Data Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAQ6klEQVR4nO3dfYxcV3nH8e9TB1QKlABZCPiFjZFLHDWQ0m1CRVE3oVAnBNxKIBK3kKAiy2pcgUTVOJUKbhGSK0pKUQ2WlboxguJSEcAiLilKcElF09pp0ziOE7pdTLyYYCchaSESkeHpHzO7GU9md+6u7+zsnPl+pNXel+M7z1l7f3N85s6ZyEwkSYPvZ/pdgCSpHga6JBXCQJekQhjoklQIA12SCnFWvx74nHPOydHR0X49vCQNpLvvvvuRzBzpdK5vgT46OsrBgwf79fCSNJAi4juznXPKRZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQnQN9IjYFREnIuK+Wc5HRHwiIiYi4t6IeG39ZUqSuqkyQr8ZWDfH+cuBNc2vjcCnzrwsSdJ8dQ30zPwG8NgcTdYDn86Gu4CzI+JldRWowTU+Ps74+PgZH6/rcUuwkL716+fa6/Z6pjrm0JcDx1r2p5rHpJ6qMzD69SRT1+PWGYa9DtbF6FtdP+9eH69bHWu5RIdjHT/XLiI20piWYdWqVTU8tBbT9D/I/fv3Vzq+lBw5f+3MXOCR89cCsPaBI7O2/9g7rwTg2P2HZvY/8PdfmTm/lPp84e4LmXx4cmYb4NA1h+Z9nW59rsO5X78HPvTxp7eBhy+9aPY/sPUFcPRHT28DbH1i3o+7fdMdALzj/A/O7F+347KZPr/1pc8DqvV569atHD16dGZ7+vvtd7ySP2lcntvveCUAb7zsf+Zd65moI9CngJUt+yuA450aZuZOYCfA2NjYwH6Y6VL6Za7bwPSt5l/0737r8Zn963ZcNnP+9y/91Xldr9PP7/Y7XskPHj8+sz39S37u1+/hscd/OLMNjXAb3XIrD08+CsDollsBOLrtLXM+7uobVs+rzlJNP1k/+dB3ZvbneuKey9SWO/nx5BMz2yu2vWHm3LXXXntmhfZIHYG+F9gcEXuAS4AnMvN7NVx3qM03WAcmiGfRHm5zjtqG1TyfxBYabu1PYlNb7gSYNdzqMNeT2P5rnzvv6+1e9Ypa6vqHDZ+o5TqLpWugR8TngHHgnIiYAj4EPAsgM3cA+4ArgAngSeA9vSpWg236v6GdRqt1Wej0w/vfdmOtdbS78caXdzz+or+8qePxczds62U5tRr0wURJugZ6Zl7d5XwC19VWkRZFCb+ECxm5SYtttqm1XujbB1yooYRg1eLp55PYoE0/DCMDXQtW99zqbNMPPumdmbrmk+ej/YVgWPw7PoaRgT5PH3vnlafd1gXUfmvXoJu+lav11q7pY1o6ltoT5SC9brBUGehadLO9QFgXb+HTsDLQF8lSGw11sn3THafdjw2cdk+2FucWPpVntunEuhnoNetLcM9xf/IgPJGUwGkmLQWuhy5JhXCEPofZRrfzfTv4oFuMNT4knTkDXWfM+5OHV/tCVa3TTL1+8VvPZKAPoSPnrz1tfQ+Ye+VBSYPBOXRJKoQj9CVmeq2HElYeXKpLjEqlMtD7ZDFWHlyIXq86WApfN9BSZKAXwpUHlwb/V6J+cg5dkgrhCF0aMt0+dk+Dy0DXkraQz9eUhpWB3mOu8SFpsRQZ6C5IJS0eXwheOooMdHW3kE+xGbY1bKRBY6DTeUQ/37XBXSdbUr9526IkFcIResEu3H0hkw9PzmwDHLrmUD9LktRDjtAlqRCO0PvMNaMl1cVAX6IW60NlJZXDKRdJKoSBLkmFGKopF99BKj3Nte/L4wh9wIyPj888MUlSKwNdkgoxVFMu/eQCRpJ6rVKgR8Q64K+AZcBNmbmt7fwLgM8Aq5rX/IvM/Nuaa9WQOnfDtu6NJHUP9IhYBmwH3gRMAQciYm9m3t/S7Drg/sx8a0SMAA9GxGcz86meVL1IfNFI0iCpMkK/GJjIzEmAiNgDrAdaAz2B50dEAM8DHgNO1VzrQPDT4CX1S5VAXw4ca9mfAi5pa/PXwF7gOPB84J2Z+dNaKuyxI+ev5cmHvjOzDbD2gSP9LEmSFqTKXS7R4Vi27f8mcA/wcuAi4K8j4uefcaGIjRFxMCIOnjx5ct7FSpJmVyXQp4CVLfsraIzEW70HuCUbJoBvA+e3Xygzd2bmWGaOjYyMLLRmSVIHVaZcDgBrIuI84LvAVcCGtjYPAW8E7oyIlwKvAibrLLSyrS+Aoz96envrE30pY6lYfcPqfpcgaZF0DfTMPBURm4HbaNy2uCszD0fEpub5HcCHgZsj4hCNKZrrM/ORHtYtSWpT6T70zNwH7Gs7tqNl+zjw5npLkyTNh2/9l6RCGOgDZHTLrdw1+Sh3TT7K6JZb+12OpCVmaNZy8QOTJZXOEbokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqxNDctjiX3ate0e8SJOmMOUKXpEIY6JJUCANdkgphoEtSIXxRdMCcu2Fbv0uQtEQ5QpekQhjoklSIIqdc9l/73H6XIEmLrshAn40fmCypZE65SFIhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQY60MfHxxkfH+93GZK0JAx0oEuSnmagS1IhDHRJKkSlQI+IdRHxYERMRMSWWdqMR8Q9EXE4Iv653jIl6Zl8He10XZfPjYhlwHbgTcAUcCAi9mbm/S1tzgY+CazLzIci4iW9KliS1FmVEfrFwERmTmbmU8AeYH1bmw3ALZn5EEBmnqi3TElSN1UCfTlwrGV/qnms1S8AL4yI/RFxd0S8u64CJUnVVPnEouhwLDtc55eBNwLPAf41Iu7KzG+ddqGIjcBGgFWrVs2/2qbRLbcC8PDkozP7R7e9ZcHXk6QSVBmhTwErW/ZXAMc7tPlqZv4oMx8BvgG8pv1CmbkzM8cyc2xkZGShNUuSOqgS6AeANRFxXkQ8G7gK2NvW5svAGyLirIj4OeAS4Ei9pUqS5tJ1yiUzT0XEZuA2YBmwKzMPR8Sm5vkdmXkkIr4K3Av8FLgpM+/rZeGShteFuy8EYPLhyZn9Q9cc6mdJS0KVOXQycx+wr+3Yjrb9jwIfra80SdJ8+E5RSSqEgS5Jhag05bJUnbthW79LkKQlwxG6JBVioEfokobb6htW97uEJcURuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgpRKdAjYl1EPBgRExGxZY52vxIRP4mIt9dXoiSpiq6BHhHLgO3A5cAFwNURccEs7f4cuK3uIiVJ3VUZoV8MTGTmZGY+BewB1ndo9wfAF4ATNdYnSaqoSqAvB4617E81j82IiOXAbwM75rpQRGyMiIMRcfDkyZPzrVWSNIcqgR4djmXb/seB6zPzJ3NdKDN3ZuZYZo6NjIxUrVGSVMFZFdpMAStb9lcAx9vajAF7IgLgHOCKiDiVmV+qpUpJUldVAv0AsCYizgO+C1wFbGhtkJnnTW9HxM3AVwxzSVpcXQM9M09FxGYad68sA3Zl5uGI2NQ8P+e8uSRpcVQZoZOZ+4B9bcc6BnlmXnvmZUmS5st3ikpSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKUSnQI2JdRDwYERMRsaXD+d+JiHubX9+MiNfUX6okaS5dAz0ilgHbgcuBC4CrI+KCtmbfBn49M18NfBjYWXehkqS5VRmhXwxMZOZkZj4F7AHWtzbIzG9m5g+au3cBK+otU5LUTZVAXw4ca9mfah6bze8B/9jpRERsjIiDEXHw5MmT1auUJHVVJdCjw7Hs2DDiUhqBfn2n85m5MzPHMnNsZGSkepWSpK7OqtBmCljZsr8CON7eKCJeDdwEXJ6Zj9ZTniSpqioj9APAmog4LyKeDVwF7G1tEBGrgFuAd2Xmt+ovU5LUTdcRemaeiojNwG3AMmBXZh6OiE3N8zuADwIvBj4ZEQCnMnOsd2VLktpVmXIhM/cB+9qO7WjZfi/w3npLkyTNh+8UlaRCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBWiUqBHxLqIeDAiJiJiS4fzERGfaJ6/NyJeW3+pkqS5dA30iFgGbAcuBy4Aro6IC9qaXQ6saX5tBD5Vc52SpC6qjNAvBiYyczIznwL2AOvb2qwHPp0NdwFnR8TLaq5VkjSHyMy5G0S8HViXme9t7r8LuCQzN7e0+QqwLTP/pbl/O3B9Zh5su9ZGGiN4gFcBD9bQh3OAR2q4ziCxz8PBPg+H+fb5FZk50unEWRX+cHQ41v4sUKUNmbkT2FnhMSuLiIOZOVbnNZc6+zwc7PNwqLPPVaZcpoCVLfsrgOMLaCNJ6qEqgX4AWBMR50XEs4GrgL1tbfYC727e7fI64InM/F7NtUqS5tB1yiUzT0XEZuA2YBmwKzMPR8Sm5vkdwD7gCmACeBJ4T+9KfoZap3AGhH0eDvZ5ONTW564vikqSBoPvFJWkQhjoklSIgQ30bssRlCIidkXEiYi4r+XYiyLiaxHx383vL+xnjXWKiJUR8fWIOBIRhyPifc3jxfYZICJ+NiL+PSL+q9nvP20eL73fyyLiP5vvZSm+vwARcTQiDkXEPRFxsHmsln4PZKBXXI6gFDcD69qObQFuz8w1wO3N/VKcAj6QmWuB1wHXNf9uS+4zwI+ByzLzNcBFwLrmHWOl9/t9wJGW/dL7O+3SzLyo5f7zWvo9kIFOteUIipCZ3wAeazu8Htjd3N4N/NaiFtVDmfm9zPyP5vb/0fhlX07BfQZoLpvxw+bus5pfScH9jogVwFuAm1oOF9vfLmrp96AG+nLgWMv+VPPYsHjp9H3+ze8v6XM9PRERo8AvAf/GEPS5Of1wD3AC+Fpmlt7vjwN/BPy05VjJ/Z2WwD9FxN3N5VCgpn5Xeev/UlRpqQENroh4HvAF4P2Z+b8Rnf7Ky5KZPwEuioizgS9GxC/2u6ZeiYgrgROZeXdEjPe7nkX2+sw8HhEvAb4WEQ/UdeFBHaEP+1ID359ezbL5/USf66lVRDyLRph/NjNvaR4uus+tMvNxYD+N105K7ffrgbdFxFEaU6aXRcRnKLe/MzLzePP7CeCLNKaQa+n3oAZ6leUISrYXuKa5fQ3w5T7WUqtoDMX/BjiSmTe2nCq2zwARMdIcmRMRzwF+A3iAQvudmTdk5orMHKXx+3tHZv4uhfZ3WkQ8NyKeP70NvBm4j5r6PbDvFI2IK2jMwU0vR/CRPpfUExHxOWCcxhKb3wc+BHwJ+DywCngIeEdmtr9wOpAi4teAO4FDPD23+sc05tGL7DNARLyaxothy2gMtD6fmX8WES+m4H4DNKdc/jAzryy9vxGxmsaoHBpT3n+XmR+pq98DG+iSpNMN6pSLJKmNgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIK8f+puZMJmizXeQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for j in range(50):\n",
    "    data=[]\n",
    "    for i in range(10):\n",
    "        data.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNED.csv').iloc[:,1].values)\n",
    "    plt.bar(j,np.mean(data), yerr=np.std(data))\n",
    "plt.show()\n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for j in range(15):\n",
    "    data=[]\n",
    "    for i in range(10):\n",
    "        data.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RotF.csv').iloc[:,1].values)\n",
    "    plt.bar(j,np.mean(data), yerr=np.std(data))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot Together"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "euclid_dist = pd.read_csv('Data/distances_euc.csv').iloc[:,1].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for j in range(15):\n",
    "    data_ED=[]\n",
    "    data_RotF=[]\n",
    "    data_RandF=[]\n",
    "    data_DTW1=[]\n",
    "    for i in range(7):\n",
