Add UA-SQuAD dataset for Ukrainian language

[nmeln]

This is a Ukrainian version of Stanford Question Answering Dataset (SQuAD). It is a QA dataset with MIT license. While the page says it's a WIP, the dataset still has a good amount of data. I will try reaching out to them to find out if they have a more complete version.

Github link: https://github.com/fido-ai/ua-datasets/tree/main/ua_datasets/src/question_answering

The repo also contains Text Classification and Token Classification datasets, but I'm not sure if they are useful for OA. https://github.com/fido-ai/ua-datasets

Info about dataset:

Number of samples: 13 859 Number of questions without answers: 2 927 File size: 17.1 MB

Link to huggingface dataset: https://huggingface.co/datasets/FIdo-AI/ua-squad

[huu4ontocord]

than you.

[nmeln]

@ontocord

Should I proceed to add it here? https://github.com/LAION-AI/Open-Assistant/blob/main/model/model_training/custom_datasets/qa_datasets.py