Refresh trainer README

Add observation context features to trainer

docs: Update README with scheduler, model strategy and targets

Frontend

Align trainer feature flow and metadata

Defer trainer startup work

Revert trainer UI and startup experiment

Use JSON outputs in trainer UI

Simplify trainer startup path

Lazy-load sklearn metrics for startup

Quote Python version in space config

Pin Python and use CPU-only XGBoost

Defer xgboost import and trim deps

Add startup and runtime diagnostics

Handle empty model registry gracefully

Downgrade Gradio to 5.47.2 for HF compatibility

Fix: server_name, activate scheduler, proper main block

Update: disable ssr_mode, fix scheduler

Train Aarhus multi-signal model bundles

Update app.py

Update app.py

Update app.py

Update app.py

Update requirements.txt

Update app.py

Update app.py

Update requirements.txt

Update app.py

Update app.py

Update requirements.txt

Update requirements.txt

Update app.py

Create app.py

Create requirements.txt

initial commit