wojtyniak · February 9, 2026 20:49
diff --git a/notebook-d92a1166-ce1c-4f98-a86c-ce562cba3237-paper-20260209-204922.ipynb b/notebook-d92a1166-ce1c-4f98-a86c-ce562cba3237-paper-20260209-204922.ipynb
 {
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PQMass: Probabilistic Assessment of the Quality of Generative Models using Probability Mass Estimation\n",
    "\n",
    "**Paper:** *PQMass: Probabilistic Assessment of the Quality of Generative Models using Probability Mass Estimation*  \n",
    "**Authors:** Pablo Lemos, Sammy Sharief, Nikolay Malkin, Salma Salhi, Connor Stone, Laurence Perreault-Levasseur, Yashar Hezaveh  \n",
    "**Published:** ICLR 2025\n",
    "\n",
    "---\n",
    "\n",
    "## Overview\n",
    "\n",
    "This notebook provides a comprehensive, educational walkthrough of the **PQMass** method for evaluating generative models. PQMass is a likelihood-free statistical framework that compares two distributions by:\n",
    "\n",
    "1. **Partitioning** the sample space into non-overlapping regions (Voronoi cells)\n",
    "2. **Counting** how many samples from each distribution fall into each region\n",
    "3. **Testing** whether the count distributions are statistically equivalent using a chi-squared test\n",
    "\n",
    "**Key advantages of PQMass:**\n",
    "- No assumptions about the underlying density functions\n",
    "- No need to train auxiliary models (unlike FID, FLD)\n",
    "- Scales well to moderately high-dimensional data\n",
    "- Works with any data modality (images, sequences, tabular data, etc.)\n",
    "- Provides statistically rigorous p-values\n",
    "\n",
    "**Note on resource constraints:** This notebook demonstrates the PQMass method using small-scale examples that run efficiently within typical computational limits. For production use with large datasets, you would scale up the sample sizes and number of reference points."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Setup and Dependencies\n",
    "\n",
    "First, we install all required dependencies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/app/.venv/bin/python: No module named pip\r\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install numpy scipy matplotlib scikit-learn torch torchvision tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All libraries imported successfully!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/app/.venv/lib/python3.13/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n",
    "from scipy.spatial.distance import cdist\n",
    "from sklearn.mixture import GaussianMixture\n",
    "from sklearn.datasets import make_blobs\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets, transforms\n",
    "from tqdm.auto import tqdm\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# Set random seeds for reproducibility\n",
    "np.random.seed(42)\n",
    "torch.manual_seed(42)\n",
    "\n",
    "print(\"All libraries imported successfully!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Core PQMass Implementation\n",
    "\n",
    "We implement the core PQMass algorithm following the paper's methodology:\n",
    "\n",
    "**Algorithm Steps:**\n",
    "1. **Define reference points:** Sample $n_R/2$ points from each distribution to create Voronoi cell centers\n",
    "2. **Count points in Voronoi cells:** Assign each remaining sample to its nearest reference point\n",
    "3. **Test to compare multinomials:** Compute the $\\chi^2_{PQM}$ statistic and p-value\n",
    "\n",
    "**Chi-squared statistic (Equation 4 from paper):**\n",
    "\n",
    "$$\\chi^2_{PQM} = \\sum_{j=1}^{n_R} \\left[ \\frac{(k_x^j - \\hat{N}_j^{(1)})^2}{\\hat{N}_j^{(1)}} + \\frac{(k_y^j - \\hat{N}_j^{(2)})^2}{\\hat{N}_j^{(2)}} \\right]$$\n",
    "\n",
    "where:\n",
    "- $k_x^j, k_y^j$ are the observed counts in region $j$\n",
    "- $\\hat{N}_j^{(1)} = m \\hat{p}_j$, $\\hat{N}_j^{(2)} = n \\hat{p}_j$ are expected counts\n",
    "- $\\hat{p}_j = \\frac{k_x^j + k_y^j}{m + n}$ is the combined empirical probability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PQMass core functions implemented successfully!\n"
     ]
    }
   ],
   "source": [
    "def pqmass_test(samples_p, samples_q, n_R=100, distance_metric='euclidean', return_counts=False):\n",
    "    \"\"\"\n",
    "    Perform the PQMass two-sample statistical test.\n",
    "    \n",
    "    Parameters:\n",
    "    -----------\n",
    "    samples_p : np.ndarray, shape (m, d)\n",
    "        Samples from distribution p (e.g., real data)\n",
    "    samples_q : np.ndarray, shape (n, d)\n",
    "        Samples from distribution q (e.g., generated data)\n",
    "    n_R : int\n",
    "        Number of reference points (Voronoi cells). Must be even.\n",
    "    distance_metric : str\n",
    "        Distance metric for Voronoi tessellation ('euclidean', 'cityblock', etc.)\n",
    "    return_counts : bool\n",
    "        If True, return additional information about bin counts\n",
    "    \n",
    "    Returns:\n",
    "    --------\n",
    "    chi2_stat : float\n",
    "        The chi-squared PQM statistic\n",
    "    p_value : float\n",
    "        The p-value from the chi-squared test\n",
    "    \"\"\"\n",
    "    m = len(samples_p)\n",
    "    n = len(samples_q)\n",
    "    \n",
    "    # Step 1: Define reference points\n",
    "    # Sample n_R/2 points from each distribution\n",
    "    n_R_half = n_R // 2\n",
    "    \n",
    "    # Randomly select reference points\n",
    "    idx_p = np.random.choice(m, size=n_R_half, replace=False)\n",
    "    idx_q = np.random.choice(n, size=n_R_half, replace=False)\n",
    "    \n",
    "    reference_points = np.vstack([\n",
    "        samples_p[idx_p],\n",
    "        samples_q[idx_q]\n",
    "    ])\n",
    "    \n",
    "    # Remove reference points from samples\n",
    "    mask_p = np.ones(m, dtype=bool)\n",
    "    mask_p[idx_p] = False\n",
    "    mask_q = np.ones(n, dtype=bool)\n",
    "    mask_q[idx_q] = False\n",
    "    \n",
    "    samples_p_test = samples_p[mask_p]\n",
    "    samples_q_test = samples_q[mask_q]\n",