    "        data_ED.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNED.csv').iloc[:,1].values)\n",
    "        data_RotF.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RotF.csv').iloc[:,1].values)\n",
    "        data_RandF.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RandF.csv').iloc[:,1].values)\n",
    "        data_DTW1.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW1.csv').iloc[:,1].values)\n",
    "    plt.scatter(euclid_dist[j],np.mean(data_ED),color='C0')\n",
    "    plt.scatter(euclid_dist[j],np.mean(data_DTW1),color='C1')\n",
    "    plt.scatter(euclid_dist[j],np.mean(data_RotF),color='C2')\n",
    "    plt.scatter(euclid_dist[j],np.mean(data_RandF),color='C3')\n",
    "plt.legend(['1NN ED','1NN DTW w=1','RotF','RandF'], loc='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparing Classifier Performance "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0.650585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0.583333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0.597953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0.644737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0.630117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>0.616959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>0.612573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>0.619883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>0.640351</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0         0\n",
       "0           0  0.650585\n",
       "1           1  0.583333\n",
       "2           2  0.597953\n",
       "3           3  0.644737\n",
       "4           4  0.630117\n",
       "5           5  0.616959\n",
       "6           6  0.612573\n",
       "7           7  0.619883\n",
       "8           8  0.640351"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_csv('Results/noise'+str(0)+'/duo'+str(0)+'/RotF.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7875313843467857 0.7687169803289499\n",
      "0.9325501867360855 0.9300091667351034\n",
      "0.9561070649015224 0.9505736945184672\n",
      "0.9848545562714554 0.9832564919067356\n",
      "0.9910293777415284 0.9897942181335792\n",
      "0.9917925232697924 0.9925738468535097\n",
      "0.9923745740958982 0.9930444851527389\n",
      "0.9927065653581278 0.9922628743486095\n",
      "0.9922789396447563 0.9917477468179002\n",
      "0.9880861476396194 0.987805981010591\n",
      "0.9817407946743576 0.9767470892101388\n",
      "0.9934783733532038 0.9931250175715489\n",
      "0.9822266781876309 0.9830771980310321\n",
      "0.9944155106178632 0.9938483484194048\n",
      "0.9954441152764048 0.9956968375090882\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.style.use('grayscale')\n",
    "fig = plt.figure()\n",
    "\n",
    "for j in range(15):\n",
    "    data_ED=[]\n",
    "    data_RotF=[]\n",
    "    data_RandF=[]\n",
    "    data_DTW1=[]\n",
    "    for i in range(10):\n",
    "        data_ED.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNED.csv').iloc[:,1].values)\n",
    "        data_RotF.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RotF50.csv').iloc[:,1].values)\n",
    "        data_RandF.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RandF.csv').iloc[:,1].values)\n",
    "        data_DTW1.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW1.csv').iloc[:,1:].values)\n",
    "\n",
    "    plt.bar(j-.3,np.mean(data_ED),color='C0', width=0.2, yerr=np.std(data_ED))\n",
    "    plt.bar(j-.1,np.mean(data_DTW1),color='C1', width=0.2, yerr=np.std(data_DTW1))\n",
    "    plt.bar(j+.1,np.mean(data_RotF),color='C2', width=0.2, yerr=np.std(data_RotF))\n",
    "    plt.bar(j+.3,np.mean(data_RandF),color='C3', width=0.2, yerr=np.std(data_RandF))\n",
    "#     print(np.mean(data_ED), np.mean(data_DTW1))\n",
    "    print(np.mean(data_RotF), np.mean(data_RandF))\n",
    "\n",
    "    \n",
    "plt.legend(['1NN ED','1NN DTW w=1','RotF','RandF'], loc='lower right')\n",
    "plt.xticks(np.arange(15), np.arange(1,16))\n",
    "plt.axhline(y=0.9, color='black', linestyle='dashed', linewidth=1)\n",
    "plt.ylim(0.5,1.05)\n",
    "plt.title('Classification Performance (10-Noise Models)')\n",
    "plt.xlabel('Organism Pair')\n",
    "plt.ylabel('Classification Accuracy')\n",
    "fig.patch.set_facecolor('none')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "times\n",
    "1NN ED 50: 1405.1696496009827\n",
    "RandF 15: 1479.9977066516876\n",
    "RotF 10: 29483.410986423492\n",
    "1NN DTW w=1: 13631.643\n",
    "1NN DTW w=2: 15998.342 \n",
    "1NN DTW w=3: 15591.04\n",
    "1NN DTW w=4: 15744.394\n",
    "1NN DTW w=5: 16604.305"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for j in range(15):\n",
    "    data_DTW1=[]\n",
    "    data_DTW2=[]\n",
    "    data_DTW3=[]\n",
    "    data_DTW4=[]\n",
    "    data_DTW5=[]\n",
    "\n",
    "    for i in range(10):\n",
    "        data_DTW1.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW1.csv').iloc[:,1:].values)\n",
    "        data_DTW2.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW2.csv').iloc[:,1:].values)\n",
    "        data_DTW3.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW3.csv').iloc[:,1:].values)\n",
    "        data_DTW4.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW4.csv').iloc[:,1:].values)\n",
    "        data_DTW5.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW5.csv').iloc[:,1:].values)\n",
    "\n",
    "    plt.bar(j-.2,np.mean(data_DTW1),color='C0', width=0.1)\n",
    "    plt.bar(j-.1,np.mean(data_DTW2),color='C1', width=0.1)\n",
    "    plt.bar(j,np.mean(data_DTW3),color='C2', width=0.1)\n",
    "    plt.bar(j+.1,np.mean(data_DTW4),color='C3', width=0.1)\n",
    "    plt.bar(j+.2,np.mean(data_DTW5),color='C4', width=0.1)\n",
    "\n",
    "plt.legend(['1NN DTW w=1','1NN DTW w=2','1NN DTW w=3','1NN DTW w=4','1NN DTW w=5'], loc='lower right')\n",
    "plt.hlines(y=0.9, xmin=-1, xmax=15, color='red', linestyle='dashed')\n",
    "plt.xticks(np.arange(15), np.arange(1,16))\n",
    "plt.title('Cross Validation of DTW Window')\n",
    "plt.xlabel('Organism Pair')\n",
    "plt.ylabel('Classification Accuracy')\n",
    "plt.ylim(0.5,1.01)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparing Noise Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "\n",
    "all_data_ED=[]\n",
    "all_data_RotF=[]\n",
    "all_data_RandF=[]\n",
    "all_data_DTW1=[]\n",
    "for i in range(10):\n",
    "    data_ED=[]\n",
    "    data_RotF=[]\n",
    "    data_RandF=[]\n",
    "    data_DTW1=[]\n",
    "    for j in range(15):\n",
    "        data_ED.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNED.csv').iloc[:,1].values)\n",
    "        data_RotF.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RotF50.csv').iloc[:,1].values)\n",
    "        data_RandF.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/RandF.csv').iloc[:,1].values)\n",
    "        data_DTW1.append(pd.read_csv('Results/noise'+str(i)+'/duo'+str(j)+'/1NNDTW1.csv').iloc[:,1:].values)\n",
    "\n",
    "    plt.bar(i-.3,np.mean(data_ED),color='C0', width=0.2)\n",
    "    plt.bar(i-.1,np.mean(data_DTW1),color='C1', width=0.2)\n",
    "    plt.bar(i+.1,np.mean(data_RotF),color='C2', width=0.2)\n",
    "    plt.bar(i+.3,np.mean(data_RandF),color='C3', width=0.2)\n",
    "#     plt.bar(i-.3,np.mean(data_ED),color='C0', width=0.2, yerr=np.std(data_ED))\n",
    "#     plt.bar(i-.1,np.mean(data_DTW1),color='C1', width=0.2, yerr=np.std(data_DTW1))\n",
    "#     plt.bar(i+.1,np.mean(data_RotF),color='C2', width=0.2, yerr=np.std(data_RotF))\n",
    "#     plt.bar(i+.3,np.mean(data_RandF),color='C3', width=0.2, yerr=np.std(data_RandF))\n",
    "    all_data_ED.append(data_ED)\n",
    "    all_data_RotF.append(data_RotF)\n",
    "    all_data_RandF.append(data_RandF)\n",
    "    all_data_DTW1.append(data_DTW1)\n",
    "    \n",
    "plt.legend(['1NN ED','1NN DTW w=1','RotF','RandF'], loc='lower left')\n",
    "plt.xticks(np.arange(10), np.arange(1,11))\n",
    "plt.axhline(y=np.mean(all_data_ED), color='C0', linestyle='dashed', linewidth=1)\n",
    "plt.axhline(y=np.mean(all_data_DTW1), color='C1', linestyle='dashed', linewidth=1)\n",
    "plt.axhline(y=np.mean(all_data_RotF), color='C2', linestyle='dashed', linewidth=1)\n",
    "plt.axhline(y=np.mean(all_data_RandF), color='C3', linestyle='dashed', linewidth=1)\n",
    "\n",
    "plt.text(5.7, 1, '**', color='C1', fontsize=12, fontweight='bold')\n",
    "plt.text(6.6, 1, '*', color='C0', fontsize=12, fontweight='bold')\n",
    "plt.text(7, 1, '*', color='C2', fontsize=12, fontweight='bold')\n",
    "# plt.text(7.2, 1, '**', color='C3', fontsize=12, fontweight='bold')\n",
    "plt.text(7.12, .975, '**', color='C3', fontsize=12, fontweight='bold') #grayscale\n",
    "\n",
    "\n",
    "plt.ylim(0.5,1.05)\n",
    "plt.title('Noise Model Comparison')\n",
    "plt.xlabel('Noise Model')\n",
    "plt.ylabel('Classification Accuracy')\n",
    "fig.patch.set_facecolor('none')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8578503992586496"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(all_data_ED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Statistical Significance\n",
    "https://medium.com/analytics-vidhya/using-the-corrected-paired-students-t-test-for-comparing-the-performance-of-machine-learning-dc6529eaa97f"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "z10 = 1.28\n",
    "z05 = 1.645\n",
    "z025 = 1.96\n",
    "z01 = 2.33\n",
    "z005 = 2.58"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "zs=[]\n",
    "for i in range(10):\n",
    "    zn=[]\n",
    "    for j in range(10):\n",
    "        acc1 = np.mean(all_data_RandF[i])\n",
    "        acc2 = np.mean(all_data_RandF[j])\n",
    "        n1 = lengths_train[i]\n",
    "        n2 = lengths_train[j]\n",
    "\n",
    "        acc12 = (acc1*n1+acc2*n2)/(n1+n2)\n",
    "        sig12 = np.sqrt(acc12*(1-acc12)/(n1+n2))\n",
    "        z = abs((acc1-acc2)/sig12)\n",
    "        zn.append(z)\n",
    "#         print(abs(z))\n",
    "    zs.append(zn)\n",
    "# a = pd.DataFrame(np.triu(np.array(zs)))\n",
    "a = pd.DataFrame((np.array(zs)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://www3.nd.edu/~rjohns15/cse40647.sp14/www/content/lectures/28%20-%20Classifier%20Comparisons.pdf\n",
    "https://www2.southeastern.edu/Academics/Faculty/dgurney/Math241/StatTopics/ZAlpha.htm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandF: pass\n"
     ]
    }
   ],
   "source": [
    "methods = [all_data_ED, all_data_DTW1, all_data_RotF,all_data_RandF]\n",
    "methods_names = ['1NN ED','1NN DTW w=1','RotF','RandF']\n",
    "z90 = 1.645\n",
    "z95 = 1.960\n",
    "z98 = 2.326\n",
    "z99 = 2.576\n",
    "z999 = 3.2905267283\n",
    "z = z99\n",
    "res=[]\n",
    "for k,m in enumerate(methods):\n",
    "    zs=[]\n",
    "    for i in range(10):\n",
    "        zn=[]\n",
    "        for j in range(10):\n",
    "            e1 = 1-np.mean(m[i])\n",
    "    #         print(acc1)\n",
    "            e2 = 1-np.mean(m[j])\n",
    "            n1 = lengths_train[i]\n",
    "            n2 = lengths_train[j]\n",
    "            sig = e1*(1-e1)/n1+e2*(1-e2)/n2\n",
    "            d = abs(e1-e2)-z*np.sqrt(sig)\n",
    "            zn.append(d)\n",
    "    #         print(abs(z))\n",
    "        zs.append(zn)\n",
    "    # a = pd.DataFrame(np.triu(np.array(zs)))\n",
    "    a = pd.DataFrame(np.array(zs))\n",
    "    t = np.hstack([a[7][:7], a[7][8:]])\n",
    "    if any(x<0 for x in t) == False:\n",
    "        print(methods_names[k]+': pass')\n",
    "    res.append(a)\n",
    "#     print(a.where(a>0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.006903</td>\n",
       "      <td>0.009942</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.011974</td>\n",
       "      <td>0.017918</td>\n",
       "      <td>0.009083</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.005656</td>\n",
       "      <td>0.014237</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.006903</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.001032</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.031225</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.009942</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.004225</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.033979</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.001032</td>\n",
       "      <td>0.004225</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.006272</td>\n",
       "      <td>0.012256</td>\n",
       "      <td>0.016662</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.011974</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.006272</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.035982</td>\n",
       "      <td>0.001411</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.017918</td>\n",
       "      <td>0.005656</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.012256</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.041858</td>\n",
       "      <td>0.007414</td>\n",
