Datasets:
This file doesn't have the proper format for the Amharic part.
#2
by
Henok
- opened
No description provided.
can you give more details? It looks fine to me:
Actually the issue is from the jsonl(https://huggingface.co/datasets/google/smol/raw/main/smolsent/am_en.jsonl) raw file and not on the Parquet
Have you tried to load and print this data in Python? I believe this is correctly encoded, but when you look at the json in this view it shows you the unicode values, not the Amharic text. For example:
>>> print("\u1235")
ስ
Okay got it, thanks