Quickstart#

Get from zero to your first pipeline run in 5 minutes.

Note

This guide uses a local filesystem config with minimal data. No S3 or CDS credentials needed.

1. Install#

git clone https://github.com/dsih-artpark/acestor.git
cd acestor
uv sync --extra dengue

See also

Need help with prerequisites (Python, geospatial libs, env vars)? See Installation.

2. Copy the example config#

cp configs-example/gba_docker_test.yaml configs/my-first-run.yaml

Open configs/my-first-run.yaml and update the paths to match your local setup:

storages:
  artifacts:
    kind: filesystem
    filesystem:
      base_path: "./artifacts"   # where outputs will be written

data:
  geojson:
    base_path: "datasets/geojsons"   # path to your GeoJSON files

  case_download:
    source_path: "datasets/raw_linelist_data/..."   # path to case data

Tip

Set debug: true in your config for the first run — it limits processing to the last 3 years of data, making the run much faster.

3. Run#

uv run python -m acestor.run \
  --pipeline pipelines.dengue.pipeline:build_pipeline \
  --config configs/my-first-run.yaml \
  --run-id my-first-run

Watch the logs as each step completes. Exit code 0 = success.

4. Check outputs#

ls artifacts/my-first-run/

You should see:

artifacts/my-first-run/
├── predictions/      ← dengue case predictions (CSV)
├── plots/            ← choropleth map PNGs
├── reports/          ← JSON + LaTeX report bundle
└── logs/

Next Steps#