Add Hugging Face Spaces (Docker SDK) deploy config for the live demo

- Dockerfile runs the Streamlit app on port 8501 (HF create API only accepts
gradio|docker|static, so Streamlit is served via Docker)
- README.md front-matter: sdk: docker, app_port: 8501
- en_core_web_lg model wheel in requirements.txt (installs during image build)
- DEPLOY_HF_SPACES.md with corrected hf-CLI + Docker steps

Deploy branch only — push to the Space's main, don't merge into main.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

DEPLOY_HF_SPACES.md

@@ -0,0 +1,44 @@
+# Deploy the live demo on Hugging Face Spaces
+
+This branch (`feat/hf-spaces-demo`) is the deploy branch. HF's create API only accepts
+`gradio | docker | static` as the SDK, so the Streamlit app runs as a **Docker Space**. It adds:
+
+1. **`Dockerfile`** — runs `streamlit run app/streamlit_app.py` on port 8501.
+2. **`README.md` front-matter** — `sdk: docker`, `app_port: 8501`.
+3. **The spaCy model in `requirements.txt`** — `en_core_web_lg` installs during the image build.
+
+Everything else (the `noteguard/` package, `app/streamlit_app.py`) already works as-is. The Space
+downloads the NHS dataset from Hugging Face at first run; if that fails it falls back to paste-only mode.
+
+## One-time deploy
+
+```bash
+# 1) Log in (needs a HF token with WRITE scope: https://huggingface.co/settings/tokens)
+hf auth login
+
+# 2) Create the Space (Docker SDK, free cpu-basic). Or make it at https://huggingface.co/new-space
+hf repos create <HF_USERNAME>/noteguard --repo-type space --space-sdk docker
+
+# 3) Push THIS branch to the Space's main branch
+git remote add space https://huggingface.co/spaces/<HF_USERNAME>/noteguard
+git push space feat/hf-spaces-demo:main
+```
+
+> Uses the `hf` CLI from `huggingface_hub` >= 1.0 (the old `huggingface-cli` still works as a
+> deprecated alias). `hf auth login` will prompt for your write token.
+
+The Space builds (installs Presidio + spaCy + `en_core_web_lg`, ~3–5 min the first time), then serves
+the app. Your **Live Demo Link** is:
+
+```
+https://huggingface.co/spaces/<HF_USERNAME>/noteguard
+```
+
+Paste that into the submission form's *Live Demo Link* field.
+
+## Notes
+- **Resource-constrained?** If the build is slow or the Space OOMs, swap the model line in
+  `requirements.txt` to `en_core_web_sm-3.8.0` (faster, lighter; slightly lower name recall).
+- **Updating the demo:** push again with `git push space feat/hf-spaces-demo:main` (the Space rebuilds).
+- **Keep this branch deploy-only** — don't merge the README front-matter into `main` (it's only for the
+  Space card).

Dockerfile

@@ -0,0 +1,24 @@
+# Hugging Face Spaces (Docker SDK) — serves the NoteGuard Streamlit app.
+# HF's create API only accepts gradio|docker|static, so Streamlit runs via Docker.
+FROM python:3.11-slim
+
+# Non-root user (HF Spaces convention) → gives a writable HOME for model/dataset caches.
+RUN useradd -m -u 1000 user
+USER user
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH \
+    PYTHONUNBUFFERED=1
+
+WORKDIR $HOME/app
+
+# Install deps first for layer caching. requirements.txt pins the spaCy model wheel.
+COPY --chown=user requirements.txt .
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r requirements.txt
+
+# App code (data/ and .venv are gitignored, so the app downloads the dataset at runtime).
+COPY --chown=user . .
+
+EXPOSE 8501
+CMD ["streamlit", "run", "app/streamlit_app.py", \
+     "--server.port=8501", "--server.address=0.0.0.0", "--server.headless=true"]

README.md

@@ -1,3 +1,15 @@
+---
+title: NoteGuard — NHS De-Identification Gate
+emoji: 🛡️
+colorFrom: blue
+colorTo: indigo
+sdk: docker
+app_port: 8501
+pinned: false
+license: mit
+short_description: Sanitise-at-source PII removal for NHS notes
+---
+
 # 🛡️ NoteGuard
 
 **Automatic PII sanitisation for NHS clinical notes — clean data in, no identifiers out.**

requirements.txt

@@ -7,3 +7,7 @@ huggingface_hub
 pandas
 streamlit
 pytest
+
+# spaCy model (installed at build time on Hugging Face Spaces; matches detect.py default).
+# Swap to en_core_web_sm-3.8.0 for a faster/lighter build if the Space is resource-constrained.
+https://github.com/explosion/spacy-models/releases/download/en_core_web_lg-3.8.0/en_core_web_lg-3.8.0-py3-none-any.whl