Streamlit app for the Benchmark

How to run it

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