charing/cuca_env repository overview

⁠CUCA Docker Image Repository

⁠Overview

CUCA is a project designed to address identifying fine-grained cell types from histology images. This Docker image repository provides a convenient containerized deployment solution for the CUCA project. By using Docker images, users can quickly set up the runtime environment for the CUCA project, avoiding the cumbersome process of installing and configuring dependencies, and ensuring consistent runtime results across different environments.

Chaoyang Yan. (2025). lyotvincent/CUCA: CUCA_software_v1_1 (cuca_v1.1). Zenodo. https://doi.org/10.5281/zenodo.15087256⁠

GitHub: https://github.com/lyotvincent/CUCA⁠

⁠Key Features

• Rapid Deployment: With Docker images, users can complete the setup of the CUCA project environment in a short time without manually installing various dependency libraries and tools.

• Environment Consistency: Docker images encapsulate all the dependencies required by the CUCA project, ensuring consistent runtime results across different hardware environments.

• Resource Isolation: Leveraging Docker's containerization technology, the CUCA project can run in an independent container, avoiding conflicts with other applications on the host machine.

⁠Image Versions

Currently, the repository provides a main version of the image: charing/cuca_env:v1_0. This image includes all the dependencies required to run the CUCA project, including Python 3.9 and PyTorch 2.4.1.

⁠Usage Instructions

⁠Pulling the Image

You can directly pull the pre-built image from DockerHub Image Registry:

docker pull charing/cuca_env:v1_0

if request canceled due to network issue, you can pull the pre-built image from Alibaba Cloud Container Registry

docker pull crpi-vgk1sxg5ba2a8sff.cn-hangzhou.personal.cr.aliyuncs.com/cuca_proj_env/cuca_env:v1_0

⁠Building the Image

If you want to build the image yourself, you can use the Dockerfile provided in the GitHub project⁠:

docker build -t cuca_env:v1_0 .
docker images

⁠Starting the Container

Once the image is ready, you can start the Docker container using the following command:

docker run -dit \
    --name cuca_proj_env \
    --gpus all \
    --cpus 64 \
    --mount type=bind,source=/your_proj_absolute_path/CUCA/,target=/home/appuser/CUCA \
    --shm-size 128g \
    -p 6009:6009 \
    cuca_env:v1_0 \
    /bin/bash -c 'while true; do echo `date`; sleep 600; done'

Please replace /your_proj_absolute_path/CUCA/ with the absolute path of your local CUCA project.

⁠Entering the Container

After starting the container, you can enter the container using the following command:

docker exec -it cuca_proj_env /bin/bash

⁠Notes

• When using the docker run command, ensure that you have correctly configured the GPU driver on the host machine so that the container can use GPU resources normally.
• If you encounter any issues during the process of building or using the image, please refer to the GitHub repository or submit an issue.

⁠Contribution and Feedback

If you have any suggestions or find issues with this Docker image repository, please feel free to submit an issue or a pull request on the GitHub repository. Your contributions and feedback are highly appreciated!