I have created a private package and i want to use a module from that package on one script. It is possible to install only the necessary dependencies to run that module or I have to install the whole package always?
I will use my script inside a docker container. I can extract the code from the library, but this will make my code not maintainable in the long term.
For example, if I want to use a linear regression from scikit-learn it will be an overkill to install the whole library. The image will be so heavy!