       "      <td>0.002078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.009083</td>\n",
       "      <td>0.014237</td>\n",
       "      <td>0.031225</td>\n",
       "      <td>0.033979</td>\n",
       "      <td>0.016662</td>\n",
       "      <td>0.035982</td>\n",
       "      <td>0.041858</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.022300</td>\n",
       "      <td>0.024374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.001411</td>\n",
       "      <td>0.007414</td>\n",
       "      <td>0.022300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.002078</td>\n",
       "      <td>0.024374</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          0         1         2         3         4         5         6  \\\n",
       "0       NaN       NaN  0.006903  0.009942       NaN  0.011974  0.017918   \n",
       "1       NaN       NaN       NaN       NaN       NaN       NaN  0.005656   \n",
       "2  0.006903       NaN       NaN       NaN  0.001032       NaN       NaN   \n",
       "3  0.009942       NaN       NaN       NaN  0.004225       NaN       NaN   \n",
       "4       NaN       NaN  0.001032  0.004225       NaN  0.006272  0.012256   \n",
       "5  0.011974       NaN       NaN       NaN  0.006272       NaN       NaN   \n",
       "6  0.017918  0.005656       NaN       NaN  0.012256       NaN       NaN   \n",
       "7  0.009083  0.014237  0.031225  0.033979  0.016662  0.035982  0.041858   \n",
       "8       NaN       NaN       NaN       NaN       NaN  0.001411  0.007414   \n",
       "9       NaN       NaN       NaN       NaN       NaN       NaN  0.002078   \n",
       "\n",
       "          7         8         9  \n",
       "0  0.009083       NaN       NaN  \n",
       "1  0.014237       NaN       NaN  \n",
       "2  0.031225       NaN       NaN  \n",
       "3  0.033979       NaN       NaN  \n",
       "4  0.016662       NaN       NaN  \n",
       "5  0.035982  0.001411       NaN  \n",
       "6  0.041858  0.007414  0.002078  \n",
       "7       NaN  0.022300  0.024374  \n",
       "8  0.022300       NaN       NaN  \n",
       "9  0.024374       NaN       NaN  "
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "i=3\n",
    "res[i][res[i]>0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1368, 560, 756, 1498, 1508, 1496, 1352, 1336, 1468, 836] [684, 282, 380, 750, 756, 748, 676, 670, 734, 418]\n"
     ]
    }
   ],
   "source": [
    "lengths_train = []\n",
    "lengths_test = []\n",
    "for i in range(10):\n",
    "        lengths_train.append(len(pd.read_csv('Results/noise'+str(i)+'/duo'+str(0)+'/split'+str(0)+'/x_train'+str(0)+'.csv')))\n",
    "        lengths_test.append(len(pd.read_csv('Results/noise'+str(i)+'/duo'+str(0)+'/split'+str(0)+'/x_test'+str(0)+'.csv')))\n",
    "print(lengths_train, lengths_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Try t-test as well"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "reject the null hypothesis of equal averages.\n"
     ]
    }
   ],
   "source": [
    "from scipy import stats\n",
    "\n",
    "if(stats.ttest_ind(np.vstack(all_data_ED[0]).ravel(),np.vstack(all_data_ED[1]).ravel(),equal_var = False)[1] < 0.05):\n",
    "    print('noise models are different.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy import stats\n",
    "import statsmodels.api as sm\n",
    "methods = [all_data_ED, all_data_DTW1, all_data_RotF,all_data_RandF]\n",
    "\n",
    "res=[]\n",
    "for k,m in enumerate(methods):\n",
    "    zs=[]\n",
    "    for i in range(10):\n",
    "        zn=[]\n",
    "        for j in range(10):\n",
    "            #Welch’s t-test, which does not assume equal population variance\n",
    "#             if(stats.ttest_ind(np.vstack(m[i]).ravel(),np.vstack(m[j]).ravel(),equal_var = False)[1] < 0.05):\n",
    "            #Welch’s t-test, but taken mean of 9-fold CV first, to get rid of dependence\n",
    "            if(stats.ttest_ind(np.mean(np.vstack(m[i]),axis=1),np.mean(np.vstack(m[j]),axis=1),equal_var = False)[1] < 0.05):                 \n",
    "            #Z-test        \n",
    "#             if(sm.stats.CompareMeans.from_data(np.vstack(m[i]).ravel(),np.vstack(m[j]).ravel()).ztest_ind(usevar = 'unequal')[1] < 0.05):\n",
    "#                 print('noise models are different.')\n",
    "                zn.append(1)\n",
    "            else:\n",
    "                zn.append(0)\n",
    "        zs.append(zn)\n",
    "    res.append(zs)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(array([], dtype=int64),)\n",
      "(array([], dtype=int64),)\n",
      "(array([], dtype=int64),)\n",
      "(array([], dtype=int64),)\n"
     ]
    }
   ],
   "source": [
    "for r in res:\n",
    "    print(np.where(np.sum(r,axis=0) == 9))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to this test, noise model 8 (7+1) is different for all classifiers\n",
    "but doesn't take into account number of curves in dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.591131</td>\n",
       "      <td>0.552403</td>\n",
       "      <td>0.624269</td>\n",
       "      <td>0.605778</td>\n",
       "      <td>0.588771</td>\n",
       "      <td>0.629679</td>\n",
       "      <td>0.634944</td>\n",
       "      <td>0.569320</td>\n",
       "      <td>0.584469</td>\n",
       "      <td>0.619883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.927388</td>\n",
       "      <td>0.786446</td>\n",
       "      <td>0.861696</td>\n",
       "      <td>0.837333</td>\n",
       "      <td>0.807025</td>\n",
       "      <td>0.871658</td>\n",
       "      <td>0.821828</td>\n",
       "      <td>0.618574</td>\n",
       "      <td>0.637754</td>\n",
       "      <td>0.714514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.742528</td>\n",
       "      <td>0.672971</td>\n",
       "      <td>0.785673</td>\n",
       "      <td>0.753481</td>\n",
       "      <td>0.733833</td>\n",
       "      <td>0.846999</td>\n",
       "      <td>0.906147</td>\n",
       "      <td>0.672471</td>\n",
       "      <td>0.703906</td>\n",
       "      <td>0.809942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.806368</td>\n",
       "      <td>0.738771</td>\n",
       "      <td>0.844152</td>\n",
       "      <td>0.790667</td>\n",
       "      <td>0.773369</td>\n",
       "      <td>0.879679</td>\n",
       "      <td>0.932446</td>\n",
       "      <td>0.691874</td>\n",
       "      <td>0.734332</td>\n",
       "      <td>0.846624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.896361</td>\n",
       "      <td>0.778172</td>\n",
