TakeshiHMM-ROBOT-NoiseODOM-Dim_reduction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "IPython.notebook.set_autosave_interval(0)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Autosave disabled\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "from sklearn.cluster import  MiniBatchKMeans\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from joblib import dump, load\n",
    "\n",
    "%autosave 0 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "class HMM (object):\n",
    "             def __init__(self,A,B,PI):\n",
    "                 self.A=A\n",
    "                 self.B=B\n",
    "                 self.PI=PI   \n",
    "def viterbi(obs,Modelo1,PI):\n",
    "    \n",
    "    delta=np.zeros((len(obs)+1,len(Modelo1.A)))\n",
    "    phi=np.zeros((len(obs)+1,len(A)))+666\n",
    "    path =np.zeros(len(obs)+1)\n",
    "    T=len(obs)\n",
    "    Modelo1.PI = PI\n",
    "    delta[0,:]= Modelo1.PI * Modelo1.B[:,obs[0]]\n",
    "    phi[0,:]=666\n",
    "    for t in range(len(obs)):\n",
    "        for j in range(delta.shape[1]):\n",
    "\n",
    "            delta [t+1,j]=np.max(delta[t] * A[:,j]) * B[j,obs[t]]\n",
    "            phi[t+1,j]= np.argmax(delta[t] * A[:,j])\n",
    "    path[T]=int(np.argmax(delta[T,:]))\n",
    "    for i in np.arange(T-1,0,-1):\n",
    "        #print (i,phi[i+1,int(path[i+1])])\n",
    "        path[i]=phi[i+1,int(path[i+1])]\n",
    "    return(path)\n",
    "def cuantizar_xy(xy, cc):\n",
    "    xycuant=cc\n",
    "    out=np.power(xycuant-xy,2).sum(axis=1).argmin()\n",
    "    return out\n",
    "    \n",
    "def path_to_xy(path,ccxy):\n",
    "    estimated= pd.DataFrame(path.astype(int).T)\n",
    "    estimated.columns=[['Path_vit']]\n",
    "    estimated['xcuant'] = estimated['Path_vit'].apply(lambda x: ccxy[x,0])\n",
    "    estimated['ycuant'] = estimated['Path_vit'].apply(lambda x: ccxy[x,1])                                \n",
    "    return (estimated)\n",
    "def quantized(xyth,ccxyth):\n",
    "    xythcuant=np.argmin(np.linalg.norm(xyth-ccxyth,axis=1))\n",
    "    x,y=ccxyth[xythcuant,:2]\n",
    "    return ((x,y),(xythcuant))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "def Markov_A_2_grafo(A,ccxyth):\n",
    "    dists=np.zeros(A.shape)\n",
    "    for i in range(A.shape[0]):\n",
    "        for j in range (A.shape[1]):\n",
    "            if A[i,j]!=0 :\n",
    "                dists[i,j]= np.linalg.norm(ccxyth[i]-ccxyth[j])    \n",
    "    \n",
    "    \n",
    "    con = np.where(dists==0,np.inf,dists)\n",
    "    graphe2=grafo(ccxyth,con)\n",
    "    return graphe2\n",
    "\n",
    "\n",
    "class node(object):\n",
    "    def __init__(self,x,y):\n",
    "        self.x=x\n",
    "        self.y=y\n",
    "        \n",
    "        \n",
    "class grafo (object):\n",
    "             def __init__(self,nodos,conec):\n",
    "                 self.nodos=nodos\n",
    "                 self.conec=conec        \n",
    "\n",
    "def dijkstra(nodoinicial,nodofinal,graphe):\n",
    "    \n",
    "\n",
    "    numnodos= len(graphe.nodos)\n",
    "    con = graphe.conec\n",
    "    D= np.ones(numnodos)*np.inf\n",
    "    Prv= np.ones(numnodos)*np.inf\n",
    "    V= np.zeros(numnodos)\n",
    "    a = nodoinicial\n",
    "    D[a]=0\n",
    "    Prv[a]=0\n",
    "    Prv[np.where(con[a]!=np.inf)]=a\n",
    "    V[a]=1\n",
    "    Dacc=D[a]\n",
    "    ########\n",
    "    D=np.minimum(D,con[a]+D[a])\n",
    "    cont=0\n",
    "    sucess=False\n",
    "    while(sucess==False):\n",
    "        a = np.argmin(D+np.where (V==1,np.inf, V))\n",
    "        Dacc=D[a]\n",
    "        Prv[np.where(D>(con[a]+Dacc) )]=a\n",
    "        V[a]=1\n",
    "        D=np.minimum(D,con[a]+Dacc)\n",
    "        if (a== nodofinal):\n",
    "            print(\"RUTA CALCULADA \")\n",
    "            sucess=True\n",
    "    rutainv=[]\n",
    "    rutainv.append(nodofinal)\n",
    "    while(rutainv[-1]!=nodoinicial):\n",
    "        prv=Prv[int(rutainv[-1])]\n",
    "        rutainv.append(prv)\n",
    "\n",
    "    ruta=[]\n",
    "    for n in reversed(rutainv):\n",
    "        ruta.append(n)\n",
    "    return(ruta)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "ccxyth,ccvk=np.load('ccxyth.npy'),np.load('ccvk.npy')\n",
    "A,B,PI = np.load('A.npy'),np.load('B.npy'),np.load('PI.npy')\n",
    "Modelo= HMM(A,B,PI)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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",
       "      <th>...</th>\n",
       "      <th>953</th>\n",
       "      <th>954</th>\n",
       "      <th>955</th>\n",
       "      <th>956</th>\n",
       "      <th>957</th>\n",
       "      <th>958</th>\n",
       "      <th>959</th>\n",
       "      <th>960</th>\n",
       "      <th>961</th>\n",
       "      <th>962</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>264.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.743439</td>\n",
       "      <td>1.744909</td>\n",
       "      <td>1.754330</td>\n",
       "      <td>1.757133</td>\n",
       "      <td>1.748436</td>\n",
       "      <td>1.757432</td>\n",
       "      <td>1.765754</td>\n",
       "      <td>1.765348</td>\n",
       "      <td>1.780917</td>\n",
       "      <td>1.792958</td>\n",
       "      <td>...</td>\n",
       "      <td>2.154606</td>\n",
       "      <td>2.104496</td>\n",
       "      <td>2.087000</td>\n",
       "      <td>2.073973</td>\n",
       "      <td>2.067409</td>\n",
       "      <td>2.052140</td>\n",
       "      <td>2.047485</td>\n",
       "      <td>2.034754</td>\n",
       "      <td>2.019341</td>\n",
       "      <td>2.020114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.017095</td>\n",
       "      <td>1.016295</td>\n",
       "      <td>1.016774</td>\n",
       "      <td>1.020715</td>\n",
       "      <td>1.024684</td>\n",
       "      <td>1.021947</td>\n",
       "      <td>1.020083</td>\n",
       "      <td>1.017688</td>\n",
       "      <td>1.012653</td>\n",
       "      <td>1.000869</td>\n",
       "      <td>...</td>\n",
       "      <td>0.712857</td>\n",
       "      <td>0.791964</td>\n",
       "      <td>0.821966</td>\n",
       "      <td>0.824994</td>\n",
       "      <td>0.824693</td>\n",
       "      <td>0.848076</td>\n",
       "      <td>0.862156</td>\n",
       "      <td>0.854418</td>\n",
       "      <td>0.860263</td>\n",
       "      <td>0.855498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.194000</td>\n",
       "      <td>0.340000</td>\n",
       "      <td>0.535000</td>\n",
       "      <td>0.526000</td>\n",
       "      <td>0.415000</td>\n",
       "      <td>0.415000</td>\n",
       "      <td>0.541000</td>\n",
       "      <td>0.541000</td>\n",
       "      <td>0.553000</td>\n",
       "      <td>0.555000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.641000</td>\n",
       "      <td>0.626000</td>\n",
       "      <td>0.598000</td>\n",
       "      <td>0.590000</td>\n",
       "      <td>0.595000</td>\n",
       "      <td>0.547000</td>\n",
       "      <td>0.538000</td>\n",
       "      <td>0.541000</td>\n",
       "      <td>0.534000</td>\n",
       "      <td>0.519000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.606750</td>\n",
       "      <td>0.602750</td>\n",
       "      <td>0.603000</td>\n",
       "      <td>0.604750</td>\n",
       "      <td>0.605000</td>\n",
       "      <td>0.607750</td>\n",
       "      <td>0.619000</td>\n",
       "      <td>0.633500</td>\n",
       "      <td>0.650000</td>\n",
       "      <td>0.660000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.574750</td>\n",
       "      <td>1.576750</td>\n",
       "      <td>1.587250</td>\n",
       "      <td>1.588000</td>\n",
       "      <td>1.581750</td>\n",
       "      <td>1.584500</td>\n",
       "      <td>1.577250</td>\n",
       "      <td>1.578750</td>\n",
       "      <td>1.577750</td>\n",
       "      <td>1.581500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.076000</td>\n",
       "      <td>2.090500</td>\n",
       "      <td>2.073500</td>\n",
       "      <td>2.088500</td>\n",
       "      <td>2.110500</td>\n",
       "      <td>2.100000</td>\n",
       "      <td>2.107500</td>\n",
       "      <td>2.104000</td>\n",
       "      <td>2.120000</td>\n",
       "      <td>2.117500</td>\n",
       "      <td>...</td>\n",
       "      <td>2.124500</td>\n",
       "      <td>2.114000</td>\n",
       "      <td>2.103500</td>\n",
       "      <td>2.106500</td>\n",
       "      <td>2.067500</td>\n",
       "      <td>2.086500</td>\n",
       "      <td>2.065000</td>\n",
       "      <td>2.052500</td>\n",
       "      <td>2.057500</td>\n",
       "      <td>2.052500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.752000</td>\n",
       "      <td>2.768250</td>\n",
       "      <td>2.866500</td>\n",
       "      <td>2.807000</td>\n",
       "      <td>2.845000</td>\n",
       "      <td>2.851250</td>\n",
       "      <td>2.863250</td>\n",
       "      <td>2.852500</td>\n",
       "      <td>2.843500</td>\n",
       "      <td>2.837750</td>\n",
       "      <td>...</td>\n",
       "      <td>2.946250</td>\n",
       "      <td>2.968500</td>\n",
       "      <td>2.989250</td>\n",
       "      <td>2.945000</td>\n",
       "      <td>2.939500</td>\n",
       "      <td>2.939250</td>\n",
       "      <td>2.964000</td>\n",
       "      <td>2.926500</td>\n",
       "      <td>2.902000</td>\n",
       "      <td>2.875000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 963 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              0           1           2           3           4           5    \\\n",
       "count  264.000000  264.000000  264.000000  264.000000  264.000000  264.000000   \n",
       "mean     1.743439    1.744909    1.754330    1.757133    1.748436    1.757432   \n",
       "std      1.017095    1.016295    1.016774    1.020715    1.024684    1.021947   \n",
       "min      0.194000    0.340000    0.535000    0.526000    0.415000    0.415000   \n",
       "25%      0.606750    0.602750    0.603000    0.604750    0.605000    0.607750   \n",
       "50%      2.076000    2.090500    2.073500    2.088500    2.110500    2.100000   \n",
       "75%      2.752000    2.768250    2.866500    2.807000    2.845000    2.851250   \n",
       "max      3.000000    3.000000    3.000000    3.000000    3.000000    3.000000   \n",
       "\n",
       "              6           7           8           9    ...         953  \\\n",
       "count  264.000000  264.000000  264.000000  264.000000  ...  264.000000   \n",
       "mean     1.765754    1.765348    1.780917    1.792958  ...    2.154606   \n",
       "std      1.020083    1.017688    1.012653    1.000869  ...    0.712857   \n",
       "min      0.541000    0.541000    0.553000    0.555000  ...    0.641000   \n",
       "25%      0.619000    0.633500    0.650000    0.660000  ...    1.574750   \n",
       "50%      2.107500    2.104000    2.120000    2.117500  ...    2.124500   \n",
       "75%      2.863250    2.852500    2.843500    2.837750  ...    2.946250   \n",
       "max      3.000000    3.000000    3.000000    3.000000  ...    3.000000   \n",
       "\n",
       "              954         955         956         957         958         959  \\\n",
       "count  264.000000  264.000000  264.000000  264.000000  264.000000  264.000000   \n",
       "mean     2.104496    2.087000    2.073973    2.067409    2.052140    2.047485   \n",
