Abdallah Nasir
Abdallah Nasir
Hello, Is there a plan to implement fully CNN seq2seq models? like this paper from Facebook https://arxiv.org/pdf/1705.03122.pdf I can see that there is a CNN encoder, but still not a...
Lately, a paper by Google released about using AdvGen to make robust NMT. [here](https://ai.googleblog.com/2019/07/robust-neural-machine-translation.html) and [here](https://arxiv.org/abs/1906.02443) Is it possible to implement this as an option when training a translator? Thanks
## What does this PR do? When translating using a model from hub_utils in the production setting, if skip_invalid_size_inputs is set to True, then the output list of translations may...