diff --git a/.gitignore b/.gitignore
index 4f3ef8b..f058a79 100644
--- a/.gitignore
+++ b/.gitignore
@@ -318,3 +318,4 @@ _doc/notebooks/sessions/A201612*.csv
 _doc/notebooks/sessions/A201612*.gz
 _doc/notebooks/sessions/bigdata/A201612*.gz
 
+_doc/notebooks/sessions/tweets_macron_sijetaispresident_201609.txt
diff --git a/_doc/notebooks/sessions/2017_session6.ipynb b/_doc/notebooks/sessions/2017_session6.ipynb
index aec581f..ac27262 100644
--- a/_doc/notebooks/sessions/2017_session6.ipynb
+++ b/_doc/notebooks/sessions/2017_session6.ipynb
@@ -522,10 +522,10 @@
             "text/plain": [
               "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
               "           max_features='auto', max_leaf_nodes=None,\n",
-              "           min_impurity_split=1e-07, min_samples_leaf=1,\n",
-              "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
-              "           n_estimators=10, n_jobs=1, oob_score=False, random_state=None,\n",
-              "           verbose=0, warm_start=False)"
+              "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
+              "           min_samples_leaf=1, min_samples_split=2,\n",
+              "           min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n",
+              "           oob_score=False, random_state=None, verbose=0, warm_start=False)"
             ]
           },
           "execution_count": null,
@@ -547,9 +547,9 @@
         {
           "data": {
             "text/plain": [
-              "array([ 0.05037649,  0.00081022,  0.00589337,  0.0006357 ,  0.01317193,\n",
-              "        0.48711232,  0.01352733,  0.05622799,  0.0044982 ,  0.01184547,\n",
-              "        0.02113995,  0.01039441,  0.32436663])"
+              "array([ 0.03571467,  0.00130566,  0.00606471,  0.00075703,  0.02778104,\n",
+              "        0.42221682,  0.01557762,  0.05941969,  0.00273247,  0.01327318,\n",
+              "        0.01474776,  0.01033679,  0.39007255])"
             ]
           },
           "execution_count": null,
@@ -587,9 +587,9 @@
       "outputs": [
         {
           "data": {
-            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsYAAAEICAYAAABcYjLsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsYAAAEICAYAAABcYjLsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
             "text/plain": [
-              "<matplotlib.figure.Figure at 0x17cf8d6ed68>"
+              "<matplotlib.figure.Figure at 0x1dbe5631080>"
             ]
           },
           "metadata": {},
@@ -621,7 +621,7 @@
         {
           "data": {
             "text/plain": [
-              "0.97587372510149317"
+              "0.98035593861463521"
             ]
           },
           "execution_count": null,
@@ -642,7 +642,7 @@
         {
           "data": {
             "text/plain": [
-              "0.97285694938662703"
+              "0.97561991735577791"
             ]
           },
           "execution_count": null,
@@ -709,19 +709,19 @@
         {
           "data": {
             "text/plain": [
-              "CRIM      -0.059532\n",
-              "ZN         0.053210\n",
-              "INDUS     -0.011778\n",
-              "CHAS       2.935796\n",
-              "NOX       -4.250586\n",
-              "RM         5.652168\n",
-              "AGE        0.011217\n",
-              "DIS       -0.756890\n",
-              "RAD        0.151340\n",
-              "TAX       -0.009599\n",
-              "PTRATIO   -0.387234\n",
-              "B          0.015458\n",
-              "LSTAT     -0.402584\n",
+              "CRIM      -0.035221\n",
+              "ZN         0.062151\n",
+              "INDUS     -0.083259\n",
+              "CHAS       3.527664\n",
+              "NOX       -4.892070\n",
+              "RM         5.554238\n",
+              "AGE       -0.006185\n",
+              "DIS       -1.342893\n",
+              "RAD        0.125101\n",
+              "TAX       -0.006814\n",
+              "PTRATIO   -0.163962\n",
+              "B          0.017828\n",
+              "LSTAT     -0.453238\n",
               "dtype: float64"
             ]
           },
