feat: add Docker image and container usage

Author: Kyle Spengler

Dockerfile

@@ -1,10 +1,25 @@
 FROM python:3.11-slim
+
+# 1) base setup
 WORKDIR /app
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
+
+# 2) copy source
 COPY serving_app ./serving_app
-COPY training ./training
-RUN python -m training.train || true
+COPY training    ./training
+
+# 3) model artifact location (inside the image)
+ENV MODEL_PATH=/app/models/model.pkl
+RUN mkdir -p /app/models
+
+# 4) train during build so the model is baked into the image
+RUN python -m training.train
+
+# 5) runtime config
 ENV PORT=8000
 EXPOSE 8000
-CMD ["uvicorn","serving_app.main:app","--host","0.0.0.0","--port","8000"]
+
+# 6) single CMD that respects $PORT
+CMD ["sh", "-c", "uvicorn serving_app.main:app --host 0.0.0.0 --port ${PORT:-8000}"]
+