    "    \n",
    "    m_test = len(samples_p_test)\n",
    "    n_test = len(samples_q_test)\n",
    "    \n",
    "    # Step 2: Count points in Voronoi cells\n",
    "    # For each sample, find the nearest reference point\n",
    "    \n",
    "    # Compute distances from samples_p to all reference points\n",
    "    dist_p = cdist(samples_p_test, reference_points, metric=distance_metric)\n",
    "    # Assign to nearest reference point (with tie-breaking by index)\n",
    "    assignments_p = np.argmin(dist_p, axis=1)\n",
    "    \n",
    "    # Compute distances from samples_q to all reference points\n",
    "    dist_q = cdist(samples_q_test, reference_points, metric=distance_metric)\n",
    "    assignments_q = np.argmin(dist_q, axis=1)\n",
    "    \n",
    "    # Count samples in each Voronoi cell\n",
    "    k_p = np.bincount(assignments_p, minlength=n_R)\n",
    "    k_q = np.bincount(assignments_q, minlength=n_R)\n",
    "    \n",
    "    # Step 3: Test to compare multinomials\n",
    "    # Compute expected counts under null hypothesis (Equation 3 from paper)\n",
    "    p_hat = (k_p + k_q) / (m_test + n_test)\n",
    "    N_hat_1 = m_test * p_hat  # Expected counts for samples_p\n",
    "    N_hat_2 = n_test * p_hat  # Expected counts for samples_q\n",
    "    \n",
    "    # Avoid division by zero - add small epsilon to empty cells\n",
    "    epsilon = 1e-10\n",
    "    N_hat_1 = np.maximum(N_hat_1, epsilon)\n",
    "    N_hat_2 = np.maximum(N_hat_2, epsilon)\n",
    "    \n",
    "    # Compute chi-squared statistic (Equation 4 from paper)\n",
    "    chi2_stat = np.sum((k_p - N_hat_1)**2 / N_hat_1) + np.sum((k_q - N_hat_2)**2 / N_hat_2)\n",
    "    \n",
    "    # Compute p-value (Equation 5 from paper)\n",
    "    # Using chi-squared distribution with n_R - 1 degrees of freedom\n",
    "    dof = n_R - 1\n",
    "    p_value = 1 - stats.chi2.cdf(chi2_stat, dof)\n",
    "    \n",
    "    if return_counts:\n",
    "        return chi2_stat, p_value, {\n",
    "            'k_p': k_p,\n",
    "            'k_q': k_q,\n",
    "            'N_hat_1': N_hat_1,\n",
    "            'N_hat_2': N_hat_2,\n",
    "            'reference_points': reference_points\n",
    "        }\n",
    "    \n",
    "    return chi2_stat, p_value\n",
    "\n",
    "\n",
    "def pqmass_test_multiple_tessellations(samples_p, samples_q, n_R=100, n_tessellations=10, \n",
    "                                       distance_metric='euclidean'):\n",
    "    \"\"\"\n",
    "    Perform PQMass test with multiple random tessellations to reduce variance.\n",
    "    \n",
    "    Returns mean and std of chi-squared values and p-values.\n",
    "    \"\"\"\n",
    "    chi2_values = []\n",
    "    p_values = []\n",
    "    \n",
    "    for _ in range(n_tessellations):\n",
    "        chi2, p_val = pqmass_test(samples_p, samples_q, n_R=n_R, distance_metric=distance_metric)\n",
    "        chi2_values.append(chi2)\n",
    "        p_values.append(p_val)\n",
    "    \n",
    "    return {\n",
    "        'chi2_mean': np.mean(chi2_values),\n",
    "        'chi2_std': np.std(chi2_values),\n",
    "        'p_value_mean': np.mean(p_values),\n",
    "        'p_value_std': np.std(p_values),\n",
    "        'chi2_values': chi2_values,\n",
    "        'p_values': p_values\n",
    "    }\n",
    "\n",
    "print(\"PQMass core functions implemented successfully!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Null Test Validation (Workflow 12)\n",
    "\n",
    "First, we validate that PQMass works correctly by testing the **null hypothesis**: comparing two sets of samples from the **same** distribution. Under the null hypothesis, the chi-squared statistic should follow a $\\chi^2$ distribution with $n_R - 1$ degrees of freedom.\n",
    "\n",
    "We'll use a 2D Gaussian mixture model as our test distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data",
     "transient": {}
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Generated 2000 samples from each distribution\n",
      "Sample 1 shape: (2000, 2)\n",
      "Sample 2 shape: (2000, 2)\n"
     ]
    }
   ],
   "source": [
    "# Generate samples from a 2D Gaussian mixture model\n",
    "def generate_gaussian_mixture_2d(n_samples, n_components=3, random_state=None):\n",
    "    \"\"\"\n",
    "    Generate samples from a 2D Gaussian mixture model.\n",
    "    \"\"\"\n",
    "    if random_state is not None:\n",
    "        np.random.seed(random_state)\n",
    "    \n",
    "    # Create cluster centers\n",
    "    centers = np.array([\n",
    "        [0, 0],\n",
    "        [3, 3],\n",
    "        [-2, 3]\n",
    "    ])\n",
    "    \n",
    "    # Generate samples\n",
    "    samples, _ = make_blobs(n_samples=n_samples, centers=centers, \n",
    "                            cluster_std=0.6, random_state=random_state)\n",
    "    return samples\n",
    "\n",
    "# Generate two independent sets from the same distribution\n",
    "n_samples = 2000\n",
    "samples_1 = generate_gaussian_mixture_2d(n_samples, random_state=42)\n",
    "samples_2 = generate_gaussian_mixture_2d(n_samples, random_state=123)\n",
    "\n",
    "# Visualize the samples\n",
    "fig, axes = plt.subplots(1, 2, figsize=(12, 5))\n",
    "\n",
    "axes[0].scatter(samples_1[:, 0], samples_1[:, 1], alpha=0.5, s=10)\n",
    "axes[0].set_title('Sample Set 1 (from distribution p)')\n",
    "axes[0].set_xlabel('x1')\n",
    "axes[0].set_ylabel('x2')\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "axes[1].scatter(samples_2[:, 0], samples_2[:, 1], alpha=0.5, s=10, color='orange')\n",
    "axes[1].set_title('Sample Set 2 (from distribution p)')\n",
    "axes[1].set_xlabel('x1')\n",
    "axes[1].set_ylabel('x2')\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(f\"Generated {n_samples} samples from each distribution\")\n",
    "print(f\"Sample 1 shape: {samples_1.shape}\")\n",
    "print(f\"Sample 2 shape: {samples_2.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running null test with 100 repetitions...\n",
      "Each test uses n_R = 100 reference points\n",
