Skip to content

Conversation

@FlorentijnD
Copy link
Contributor

This PR aims to:

  • Condense our learnings from all projects and research we've conducted so far in the realm of causality & time series.
  • Port the codebase from our internal ml-guidelines repository and obfuscating project work under NDA.
  • Work with the latest versions of PCMCI - tigramite==5.0.0.3 and scipy==1.8.1

@FlorentijnD FlorentijnD added enhancement New feature or request sd Structured Data chapter labels Jun 2, 2022
@FlorentijnD FlorentijnD self-assigned this Jun 2, 2022
Copy link
Contributor

@julienschuermans julienschuermans left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Really sorry for the delay. I have some tiny remarks, but other than that this PR looks good to go:

  • Storing the data in a .pickle file is probably not the best idea.
  • By copying pycausality and adding wrappers this PR adds a lot of code. Next time we should check if a "quick tip" can be smaller in size. The reusable code could be hosted in a separate codebase.
  • There are a lot of items in the requirements.txt. I guess most of those are "secondary"? Maybe next time let's only add the direct dependencies.

## Purpose of this folder
We advocate to use the [PyCausality](https://pypi.org/project/PyCausality/) package when working with Transfer Entropy.
The code itself is not maintained anymore but still works well.
Because the pip package might become/is unstable, we copied the source code and created a wrapper around it.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the pip package actually unstable? Or is the main reason here that our wrapper adds a lot of value?

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

Labels

enhancement New feature or request sd Structured Data chapter

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants