1 repositorio
Processes raw interaction files into standardized internal representations for ML models.
Distinct from Data Transformations: Shortlist candidates are too specific to binary files, pixels, or caching, missing general ML data transformation.
Explore 1 awesome GitHub repository matching data & databases · Interaction Log Transformations. Refine with filters or upvote what's useful.
RecBole is a PyTorch-based recommendation framework designed for building, training, and evaluating a wide variety of recommendation algorithms. It serves as a standardized benchmark environment that allows for the comparison of different model architectures using public datasets and consistent evaluation metrics. The project provides specialized toolkits for sequential recommendation and knowledge-graph integration, enabling the prediction of item sequences based on user history or the incorporation of structured external knowledge. It includes a dedicated hyperparameter optimization engine
Transforms raw input files into a unified internal representation for standardized algorithmic consumption.