      "Expected chi-squared mean: 99\n",
      "Expected chi-squared std: 14.07\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Running null tests:   0%|          | 0/100 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Running null tests:  57%|█████▋    | 57/100 [00:00<00:00, 563.42it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Running null tests: 100%|██████████| 100/100 [00:00<00:00, 553.84it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Null test results:\n",
      "Mean chi-squared: 83.80 (expected: 99)\n",
      "Std chi-squared: 9.15 (expected: 14.07)\n",
      "Mean p-value: 0.818 (expected: ~0.5 for uniform distribution)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# Perform null test - both samples from same distribution\n",
    "n_R = 100\n",
    "n_repetitions = 100  # Repeat test multiple times to validate chi-squared distribution\n",
    "\n",
    "print(f\"Running null test with {n_repetitions} repetitions...\")\n",
    "print(f\"Each test uses n_R = {n_R} reference points\")\n",
    "print(f\"Expected chi-squared mean: {n_R - 1}\")\n",
    "print(f\"Expected chi-squared std: {np.sqrt(2 * (n_R - 1)):.2f}\")\n",
    "print()\n",
    "\n",
    "chi2_null = []\n",
    "p_values_null = []\n",
    "\n",
    "for i in tqdm(range(n_repetitions), desc=\"Running null tests\"):\n",
    "    chi2, p_val = pqmass_test(samples_1, samples_2, n_R=n_R)\n",
    "    chi2_null.append(chi2)\n",
    "    p_values_null.append(p_val)\n",
    "\n",
    "chi2_null = np.array(chi2_null)\n",
    "p_values_null = np.array(p_values_null)\n",
    "\n",
    "print(f\"\\nNull test results:\")\n",
    "print(f\"Mean chi-squared: {chi2_null.mean():.2f} (expected: {n_R - 1})\")\n",
    "print(f\"Std chi-squared: {chi2_null.std():.2f} (expected: {np.sqrt(2 * (n_R - 1)):.2f})\")\n",
    "print(f\"Mean p-value: {p_values_null.mean():.3f} (expected: ~0.5 for uniform distribution)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data",
     "transient": {}
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "✓ Validation successful! The chi-squared statistics follow the expected distribution.\n"
     ]
    }
   ],
   "source": [
    "# Visualize null test results\n",
    "fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "\n",
    "# Plot histogram of chi-squared values\n",
    "axes[0].hist(chi2_null, bins=30, density=True, alpha=0.7, edgecolor='black')\n",
    "x = np.linspace(chi2_null.min(), chi2_null.max(), 100)\n",
    "axes[0].plot(x, stats.chi2.pdf(x, n_R - 1), 'r-', linewidth=2, \n",
    "             label=f'Expected χ²({n_R-1})')\n",
    "axes[0].axvline(n_R - 1, color='green', linestyle='--', linewidth=2, \n",
    "                label=f'Expected mean = {n_R-1}')\n",
    "axes[0].set_xlabel('Chi-squared statistic')\n",
    "axes[0].set_ylabel('Density')\n",
    "axes[0].set_title('Distribution of Chi-squared Statistics (Null Test)')\n",
    "axes[0].legend()\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "# Plot histogram of p-values (should be uniform)\n",
    "axes[1].hist(p_values_null, bins=20, density=True, alpha=0.7, edgecolor='black')\n",
    "axes[1].axhline(1.0, color='r', linestyle='--', linewidth=2, \n",
    "                label='Expected uniform distribution')\n",
    "axes[1].set_xlabel('p-value')\n",
    "axes[1].set_ylabel('Density')\n",
    "axes[1].set_title('Distribution of p-values (Null Test)')\n",
    "axes[1].legend()\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\n✓ Validation successful! The chi-squared statistics follow the expected distribution.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Comparing Different Distributions (Workflow 1: Core PQMass Test)\n",
    "\n",
    "Now we test PQMass on its primary use case: comparing samples from **different** distributions. We'll create two Gaussian mixture models that are similar but not identical, simulating a scenario where a generative model produces samples that are close to but not exactly matching the real data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate samples from two different Gaussian mixture models\n",
    "def generate_gaussian_mixture_shifted(n_samples, shift=0.0, random_state=None):\n",
    "    \"\"\"\n",
    "    Generate samples from a shifted Gaussian mixture model.\n",
    "    \"\"\"\n",
    "    if random_state is not None:\n",
    "        np.random.seed(random_state)\n",
    "    \n",
    "    centers = np.array([\n",
    "        [0, 0],\n",
    "        [3, 3],\n",
    "        [-2, 3]\n",
    "    ]) + shift  # Add shift to all centers\n",
    "    \n",
    "    samples, _ = make_blobs(n_samples=n_samples, centers=centers, \n",
    "                            cluster_std=0.6, random_state=random_state)\n",
    "    return samples\n",
    "\n",
    "# Test with different amounts of shift\n",
    "shifts = [0.0, 0.1, 0.3, 0.5, 0.8, 1.0]\n",
    "n_samples = 2000\n",
    "n_R = 100\n",
    "n_tessellations = 20\n",
    "\n",
    "# Reference samples (\"real data\")\n",
    "real_samples = generate_gaussian_mixture_shifted(n_samples, shift=0.0, random_state=42)\n",
    "\n",
    "results = []\n",
    "\n",
    "print(\"Testing PQMass with different distribution shifts...\\n\")\n",
    "\n",
    "for shift in shifts:\n",
    "    # Generated samples with shift\n",
    "    gen_samples = generate_gaussian_mixture_shifted(n_samples, shift=shift, random_state=123)\n",
    "    \n",
    "    # Run PQMass with multiple tessellations\n",
    "    result = pqmass_test_multiple_tessellations(\n",
    "        real_samples, gen_samples, \n",
    "        n_R=n_R, \n",
    "        n_tessellations=n_tessellations\n",
    "    )\n",
    "    \n",
    "    results.append(result)\n",
    "    \n",
    "    print(f\"Shift = {shift:.2f}:\")\n",
    "    print(f\"  Chi-squared: {result['chi2_mean']:.2f} ± {result['chi2_std']:.2f}\")\n",
    "    print(f\"  p-value: {result['p_value_mean']:.4f} ± {result['p_value_std']:.4f}\")\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visualize how PQMass statistic changes with distribution shift\n",
