Learning to Rank in PyTorch
The recent representative methods (such as MO4SRD and DALETOR) for Search Result Diversification by directly optimizing the evaluation metric (e.g., alpha-nDCG) have been added. (02/22/2022)
Different types of neural scoring functions are supported now, namely pointwise neural scoring function (mainly consists of feedforward layers) and listwise neural scoring function (mainly builds upon multi-head self-attention Layer). (02/22/2022)
This open-source project, referred to as PTRanking (Learning-to-Rank in PyTorch) aims to provide scalable and extendable implementations of typical learning-to-rank methods based on PyTorch. On one hand, this project enables a uniform comparison over several benchmark datasets leading to an in-depth understanding of previous learning-to-rank methods. On the other hand, this project makes it easy to develop and incorporate newly proposed models, so as to expand the territory of techniques on learning-to-rank.
Key Features:
Please refer to the documentation site for more details.