Sinapsis Hugging Face Grounding DINO

Templates for seamless integration with Grounding DINO models

🐍 Installation • 📦 Features • ▶️ Example Usage • 🌐 Webapp • 📙 Documentation • 🔍 License

🐍 Installation

Install using your package manager of choice. We encourage the use of uv

Example with uv:

  uv pip install sinapsis-huggingface-grounding-dino --extra-index-url https://pypi.sinapsis.tech

or with raw pip:

  pip install sinapsis-huggingface-grounding-dino --extra-index-url https://pypi.sinapsis.tech

[!IMPORTANT] Templates may require extra optional dependencies. For development, we recommend installing the package with all the optional dependencies:

with uv:

  uv pip install sinapsis-huggingface-grounding-dino[all] --extra-index-url https://pypi.sinapsis.tech

or with raw pip:

  pip install sinapsis-huggingface-grounding-dino[all] --extra-index-url https://pypi.sinapsis.tech

📦 Features

The templates in this package include functionality for different Grounding DINO-based tasks:

• GroundingDINO: Detect objects with bounding boxes based on text prompts (zero-shot object detection).
• GroundingDINOClassification: Classify images using predefined classes or text prompts, handling as many classes as possible within token limits.
• GroundingDINOFineTuning: Fine-tune Grounding DINO checkpoints on custom datasets.

▶️ Example Usage

Below is an example YAML configuration for zero-shot object detection using Grounding DINO.

agent:
  name: grounding_dino_detection

templates:
  - template_name: InputTemplate
    class_name: InputTemplate
    attributes: {}

  - template_name: FolderImageDatasetCV2
    class_name: FolderImageDatasetCV2
    template_input: InputTemplate
    attributes:
      data_dir: my_dataset

  - template_name: GroundingDINO
    class_name: GroundingDINO
    template_input: FolderImageDatasetCV2
    attributes:
      model_path: IDEA-Research/grounding-dino-base
      inference_mode: zero_shot
      text_input: a person.
      device: cuda
      threshold: 0.2
      text_threshold: 0.3

  - template_name: BBoxDrawer
    class_name: BBoxDrawer
    template_input: GroundingDINO
    attributes:
      overwrite: true
      randomized_color: false

  - template_name: ImageSaver
    class_name: ImageSaver
    template_input: BBoxDrawer
    attributes:
      save_dir: ./output_dir
      extension: png

[!IMPORTANT] The FolderImageDatasetCV2, BBoxDrawer and ImageSaver templates correspond to the sinapsis-data-readers, sinapsis-data-visualization and sinapsis-data-writers packages respectively. If you want to use the example, please make sure you install these packages.

To run the config, use the CLI:

sinapsis run name_of_config.yml

🌐 Webapp

Th Sinapsis web applications provide an interactive way to explore and experiment with AI models. They allow users to generate outputs, test different inputs, and visualize results in real time, making it easy to experience the capabilities of each model. Below are the available webapps and instructions to launch them.

[!IMPORTANT] To run any of the apps, you first need to clone this repo:

git clone git@github.com:Sinapsis-ai/sinapsis-huggingface.git
cd sinapsis-huggingface

[!NOTE] If you'd like to enable external app sharing in Gradio, export GRADIO_SHARE_APP=True.

[!NOTE] Agent configuration can be changed through the AGENT_CONFIG_PATH env var. You can check the available configurations in each package configs folder.

docker compose -f docker/compose.yaml build

docker compose -f docker/compose_vision.yaml up sinapsis-huggingface-vision-gradio -d

docker logs -f sinapsis-huggingface-vision-gradio

Running on local URL:  http://127.0.0.1:7860

docker compose -f docker/compose_vision.yaml down

uv sync --frozen

uv pip install sinapsis-huggingface[all] --extra-index-url https://pypi.sinapsis.tech

uv run webapps/vision_demo.py

Running on local URL:  http://127.0.0.1:7860

📙 Documentation

Documentation is available on the sinapsis website

Tutorials for different projects within sinapsis are available at sinapsis tutorials page

🔍 License

This project is licensed under the AGPLv3 license, which encourages open collaboration and sharing. For more details, please refer to the LICENSE file.

For commercial use, please refer to our official Sinapsis website for information on obtaining a commercial license.