@@ -745,25 +745,25 @@
               "<table class=\"simpletable\">\n",
               "<caption>OLS Regression Results</caption>\n",
               "<tr>\n",
-              "  <th>Dep. Variable:</th>          <td>MEDV</td>       <th>  R-squared:         </th> <td>   0.957</td> \n",
+              "  <th>Dep. Variable:</th>          <td>MEDV</td>       <th>  R-squared:         </th> <td>   0.956</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.955</td> \n",
+              "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.954</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   558.6</td> \n",
+              "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   544.4</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Date:</th>             <td>Mon, 26 Jun 2017</td> <th>  Prob (F-statistic):</th> <td>6.95e-214</td>\n",
+              "  <th>Date:</th>             <td>Mon, 28 Aug 2017</td> <th>  Prob (F-statistic):</th> <td>3.86e-212</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Time:</th>                 <td>23:01:24</td>     <th>  Log-Likelihood:    </th> <td> -1023.8</td> \n",
+              "  <th>Time:</th>                 <td>11:57:18</td>     <th>  Log-Likelihood:    </th> <td> -1033.7</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>No. Observations:</th>      <td>   339</td>      <th>  AIC:               </th> <td>   2074.</td> \n",
+              "  <th>No. Observations:</th>      <td>   339</td>      <th>  AIC:               </th> <td>   2093.</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Df Residuals:</th>          <td>   326</td>      <th>  BIC:               </th> <td>   2123.</td> \n",
+              "  <th>Df Residuals:</th>          <td>   326</td>      <th>  BIC:               </th> <td>   2143.</td> \n",
               "</tr>\n",
               "<tr>\n",
               "  <th>Df Model:</th>              <td>    13</td>      <th>                     </th>     <td> </td>    \n",
@@ -777,57 +777,57 @@
               "     <td></td>        <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>CRIM</th>    <td>   -0.0595</td> <td>    0.040</td> <td>   -1.493</td> <td> 0.136</td> <td>   -0.138</td> <td>    0.019</td>\n",
+              "  <th>CRIM</th>    <td>   -0.0352</td> <td>    0.057</td> <td>   -0.615</td> <td> 0.539</td> <td>   -0.148</td> <td>    0.077</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>ZN</th>      <td>    0.0532</td> <td>    0.019</td> <td>    2.828</td> <td> 0.005</td> <td>    0.016</td> <td>    0.090</td>\n",
+              "  <th>ZN</th>      <td>    0.0622</td> <td>    0.019</td> <td>    3.241</td> <td> 0.001</td> <td>    0.024</td> <td>    0.100</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>INDUS</th>   <td>   -0.0118</td> <td>    0.082</td> <td>   -0.144</td> <td> 0.886</td> <td>   -0.173</td> <td>    0.149</td>\n",
+              "  <th>INDUS</th>   <td>   -0.0833</td> <td>    0.085</td> <td>   -0.976</td> <td> 0.330</td> <td>   -0.251</td> <td>    0.085</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>CHAS</th>    <td>    2.9358</td> <td>    1.198</td> <td>    2.450</td> <td> 0.015</td> <td>    0.578</td> <td>    5.293</td>\n",
+              "  <th>CHAS</th>    <td>    3.5277</td> <td>    1.075</td> <td>    3.282</td> <td> 0.001</td> <td>    1.413</td> <td>    5.642</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>NOX</th>     <td>   -4.2506</td> <td>    4.178</td> <td>   -1.017</td> <td> 0.310</td> <td>  -12.469</td> <td>    3.968</td>\n",
+              "  <th>NOX</th>     <td>   -4.8921</td> <td>    4.106</td> <td>   -1.192</td> <td> 0.234</td> <td>  -12.969</td> <td>    3.185</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>RM</th>      <td>    5.6522</td> <td>    0.402</td> <td>   14.058</td> <td> 0.000</td> <td>    4.861</td> <td>    6.443</td>\n",
+              "  <th>RM</th>      <td>    5.5542</td> <td>    0.403</td> <td>   13.784</td> <td> 0.000</td> <td>    4.762</td> <td>    6.347</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>AGE</th>     <td>    0.0112</td> <td>    0.018</td> <td>    0.624</td> <td> 0.533</td> <td>   -0.024</td> <td>    0.047</td>\n",
