graphics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x81 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x81 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x81 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x81 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.patches import Rectangle\n",
    "from matplotlib.text import Text\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "%matplotlib inline\n",
    "\n",
    "scale = .75\n",
    "\n",
    "for run in (0, 1, 2, 3):\n",
    "    fig = plt.figure(figsize=(scale * 8, scale * 1.5))\n",
    "    ax = plt.subplot(1, 1, 1)\n",
    "    ax.set_axis_off()\n",
    "\n",
    "    for p in (0, 1, 2, 3):\n",
    "        x0 = .26 * p * scale\n",
    "        color = \"chocolate\" if p == run else \"steelblue\"\n",
    "        r = Rectangle((x0+ .01, 0.01), .23 * scale, .98, figure=fig, facecolor=\"w\", edgecolor=color, linewidth=1)\n",
    "        ax.add_patch(r)\n",
    "        plt.text(.09 * scale + x0, .55 * scale, \"P\" + str(p + 1), fontsize=20 * scale)\n",
    "        \n",
    "    plt.savefig(\"cross_val_%d.svg\" % run)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x129.6 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(scale * 8, scale * 2.4))\n",
    "ax = plt.subplot(1, 1, 1)\n",
    "ax.set_axis_off()\n",
    "\n",
    "ax.set_ylim([0, 1.5])\n",
    "\n",
    "for p in (0, 1, 2, 3):\n",
    "    x0 = .24 * p * scale\n",
    "    color = \"steelblue\"\n",
    "    r = Rectangle((x0 + 0.02, 0.1), .22 * scale, .8 , figure=fig, facecolor=\"w\", edgecolor=color, linewidth=1)\n",
    "    ax.add_patch(r)\n",
    "    plt.text(x0 + scale * 0.1, .54 * scale, \"P\" + str(p + 1), fontsize=20 * scale)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "r = Rectangle((0.005, 0.02), 0.738, .98, figure=fig, alpha=0.5, facecolor=\"w\", edgecolor=\"k\", linewidth=1)\n",
    "ax.add_patch(r);\n",
    "\n",
    "\n",
    "r = Rectangle((.8, 0.1), .22 * scale, .8, figure=fig, facecolor=\"w\", edgecolor=\"g\", linewidth=1)\n",
    "ax.add_patch(r);\n",
    "\n",
    "plt.text(.87, .54 * scale, \"V\" , fontsize=20 * scale)\n",
    "plt.text(0.25, 1.2, \"crosseval\", fontsize=20 * scale)\n",
    "plt.text(.78, 1.2, \"validation\", fontsize=20 * scale)\n",
    "\n",
    "plt.savefig(\"cross_eval_and_test.svg\", bbox=\"tight\")\n",
    "\n",
    "plt.savefig(\"cross_eval_and_test.png\", bbox=\"tight\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rectangle(xy=(0.8, 0.1), width=0.165, height=0.8, angle=0)\n"
     ]
    }
   ],
   "source": [
    "print(r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x129.6 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x129.6 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def create_plot(x0, name):\n",
    "    fig = plt.figure(figsize=(scale * 8, scale * 2.4))\n",
    "\n",
    "    ax = plt.subplot(1, 1, 1)\n",
    "    ax.set_title(\"higher recall, lower precision\")\n",
    "    ax.set_axis_off()\n",
    "    ax.set_ylim([0, 0.85])\n",
    "    ax.set_xlim([0, 2])\n",
    "\n",
    "    ax.vlines(x0, 0.2, 0.8)\n",
    "    ax.arrow(1, 0.5, -0.8, 0, head_width=0.08, head_length=0.05, fc='k', ec='k')\n",
    "    ax.arrow(1, 0.5, 0.8, 0, head_width=0.08, head_length=0.05, fc='k', ec='k')\n",
    "    ax.text(0.25, 0.2, \"irrelevant results\")\n",
    "    ax.text(1.3, 0.2, \"relevant results\");\n",
    "\n",
    "    plt.savefig(name)\n",
    "    \n",
    "create_plot(0.85, \"precision_low_recall_high.svg\")\n",
    "create_plot(1.15, \"precision_high_recall_low.svg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "x = np.arange(1, 13, .2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x324 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train =  (1  - np.exp(-x / 2)) \n",
    "y_eval =  - np.exp(-(x - 7) ** 2)\n",
    "w = 1.2 * ( np.exp(-(x / 18) ** 2 / 3)) - .25\n",
    "\n",
    "scale = .75\n",
    "plt.figure(figsize=(scale * 8, scale * 6))\n",
    "plt.plot(x, y_train, label=\"training\", color=\"steelblue\", linewidth=2)\n",
    "# plt.plot(x, w) # plt.plot(x, y_eval)\n",
    "plt.plot(x, y_train * w, label=\"eval\", color=\"chocolate\", linewidth=2)\n",
    "plt.legend()\n",
    "\n",
    "plt.title(\"accuracy training vs eval\")\n",
    "plt.xlabel(\"model complexity\")\n",
    "plt.ylabel(\"accuracy\")\n",
    "plt.savefig(\"accuracy_training_vs_eval.svg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1440x504 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "pd.set_option('precision', 3)\n",
    "\n",
    "import seaborn as sns\n",
    "sns.set(style=\"ticks\")\n",
    "\n",
    "beer_data = pd.read_csv(\"beers.csv\")\n",
    "\n",
    "\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "\n",
    "fig = plt.figure(figsize=(20, 7))\n",
    "\n",
    "xv = beer_data[\"alcohol_content\"]\n",
    "yv = beer_data[\"darkness\"]\n",
    "zv = beer_data[\"bitterness\"]\n",
    "\n",
    "colors = [[\"steelblue\", \"chocolate\"][i] for i in beer_data[\"is_yummy\"]]\n",
    "\n",
    "def plot3d(ax):\n",
    "    ax.scatter(xv, yv, zv, c=colors, marker='o') \n",
    "    \n",
    "    ax.set_xlabel('alcohol_content')\n",
    "    ax.set_ylabel('darkness')\n",
    "    ax.set_zlabel('bitterness');\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 175 180 185 190 195 200 205 210 215 220 225 230 235 240 245 250 255 260 265 270 275 280 285 290 295 300 305 310 315 320 325 330 335 340 345 350 355 "
     ]
    }
   ],
   "source": [
    "for a in range(0, 360, 5):\n",
    "    fig = plt.figure(figsize=(9, 7))\n",
    "    print(a, end=\" \")   \n",
    "    ax = fig.add_subplot(111, projection='3d')\n",
    "    ax.set_axis_off()\n",
    "    plot3d(ax)\n",
    "    # ax.set_title(str(a))\n",
    "    ax.view_init(20, a)\n",
    "    plt.savefig(\"images/{:03d}.png\".format(a))\n",
    "    plt.close()\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/rotated.gif?xx\" />"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.7.2"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autoclose": false,
   "autocomplete": true,
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}