Max Berrendorf
Max Berrendorf
This small PR changes the calculation of rank values from a `argsort` based implementation to a faster version based on counting `>=` / `>` comparisons between paired positive and negative...
Hi, first of all thanks for the great package and for providing pre-built binaries. Is there already an estimate when binaries compatible with the [latest PyTorch release, 1.12](https://github.com/pytorch/pytorch/releases/tag/v1.12.0), will be...
Add an evaluator to call OGB evaluation from within the pipeline. Currently blocked by `Evaluator.evaluate` receiving multiple `kwargs` (including `additional_filter_triples`) to be passed to `evaluate`, which are not supported by...
Reference Implementation: https://github.com/LongYu-360/TripleRE-Add-NodePiece/blob/master/TripleRE%2BNodepiece/ogb_wikikg2/model.py _Note_: this PR also renames `NodePiece`'s `token_representation_kwargs` to `token_representations_kwargs` for consistency. #### Tasks: * [x] support choosing `"mlp"` aggregation for `NodePieceRepresentation` (moved from `NodePiece` model) * [x]...
This PR works towards making the evaluation in the `pipeline` function optional.
**Describe the benchmark KG dataset**: The dataset is shipped as part of [DGL-KE](https://github.com/awslabs/dgl-ke/tree/master/examples). It is a subset of the Freebase KG. It comprises around 86M entities, 15k relations and 338M...
This PR aims at adding fast sLCWA training to PyKEEN. For this, we exploit that negative samples are created by corruption, and thus share some parts of the triple, e.g.,...
Fixes #749 #### Tasks: * [x] remove `inverse_triples` flags from `TriplesFactory` and `Datasets`. * [x] add inverse triples handling to Model * [ ] update training loop / instances ####...
This PR adds a `get_config` method to `Loss` and its subclasses. The method is inspired by https://keras.io/api/layers/base_layer/#getconfig-method and can be used to obtain all parameters necessary to re-instantiate an equivalent...
This PR adds the HAKE and ModE interaction functions and modules to PyKEEN. **Learning Hierarchy-Aware Knowledge Graph Embeddings for Link Prediction.** _Zhanqiu Zhang, Jianyu Cai, Yongdong Zhang, Jie Wang._ AAAI...