README.md

March 16, 2021 ยท View on GitHub

Introduction

The source code and models for our paper Jo-SRC: A Contrastive Approach for Combating Noisy Labels

Framework

framework

Installation

After creating a virtual environment of python 3.6, run pip install -r requirements.txt to install all dependencies

How to use

The code is currently tested only on GPU.

  • Data preparation

    Created a folder Datasets and download cifar100/clothing1m/food101n dataset into this folder.

  • Source code

    • If you want to train the whole model from beginning using the source code, please follow subsequent steps:
      • Prepare data
      • Modify GPU device in the corresponding train script xxx.sh in scripts folder
      • Activate virtual environment (e.g. conda) and then run
      bash scripts/xxx.sh

  • Demo

    • If you just want to do a quick test on the model, please follow subsequent steps:
      • Prepare data
      • Download one of the following trained model 
        wget https://josrc.oss-cn-shanghai.aliyuncs.com/clothing1m_r18_71.78.pth
wget https://josrc.oss-cn-shanghai.aliyuncs.com/food101n_r50_86.66.pth
      • Modify GPU, ARCH, MODEL, DATASET, and NCLASSES accordingly in the demo script demo.sh in scripts folder
      • Activate virtual environment (e.g. conda) and then run 
        bash scripts/demo.sh