metadata
language: en
license: mit
authors:
- Hengyu Shi
tags:
- clip
- vision
- text
- multimodal
Authors
- Hengyu Shi
- Boynn
Fine-tuned CLIP-ViT-bigG-14 Model
This model is a fine-tuned version based on laion/CLIP-ViT-bigG-14-laion2B-39B-b160k.
Usage Method
base_model = CLIPTextModelWithProjection.from_pretrained("Boynn/CLIP-ViT-bigG-14-laion2B-39B-b160k-sft")