       "      <td>0.907602</td>\n",
       "      <td>0.915704</td>\n",
       "      <td>0.885949</td>\n",
       "      <td>0.946078</td>\n",
       "      <td>0.950526</td>\n",
       "      <td>0.771476</td>\n",
       "      <td>0.806085</td>\n",
       "      <td>0.902711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.934048</td>\n",
       "      <td>0.967691</td>\n",
       "      <td>0.996491</td>\n",
       "      <td>0.960444</td>\n",
       "      <td>0.941505</td>\n",
       "      <td>0.972965</td>\n",
       "      <td>0.997535</td>\n",
       "      <td>0.915755</td>\n",
       "      <td>0.973358</td>\n",
       "      <td>0.995215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.876056</td>\n",
       "      <td>0.823877</td>\n",
       "      <td>0.950585</td>\n",
       "      <td>0.875556</td>\n",
       "      <td>0.858466</td>\n",
       "      <td>0.941176</td>\n",
       "      <td>0.970414</td>\n",
       "      <td>0.762023</td>\n",
       "      <td>0.781259</td>\n",
       "      <td>0.935673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.880929</td>\n",
       "      <td>0.838849</td>\n",
       "      <td>0.961404</td>\n",
       "      <td>0.893185</td>\n",
       "      <td>0.870370</td>\n",
       "      <td>0.949346</td>\n",
       "      <td>0.972715</td>\n",
       "      <td>0.778109</td>\n",
       "      <td>0.803815</td>\n",
       "      <td>0.946039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.888564</td>\n",
       "      <td>0.886525</td>\n",
       "      <td>0.975439</td>\n",
       "      <td>0.920593</td>\n",
       "      <td>0.907995</td>\n",
       "      <td>0.966132</td>\n",
       "      <td>0.984714</td>\n",
       "      <td>0.807131</td>\n",
       "      <td>0.836512</td>\n",
       "      <td>0.970494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.930962</td>\n",
       "      <td>0.852640</td>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.970074</td>\n",
       "      <td>0.952087</td>\n",
       "      <td>0.978313</td>\n",
       "      <td>0.978632</td>\n",
       "      <td>0.797512</td>\n",
       "      <td>0.818498</td>\n",
       "      <td>0.958799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.757960</td>\n",
       "      <td>0.708826</td>\n",
       "      <td>0.793860</td>\n",
       "      <td>0.753926</td>\n",
       "      <td>0.752352</td>\n",
       "      <td>0.835710</td>\n",
       "      <td>0.890039</td>\n",
       "      <td>0.668325</td>\n",
       "      <td>0.727066</td>\n",
       "      <td>0.791334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.902859</td>\n",
       "      <td>0.941292</td>\n",
       "      <td>0.993567</td>\n",
       "      <td>0.945037</td>\n",
       "      <td>0.937243</td>\n",
       "      <td>0.976084</td>\n",
       "      <td>0.995069</td>\n",
       "      <td>0.872471</td>\n",
       "      <td>0.940963</td>\n",
       "      <td>0.992026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.899285</td>\n",
       "      <td>0.778960</td>\n",
       "      <td>0.908772</td>\n",
       "      <td>0.919407</td>\n",
       "      <td>0.879630</td>\n",
       "      <td>0.949643</td>\n",
       "      <td>0.952498</td>\n",
       "      <td>0.728358</td>\n",
       "      <td>0.759612</td>\n",
       "      <td>0.905369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.921540</td>\n",
       "      <td>0.854216</td>\n",
       "      <td>0.968129</td>\n",
       "      <td>0.963111</td>\n",
       "      <td>0.953410</td>\n",
       "      <td>0.978313</td>\n",
       "      <td>0.978468</td>\n",
       "      <td>0.840464</td>\n",
       "      <td>0.892825</td>\n",
       "      <td>0.955077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.935673</td>\n",
       "      <td>0.900709</td>\n",
       "      <td>0.980409</td>\n",
       "      <td>0.982370</td>\n",
       "      <td>0.975162</td>\n",
       "      <td>0.987077</td>\n",
       "      <td>0.986686</td>\n",
       "      <td>0.870149</td>\n",
       "      <td>0.933394</td>\n",
       "      <td>0.977937</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           0         1         2         3         4         5         6  \\\n",
       "0   0.591131  0.552403  0.624269  0.605778  0.588771  0.629679  0.634944   \n",
       "1   0.927388  0.786446  0.861696  0.837333  0.807025  0.871658  0.821828   \n",
       "2   0.742528  0.672971  0.785673  0.753481  0.733833  0.846999  0.906147   \n",
       "3   0.806368  0.738771  0.844152  0.790667  0.773369  0.879679  0.932446   \n",
       "4   0.896361  0.778172  0.907602  0.915704  0.885949  0.946078  0.950526   \n",
       "5   0.934048  0.967691  0.996491  0.960444  0.941505  0.972965  0.997535   \n",
       "6   0.876056  0.823877  0.950585  0.875556  0.858466  0.941176  0.970414   \n",
       "7   0.880929  0.838849  0.961404  0.893185  0.870370  0.949346  0.972715   \n",
       "8   0.888564  0.886525  0.975439  0.920593  0.907995  0.966132  0.984714   \n",
       "9   0.930962  0.852640  0.966667  0.970074  0.952087  0.978313  0.978632   \n",
       "10  0.757960  0.708826  0.793860  0.753926  0.752352  0.835710  0.890039   \n",
       "11  0.902859  0.941292  0.993567  0.945037  0.937243  0.976084  0.995069   \n",
       "12  0.899285  0.778960  0.908772  0.919407  0.879630  0.949643  0.952498   \n",
       "13  0.921540  0.854216  0.968129  0.963111  0.953410  0.978313  0.978468   \n",
       "14  0.935673  0.900709  0.980409  0.982370  0.975162  0.987077  0.986686   \n",
       "\n",
       "           7         8         9  \n",
       "0   0.569320  0.584469  0.619883  \n",
       "1   0.618574  0.637754  0.714514  \n",
       "2   0.672471  0.703906  0.809942  \n",
       "3   0.691874  0.734332  0.846624  \n",
       "4   0.771476  0.806085  0.902711  \n",
       "5   0.915755  0.973358  0.995215  \n",
       "6   0.762023  0.781259  0.935673  \n",
       "7   0.778109  0.803815  0.946039  \n",
       "8   0.807131  0.836512  0.970494  \n",
       "9   0.797512  0.818498  0.958799  \n",
       "10  0.668325  0.727066  0.791334  \n",
       "11  0.872471  0.940963  0.992026  \n",
       "12  0.728358  0.759612  0.905369  \n",
       "13  0.840464  0.892825  0.955077  \n",
       "14  0.870149  0.933394  0.977937  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = pd.DataFrame([np.mean(np.vstack(all_data_ED[i]),axis=1) for i in range(10)]).T\n",
