• Install Pytorch 1.11.0 and CUDA 10.0
• Clone this repo

git clone https://github.com/rxy1234/CCRL
cd CLSD

• Download CRAG dataset
  
• Download ISC dataset
  
• Put the data under ./data/
• Divide the dataset into six folders myTest_Data myTest_Label myTraining_Data myTraining_Label myValid_Data myValid_Label
• Fill the folder myTest_Label with images with pixel values of all 0 corresponding to the folder myTest_Data

• Open ema_train.txt in the folder scripts
• Copy content after modifying parameters
• Open the terminal and cd 'the path of the current project'
• Paste the copied operation command and run it

• Download Test dataset
  
• Put the test data under ./data/
• Download Parameter
  
• Put the parameter under ./exp/
• Open ema_test.txt in the folder scripts
• Copy content after modifying parameters
• Open the terminal and cd 'the path of the current project'
• Paste the copied operation command and run it

Some code is reused from the Pytorch implementation of mean teacher.