ksample/3samp_vs_epsilon.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "particular-joyce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#!/usr/bin/env python\n",
    "# coding: utf-8\n",
    "\n",
    "# # 3 Sample Tests Power over Increasing Epsilon\n",
    "\n",
    "# These are same useful functions to import. Since we are calculating the statistical power over all the tests for all the simulations, we can just use a wild card import from the respective modules\n",
    "\n",
    "# In[1]:\n",
    "\n",
    "\n",
    "import os\n",
    "import sys\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from hyppo.independence import MGC, Dcorr, Hsic\n",
    "from hyppo.sims import gaussian_3samp\n",
    "from joblib import Parallel, delayed\n",
    "from matplotlib.legend import Legend\n",
    "\n",
    "from power_3samp import power_3samp_epsweight\n",
    "\n",
    "sys.path.append(os.path.realpath(\"..\"))\n",
    "\n",
    "\n",
    "# In[4]:\n",
    "\n",
    "\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set(color_codes=True, style=\"white\", context=\"talk\", font_scale=2)\n",
    "PALETTE = sns.color_palette(\"Set1\")\n",
    "sns.set_palette(PALETTE[3:])\n",
    "\n",
    "\n",
    "# In[5]:\n",
    "\n",
    "\n",
    "# from statsmodels.multivariate.manova import MANOVA\n",
    "\n",
    "\n",
    "# class PyManova:\n",
    "#     r\"\"\"\n",
    "#     Wrapper of statsmodels MANOVA\n",
    "#     \"\"\"\n",
    "\n",
    "#     def _statistic(self, x, y):\n",
    "#         r\"\"\"\n",
    "#         Helper function to calculate the test statistic\n",
    "#         \"\"\"\n",
    "#         stat = MANOVA(x, y).mv_test().results['x0']['stat'].values[1, 0]\n",
    "#         self.stat = stat\n",
    "\n",
    "#         return stat\n",
    "\n",
    "\n",
    "# In[6]:\n",
    "\n",
    "\n",
    "from rpy2.robjects import Formula, numpy2ri\n",
    "from rpy2.robjects.packages import importr\n",
    "\n",
    "\n",
    "class Manova:\n",
    "    r\"\"\"\n",
    "    Wrapper of R MANOVA\n",
    "    \"\"\"\n",
    "\n",
    "    def __init__(self):\n",
    "        self.stats = importr(\"stats\")\n",
    "        self.r_base = importr(\"base\")\n",
    "\n",
    "        numpy2ri.activate()\n",
    "\n",
    "        self.formula = Formula(\"X ~ Y\")\n",
    "        self.env = self.formula.environment\n",
    "\n",
    "    def statistic(self, x, y):\n",
    "        r\"\"\"\n",
    "        Helper function to calculate the test statistic\n",
    "        \"\"\"\n",
    "        self.env[\"Y\"] = y\n",
    "        self.env[\"X\"] = x\n",
    "\n",
    "        stat = self.r_base.summary(self.stats.manova(self.formula), test=\"Pillai\")[3][4]\n",
    "\n",
    "        return stat\n",
    "\n",
    "\n",
    "# These are some constants that are used in this notebook. If running these notebook, please only manipulate these constants if you are not running more tests. They define the epsilons (distance from 0 for the center of each gaussian cluster) tested upon and the number of replications. The simulations tested over and the independence tests tested over are defined also.\n",
    "\n",
    "# In[7]:\n",
    "\n",
    "\n",
    "MAX_EPSILON = 1.0  # 0.6\n",
    "STEP_SIZE = 0.05\n",
    "EPSILONS = np.arange(0, MAX_EPSILON + STEP_SIZE, STEP_SIZE)\n",
    "WEIGHTS = EPSILONS\n",
    "POWER_REPS = 5\n",
    "REPS = 1000\n",
    "n_jobs = 5\n",
    "workers = 5\n",
    "ONEWAY_EPSILON = 0.3\n",
    "\n",
    "run = False  # True#\n",
    "plot = True  # False#\n",
    "\n",
    "# In[8]:\n",
    "\n",
    "tests = [\n",
    "    Dcorr,\n",
    "    MGC,\n",
    "    Hsic,\n",
    "    # PyManova,\n",
    "    Manova,\n",
    "]\n",
    "\n",
    "multiway_tests = [\n",
    "    # Dcorr,\n",
    "    # MGC,\n",
    "]\n",
    "\n",
    "cases = [\n",
    "    1,\n",
    "    2,\n",
    "    3,\n",
    "    # 4,\n",
    "    # 5,\n",
    "    # 6,\n",
    "]\n",
    "\n",
    "multiway_cases = [\n",
    "    # 1,\n",
    "    # 6,\n",
    "]\n",
    "\n",
    "# The following function calculates the estimated power ``POWER_REPS`` number off times and averages them. It does this iterating over the number of sample sizes.\n",
    "#\n",
    "# **Note: We only recommend running this code if running the next 2 cells ONCE to generate the csv files used to visualize the plots. This code takes a very long time to run and if running, we recommend using a machine with many cores.**\n",
    "\n",
    "# In[ ]:\n",
    "\n",
    "\n",
    "def estimate_power(case, test, multiway=False):\n",
    "    if test == Manova:\n",
    "        ws = 1\n",
    "    else:\n",
    "        ws = workers\n",
    "\n",
    "    if case in [4, 5]:\n",
    "        est_power = np.array(\n",
    "            [\n",
    "                np.mean(\n",