       "std      0.791964    0.821966    0.824994    0.824693    0.848076    0.862156   \n",
       "min      0.626000    0.598000    0.590000    0.595000    0.547000    0.538000   \n",
       "25%      1.576750    1.587250    1.588000    1.581750    1.584500    1.577250   \n",
       "50%      2.114000    2.103500    2.106500    2.067500    2.086500    2.065000   \n",
       "75%      2.968500    2.989250    2.945000    2.939500    2.939250    2.964000   \n",
       "max      3.000000    3.000000    3.000000    3.000000    3.000000    3.000000   \n",
       "\n",
       "              960         961         962  \n",
       "count  264.000000  264.000000  264.000000  \n",
       "mean     2.034754    2.019341    2.020114  \n",
       "std      0.854418    0.860263    0.855498  \n",
       "min      0.541000    0.534000    0.519000  \n",
       "25%      1.578750    1.577750    1.581500  \n",
       "50%      2.052500    2.057500    2.052500  \n",
       "75%      2.926500    2.902000    2.875000  \n",
       "max      3.000000    3.000000    3.000000  \n",
       "\n",
       "[8 rows x 963 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "data=pd.read_csv('RealROBOT/lecturasconodometria-noisy-a.txt')\n",
    "#data=data.iloc[:40,:]\n",
    "\n",
    "cols=['x','y','theta','y_noise','x_noise','theta_noise']\n",
    "n_reads=data.shape[1]-6\n",
    "etiquetas=[]\n",
    "for i in range (n_reads):\n",
    "    etiquetas.append('R'+str(i))\n",
    "\n",
    "        \n",
    "for lab in cols:\n",
    "    etiquetas.append(lab)\n",
    "\n",
    "\n",
    "data.columns=etiquetas\n",
    "\n",
    "\"\"\"if (only_nice!=True):\n",
    "    print(\"NOT ONLY NICE\")\n",
    "    data2=pd.read_csv('lecturasconodometria.txt')\n",
    "    data2.columns=etiquetas\n",
    "    data3= pd.concat((data,data2))\n",
    "    data=data3\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "\n",
    "\n",
    "####\n",
    "#index = data[data.y<-1].index\n",
    "#data=data.drop(index)\n",
    "\n",
    "\n",
    "\n",
    "#####\n",
    "\n",
    "xyth=data[['x','y','theta']]\n",
    "\n",
    "lecs=data.iloc[:,0:n_reads]\n",
    "lecs=pd.DataFrame(np.clip(lecs.values,0,3))\n",
    "data.y_noise=data.y_noise*-1\n",
    "\n",
    "lecs.describe()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(264, 45)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "from sklearn.cluster import FeatureAgglomeration\n",
    "\n",
    "from sklearn.decomposition import PCA, KernelPCA\n",
    "\n",
    "F_agg= FeatureAgglomeration(n_clusters=45).fit(lecs)\n",
    "pca =PCA(n_components=45).fit(lecs)\n",
    "\n",
    "\n",
    "\n",
    "lecs_pca=pd.DataFrame(pca.transform(lecs))\n",
    "\n",
    "lecs_fag=pd.DataFrame(F_agg.transform(lecs))\n",
    "lecs_fag.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "rot_ang=3.1416/8\n",
    "R= np.ones((2,2))\n",
    "R[0,0]= np.cos(rot_ang)\n",
    "R[0,1]= -np.sin(rot_ang)\n",
    "R[1,0]= np.sin(rot_ang)\n",
    "R[1,1]= np.cos(rot_ang)\n",
    "rotated=[]\n",
    "for xy in data[['x_noise','y_noise']].values:\n",
    "    xy_rotated=np.matmul(R,xy)\n",
    "    rotated.append(xy_rotated)\n",
    "rotated=np.asarray(rotated)\n",
    "data['x_rot']=rotated[:,0]\n",
    "data['y_rot']=rotated[:,1]\n",
    "tx=data.x.iloc[0]-data.x_rot.iloc[0]\n",
    "ty=data.y.iloc[0]-data.y_rot.iloc[0]\n",
    "tx,ty\n",
    "data.x_rot=data.x_rot+tx\n",
    "data.y_rot=data.y_rot+ty\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "delta_odom= np.linalg.norm(data[['x','y']].values-data[['x_rot','y_rot']].values,axis=1)\n",
    "data['delta_odom'] =delta_odom"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    264.000000\n",
       "mean       0.944422\n",
       "std        0.801651\n",
       "min        0.000000\n",
       "25%        0.094187\n",
       "50%        0.748308\n",
       "75%        1.865282\n",
       "max        1.960854\n",
       "Name: delta_odom, dtype: float64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.delta_odom.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_1,y_1,th_1= 0,0,0\n",
    "delta_mov=[]\n",
    "delta_th=[]\n",
    "for x,y,th in zip (data.x,data.y,data.theta):\n",
    "    \n",
    "    \n",
    "    delta_mov.append(np.array((x,y))-np.array((x_1,y_1)  ))\n",
    "    if (th-th_1)> 6:\n",
    "        delta_th.append(th-th_1-(2*3.1416))\n",
    "    if (th-th_1)<- 6:\n",
    "        delta_th.append(th-th_1+(2*3.1416))\n",
    "    \n",
    "    if ((th-th_1)>- 6 and (th-th_1)< 6  ):\n",
    "        delta_th.append(th-th_1)\n",
    "    x_1,y_1,th_1=x,y,th\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_1,y_1,th_1= 0,0,0\n",
    "delta_mov_rot=[]\n",
    "delta_th_rot=[]\n",
    "for x,y,th in zip (data.x_rot,data.y_rot,data.theta_noise):\n",
    "    \n",
    "    \n",
    "    delta_mov_rot.append(np.array((x,y))-np.array((x_1,y_1)  ))\n",
    "    if (th-th_1)> 6:\n",
    "        delta_th_rot.append(th-th_1-(2*3.1416))\n",
    "    if (th-th_1)<- 6:\n",
    "        delta_th_rot.append(th-th_1+(2*3.1416))\n",
    "    \n",
    "    if ((th-th_1)>- 6 and (th-th_1)< 6  ):\n",
    "        delta_th_rot.append(th-th_1)\n",
    "    x_1,y_1,th_1=x,y,th\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "delta_th[0],delta_th_rot[0]=0,0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ThetaNoise ,   N(\\mu ,\\sigma) $$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$XY Noise ,   N(\\mu ,\\sigma) $$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "xyth_od=data[['x_rot','y_rot','theta_noise']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "if True:\n",
    "    \n",
    "    K_corpus=128\n",
    "    N=16\n",
    "    \n",
    "    kmedias=MiniBatchKMeans(init='k-means++', n_clusters=K_corpus, batch_size=1000,\n",
    "                      n_init=10, max_no_improvement=10, verbose=0)\n",
    "    kmedias.fit(lecs_fag)  ########CHOOSE LECS OR DIM REDUCED LECS\n",
    "    ccvk=kmedias.cluster_centers_\n",
    "    mbk = MiniBatchKMeans(init='k-means++', n_clusters=N, batch_size=1000,\n",
    "                      n_init=10, max_no_improvement=10, verbose=0)\n",
    "    mbk.fit(xyth)\n",
    "    ccxyth=mbk.cluster_centers_\n",
    "aus=[]\n",
    "aus_odom=[]\n",
    "auvks=[]\n",
    "for i,lec,j in zip(xyth.values, lecs_fag.values,xyth_od.values):########CHOOSE LECS OR DIM REDUCED LECS\n",
    "    _, au= quantized(i , ccxyth)\n",
    "    _, au_od= quantized(i , ccxyth)\n",
    "    aus.append(au)\n",
    "    aus_odom.append(au_od)\n",
    "\n",
    "    auvk= np.power(lec.T-ccvk,2).sum(axis=1,keepdims=True).argmax()\n",
    "    auvks.append(auvk)\n",
    "\n",
    "data['Y']=aus\n",
    "data['Y_odom']=aus_odom\n",
    "data['Vk']=auvks\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import AffinityPropagation\n",
    "\n",
    "#data['Vk_aff'] = AffinityPropagation().fit_predict(lecs_fag)\n",
    "data['Vk_aff']=data['Vk']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f8d57a6bfd0>"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "fig = plt.figure(figsize=(10,10))\n",
    "fig.suptitle('Quantized Read (HMM Observation)', fontsize=16)\n",
    "\n",
    "ax1 = fig.add_subplot(111)\n",
    "\n",
    "ax1.scatter(x=data.x_rot,y=data.y_rot,marker='+' ,c=data.Vk, label='Odom' )\n",
    "ax1.scatter(x=data.x,y=data.y ,marker='*', c=data.Vk,label='Real')\n",
    "ax1.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f8d57a98390>"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAERFJREFUeJzt3W2MXOV5h/HrLs6Lw7YYAh25Nuq6CkpF2QbCihKlqnYhaQmgwAcUEVmprbraD01a2rpKTaN+iNRKRC15KWpTWSHFrdwslEBtQZKGOmyjSg0JTlIMOBRDTINlvEn90ixFSba5+2GO08V4PGdmZ3Zmnlw/abVzzjxn5r7P2fn7zDMvjsxEkjT6fmLQBUiSesNAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBVi1Ure2fnnn5/j4+Ndbfviiy9y9tln97agASihjxJ6gDL6sIfh0c8+9u7d+53MvKDduBUN9PHxcR599NGutp2bm2Nqaqq3BQ1ACX2U0AOU0Yc9DI9+9hERz9UZ55SLJBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVYkU/KSpJgzS+7cG+3fbWiUU2t7j9g7dd17f7XcozdEkqhIEuSYVwymWI9fPp4Zms1NNDSb3lGbokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKkStr8+NiIPAd4H/BRYzczIizgPuBsaBg8C7MvNYf8qUJLXTyRn6dGZempmT1fI2YE9mXgTsqZYlSQOynCmXG4Ad1eUdwI3LL0eS1K26gZ7A5yNib0TMVOsamXm4uvwC0Oh5dZKk2iIz2w+KWJeZhyLip4GHgN8GdmfmmiVjjmXmuafZdgaYAWg0GpfPzs52Vej80RMceamrTZdlYt05Pb29hYUFxsbGao3dd+hET++7rnY9d9LDMCuhD3voTD8fU43VtMyo5ebI9PT03iXT3S3VelE0Mw9Vv+cj4n7gCuBIRKzNzMMRsRaYb7HtdmA7wOTkZE5NTdVs4eXu2LmL2/et/H+BenDjVE9vb25ujrr7YPOg/k/RNj130sMwK6EPe+hMPx9TWycWW2ZUr3OklbZTLhFxdkT85MnLwK8CjwO7gU3VsE3Arn4VKUlqr84pbwO4PyJOjv/7zPxcRHwFuCcitgDPAe/qX5mSpHbaBnpmPgu86TTr/wu4uh9FSZI65ydFJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVIjagR4RZ0XE1yLigWp5Q0Q8EhEHIuLuiHh1/8qUJLXTyRn6LcD+JcsfAj6SmW8AjgFbelmYJKkztQI9ItYD1wGfqJYDuAq4txqyA7ixHwVKkuqpe4b+UeD9wA+r5dcDxzNzsVp+HljX49okSR2IzDzzgIjrgWsz87ciYgr4A2Az8KVquoWIuBD4bGZecprtZ4AZgEajcfns7GxXhc4fPcGRl7radFkm1p3T09tbWFhgbGys1th9h0709L7ratdzJz0MsxL6sIfO9PMx1VhNy4xabo5MT0/vzczJduNW1bittwLvjIhrgdcCPwV8DFgTEauqs/T1wKHTbZyZ24HtAJOTkzk1NVWvg1PcsXMXt++rU25vHdw41dPbm5ubo+4+2LztwZ7ed13teu6kh2FWQh/20Jl+Pqa2Tiy2zKhe50grbadcMvPWzFyfmePAzcAXMnMj8DBwUzVsE7Crb1VKktpazvvQ/xD4/Yg4QHNO/c7elCRJ6kZHcxiZOQfMVZefBa7ofUmSpG74SVFJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVom2gR8RrI+LLEfHvEfFERHywWr8hIh6JiAMRcXdEvLr/5UqSWqlzhv494KrMfBNwKXBNRFwJfAj4SGa+ATgGbOlfmZKkdtoGejYtVIuvqn4SuAq4t1q/A7ixLxVKkmqpNYceEWdFxNeBeeAh4BngeGYuVkOeB9b1p0RJUh2RmfUHR6wB7gf+GLirmm4hIi4EPpuZl5xmmxlgBqDRaFw+OzvbVaHzR09w5KWuNl2WiXXn9PT2FhYWGBsbqzV236ETPb3vutr13EkPw6yEPuyhM/18TDVW0zKjlpsj09PTezNzst24VZ3caGYej4iHgbcAayJiVXWWvh441GKb7cB2gMnJyZyamurkLn/kjp27uH1fR+X2xMGNUz29vbm5Oerug83bHuzpfdfVrudOehhmJfRhD53p52Nq68Riy4zqdY60UuddLhdUZ+ZExGrg7cB+4GHgpmrYJmBXv4qUJLVX55R3LbAjIs6i+Q/APZn5QEQ8CcxGxJ8AXwPu7GOdkqQ22gZ6Zj4GXHaa9c8CV/SjqGEy3uOnaFsnFgc2lSKpbH5SVJIKYaBLUiEMdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqhIEuSYVoG+gRcWFEPBwRT0bEExFxS7X+vIh4KCKern6f2/9yJUmt1DlDXwS2ZubFwJXAeyPiYmAbsCczLwL2VMuSpAFpG+iZeTgzv1pd/i6wH1gH3ADsqIbtAG7sV5GSpPY6mkOPiHHgMuARoJGZh6urXgAaPa1MktSRyMx6AyPGgH8B/jQz74uI45m5Zsn1xzLzFfPoETEDzAA0Go3LZ2dnuyp0/ugJjrzU1aZDpbGaoe9jYt05Z7x+YWGBsbGxFaqmf0rowx46s+/Qib7d9pke2+0eU+1MT0/vzczJduNW1bmxiHgV8GlgZ2beV60+EhFrM/NwRKwF5k+3bWZuB7YDTE5O5tTUVJ27fIU7du7i9n21yh1qWycWh76Pgxunznj93Nwc3R7HYVJCH/bQmc3bHuzbbZ/psd3uMdUrdd7lEsCdwP7M/PCSq3YDm6rLm4BdvS9PklRXnVPFtwLvAfZFxNerdX8E3AbcExFbgOeAd/WnRElSHW0DPTP/FYgWV1/d23IkSd3yk6KSVAgDXZIKYaBLUiEMdEkqhIEuSYUw0CWpEMP9kUX92Bnv4yf5TrV1YvFHnxw8eNt1K3a/Ur94hi5JhTDQJakQBrokFcI5dGnAlvO6wdLXATrl6wbl8QxdkgphoEtSIZxy0Su0mwJYztN8Sf3jGbokFcJAl6RCGOiSVAjn0KUfUyv5NQun8i2T/eEZuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCtA30iPhkRMxHxONL1p0XEQ9FxNPV73P7W6YkqZ06Z+h3Adecsm4bsCczLwL2VMuSpAFqG+iZ+UXg6CmrbwB2VJd3ADf2uC5JUoe6nUNvZObh6vILQKNH9UiSuhSZ2X5QxDjwQGZeUi0fz8w1S64/lpmnnUePiBlgBqDRaFw+OzvbVaHzR09w5KWuNh0qjdWMfB8l9AAv72Ni3TkDq2PfoRNdbzuqx2Lp/l5YWGBsbGxF7nc5+7qdMx2L5f59TU9P783MyXbjuv0+9CMRsTYzD0fEWmC+1cDM3A5sB5icnMypqamu7vCOnbu4fd/of3371onFke+jhB7g5X0c3Dg1sDqW8/+zjuqxWLq/5+bm6DYXOtXP/wv3TMdipf6+up1y2Q1sqi5vAnb1phxJUrfqvG3xU8C/AW+MiOcjYgtwG/D2iHgaeFu1LEkaoLbP1TLz3S2uurrHtUiSlsFPikpSIQx0SSqEgS5JhTDQJakQBrokFcJAl6RCGOiSVAgDXZIKYaBLUiEMdEkqxOh9TZvUB+N9/BY+vdLS/b11YrGv34L448QzdEkqhIEuSYUw0CWpEAa6JBXCQJekQhjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQVwkCXpEIY6JJUCANdkgphoEtSIQx0SSrEsgI9Iq6JiKci4kBEbOtVUZKkznUd6BFxFvCXwDuAi4F3R8TFvSpMktSZ5ZyhXwEcyMxnM/P7wCxwQ2/KkiR1ajmBvg741pLl56t1kqQBiMzsbsOIm4BrMvM3q+X3AL+Ume87ZdwMMFMtvhF4qstazwe+0+W2w6SEPkroAcrowx6GRz/7+NnMvKDdoFXLuINDwIVLltdX614mM7cD25dxPwBExKOZObnc2xm0EvoooQcoow97GB7D0Mdyply+AlwUERsi4tXAzcDu3pQlSepU12fombkYEe8D/gk4C/hkZj7Rs8okSR1ZzpQLmfkZ4DM9qqWdZU/bDIkS+iihByijD3sYHgPvo+sXRSVJw8WP/ktSIUYi0EfxKwYi4sKIeDginoyIJyLilmr9eRHxUEQ8Xf0+d9C1thMRZ0XE1yLigWp5Q0Q8Uh2Pu6sXxYdaRKyJiHsj4hsRsT8i3jJqxyIifq/6W3o8Ij4VEa8dhWMREZ+MiPmIeHzJutPu+2j6i6qfxyLizYOr/OVa9PFn1d/UYxFxf0SsWXLdrVUfT0XEr61EjUMf6CP8FQOLwNbMvBi4EnhvVfc2YE9mXgTsqZaH3S3A/iXLHwI+kplvAI4BWwZSVWc+BnwuM38eeBPNfkbmWETEOuB3gMnMvITmGxFuZjSOxV3ANaesa7Xv3wFcVP3MAB9foRrruItX9vEQcElm/iLwH8CtANVj/WbgF6pt/qrKsr4a+kBnRL9iIDMPZ+ZXq8vfpRkg62jWvqMatgO4cTAV1hMR64HrgE9UywFcBdxbDRmFHs4BfgW4EyAzv5+ZxxmxY0HzTQyrI2IV8DrgMCNwLDLzi8DRU1a32vc3AH+bTV8C1kTE2pWp9MxO10dmfj4zF6vFL9H8PA40+5jNzO9l5jeBAzSzrK9GIdBH/isGImIcuAx4BGhk5uHqqheAxoDKquujwPuBH1bLrweOL/kjHoXjsQH4NvA31dTRJyLibEboWGTmIeDPgf+kGeQngL2M3rE4qdW+H+XH+28An60uD6SPUQj0kRYRY8Cngd/NzP9eel0232I0tG8ziojrgfnM3DvoWpZpFfBm4OOZeRnwIqdMr4zAsTiX5lnfBuBngLN55dP/kTTs+76OiPgAzWnWnYOsYxQCvdZXDAyjiHgVzTDfmZn3VauPnHwKWf2eH1R9NbwVeGdEHKQ51XUVzbnoNdXTfhiN4/E88HxmPlIt30sz4EfpWLwN+GZmfjszfwDcR/P4jNqxOKnVvh+5x3tEbAauBzbm/78PfCB9jEKgj+RXDFRzzXcC+zPzw0uu2g1sqi5vAnatdG11Zeatmbk+M8dp7vcvZOZG4GHgpmrYUPcAkJkvAN+KiDdWq64GnmSEjgXNqZYrI+J11d/WyR5G6lgs0Wrf7wZ+vXq3y5XAiSVTM0MnIq6hOSX5zsz8nyVX7QZujojXRMQGmi/yfrnvBWXm0P8A19J8BfkZ4AODrqdmzb9M82nkY8DXq59rac5B7wGeBv4ZOG/QtdbsZwp4oLr8c9Uf5wHgH4DXDLq+GvVfCjxaHY9/BM4dtWMBfBD4BvA48HfAa0bhWACfojnv/wOaz5a2tNr3QNB8V9szwD6a7+oZeA9n6OMAzbnyk4/xv14y/gNVH08B71iJGv2kqCQVYhSmXCRJNRjoklQIA12SCmGgS1IhDHRJKoSBLkmFMNAlqRAGuiQV4v8AfCOgFzK7kAUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.Vk_aff.