GPU_MANAGER_PACKAGE: Developed a Python package to automate GPU usage, including allocation and release of memory specifically for Machine Learning tasks. Implemented for use on a DGX A100 server with 4 GPUs, each with 40GB memory, within a JupyterHub server environment. It can be installed on your ubuntu systems using pip.
GPU_MANAGER PACKAGE
Steps to setup the package and install: Download the acquire_release.py and setup.py files. Create a folder 'gpu_magaer' and place the acquire_release.py in it. Create another folder shared inside the 'gpu_magaer' folder. Create an empty file init.py file inside shared and also at acquire_release.py file path. setup.py should be outside the 'gpu_magaer' folder.
Run the below command in the terminal with the gpu_manager folder and setup.py file path. python setup.py sdist
After running the above command gpu_manager.egg-info and dist folders are created automatically in the same path.
In the dist folder, the gpu_manager-1.0.0.tar.gz file will be there, using this file we can install the package with the following terminal command.
Sudo pip3 install gpu_manager-1.0.0.tar.gz