Scruff AI
56c71c87b2
- cuda/ — main LBM kernel (khra_gixx_1024_v5.cu) - navigator/ — lattice_observer, golden_weave, bridges, mock daemon - scripts/ — compile, start, launch, setup (paths updated) - docs/ — system manual - archive/ — everything else (old kernels, inquiries, experiments) - README.md — full setup guide: requirements, quick start, use your own LLM - removed training/ entirely (broken LoRA scripts + datasets) - .gitignore: exclude build/ logs/ training/ *.jsonl
27 lines
718 B
Bash
27 lines
718 B
Bash
#!/bin/bash
|
|
# Start CUDA daemon + lattice observer
|
|
REPO_ROOT="$(cd "$(dirname "$0")/.." && pwd)"
|
|
cd "$REPO_ROOT"
|
|
|
|
mkdir -p build logs
|
|
|
|
echo "[LAUNCHER] Starting khra_gixx_1024_v5 daemon..."
|
|
nohup ./build/khra_gixx_1024_v5 > logs/v5_stdout.log 2> logs/v5_stderr.log &
|
|
DAEMON_PID=$!
|
|
echo "[LAUNCHER] Daemon PID: $DAEMON_PID"
|
|
|
|
# Wait for ZMQ ports to bind
|
|
sleep 3
|
|
|
|
# Verify daemon is running
|
|
if kill -0 $DAEMON_PID 2>/dev/null; then
|
|
echo "[LAUNCHER] Daemon is running."
|
|
else
|
|
echo "[LAUNCHER] ERROR: Daemon failed to start. Check logs/v5_stderr.log"
|
|
cat logs/v5_stderr.log
|
|
exit 1
|
|
fi
|
|
|
|
echo "[LAUNCHER] Starting lattice_observer.py..."
|
|
exec python3 navigator/lattice_observer.py 2>&1 | tee logs/observer.log
|