HUANG Fei

@sudanl Can you run the script with only one GPU (single process)? It only says that the Cuda program is not correctly compiled but not shows the real errors.

Thanks for your interest. Unfortuantely, we did not try architectures other than transformer-base. In my intuition, both the encoder and decoder are important in capturing the data information. Especially, a...

@JunchengYao The problems you mentioned are very common in NAT models. They are caused by the nature of parallel prediction nature and the conditional independent assumption. Many recent studies (including...

不好意思，目前没有出电子书的计划

https://item.jd.com/10026041769006.html 这个是和我们合作的出版社京东官方店

@sdws258 Try removing this line https://github.com/thu-coai/DA-Transformer/blob/8ef9c8f43f239d7004be004d5d2778cf080d4b2f/fs_plugins/custom_ops/dag_loss.py#L54 If it works well, I will update a patch to fix the issue

``max-source-positions`` and ``max-target-positions`` specify the max length of the samples. You should set it according to your dataset. If you want to train with a sample whose target length is...

I just restarted the space and it works now. @Tangent-90C

Try filtering out samples that satisfying ``target length / source length > src_upsample_scale`` in your dataset. If you still have problem, please you specify the dataset and your training script?

@MeWannaSleep ``--src-upsample-scale`` is an argument in previous version. You can now use ``--upsample-scale`` and ``--decode-upsample-scale`` to control the upsampling rate in training and inference, respectively. If you met the same...