+              "  <th>AGE</th>     <td>   -0.0062</td> <td>    0.018</td> <td>   -0.351</td> <td> 0.726</td> <td>   -0.041</td> <td>    0.028</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>DIS</th>     <td>   -0.7569</td> <td>    0.245</td> <td>   -3.088</td> <td> 0.002</td> <td>   -1.239</td> <td>   -0.275</td>\n",
+              "  <th>DIS</th>     <td>   -1.3429</td> <td>    0.267</td> <td>   -5.028</td> <td> 0.000</td> <td>   -1.868</td> <td>   -0.818</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>RAD</th>     <td>    0.1513</td> <td>    0.087</td> <td>    1.741</td> <td> 0.083</td> <td>   -0.020</td> <td>    0.322</td>\n",
+              "  <th>RAD</th>     <td>    0.1251</td> <td>    0.087</td> <td>    1.431</td> <td> 0.153</td> <td>   -0.047</td> <td>    0.297</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>TAX</th>     <td>   -0.0096</td> <td>    0.005</td> <td>   -1.832</td> <td> 0.068</td> <td>   -0.020</td> <td>    0.001</td>\n",
+              "  <th>TAX</th>     <td>   -0.0068</td> <td>    0.005</td> <td>   -1.306</td> <td> 0.192</td> <td>   -0.017</td> <td>    0.003</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>PTRATIO</th> <td>   -0.3872</td> <td>    0.137</td> <td>   -2.832</td> <td> 0.005</td> <td>   -0.656</td> <td>   -0.118</td>\n",
+              "  <th>PTRATIO</th> <td>   -0.1640</td> <td>    0.146</td> <td>   -1.124</td> <td> 0.262</td> <td>   -0.451</td> <td>    0.123</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>B</th>       <td>    0.0155</td> <td>    0.003</td> <td>    4.926</td> <td> 0.000</td> <td>    0.009</td> <td>    0.022</td>\n",
+              "  <th>B</th>       <td>    0.0178</td> <td>    0.004</td> <td>    5.035</td> <td> 0.000</td> <td>    0.011</td> <td>    0.025</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>LSTAT</th>   <td>   -0.4026</td> <td>    0.067</td> <td>   -6.039</td> <td> 0.000</td> <td>   -0.534</td> <td>   -0.271</td>\n",
+              "  <th>LSTAT</th>   <td>   -0.4532</td> <td>    0.065</td> <td>   -6.935</td> <td> 0.000</td> <td>   -0.582</td> <td>   -0.325</td>\n",
               "</tr>\n",
               "</table>\n",
               "<table class=\"simpletable\">\n",
               "<tr>\n",
-              "  <th>Omnibus:</th>       <td>118.166</td> <th>  Durbin-Watson:     </th> <td>   2.133</td> \n",
+              "  <th>Omnibus:</th>       <td>136.429</td> <th>  Durbin-Watson:     </th> <td>   2.020</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 678.319</td> \n",
+              "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 826.705</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Skew:</th>          <td> 1.331</td>  <th>  Prob(JB):          </th> <td>5.07e-148</td>\n",
+              "  <th>Skew:</th>          <td> 1.559</td>  <th>  Prob(JB):          </th> <td>3.04e-180</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Kurtosis:</th>      <td> 9.398</td>  <th>  Cond. No.          </th> <td>8.52e+03</td> \n",
+              "  <th>Kurtosis:</th>      <td> 9.986</td>  <th>  Cond. No.          </th> <td>8.22e+03</td> \n",
               "</tr>\n",
               "</table>"
             ],
@@ -836,41 +836,41 @@
               "\"\"\"\n",
               "                            OLS Regression Results                            \n",
               "==============================================================================\n",
-              "Dep. Variable:                   MEDV   R-squared:                       0.957\n",
-              "Model:                            OLS   Adj. R-squared:                  0.955\n",
-              "Method:                 Least Squares   F-statistic:                     558.6\n",
-              "Date:                Mon, 26 Jun 2017   Prob (F-statistic):          6.95e-214\n",
-              "Time:                        23:01:24   Log-Likelihood:                -1023.8\n",
-              "No. Observations:                 339   AIC:                             2074.\n",
-              "Df Residuals:                     326   BIC:                             2123.\n",
+              "Dep. Variable:                   MEDV   R-squared:                       0.956\n",
+              "Model:                            OLS   Adj. R-squared:                  0.954\n",
+              "Method:                 Least Squares   F-statistic:                     544.4\n",
+              "Date:                Mon, 28 Aug 2017   Prob (F-statistic):          3.86e-212\n",
+              "Time:                        11:57:18   Log-Likelihood:                -1033.7\n",
+              "No. Observations:                 339   AIC:                             2093.\n",
+              "Df Residuals:                     326   BIC:                             2143.\n",
               "Df Model:                          13                                         \n",
               "Covariance Type:            nonrobust                                         \n",
               "==============================================================================\n",
               "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
               "------------------------------------------------------------------------------\n",
-              "CRIM          -0.0595      0.040     -1.493      0.136      -0.138       0.019\n",
-              "ZN             0.0532      0.019      2.828      0.005       0.016       0.090\n",
-              "INDUS         -0.0118      0.082     -0.144      0.886      -0.173       0.149\n",
-              "CHAS           2.9358      1.198      2.450      0.015       0.578       5.293\n",
-              "NOX           -4.2506      4.178     -1.017      0.310     -12.469       3.968\n",
-              "RM             5.6522      0.402     14.058      0.000       4.861       6.443\n",
-              "AGE            0.0112      0.018      0.624      0.533      -0.024       0.047\n",
-              "DIS           -0.7569      0.245     -3.088      0.002      -1.239      -0.275\n",
-              "RAD            0.1513      0.087      1.741      0.083      -0.020       0.322\n",
-              "TAX           -0.0096      0.005     -1.832      0.068      -0.020       0.001\n",
-              "PTRATIO       -0.3872      0.137     -2.832      0.005      -0.656      -0.118\n",
-              "B              0.0155      0.003      4.926      0.000       0.009       0.022\n",
-              "LSTAT         -0.4026      0.067     -6.039      0.000      -0.534      -0.271\n",
+              "CRIM          -0.0352      0.057     -0.615      0.539      -0.148       0.077\n",
+              "ZN             0.0622      0.019      3.241      0.001       0.024       0.100\n",
+              "INDUS         -0.0833      0.085     -0.976      0.330      -0.251       0.085\n",
+              "CHAS           3.5277      1.075      3.282      0.001       1.413       5.642\n",
+              "NOX           -4.8921      4.106     -1.192      0.234     -12.969       3.185\n",
+              "RM             5.5542      0.403     13.784      0.000       4.762       6.347\n",
+              "AGE           -0.0062      0.018     -0.351      0.726      -0.041       0.028\n",
+              "DIS           -1.3429      0.267     -5.028      0.000      -1.868      -0.818\n",
+              "RAD            0.1251      0.087      1.431      0.153      -0.047       0.297\n",
+              "TAX           -0.0068      0.005     -1.306      0.192      -0.017       0.003\n",
+              "PTRATIO       -0.1640      0.146     -1.124      0.262      -0.451       0.123\n",
+              "B              0.0178      0.004      5.035      0.000       0.011       0.025\n",
+              "LSTAT         -0.4532      0.065     -6.935      0.000      -0.582      -0.325\n",
               "==============================================================================\n",
-              "Omnibus:                      118.166   Durbin-Watson:                   2.133\n",
-              "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              678.319\n",
-              "Skew:                           1.331   Prob(JB):                    5.07e-148\n",
-              "Kurtosis:                       9.398   Cond. No.                     8.52e+03\n",
