---
language:
- en
license:
- cc-by-4.0
multilinguality:
- monolingual
---

# CLIP-BERT training data

This data was used to train the CLIP-BERT model first described in [this paper](https://arxiv.org/abs/2109.11321). 

The dataset is based on text and images from MS COCO, SBU Captions, Visual Genome QA and Conceptual Captions.

The image features have been extracted using the CLIP model [openai/clip-vit-base-patch32](https://huggingface.co/openai/clip-vit-base-patch32) available on Huggingface.