Instructions for running the model:  

1) Install the required packages using requirements.txt 
2) Replace the Dataset PATH to where it is stored locally. 
3) Replace the Probability Matrix to the place where it is locally stored. 
4) Set the Curriculum threshold values.
5) Change bert-base-uncased to bert-base-multilingual-uncased incase running for languages other than english. 
6) Replace the num_label with the number of labels in the dataset 
7) Number of training samples in the dataframe 


We have used some standard GLUE Datasets that could be downloaded using the transformer dataset or their official webpage.

For the Matrix, run the code given in matrix.py.

The code is well documented for further explanation.