Introduction

source code for our paper GASAnet: Generative Adversarial and Self Attention Based Fine-Grained Cross-Modal Retrieval

Network Architecture

Installation

Requirement

• pytorch, tested on [v1.0]
• CUDA, tested on v9.0
• Language: Python 3.6

How to use

The code is currently tested only on GPU.

• Download dataset
  Please visit dataset.
• Prepare audio data
  python audio.py
• Training
  • If you want to train the whole model from beginning using the source code, please follow the subsequent steps.
    • Download dataset to the dataset folder.
    • In main_gan_lstm_resnet.py

    • modify lr in params1 to 0.001, lr in params2 and lr in discriminator to 1.
    • modify model_path to the path where you want to save your parameters of networks.

    • Activate virtual environment (e.g. conda) and then run the script
      python main_gan_lstm_resnet.py
• Testing
  • If you just want to do a quick test on the model and check the final retrieval performance, please follow the subsequent steps.
    • The trained models of our work can be downloaded from Baidu Cloud, and the extraction code is v99c.
    • Activate virtual environment (e.g. conda) and then run the script
      python test_gan.py