---
license: cc-by-4.0
tags:
- active
- adaptive
- code
- domain-specific
- manual
- randaugment
- stratified-90-10
- video-text
- webdataset
---

# dataset_135359330_code_video_text.py

## Dataset Summary

A **code** dataset with **video text** modality, stored in **webdataset** format.

## Preprocessing & Augmentation

- **Preprocessing**: domain specific
- **Augmentation**: randaugment

## Splits & Sampling

- **Split strategy**: stratified 90 10
- **Sampling**: active

## Quality & Labeling

- **Quality filtering**: adaptive
- **Labeling**: manual

## Files

- `dataset_135359330_code_video_text.py` — main artifact of this repository

## License

See the license field above.