    "                    [\n",
    "                        power_3samp_epsweight(\n",
    "                            test,\n",
    "                            case=case,\n",
    "                            weight=i,\n",
    "                            workers=ws,\n",
    "                            c=ONEWAY_EPSILON,\n",
    "                            multiway=multiway,\n",
    "                            reps=REPS,\n",
    "                        )\n",
    "                        for _ in range(POWER_REPS)\n",
    "                    ]\n",
    "                )\n",
    "                for i in WEIGHTS\n",
    "            ]\n",
    "        )\n",
    "    else:\n",
    "        est_power = np.array(\n",
    "            [\n",
    "                np.mean(\n",
    "                    [\n",
    "                        power_3samp_epsweight(\n",
    "                            test,\n",
    "                            case=case,\n",
    "                            epsilon=i,\n",
    "                            workers=ws,\n",
    "                            c=ONEWAY_EPSILON,\n",
    "                            multiway=multiway,\n",
    "                            reps=REPS,\n",
    "                        )\n",
    "                        for _ in range(POWER_REPS)\n",
    "                    ]\n",
    "                )\n",
    "                for i in EPSILONS\n",
    "            ]\n",
    "        )\n",
    "    if multiway:\n",
    "        np.savetxt(\n",
    "            \"../ksample/3samp_vs_epsilon/{}_{}_multiway.csv\".format(\n",
    "                case, test.__name__\n",
    "            ),\n",
    "            est_power,\n",
    "            delimiter=\",\",\n",
    "        )\n",
    "    else:\n",
    "        np.savetxt(\n",
    "            \"../ksample/3samp_vs_epsilon/{}_{}.csv\".format(case, test.__name__),\n",
    "            est_power,\n",
    "            delimiter=\",\",\n",
    "        )\n",
    "\n",
    "    return est_power\n",
    "\n",
    "\n",
    "# In[ ]:\n",
    "\n",
    "\n",
    "if run:\n",
    "    outputs = Parallel(n_jobs=n_jobs, verbose=100)(\n",
    "        [delayed(estimate_power)(case, test) for case in cases for test in tests]\n",
    "    )\n",
    "\n",
    "    outputs = Parallel(n_jobs=n_jobs, verbose=100)(\n",
    "        [\n",
    "            delayed(estimate_power)(case, test, multiway=True)\n",
    "            for case in multiway_cases\n",
    "            for test in multiway_tests\n",
    "        ]\n",
    "    )\n",
    "\n",
    "\n",
    "# The following code loops over each saved independence test file and generates absolute power curves for each test and for each simulation modality.\n",
    "\n",
    "# In[9]:\n",
    "\n",
    "\n",
    "FONTSIZE = 20\n",
    "\n",
    "\n",
    "def plot_power():\n",
    "    fig, ax = plt.subplots(nrows=2, ncols=3, figsize=(12, 6))\n",
    "\n",
    "    sim_title = [\n",
    "        \"None Different\",\n",
    "        \"One Different\",\n",
    "        \"All Different\",\n",
    "        # \"One Not Gaussian\",\n",
    "        # \"None Gaussian\",\n",
    "        # \"More Different\",\n",
    "    ]\n",
    "\n",
    "    for i, row in enumerate(ax):\n",
    "        for j, col in enumerate(row):\n",
    "            case = cases[j]\n",
    "            if i == 0:\n",
    "                sims = gaussian_3samp(100, epsilon=4, weight=0.9, case=case)\n",
    "\n",
    "                sim_markers = [\"1\", \"+\", \"x\"]\n",
    "                custom_color = [\"#d9d9d9\", \"#969696\", \"#525252\"]\n",
    "\n",
    "                count = 0\n",
    "                scatters = []\n",
    "                for sim in sims:\n",
    "                    x, y = np.hsplit(sim, 2)\n",
    "                    scatters.append(\n",
    "                        col.scatter(\n",
    "                            x, y, marker=sim_markers[count], color=custom_color[count]\n",
    "                        )\n",
    "                    )\n",
    "\n",
    "                    if case not in [6]:\n",
    "                        col.set_xlim(-5, 5)\n",
    "                    if case not in [2, 4, 6]:\n",
    "                        col.set_ylim(-5, 5)\n",
    "                    else:\n",
    "                        col.set_ylim(-3, 7)\n",
    "                    col.set_xticks([])\n",
    "                    col.set_yticks([])\n",
    "                    col.set_title(sim_title[j], fontsize=FONTSIZE)\n",
    "                    count += 1\n",
    "            else:\n",
    "                for test in tests:\n",
    "                    custom_color = {\n",
    "                        \"Dcorr\": \"#377eb8\",\n",
    "                        \"Hsic\": \"#4daf4a\",\n",
