Datasets:
mteb
/

Modalities:
Text
Formats:
arrow
Languages:
German
ArXiv:
Libraries:
Datasets
License:
rasdani commited on
Commit
1661468
1 Parent(s): 6c37e5c

update README

Browse files
Files changed (1) hide show
  1. README.md +10 -0
README.md CHANGED
@@ -9,4 +9,14 @@ configs:
9
  license: cc-by-4.0
10
  language:
11
  - de
 
 
12
  ---
 
 
 
 
 
 
 
 
 
9
  license: cc-by-4.0
10
  language:
11
  - de
12
+ source_datasets:
13
+ - "deepset/germanquad"
14
  ---
15
+
16
+ This dataset is derived from the [GermanQuAD](https://www.deepset.ai/germanquad) dataset.
17
+
18
+ This dataset takes the testset and represents it as a corpus in the [BEIR](https://github.com/beir-cellar/beir) information retrieval benchmark format.
19
+ The corresponding qrels can be found [here](hf.co/datasets/rasdani/germanquad-retrieval-qrels).
20
+
21
+
22
+ Full credit for the original dataset goes to the [authors](https://arxiv.org/abs/2104.12741) of the GermanQuAD [dataset](https://huggingface.co/datasets/deepset/germandpr).