Datasets:

Modalities:
Text
Formats:
parquet
Libraries:
Datasets
Dask
License:

Convert dataset to Parquet

#2
by albertvillanova HF staff - opened
Language Technology Research Group at the University of Helsinki org
No description provided.
albertvillanova changed pull request status to merged

This conversion is incomplete, I'm afraid.

There are many more possible languages and language combinations. The conversion covered only the example pairs from the original dataset script, i.e.:

# Please note that only few pairs are shown here. You can use config to generate data for all language pairs
_LANGUAGE_PAIRS = [
    ("el", "en"),
    ("en", "km"),
    ("en", "so"),
    ("de", "pl"),
    ("fr", "nl"),
    ("en", "sw"),
    ("en", "tl"),
    ("es", "gl"),
]

The possible options can be seen here: http://opus.nlpl.eu/ParaCrawl.php

With only a reduced set of languages covered after the conversion, most data in this dataset is now inaccessible through HF datasets.

Sign up or log in to comment