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f8d57a7ea90>"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "data.Vk.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f8d579c0550>"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "fig = plt.figure(figsize=(10,10))\n",
    "ax1 = fig.add_subplot(111)\n",
    "fig.suptitle('Quantized POSE (HMM State)', fontsize=16)\n",
    "\n",
    "ax1.scatter(x=data.x_rot,y=data.y_rot,marker='*' ,c=data.Y_odom  ,label='ODOM')\n",
    "ax1.scatter(x=data.x,y=data.y , c=data.Y,label='REAL')\n",
    "ax1.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "save=False\n",
    "est,obs,obs2= data.Y.values,data.Vk.values,data.Vk_aff.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL NOT SAVED\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(16, 124)"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "####BAUM WELCH APPROX###############???!?!?!\n",
    "\n",
    "\n",
    "TRANS=np.zeros((N,N))\n",
    "for i in range (N):\n",
    "    fromTO=[]\n",
    "\n",
    "\n",
    "    indices=np.asarray(np.where(est==i)).ravel()\n",
    "   \n",
    "\n",
    "    for indice in indices:\n",
    "\n",
    "        if (int(indice)==len(est)-1):\n",
    "            \n",
    "            indice=indice-1\n",
    "\n",
    "        fromTO.append(est[indice+1])\n",
    "    fromTO.append(0)\n",
    "    fromTO.append(N-1)\n",
    "    aux=np.asarray(np.bincount(fromTO    ))\n",
    "    aux[0]=aux[0]-1\n",
    "    aux[N-1]=aux[N-1]-1\n",
    "    \n",
    "    TRANS[i,:]=aux\n",
    "TRANS=TRANS/TRANS.sum(axis=1)\n",
    "#####################################3\n",
    "EMIS= np.zeros((N,K_corpus))\n",
    "for Vk in range (K_corpus):\n",
    "\n",
    "    indices=np.asarray(np.where(obs==Vk)).ravel()\n",
    "    estconvk= est[indices]\n",
    "\n",
    "    numestenvk=np.bincount(estconvk)\n",
    "    if (N-len(numestenvk)>0):\n",
    "        numestenvk=np.append(numestenvk, np.zeros(N-len(numestenvk)))\n",
    "    EMIS[:,Vk]=(numestenvk/np.bincount(est))\n",
    "\n",
    "A,B=TRANS,EMIS\n",
    "PI=np.ones(N)/N\n",
    "Modelo1=HMM(A,B,PI)\n",
    "\n",
    "\n",
    "if save:\n",
    "\n",
    "    #np.linalg.norm(Modelo1.A-Modelo2.A)\n",
    "    np.save('A.npy',Modelo1.A)\n",
    "    np.save('B.npy',Modelo1.B)\n",
    "    np.save('PI.npy',Modelo1.PI)\n",
    "    print(\"MODEL SAVED\")\n",
    "else:\n",
    "    print(\"MODEL NOT SAVED\")\n",
    "B.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MODEL NOT SAVED\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(16, 124)"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "####BAUM WELCH APPROX###############???!?!?!\n",
    "\n",
    "\n",
    "K_corpus2= obs2.max()+1\n",
    "TRANS=np.zeros((N,N))\n",
    "for i in range (N):\n",
    "    fromTO=[]\n",
    "\n",
    "\n",
    "    indices=np.asarray(np.where(est==i)).ravel()\n",
    "   \n",
    "\n",
    "    for indice in indices:\n",
    "\n",
    "        if (int(indice)==len(est)-1):\n",
    "            \n",
    "            indice=indice-1\n",
    "\n",
    "        fromTO.append(est[indice+1])\n",
    "    fromTO.append(0)\n",
    "    fromTO.append(N-1)\n",
    "    aux=np.asarray(np.bincount(fromTO    ))\n",
    "    aux[0]=aux[0]-1\n",
    "    aux[N-1]=aux[N-1]-1\n",
    "    \n",
    "    TRANS[i,:]=aux\n",
    "TRANS=TRANS/TRANS.sum(axis=1)\n",
    "#####################################3\n",
    "EMIS= np.zeros((N,K_corpus2))\n",
    "for Vk in range (K_corpus2):\n",
    "\n",
    "    indices=np.asarray(np.where(obs2==Vk)).ravel()\n",
    "    estconvk= est[indices]\n",
    "\n",
    "    numestenvk=np.bincount(estconvk)\n",
    "    if (N-len(numestenvk)>0):\n",
    "        numestenvk=np.append(numestenvk, np.zeros(N-len(numestenvk)))\n",
    "    EMIS[:,Vk]=(numestenvk/np.bincount(est))\n",
    "\n",
    "\n",
    "PI=np.ones(N)/N\n",
    "Modelo2=HMM(TRANS,EMIS,PI)\n",
    "\n",
    "\n",
    "if save:\n",
    "\n",
    "    #np.linalg.norm(Modelo1.A-Modelo2.A)\n",
    "    np.save('A.npy',Modelo2.A)\n",
    "    np.save('B.npy',Modelo2.B)\n",
    "    np.save('PI.npy',Modelo2.PI)\n",
    "    print(\"MODEL SAVED\")\n",
    "else:\n",
    "    print(\"MODEL NOT SAVED\")\n",
    "EMIS.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "def accuracy(yhat,ytrue):\n",
    "    weights=[.5,.2,.1,.05,.05]\n",
    "    if len (yhat)>=5:\n",
    "        yhat=yhat[-5:]\n",
    "        ytrue=ytrue[-5:]\n",
    "        accuracy=np.dot((yhat==ytrue),weights)\n",
    "        return accuracy\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "216\n",
      "0.3 0.3\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "buf_vit=10\n",
    "indice= int(np.random.rand()*(len(obs)-2*buf_vit))\n",
    "for i in range(1):\n",
    "    \n",
    "   \n",
    "    obs_windowed= obs[indice:indice+buf_vit]\n",
    "    obs2_windowed= obs2[indice:indice+buf_vit]\n",
    "    est_windowed= est[indice:indice+buf_vit]\n",
    "    odomreal= data[['x','y']].iloc[indice:indice+buf_vit]\n",
    "    \n",
    "\n",
    "    print(indice)\n",
    "    #PATH from Model 1 KMEANS CC's obs\n",
    "    path= viterbi(obs_windowed,Modelo1,Modelo1.PI)\n",
    "    \n",
    "    ##PATH FROM MODEL 2 WICH NEEDS AFF PORP QUANT OBS\n",
    "    path2= viterbi(obs2_windowed,Modelo2,Modelo2.PI)\n",
    "    yhat=np.zeros(len(path)-1)\n",
    "    yhat2=np.zeros(len(path2)-1)\n",
    "    ytrue=np.zeros(len(path)-1)\n",
    "    for i in range(len(path)-1):\n",
    "        yhat[i]=int(path[i])\n",
    "        yhat2[i]=int(path2[i])\n",
    "        ytrue[i]=int(est_windowed[i])\n",
    "    print(accuracy(ytrue,yhat), accuracy(ytrue,yhat2))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f8d5792eb00>"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cc=ccxyth\n",
    "paths= viterbi(obs_windowed,Modelo1,Modelo1.PI)  \n",
    "path=paths[int(-len(paths)/2):]\n",
    "path_bigdelay = paths[:int(-len(paths)/2)]\n",
    "cords= path_to_xy(path,cc)\n",
    "cords_bigdelay = path_to_xy(path_bigdelay,cc)\n",
    "cords1=path_to_xy(est_windowed.T[int(-len(paths)/2):],cc)\n",
    "cords1delay=path_to_xy(est_windowed.T[:int(-len(paths)/2)],cc)\n",
    "\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(10,10))\n",
    "fig.suptitle('Test Run Buf Vit=10 Model kmeans 128', fontsize=16)\n",
    "\n",
    "ax1 = fig.add_subplot(111)\n",
    "ax1.scatter(x= data.x, y=data.y,marker='+',c=data.Y,alpha=.00313)\n",
    "ax1.scatter(x=cords.xcuant, y=cords.ycuant, marker='*',c='r',s=52 )\n",
    "ax1.scatter(x=cords_bigdelay.xcuant, y=cords_bigdelay.ycuant, marker='*',c='m',s=52 )\n",
    "ax1.scatter(x=cords1.xcuant, y=cords1.ycuant, marker='o',c='b' ,alpha=.5)\n",
    "ax1.scatter(x=cords1delay.xcuant, y=cords1delay.ycuant, marker='o',c='c' ,alpha=.1)\n",
    "\n",
    "\n",
    "\n",
    "ax1.scatter(x= odomreal.x[int(-len(paths)/2):], y=odomreal.y[int(-len(paths)/2):],marker='.',s=100,c='m')\n",
    "ax1.scatter(x= odomreal.x[:int(-len(paths)/2)], y=odomreal.y[:int(-len(paths)/2)],marker='.',s=100,c='y',alpha=.25)\n",
    "cords3=path_to_xy(np.arange(0,Modelo1.A.shape[0]),cc)\n",
    "ax1.scatter(x= cords3.xcuant, y=cords3.ycuant ,marker='.',s=40,c='g')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.3\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cc=ccxyth\n",
    "paths= viterbi(obs2_windowed,Modelo2,Modelo2.PI)  \n",
    "path=paths[int(-len(paths)/2):]\n",
    "path_bigdelay = paths[:int(-len(paths)/2)]\n",
    "cords= path_to_xy(path,cc)\n",
    "cords_bigdelay = path_to_xy(path_bigdelay,cc)\n",
    "cords1=path_to_xy(est_windowed.T[int(-len(paths)/2):],cc)\n",
    "cords1delay=path_to_xy(est_windowed.T[:int(-len(paths)/2)],cc)\n",
    "\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(10,10))\n",
    "fig.suptitle('Test Run Buf Vit=10 Aff Prop (PCA) ', fontsize=16)\n",
    "\n",
    "ax1 = fig.add_subplot(111)\n",
    "ax1.scatter(x= data.x, y=data.y,marker='+',c=data.Y,alpha=.00313)\n",
    "ax1.scatter(x=cords.xcuant, y=cords.ycuant, marker='*',c='r',s=52 )\n",
    "ax1.scatter(x=cords_bigdelay.xcuant, y=cords_bigdelay.ycuant, marker='*',c='m',s=52 )\n",
    "ax1.scatter(x=cords1.xcuant, y=cords1.ycuant, marker='o',c='b' ,alpha=.5)\n",
    "ax1.scatter(x=cords1delay.xcuant, y=cords1delay.ycuant, marker='o',c='c' ,alpha=.1)\n",
    "\n",
    "\n",
    "\n",
    "ax1.scatter(x= odomreal.x[int(-len(paths)/2):], y=odomreal.y[int(-len(paths)/2):],marker='.',s=13,c='y')\n",
    "ax1.scatter(x= odomreal.x[:int(-len(paths)/2)], y=odomreal.y[:int(-len(paths)/2)],marker='.',s=13,c='m')\n",
    "cords3=path_to_xy(np.arange(0,Modelo1.A.shape[0]),cc)\n",
    "ax1.scatter(x= cords3.xcuant, y=cords3.ycuant ,marker='.',s=40,c='g')\n",
    "\n",
    "print(accuracy(ytrue,yhat2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(264, 15)"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.feature_selection import SelectKBest\n",
    "lecs_kbest = pd.DataFrame(SelectKBest( k=15).fit_transform(lecs, data.Y))\n",
    "lecs_kbest.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 5., 5., 3., 3., 3., 6., 6., 1., 1., 1.])"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((16, 128), (16, 39))"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Modelo1.B.shape,Modelo2.B.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "38"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.max(obs2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>R0</th>\n",
       "      <th>R1</th>\n",
       "      <th>R2</th>\n",
       "      <th>R3</th>\n",
       "      <th>R4</th>\n",
       "      <th>R5</th>\n",
       "      <th>R6</th>\n",
       "      <th>R7</th>\n",
       "      <th>R8</th>\n",
       "      <th>R9</th>\n",
       "      <th>...</th>\n",
       "      <th>y_noise</th>\n",
       "      <th>x_noise</th>\n",
       "      <th>theta_noise</th>\n",
       "      <th>x_rot</th>\n",
       "      <th>y_rot</th>\n",
