How to run this project?

[theova]

Hello,

Thank you for this project!
I really enjoyed reading the paper and I want to replicate the results to use it in a next project.

More precisely, I wonder how I can use this repository to build together the pipeline depicted in Figure 2:

Could you give me some advice on that?

Thank you very much!

[KirtusLeyba]

Hi theova, sorry that the documentation is basically non-existent. It's one of my goals to freshen up this repo and make it more clear what needs to be executed in which order, but I'm the only maintainer and I have a lot of research code I'm working on. I'll try to remember to let you know when I revisit this.

Right now the best answer I can give you is that you need to use one of the observation counting methods, either parallel_counting or parallel_counting_task_based should work, and then feed those observations into class_based_em. I'll produce some better docs when I have time to revisit this project. Best wishes.