henry9zhr/QANet-pytorch

A PyTorch implementation of QANet.

What it solves

This project provides a PyTorch implementation of QANet, a model designed for Question Answering (QA) tasks, allowing users to train and test the model on datasets like SQuAD.

How it works

The system uses a neural network architecture defined in models.py to process question and context text. It includes a preprocessing step (preproc.py) to convert raw datasets into tensors, a training loop in main.py, and a configuration file for hyperparameters.

Who it’s for

Researchers and developers interested in implementing or replicating the paper "QANet: Convolutional Question Answering Networks" using PyTorch.

Highlights

  • PyTorch implementation of the QANet architecture.
  • Includes scripts for dataset downloading and preprocessing.
  • Support for training and testing modes.
  • Community-driven bug fixes regarding positional encodings and weight sharing.

Related

  • Project
    mit-han-lab/torchquantum

    A PyTorch-based framework for simulating quantum computations on classical hardware, enabling the training of parameterized quantum circuits and hybrid classical-quantum models.

  • 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
    Talmaj/onnx2pytorch

    A library that transforms ONNX models into PyTorch models, allowing users to utilize PyTorch's backend and neural network manipulation tools.

  • Project
    k2-fsa/sherpa

    An open-source speech-to-text inference framework using PyTorch for deploying pre-trained transducer and CTC-based models.

  • Project
    qiskit-community/qiskit-machine-learning

    A library providing fundamental building blocks for quantum machine learning, enabling the implementation of quantum kernels, neural networks, and hybrid quantum-classical models.