|
Download README.md from IshaanKaur/bert-tokenizer: direct link, hf CLI and curl.
- Browser
- Download file 650 Bytes
-
https://huggingface.co/datasets/IshaanKaur/bert-tokenizer/resolve/main/README.md
- Command line
-
hf download hf://datasets/IshaanKaur/bert-tokenizer/README.md
-
curl -L -o README.md https://huggingface.co/datasets/IshaanKaur/bert-tokenizer/resolve/main/README.md
650 Bytes
metadata
license: apache-2.0
tags:
- active
- audio-text
- auto-ml
- lenient
- news-media
- none
- parquet
- semi-auto
- temporal
dataloader.py
Dataset Summary
A news media dataset with audio text modality, stored in parquet format.
Preprocessing & Augmentation
- Preprocessing: auto ml
- Augmentation: none
Splits & Sampling
- Split strategy: temporal
- Sampling: active
Quality & Labeling
- Quality filtering: lenient
- Labeling: semi auto
Files
dataloader.py— main artifact of this repository
License
See the license field above.