Quick Start¶
Export-first path: install → research export → optional policy triage → optional holdout. The interactive CLI wizard is optional and documented at the bottom.
1. Install¶
pip install monte-neo
# Apple Silicon extras (MLX / Metal bindings):
pip install "monte-neo[apple]"
Isolated CLI: pipx install "monte-neo[apple]".
Full matrix: Installation.
2. Export API (primary)¶
Fee-aware SMA sweep on synthetic OHLC — same shape as
examples/export_sma_sweep_quickstart.py:
from monte_neo.backtest import ExecutionModel, export_sma_sweep, synthetic_ohlcv
ohlc = synthetic_ohlcv(100_000, seed=42)
model = ExecutionModel(commission_bps=5.0, slippage_bps=5.0, warmup_bars=50)
out = export_sma_sweep(
ohlc["open"],
ohlc["high"],
ohlc["low"],
ohlc["close"],
combos=16,
model=model,
device="auto", # Metal when safe on Apple Silicon; else cpu_numba
)
print(
{
"device": out.get("device"),
"fallback_reason": out.get("fallback_reason"),
"combos": out.get("combos"),
"ok": out.get("ok"),
}
)
Aligned with the Home snippet. More detail: Research export API.
3. Policy triage (optional)¶
After you save an export_sma_sweep / export_batch JSON:
monte-neo --policy-triage path/to/export.json
See Research policy.
4. Holdout smoke (optional)¶
monte-neo --holdout-sma --holdout-bars 20000 --holdout-combos 32
See Holdout.
CLI wizard (optional)¶
The interactive menu is still available if you prefer a guided loop.
uv run monte-neo
# or: monte-neo
Typical flow:
- Download Market Data — searchable symbols (e.g.
BTCUSDT), timeframe, history length - Set Target Metrics — e.g. profit factor threshold
- Generate Indicator — search / optimize candidates
Dynamic Mode (evolutionary)¶
- In configuration, select
dynamicamong indicator types - Random search finds candidates first
- With enough candidates, evolutionary optimization cross-breeds / mutates survivors
- Results can be checked with Monte Carlo helpers
Tips: arrow keys navigate; Space toggles checkboxes. Deep dive: Dynamic Indicators.