Linhao Luo
Linhao Luo
I did not face this problem. But maybe this note can help > Note: Please make sure `dgl==0.9.0` and use CUDA, our codes rely on a small [bug](https://github.com/dmlc/dgl/issues/4512#issuecomment-1250642930) of dgl...
您好,感谢您对我们工作的兴趣。相同现象我们也有发现,由于neural process需要从分布中采样,所以在训练过程中会导致产生一定波动。根据我们的经验,在当前种子设置下,运行5次左右应该能达到论文汇报的结果。同时,您也可以采用我们公开的checkpoint进行结果复现。
Thanks for the notice! We will fix them later.
Thanks for pointing out. I have updated the script. The alpaca model we used is from the official weight diff. I am not sure it is okay to replace it...
Could you try using Python 3.10 to have a try?
Hi, thanks for your interest in our project. Currently, the GFM-RAG has not been trained to handle Chinese text. In order to do so, we need to switch to a...