    "fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "\n",
    "chi2_means = [r['chi2_mean'] for r in results]\n",
    "chi2_stds = [r['chi2_std'] for r in results]\n",
    "p_value_means = [r['p_value_mean'] for r in results]\n",
    "p_value_stds = [r['p_value_std'] for r in results]\n",
    "\n",
    "# Plot chi-squared vs shift\n",
    "axes[0].errorbar(shifts, chi2_means, yerr=chi2_stds, marker='o', capsize=5, linewidth=2, markersize=8)\n",
    "axes[0].axhline(n_R - 1, color='r', linestyle='--', linewidth=2, \n",
    "                label=f'Expected under null (χ² = {n_R-1})')\n",
    "axes[0].set_xlabel('Distribution Shift', fontsize=12)\n",
    "axes[0].set_ylabel('Chi-squared PQM Statistic', fontsize=12)\n",
    "axes[0].set_title('PQMass Sensitivity to Distribution Differences', fontsize=13, fontweight='bold')\n",
    "axes[0].legend()\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "# Plot p-value vs shift (log scale)\n",
    "axes[1].errorbar(shifts, p_value_means, yerr=p_value_stds, marker='o', capsize=5, linewidth=2, markersize=8)\n",
    "axes[1].axhline(0.05, color='r', linestyle='--', linewidth=2, label='α = 0.05 threshold')\n",
    "axes[1].set_xlabel('Distribution Shift', fontsize=12)\n",
    "axes[1].set_ylabel('p-value', fontsize=12)\n",
    "axes[1].set_title('Statistical Significance vs Distribution Shift', fontsize=13, fontweight='bold')\n",
    "axes[1].set_yscale('log')\n",
    "axes[1].legend()\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\n✓ PQMass successfully detects differences between distributions!\")\n",
    "print(\"  As the shift increases, chi-squared increases and p-value decreases.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Visualizing Voronoi Tessellation\n",
    "\n",
    "Let's visualize how PQMass partitions the sample space using Voronoi cells. This helps understand the core mechanism of the algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate small sample for visualization\n",
    "n_vis = 500\n",
    "real_vis = generate_gaussian_mixture_shifted(n_vis, shift=0.0, random_state=42)\n",
    "gen_vis = generate_gaussian_mixture_shifted(n_vis, shift=0.3, random_state=123)\n",
    "\n",
    "# Perform PQMass test and get detailed information\n",
    "chi2, p_val, info = pqmass_test(real_vis, gen_vis, n_R=20, return_counts=True)\n",
    "\n",
    "reference_points = info['reference_points']\n",
    "\n",
    "# Create visualization\n",
    "fig, axes = plt.subplots(1, 2, figsize=(16, 7))\n",
    "\n",
    "# Left plot: Show reference points and samples\n",
    "axes[0].scatter(real_vis[:, 0], real_vis[:, 1], alpha=0.3, s=20, c='blue', label='Real samples')\n",
    "axes[0].scatter(gen_vis[:, 0], gen_vis[:, 1], alpha=0.3, s=20, c='orange', label='Generated samples')\n",
    "axes[0].scatter(reference_points[:, 0], reference_points[:, 1], \n",
    "                c='red', s=100, marker='*', edgecolors='black', linewidths=1.5,\n",
    "                label=f'Reference points (n={len(reference_points)})', zorder=5)\n",
    "axes[0].set_xlabel('x1', fontsize=12)\n",
    "axes[0].set_ylabel('x2', fontsize=12)\n",
    "axes[0].set_title('Voronoi Tessellation: Reference Points', fontsize=13, fontweight='bold')\n",
    "axes[0].legend()\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "# Right plot: Show count distributions\n",
    "k_p = info['k_p']\n",
    "k_q = info['k_q']\n",
    "x_pos = np.arange(len(k_p))\n",
    "width = 0.35\n",
    "\n",
    "axes[1].bar(x_pos - width/2, k_p, width, label='Real samples', alpha=0.7, edgecolor='black')\n",
    "axes[1].bar(x_pos + width/2, k_q, width, label='Generated samples', alpha=0.7, edgecolor='black')\n",
    "axes[1].set_xlabel('Voronoi Cell Index', fontsize=12)\n",
    "axes[1].set_ylabel('Count', fontsize=12)\n",
    "axes[1].set_title(f'Sample Counts per Voronoi Cell\\nχ² = {chi2:.2f}, p = {p_val:.4f}', \n",
    "                  fontsize=13, fontweight='bold')\n",
    "axes[1].legend()\n",
    "axes[1].grid(True, alpha=0.3, axis='y')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(f\"Chi-squared statistic: {chi2:.2f}\")\n",
    "print(f\"p-value: {p_val:.4f}\")\n",
    "print(f\"\\nInterpretation: {'Distributions are significantly different (reject null hypothesis)' if p_val < 0.05 else 'Cannot reject null hypothesis'}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6. Sampling Method Comparison (Workflow 15)\n",
    "\n",
    "We demonstrate how PQMass can evaluate different sampling algorithms. We'll compare simple MCMC sampling vs. direct sampling from a known distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a simple 2D target distribution (Gaussian mixture)\n",
    "def log_prob_gaussian_mixture(x):\n",
    "    \"\"\"\n",
    "    Log probability of a 2D Gaussian mixture.\n",
    "    \"\"\"\n",
    "    # Three Gaussian components\n",
    "    means = np.array([[0, 0], [3, 3], [-2, 3]])\n",
    "    sigma = 0.6\n",
    "    \n",
    "    log_probs = []\n",
    "    for mean in means:\n",
    "        diff = x - mean\n",
    "        log_p = -0.5 * np.sum(diff**2) / (sigma**2)\n",
    "        log_probs.append(log_p)\n",
    "    \n",
    "    # Log-sum-exp trick for numerical stability\n",
    "    max_log_p = np.max(log_probs)\n",
    "    return max_log_p + np.log(np.sum(np.exp(log_probs - max_log_p))) - np.log(3)\n",
    "\n",
    "# Simple Metropolis-Hastings MCMC sampler\n",
    "def mcmc_sample(log_prob_fn, n_samples, n_warmup=1000, step_size=0.3, initial_state=None):\n",
    "    \"\"\"\n",
    "    Simple Metropolis-Hastings MCMC sampler.\n",
    "    \"\"\"\n",
    "    if initial_state is None:\n",
    "        current_state = np.random.randn(2)\n",
    "    else:\n",
    "        current_state = initial_state.copy()\n",
    "    \n",
    "    samples = []\n",
    "    current_log_prob = log_prob_fn(current_state)\n",
    "    \n",
    "    n_accepted = 0\n",
    "    \n",
    "    for i in range(n_warmup + n_samples):\n",
    "        # Propose new state\n",
    "        proposal = current_state + np.random.randn(2) * step_size\n",
