This repository contains the code supporting the MetaCLIP base model for use with Autodistill.
MetaCLIP, developed by Meta AI Research, is a computer vision model trained using pairs of images and text. The model was described in the Demystifying CLIP Data paper. You can use MetaCLIP with autodistill for image classification.
Read the full Autodistill documentation.
Read the MetaCLIP Autodistill documentation.
To use MetaCLIP with autodistill, you need to install the following dependency:
pip3 install autodistill-metaclipfrom autodistill_metaclip import MetaCLIP
# define an ontology to map class names to our MetaCLIP prompt
# the ontology dictionary has the format {caption: class}
# where caption is the prompt sent to the base model, and class is the label that will
# be saved for that caption in the generated annotations
# then, load the model
base_model = MetaCLIP(
ontology=CaptionOntology(
{
"person": "person",
"a forklift": "forklift"
}
)
)
results = base_model.predict("./image.png")
print(results)from autodistill_metaclip import MetaCLIP
base_model = MetaCLIP(None)
text = base_model.embed_text("coffee")
image = base_model.embed_image("coffeeshop.jpg")
print(base_model.compare(text, image))This project was licensed under a Creative Commons Attribution-NonCommercial 4.0 International.
We love your input! Please see the core Autodistill contributing guide to get started. Thank you 🙏 to all our contributors!