Skip to content

feat: gate optimization by dof multiplier #116

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
AlexanderPietsch wants to merge 2 commits into from
Closed

feat: gate optimization by dof multiplier #116

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
80d7822
feat: gate optimization by dof multiplier
AlexanderPietsch Feb 22, 2026
4ac86ed
feat: make min bars configurable
AlexanderPietsch Feb 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 20 additions & 0 deletions src/backtest/runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -514,6 +514,26 @@ def run_all(self, only_cached: bool = False) -> list[BestResult]:
else:
search_space[name] = options

n_params = len(search_space)
dof_multiplier = self.cfg.param_dof_multiplier
min_bars_floor = self.cfg.param_min_bars
min_bars_for_optimization = max(min_bars_floor, dof_multiplier * n_params)
if search_space and len(df) < min_bars_for_optimization:
self.logger.info(
"skipping optimization due to insufficient bars",
collection=col.name,
symbol=symbol,
timeframe=timeframe,
bars=len(df),
min_bars=min_bars_for_optimization,
n_params=n_params,
dof_multiplier=dof_multiplier,
min_bars_floor=min_bars_floor,
strategy=strat.name,
search_method=search_method,
)
search_space = {}
Copy link

@cursor cursor bot Feb 22, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Search space parameters silently dropped when gate triggers

High Severity

When the data-sufficiency gate triggers and clears search_space = {}, parameters that had multiple options are silently lost. They are never transferred into fixed_params, so the subsequent evaluate({}) call constructs full_params without them. The strategy then runs with missing parameters, likely producing incorrect results or failures. The search-space parameters need default values (e.g., first option from each list) added to fixed_params before clearing search_space.

Fix in Cursor Fix in Web

Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot Feb 22, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Keep tunable params when skipping optimization

When data is below the new threshold, search_space is cleared entirely, so run_all() later calls evaluate({}) and drops every parameter that originally had multiple candidate values. For strategies that require those parameters (e.g., external adapters/classes expecting constructor args from the grid), this turns a previously valid run into generate_signals failures or silently changes behavior on short datasets. Instead of erasing the space, the gate should collapse each tunable parameter to a deterministic single value (such as the first option) so required keys are still passed.

Useful? React with 👍 / 👎.


best_val = -np.inf
best_params: dict[str, Any] | None = None
best_stats: dict[str, Any] | None = None
Expand Down
4 changes: 4 additions & 0 deletions src/config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,6 +50,8 @@ class Config:
engine: str = "pybroker" # pybroker engine
param_search: str = "grid" # grid | optuna
param_trials: int = 25
param_dof_multiplier: int = 100
param_min_bars: int = 2000
max_workers: int = 1
asset_workers: int = 1
param_workers: int = 1
Expand Down Expand Up @@ -127,6 +129,8 @@ def load_config(path: str | Path) -> Config:
engine=str(raw.get("engine", "pybroker")).lower(),
param_search=str(raw.get("param_search", raw.get("param_optimizer", "grid"))).lower(),
param_trials=int(raw.get("param_trials", raw.get("opt_trials", 25))),
param_dof_multiplier=int(raw.get("param_dof_multiplier", 100)),
param_min_bars=int(raw.get("param_min_bars", 2000)),
max_workers=int(raw.get("max_workers", raw.get("asset_workers", 1))),
asset_workers=int(raw.get("asset_workers", raw.get("max_workers", 1))),
param_workers=int(raw.get("param_workers", 1)),
Expand Down