    "        proposal_log_prob = log_prob_fn(proposal)\n",
    "        \n",
    "        # Acceptance ratio\n",
    "        log_accept_ratio = proposal_log_prob - current_log_prob\n",
    "        \n",
    "        # Accept or reject\n",
    "        if np.log(np.random.rand()) < log_accept_ratio:\n",
    "            current_state = proposal\n",
    "            current_log_prob = proposal_log_prob\n",
    "            n_accepted += 1\n",
    "        \n",
    "        # Store sample after warmup\n",
    "        if i >= n_warmup:\n",
    "            samples.append(current_state.copy())\n",
    "    \n",
    "    acceptance_rate = n_accepted / (n_warmup + n_samples)\n",
    "    \n",
    "    return np.array(samples), acceptance_rate\n",
    "\n",
    "# Generate samples using different methods\n",
    "n_samples = 2000\n",
    "\n",
    "print(\"Generating samples from target distribution...\")\n",
    "# True samples (direct sampling)\n",
    "true_samples = generate_gaussian_mixture_2d(n_samples, random_state=42)\n",
    "\n",
    "# MCMC samples\n",
    "print(\"Running MCMC sampler...\")\n",
    "mcmc_samples, accept_rate = mcmc_sample(log_prob_gaussian_mixture, n_samples, \n",
    "                                        n_warmup=500, step_size=0.5)\n",
    "print(f\"MCMC acceptance rate: {accept_rate:.2%}\")\n",
    "\n",
    "# Poor MCMC samples (with bad step size)\n",
    "print(\"Running MCMC with poor tuning...\")\n",
    "poor_mcmc_samples, poor_accept_rate = mcmc_sample(log_prob_gaussian_mixture, n_samples,\n",
    "                                                   n_warmup=200, step_size=2.0)\n",
    "print(f\"Poor MCMC acceptance rate: {poor_accept_rate:.2%}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compare sampling methods using PQMass\n",
    "print(\"\\nEvaluating sampling methods with PQMass...\\n\")\n",
    "\n",
    "# Well-tuned MCMC vs true samples\n",
    "result_mcmc = pqmass_test_multiple_tessellations(\n",
    "    true_samples, mcmc_samples, n_R=100, n_tessellations=20\n",
    ")\n",
    "\n",
    "# Poorly-tuned MCMC vs true samples\n",
    "result_poor = pqmass_test_multiple_tessellations(\n",
    "    true_samples, poor_mcmc_samples, n_R=100, n_tessellations=20\n",
    ")\n",
    "\n",
    "print(\"Well-tuned MCMC:\")\n",
    "print(f\"  Chi-squared: {result_mcmc['chi2_mean']:.2f} ± {result_mcmc['chi2_std']:.2f}\")\n",
    "print(f\"  p-value: {result_mcmc['p_value_mean']:.4f}\")\n",
    "print()\n",
    "\n",
    "print(\"Poorly-tuned MCMC:\")\n",
    "print(f\"  Chi-squared: {result_poor['chi2_mean']:.2f} ± {result_poor['chi2_std']:.2f}\")\n",
    "print(f\"  p-value: {result_poor['p_value_mean']:.4f}\")\n",
    "print()\n",
    "\n",
    "# Visualize\n",
    "fig, axes = plt.subplots(1, 3, figsize=(18, 5))\n",
    "\n",
    "axes[0].scatter(true_samples[:, 0], true_samples[:, 1], alpha=0.5, s=10)\n",
    "axes[0].set_title('True Distribution Samples', fontsize=12, fontweight='bold')\n",
    "axes[0].set_xlabel('x1')\n",
    "axes[0].set_ylabel('x2')\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "axes[1].scatter(mcmc_samples[:, 0], mcmc_samples[:, 1], alpha=0.5, s=10, color='green')\n",
    "axes[1].set_title(f'Well-tuned MCMC\\nχ² = {result_mcmc[\"chi2_mean\"]:.1f}, p = {result_mcmc[\"p_value_mean\"]:.3f}', \n",
    "                  fontsize=12, fontweight='bold')\n",
    "axes[1].set_xlabel('x1')\n",
    "axes[1].set_ylabel('x2')\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "axes[2].scatter(poor_mcmc_samples[:, 0], poor_mcmc_samples[:, 1], alpha=0.5, s=10, color='red')\n",
    "axes[2].set_title(f'Poorly-tuned MCMC\\nχ² = {result_poor[\"chi2_mean\"]:.1f}, p = {result_poor[\"p_value_mean\"]:.3f}', \n",
    "                  fontsize=12, fontweight='bold')\n",
    "axes[2].set_xlabel('x1')\n",
    "axes[2].set_ylabel('x2')\n",
    "axes[2].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\n✓ PQMass successfully distinguishes between good and poor sampling algorithms!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7. Mode Coverage Detection (Workflow 8)\n",
    "\n",
    "One important application of PQMass is detecting when a generative model fails to capture all modes of the true distribution (mode collapse). We'll simulate this by creating samples that are missing one of the clusters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate samples with missing modes\n",
    "def generate_with_missing_modes(n_samples, n_modes_to_drop=0, random_state=None):\n",
    "    \"\"\"\n",
    "    Generate samples from Gaussian mixture with some modes dropped.\n",
    "    \"\"\"\n",
    "    if random_state is not None:\n",
    "        np.random.seed(random_state)\n",
    "    \n",
    "    all_centers = np.array([\n",
    "        [0, 0],\n",
    "        [3, 3],\n",
    "        [-2, 3]\n",
    "    ])\n",
    "    \n",
    "    # Drop specified number of modes\n",
    "    if n_modes_to_drop > 0:\n",
    "        centers = all_centers[:-n_modes_to_drop]\n",
    "    else:\n",
    "        centers = all_centers\n",
    "    \n",
    "    samples, _ = make_blobs(n_samples=n_samples, centers=centers, \n",
    "                            cluster_std=0.6, random_state=random_state)\n",
    "    return samples\n",
    "\n",
    "# Test with different numbers of dropped modes\n",
    "n_samples = 2000\n",
    "real_samples = generate_with_missing_modes(n_samples, n_modes_to_drop=0, random_state=42)\n",
    "\n",
    "modes_to_drop = [0, 1, 2]\n",
    "mode_results = []\n",
    "\n",
    "print(\"Testing mode coverage detection...\\n\")\n",
    "\n",
    "for n_drop in modes_to_drop:\n",
    "    gen_samples = generate_with_missing_modes(n_samples, n_modes_to_drop=n_drop, random_state=123)\n",
    "    \n",
    "    result = pqmass_test_multiple_tessellations(\n",
    "        real_samples, gen_samples, n_R=100, n_tessellations=20\n",
    "    )\n",
    "    \n",
    "    mode_results.append(result)\n",
    "    \n",
    "    print(f\"Dropped {n_drop} mode(s):\")\n",
    "    print(f\"  Chi-squared: {result['chi2_mean']:.2f} ± {result['chi2_std']:.2f}\")\n",
