase before adding anything new.
The Input Gate
The input gate decides how much of the fresh candidate information should be written into the cell state at this step.
The Output Gate
The output gate controls how much of the updated cell state becomes the visible hidden state passed to the next step. 🚪
Why Gates Help Gradients
Because the cell state changes gently, gradients flow back across many steps without vanishing, so the model learns long-range patterns.
Meet the GRU
The GRU is a lighter cousin: it merges gates and drops the separate cell state, giving similar power with fewer parameters.
GRU's Two Gates
A GRU uses just two gates, a reset gate and an update gate, to balance old memory against new input each step.
Drop-In in PyTorch
Both are one-liners in PyTorch. Swap nn.LSTM or nn.GRU for nn.RNN and keep almost the same training code.
lstm = nn.LSTM(input_size=10, hidden_size=20)Which to Choose?
GRUs are faster and often match LSTMs; LSTMs can edge ahead on the hardest long sequences. Try both and let your validation score decide.
Quick Check
What is the job of the forget gate in an LSTM?
Recap
LSTMs and GRUs use gates to control memory, letting gradients survive long sequences and capturing far-apart dependencies. ✅
常见问题解答
「LSTM 与 GRU 门控」课时是免费的吗？
是的 — 「LSTM 与 GRU 门控」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。
「LSTM 与 GRU 门控」这节课中我会学到什么？
记住长距离依赖关系 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。
「LSTM 与 GRU 门控」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。