    "d"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://reneshbedre.github.io/blog/anova.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Multiple Comparison of Means - Tukey HSD, FWER=0.05 \n",
      "====================================================\n",
      "group1 group2 meandiff p-adj   lower   upper  reject\n",
      "----------------------------------------------------\n",
      "     0      1   -0.054    0.9  -0.175  0.0671  False\n",
      "     0      2   0.0418    0.9 -0.0792  0.1628  False\n",
      "     0      3    0.013    0.9  -0.108   0.134  False\n",
      "     0      4   -0.005    0.9  -0.126   0.116  False\n",
      "     0      5   0.0545    0.9 -0.0665  0.1755  False\n",
      "     0      6   0.0707 0.6593 -0.0503  0.1917  False\n",
      "     0      7  -0.1018 0.1813 -0.2229  0.0192  False\n",
      "     0      8  -0.0639 0.7695 -0.1849  0.0572  False\n",
      "     0      9   0.0287    0.9 -0.0923  0.1497  False\n",
      "     1      2   0.0958 0.2547 -0.0253  0.2168  False\n",
      "     1      3    0.067 0.7198 -0.0541   0.188  False\n",
      "     1      4    0.049    0.9  -0.072    0.17  False\n",
      "     1      5   0.1084 0.1214 -0.0126  0.2294  False\n",
      "     1      6   0.1247 0.0378  0.0037  0.2457   True\n",
      "     1      7  -0.0479    0.9 -0.1689  0.0731  False\n",
      "     1      8  -0.0099    0.9 -0.1309  0.1111  False\n",
      "     1      9   0.0826 0.4675 -0.0384  0.2036  False\n",
      "     2      3  -0.0288    0.9 -0.1498  0.0922  False\n",
      "     2      4  -0.0468    0.9 -0.1678  0.0742  False\n",
      "     2      5   0.0127    0.9 -0.1083  0.1337  False\n",
      "     2      6   0.0289    0.9 -0.0921  0.1499  False\n",
      "     2      7  -0.1436 0.0075 -0.2647 -0.0226   True\n",
      "     2      8  -0.1057  0.144 -0.2267  0.0154  False\n",
      "     2      9  -0.0131    0.9 -0.1342  0.1079  False\n",
      "     3      4   -0.018    0.9  -0.139   0.103  False\n",
      "     3      5   0.0415    0.9 -0.0795  0.1625  False\n",
      "     3      6   0.0577 0.8674 -0.0633  0.1787  False\n",
      "     3      7  -0.1148 0.0788 -0.2359  0.0062  False\n",
      "     3      8  -0.0769 0.5613 -0.1979  0.0442  False\n",
      "     3      9   0.0157    0.9 -0.1053  0.1367  False\n",
      "     4      5   0.0594   0.84 -0.0616  0.1805  False\n",
      "     4      6   0.0757 0.5798 -0.0453  0.1967  False\n",
      "     4      7  -0.0969 0.2398 -0.2179  0.0241  False\n",
      "     4      8  -0.0589 0.8489 -0.1799  0.0621  False\n",
      "     4      9   0.0336    0.9 -0.0874  0.1546  False\n",
      "     5      6   0.0163    0.9 -0.1048  0.1373  False\n",
      "     5      7  -0.1563 0.0022 -0.2773 -0.0353   True\n",
      "     5      8  -0.1183 0.0612 -0.2393  0.0027  False\n",
      "     5      9  -0.0258    0.9 -0.1468  0.0952  False\n",
      "     6      7  -0.1726  0.001 -0.2936 -0.0516   True\n",
      "     6      8  -0.1346 0.0168 -0.2556 -0.0136   True\n",
      "     6      9  -0.0421    0.9 -0.1631  0.0789  False\n",
      "     7      8    0.038    0.9  -0.083   0.159  False\n",
      "     7      9   0.1305 0.0236  0.0095  0.2515   True\n",
      "     8      9   0.0925 0.3007 -0.0285  0.2135  False\n",
      "----------------------------------------------------\n",
      "0.9086856245994568 4.218832927449512e-08\n",
      "Multiple Comparison of Means - Tukey HSD, FWER=0.05 \n",
      "====================================================\n",
      "group1 group2 meandiff p-adj   lower   upper  reject\n",
      "----------------------------------------------------\n",
      "     0      1  -0.0474    0.9 -0.1657  0.0708  False\n",
      "     0      2    0.041    0.9 -0.0773  0.1592  False\n",
      "     0      3  -0.0102    0.9 -0.1285  0.1081  False\n",
      "     0      4  -0.0236    0.9 -0.1418  0.0947  False\n",
      "     0      5   0.0597 0.8145 -0.0586  0.1779  False\n",
      "     0      6   0.0915 0.2853 -0.0268  0.2097  False\n",
      "     0      7  -0.1003 0.1734 -0.2186  0.0179  False\n",
      "     0      8  -0.0759 0.5484 -0.1942  0.0424  False\n",
      "     0      9   0.0412    0.9 -0.0771  0.1595  False\n",
      "     1      2   0.0884 0.3322 -0.0299  0.2067  False\n",
      "     1      3   0.0372    0.9  -0.081  0.1555  False\n",
      "     1      4   0.0239    0.9 -0.0944  0.1422  False\n",
      "     1      5   0.1071 0.1125 -0.0112  0.2254  False\n",
      "     1      6   0.1389 0.0086  0.0206  0.2572   True\n",
      "     1      7  -0.0529    0.9 -0.1712  0.0654  False\n",
      "     1      8  -0.0285    0.9 -0.1467  0.0898  False\n",
      "     1      9   0.0887 0.3279 -0.0296  0.2069  False\n",
      "     2      3  -0.0512    0.9 -0.1694  0.0671  False\n",
      "     2      4  -0.0645  0.735 -0.1828  0.0538  False\n",
      "     2      5   0.0187    0.9 -0.0996   0.137  False\n",
      "     2      6   0.0505    0.9 -0.0678  0.1688  False\n",
      "     2      7  -0.1413 0.0069 -0.2595  -0.023   True\n",
      "     2      8  -0.1169 0.0558 -0.2351  0.0014  False\n",
      "     2      9   0.0003    0.9  -0.118  0.1185  False\n",
      "     3      4  -0.0133    0.9 -0.1316  0.1049  False\n",
      "     3      5   0.0699 0.6472 -0.0484  0.1881  False\n",
      "     3      6   0.1017 0.1595 -0.0166  0.2199  False\n",
      "     3      7  -0.0901 0.3055 -0.2084  0.0282  False\n",
      "     3      8  -0.0657 0.7157  -0.184  0.0526  False\n",
      "     3      9   0.0514    0.9 -0.0668  0.1697  False\n",
      "     4      5   0.0832 0.4234 -0.0351  0.2015  False\n",
      "     4      6    0.115 0.0642 -0.0033  0.2333  False\n",
      "     4      7  -0.0768 0.5342  -0.195  0.0415  False\n",
      "     4      8  -0.0523    0.9 -0.1706  0.0659  False\n",
      "     4      9   0.0648 0.7307 -0.0535   0.183  False\n",
      "     5      6   0.0318    0.9 -0.0865  0.1501  False\n",
      "     5      7    -0.16 0.0011 -0.2782 -0.0417   True\n",
      "     5      8  -0.1356 0.0117 -0.2538 -0.0173   True\n",
      "     5      9  -0.0184    0.9 -0.1367  0.0998  False\n",
      "     6      7  -0.1918  0.001   -0.31 -0.0735   True\n",
      "     6      8  -0.1674  0.001 -0.2856 -0.0491   True\n",
      "     6      9  -0.0502    0.9 -0.1685   0.068  False\n",
      "     7      8   0.0244    0.9 -0.0938  0.1427  False\n",
      "     7      9   0.1415 0.0067  0.0233  0.2598   True\n",
      "     8      9   0.1171 0.0547 -0.0011  0.2354  False\n",
      "----------------------------------------------------\n",
      "0.9265771508216858 5.777276328444714e-07\n",
