This is a Streamlit app for exploring student feedback, comparing sentiment models, and predicting sentiment for new feedback.

streamlit run app.py

The app includes a small repository-local sample dataset, so it opens immediately after installation. Use the sidebar to upload a CSV with these columns:

preprocessed_text, sentiment_label, emotion_tag, subject_specific_context

• The app uses a repository-local dataset at data/Labelled_stories.txt
• The deployment entry point is app.py
• The hosting command is defined in Procfile
• Set OPENROUTER_API_KEY in the host's environment variables to enable AI-generated summaries; the core analytics work without it

Create a new Web Service from this repository with:

• Build command: pip install -r requirements.txt
• Start command: streamlit run app.py --server.port $PORT --server.address 0.0.0.0
• Environment variable: OPENROUTER_API_KEY (optional)

The service will provide a public URL after the first successful deploy. Keep the repository and dataset available in the deployed service because the default dashboard reads the bundled sample file. Once deployed, the URL: you never commit it, Streamlit generates it for you.