# Compute metrics from predictions + actuals
python scripts/compute_metrics.py --predictions preds.npy --actuals actuals.npy --timestamps ts.npy

# Generate full evaluation report
python scripts/generate_report.py --results folds.jsonl --output report.md

pip install -r requirements.txt
# Or: pip install numpy>=1.24 pandas>=2.0 scipy>=1.10

def weekly_sharpe(pnl: np.ndarray, timestamps: np.ndarray) -> float:
    """Sharpe with daily aggregation for range bars."""
    daily_pnl = _group_by_day(pnl, timestamps)  # Sum PnL per calendar day
    if len(daily_pnl) < 2 or np.std(daily_pnl) == 0:
        return 0.0
    daily_sharpe = np.mean(daily_pnl) / np.std(daily_pnl)
    # For crypto (7-day week): sqrt(7). For equities: sqrt(5)
    return daily_sharpe * np.sqrt(7)  # Crypto default

from scipy.stats import spearmanr

def information_coefficient(predictions: np.ndarray, actuals: np.ndarray) -> float:
    """Spearman rank IC - captures magnitude alignment."""
    ic, _ = spearmanr(predictions, actuals)
    return ic  # Range: [-1, 1]. >0.02 acceptable, >0.05 good, >0.10 excellent

from scipy.stats import norm

def psr(sharpe: float, se: float, benchmark: float = 0.0) -> float:
    """P(true Sharpe > benchmark)."""
    return norm.cdf((sharpe - benchmark) / se)

go_criteria:
  - positive_sharpe_rate > 0.55
  - mean_weekly_sharpe > 0
  - cv_fold_returns < 1.5
  - mean_hit_rate > 0.50

publication_criteria:
  - binomial_pvalue < 0.05
  - psr > 0.85
  - dsr > 0.50 # If n_trials > 1

# ALWAYS check for model collapse after prediction
pred_std = np.std(predictions)
if pred_std < 1e-6:
    logger.warning(
        f"Constant predictions detected (std={pred_std:.2e}). "
        "Model collapsed to mean - check architecture."
    )

# BEFORE (causes collapse on range bars)
HIDDEN_SIZE = 16
DROPOUT = 0.5

# AFTER (prevents collapse)
HIDDEN_SIZE = 48  # Triple capacity
DROPOUT = 0.3     # Less aggressive regularization