Datasets:

community-datasets
/

cawac

Tasks:

Text Generation

Modalities:

Formats:

Sub-tasks:

language-modeling

masked-language-modeling

Languages:

Size:

Libraries:

License:

Dataset card Data Studio Files Files and versions

albertvillanova HF Staff commited on Jan 16, 2024

Commit

914509d

·

verified ·

1 Parent(s): d80d4a7

Update loading script

Files changed (1) hide show

cawac.py +4 -3

cawac.py CHANGED Viewed

@@ -50,8 +50,9 @@ caWaC is a 780-million-token web corpus of Catalan built from the .cat top-level
 _LICENSE = "CC BY-SA 3.0"
-_URL = "http://nlp.ffzg.hr/resources/corpora/cawac/"
-_URLS = "http://nlp.ffzg.hr/data/corpora/cawac.uniq.sortr.gz"
 class Cawac(datasets.GeneratorBasedBuilder):
@@ -66,7 +67,7 @@ class Cawac(datasets.GeneratorBasedBuilder):
                 }
             ),
             supervised_keys=None,
-            homepage=_URL,
             license=_LICENSE,
             citation=_CITATION,
         )

 _LICENSE = "CC BY-SA 3.0"
+_HOMEPAGE = "http://nlp.ffzg.hr/resources/corpora/cawac/"
+# Source: http://nlp.ffzg.hr/data/corpora/cawac.uniq.sortr.gz
+_URLS = "data/cawac.uniq.sortr.gz"
 class Cawac(datasets.GeneratorBasedBuilder):
                 }
             ),
             supervised_keys=None,
+            homepage=_HOMEPAGE,
             license=_LICENSE,
             citation=_CITATION,
         )