harry/ForexBot2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Audit fixes: mixed-regime bias, live==backtest session parity, parallel replay

Browse Source 
Fix 1 — mixed-"trending" regime (ADX strong, EMAs unaligned) was
hard-coded bullish in 5 sites: RSI reinterpretation used the uptrend
playbook (RSI 38-55 -> STRONG_BUY conf 75), BUY-only confidence
bonuses, BUY-only extension penalties, no counter-trend penalty.
Now: extremes-only RSI (as transitioning), no directional bonuses,
symmetric extension penalties. 6yr validation: H1 +67.2->+61.9R
(mixed skew 16B/11S -> 4B/7S), H4 +31.6->+27.5R (the 4 bias BUYs),
M15 +152.6->+154.7R, M5 +207.1->+206.1R with maxDD 28.7->24.3R.
The ~8R aggregate give-back is bull-market subsidy from the bug.

Fix 2 — live session filter now applies the EXACT replay-loop
predicate (session window + blocked hours on the signal bar's CLOSE
time, from the bar timestamp not wall clock). Replaces a wall-clock
filter gated on a "XAU STRATEGY" marker that no code writes, which
made live diverge from every validated book.

Fix 3 — delete dead _check_xau_oversold_bounce and
_generate_xau_trade_setup (unreachable; the latter hardcoded BUY and
skipped the spread/RR gates).

Backtester: parallel bar replay via forked workers (--workers,
default auto = cores-2 capped 12). Per-bar evaluation is pure; the
entry cooldown is applied in an ordered phase-2 pass. Proven
byte-identical to sequential on the same window; ~300 -> 1030-1680
bars/s (6yr M5: ~26min -> 5-8min). --workers 1 = sequential path.

Tests: 179 passed (new mixed-regime RSI regression test).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Harwinder Sekhon
2026-07-11 22:57:37 -04:00
parent f3ba4d56a3
commit 53d8bc4390
5 changed files with 282 additions and 287 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CLAUDE.md
+1 -1
View File
@@ -116,7 +116,7 @@ The `backtest/` directory contains `gpu_backtest.py` — a GPU-accelerated backt
Fill realism: charges a round-trip spread per trade (`--spread`; defaults to the pair's `default_spread` in PAIR_INFO), fills gap-through SL/TP at the bar open, and resolves trades against the RAW price path (the parquet cache stores raw bars; validation/cleaning runs on load — signals see cleaned data, trade resolution sees real spikes). Entry is at the next bar's open after the signal bar. Weekend/session filters evaluate bar-close (decision) time to match the live bot's wall clock.
GPU note: CuPy needs the NVIDIA runtime pip wheels (`nvidia-cuda-runtime-cu12`, `nvidia-cuda-nvrtc-cu12`, `nvidia-curand-cu12`, `nvidia-nvjitlink-cu12`) or it silently falls back to NumPy — check the "Backend:" line in the run log. The bar-replay loop itself is single-threaded Python either way (~340 bars/s); GPU only accelerates indicator precompute and SL/TP scanning.
GPU note: CuPy needs the NVIDIA runtime pip wheels (`nvidia-cuda-runtime-cu12`, `nvidia-cuda-nvrtc-cu12`, `nvidia-curand-cu12`, `nvidia-nvjitlink-cu12`) or it silently falls back to NumPy — check the "Backend:" line in the run log. GPU accelerates indicator precompute and SL/TP scanning; the bar-replay loop is Python (~300 bars/s/core) but fans out across forked worker processes (`--workers`, default auto = cores−2 capped at 12, ~1700 bars/s at 10). Per-bar signal evaluation is pure, so parallel and sequential books are identical (the entry cooldown is applied in an ordered pass afterwards); `--workers 1` forces the sequential path.
## Code Conventions