Streamlit app for the Benchmark

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
nltk.download('punkt')
source .env
streamlit run app.py

ToDo

remove table of uploads
average token lenghts
model selection
add to the blue score
add recall, precision and f1
paragraph similarity -> sentence transformers / sentencepiece library -> https: //sbert.net/ --> cosine