    "                        \"MGC\": \"#e41a1c\",\n",
    "                    }\n",
    "\n",
    "                    for multiway in [True, False]:\n",
    "                        if multiway and (test not in multiway_tests or case not in [6]):\n",
    "                            continue\n",
    "                        elif multiway:\n",
    "                            power = np.genfromtxt(\n",
    "                                \"../ksample/3samp_vs_epsilon/{}_{}_multiway.csv\".format(\n",
    "                                    case, test.__name__\n",
    "                                ),\n",
    "                                delimiter=\",\",\n",
    "                            )\n",
    "\n",
    "                            ls = \"--\"\n",
    "                            label = f\"Multiway {test.__name__}\"\n",
    "                        else:\n",
    "                            power = np.genfromtxt(\n",
    "                                \"../ksample/3samp_vs_epsilon/{}_{}.csv\".format(\n",
    "                                    case, test.__name__\n",
    "                                ),\n",
    "                                delimiter=\",\",\n",
    "                            )\n",
    "\n",
    "                            ls = \"-\"\n",
    "                            label = test.__name__\n",
    "                        if test.__name__ in custom_color.keys():\n",
    "                            if test.__name__ == \"MGC\":\n",
    "                                col.plot(\n",
    "                                    EPSILONS,\n",
    "                                    power,\n",
    "                                    custom_color[test.__name__],\n",
    "                                    label=label,\n",
    "                                    lw=3,\n",
    "                                    ls=ls,\n",
    "                                )\n",
    "                            else:\n",
    "                                col.plot(\n",
    "                                    EPSILONS,\n",
    "                                    power,\n",
    "                                    custom_color[test.__name__],\n",
    "                                    label=label,\n",
    "                                    lw=2,\n",
    "                                    ls=ls,\n",
    "                                )\n",
    "                        else:\n",
    "                            col.plot(EPSILONS, power, label=label, lw=2, ls=ls)\n",
    "                    col.tick_params(labelsize=FONTSIZE)\n",
    "                    col.set_xticks([EPSILONS[0], EPSILONS[-1]])\n",
    "                    col.set_ylim(0, 1.05)\n",
    "                    col.set_yticks([])\n",
    "                    if j == 0:\n",
    "                        col.set_yticks([0, 1])\n",
    "            col.set_aspect(np.diff(col.get_xlim()) / np.diff(col.get_ylim()))\n",
    "\n",
    "    fig.text(0.5, 0, \"Cluster Separation\", ha=\"center\", fontsize=FONTSIZE)\n",
    "    #     fig.text(0.75, 0, 'Increasing Weight', ha='center')\n",
    "    fig.text(0.11, 0.3, \"Power\", va=\"center\", rotation=\"vertical\", fontsize=FONTSIZE)\n",
    "    fig.text(\n",
    "        0.11, 0.7, \"Scatter Plots\", va=\"center\", rotation=\"vertical\", fontsize=FONTSIZE\n",
    "    )\n",
    "\n",
    "    leg = plt.legend(\n",
    "        bbox_to_anchor=(1.04, 0.5),\n",
    "        bbox_transform=plt.gcf().transFigure,\n",
    "        ncol=1,\n",
    "        loc=\"upper center\",\n",
    "        fontsize=FONTSIZE,\n",
    "    )\n",
    "    leg.get_frame().set_linewidth(0.0)\n",
    "    for legobj in leg.legendHandles:\n",
    "        legobj.set_linewidth(5.0)\n",
    "    plt.subplots_adjust(hspace=0.20)\n",
    "    leg = Legend(\n",
    "        fig,\n",
    "        scatters,\n",
    "        [\"Cluster 1\", \"Cluster 2\", \"Cluster 3\"],\n",
    "        loc=\"upper center\",\n",
    "        frameon=False,\n",
    "        ncol=1,\n",
    "        bbox_transform=plt.gcf().transFigure,\n",
    "        bbox_to_anchor=(1.03, 0.9),\n",
    "        fontsize=FONTSIZE,\n",
    "    )\n",
    "    fig.add_artist(leg)\n",
    "    for legobj in leg.legendHandles:\n",
    "        legobj.set_linewidth(3.0)\n",
    "    plt.suptitle(\"Power vs. increasing cluster separation\", fontsize=FONTSIZE, y=1.03)\n",
    "    plt.savefig(\n",
    "        \"../ksample/figs/3samp_power_epsilon.pdf\",\n",
    "        transparent=True,\n",
    "        bbox_inches=\"tight\",\n",
    "    )\n",
    "\n",
    "\n",
    "if plot:\n",
    "    plot_power()\n",
    "\n",
    "# In[10]:\n",
    "\n",
    "\n",
    "# In[ ]:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "soviet-customs",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}