Bark and Ambrosia Beetle Gallery
A dedicated web platform for storing, browsing, and managing large datasets of annotated images for Bark and Ambrosia Beetles.
Documentation
Developer setup, the daily dev workflow, git/PR conventions, and the production deployment guide all live in the project wiki:
- Contributing — how to contribute, from issue to merged PR, and the house rules.
- Getting Started — prerequisites, first-time setup, running the app locally, editing code/CSS/models, running tests.
- Git Workflow — how we branch, commit, and open PRs.
- Deployment — the production server, logs, and how deploys happen.
Quick Start
The site runs only in Docker. You need Git and Docker with Docker Compose (on Windows or Mac, Docker Desktop must be
running); docker compose also loads docker-compose.override.yml, which adds a local mail inbox.
git clone https://github.com/ChristopherMarais/barkandambrosiagallery.git
cd barkandambrosiagallery
echo DJANGO_DEBUG=True > .env
docker compose build
docker compose run --rm web pixi run install-js
docker compose up -d --wait db
docker compose exec db psql -U beetles_user -d beetles_db -c "CREATE EXTENSION IF NOT EXISTS pg_trgm;"
docker compose run --rm web pixi run migrate
docker compose run --rm web pixi run python manage.py createsuperuser
docker compose up -d
Then open http://localhost:8000. See Getting Started in the wiki for the full walkthrough and the daily dev loop.
Licence
The platform code is MIT-licensed (see LICENSE).
The AI service (beetlesgallery/tools/modal_ibbi_api.py) is deployed separately from the web app. It installs the
ibbi package together with
Ultralytics, which is AGPL-3.0, and runs IBBI model weights under their
own licences: AGPL-3.0 for the Ultralytics-trained detectors, the DINOv3 License for the DINOv3 classifiers and MIT for
the BioCLIP 2 classifier (see IBBI's licences).
Images in the gallery keep the licence given by their contributor, shown on each specimen record.
