Datasets:
Dataset viewer error
I had uploaded each language as a split. But, the dataset viewer doesn't seem to support more than 32 splits for the dataset to be viewed. is there a way it can be fixed?
Lang could be a column, can be filtered based on col to get specific languages or combos, and there's neat features like the new SQL console that you could leverage with lang in col
@rwightman I was wondering if it was possible to download the entire dataset on my laptop and fix the file names from data1, data2 to something meaningful and then re-upload it?
if it's possible can you share the Syntax for it? Cuz I saw HF datasets are saved on disk as pyarrow files which can't be opened with pyarrow Library.
also had another question: At the moment I am not sure how can I commit/upload my dataset as a single config like English, German, Polish etc.
I can change the YAML file and do it later but I want to upload them as respective Configs.
Do you know the Syntax for that?