AI Apps Tutorial

Minimal Streamlit apps for the course "Software Tools and Techniques for AI".

Each file adds one new idea, so students can run the versions in order:

pip install -r requirements.txt
streamlit run v1.py
streamlit run v2.py
...
streamlit run v10.py

The final deployable app is app.py, which imports v10.py.

What each version teaches

File	New idea
v1.py	title, text, basic app structure
v2.py	text input and button
v3.py	slider, selectbox, radio, simple logic
v4.py	sidebar, columns, metrics, progress
v5.py	dataframe and charts
v6.py	file upload and sample CSV download
v7.py	text area and simple text analysis
v8.py	checkbox, markdown, and LaTeX
v9.py	a simple sklearn classifier with cached loading
v10.py	a final sklearn app with charts, tabs, upload, markdown, and export

Suggested lab flow

1. Run v1.py to show how quickly a web app appears.
2. Move to v2.py to explain reruns and buttons.
3. Use v3.py and v4.py to cover widgets and layout.
4. Use v5.py and v6.py for data display and CSV upload.
5. Use v7.py and v8.py for text display, markdown, and math.
6. Use v9.py to show how sklearn fits into a Streamlit app.
7. End with v10.py, then deploy app.py.

Local setup

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
streamlit run app.py

Run Streamlit from the repository root so local paths behave the same way as deployment.

Deployment

• Streamlit Community Cloud: see deploy/streamlit-community-cloud.md
• Hugging Face Spaces: use the separate hf-spaces-deploy-tutorial repo

Notes for teaching

• The apps stay self-contained so students can open one file and understand it.
• The examples are student-friendly and now include a minimal sklearn model.
• Markdown, LaTeX, checkbox widgets, charts, and CSV upload are all covered before deployment.