Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Languages:
French
Size:
10K - 100K
ArXiv:
License:
Update README.md
Browse files
README.md
CHANGED
@@ -35,7 +35,7 @@ For more information on the revision details, please refer to our paper [WikiNER
|
|
35 |
|
36 |
The dataset is available in two formats.
|
37 |
The CoNLL version contains three columns: text, POS and NER.
|
38 |
-
The
|
39 |
Originally conceived as a test set, there is no recommended train/dev/test split.
|
40 |
The downloaded dataset is by default labeled `train`.
|
41 |
|
|
|
35 |
|
36 |
The dataset is available in two formats.
|
37 |
The CoNLL version contains three columns: text, POS and NER.
|
38 |
+
The Parquet version is downloadable using the `datasets` library.
|
39 |
Originally conceived as a test set, there is no recommended train/dev/test split.
|
40 |
The downloaded dataset is by default labeled `train`.
|
41 |
|