bioptimus

/

H-optimus-0

like 22

Description

This PR adds the processor_config.json and updates the config.json to include the model_type="dinov2" so that the model can be used within the pipeline("image-feature-extraction", ...) from 🤗 Transformers, not sure who to link for review but maybe @rwightman may know if this is the right PR, I've tried already and seems to be working fine.

See the linked code below:

import requests

from PIL import Image
from transformers import pipeline

pipe = pipeline(
    "image-feature-extraction",
    model="bioptimus/H-optimus-0",
    image_processor="bioptimus/H-optimus-0",
    revision="25ce2c7450a358dc4e135177f9ac80cde6302880",
    device="cuda",
)

image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/cats.png"
image = Image.open(requests.get(image_url, stream=True).raw).convert("RGB")

outputs = pipe([image])