| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| 数据操作 |
|
70 | 70059 | February 6, 2026 |
| 多层感知机的从零实现 |
|
2 | 3028 | February 2, 2026 |
| 线性回归的从零开始实现 |
|
159 | 75848 | February 2, 2026 |
| 注意力汇聚：Nadaraya-Watson 核回归 |
|
54 | 21208 | January 28, 2026 |
| 关于可能的报错 |
|
0 | 559 | January 24, 2026 |
| 语言模型和数据集 |
|
39 | 17341 | January 20, 2026 |
| 序列模型 |
|
37 | 15353 | January 18, 2026 |
| 层和块 |
|
46 | 18760 | January 17, 2026 |
| Gaussian Process Priors |
|
2 | 1924 | January 15, 2026 |
| Object-Oriented Design for Implementation |
|
2 | 2736 | January 13, 2026 |
| Softmax回归 |
|
79 | 60867 | January 11, 2026 |
| Colab安装d2l报错 |
|
0 | 603 | January 6, 2026 |
| 2026外网梯子推荐︱最佳翻墙机场︱科学上网VPN加速器节点 |
|
0 | 10271 | December 28, 2025 |
| Gaussian Process Inference |
|
3 | 2245 | December 23, 2025 |
| Multilayer Perceptrons |
|
14 | 8497 | December 16, 2025 |
| The Dataset for Pretraining BERT |
|
7 | 3631 | December 16, 2025 |
| 束搜索 |
|
21 | 10789 | December 12, 2025 |
| 为什么在 d2l 的 torch 包中没有 train_ch3 这个函数？ |
|
6 | 3115 | December 8, 2025 |
| 从全连接层到卷积 |
|
42 | 23060 | December 4, 2025 |
| 多层感知机 |
|
34 | 24100 | December 3, 2025 |
| 目标检测和边界框 |
|
14 | 8248 | December 3, 2025 |
| 数值稳定性和模型初始化 |
|
17 | 10513 | December 2, 2025 |
| 参数管理 |
|
26 | 17106 | November 30, 2025 |
| 风格迁移 |
|
11 | 5644 | November 27, 2025 |
| Language Models |
|
16 | 6207 | August 6, 2024 |
| Object-Oriented Design for Implementation |
|
15 | 7296 | November 25, 2025 |
| 引言 |
|
48 | 49692 | November 21, 2025 |
| 通过时间反向传播 |
|
22 | 10400 | November 14, 2025 |
| 含并行连结的网络（GoogLeNet） |
|
28 | 10709 | November 13, 2025 |
| 自然语言推断：使用注意力 |
|
3 | 1990 | November 12, 2025 |