2026-08-18 02:49:20 +02:00
|
|
|
"""The gate: no speed number is published for a workload the engines disagree on.
|
|
|
|
|
|
2026-08-20 16:16:33 +02:00
|
|
|
A benchmark between backtesters is only a benchmark if they did the same work.
|
|
|
|
|
This module compares what a challenger produced against the reference and
|
|
|
|
|
classifies the result, and ``bench.py`` refuses to report a timing for anything
|
|
|
|
|
it classifies as a failure.
|
|
|
|
|
|
|
|
|
|
Every comparison is one challenger against the reference. Challengers are never
|
|
|
|
|
joined to each other: three engines make three pairs, and a table of pairs is a
|
|
|
|
|
matrix, not a benchmark. It also would not add anything, since agreement with
|
|
|
|
|
the reference is transitive enough for the only question being asked here, which
|
|
|
|
|
is whether a published timing describes the same simulation.
|
2026-08-18 02:49:20 +02:00
|
|
|
|
|
|
|
|
Three verdicts:
|
|
|
|
|
|
|
|
|
|
``exact``
|
|
|
|
|
Agreement down to float-reordering noise. The timing is publishable.
|
|
|
|
|
``documented``
|
2026-08-20 16:16:33 +02:00
|
|
|
The two disagree, the workload declared it in advance for this specific
|
|
|
|
|
engine, and the reason is written down. The timing goes to the annex with
|
|
|
|
|
the reason attached.
|
2026-08-18 02:49:20 +02:00
|
|
|
``failed``
|
2026-08-20 16:16:33 +02:00
|
|
|
They disagree and nobody predicted it. That is a finding about the engines,
|
|
|
|
|
not about their speed: the timing is withheld.
|
2026-08-18 02:49:20 +02:00
|
|
|
"""
|
|
|
|
|
from __future__ import annotations
|
|
|
|
|
|
|
|
|
|
from typing import Any, Dict
|
|
|
|
|
|
2026-08-20 16:16:33 +02:00
|
|
|
from engines import ENGINES, REFERENCE
|
|
|
|
|
from workloads import CAPITAL, expectation, why
|
2026-08-18 02:49:20 +02:00
|
|
|
|
|
|
|
|
# Float reordering across two implementations of the same arithmetic lands
|
|
|
|
|
# around 1e-13 of the account on a million bars. Anything above this is a
|
|
|
|
|
# different decision somewhere, not a different summation order.
|
|
|
|
|
REL_TOL = 1e-9
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _rel(a: float, b: float) -> float:
|
|
|
|
|
"""Plain relative difference, for quantities that are not money."""
|
|
|
|
|
return abs(a - b) / max(1e-12, abs(b))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _vs_capital(a: float, b: float) -> float:
|
|
|
|
|
"""Difference as a fraction of the money at risk, not of the result itself.
|
|
|
|
|
|
|
|
|
|
Anchoring on the result breaks exactly when the result is interesting: a
|
|
|
|
|
strategy that ends near zero equity, or near zero return, turns a difference
|
|
|
|
|
of a hundredth of a cent into a 1% relative error and fails a comparison the
|
|
|
|
|
engines actually passed. The account size is the stable yardstick."""
|
|
|
|
|
return abs(a - b) / CAPITAL
|
|
|
|
|
|
|
|
|
|
|
2026-08-20 16:16:33 +02:00
|
|
|
def compare(ref: Dict[str, Any], other: Dict[str, Any], key: str, engine: str) -> Dict[str, Any]:
|
|
|
|
|
expected = expectation(key, engine)
|
2026-08-18 02:49:20 +02:00
|
|
|
|
|
|
|
|
diffs = {
|
2026-08-20 16:16:33 +02:00
|
|
|
"final_equity_vs_capital": _vs_capital(ref["final_equity"], other["final_equity"]),
|
|
|
|
|
"round_trips_delta": ref["round_trips"] - other["round_trips"],
|
|
|
|
|
"total_fees_vs_capital": _vs_capital(ref["total_fees"], other["total_fees"]),
|
2026-08-18 02:49:20 +02:00
|
|
|
}
|
|
|
|
|
agrees = (
|
|
|
|
|
diffs["final_equity_vs_capital"] <= REL_TOL
|
|
|
|
|
and diffs["round_trips_delta"] == 0
|
|
|
|
|
and diffs["total_fees_vs_capital"] <= REL_TOL
|
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
# Workloads that also produce a performance summary are gated on the
|
2026-08-20 16:16:33 +02:00
|
|
|
# drawdown, which every engine here computes at full bar resolution and
|
|
|
|
|
# which must match. The ratios are reported but not gated: the reference
|
|
|
|
|
# buckets its daily returns slightly differently from vectorbt, a difference
|
|
|
|
|
# worth stating rather than hiding, and worth nothing at all as an argument
|
|
|
|
|
# about speed.
|
|
|
|
|
if other.get("max_drawdown") is not None and ref.get("max_drawdown") is not None:
|
|
|
|
|
diffs["max_drawdown_rel"] = _rel(ref["max_drawdown"], other["max_drawdown"])
|
2026-08-18 02:49:20 +02:00
|
|
|
agrees = agrees and diffs["max_drawdown_rel"] <= REL_TOL
|
2026-08-20 16:16:33 +02:00
|
|
|
# Only between engines that annualise the same way. raptorbt returns its
|
|
|
|
|
# ratios on its own basis, and subtracting those from the reference's
|
|
|
|
|
# would publish a units mismatch as a disagreement (measured: Sharpe
|
|
|
|
|
# 0.21 against 8.14 on a run whose equity curve is bit-identical).
|
|
|
|
|
if ENGINES[engine].ratio_basis == ENGINES[REFERENCE].ratio_basis:
|
|
|
|
|
diffs["advisory_ratio_rel"] = {
|
|
|
|
|
name: _rel(ref[name], other[name])
|
|
|
|
|
for name in ("sharpe", "sortino", "volatility")
|
|
|
|
|
if ref.get(name) is not None and other.get(name) is not None
|
|
|
|
|
}
|
|
|
|
|
else:
|
|
|
|
|
diffs["ratio_basis"] = ENGINES[engine].ratio_basis
|
2026-08-18 02:49:20 +02:00
|
|
|
|
|
|
|
|
if agrees:
|
|
|
|
|
status = "exact"
|
|
|
|
|
elif expected == "documented":
|
|
|
|
|
status = "documented"
|
|
|
|
|
else:
|
|
|
|
|
status = "failed"
|
|
|
|
|
|
|
|
|
|
return {
|
|
|
|
|
"status": status,
|
|
|
|
|
"expected": expected,
|
|
|
|
|
"publishable": status == "exact",
|
|
|
|
|
"diffs": diffs,
|
2026-08-20 16:16:33 +02:00
|
|
|
"metrics": {REFERENCE: ref, engine: other},
|
|
|
|
|
"note": why(key, engine) if status == "documented" else "",
|
2026-08-18 02:49:20 +02:00
|
|
|
}
|