       "      <th>delta_odom</th>\n",
       "      <th>Y</th>\n",
       "      <th>Y_odom</th>\n",
       "      <th>Vk</th>\n",
       "      <th>Vk_aff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.593</td>\n",
       "      <td>0.588</td>\n",
       "      <td>0.719</td>\n",
       "      <td>0.719</td>\n",
       "      <td>0.623</td>\n",
       "      <td>0.623</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.636</td>\n",
       "      <td>0.653</td>\n",
       "      <td>0.672</td>\n",
       "      <td>...</td>\n",
       "      <td>1.035714</td>\n",
       "      <td>1.350020</td>\n",
       "      <td>1.750313</td>\n",
       "      <td>6.641049</td>\n",
       "      <td>3.250945</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>121</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6.672</td>\n",
       "      <td>6.758</td>\n",
       "      <td>6.963</td>\n",
       "      <td>7.152</td>\n",
       "      <td>7.359</td>\n",
       "      <td>7.423</td>\n",
       "      <td>7.593</td>\n",
       "      <td>9.744</td>\n",
       "      <td>9.743</td>\n",
       "      <td>9.756</td>\n",
       "      <td>...</td>\n",
       "      <td>1.032806</td>\n",
       "      <td>1.344383</td>\n",
       "      <td>1.539686</td>\n",
       "      <td>6.636953</td>\n",
       "      <td>3.246101</td>\n",
       "      <td>0.004644</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>81</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.300</td>\n",
       "      <td>3.361</td>\n",
       "      <td>3.344</td>\n",
       "      <td>3.373</td>\n",
       "      <td>3.373</td>\n",
       "      <td>6.802</td>\n",
       "      <td>6.803</td>\n",
       "      <td>6.789</td>\n",
       "      <td>6.772</td>\n",
       "      <td>6.763</td>\n",
       "      <td>...</td>\n",
       "      <td>1.028448</td>\n",
       "      <td>1.340108</td>\n",
       "      <td>1.325055</td>\n",
       "      <td>6.634671</td>\n",
       "      <td>3.240439</td>\n",
       "      <td>0.016999</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.150</td>\n",
       "      <td>2.151</td>\n",
       "      <td>2.130</td>\n",
       "      <td>2.136</td>\n",
       "      <td>2.144</td>\n",
       "      <td>2.147</td>\n",
       "      <td>2.135</td>\n",
       "      <td>2.129</td>\n",
       "      <td>2.126</td>\n",
       "      <td>2.121</td>\n",
       "      <td>...</td>\n",
       "      <td>1.029537</td>\n",
       "      <td>1.338056</td>\n",
       "      <td>1.112569</td>\n",
       "      <td>6.632358</td>\n",
       "      <td>3.240660</td>\n",
       "      <td>0.016266</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>93</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.332</td>\n",
       "      <td>2.327</td>\n",
       "      <td>2.323</td>\n",
       "      <td>2.309</td>\n",
       "      <td>2.290</td>\n",
       "      <td>2.209</td>\n",
       "      <td>2.216</td>\n",
       "      <td>2.244</td>\n",
       "      <td>2.240</td>\n",
       "      <td>2.259</td>\n",
       "      <td>...</td>\n",
       "      <td>1.022495</td>\n",
       "      <td>1.335604</td>\n",
       "      <td>0.914470</td>\n",
       "      <td>6.632788</td>\n",
       "      <td>3.233215</td>\n",
       "      <td>0.019450</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>93</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2.665</td>\n",
       "      <td>2.662</td>\n",
       "      <td>2.650</td>\n",
       "      <td>2.657</td>\n",
       "      <td>2.644</td>\n",
       "      <td>2.649</td>\n",
       "      <td>2.652</td>\n",
       "      <td>2.671</td>\n",
       "      <td>2.668</td>\n",
       "      <td>2.662</td>\n",
       "      <td>...</td>\n",
       "      <td>1.023280</td>\n",
       "      <td>1.333553</td>\n",
       "      <td>0.674101</td>\n",
       "      <td>6.630592</td>\n",
       "      <td>3.233155</td>\n",
       "      <td>0.029568</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>81</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2.184</td>\n",
       "      <td>2.184</td>\n",
       "      <td>2.193</td>\n",
       "      <td>2.185</td>\n",
       "      <td>2.205</td>\n",
       "      <td>2.208</td>\n",
       "      <td>2.216</td>\n",
       "      <td>2.214</td>\n",
       "      <td>2.220</td>\n",
       "      <td>2.225</td>\n",
       "      <td>...</td>\n",
       "      <td>1.021649</td>\n",
       "      <td>1.329625</td>\n",
       "      <td>0.477713</td>\n",
       "      <td>6.627588</td>\n",
       "      <td>3.230146</td>\n",
       "      <td>0.033285</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>93</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3.280</td>\n",
       "      <td>3.288</td>\n",
       "      <td>3.295</td>\n",
       "      <td>3.295</td>\n",
       "      <td>3.306</td>\n",
       "      <td>3.314</td>\n",
       "      <td>3.319</td>\n",
       "      <td>3.328</td>\n",
       "      <td>3.342</td>\n",
       "      <td>3.342</td>\n",
       "      <td>...</td>\n",
       "      <td>1.021730</td>\n",
       "      <td>1.326667</td>\n",
       "      <td>0.274111</td>\n",
       "      <td>6.624824</td>\n",
       "      <td>3.229088</td>\n",
       "      <td>0.037496</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>93</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2.689</td>\n",
       "      <td>2.685</td>\n",
       "      <td>2.692</td>\n",
       "      <td>2.695</td>\n",
       "      <td>2.714</td>\n",
       "      <td>2.740</td>\n",
       "      <td>2.861</td>\n",
       "      <td>3.315</td>\n",
       "      <td>3.315</td>\n",
       "      <td>2.589</td>\n",
       "      <td>...</td>\n",
       "      <td>1.022636</td>\n",
       "      <td>1.325774</td>\n",
       "      <td>0.048981</td>\n",
       "      <td>6.623653</td>\n",
       "      <td>3.229584</td>\n",
       "      <td>0.037064</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>81</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2.838</td>\n",
       "      <td>2.876</td>\n",
       "      <td>2.871</td>\n",
       "      <td>2.839</td>\n",
       "      <td>2.845</td>\n",
       "      <td>3.188</td>\n",
       "      <td>3.257</td>\n",
       "      <td>3.257</td>\n",
       "      <td>3.223</td>\n",
       "      <td>2.748</td>\n",
       "      <td>...</td>\n",
       "      <td>1.019186</td>\n",
       "      <td>1.332234</td>\n",
       "      <td>-0.153986</td>\n",
       "      <td>6.630941</td>\n",
       "      <td>3.228868</td>\n",
       "      <td>0.030765</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>93</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.612</td>\n",
       "      <td>0.594</td>\n",
       "      <td>0.594</td>\n",
       "      <td>0.590</td>\n",
       "      <td>0.651</td>\n",
       "      <td>3.119</td>\n",
       "      <td>3.086</td>\n",
       "      <td>2.890</td>\n",
       "      <td>2.805</td>\n",
       "      <td>2.782</td>\n",
       "      <td>...</td>\n",
       "      <td>0.989331</td>\n",
       "      <td>1.324199</td>\n",
       "      <td>-0.374625</td>\n",
       "      <td>6.634942</td>\n",
       "      <td>3.198211</td>\n",
       "      <td>0.066363</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.565</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.628</td>\n",
       "      <td>0.614</td>\n",
       "      <td>0.603</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.627</td>\n",
       "      <td>0.647</td>\n",
       "      <td>...</td>\n",
       "      <td>0.957698</td>\n",
       "      <td>1.304092</td>\n",
       "      <td>-0.590330</td>\n",
       "      <td>6.628472</td>\n",
       "      <td>3.161292</td>\n",
       "      <td>0.112827</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>107</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.607</td>\n",
       "      <td>0.576</td>\n",
       "      <td>0.605</td>\n",
       "      <td>0.605</td>\n",
       "      <td>0.606</td>\n",
       "      <td>0.617</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.645</td>\n",
       "      <td>0.664</td>\n",
       "      <td>0.669</td>\n",
       "      <td>...</td>\n",
       "      <td>0.920562</td>\n",
       "      <td>1.269147</td>\n",
       "      <td>-0.805200</td>\n",
       "      <td>6.610398</td>\n",
       "      <td>3.113610</td>\n",
       "      <td>0.179511</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>110</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>3.686</td>\n",
       "      <td>3.633</td>\n",
       "      <td>3.624</td>\n",
       "      <td>3.590</td>\n",
       "      <td>3.582</td>\n",
       "      <td>3.548</td>\n",
       "      <td>3.543</td>\n",
       "      <td>3.526</td>\n",
       "      <td>3.515</td>\n",
       "      <td>3.487</td>\n",
       "      <td>...</td>\n",
       "      <td>0.877029</td>\n",
       "      <td>1.213013</td>\n",
       "      <td>-1.014855</td>\n",
       "      <td>6.575196</td>\n",
       "      <td>3.051908</td>\n",
       "      <td>0.262562</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>54</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2.166</td>\n",
       "      <td>2.180</td>\n",
       "      <td>2.222</td>\n",
       "      <td>2.645</td>\n",
       "      <td>2.722</td>\n",
       "      <td>2.756</td>\n",
       "      <td>2.810</td>\n",
       "      <td>2.811</td>\n",
       "      <td>2.831</td>\n",
       "      <td>2.837</td>\n",
       "      <td>...</td>\n",
       "      <td>0.836900</td>\n",
       "      <td>1.131391</td>\n",
       "      <td>-1.224845</td>\n",
       "      <td>6.515144</td>\n",
       "      <td>2.983599</td>\n",
       "      <td>0.372451</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1.994</td>\n",
       "      <td>1.833</td>\n",
       "      <td>1.818</td>\n",
       "      <td>1.812</td>\n",
       "      <td>1.801</td>\n",
       "      <td>1.804</td>\n",
       "      <td>1.799</td>\n",
       "      <td>1.828</td>\n",
       "      <td>1.907</td>\n",
       "      <td>2.059</td>\n",
       "      <td>...</td>\n",
       "      <td>0.809691</td>\n",
       "      <td>1.017909</td>\n",
       "      <td>-1.428236</td>\n",
       "      <td>6.420713</td>\n",
       "      <td>2.915033</td>\n",
       "      <td>0.507970</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>107</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2.454</td>\n",
       "      <td>2.444</td>\n",
       "      <td>2.445</td>\n",
       "      <td>2.429</td>\n",
       "      <td>2.323</td>\n",
       "      <td>2.241</td>\n",
       "      <td>2.232</td>\n",
       "      <td>2.225</td>\n",
       "      <td>2.215</td>\n",
       "      <td>2.224</td>\n",
       "      <td>...</td>\n",
       "      <td>0.802515</td>\n",
       "      <td>0.852086</td>\n",
       "      <td>-1.637951</td>\n",
       "      <td>6.270259</td>\n",
       "      <td>2.844946</td>\n",
       "      <td>0.691832</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.631</td>\n",
       "      <td>0.631</td>\n",
       "      <td>0.600</td>\n",
       "      <td>0.585</td>\n",
       "      <td>0.592</td>\n",
       "      <td>0.601</td>\n",
       "      <td>0.652</td>\n",
       "      <td>0.692</td>\n",
       "      <td>0.675</td>\n",
       "      <td>0.593</td>\n",
       "      <td>...</td>\n",
       "      <td>0.823411</td>\n",
       "      <td>0.458046</td>\n",