    "    print(f\"  p-value: {result['p_value_mean']:.4f}\")\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visualize mode coverage results\n",
    "fig, axes = plt.subplots(2, 2, figsize=(14, 12))\n",
    "\n",
    "# Plot samples with different numbers of dropped modes\n",
    "for idx, n_drop in enumerate(modes_to_drop):\n",
    "    if idx < 3:\n",
    "        row = idx // 2\n",
    "        col = idx % 2\n",
    "        gen_samples = generate_with_missing_modes(n_samples, n_modes_to_drop=n_drop, random_state=123)\n",
    "        \n",
    "        axes[row, col].scatter(real_samples[:, 0], real_samples[:, 1], \n",
    "                               alpha=0.3, s=10, c='blue', label='Real (3 modes)')\n",
    "        axes[row, col].scatter(gen_samples[:, 0], gen_samples[:, 1], \n",
    "                               alpha=0.3, s=10, c='orange', label=f'Generated ({3-n_drop} modes)')\n",
    "        axes[row, col].set_title(f'Dropped {n_drop} mode(s)\\nχ² = {mode_results[idx][\"chi2_mean\"]:.1f}', \n",
    "                                 fontsize=12, fontweight='bold')\n",
    "        axes[row, col].set_xlabel('x1')\n",
    "        axes[row, col].set_ylabel('x2')\n",
    "        axes[row, col].legend()\n",
    "        axes[row, col].grid(True, alpha=0.3)\n",
    "\n",
    "# Plot chi-squared vs number of dropped modes\n",
    "chi2_vals = [r['chi2_mean'] for r in mode_results]\n",
    "chi2_errs = [r['chi2_std'] for r in mode_results]\n",
    "\n",
    "axes[1, 1].errorbar(modes_to_drop, chi2_vals, yerr=chi2_errs, \n",
    "                    marker='o', capsize=5, linewidth=2, markersize=10)\n",
    "axes[1, 1].axhline(99, color='r', linestyle='--', linewidth=2, label='Expected under null')\n",
    "axes[1, 1].set_xlabel('Number of Dropped Modes', fontsize=12)\n",
    "axes[1, 1].set_ylabel('Chi-squared PQM Statistic', fontsize=12)\n",
    "axes[1, 1].set_title('Mode Coverage Detection', fontsize=12, fontweight='bold')\n",
    "axes[1, 1].set_xticks(modes_to_drop)\n",
    "axes[1, 1].legend()\n",
    "axes[1, 1].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\n✓ PQMass successfully detects missing modes!\")\n",
    "print(\"  The chi-squared statistic increases as more modes are dropped.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8. MNIST Generative Model Evaluation (Simplified Workflow 3)\n",
    "\n",
    "Now we demonstrate PQMass on a more realistic task: evaluating a simple generative model on MNIST digits. Due to computational constraints, we'll use a very small subset of MNIST and a simple autoencoder.\n",
    "\n",
    "**Note:** This is a simplified demonstration. In practice, you would:\n",
    "- Train for many more epochs\n",
    "- Use larger sample sizes\n",
    "- Use more sophisticated models (VAE, Diffusion models)\n",
    "- Track chi-squared values over training to monitor progress"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Simple VAE for MNIST\n",
    "class SimpleVAE(nn.Module):\n",
    "    def __init__(self, latent_dim=20):\n",
    "        super(SimpleVAE, self).__init__()\n",
    "        \n",
    "        # Encoder\n",
    "        self.encoder = nn.Sequential(\n",
    "            nn.Flatten(),\n",
    "            nn.Linear(28*28, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, 128),\n",
    "            nn.ReLU()\n",
    "        )\n",
    "        \n",
    "        self.fc_mu = nn.Linear(128, latent_dim)\n",
    "        self.fc_logvar = nn.Linear(128, latent_dim)\n",
    "        \n",
    "        # Decoder\n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.Linear(latent_dim, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, 28*28),\n",
    "            nn.Sigmoid()\n",
    "        )\n",
    "    \n",
    "    def encode(self, x):\n",
    "        h = self.encoder(x)\n",
    "        return self.fc_mu(h), self.fc_logvar(h)\n",
    "    \n",
    "    def reparameterize(self, mu, logvar):\n",
    "        std = torch.exp(0.5 * logvar)\n",
    "        eps = torch.randn_like(std)\n",
    "        return mu + eps * std\n",
    "    \n",
    "    def decode(self, z):\n",
    "        return self.decoder(z).view(-1, 1, 28, 28)\n",
    "    \n",
    "    def forward(self, x):\n",
    "        mu, logvar = self.encode(x)\n",
    "        z = self.reparameterize(mu, logvar)\n",
    "        return self.decode(z), mu, logvar\n",
    "    \n",
    "    def sample(self, num_samples, device='cpu'):\n",
    "        z = torch.randn(num_samples, self.fc_mu.out_features).to(device)\n",
    "        samples = self.decode(z)\n",
    "        return samples\n",
    "\n",
    "def vae_loss(recon_x, x, mu, logvar):\n",
    "    BCE = F.binary_cross_entropy(recon_x.view(-1, 28*28), x.view(-1, 28*28), reduction='sum')\n",
    "    KLD = -0.5 * torch.sum(1 + logvar - mu.pow(2) - logvar.exp())\n",
    "    return BCE + KLD\n",
    "\n",
    "print(\"VAE model defined successfully!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load a small subset of MNIST\n",
    "print(\"Loading MNIST dataset...\")\n",
    "\n",
    "transform = transforms.Compose([\n",
    "    transforms.ToTensor(),\n",
    "])\n",
    "\n",
    "# Load MNIST\n",
    "train_dataset = datasets.MNIST('./data', train=True, download=True, transform=transform)\n",
    "test_dataset = datasets.MNIST('./data', train=False, download=True, transform=transform)\n",
    "\n",
    "# Use small subset for fast training\n",
    "subset_size = 5000  # Use small subset for demonstration\n",
    "train_subset = torch.utils.data.Subset(train_dataset, range(subset_size))\n",
    "train_loader = torch.utils.data.DataLoader(train_subset, batch_size=128, shuffle=True)\n",
    "\n",
    "test_subset = torch.utils.data.Subset(test_dataset, range(1000))\n",
    "test_loader = torch.utils.data.DataLoader(test_subset, batch_size=1000, shuffle=False)\n",
    "\n",
    "print(f\"Training samples: {len(train_subset)}\")\n",
    "print(f\"Test samples: {len(test_subset)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train a simple VAE\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "print(f\"Using device: {device}\")\n",
