pytorch-patterns

# PyTorch 开发模式 构建稳健、高效和可复现深度学习应用的 PyTorch 惯用模式与最佳实践。 ## 何时使用 * 编写新的 PyTorch 模型或训练脚本时 * 评审深度学习代码时 * 调试训练循环或数据管道时 * 优化 GPU 内存使用或训练速度时 * 设置可复现实验时 ## 核心原则 ### 1. 设备无关代码 始终编写能在 CPU 和 GPU 上运行且不硬编码设备的代码。 ```python # Good: Device-agnostic device = torch.device("cuda" if torch.cuda.is_available() else "cpu") model = MyModel().to(device) data = data.to(device) # Bad: Hardcoded device model = MyModel().cuda() # Crashes if no GPU data = data.cuda() ``` ### 2. 可复现性优先 设置所有随机种子以获得可复现的结果。 ```python # Good: Full reproducibility setup def set_seed(seed: int = 42) -> None: torch.manual_seed(seed) torch.cuda.manual_seed_all(seed) np.random.seed(seed) random.seed(seed) torch.backends.cudnn.deterministic = True torch.backends.cudnn.benchmark = False # Bad: No seed control model = MyModel() # Different weights every run ``` ### 3. 显式形状管理 始终记录并验证张量形状。 ```python # Good: Shape-annotated forward pass def forward(self, x: torch.Tensor) -> torch.Tensor: # x: (batch_size, channels, height, width) x = self.conv1(x) # -> (batch_size, 32, H, W) x = self.pool(x) # -> (batch_size, 32, H//2, W//2) x = x.view(x.size(0), -1) # -> (batch_size, 32*H//2*W//2) return self.fc(x) # -> (batch_size, num_classes) # Bad: No shape tracking def forward(self, x): x = self.conv1(x) x = self.pool(x) x = x.view(x.size(0), -1) # What size is this? return self.fc(x) # Will this even work? ``` ## 模型架构模式 ### 清晰的 nn.Module 结构 ```pyt...