       "      <td>-1.503236</td>\n",
       "      <td>5.898217</td>\n",
       "      <td>2.713458</td>\n",
       "      <td>1.150678</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>114</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.635</td>\n",
       "      <td>0.637</td>\n",
       "      <td>0.629</td>\n",
       "      <td>0.567</td>\n",
       "      <td>0.566</td>\n",
       "      <td>0.566</td>\n",
       "      <td>0.565</td>\n",
       "      <td>0.651</td>\n",
       "      <td>0.651</td>\n",
       "      <td>0.696</td>\n",
       "      <td>...</td>\n",
       "      <td>0.798090</td>\n",
       "      <td>0.098752</td>\n",
       "      <td>-1.606075</td>\n",
       "      <td>5.575963</td>\n",
       "      <td>2.552569</td>\n",
       "      <td>1.595312</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>55</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2.216</td>\n",
       "      <td>2.429</td>\n",
       "      <td>3.291</td>\n",
       "      <td>3.293</td>\n",
       "      <td>3.293</td>\n",
       "      <td>3.313</td>\n",
       "      <td>3.322</td>\n",
       "      <td>3.315</td>\n",
       "      <td>3.315</td>\n",
       "      <td>3.347</td>\n",
       "      <td>...</td>\n",
       "      <td>0.823886</td>\n",
       "      <td>-0.202639</td>\n",
       "      <td>-1.570507</td>\n",
       "      <td>5.287642</td>\n",
       "      <td>2.461063</td>\n",
       "      <td>1.954704</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>56</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>3.653</td>\n",
       "      <td>3.251</td>\n",
       "      <td>3.150</td>\n",
       "      <td>3.184</td>\n",
       "      <td>3.558</td>\n",
       "      <td>3.728</td>\n",
       "      <td>3.708</td>\n",
       "      <td>3.694</td>\n",
       "      <td>3.708</td>\n",
       "      <td>3.734</td>\n",
       "      <td>...</td>\n",
       "      <td>0.825146</td>\n",
       "      <td>-0.205170</td>\n",
       "      <td>-1.360718</td>\n",
       "      <td>5.284822</td>\n",
       "      <td>2.461259</td>\n",
       "      <td>1.956511</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>28</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2.967</td>\n",
       "      <td>2.939</td>\n",
       "      <td>2.922</td>\n",
       "      <td>2.893</td>\n",
       "      <td>2.888</td>\n",
       "      <td>2.859</td>\n",
       "      <td>2.838</td>\n",
       "      <td>2.816</td>\n",
       "      <td>2.792</td>\n",
       "      <td>2.762</td>\n",
       "      <td>...</td>\n",
       "      <td>0.821483</td>\n",
       "      <td>-0.205713</td>\n",
       "      <td>-1.153256</td>\n",
       "      <td>5.285722</td>\n",
       "      <td>2.457667</td>\n",
       "      <td>1.960754</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>73</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2.107</td>\n",
       "      <td>2.105</td>\n",
       "      <td>2.102</td>\n",
       "      <td>2.100</td>\n",
       "      <td>2.102</td>\n",
       "      <td>2.102</td>\n",
       "      <td>2.097</td>\n",
       "      <td>2.065</td>\n",
       "      <td>1.960</td>\n",
       "      <td>1.936</td>\n",
       "      <td>...</td>\n",
       "      <td>0.825251</td>\n",
       "      <td>-0.205470</td>\n",
       "      <td>-0.931603</td>\n",
       "      <td>5.284504</td>\n",
       "      <td>2.461241</td>\n",
       "      <td>1.960854</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>38</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.644</td>\n",
       "      <td>0.597</td>\n",
       "      <td>0.566</td>\n",
       "      <td>0.582</td>\n",
       "      <td>0.604</td>\n",
       "      <td>0.648</td>\n",
       "      <td>0.603</td>\n",
       "      <td>0.618</td>\n",
       "      <td>0.771</td>\n",
       "      <td>0.774</td>\n",
       "      <td>...</td>\n",
       "      <td>0.829827</td>\n",
       "      <td>-0.204245</td>\n",
       "      <td>-0.718095</td>\n",
       "      <td>5.283885</td>\n",
       "      <td>2.465938</td>\n",
       "      <td>1.957838</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>124</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1.668</td>\n",
       "      <td>1.686</td>\n",
       "      <td>1.724</td>\n",
       "      <td>1.734</td>\n",
       "      <td>1.759</td>\n",
       "      <td>1.779</td>\n",
       "      <td>1.811</td>\n",
       "      <td>1.827</td>\n",
       "      <td>1.859</td>\n",
       "      <td>1.870</td>\n",
       "      <td>...</td>\n",
       "      <td>0.830810</td>\n",
       "      <td>-0.200426</td>\n",
       "      <td>-0.506446</td>\n",
       "      <td>5.287037</td>\n",
       "      <td>2.468307</td>\n",
       "      <td>1.955938</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>11</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0.601</td>\n",
       "      <td>0.606</td>\n",
       "      <td>0.614</td>\n",
       "      <td>0.596</td>\n",
       "      <td>0.597</td>\n",
       "      <td>0.615</td>\n",
       "      <td>0.617</td>\n",
       "      <td>0.675</td>\n",
       "      <td>0.770</td>\n",
       "      <td>2.156</td>\n",
       "      <td>...</td>\n",
       "      <td>0.832573</td>\n",
       "      <td>-0.198906</td>\n",
       "      <td>-0.308557</td>\n",
       "      <td>5.287766</td>\n",
       "      <td>2.470518</td>\n",
       "      <td>1.951903</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>70</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.563</td>\n",
       "      <td>0.572</td>\n",
       "      <td>0.572</td>\n",
       "      <td>0.581</td>\n",
       "      <td>0.573</td>\n",
       "      <td>0.596</td>\n",
       "      <td>0.612</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.617</td>\n",
       "      <td>0.616</td>\n",
       "      <td>...</td>\n",
       "      <td>0.837799</td>\n",
       "      <td>-0.195681</td>\n",
       "      <td>-0.098863</td>\n",
       "      <td>5.288745</td>\n",
       "      <td>2.476580</td>\n",
       "      <td>1.945966</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>88</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>0.637</td>\n",
       "      <td>0.598</td>\n",
       "      <td>0.581</td>\n",
       "      <td>0.567</td>\n",
       "      <td>0.591</td>\n",
       "      <td>0.603</td>\n",
       "      <td>0.609</td>\n",
       "      <td>0.611</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.639</td>\n",
       "      <td>...</td>\n",
       "      <td>0.837069</td>\n",
       "      <td>-0.193857</td>\n",
       "      <td>0.115163</td>\n",
       "      <td>5.290710</td>\n",
       "      <td>2.476603</td>\n",
       "      <td>1.942955</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>38</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>0.609</td>\n",
       "      <td>0.601</td>\n",
       "      <td>0.601</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.614</td>\n",
       "      <td>0.622</td>\n",
       "      <td>0.622</td>\n",
       "      <td>0.636</td>\n",
       "      <td>0.595</td>\n",
       "      <td>0.625</td>\n",
       "      <td>...</td>\n",
       "      <td>0.832844</td>\n",
       "      <td>-0.188633</td>\n",
       "      <td>0.342044</td>\n",
       "      <td>5.297154</td>\n",
       "      <td>2.474700</td>\n",
       "      <td>1.933763</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>85</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0.609</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.603</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.626</td>\n",
       "      <td>0.661</td>\n",
       "      <td>0.650</td>\n",
       "      <td>0.621</td>\n",
       "      <td>0.650</td>\n",
       "      <td>0.690</td>\n",
       "      <td>...</td>\n",
       "      <td>0.817664</td>\n",
       "      <td>-0.172010</td>\n",
       "      <td>0.560894</td>\n",
       "      <td>5.318321</td>\n",
       "      <td>2.467036</td>\n",
       "      <td>1.914236</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>126</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>234</th>\n",
       "      <td>0.635</td>\n",
       "      <td>0.630</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.602</td>\n",
       "      <td>0.599</td>\n",
       "      <td>0.599</td>\n",
       "      <td>0.620</td>\n",
       "      <td>0.627</td>\n",
       "      <td>0.674</td>\n",
       "      <td>0.634</td>\n",
       "      <td>...</td>\n",
       "      <td>0.919114</td>\n",
       "      <td>-0.147631</td>\n",
       "      <td>0.443644</td>\n",
       "      <td>5.302020</td>\n",
       "      <td>2.570093</td>\n",
       "      <td>1.837659</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>126</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>0.543</td>\n",
       "      <td>0.582</td>\n",
       "      <td>0.604</td>\n",
       "      <td>0.626</td>\n",
       "      <td>0.607</td>\n",
       "      <td>0.570</td>\n",
       "      <td>0.555</td>\n",
       "      <td>0.576</td>\n",
       "      <td>0.638</td>\n",
       "      <td>0.666</td>\n",
       "      <td>...</td>\n",
       "      <td>0.893033</td>\n",
       "      <td>-0.127451</td>\n",
       "      <td>0.663867</td>\n",
       "      <td>5.330645</td>\n",
       "      <td>2.553720</td>\n",
       "      <td>1.814434</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>7</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236</th>\n",
       "      <td>0.533</td>\n",
       "      <td>0.536</td>\n",
       "      <td>0.555</td>\n",
       "      <td>0.540</td>\n",
       "      <td>0.549</td>\n",
       "      <td>0.549</td>\n",
       "      <td>0.588</td>\n",
       "      <td>0.634</td>\n",
       "      <td>0.635</td>\n",
       "      <td>0.633</td>\n",
       "      <td>...</td>\n",
       "      <td>0.867843</td>\n",
       "      <td>-0.097606</td>\n",
       "      <td>0.880158</td>\n",
       "      <td>5.367858</td>\n",
       "      <td>2.541869</td>\n",
       "      <td>1.781405</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>107</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>0.552</td>\n",
       "      <td>0.547</td>\n",
       "      <td>0.567</td>\n",
       "      <td>0.557</td>\n",
       "      <td>0.554</td>\n",
       "      <td>0.554</td>\n",
       "      <td>0.575</td>\n",
       "      <td>0.582</td>\n",
       "      <td>0.579</td>\n",
       "      <td>0.577</td>\n",
       "      <td>...</td>\n",
       "      <td>0.838276</td>\n",
       "      <td>-0.057815</td>\n",
       "      <td>1.073008</td>\n",
       "      <td>5.415935</td>\n",
       "      <td>2.529780</td>\n",
       "      <td>1.738179</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>238</th>\n",
       "      <td>0.668</td>\n",
       "      <td>0.639</td>\n",
       "      <td>0.645</td>\n",
       "      <td>0.616</td>\n",
       "      <td>0.616</td>\n",
       "      <td>0.624</td>\n",
       "      <td>0.662</td>\n",
       "      <td>0.664</td>\n",
       "      <td>0.675</td>\n",
       "      <td>0.756</td>\n",
       "      <td>...</td>\n",
       "      <td>0.809922</td>\n",
       "      <td>0.010134</td>\n",
       "      <td>1.301265</td>\n",
       "      <td>5.489562</td>\n",
       "      <td>2.529587</td>\n",
       "      <td>1.667617</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>0.662</td>\n",
       "      <td>0.669</td>\n",
       "      <td>0.677</td>\n",
       "      <td>0.691</td>\n",
       "      <td>0.694</td>\n",
       "      <td>0.698</td>\n",
       "      <td>0.609</td>\n",
       "      <td>0.654</td>\n",
       "      <td>0.707</td>\n",
       "      <td>0.752</td>\n",
       "      <td>...</td>\n",
       "      <td>0.797439</td>\n",
       "      <td>0.107468</td>\n",
       "      <td>1.515579</td>\n",
       "      <td>5.584264</td>\n",
       "      <td>2.555302</td>\n",
       "      <td>1.558568</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>240</th>\n",
       "      <td>5.664</td>\n",
       "      <td>5.689</td>\n",
       "      <td>5.718</td>\n",
       "      <td>5.806</td>\n",
       "      <td>5.827</td>\n",
       "      <td>5.863</td>\n",
       "      <td>5.856</td>\n",
       "      <td>4.437</td>\n",
       "      <td>4.395</td>\n",
       "      <td>4.579</td>\n",
       "      <td>...</td>\n",
       "      <td>0.807330</td>\n",
       "      <td>0.241779</td>\n",
       "      <td>1.718374</td>\n",
       "      <td>5.704566</td>\n",
       "      <td>2.615840</td>\n",
       "      <td>1.398309</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>93</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>5.294</td>\n",
       "      <td>5.374</td>\n",
       "      <td>5.358</td>\n",
       "      <td>5.403</td>\n",
       "      <td>5.370</td>\n",
       "      <td>5.483</td>\n",
       "      <td>5.567</td>\n",
       "      <td>5.613</td>\n",
       "      <td>5.021</td>\n",
       "      <td>4.718</td>\n",
       "      <td>...</td>\n",
       "      <td>0.863128</td>\n",
       "      <td>0.562906</td>\n",
       "      <td>1.670150</td>\n",
       "      <td>5.979896</td>\n",
       "      <td>2.790280</td>\n",
       "      <td>1.022240</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>91</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>4.907</td>\n",
       "      <td>4.719</td>\n",
       "      <td>6.190</td>\n",
       "      <td>6.283</td>\n",
       "      <td>6.345</td>\n",
       "      <td>7.166</td>\n",
       "      <td>7.134</td>\n",
       "      <td>6.903</td>\n",
       "      <td>6.536</td>\n",
       "      <td>6.359</td>\n",
       "      <td>...</td>\n",
       "      <td>0.870681</td>\n",
       "      <td>0.957775</td>\n",
       "      <td>1.615743</td>\n",
       "      <td>6.341816</td>\n",
       "      <td>2.948368</td>\n",
       "      <td>0.558188</td>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "      <td>58</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>3.491</td>\n",
       "      <td>3.499</td>\n",
       "      <td>3.512</td>\n",
       "      <td>3.517</td>\n",
       "      <td>3.526</td>\n",
       "      <td>3.534</td>\n",
       "      <td>3.542</td>\n",
       "      <td>3.542</td>\n",
       "      <td>3.558</td>\n",
       "      <td>3.559</td>\n",
       "      <td>...</td>\n",
       "      <td>0.912911</td>\n",
       "      <td>1.257845</td>\n",
       "      <td>1.831761</td>\n",
       "      <td>6.602884</td>\n",
       "      <td>3.102216</td>\n",
       "      <td>0.236772</td>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "      <td>23</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>244</th>\n",
       "      <td>6.599</td>\n",
       "      <td>6.628</td>\n",
       "      <td>6.630</td>\n",
       "      <td>6.627</td>\n",
       "      <td>6.637</td>\n",
       "      <td>6.694</td>\n",
       "      <td>6.893</td>\n",
       "      <td>6.998</td>\n",
       "      <td>7.324</td>\n",
       "      <td>7.381</td>\n",
       "      <td>...</td>\n",
       "      <td>0.947360</td>\n",
       "      <td>1.405765</td>\n",
       "      <td>1.410013</td>\n",
       "      <td>6.726362</td>\n",
       "      <td>3.190650</td>\n",
       "      <td>0.093264</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>81</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245</th>\n",
       "      <td>3.337</td>\n",
       "      <td>3.352</td>\n",
       "      <td>3.372</td>\n",
       "      <td>3.395</td>\n",
       "      <td>6.787</td>\n",
       "      <td>6.785</td>\n",
       "      <td>6.775</td>\n",
       "      <td>6.755</td>\n",
       "      <td>6.740</td>\n",
       "      <td>6.711</td>\n",
       "      <td>...</td>\n",
       "      <td>0.946513</td>\n",
       "      <td>1.404663</td>\n",
       "      <td>1.208116</td>\n",
       "      <td>6.725667</td>\n",
       "      <td>3.189445</td>\n",
       "      <td>0.094116</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>2.126</td>\n",
       "      <td>2.121</td>\n",
       "      <td>2.123</td>\n",
       "      <td>2.121</td>\n",
       "      <td>2.122</td>\n",
       "      <td>2.118</td>\n",
       "      <td>2.118</td>\n",
       "      <td>2.118</td>\n",
       "      <td>2.120</td>\n",
       "      <td>2.100</td>\n",
       "      <td>...</td>\n",
       "      <td>0.947416</td>\n",
       "      <td>1.406636</td>\n",
       "      <td>0.999418</td>\n",
       "      <td>6.727144</td>\n",
       "      <td>3.191034</td>\n",
       "      <td>0.098167</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>93</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>247</th>\n",
       "      <td>2.216</td>\n",
       "      <td>2.215</td>\n",
       "      <td>2.245</td>\n",
       "      <td>2.265</td>\n",
       "      <td>2.269</td>\n",
       "      <td>2.228</td>\n",
       "      <td>2.234</td>\n",
       "      <td>2.233</td>\n",
       "      <td>2.225</td>\n",
       "      <td>2.235</td>\n",
       "      <td>...</td>\n",
       "      <td>0.941252</td>\n",
       "      <td>1.406220</td>\n",
       "      <td>0.817502</td>\n",
       "      <td>6.729119</td>\n",
       "      <td>3.185180</td>\n",
       "      <td>0.103568</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>93</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>2.665</td>\n",
       "      <td>2.647</td>\n",
       "      <td>2.653</td>\n",
       "      <td>2.655</td>\n",
       "      <td>2.657</td>\n",
       "      <td>2.661</td>\n",
       "      <td>2.665</td>\n",
       "      <td>2.649</td>\n",
       "      <td>2.630</td>\n",
       "      <td>2.539</td>\n",
       "      <td>...</td>\n",
       "      <td>0.942643</td>\n",
       "      <td>1.404869</td>\n",
       "      <td>0.542955</td>\n",
       "      <td>6.727339</td>\n",
       "      <td>3.185948</td>\n",
       "      <td>0.094211</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>81</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>249</th>\n",
       "      <td>2.205</td>\n",
       "      <td>2.205</td>\n",
       "      <td>2.204</td>\n",
       "      <td>2.210</td>\n",
       "      <td>2.217</td>\n",
       "      <td>2.219</td>\n",
       "      <td>2.235</td>\n",
       "      <td>2.235</td>\n",
       "      <td>2.239</td>\n",
       "      <td>2.250</td>\n",
       "      <td>...</td>\n",
       "      <td>0.940940</td>\n",
       "      <td>1.402126</td>\n",
       "      <td>0.344362</td>\n",
       "      <td>6.725456</td>\n",
       "      <td>3.183325</td>\n",
       "      <td>0.095920</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>93</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250</th>\n",
       "      <td>3.290</td>\n",
       "      <td>3.290</td>\n",
       "      <td>3.292</td>\n",
       "      <td>3.299</td>\n",
       "      <td>3.303</td>\n",
       "      <td>3.317</td>\n",
       "      <td>3.329</td>\n",
       "      <td>3.336</td>\n",
       "      <td>3.341</td>\n",
       "      <td>3.344</td>\n",
       "      <td>...</td>\n",
       "      <td>0.941150</td>\n",
       "      <td>1.398761</td>\n",
       "      <td>0.136874</td>\n",
       "      <td>6.722267</td>\n",
       "      <td>3.182232</td>\n",
       "      <td>0.090962</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>93</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>251</th>\n",
       "      <td>2.703</td>\n",
       "      <td>2.709</td>\n",
       "      <td>2.712</td>\n",
       "      <td>2.712</td>\n",
       "      <td>2.718</td>\n",
       "      <td>2.720</td>\n",
       "      <td>2.770</td>\n",
       "      <td>2.863</td>\n",
       "      <td>3.344</td>\n",
       "      <td>3.352</td>\n",
       "      <td>...</td>\n",
       "      <td>0.935414</td>\n",
       "      <td>1.399125</td>\n",
       "      <td>-0.060493</td>\n",
       "      <td>6.724799</td>\n",
       "      <td>3.177072</td>\n",
       "      <td>0.096892</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>81</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>2.070</td>\n",
       "      <td>2.105</td>\n",
       "      <td>2.291</td>\n",
       "      <td>2.302</td>\n",
       "      <td>2.306</td>\n",
       "      <td>2.325</td>\n",
       "      <td>2.338</td>\n",
       "      <td>2.393</td>\n",
       "      <td>2.788</td>\n",
       "      <td>2.776</td>\n",
       "      <td>...</td>\n",
       "      <td>0.919356</td>\n",
       "      <td>1.395268</td>\n",
       "      <td>-0.293647</td>\n",
       "      <td>6.727380</td>\n",
       "      <td>3.160759</td>\n",
       "      <td>0.112173</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>93</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>253</th>\n",
       "      <td>0.642</td>\n",
       "      <td>0.675</td>\n",
       "      <td>1.701</td>\n",
       "      <td>3.071</td>\n",
       "      <td>3.109</td>\n",
       "      <td>3.143</td>\n",
       "      <td>3.143</td>\n",
       "      <td>2.894</td>\n",
       "      <td>2.832</td>\n",
       "      <td>2.815</td>\n",
       "      <td>...</td>\n",
       "      <td>0.895171</td>\n",
       "      <td>1.384946</td>\n",
       "      <td>-0.488721</td>\n",
       "      <td>6.727099</td>\n",
       "      <td>3.134466</td>\n",
       "      <td>0.144158</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>254</th>\n",
       "      <td>0.584</td>\n",
       "      <td>0.586</td>\n",
       "      <td>0.599</td>\n",
       "      <td>0.624</td>\n",
       "      <td>0.629</td>\n",
       "      <td>0.618</td>\n",
       "      <td>0.597</td>\n",
       "      <td>0.623</td>\n",
       "      <td>0.669</td>\n",
       "      <td>0.706</td>\n",
       "      <td>...</td>\n",
       "      <td>0.872728</td>\n",
       "      <td>1.368373</td>\n",
       "      <td>-0.670040</td>\n",
       "      <td>6.720376</td>\n",
       "      <td>3.107389</td>\n",
       "      <td>0.184990</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>81</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>255</th>\n",
       "      <td>0.632</td>\n",
       "      <td>0.601</td>\n",
       "      <td>0.601</td>\n",
       "      <td>0.603</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.624</td>\n",
       "      <td>0.663</td>\n",
       "      <td>0.662</td>\n",
       "      <td>0.701</td>\n",
       "      <td>...</td>\n",
       "      <td>0.834508</td>\n",
       "      <td>1.323827</td>\n",
       "      <td>-0.928808</td>\n",
       "      <td>6.693848</td>\n",
       "      <td>3.055031</td>\n",
       "      <td>0.239698</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>110</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256</th>\n",
       "      <td>3.636</td>\n",
       "      <td>3.609</td>\n",
       "      <td>3.582</td>\n",
       "      <td>3.580</td>\n",
       "      <td>3.565</td>\n",
       "      <td>3.547</td>\n",
       "      <td>3.462</td>\n",
