Skip to content

README.md

Latest commit

 

History

History
91 lines (64 loc) · 2.61 KB

README.md

File metadata and controls

91 lines (64 loc) · 2.61 KB
Raw

Cloud Service

This directory contains the canonical cloud-side source for the current Wand Brain deployment.

Main Pieces

  • main.py repo-level FastAPI entrypoint that wraps the live runtime with persistence, leaderboards, node control, and the frontend.
  • start_script.sh preferred launcher for the current cloud app.
  • frontend/ single-page polling dashboard served by the same FastAPI process.
  • database/ SQLAlchemy configuration and persisted models.
  • backend/ live UDP ingest, rendering, and scoring runtime packages.
  • data/ local runtime files such as the SQLite database, node-control state, and generated output images.

What This Subtree Delivers

The cloud side is not just a packet sink. In the final system it provides:

  • UDP ingest for wand point packets
  • live reconstruction of active strokes
  • rasterized drawing previews
  • template scoring and result persistence
  • node control over HTTP
  • a single-page live console for operation and debugging
  • leaderboard and recent-attempt views

Canonical vs Runtime

Edit the source in this directory tree for GitHub.

Do not treat EC2 deployment copies such as cloud/alt_live_console as the long-term source of truth. Those are runtime artifacts of deployment.

Start Here

For most contributors, the most useful files are:

Quick Run

To run the cloud service locally:

cd software/cloud
bash start_script.sh --install-deps

Then open:

http://127.0.0.1:8000/

For a step-by-step rerun path, see:

API Documentation

The detailed API guides live under the protocol docs:

Those documents explain the served Brain features by API family, including:

  • live plotting and attempt images
  • templates and scoring
  • node control
  • leaderboards and persistence

Architecture Reports

More formal backend reports live under: