Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
json
Sub-tasks:
fact-checking
Languages:
Slovak
Size:
10K - 100K
License:
Download data/train.jsonl from NaiveNeuron/DemagogSK: direct link, hf CLI and curl.
- Browser
- Download file 39 MB
-
https://huggingface.co/datasets/NaiveNeuron/DemagogSK/resolve/main/data/train.jsonl
- Command line
-
hf download hf://datasets/NaiveNeuron/DemagogSK/data/train.jsonl
-
curl -L -o train.jsonl https://huggingface.co/datasets/NaiveNeuron/DemagogSK/resolve/main/data/train.jsonl
39 MB
- Xet hash:
- 1191f28fb29370ff2082b96afb4bb0dc3e9ca062673c75b45957577844d44a64
- Size of remote file:
- 39 MB
- SHA256:
- 9514b43878794844a3832de5628f44313e21642a08d13ea05d370ce15ca9c076
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.