       "      <td>3.374</td>\n",
       "      <td>3.267</td>\n",
       "      <td>3.269</td>\n",
       "      <td>...</td>\n",
       "      <td>0.804334</td>\n",
       "      <td>1.274169</td>\n",
       "      <td>-1.124658</td>\n",
       "      <td>6.659517</td>\n",
       "      <td>3.008151</td>\n",
       "      <td>0.300622</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>54</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>257</th>\n",
       "      <td>2.189</td>\n",
       "      <td>2.188</td>\n",
       "      <td>2.199</td>\n",
       "      <td>2.218</td>\n",
       "      <td>2.228</td>\n",
       "      <td>2.236</td>\n",
       "      <td>2.750</td>\n",
       "      <td>2.750</td>\n",
       "      <td>2.795</td>\n",
       "      <td>2.852</td>\n",
       "      <td>...</td>\n",
       "      <td>0.774462</td>\n",
       "      <td>1.187015</td>\n",
       "      <td>-1.348455</td>\n",
       "      <td>6.590428</td>\n",
       "      <td>2.947200</td>\n",
       "      <td>0.391515</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258</th>\n",
       "      <td>2.177</td>\n",
       "      <td>2.180</td>\n",
       "      <td>2.174</td>\n",
       "      <td>2.157</td>\n",
       "      <td>2.139</td>\n",
       "      <td>2.026</td>\n",
       "      <td>1.935</td>\n",
       "      <td>1.860</td>\n",
       "      <td>1.838</td>\n",
       "      <td>1.826</td>\n",
       "      <td>...</td>\n",
       "      <td>0.761763</td>\n",
       "      <td>1.083811</td>\n",
       "      <td>-1.538943</td>\n",
       "      <td>6.499940</td>\n",
       "      <td>2.895973</td>\n",
       "      <td>0.504797</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>107</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>259</th>\n",
       "      <td>2.434</td>\n",
       "      <td>2.428</td>\n",
       "      <td>2.417</td>\n",
       "      <td>2.413</td>\n",
       "      <td>2.447</td>\n",
       "      <td>2.446</td>\n",
       "      <td>2.456</td>\n",
       "      <td>2.453</td>\n",
       "      <td>2.453</td>\n",
       "      <td>2.403</td>\n",
       "      <td>...</td>\n",
       "      <td>0.775990</td>\n",
       "      <td>0.927739</td>\n",
       "      <td>-1.757301</td>\n",
       "      <td>6.350304</td>\n",
       "      <td>2.849391</td>\n",
       "      <td>0.653228</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>260</th>\n",
       "      <td>0.598</td>\n",
       "      <td>0.612</td>\n",
       "      <td>0.592</td>\n",
       "      <td>0.590</td>\n",
       "      <td>0.631</td>\n",
       "      <td>0.582</td>\n",
       "      <td>0.575</td>\n",
       "      <td>0.617</td>\n",
       "      <td>0.628</td>\n",
       "      <td>0.628</td>\n",
       "      <td>...</td>\n",
       "      <td>0.850278</td>\n",
       "      <td>0.569604</td>\n",
       "      <td>-1.653426</td>\n",
       "      <td>5.991002</td>\n",
       "      <td>2.780971</td>\n",
       "      <td>1.009725</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>261</th>\n",
       "      <td>0.643</td>\n",
       "      <td>0.648</td>\n",
       "      <td>0.635</td>\n",
       "      <td>0.635</td>\n",
       "      <td>0.586</td>\n",
       "      <td>0.639</td>\n",
       "      <td>0.665</td>\n",
       "      <td>0.689</td>\n",
       "      <td>0.702</td>\n",
       "      <td>0.709</td>\n",
       "      <td>...</td>\n",
       "      <td>0.867695</td>\n",
       "      <td>0.182198</td>\n",
       "      <td>-1.697039</td>\n",
       "      <td>5.626419</td>\n",
       "      <td>2.648809</td>\n",
       "      <td>1.447297</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>113</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262</th>\n",
       "      <td>2.074</td>\n",
       "      <td>2.080</td>\n",
       "      <td>2.227</td>\n",
       "      <td>2.485</td>\n",
       "      <td>3.312</td>\n",
       "      <td>3.327</td>\n",
       "      <td>3.333</td>\n",
       "      <td>3.328</td>\n",
       "      <td>3.321</td>\n",
       "      <td>3.324</td>\n",
       "      <td>...</td>\n",
       "      <td>0.940873</td>\n",
       "      <td>-0.173421</td>\n",
       "      <td>-1.704811</td>\n",
       "      <td>5.269867</td>\n",
       "      <td>2.580326</td>\n",
       "      <td>1.845325</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>56</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>263</th>\n",
       "      <td>3.738</td>\n",
       "      <td>3.741</td>\n",
       "      <td>3.715</td>\n",
       "      <td>3.188</td>\n",
       "      <td>3.169</td>\n",
       "      <td>3.207</td>\n",
       "      <td>3.503</td>\n",
       "      <td>3.727</td>\n",
       "      <td>3.737</td>\n",
       "      <td>3.735</td>\n",
       "      <td>...</td>\n",
       "      <td>0.942737</td>\n",
       "      <td>-0.176349</td>\n",
       "      <td>-1.486557</td>\n",
       "      <td>5.266448</td>\n",
       "      <td>2.580928</td>\n",
       "      <td>1.850523</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>56</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>264 rows × 976 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        R0     R1     R2     R3     R4     R5     R6     R7     R8     R9  \\\n",
       "0    0.593  0.588  0.719  0.719  0.623  0.623  0.608  0.636  0.653  0.672   \n",
       "1    6.672  6.758  6.963  7.152  7.359  7.423  7.593  9.744  9.743  9.756   \n",
       "2    3.300  3.361  3.344  3.373  3.373  6.802  6.803  6.789  6.772  6.763   \n",
       "3    2.150  2.151  2.130  2.136  2.144  2.147  2.135  2.129  2.126  2.121   \n",
       "4    2.332  2.327  2.323  2.309  2.290  2.209  2.216  2.244  2.240  2.259   \n",
       "5    2.665  2.662  2.650  2.657  2.644  2.649  2.652  2.671  2.668  2.662   \n",
       "6    2.184  2.184  2.193  2.185  2.205  2.208  2.216  2.214  2.220  2.225   \n",
       "7    3.280  3.288  3.295  3.295  3.306  3.314  3.319  3.328  3.342  3.342   \n",
       "8    2.689  2.685  2.692  2.695  2.714  2.740  2.861  3.315  3.315  2.589   \n",
       "9    2.838  2.876  2.871  2.839  2.845  3.188  3.257  3.257  3.223  2.748   \n",
       "10   0.612  0.594  0.594  0.590  0.651  3.119  3.086  2.890  2.805  2.782   \n",
       "11   0.565  0.625  0.628  0.614  0.603  0.608  0.608  0.608  0.627  0.647   \n",
       "12   0.607  0.576  0.605  0.605  0.606  0.617  0.625  0.645  0.664  0.669   \n",
       "13   3.686  3.633  3.624  3.590  3.582  3.548  3.543  3.526  3.515  3.487   \n",
       "14   2.166  2.180  2.222  2.645  2.722  2.756  2.810  2.811  2.831  2.837   \n",
       "15   1.994  1.833  1.818  1.812  1.801  1.804  1.799  1.828  1.907  2.059   \n",
       "16   2.454  2.444  2.445  2.429  2.323  2.241  2.232  2.225  2.215  2.224   \n",
       "17   0.631  0.631  0.600  0.585  0.592  0.601  0.652  0.692  0.675  0.593   \n",
       "18   0.635  0.637  0.629  0.567  0.566  0.566  0.565  0.651  0.651  0.696   \n",
       "19   2.216  2.429  3.291  3.293  3.293  3.313  3.322  3.315  3.315  3.347   \n",
       "20   3.653  3.251  3.150  3.184  3.558  3.728  3.708  3.694  3.708  3.734   \n",
       "21   2.967  2.939  2.922  2.893  2.888  2.859  2.838  2.816  2.792  2.762   \n",
       "22   2.107  2.105  2.102  2.100  2.102  2.102  2.097  2.065  1.960  1.936   \n",
       "23   0.644  0.597  0.566  0.582  0.604  0.648  0.603  0.618  0.771  0.774   \n",
       "24   1.668  1.686  1.724  1.734  1.759  1.779  1.811  1.827  1.859  1.870   \n",
       "25   0.601  0.606  0.614  0.596  0.597  0.615  0.617  0.675  0.770  2.156   \n",
       "26   0.563  0.572  0.572  0.581  0.573  0.596  0.612  0.625  0.617  0.616   \n",
       "27   0.637  0.598  0.581  0.567  0.591  0.603  0.609  0.611  0.625  0.639   \n",
       "28   0.609  0.601  0.601  0.608  0.614  0.622  0.622  0.636  0.595  0.625   \n",
       "29   0.609  0.608  0.603  0.608  0.626  0.661  0.650  0.621  0.650  0.690   \n",
       "..     ...    ...    ...    ...    ...    ...    ...    ...    ...    ...   \n",
       "234  0.635  0.630  0.608  0.602  0.599  0.599  0.620  0.627  0.674  0.634   \n",
       "235  0.543  0.582  0.604  0.626  0.607  0.570  0.555  0.576  0.638  0.666   \n",
       "236  0.533  0.536  0.555  0.540  0.549  0.549  0.588  0.634  0.635  0.633   \n",
       "237  0.552  0.547  0.567  0.557  0.554  0.554  0.575  0.582  0.579  0.577   \n",
       "238  0.668  0.639  0.645  0.616  0.616  0.624  0.662  0.664  0.675  0.756   \n",
       "239  0.662  0.669  0.677  0.691  0.694  0.698  0.609  0.654  0.707  0.752   \n",
       "240  5.664  5.689  5.718  5.806  5.827  5.863  5.856  4.437  4.395  4.579   \n",
       "241  5.294  5.374  5.358  5.403  5.370  5.483  5.567  5.613  5.021  4.718   \n",
       "242  4.907  4.719  6.190  6.283  6.345  7.166  7.134  6.903  6.536  6.359   \n",
       "243  3.491  3.499  3.512  3.517  3.526  3.534  3.542  3.542  3.558  3.559   \n",
       "244  6.599  6.628  6.630  6.627  6.637  6.694  6.893  6.998  7.324  7.381   \n",
       "245  3.337  3.352  3.372  3.395  6.787  6.785  6.775  6.755  6.740  6.711   \n",
       "246  2.126  2.121  2.123  2.121  2.122  2.118  2.118  2.118  2.120  2.100   \n",
       "247  2.216  2.215  2.245  2.265  2.269  2.228  2.234  2.233  2.225  2.235   \n",
       "248  2.665  2.647  2.653  2.655  2.657  2.661  2.665  2.649  2.630  2.539   \n",
       "249  2.205  2.205  2.204  2.210  2.217  2.219  2.235  2.235  2.239  2.250   \n",
       "250  3.290  3.290  3.292  3.299  3.303  3.317  3.329  3.336  3.341  3.344   \n",
       "251  2.703  2.709  2.712  2.712  2.718  2.720  2.770  2.863  3.344  3.352   \n",
       "252  2.070  2.105  2.291  2.302  2.306  2.325  2.338  2.393  2.788  2.776   \n",
       "253  0.642  0.675  1.701  3.071  3.109  3.143  3.143  2.894  2.832  2.815   \n",
       "254  0.584  0.586  0.599  0.624  0.629  0.618  0.597  0.623  0.669  0.706   \n",
       "255  0.632  0.601  0.601  0.603  0.625  0.625  0.624  0.663  0.662  0.701   \n",
       "256  3.636  3.609  3.582  3.580  3.565  3.547  3.462  3.374  3.267  3.269   \n",
       "257  2.189  2.188  2.199  2.218  2.228  2.236  2.750  2.750  2.795  2.852   \n",
       "258  2.177  2.180  2.174  2.157  2.139  2.026  1.935  1.860  1.838  1.826   \n",
       "259  2.434  2.428  2.417  2.413  2.447  2.446  2.456  2.453  2.453  2.403   \n",
       "260  0.598  0.612  0.592  0.590  0.631  0.582  0.575  0.617  0.628  0.628   \n",
       "261  0.643  0.648  0.635  0.635  0.586  0.639  0.665  0.689  0.702  0.709   \n",
       "262  2.074  2.080  2.227  2.485  3.312  3.327  3.333  3.328  3.321  3.324   \n",
       "263  3.738  3.741  3.715  3.188  3.169  3.207  3.503  3.727  3.737  3.735   \n",
       "\n",
       "     ...   y_noise   x_noise  theta_noise     x_rot     y_rot  delta_odom   Y  \\\n",
       "0    ...  1.035714  1.350020     1.750313  6.641049  3.250945    0.000000   0   \n",
       "1    ...  1.032806  1.344383     1.539686  6.636953  3.246101    0.004644   0   \n",
       "2    ...  1.028448  1.340108     1.325055  6.634671  3.240439    0.016999   0   \n",
       "3    ...  1.029537  1.338056     1.112569  6.632358  3.240660    0.016266   9   \n",
       "4    ...  1.022495  1.335604     0.914470  6.632788  3.233215    0.019450   9   \n",
       "5    ...  1.023280  1.333553     0.674101  6.630592  3.233155    0.029568   9   \n",
       "6    ...  1.021649  1.329625     0.477713  6.627588  3.230146    0.033285   3   \n",
       "7    ...  1.021730  1.326667     0.274111  6.624824  3.229088    0.037496   3   \n",
