notebooks/Optimizations/ROCK_Hyperparameter_Search.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Optimizing Rock's hyperparameter $t_{max}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "UbvqVKulcAwD"
   },
   "outputs": [],
   "source": [
    "import sys  \n",
    "sys.path.insert(0, '../..')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 485,
     "status": "ok",
     "timestamp": 1621160417511,
     "user": {
      "displayName": "Isdm Hueni",
      "photoUrl": "",
      "userId": "05072038424951783895"
     },
     "user_tz": -120
    },
    "id": "lWMfMVRSeO88",
    "outputId": "964a462f-e01f-4e82-c8ef-bc9ba8a26852"
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 389
    },
    "executionInfo": {
     "elapsed": 948,
     "status": "error",
     "timestamp": 1621259135300,
     "user": {
      "displayName": "Isdm Hueni",
      "photoUrl": "",
      "userId": "05072038424951783895"
     },
     "user_tz": -120
    },
    "id": "EXvc2zQ-csmQ",
    "outputId": "82654e58-c034-4012-92df-17764dfcaad0"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.datasets import make_blobs, make_moons, make_circles, load_iris, load_wine, load_breast_cancer, load_digits, fetch_covtype    \n",
    "from sklearn.cluster import KMeans, DBSCAN, SpectralClustering, MeanShift, estimate_bandwidth\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import adjusted_mutual_info_score as AMI\n",
    "from rock import ROCK\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import logging\n",
    "import sys\n",
    "from tqdm.notebook import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "sfM20A23iHzF"
   },
   "outputs": [],
   "source": [
    "experiment = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "dZTxToxqaBUt"
   },
   "outputs": [],
   "source": [
    "def getBlobDensities(n_centers):\n",
    "  return np.round([3/n_centers * x for x in range(1, n_centers + 1)], decimals=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "FLtgfMjY5C_j"
   },
   "outputs": [],
   "source": [
    "blobs = make_blobs(n_samples=300, n_features=3, cluster_std=getBlobDensities(2), centers=2, random_state=42)\n",
    "moons = make_moons(n_samples=1000, noise=0.15, random_state=42)\n",
    "circles = make_circles(n_samples=1600, noise=0.02, random_state=42)\n",
    "iris = load_iris(return_X_y=True)\n",
    "wine = load_wine(return_X_y=True)\n",
    "cancer = load_breast_cancer(return_X_y=True)\n",
    "digits = load_digits(return_X_y=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "executionInfo": {
     "elapsed": 757,
     "status": "ok",
     "timestamp": 1621259125177,
     "user": {
      "displayName": "Isdm Hueni",
      "photoUrl": "",
      "userId": "05072038424951783895"
     },
     "user_tz": -120
    },
    "id": "SpOu6jOr43i3"
   },
   "outputs": [],
   "source": [
    "datasets = [blobs, moons, circles, iris, wine, cancer, digits]\n",
    "data_names = ['Blobs', 'Moons', 'Circles', 'Iris', 'Wine', 'Breast Cancer', 'Digits']\n",
    "tmaxs = range(1, 31)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 244
    },
    "executionInfo": {
     "elapsed": 676,
     "status": "error",
     "timestamp": 1621259127305,
     "user": {
      "displayName": "Isdm Hueni",
      "photoUrl": "",
      "userId": "05072038424951783895"
     },
     "user_tz": -120
    },
    "id": "q752xexB41rz",
    "outputId": "3959aa6e-3b96-4bb4-8895-2e79f08997c0"
   },
   "outputs": [],
   "source": [
    "for d in tqdm(range(0, len(datasets))):\n",
    "    for t in tqdm(range(0, len(tmaxs))):\n",
    "      dataset = datasets[d]\n",
    "      run = {}\n",
    "      scaler = StandardScaler()\n",
    "      data = scaler.fit_transform(dataset[0])\n",
    "      data = dataset[0]\n",
    "      rock = ROCK(tmax=tmaxs[t]).fit(data).labels_\n",
    "      run['Dataset'] = data_names[d]\n",
    "      gt = dataset[1]\n",
    "      run['AMI'] = AMI(rock, gt)\n",
    "      run['tmax'] = tmaxs[t]\n",
    "      \n",
    "      experiment.append(run)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "U3FnCVEum_Vi"
   },
   "outputs": [],
   "source": [
    "results = pd.DataFrame(experiment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 297
    },
    "executionInfo": {
     "elapsed": 1160,
     "status": "ok",
     "timestamp": 1621256483676,
     "user": {
      "displayName": "Isdm Hueni",
      "photoUrl": "",
      "userId": "05072038424951783895"
     },
     "user_tz": -120
    },
    "id": "HoPbzzkaJeTI",
    "outputId": "e6512de8-068d-4a93-de19-1e4127f72d7c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 396.851x255.118 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.rcParams['axes.grid'] = True\n",
    "plt.rcParams['axes.spines.top'] = False\n",
    "plt.rcParams['axes.spines.right'] = False\n",
    "plt.rcParams['axes.spines.bottom'] = False\n",
    "plt.rcParams['axes.spines.left'] = False\n",
    "plt.rcParams['figure.figsize'] = (14*0.393701, 9*0.393701)\n",
    "\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Moons']['AMI'], linestyle='--', marker='P', label='Moons')\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Blobs']['AMI'], linestyle=\":\", marker='^', label='Blobs')\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Circles']['AMI'], linestyle='-.', marker='p', label='Rings')\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Digits']['AMI'], linestyle='--', marker='*',  label='Digits')\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Wine']['AMI'], linestyle='-', marker='o', label='Wine')\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Iris']['AMI'], linestyle='--', marker='s', label='Iris')\n",
    "plt.plot(tmaxs, results[results['Dataset'] == 'Breast Cancer']['AMI'], linestyle=':', marker='X', label='Breast Cancer')\n",
    "plt.xlabel('tmax')\n",
    "plt.ylabel('AMI')\n",
    "plt.tight_layout()\n",
    "plt.ylim(-0.01, 1.1)\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "plt.savefig('../../figures/ROCK_hyper_search.png', dpi = 600, bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "tVPJ0KOPnALz"
   },
   "outputs": [],
   "source": [
    "results.to_csv('../../results/optimization/ROCK_hyper.csv')"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyO/7JeYJfy84vAcNbpgiSOi",
   "collapsed_sections": [],
   "mount_file_id": "1vi3vjOwPDsYrLGNN2_krA5SkV1Wty3bn",
   "name": "ROCK_Hyperparameter_Search.ipynb",
   "provenance": [
    {
     "file_id": "1yZZ4GfxaXLw7kQQ-ufKegMhT-sh93n-d",
     "timestamp": 1620305717278
    },
    {
     "file_id": "1ovm-C9cOxmlH44OzWKrLSfVgtP3HqT18",
     "timestamp": 1620113188233
    }
   ]
  },
  "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": "3.9.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}