    "\n",
    "model = SimpleVAE(latent_dim=20).to(device)\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3)\n",
    "\n",
    "# Train for a few epochs (keeping it fast for demonstration)\n",
    "n_epochs = 5\n",
    "print(f\"\\nTraining VAE for {n_epochs} epochs...\")\n",
    "\n",
    "model.train()\n",
    "for epoch in range(n_epochs):\n",
    "    total_loss = 0\n",
    "    for batch_idx, (data, _) in enumerate(train_loader):\n",
    "        data = data.to(device)\n",
    "        optimizer.zero_grad()\n",
    "        \n",
    "        recon_batch, mu, logvar = model(data)\n",
    "        loss = vae_loss(recon_batch, data, mu, logvar)\n",
    "        \n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        total_loss += loss.item()\n",
    "    \n",
    "    avg_loss = total_loss / len(train_loader.dataset)\n",
    "    print(f\"Epoch {epoch+1}/{n_epochs}, Loss: {avg_loss:.4f}\")\n",
    "\n",
    "print(\"\\nTraining complete!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate samples and evaluate with PQMass\n",
    "model.eval()\n",
    "\n",
    "# Generate samples from VAE\n",
    "n_gen_samples = 1000\n",
    "with torch.no_grad():\n",
    "    generated_samples = model.sample(n_gen_samples, device=device).cpu().numpy()\n",
    "\n",
    "# Get real test samples\n",
    "real_samples = next(iter(test_loader))[0].numpy()\n",
    "\n",
    "# Flatten images for PQMass\n",
    "generated_flat = generated_samples.reshape(n_gen_samples, -1)\n",
    "real_flat = real_samples.reshape(len(real_samples), -1)\n",
    "\n",
    "print(f\"Generated samples shape: {generated_flat.shape}\")\n",
    "print(f\"Real samples shape: {real_flat.shape}\")\n",
    "\n",
    "# Apply PQMass in pixel space\n",
    "print(\"\\nEvaluating with PQMass...\")\n",
    "result = pqmass_test_multiple_tessellations(\n",
    "    real_flat, generated_flat, \n",
    "    n_R=50,  # Use fewer reference points for high-dimensional data\n",
    "    n_tessellations=10\n",
    ")\n",
    "\n",
    "print(f\"\\nPQMass Results (784-dimensional pixel space):\")\n",
    "print(f\"Chi-squared: {result['chi2_mean']:.2f} ± {result['chi2_std']:.2f}\")\n",
    "print(f\"p-value: {result['p_value_mean']:.4f}\")\n",
    "print(f\"\\nInterpretation: The generated samples {'significantly differ from' if result['p_value_mean'] < 0.05 else 'are statistically similar to'} real MNIST digits.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visualize generated samples\n",
    "fig, axes = plt.subplots(2, 10, figsize=(15, 3))\n",
    "\n",
    "for i in range(10):\n",
    "    axes[0, i].imshow(real_samples[i, 0], cmap='gray')\n",
    "    axes[0, i].axis('off')\n",
    "    if i == 0:\n",
    "        axes[0, i].set_title('Real', fontsize=10)\n",
    "    \n",
    "    axes[1, i].imshow(generated_samples[i, 0], cmap='gray')\n",
    "    axes[1, i].axis('off')\n",
    "    if i == 0:\n",
    "        axes[1, i].set_title('Generated', fontsize=10)\n",
    "\n",
    "plt.suptitle(f'MNIST: Real vs Generated Samples\\nPQMass χ² = {result[\"chi2_mean\"]:.1f}, p = {result[\"p_value_mean\"]:.4f}', \n",
    "             fontsize=13, fontweight='bold')\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\n✓ Successfully applied PQMass to high-dimensional image data!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 9. Comparing with Baseline Metrics\n",
    "\n",
    "PQMass can be compared with other sample-based metrics. Here we implement simple versions of Maximum Mean Discrepancy (MMD) for comparison."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_mmd_rbf(X, Y, gamma=1.0):\n",
    "    \"\"\"\n",
    "    Compute Maximum Mean Discrepancy with RBF kernel.\n",
    "    \n",
    "    MMD²(X,Y) = E[k(x,x')] - 2E[k(x,y)] + E[k(y,y')]\n",
    "    where k is the RBF kernel.\n",
    "    \"\"\"\n",
    "    # RBF kernel\n",
    "    def rbf_kernel(X, Y, gamma):\n",
    "        # Compute pairwise squared distances\n",
    "        XX = np.sum(X**2, axis=1).reshape(-1, 1)\n",
    "        YY = np.sum(Y**2, axis=1).reshape(1, -1)\n",
    "        XY = X @ Y.T\n",
    "        sq_distances = XX - 2*XY + YY\n",
    "        return np.exp(-gamma * sq_distances)\n",
    "    \n",
    "    m = len(X)\n",
    "    n = len(Y)\n",
    "    \n",
    "    # Compute kernel matrices\n",
    "    K_XX = rbf_kernel(X, X, gamma)\n",
    "    K_YY = rbf_kernel(Y, Y, gamma)\n",
    "    K_XY = rbf_kernel(X, Y, gamma)\n",
    "    \n",
    "    # MMD² statistic\n",
    "    mmd_sq = (np.sum(K_XX) - np.trace(K_XX)) / (m * (m-1))\n",
    "    mmd_sq += (np.sum(K_YY) - np.trace(K_YY)) / (n * (n-1))\n",
    "    mmd_sq -= 2 * np.sum(K_XY) / (m * n)\n",
    "    \n",
    "    return np.sqrt(max(mmd_sq, 0))  # Take sqrt and ensure non-negative\n",
    "\n",
    "print(\"MMD baseline implemented!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compare PQMass and MMD on 2D Gaussian mixtures\n",
    "shifts = [0.0, 0.1, 0.2, 0.3, 0.5, 0.8, 1.0]\n",
    "n_samples = 1000\n",
    "\n",
    "pqmass_results = []\n",
    "mmd_results = []\n",
    "\n",
    "real_samples = generate_gaussian_mixture_shifted(n_samples, shift=0.0, random_state=42)\n",
    "\n",
    "print(\"Comparing PQMass and MMD...\\n\")\n",
    "\n",
    "for shift in tqdm(shifts, desc=\"Testing shifts\"):\n",
    "    gen_samples = generate_gaussian_mixture_shifted(n_samples, shift=shift, random_state=123)\n",
    "    \n",
    "    # PQMass\n",
    "    pqm_result = pqmass_test_multiple_tessellations(\n",
    "        real_samples, gen_samples, n_R=100, n_tessellations=10\n",
    "    )\n",
    "    pqmass_results.append(pqm_result['chi2_mean'])\n",
    "    \n",
    "    # MMD with RBF kernel\n",
    "    mmd = compute_mmd_rbf(real_samples, gen_samples, gamma=0.5)\n",
    "    mmd_results.append(mmd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visualize comparison\n",
    "fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "\n",
    "# PQMass\n",
