Add Documentation for seq_flow_lite

[jasonw247]

Prerequisites

Please answer the following question for yourself before submitting an issue.

• I checked to make sure that this feature has not been requested already.

1. The entire URL of the file you are using

https://github.com/tensorflow/models/tree/master/research/seq_flow_lite

2. Describe the feature you request

Would it be possible to document or provide the process for pre-training ByteQRNN on multilingual data? Additionally, are there any plans to release model checkpoints?

I read the blog post here which mentions that pre-trained and finetuned ByteQRNN models were evaluated against BERT on the civil_comments dataset and that the models were pre-trained using multilingual data. I would be interested in trying to reproduce this, as well as using the pre-trained models for transfer learning on my own multilingual data (classification + NER).

3. Additional context

N/A

4. Are you willing to contribute it? (Yes or No)

Yes