bushlab/ml_bushlab repository overview

⁠Container Information

This container was built via condaforge, using the mambaforge installer, and it contains :

• Tensorflow 2 (with CUDA 11.8 compatibility)
• Pytorch (with CUDA compatibility)
• Jupyterlab
• R (along with an R kernel for Jupyterlab)

⁠Running the container

To run the container, use the following Docker command:

docker run -it  --rm -p 8888:8888 bushlab/ml_bushlab:v1.0

⁠Activating the torch_env environment

You may need to activate the torch_env environment with the following command:

mamba activate torch_env

⁠Launching Jupyter lab

Once inside the container, run:

jupyter lab --allow-root --ip=0.0.0.0 --port=8888

⁠Opening Jupyter Lab

Open your web browser and enter the following URL: http://localhost:8888. Copy the token provided to access Jupyter Lab.

⁠Note

This container is specifically designed for use with NVIDIA GPUs that have CUDA 12.2 Compute Capability and relies on NVIDIA_SMI version 535.54.03. It was primarily created for Singularity on Case Western Reserve University's HPC server.