Install the required dependencies:

pip install -r ./requirements.txt

Download the original DEFT Corpus inside the project folder:

git clone https://github.com/adobe-research/deft_corpus

Perform the preprocessing step (preprocessed data will be found inside ./dataset/preprocessed/):

cd ./dataset
python ./preprocessing.py

Perform the augmentation step (preprocessed data will be found inside ./dataset/augmented/):

cd ./dataset
python ./augmentation.py

Run the Jupyter Notebook (./TL_SimonePersiani.ipynb).

The original dataset is attributed to AdobeResearch.