Step 1: Install MemBrain-Seg

Clone the GitHub repository and follow the instructions in the README to install dependencies and set up the Python environment:

Step 2: Select a Specific GPU

To specify which GPU to use, prepend your MemBrain command with the CUDA_VISIBLE_DEVICES environment variable. This variable tells PyTorch which GPU(s) are visible to the program. You can run multiple MemBrain commands in parallel on different GPUs to process several tomograms simultaneously.

CUDA_VISIBLE_DEVICES=1 membrain segment --tomogram-path lamella4_ts_001_full_rec.mrc --ckpt-path /home/csun/Membrain-model/MemBrain_seg_v10_alpha.ckpt --store-connected-components