Guolin Ke

@jameslamb Thank you! ping me when you need reviews.

@NicolasHug I think you used the wrong parameter name in that code.

@DimaMolod did it happen at the beginning, or already in-progress?

We will report the issues to the modelscope. And yes, the multimer data is included.

The problem is due to the unstable network, as the data is hosted in China. The modelscope team promised they would fix it in the next 2 weeks.

@DimaMolod The data generation code is almost the same as the one used in inference, except for the label extraction from mmcif. @ZiyaoLi maybe we can add a script for...

Hi @suncanghuai , thank you very much for the detailed plan and careful thoughts! The scheme looks good to me, I think option 1 is better for now; option 2...

you can 1. pull unimol's image, or 2. pull uni-core's image and install rdkit by yourself.

这是新手常见的错误，变量如果没有detach的话，计算图会保留，不会释放，导致内存越来越大。 refer to https://discuss.pytorch.org/t/memory-leak-when-appending-tensors-to-a-list/25937

You can refer to our test for c_api: https://github.com/microsoft/LightGBM/blob/master/tests/c_api_test/test_.py You can also check how we use these apis in python/R packages.