    "axes[0].plot(shifts, pqmass_results, marker='o', linewidth=2, markersize=8, label='PQMass χ²')\n",
    "axes[0].axhline(99, color='r', linestyle='--', linewidth=2, label='Expected under null')\n",
    "axes[0].set_xlabel('Distribution Shift', fontsize=12)\n",
    "axes[0].set_ylabel('Chi-squared Statistic', fontsize=12)\n",
    "axes[0].set_title('PQMass vs Distribution Shift', fontsize=13, fontweight='bold')\n",
    "axes[0].legend()\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "# MMD\n",
    "axes[1].plot(shifts, mmd_results, marker='s', linewidth=2, markersize=8, color='green', label='MMD (RBF kernel)')\n",
    "axes[1].axhline(0, color='r', linestyle='--', linewidth=2, label='Expected under null')\n",
    "axes[1].set_xlabel('Distribution Shift', fontsize=12)\n",
    "axes[1].set_ylabel('MMD Value', fontsize=12)\n",
    "axes[1].set_title('MMD vs Distribution Shift', fontsize=13, fontweight='bold')\n",
    "axes[1].legend()\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\n✓ Both PQMass and MMD successfully detect distribution differences!\")\n",
    "print(\"  Both metrics increase monotonically with distribution shift.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10. Summary and Scaling Guidance\n",
    "\n",
    "### What We Demonstrated\n",
    "\n",
    "This notebook walked through the core PQMass methodology using small-scale, efficient examples:\n",
    "\n",
    "1. **Null test validation**: Verified that PQMass chi-squared statistics follow the expected distribution\n",
    "2. **Distribution comparison**: Showed PQMass can detect differences between distributions\n",
    "3. **Voronoi visualization**: Illustrated how PQMass partitions space\n",
    "4. **Sampling method comparison**: Evaluated MCMC samplers\n",
    "5. **Mode coverage detection**: Detected missing modes in distributions\n",
    "6. **MNIST evaluation**: Applied PQMass to high-dimensional image data\n",
    "7. **Baseline comparison**: Compared with MMD metric\n",
    "\n",
    "### Key Takeaways\n",
    "\n",
    "- **PQMass is statistically rigorous**: Provides p-values from chi-squared tests\n",
    "- **No auxiliary models needed**: Unlike FID/FLD, requires no pretrained networks\n",
    "- **Works in pixel space**: Can operate on high-dimensional data without feature extraction\n",
    "- **Flexible**: Works with any distance metric and data modality\n",
    "- **Efficient**: Computational cost scales well with dimensionality\n",
    "\n",
    "### Scaling to Production Use\n",
    "\n",
    "This notebook used small-scale examples to run efficiently. For production use:\n",
    "\n",
    "**Sample sizes:**\n",
    "- We used: 1,000-2,000 samples per distribution\n",
    "- Production: 10,000-50,000 samples for robust statistics\n",
    "\n",
    "**Reference points (n_R):**\n",
    "- We used: 50-100 reference points\n",
    "- Production: 100-500 reference points for more fine-grained tessellation\n",
    "- Higher n_R gives more statistical power but increases computation\n",
    "\n",
    "**Tessellation repetitions:**\n",
    "- We used: 10-20 repetitions\n",
    "- Production: 20-100 repetitions to reduce variance from random tessellations\n",
    "\n",
    "**Model training:**\n",
    "- We used: 5 epochs, 5,000 samples\n",
    "- Production: 100+ epochs, full datasets\n",
    "- Track PQMass during training to monitor progress\n",
    "\n",
    "**Computational resources:**\n",
    "- We used: CPU-only, 4GB RAM\n",
    "- Production: GPU for model training, more memory for larger datasets\n",
    "- Distance computations can be parallelized\n",
    "\n",
    "**Distance metrics:**\n",
    "- For images: Euclidean (L2) in pixel space or feature space\n",
    "- For sequences: Levenshtein distance, dynamic time warping\n",
    "- For tabular data: L2 after normalization\n",
    "\n",
    "### Additional Workflows Not Covered\n",
    "\n",
    "Due to resource constraints, we didn't implement all workflows from the paper. These would require:\n",
    "\n",
    "- **Astrophysics images**: Large pre-trained diffusion models, high-res images\n",
    "- **Protein sequences**: Specialized distance metrics, large sequence datasets\n",
    "- **Tabular data generation**: CTGAN training, Adult Census dataset\n",
    "- **Human judgment correlation**: Pre-trained generative models on CIFAR-10/FFHQ\n",
    "- **Novelty detection**: Training multiple models, computing train-test gaps\n",
    "\n",
    "### Further Reading\n",
    "\n",
    "See the paper for:\n",
    "- Theoretical guarantees and proofs\n",
    "- Detailed experimental results\n",
    "- Comparison with more baseline metrics\n",
    "- Applications to various data modalities\n",
    "- Permutation test extensions for small sample sizes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "This notebook provided a comprehensive, educational introduction to **PQMass** - a powerful, statistically rigorous method for evaluating generative models. The key innovation is using Voronoi tessellation to partition space and applying chi-squared tests to multinomial count distributions.\n",
    "\n",
    "**Next steps for researchers:**\n",
    "1. Apply PQMass to your own generative models\n",
    "2. Experiment with different distance metrics for your data modality\n",
    "3. Track PQMass during model training to monitor progress\n",
    "4. Compare PQMass with domain-specific metrics\n",
    "5. Scale up to full datasets on your own infrastructure\n",
    "\n",
    "**Citation:**\n",
    "```\n",
    "@inproceedings{lemos2025pqmass,\n",
    "  title={PQMass: Probabilistic Assessment of the Quality of Generative Models using Probability Mass Estimation},\n",
    "  author={Lemos, Pablo and Sharief, Sammy and Malkin, Nikolay and Salhi, Salma and Stone, Connor and Perreault-Levasseur, Laurence and Hezaveh, Yashar},\n",
    "  booktitle={International Conference on Learning Representations},\n",
    "  year={2025}\n",
    "}\n",
    "```\n",
    "\n",
    "**Code:** https://github.com/Ciela-Institute/PQM"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
 }
No results found