Skip to content

autodistill/autodistill-kosmos-2

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Autodistill Kosmos-2 Module

This repository contains the code supporting the Kosmos-2 base model for use with Autodistill.

Kosmos-2, developed by Microsoft, is a multimodal language model that you can use for zero-shot object detection. You can use Kosmos-2 with autodistill for object detection.

Read the full Autodistill documentation.

Read the Kosmos-2 Autodistill documentation.

Installation

To use Kosmos-2 with autodistill, you need to install the following dependency:

pip3 install autodistill-kosmos-2

Quickstart

from autodistill_kosmos_2 import Kosmos2

# define an ontology to map class names to our Kosmos2 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 = Kosmos2(
    ontology=CaptionOntology(
        {
            "person": "person",
            "a forklift": "forklift"
        }
    )
)

predictions = base_model.predict("./example.png")

base_model.label("./context_images", extension=".jpeg")

License

This package is implemented using the Transformers Kosmos-2 implementation. The underlying Kosmos-2 model, developed by Microsoft, is licensed under an MIT license.

🏆 Contributing

We love your input! Please see the core Autodistill contributing guide to get started. Thank you 🙏 to all our contributors!