藍+85CD

> It gives me the same error Maybe you could try creating a repo for me and others to look at?

也许可以推荐一下我的插件 [markdown-it-fff](https://fff.js.org/packages/markdown-it-fff.html) 它在读取时处理 Frontmatter（而不是提前修改），初始配置以外几乎不需要用户操作

> > 也许可以推荐一下我的插件 [markdown-it-fff](https://fff.js.org/packages/markdown-it-fff.html) 它在读取时处理 Frontmatter（而不是提前修改），初始配置以外几乎不需要用户操作 > > 看起来和 @qtqz 想要的有点区别，我猜他应该是想要一个可视化界面/交互终端脚本以修改 markdown 的 frontmatter 信息并保存原文件。 方式不同，不过我认为同样解决了标题中的 “方便用户迁移”

Since `hyper` was upgraded to 1.0, maybe `axum` should be upgraded to 0.7 as well? https://tokio.rs/blog/2023-11-27-announcing-axum-0-7-0

It looks like `actix-web` won't upgrade `http` crate until v5, and v5 has no due date...

> can you share the error messages during docker building? Maybe we can fix this problem first. :) Okay. The problem is that the installation of the dependency fails: ```dockerfile...

Update: The image builds successfully after adding `apt remove -y python3-blinker`, but it looks like still need to set up the `accelerate config` as described [here](https://github.com/intel-analytics/ipex-llm/tree/main/python/llm/example/GPU/LLM-Finetuning/axolotl#22-configures-accelerate-in-command-line-interactively).

If I try to enable DeepSpeed via `accelerate config`, it tells me that DeepSpeed is not installed. ```bash Do you want to enable dynamic shape tracing? [yes/NO]: Do you want...

Currently axolotl reports error when saving the prepared dataset: ```bash /usr/local/lib/python3.11/dist-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: ''If you don't plan on using image functionality from `torchvision.io`, you can...

> Currently axolotl reports error when saving the prepared dataset: This looks to be related to https://github.com/OpenAccess-AI-Collective/axolotl/issues/1544.