Overfitting Python Packages

monte-neo

Independent verifier for trading strategies written by AI agents and humans: catches look-ahead bias, hidden costs and overfitting. MCP server, CLI, GitHub Action, signed certificates.

ml4t-diagnostic

Signal diagnostics, statistical validation, and backtest evaluation for quantitative trading workflows.

omega-lock

Catch overfitting before you ship it: a walk-forward / held-out gate for any optimizer (Optuna, sweeps, prompt search). The highest score isn't deployable until it survives data it never saw. CLI + Python.

omegaprompt

The overfit gate for your prompts: re-test the winning prompt on held-out examples it never tuned on, and block the ship if it doesn't generalize. Sits on top of promptfoo/DSPy. CLI + MCP + CI.

spatial-kfold

spatial resampling for more robust cross validation in spatial studies

anchortest

Stop trusting a single backtest run: anchor-average across every phase, compare candidates correctly, catch blending artifacts before you ship a strategy.

falsify-edge

Gate a backtested trading edge before you trust it: multiple-testing, cost-floor, and autocorrelation checks in pure-stdlib Python.

ftdoctor

Did your fine-tune overfit? Which checkpoint should you keep? What can you delete? For any HuggingFace Trainer run, local or on the Hub: TRL, Unsloth, LLaMA-Factory, Axolotl.

quant-harness

Honest backtest harness with a measured detector: leakage-safe CPCV, Deflated Sharpe, PBO, realistic costs, mechanical PASS/KILL verdict. Strategy audit CLI + verdict API.

falsify-quant

Adversarial backtest validation for trading strategies. Runs your parameter sweep so it knows the real trial count, then spends six tests trying to prove the result is noise: lookahead detection, cost breakeven, Deflated Sharpe, PBO, permutation, regime concentration.

quanttro

Open-source Python toolkit for quantitative research on Borsa Istanbul (BIST): data, features, walk-forward backtesting, risk/portfolio, backtest robustness (CPCV/PBO), diagnostics, reports, paper-trading bridge.

nullbar

Pre-registration and honest statistics for trading research: trial ledgers, deflated Sharpe, clustered inference, leak detection, fill realism.

quantros

策略实盘就绪度证伪器——五门判决,回测能不能实盘,先过五门

oos-lab

Research-grade validation toolkit for trading strategies — detect backtest overfitting (PSR, Deflated Sharpe, PBO/CSCV, walk-forward, Harvey-Liu haircut). MIT.

holdout-first

Validation harness that fits on a small slice and demands survival on a large held-out remainder, across instruments and periods, under a hard parameter budget.

lambda-guard-boosting

Overfitting detection for Gradient Boosting — no validation set required

agent-walkforward

Walk-forward / out-of-sample validation for agent evals — detect eval-set overfitting the way quants detect backtest overfitting. Zero dependencies.

quant-integrity

Statistical tests for whether a backtest means anything: deflated Sharpe, PBO, purged CV, reality checks, attestation verification.

nullius

Falsify-first backtest diagnostics: lookahead detection, skill placebos, honest forecast evaluation, and multiple-testing deflation. In development.

sklearn-diagnose

🔍 AI-powered diagnosis for Scikit-learn models: Detect overfitting, data leakage, class imbalance & more with LLM-generated insights

sk-autod

Auto-diagnose deep learning training pathologies: overfitting, unstable LR, exploding gradients, noisy training, and early stopping issues.

perception-xalpha-lite

Backtest overfitting audit for factor research: probability of backtest overfitting (PBO), deflated Sharpe ratio, point-in-time data, purged walk-forward. Searches published factor libraries and reports what actually survived costs.

skepsis

Is your backtest result real, or overfit? Statistical diagnostics for backtest overfitting.