Datasets:

Modalities:
Text
Formats:
parquet
Size:
< 1K
ArXiv:
Libraries:
Datasets
pandas
malhajar nielsr HF Staff commited on
Commit
4d64739
·
verified ·
1 Parent(s): ab6bcf6

Add task category, link to paper (#3)

Browse files

- Add task category, link to paper (b7a650fad93b975c97f047c71479b0825b224f34)


Co-authored-by: Niels Rogge <[email protected]>

Files changed (1) hide show
  1. README.md +4 -1
README.md CHANGED
@@ -16,7 +16,10 @@ configs:
16
  data_files:
17
  - split: test
18
  path: data/test-*
 
 
19
  ---
 
20
  # GSM1K-TR-50-Samples
21
 
22
- This dataset contains 50 samples from the turkish replication of [gsm1k paper](https://arxiv.org/pdf/2405.00332). They are sampled from the complete gsm1k turkish dataset by difficulty distribution, where the difficulty's measurement method is provided in the original paper.
 
16
  data_files:
17
  - split: test
18
  path: data/test-*
19
+ task_categories:
20
+ - question-answering
21
  ---
22
+
23
  # GSM1K-TR-50-Samples
24
 
25
+ This dataset contains 50 samples from the turkish replication of [gsm1k paper](https://huggingface.co/papers/2405.00332). They are sampled from the complete gsm1k turkish dataset by difficulty distribution, where the difficulty's measurement method is provided in the original paper.