REPRODUCING.md

This Code Ocean Compute Capsule will allow you to run and reproduce the results of Mask Gen on your local machine¹. Follow the instructions below, or consult our knowledge base for more information. Don't hesitate to reach out to Support if you have any questions.

¹ You may need access to additional hardware and/or software licenses.

Prerequisites

• Docker Community Edition (CE)
• nvidia-container-runtime for code that leverages the GPU

Instructions

Download attached Data Assets

In order to fetch the Data Asset(s) this Capsule depends on, download them into the Capsule's data folder:

• mask-training-2025-05-23 should be downloaded to data/mask-training-2025-05-23
• 754615-mask-training-data should be downloaded to data/754615-mask-training-data
• mask-training-data-2025-05-13 should be downloaded to data/mask-training-data-2025-05-13
• 751474_mask-result-2025-06-02 should be downloaded to data/751474_mask-result-2025-06-02
• mask-training-data-2025-05-14 should be downloaded to data/mask-training-data-2025-05-14
• mask-training-2025-05-31 should be downloaded to data/mask-training-2025-05-31
• mask-training-2025-05-15 should be downloaded to data/mask-training-2025-05-15
• 754615_mask-result-2025-06-02 should be downloaded to data/754615_mask-result-2025-06-02
• mask-result-2025-06-02 should be downloaded to data/mask-result-2025-06-02
• mask-training-2025-05-16 should be downloaded to data/mask-training-2025-05-16

Log in to the Docker registry

In your terminal, execute the following command, providing your password or API key when prompted for it:

docker login -u cameron.arshadi@alleninstitute.org registry.codeocean.allenneuraldynamics.org

Run the Capsule to reproduce the results

In your terminal, navigate to the folder where you've extracted the Capsule and execute the following command, adjusting parameters as needed:

docker run --platform linux/amd64 --rm --gpus all \
  --workdir /code \
  --volume "$PWD/code":/code \
  --volume "$PWD/data":/data \
  --volume "$PWD/results":/results \
  --env AWS_ACCESS_KEY_ID=value \
  --env AWS_SECRET_ACCESS_KEY=value \
  --env AWS_DEFAULT_REGION=value \
  registry.codeocean.allenneuraldynamics.org/capsule/8732c547-1336-4261-a682-fd53c9f61786 \
  bash run '<S3_ZARR_PATH|TILE_JSON_PATH>' '<INFERENCE_CHANNEL>'

The first argument can be either a parent zarr path or a JSON file containing a tile_paths list. For example:

bash run '<S3_ZARR_PATH>' '<INFERENCE_CHANNEL>'
bash run /data/tile_000006.json 561

As secrets are required, replace all value occurances with your personal credentials prior to your run.