Zheng-Ning Liu

请问您的系统中是否安装了 mipcc，如果按照了，请通过 `mpicc --showme:version` 告知我们 mpicc 版本，方便核对错误

Good. Please also provide documentation for each new function.

您好，algorithm cache 是用于存储不同卷积计算方法的缓存。cache full 常出现于输入形状不确定的网络中，对训练结果没有影响。

Jittor 目前支持 arm CPU上部署。目前测试支持了国产的龙芯、飞腾、申威、鲲鹏、 Apple M1 芯片等非 X86 架构。

Jittor可以直接在上面提到的芯片架构运行，不需要额外的配置。部署时安装好 Jittor 所需的依赖即可。

Firstly, convert jittor var to numpy. Then you can use it in torch. ```python a = torch.rand(3) b = jt.float(a) c = b.data # numpy array d = torch.from_numpy(c) ```

Hi @MyEternalMoon , Thanks for hightlighting the data sample. There are a lot of holes / genues in the mesh 'bookshelf_0099'. So I tried my best to make it manifold...

Hi, @whsu2s , 1. The original datasets (SHREC11, Huambody, etc.) use off formats. The number of faces of remeshed shapes are determined by the subdivision depth and the base size,...

Hi @lykius, Could you please show some examples from your meshes so that I can have some clues? Also, you can try [genTrainData_slow.m](https://github.com/HTDerekLiu/neuralSubdiv/blob/master/genTrainData_slow.m) in [NeuralSubdivision](https://github.com/HTDerekLiu/neuralSubdiv). It may work in your...

Sorry that I forgot to further reply to your question. I have try the remeshing script to a few of your provided meshes. The algorithm seems Okay in my cases....