       "8    ...  1.022636  1.325774     0.048981  6.623653  3.229584    0.037064   3   \n",
       "9    ...  1.019186  1.332234    -0.153986  6.630941  3.228868    0.030765  11   \n",
       "10   ...  0.989331  1.324199    -0.374625  6.634942  3.198211    0.066363  11   \n",
       "11   ...  0.957698  1.304092    -0.590330  6.628472  3.161292    0.112827  11   \n",
       "12   ...  0.920562  1.269147    -0.805200  6.610398  3.113610    0.179511   6   \n",
       "13   ...  0.877029  1.213013    -1.014855  6.575196  3.051908    0.262562   6   \n",
       "14   ...  0.836900  1.131391    -1.224845  6.515144  2.983599    0.372451   6   \n",
       "15   ...  0.809691  1.017909    -1.428236  6.420713  2.915033    0.507970   1   \n",
       "16   ...  0.802515  0.852086    -1.637951  6.270259  2.844946    0.691832   1   \n",
       "17   ...  0.823411  0.458046    -1.503236  5.898217  2.713458    1.150678   1   \n",
       "18   ...  0.798090  0.098752    -1.606075  5.575963  2.552569    1.595312  10   \n",
       "19   ...  0.823886 -0.202639    -1.570507  5.287642  2.461063    1.954704  10   \n",
       "20   ...  0.825146 -0.205170    -1.360718  5.284822  2.461259    1.956511  10   \n",
       "21   ...  0.821483 -0.205713    -1.153256  5.285722  2.457667    1.960754   5   \n",
       "22   ...  0.825251 -0.205470    -0.931603  5.284504  2.461241    1.960854   5   \n",
       "23   ...  0.829827 -0.204245    -0.718095  5.283885  2.465938    1.957838   5   \n",
       "24   ...  0.830810 -0.200426    -0.506446  5.287037  2.468307    1.955938  15   \n",
       "25   ...  0.832573 -0.198906    -0.308557  5.287766  2.470518    1.951903  15   \n",
       "26   ...  0.837799 -0.195681    -0.098863  5.288745  2.476580    1.945966   4   \n",
       "27   ...  0.837069 -0.193857     0.115163  5.290710  2.476603    1.942955   4   \n",
       "28   ...  0.832844 -0.188633     0.342044  5.297154  2.474700    1.933763  14   \n",
       "29   ...  0.817664 -0.172010     0.560894  5.318321  2.467036    1.914236  14   \n",
       "..   ...       ...       ...          ...       ...       ...         ...  ..   \n",
       "234  ...  0.919114 -0.147631     0.443644  5.302020  2.570093    1.837659  14   \n",
       "235  ...  0.893033 -0.127451     0.663867  5.330645  2.553720    1.814434  14   \n",
       "236  ...  0.867843 -0.097606     0.880158  5.367858  2.541869    1.781405   2   \n",
       "237  ...  0.838276 -0.057815     1.073008  5.415935  2.529780    1.738179   2   \n",
       "238  ...  0.809922  0.010134     1.301265  5.489562  2.529587    1.667617   2   \n",
       "239  ...  0.797439  0.107468     1.515579  5.584264  2.555302    1.558568   7   \n",
       "240  ...  0.807330  0.241779     1.718374  5.704566  2.615840    1.398309   7   \n",
       "241  ...  0.863128  0.562906     1.670150  5.979896  2.790280    1.022240   7   \n",
       "242  ...  0.870681  0.957775     1.615743  6.341816  2.948368    0.558188  12   \n",
       "243  ...  0.912911  1.257845     1.831761  6.602884  3.102216    0.236772  12   \n",
       "244  ...  0.947360  1.405765     1.410013  6.726362  3.190650    0.093264   0   \n",
       "245  ...  0.946513  1.404663     1.208116  6.725667  3.189445    0.094116   0   \n",
       "246  ...  0.947416  1.406636     0.999418  6.727144  3.191034    0.098167   9   \n",
       "247  ...  0.941252  1.406220     0.817502  6.729119  3.185180    0.103568   9   \n",
       "248  ...  0.942643  1.404869     0.542955  6.727339  3.185948    0.094211   9   \n",
       "249  ...  0.940940  1.402126     0.344362  6.725456  3.183325    0.095920   3   \n",
       "250  ...  0.941150  1.398761     0.136874  6.722267  3.182232    0.090962   3   \n",
       "251  ...  0.935414  1.399125    -0.060493  6.724799  3.177072    0.096892   3   \n",
       "252  ...  0.919356  1.395268    -0.293647  6.727380  3.160759    0.112173  11   \n",
       "253  ...  0.895171  1.384946    -0.488721  6.727099  3.134466    0.144158  11   \n",
       "254  ...  0.872728  1.368373    -0.670040  6.720376  3.107389    0.184990  11   \n",
       "255  ...  0.834508  1.323827    -0.928808  6.693848  3.055031    0.239698   6   \n",
       "256  ...  0.804334  1.274169    -1.124658  6.659517  3.008151    0.300622   6   \n",
       "257  ...  0.774462  1.187015    -1.348455  6.590428  2.947200    0.391515   6   \n",
       "258  ...  0.761763  1.083811    -1.538943  6.499940  2.895973    0.504797   1   \n",
       "259  ...  0.775990  0.927739    -1.757301  6.350304  2.849391    0.653228   1   \n",
       "260  ...  0.850278  0.569604    -1.653426  5.991002  2.780971    1.009725   1   \n",
       "261  ...  0.867695  0.182198    -1.697039  5.626419  2.648809    1.447297  10   \n",
       "262  ...  0.940873 -0.173421    -1.704811  5.269867  2.580326    1.845325  10   \n",
       "263  ...  0.942737 -0.176349    -1.486557  5.266448  2.580928    1.850523  10   \n",
       "\n",
       "     Y_odom   Vk  Vk_aff  \n",
       "0         0  121      17  \n",
       "1         0   81       8  \n",
       "2         0   56      36  \n",
       "3         9   93       9  \n",
       "4         9   93       0  \n",
       "5         9   81      19  \n",
       "6         3   93      37  \n",
       "7         3   93      10  \n",
       "8         3   81      18  \n",
       "9        11   93       1  \n",
       "10       11   45       2  \n",
       "11       11  107      11  \n",
       "12        6  110      12  \n",
       "13        6   54       3  \n",
       "14        6    3       4  \n",
       "15        1  107       5  \n",
       "16        1   37       6  \n",
       "17        1  114      38  \n",
       "18       10   55      23  \n",
       "19       10   56       7  \n",
       "20       10   28      13  \n",
       "21        5   73      20  \n",
       "22        5   38      20  \n",
       "23        5  124      21  \n",
       "24       15   11      24  \n",
       "25       15   70      14  \n",
       "26        4   88      25  \n",
       "27        4   38      26  \n",
       "28       14   85      27  \n",
       "29       14  126      28  \n",
       "..      ...  ...     ...  \n",
       "234      14  126      28  \n",
       "235      14    7      29  \n",
       "236       2  107      30  \n",
       "237       2   33      31  \n",
       "238       2    3      34  \n",
       "239       7    2      15  \n",
       "240       7   93      32  \n",
       "241       7   91      16  \n",
       "242      12   58      35  \n",
       "243      12   23      22  \n",
       "244       0   81       8  \n",
       "245       0   56      36  \n",
       "246       9   93       9  \n",
       "247       9   93       0  \n",
       "248       9   81      19  \n",
       "249       3   93      37  \n",
       "250       3   93      10  \n",
       "251       3   81      18  \n",
       "252      11   93       1  \n",
       "253      11   45       2  \n",
       "254      11   81      11  \n",
       "255       6  110      12  \n",
       "256       6   54       3  \n",
       "257       6    3       4  \n",
       "258       1  107       5  \n",
       "259       1   37       6  \n",
       "260       1   37      38  \n",
       "261      10  113      23  \n",
       "262      10   56       7  \n",
       "263      10   56      13  \n",
       "\n",
       "[264 rows x 976 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.26315789, 0.10526316,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.26315789, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.36842105, 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.04761905,\n",
       "        0.33333333, 0.33333333, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.28571429],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.375     , 0.3125    , 0.        , 0.        , 0.3125    ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.33333333, 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.28571429, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.38095238, 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.5       , 0.5       , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.61111111, 0.38888889, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.31818182, 0.27272727,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.04545455, 0.36363636, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.33333333, 0.33333333, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.33333333, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.25      , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.5       , 0.        , 0.25      ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.26315789, 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.36842105,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.36842105,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.36842105, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.31578947, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.31578947, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.3       , 0.35      , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.25      , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.1       , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.125     , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.0625    , 0.        , 0.        , 0.0625    , 0.        ,\n",
       "        0.        , 0.        , 0.3125    , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.1875    , 0.        ,\n",
       "        0.25      , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.2       ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.2       , 0.        , 0.        , 0.        ,\n",
       "        0.2       , 0.        , 0.4       , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.33333333, 0.33333333, 0.33333333,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ],\n",
       "       [0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.5       ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.5       ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ]])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Modelo2.B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "path2= viterbi(obs2_windowed,Modelo2,Modelo2.PI)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([18,  1,  2, 11, 12,  3,  4,  5,  6, 38, 23,  7, 13, 20, 20, 21, 24,\n",
       "       14, 25, 26, 27, 28, 29, 30, 31, 34, 15, 32, 16, 35])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obs2_windowed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 81,  93,  45,  81, 110,  54,   3, 109,  37, 114,  28,  56,  56,\n",
       "        73, 102,   9,  11,  70,  88,  38,  85, 126,   7, 107,  33,   3,\n",
       "         2,  93,  76,   7])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obs_windowed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 3, 11, 11, 11,  6,  6,  6,  1,  1,  1, 10, 10, 10,  5,  5,  5, 15,\n",
       "       15,  4,  4, 14, 14, 14,  2,  2,  2,  7,  7,  7, 12])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "est_windowed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.        , 0.        , 0.        , 0.28571429, 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.1       , 0.0625    , 0.        , 0.        ,\n",
       "       0.        ])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Modelo2.B[:,18]"
   ]
  },
  {
   "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}