strange datasets from OSCAR corpus

From the [official site ](https://oscar-corpus.com/), the Yue Chinese dataset should have 2.2KB data.
7 training instances is obviously not a right number.
As I can read Yue Chinese, I call tell the last instance is definitely not something that would appear on Common Crawl.
And even if you don't read Yue Chinese, you can tell the first six instance are problematic.
(It is embarrassing, as the 7 training instances look exactly like something from a pornographic novel or flitting messages in a chat of a dating app)
It might not be the problem of the huggingface/datasets implementation, because when I tried to download the dataset from the official site, I found out that the zip file is corrupted.
I will try to inform the host of OSCAR corpus later.
Awy a remake about this dataset in huggingface/datasets is needed, perhaps after the host of the dataset fixes the issue.

> Hi @jerryIsHere , sorry for the late response! Sadly this is normal, the problem comes form fasttext's classifier which we used to create the original corpus. In general the classifier is not really capable of properly recognizing Yue Chineese so the file ends un being just noise from Common Crawl. Some of these problems with OSCAR were already discussed [here](https://arxiv.org/pdf/2103.12028.pdf) but we are working on explicitly documenting the problems by language on our website. In fact, could please you open an issue on [our repo](https://github.com/oscar-corpus/oscar-website/issues) as well so that we can track it?

Thanks a lot, the new post is here:
https://github.com/oscar-corpus/oscar-website/issues/11