pykt-team/pykt-toolkit

pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models

What it solves

pyKT provides a standardized way to train and benchmark deep learning-based knowledge tracing (DLKT) models. It eliminates the need to implement multiple models from scratch by providing a unified library for comparing different knowledge tracing approaches across various datasets.

How it works

Built on PyTorch, the library integrates several key components:

  • Data Preprocessing: Standardized procedures for over 7 popular datasets across different domains.
  • Model Implementations: Over 10 frequently compared deep learning knowledge tracing (DLKT) models.
  • Evaluation: Five detailed prediction scenarios to ensure transparent and extensive experimentation.

Who it’s for

Researchers and developers working on educational AI, specifically those focused on modeling student proficiency and predicting learning outcomes through knowledge tracing.

Highlights

  • Comprehensive Model Suite: Includes a wide range of DLKT approaches for easy comparison.
  • Standardized Benchmarking: Provides integrated preprocessing for multiple datasets and specific prediction scenarios.
  • PyTorch-based: Leverages the PyTorch ecosystem for deep learning training.

Related

  • Project
    neuralhydrology/neuralhydrology

    A PyTorch-based Python library designed for deep learning research in hydrology, allowing users to train neural networks via configuration files.

  • Project
    sktime/pytorch-forecasting

    A PyTorch-based library for time series forecasting that provides a high-level API and state-of-the-art deep learning architectures like Temporal Fusion Transformers and N-BEATS.

  • Project
    yihaosun1124/OfflineRL-Kit

    A PyTorch library for offline reinforcement learning that provides a standardized framework and implementations of state-of-the-art model-free and model-based algorithms.

  • Project
    open-mmlab/mmengine

    A foundational PyTorch library that serves as a training engine for deep learning models, providing standardized components for training loops, configuration, and monitoring.

  • Project
    VLARLKit/VLARLKit

    A PyTorch library for training Vision-Language-Action (VLA) models using reinforcement learning, featuring a decoupled architecture to prevent dependency conflicts between models and simulators.