Develop A New Model

Develop Your Own Learning-to-Rank Method

PT-Ranking offers deep neural networks as the basis to construct a scoring function based on PyTorch and can thus fully leverage the advantages of PyTorch. NeuralRanker is a class that represents a general learning-to-rank model. A key component of NeuralRanker is the neural scoring function. The configurable hyper-parameters include activation function, number of layers, number of neurons per layer, etc. All specific learning-to-rank models inherit NeuralRanker and mainly differ in the way of computing the training loss. The following figure shows the main step in developing a new learning-to-rank model based on Empirical Risk Minimization, where batch_preds and batch_stds correspond to outputs of the scoring function and ground-truth lables, respectively. We can observe that the main work is to define the surrogate loss function.

When incorporating a newly developed model (say ModelA), it is commonly required to develop the subclass ModelAParameter by inheriting ModelParameter and customizing the functions, such as to_para_string(), default_para_dict() and grid_search(). Please refer to Configuration for detailed description on parameter setting and LambdaRankParameter as an example.

To fully leverage PT-Ranking, one needs to be familiar with PyTorch.

For detailed introduction on learning-to-rank, please refer to the book: Learning to Rank for Information Retrieval.