      "Multiple Comparison of Means - Tukey HSD, FWER=0.05\n",
      "===================================================\n",
      "group1 group2 meandiff p-adj   lower  upper  reject\n",
      "---------------------------------------------------\n",
      "     0      1   0.0073    0.9 -0.0594 0.0741  False\n",
      "     0      2   0.0239    0.9 -0.0428 0.0906  False\n",
      "     0      3   0.0266    0.9 -0.0402 0.0933  False\n",
      "     0      4    0.014    0.9 -0.0527 0.0808  False\n",
      "     0      5   0.0269    0.9 -0.0398 0.0936  False\n",
      "     0      6   0.0317 0.8728 -0.0351 0.0984  False\n",
      "     0      7  -0.0203    0.9 -0.0871 0.0464  False\n",
      "     0      8   0.0158    0.9 -0.0509 0.0825  False\n",
      "     0      9   0.0193    0.9 -0.0475  0.086  False\n",
      "     1      2   0.0165    0.9 -0.0502 0.0833  False\n",
      "     1      3   0.0192    0.9 -0.0475 0.0859  False\n",
      "     1      4   0.0067    0.9   -0.06 0.0734  False\n",
      "     1      5   0.0195    0.9 -0.0472 0.0863  False\n",
      "     1      6   0.0243    0.9 -0.0424  0.091  False\n",
      "     1      7  -0.0277    0.9 -0.0944  0.039  False\n",
      "     1      8   0.0084    0.9 -0.0583 0.0752  False\n",
      "     1      9   0.0119    0.9 -0.0548 0.0786  False\n",
      "     2      3   0.0027    0.9 -0.0641 0.0694  False\n",
      "     2      4  -0.0099    0.9 -0.0766 0.0569  False\n",
      "     2      5    0.003    0.9 -0.0637 0.0697  False\n",
      "     2      6   0.0078    0.9  -0.059 0.0745  False\n",
      "     2      7  -0.0442 0.5077  -0.111 0.0225  False\n",
      "     2      8  -0.0081    0.9 -0.0748 0.0586  False\n",
      "     2      9  -0.0046    0.9 -0.0714 0.0621  False\n",
      "     3      4  -0.0125    0.9 -0.0793 0.0542  False\n",
      "     3      5   0.0003    0.9 -0.0664  0.067  False\n",
      "     3      6   0.0051    0.9 -0.0616 0.0718  False\n",
      "     3      7  -0.0469  0.425 -0.1136 0.0198  False\n",
      "     3      8  -0.0108    0.9 -0.0775  0.056  False\n",
      "     3      9  -0.0073    0.9  -0.074 0.0594  False\n",
      "     4      5   0.0129    0.9 -0.0539 0.0796  False\n",
      "     4      6   0.0176    0.9 -0.0491 0.0844  False\n",
      "     4      7  -0.0344 0.7941 -0.1011 0.0324  False\n",
      "     4      8   0.0018    0.9  -0.065 0.0685  False\n",
      "     4      9   0.0052    0.9 -0.0615  0.072  False\n",
      "     5      6   0.0048    0.9  -0.062 0.0715  False\n",
      "     5      7  -0.0472 0.4147 -0.1139 0.0195  False\n",
      "     5      8  -0.0111    0.9 -0.0778 0.0556  False\n",
      "     5      9  -0.0076    0.9 -0.0744 0.0591  False\n",
      "     6      7   -0.052 0.2752 -0.1187 0.0147  False\n",
      "     6      8  -0.0159    0.9 -0.0826 0.0509  False\n",
      "     6      9  -0.0124    0.9 -0.0791 0.0543  False\n",
      "     7      8   0.0361 0.7429 -0.0306 0.1028  False\n",
      "     7      9   0.0396 0.6422 -0.0271 0.1063  False\n",
      "     8      9   0.0035    0.9 -0.0633 0.0702  False\n",
      "---------------------------------------------------\n",
      "0.5513651371002197 1.531286960468851e-19\n",
      "Multiple Comparison of Means - Tukey HSD, FWER=0.05\n",
      "===================================================\n",
      "group1 group2 meandiff p-adj   lower  upper  reject\n",
      "---------------------------------------------------\n",
      "     0      1   0.0082    0.9 -0.0631 0.0795  False\n",
      "     0      2   0.0219    0.9 -0.0494 0.0932  False\n",
      "     0      3   0.0229    0.9 -0.0484 0.0942  False\n",
      "     0      4   0.0068    0.9 -0.0645 0.0781  False\n",
      "     0      5   0.0248    0.9 -0.0465 0.0961  False\n",
      "     0      6   0.0303    0.9  -0.041 0.1016  False\n",
      "     0      7  -0.0265    0.9 -0.0978 0.0448  False\n",
      "     0      8   0.0121    0.9 -0.0592 0.0834  False\n",
      "     0      9   0.0155    0.9 -0.0558 0.0868  False\n",
      "     1      2   0.0136    0.9 -0.0577 0.0849  False\n",
      "     1      3   0.0147    0.9 -0.0566  0.086  False\n",
      "     1      4  -0.0014    0.9 -0.0727 0.0699  False\n",
      "     1      5   0.0165    0.9 -0.0548 0.0878  False\n",
      "     1      6   0.0221    0.9 -0.0492 0.0934  False\n",
      "     1      7  -0.0347 0.8479  -0.106 0.0366  False\n",
      "     1      8   0.0039    0.9 -0.0674 0.0752  False\n",
      "     1      9   0.0073    0.9  -0.064 0.0786  False\n",
      "     2      3    0.001    0.9 -0.0703 0.0723  False\n",
      "     2      4  -0.0151    0.9 -0.0864 0.0562  False\n",
      "     2      5   0.0029    0.9 -0.0684 0.0742  False\n",
      "     2      6   0.0084    0.9 -0.0629 0.0797  False\n",
      "     2      7  -0.0484 0.4761 -0.1197 0.0229  False\n",
      "     2      8  -0.0098    0.9 -0.0811 0.0615  False\n",
      "     2      9  -0.0063    0.9 -0.0777  0.065  False\n",
      "     3      4  -0.0161    0.9 -0.0874 0.0552  False\n",
      "     3      5   0.0019    0.9 -0.0694 0.0732  False\n",
      "     3      6   0.0074    0.9 -0.0639 0.0787  False\n",
      "     3      7  -0.0494 0.4455 -0.1207 0.0219  False\n",
      "     3      8  -0.0108    0.9 -0.0821 0.0605  False\n",
      "     3      9  -0.0074    0.9 -0.0787 0.0639  False\n",
      "     4      5    0.018    0.9 -0.0533 0.0893  False\n",
      "     4      6   0.0235    0.9 -0.0478 0.0948  False\n",
      "     4      7  -0.0333 0.8867 -0.1046  0.038  False\n",
      "     4      8   0.0053    0.9  -0.066 0.0766  False\n",
      "     4      9   0.0087    0.9 -0.0626   0.08  False\n",
      "     5      6   0.0055    0.9 -0.0658 0.0768  False\n",
      "     5      7  -0.0513   0.39 -0.1226   0.02  False\n",
      "     5      8  -0.0127    0.9  -0.084 0.0586  False\n",
      "     5      9  -0.0093    0.9 -0.0806 0.0621  False\n",
      "     6      7  -0.0568 0.2456 -0.1281 0.0145  False\n",
      "     6      8  -0.0182    0.9 -0.0895 0.0531  False\n",
      "     6      9  -0.0148    0.9 -0.0861 0.0565  False\n",
      "     7      8   0.0386 0.7422 -0.0327 0.1099  False\n",
      "     7      9    0.042 0.6496 -0.0293 0.1133  False\n",
      "     8      9   0.0034    0.9 -0.0679 0.0747  False\n",
      "---------------------------------------------------\n",
      "0.5510478019714355 1.5091808170923128e-19\n"
     ]
    }
   ],
   "source": [
    "for m in methods:\n",
    "    d = pd.DataFrame([np.mean(np.vstack(m[i]),axis=1) for i in range(10)]).T\n",
