search

pyg-team / pytorch_geometric

Graph Neural Network Library for PyTorch
https://pyg.org
MIT License
21.08k stars 3.63k forks source link

Add `RetailHero` and `MovieLens25` bipartite datasets with causal parameters #9471

Open geopanag opened 3 months ago

geopanag commented 3 months ago

I would like to add two new bipartite networks in PyG. One is based on the RetailHero (https://ods.ai/competitions/x5-retailhero-uplift-modeling/data) (users-buy-products) with causal outcomes and interventions from a marketing campaign, and the other is based on MovieLens25 (https://grouplens.org/datasets/movielens/25m/) (movies-ratedby-users) with observational causal parameters .

I use them to test a GNN for causal outcome prediction in https://hal.science/hal-04601553/document ( ECMLPKDD 2024 ) and I believe they re useful for this line of research.

codecov[bot] commented 3 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 86.87%. Comparing base (dafbd30) to head (a63ab81). Report is 37 commits behind head on master.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #9471 +/- ## ========================================== - Coverage 88.25% 86.87% -1.38% ========================================== Files 473 463 -10 Lines 30844 30617 -227 ========================================== - Hits 27221 26599 -622 - Misses 3623 4018 +395 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

geopanag commented 1 month ago

Since there 's been some time can I ask, I am not sure if I should make any corrections e.g. for the changelog and typehints, or I should first wait for the CR. Are these checks blocking the CR?