Group: 12

Three different architectures can be found in this repo. Each of them can be run by simply installing the requirements.txt

python -m pip install -r requirements.txt

For the LSTM and GPT2, one first needs to create the dataset

cd Dataset
python create_dataset.py
cd ..

All individual models can be run by going into the respective map and running main.py

GPT2 and the LSTM were fully trained on the Google Colab servers using a Tesla T40 GPU.

A complete list of dependencies can be found in requirements.txt. Additionally, here is a list:

• https://github.com/klaudia-nazarko/nlg-text-generation (for inspiration and some code)