修复笔记：SkyReels-V2 项目中的 torch.load 警告

#工作记录

一、问题描述

在运行项目时，出现以下警告：

FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.

二、受影响的文件和行号

1. 文件：F:\PythonProjects\SkyReels-V2\skyreels_v2_infer\modules\vae.py

   • 行号：第 566 行

   • 原代码

     model.load_state_dict(torch.load(pretrained_path, map_location=device), assign=True)

2. 文件：F:\PythonProjects\SkyReels-V2\skyreels_v2_infer\modules\t5.py

   • 行号：第 437 行

   • 原代码

     model.load_state_dict(torch.load(checkpoint_path, map_location="cpu"))

三、修复过程

1. 查找相关代码：在项目中查找使用 torch.load 的代码片段。

2. 修改代码：在 torch.load 中添加 weights_only=True 参数。

3. 测试修改后的代码：重新运行脚本，确认警告是否消失。

四、修改后的代码

1. vae.py 文件第 566 行：

   model.load_state_dict(torch.load(pretrained_path, map_location=device, weights_only=True), assign=True)

2. t5.py 文件第 437 行

   model.load_state_dict(torch.load(checkpoint_path, map_location="cpu", weights_only=True))

五、验证修改

完成上述修改后，重新运行项目以确保所有问题都已解决：

python generate_video.py --resolution 540P

六、总结

通过上述步骤，成功修复了项目中的 torch.load 警告。这些修改确保了代码在未来版本的库中仍然兼容，并提高了代码的安全性。

希望这份修复笔记能帮助你解决相同的问题。