Chriskuei comments

Results 5 comments of


                                            Chriskuei

RankHingeLoss代码逻辑不合理

@wenjunyang could you open a PR to fix the bug?

> 你好，我想问下，当我使用bert训练模型时候，在最后trainer.run()这里，老是报“Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.cuda.IntTensor instead (while checking arguments for embedding)”这个错，一直没找出到底该改动哪里，能否给解决一下啊非常非常感谢~~ Please provide more details, e.g. code snippets

How to load large embedding efficiently?

Thanks for your feedback. We will fix it soon.

For drmm model, dense_output should be flipped at dim=-1?

Hi @wangcongcong123, the `pad_mode` of `text_left` and `match_histogram ` is consistent. https://github.com/NTMC-Community/MatchZoo-py/blob/49548ad4dd7da4c890ac786a09d9df9172c3af47/matchzoo/dataloader/callbacks/padding.py#L226-L239

DSSM tutorial code: RuntimeError Expected object of scalar type Float but got scalar type Double

Hi @katestfla, which version of PyTorch do you install?

Chriskuei

RankHingeLoss代码逻辑不合理

bert processor

How to load large embedding efficiently?

For drmm model, dense_output should be flipped at dim=-1?

DSSM tutorial code: RuntimeError Expected object of scalar type Float but got scalar type Double