+              "Omnibus:                      136.429   Durbin-Watson:                   2.020\n",
+              "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              826.705\n",
+              "Skew:                           1.559   Prob(JB):                    3.04e-180\n",
+              "Kurtosis:                       9.986   Cond. No.                     8.22e+03\n",
               "==============================================================================\n",
               "\n",
               "Warnings:\n",
               "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
-              "[2] The condition number is large, 8.52e+03. This might indicate that there are\n",
+              "[2] The condition number is large, 8.22e+03. This might indicate that there are\n",
               "strong multicollinearity or other numerical problems.\n",
               "\"\"\""
             ]
@@ -904,10 +904,10 @@
               "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   846.6</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Date:</th>             <td>Mon, 26 Jun 2017</td> <th>  Prob (F-statistic):</th> <td>2.38e-320</td>\n",
+              "  <th>Date:</th>             <td>Mon, 28 Aug 2017</td> <th>  Prob (F-statistic):</th> <td>2.38e-320</td>\n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Time:</th>                 <td>23:01:24</td>     <th>  Log-Likelihood:    </th> <td> -1556.1</td> \n",
+              "  <th>Time:</th>                 <td>11:57:18</td>     <th>  Log-Likelihood:    </th> <td> -1556.1</td> \n",
               "</tr>\n",
               "<tr>\n",
               "  <th>No. Observations:</th>      <td>   506</td>      <th>  AIC:               </th> <td>   3136.</td> \n",
@@ -986,8 +986,8 @@
               "Dep. Variable:                   MEDV   R-squared:                       0.954\n",
               "Model:                            OLS   Adj. R-squared:                  0.953\n",
               "Method:                 Least Squares   F-statistic:                     846.6\n",
-              "Date:                Mon, 26 Jun 2017   Prob (F-statistic):          2.38e-320\n",
-              "Time:                        23:01:24   Log-Likelihood:                -1556.1\n",
+              "Date:                Mon, 28 Aug 2017   Prob (F-statistic):          2.38e-320\n",
+              "Time:                        11:57:18   Log-Likelihood:                -1556.1\n",
               "No. Observations:                 506   AIC:                             3136.\n",
               "Df Residuals:                     494   BIC:                             3187.\n",
               "Df Model:                          12                                         \n",
@@ -1050,7 +1050,7 @@
           "name": "stderr",
           "output_type": "stream",
           "text": [
-            "c:\\Python36_x64\\lib\\site-packages\\sklearn\\cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
+            "c:\\Python36_x64\\lib\\site-packages\\sklearn\\cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
             "  \"This module will be removed in 0.20.\", DeprecationWarning)\n",
             "                                                                                                                       \r"
           ]
@@ -1059,7 +1059,7 @@
           "name": "stdout",
           "output_type": "stream",
           "text": [
-            "Generation 1 - Current best internal CV score: 8.866376470784171\n"
+            "Generation 1 - Current best internal CV score: 12.929444055404636\n"
           ]
         },
         {
@@ -1073,7 +1073,7 @@
           "name": "stdout",
           "output_type": "stream",
           "text": [
-            "Generation 2 - Current best internal CV score: 8.866376470784171\n"
+            "Generation 2 - Current best internal CV score: 12.929444055404636\n"
           ]
         },
         {
@@ -1088,8 +1088,8 @@
           "output_type": "stream",
           "text": [
             "\n",
-            "Best pipeline: XGBRegressor(input_matrix, XGBRegressor__learning_rate=0.1, XGBRegressor__max_depth=8, XGBRegressor__min_child_weight=3, XGBRegressor__n_estimators=DEFAULT, XGBRegressor__nthread=1, XGBRegressor__subsample=0.75)\n",
