5 dépôts
Processes for preparing and cleaning user interaction datasets specifically for recommendation models.
Distinct from Recommendation Datasets: Shortlist candidates are either general datasets or recommendation engines, not the engineering process
Explore 5 awesome GitHub repositories matching artificial intelligence & ml · Recommendation Data Engineering. Refine with filters or upvote what's useful.
This project is a recommendation system framework designed for building, evaluating, and operationalizing personalized item suggestion engines. It provides a comprehensive toolkit for implementing collaborative filtering and content-based algorithms, supported by an end-to-end machine learning pipeline for preparing datasets and deploying predictive models. The framework distinguishes itself through the integration of knowledge graphs to provide richer context for recommendations and the use of industry-specific patterns to accelerate system deployment. It also includes a specialized model ev
Provides a comprehensive workflow for preparing, cleaning, and splitting interaction datasets for ML models.
ParlAI is a conversational AI research framework designed for training, evaluating, and sharing dialogue models using a unified interface for datasets and agents. It functions as a PyTorch-based training platform and a dialogue data collection system, providing a centralized model zoo for the distribution of versioned pretrained agents. The project distinguishes itself through a knowledge-grounded retrieval system that combines dense and sparse indexing to ground responses in external information. It also provides a comprehensive infrastructure for gathering human-AI interaction data via inte
Provides utilities to print and preview the first few examples of a task dataset.
Gorse is a personalized recommendation engine server and machine learning pipeline designed to suggest items to users based on their behavior and preferences. It operates as a distributed system that separates training, candidate generation, and serving nodes to support high-throughput workloads. The system utilizes a multi-stage recommendation pipeline to refine results through retrieval, scoring, and reranking. It generates personalized suggestions using collaborative filtering, matrix factorization, and item-to-item similarity models, while also providing non-personalized and fallback reco
Allows retrieving and filtering users, items, and categories to verify the current state of the recommendation dataset.
dlt est un outil d'ingestion de données Python et un framework de pipeline ETL conçu pour récupérer des données depuis diverses sources et les persister dans des destinations structurées. Il fonctionne comme un moteur d'inférence de schéma qui détecte automatiquement les types de données et aplatit les structures JSON imbriquées en tables relationnelles, déplaçant les données des sources vers des lakehouses, des entrepôts ou des bases de données vectorielles. Le projet se distingue par une génération de pipeline alimentée par l'IA, utilisant de grands modèles de langage pour échafauder le code d'extraction et les connecteurs pour les API REST. Il prend également en charge le stockage vectoriel multimodal et la population spécialisée de bases de données vectorielles pour prendre en charge les applications d'IA et de machine learning. Le framework couvre un large éventail de capacités, incluant l'évolution automatique du schéma, le chargement incrémentiel de données via le suivi d'état et la validation de la qualité des données par l'application de contrats de données. Il fournit des outils pour la normalisation des données relationnelles, les transformations pré- et post-chargement, et une variété d'adaptateurs de destination pour les bases de données SQL et les magasins d'objets cloud. L'observabilité est gérée via des tableaux de bord d'exécution de pipeline, le suivi de lignage des colonnes et la vérification de version de schéma utilisant des hachages basés sur le contenu.
Allows for the inspection of destination datasets using read-only SQL queries and row count retrieval.
Neural collaborative filtering is a recommendation system framework that predicts user item preferences from implicit feedback by combining generalized matrix factorization and multi-layer perceptron networks through a shared final embedding layer. It captures both linear and non-linear interactions to model user preferences from historical data. The framework executes training and evaluation runs through a configuration-driven pipeline accessible via command-line interfaces, parsing hyperparameters such as learning rates, batch sizes, and latent dimensions. It optimizes implicit feedback mod
Organizes user interaction history into training ratings, positive test instances, and negative samples for evaluation pipelines.