Score-Based Model for Low-Rank Tensor Recovery

 

Overview

Installation

1. Download source code and dataset:

   • git clone https://github.com/CZY-Code/ScoreTR.git
   • Download datasets
     • Alog
     • Acc
     • Air
     • Click
     • Color Image
     • CAVE
     • Videos

2. Pip install dependencies:

   • OS: Ubuntu 20.04.6
   • nvidia :
     • cuda: 12.1
     • cudnn: 8.5.0
   • python == 3.9.18
   • pytorch >= 2.1.0
   • Python packages: pip install -r requirements.txt

Dataset Preparation

Unzip and move dataset into ROOT/data

Directory structure of dataset

    |- data      
    |   |- beijing.npy
    |   |- ctr_50k.npy
    |   |- acc
    |   |- alog       
    │   |- misc              
    │   |- MSIs
    |   |- Videos

Train and Test

• run ./run_scoreTR.sh

Acknowledgement

This implementation is based on / inspired by:

• EnergyTD
• reproducible-tensor-completion-state-of-the-art
• M²DMT
• HLRTF
• Continuous-Tensor-Toolbox
• DeepTensor