Charles Tapley Hoyt

The current binarize function uses a cutoff of 0.5 for binarization: https://github.com/AstraZeneca/rexmex/blob/3e266529761281ae832e49736e48d3e46f3b4af4/rexmex/utils.py#L28-L34 This is an issue for PyKEEN, where the scores that come from a model could all be on...

Adjusted mean rank

3

@mberr's adjusted mean rank address some of the problems with the mean rank, including its size dependence. Reference: https://arxiv.org/abs/2002.06914

How to resolve pathway identifiers?

1

After a closer look, I'm having issues with the data source used for pathways. ``` Pathway::PC7_2008 BMAL1:CLOCK,NPAS2 activates circadian gene expression Pathway ``` Should correspond to the reactome pathway, https://reactome.org/content/detail/R-HSA-1368108,...

Provenance for some CbP relationships is lost during condense() operation

3

I'm going through metadata in the compound-binds-gene relationships, and taking a specific look at the `actions` lists. In many examples, there are several actions, such as with [drugbank:DB00502](https://www.drugbank.ca/drugs/DB00502) binds [ncbigene:1813](https://www.ncbi.nlm.nih.gov/gene/1813)....

Similarly to the `WikidataTextRepresentation`, this uses `PyOBO` as a service for looking up labels for entities encoded with `CURIE`s appearing in biomedical knowledge graphs. Unfortunately, the semantics of all of...

🚀🫀 Extend scoring, evaluation, and inference for relation side

2

Most of the PyKEEN pipeline was built around the scoring and evaluation of head/tail corruptions. However, there were a few places where we added in functionality for training where the...

Demonstrate use of PyG message passing in R-GCN

5

Closes #435 This PR abstracts a base class for message passing layers and updates its forward signature to better match PyG. It then gives an incomplete demo on how you...

- [x] Enable passing `pathlib.Path` objects in addition to strings - [x] Fixes #451 where ERModels crash the pipeline (done in #453 and dafad330) - [ ] Create dummy ablation...

This PR is a (first/bad) attempt to incorporate MuRP (for #343). It heavily borrows from Balažević's original implementation at https://github.com/ibalazevic/multirelational-poincare and does its best to give attribution. This PR does...

PyKEEN currently implements two training loops: 1. The local closed world assumption (LCWA). People often mistakenly call this the _closed world assumption_) 2. The _stochastic_ closed world assumption (sLCWA). People...