Skip to content

Conversation

@ghost
Copy link

@ghost ghost commented Oct 3, 2018

Hey,
I am working on my bachelor thesis about "Recommender-Systems" and I use this project for working.

So I thought it might be useful to share my experiences with you and to expand the project a bit.

This PR adds a Dockerfile and the regarding documentation to this project.
With Docker it is possible to host a container with all dependencies which are requiered by this project at any OS.

You can use the Dockerfile to simply host a test-container or as a interpreter for PyCharm.
It is also possible to use a running container completly for development.

Please feel free to share your opinion with me.
Best regards
Marc Feger

@NicolasHug
Copy link
Owner

Hi Marc,
Thanks a lot for the PR! I think integrating docker is outside the scope of this repo, plus I'm not experienced with docker and I wouldn't have the time / experience to maintain it, so I'd rather not merge this PR.

Maybe a better option would be to create your own repo with the Dockerfile?

Nicolas

@ghost
Copy link
Author

ghost commented Oct 6, 2018

Hey Nicolas,

I understand that good. Docker is not easy if you have not worked with it yet.

I wanted to go through some papers for my bachelor thesis and implement RS evaluation procedures.

In addition I saw that already in the "similarities module" some procedures from Wikipedia or books were implemented.
That's why I would like to know if there is any interest in literary methods such as NDMP and the resulting Kendall tau, R score, NDCG and many others?

I wish you a nice weekend
Marc Feger

@NicolasHug
Copy link
Owner

You mean NDPM?

I think I'd be interseted in seeing what the actual functions would look like first, for example in a FAQ entry. That would help me have a clearer idea of what is actually needed: for example NDCG needs a utility function, which is quite open.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

1 participant