Visualization

Plotting utilities for confusion matrices, benchmark comparisons, and learning curves.

pitch_sequencing.evaluation.visualization

Visualization utilities for evaluation results.

plot_ablation_results(ablation_df, ablation_type='feature')

Plot ablation study results.

Parameters:

Name	Type	Description	Default
ablation_df	DataFrame	DataFrame with 'variant' and 'accuracy' columns (and optionally 'ci_low', 'ci_high').	required
ablation_type	str	Type label for chart title.	'feature'

Source code in src/pitch_sequencing/evaluation/visualization.py

def plot_ablation_results(
    ablation_df: pd.DataFrame,
    ablation_type: str = "feature",
) -> plt.Figure:
    """Plot ablation study results.

    Args:
        ablation_df: DataFrame with 'variant' and 'accuracy' columns (and optionally 'ci_low', 'ci_high').
        ablation_type: Type label for chart title.
    """
    fig, ax = plt.subplots(figsize=(10, 6))

    variants = ablation_df["variant"]
    means = ablation_df["accuracy"]

    if "ci_low" in ablation_df.columns:
        errors = np.array([
            means - ablation_df["ci_low"],
            ablation_df["ci_high"] - means,
        ])
        ax.barh(range(len(variants)), means, xerr=errors, capsize=4, color="teal", alpha=0.8)
    else:
        ax.barh(range(len(variants)), means, color="teal", alpha=0.8)

    ax.set_yticks(range(len(variants)))
    ax.set_yticklabels(variants)
    ax.set_xlabel("Accuracy")
    ax.set_title(f"Ablation Study: {ablation_type.replace('_', ' ').title()}")
    ax.grid(axis="x", alpha=0.3)
    plt.tight_layout()
    return fig

plot_benchmark_comparison(results_df, metric='accuracy')

Plot grouped bar chart comparing models with CI error bars.

Expects results_df to have columns: model, metric_mean, metric_ci_low, metric_ci_high.

Source code in src/pitch_sequencing/evaluation/visualization.py

def plot_benchmark_comparison(
    results_df: pd.DataFrame,
    metric: str = "accuracy",
) -> plt.Figure:
    """Plot grouped bar chart comparing models with CI error bars.

    Expects results_df to have columns: model, metric_mean, metric_ci_low, metric_ci_high.
    """
    fig, ax = plt.subplots(figsize=(10, 6))
    models = results_df["model"]
    means = results_df[f"{metric}_mean"]
    ci_low = results_df[f"{metric}_ci_low"]
    ci_high = results_df[f"{metric}_ci_high"]
    errors = np.array([means - ci_low, ci_high - means])

    bars = ax.bar(range(len(models)), means, yerr=errors, capsize=5, color="steelblue", alpha=0.8)
    ax.set_xticks(range(len(models)))
    ax.set_xticklabels(models, rotation=30, ha="right")
    ax.set_ylabel(metric.replace("_", " ").title())
    ax.set_title(f"Model Comparison: {metric.replace('_', ' ').title()}")
    ax.grid(axis="y", alpha=0.3)

    for bar, mean in zip(bars, means):
        ax.text(bar.get_x() + bar.get_width() / 2, bar.get_height() + 0.005,
                f"{mean:.3f}", ha="center", va="bottom", fontsize=9, fontweight="bold")

    plt.tight_layout()
    return fig

plot_confusion_matrix(y_true, y_pred, labels=None, title='Confusion Matrix')

Plot a confusion matrix heatmap.

Source code in src/pitch_sequencing/evaluation/visualization.py

def plot_confusion_matrix(
    y_true: np.ndarray,
    y_pred: np.ndarray,
    labels: Optional[List[str]] = None,
    title: str = "Confusion Matrix",
) -> plt.Figure:
    """Plot a confusion matrix heatmap."""
    fig, ax = plt.subplots(figsize=(8, 6))
    cm = confusion_matrix(y_true, y_pred)
    disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=labels)
    disp.plot(cmap="Blues", ax=ax)
    ax.set_title(title)
    plt.tight_layout()
    return fig

plot_feature_importance(importance_dict)

Plot horizontal bar chart of feature importance.

Source code in src/pitch_sequencing/evaluation/visualization.py

def plot_feature_importance(importance_dict: Dict[str, float]) -> plt.Figure:
    """Plot horizontal bar chart of feature importance."""
    fig, ax = plt.subplots(figsize=(8, 5))
    features = list(importance_dict.keys())
    values = list(importance_dict.values())
    sorted_idx = np.argsort(values)
    ax.barh([features[i] for i in sorted_idx], [values[i] for i in sorted_idx], color="coral", alpha=0.8)
    ax.set_xlabel("Importance (accuracy drop)")
    ax.set_title("Feature Importance (Leave-One-Out)")
    ax.grid(axis="x", alpha=0.3)
    plt.tight_layout()
    return fig

plot_learning_curves(history, title='')

Plot train/val loss and accuracy curves.

Source code in src/pitch_sequencing/evaluation/visualization.py

def plot_learning_curves(history: Dict[str, List[float]], title: str = "") -> plt.Figure:
    """Plot train/val loss and accuracy curves."""
    fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

    ax1.plot(history["train_losses"], label="Train Loss")
    ax1.plot(history["val_losses"], label="Val Loss")
    ax1.set_xlabel("Epoch")
    ax1.set_ylabel("Loss")
    ax1.set_title(f"Loss Curves{f' — {title}' if title else ''}")
    ax1.legend()
    ax1.grid(True, alpha=0.3)

    ax2.plot(history["val_accuracies"], label="Val Accuracy", color="green")
    ax2.set_xlabel("Epoch")
    ax2.set_ylabel("Accuracy")
    ax2.set_title(f"Validation Accuracy{f' — {title}' if title else ''}")
    ax2.legend()
    ax2.grid(True, alpha=0.3)

    plt.tight_layout()
    return fig