Florence-2-base / README.md
captainspock's picture
Create README.md
64e4a8f verified
---
base_model: microsoft/Florence-2-base
library_name: transformers.js
license: mit
pipeline_tag: image-text-to-text
tags:
- vision
- text-generation
- text2text-generation
- image-to-text
---
## Changes for WebNN
Vision Encoder: Changed data type of Pow Y = 0.5 to float32 from float16 for fp16 and q4f16 models.
Based on:
https://huggingface.co/onnx-community/Florence-2-base/
https://huggingface.co/microsoft/Florence-2-base with ONNX weights to be compatible with Transformers.js.