Running two different independent PyTorch programs on a single GPU

Question

I have a single NVIDIA GPU which has a memory of 16GB. I have to run two different (and independent; meaning, two different problems: one is a vision type task, another is N