WebVid 10M Classified (100k)

Each description from the WebVid 10M dataset is passed through Llama 3.3 70B to classify the description as either action or no_action.

If it is classified as an action, it'll be rewritten in a clearer way. Otherwise, the rewritten description will be none.

WebVid is a large-scale dataset of short videos with textual descriptions sourced from the web. The videos are diverse and rich in their content.

WebVid-10M contains:

10.7M video-caption pairs. 52K total video hours.

To use this dataset:

import tensorflow_datasets as tfds

ds = tfds.load('webvid', split='train')
for ex in ds.take(4):
 print(ex)

See the guide for more informations on tensorflow_datasets.

qingy2024/webvid-10M-pro dataset hosted on Hugging Face and contributed by the HF Datasets community

The dataset used for training the video model consists of Webvid-10M, a large-scale dataset of short videos with textual descriptions.