    "\n",
    "    fvalue, pvalue = stats.f_oneway(d[0],d[1],d[2],d[3],d[4],d[5],d[6],d[7],d[8],d[9])\n",
    "    # print(fvalue,pvalue)\n",
    "\n",
    "    # get ANOVA table as R like output\n",
    "    import statsmodels.api as sm\n",
    "    from statsmodels.formula.api import ols\n",
    "    # reshape the d dataframe suitable for statsmodels package \n",
    "    d_melt = pd.melt(d.reset_index(), id_vars=['index'], value_vars=[0,1,2,3,4,5,6,7,8,9])\n",
    "    # replace column names\n",
    "    d_melt.columns = ['index', 'noise_models', 'value']\n",
    "    # Ordinary Least Squares (OLS) model\n",
    "    model = ols('value ~ C(noise_models)', data=d_melt).fit()\n",
    "    anova_table = sm.stats.anova_lm(model, typ=2)\n",
    "    anova_table\n",
    "\n",
    "    # load packages\n",
    "    from statsmodels.stats.multicomp import pairwise_tukeyhsd\n",
    "    # perform multiple pairwise comparison (Tukey HSD)\n",
    "    m_comp = pairwise_tukeyhsd(endog=d_melt['value'], groups=d_melt['noise_models'], alpha=0.05)\n",
    "    print(m_comp)\n",
    "    \n",
    "    w, pvalue = stats.shapiro(model.resid)\n",
    "    print(w, pvalue)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparing variation due to 3 splits x 3 splits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [],
   "source": [
    "std_data_ED=[]\n",
    "std_data_RotF=[]\n",
    "std_data_RandF=[]\n",
    "std_data_DTW1=[]\n",
    "for j in range(10):\n",
    "    data_ED=[]\n",
    "    data_RotF=[]\n",
    "    data_RandF=[]\n",
    "    data_DTW1=[]\n",
    "    for i in range(15):\n",
    "        data_ED.append(pd.read_csv('Results/noise'+str(j)+'/duo'+str(i)+'/1NNED.csv').iloc[:,1].std())\n",
    "        data_RotF.append(pd.read_csv('Results/noise'+str(j)+'/duo'+str(i)+'/RotF50.csv').iloc[:,1].std())\n",
    "        data_RandF.append(pd.read_csv('Results/noise'+str(j)+'/duo'+str(i)+'/RandF.csv').iloc[:,1].std())\n",
    "        data_DTW1.append(pd.read_csv('Results/noise'+str(j)+'/duo'+str(i)+'/1NNDTW1.csv').iloc[:,1:].values.std())\n",
    "    std_data_ED.append(data_ED)\n",
    "    std_data_RotF.append(data_RotF)\n",
    "    std_data_RandF.append(data_RandF)\n",
    "    std_data_DTW1.append(data_DTW1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.37\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>Average</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1NN ED</th>\n",
       "      <td>2.37</td>\n",
       "      <td>1.93</td>\n",
       "      <td>2.21</td>\n",
       "      <td>2.02</td>\n",
       "      <td>1.56</td>\n",
       "      <td>0.71</td>\n",
       "      <td>1.41</td>\n",
       "      <td>1.31</td>\n",
       "      <td>1.15</td>\n",
       "      <td>1.03</td>\n",
       "      <td>2.32</td>\n",
       "      <td>0.93</td>\n",
       "      <td>1.51</td>\n",
       "      <td>1.08</td>\n",
       "      <td>0.82</td>\n",
       "      <td>1.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1NN DTW w=1</th>\n",
       "      <td>2.37</td>\n",
       "      <td>2.21</td>\n",
       "      <td>2.28</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.23</td>\n",
       "      <td>1.01</td>\n",
       "      <td>1.31</td>\n",
       "      <td>1.03</td>\n",
       "      <td>1.02</td>\n",
       "      <td>1.65</td>\n",
       "      <td>2.33</td>\n",
       "      <td>1.14</td>\n",
       "      <td>1.90</td>\n",
       "      <td>1.55</td>\n",
       "      <td>1.38</td>\n",
       "      <td>1.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RotF</th>\n",
       "      <td>1.74</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandF</th>\n",
       "      <td>1.64</td>\n",
       "      <td>1.03</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.32</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.27</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                1     2     3     4     5     6     7     8     9    10    11  \\\n",
       "1NN ED       2.37  1.93  2.21  2.02  1.56  0.71  1.41  1.31  1.15  1.03  2.32   \n",
       "1NN DTW w=1  2.37  2.21  2.28  2.00  2.23  1.01  1.31  1.03  1.02  1.65  2.33   \n",
       "RotF         1.74  1.00  0.95  0.50  0.38  0.38  0.30  0.28  0.30  0.52  0.53   \n",
       "RandF        1.64  1.03  0.81  0.53  0.40  0.33  0.29  0.33  0.32  0.50  0.59   \n",
       "\n",
       "               12    13    14    15  Average  \n",
       "1NN ED       0.93  1.51  1.08  0.82     1.49  \n",
       "1NN DTW w=1  1.14  1.90  1.55  1.38     1.69  \n",
       "RotF         0.38  0.49  0.31  0.25     0.55  \n",
       "RandF        0.38  0.53  0.27  0.24     0.55  "
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "std_methods = [std_data_ED, std_data_DTW1, std_data_RotF,std_data_RandF]\n",
    "std_table = pd.DataFrame(np.vstack([np.round(100*np.mean(np.array(m), axis=0),2) for m in std_methods]))\n",
    "std_table.index = ['1NN ED','1NN DTW w=1','RotF','RandF']\n",
    "std_table.columns = range(1,16)\n",
    "print(max(std_table.max()))\n",
    "std_table['Average'] = np.round(std_table.mean(axis=1),2)\n",
    "# std_table['max(\\u03C3)'] = [np.round(100*max(np.array(m).ravel()),2) for m in std_methods]\n",
    "std_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04110891703028544"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max(np.array(std_data_ED).ravel())\n",
    "#     std_data_RotF.append(data_RotF)\n",
    "#     std_data_RandF.append(data_RandF)\n",
    "#     std_data_DTW1.append(data_DTW1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrrrrrrrrrrrrrr}\n",
      "\\toprule\n",
      "{} &     1 &     2 &     3 &     4 &     5 &     6 &     7 &     8 &     9 &    10 &    11 &    12 &    13 &    14 &    15 &  Average \\\\\n",
      "\\midrule\n",
      "1NN ED      &  2.37 &  1.93 &  2.21 &  2.02 &  1.56 &  0.71 &  1.41 &  1.31 &  1.15 &  1.03 &  2.32 &  0.93 &  1.51 &  1.08 &  0.82 &     1.49 \\\\\n",
      "1NN DTW w=1 &  2.37 &  2.21 &  2.28 &  2.00 &  2.23 &  1.01 &  1.31 &  1.03 &  1.02 &  1.65 &  2.33 &  1.14 &  1.90 &  1.55 &  1.38 &     1.69 \\\\\n",
      "RotF        &  1.74 &  1.00 &  0.95 &  0.50 &  0.38 &  0.38 &  0.30 &  0.28 &  0.30 &  0.52 &  0.53 &  0.38 &  0.49 &  0.31 &  0.25 &     0.55 \\\\\n",
      "RandF       &  1.64 &  1.03 &  0.81 &  0.53 &  0.40 &  0.33 &  0.29 &  0.33 &  0.32 &  0.50 &  0.59 &  0.38 &  0.53 &  0.27 &  0.24 &     0.55 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(std_table.to_latex())"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}