-            "17.1420145535\n"
+            "Best pipeline: GradientBoostingRegressor(input_matrix, GradientBoostingRegressor__alpha=DEFAULT, GradientBoostingRegressor__learning_rate=0.1, GradientBoostingRegressor__loss=huber, GradientBoostingRegressor__max_depth=DEFAULT, GradientBoostingRegressor__max_features=0.65, GradientBoostingRegressor__min_samples_leaf=1, GradientBoostingRegressor__min_samples_split=8, GradientBoostingRegressor__n_estimators=100, GradientBoostingRegressor__subsample=0.6)\n",
+            "9.34632700796\n"
           ]
         }
       ],
@@ -1116,7 +1116,7 @@
         {
           "data": {
             "text/plain": [
-              "0.99872711294155603"
+              "0.97107575656926248"
             ]
           },
           "execution_count": null,
@@ -1136,7 +1136,7 @@
         {
           "data": {
             "text/plain": [
-              "0.80075945303150209"
+              "0.88100225324209969"
             ]
           },
           "execution_count": null,
@@ -1169,10 +1169,10 @@
             "text/plain": [
               "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
               "           max_features='auto', max_leaf_nodes=None,\n",
-              "           min_impurity_split=1e-07, min_samples_leaf=1,\n",
-              "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
-              "           n_estimators=10, n_jobs=1, oob_score=False, random_state=None,\n",
-              "           verbose=0, warm_start=False)"
+              "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
+              "           min_samples_leaf=1, min_samples_split=2,\n",
+              "           min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n",
+              "           oob_score=False, random_state=None, verbose=0, warm_start=False)"
             ]
           },
           "execution_count": null,
@@ -1207,38 +1207,38 @@
           "output_type": "stream",
           "text": [
             "Instance 0\n",
-            "Bias (trainset mean) 22.4966205534\n",
+            "Bias (trainset mean) 22.8580039526\n",
             "Feature contributions:\n",
-            "RM 18.57\n",
-            "LSTAT 5.44\n",
-            "PTRATIO -3.9\n",
-            "B -1.59\n",
-            "AGE 0.68\n",
-            "NOX 0.2\n",
-            "DIS 0.1\n",
-            "TAX 0.06\n",
-            "ZN -0.02\n",
-            "CRIM -0.01\n",
-            "INDUS 0.01\n",
-            "CHAS 0.0\n",
+            "LSTAT -4.92\n",
+            "CRIM 2.39\n",
+            "RM -2.05\n",
+            "AGE 1.61\n",
+            "NOX 0.57\n",
+            "TAX 0.53\n",
+            "PTRATIO 0.34\n",
+            "INDUS 0.19\n",
+            "B -0.18\n",
+            "CHAS 0.05\n",
+            "DIS 0.04\n",
+            "ZN 0.0\n",
             "RAD 0.0\n",
             "--------------------\n",
             "Instance 1\n",
-            "Bias (trainset mean) 22.4966205534\n",
+            "Bias (trainset mean) 22.8580039526\n",
             "Feature contributions:\n",
-            "LSTAT -4.49\n",
-            "RM -1.54\n",
-            "CRIM 1.21\n",
-            "AGE 1.18\n",
-            "NOX 1.17\n",
-            "CHAS 0.59\n",
-            "PTRATIO 0.54\n",
-            "INDUS 0.42\n",
-            "B 0.39\n",
-            "TAX 0.25\n",
-            "DIS 0.09\n",
+            "RM -5.76\n",
+            "LSTAT 3.07\n",
+            "PTRATIO 0.75\n",
+            "CRIM -0.67\n",
+            "TAX -0.29\n",
+            "DIS -0.26\n",
+            "B 0.1\n",
+            "RAD -0.06\n",
+            "AGE 0.06\n",
+            "NOX 0.05\n",
+            "INDUS 0.04\n",
             "ZN 0.0\n",
-            "RAD 0.0\n",
+            "CHAS 0.0\n",
             "--------------------\n"
           ]
         }
@@ -1429,7 +1429,7 @@
         {
           "data": {
             "text/plain": [
-              "<matplotlib.axes._subplots.AxesSubplot at 0x17cfb769f60>"
+              "<matplotlib.axes._subplots.AxesSubplot at 0x1dbe7f0a0b8>"
             ]
           },
           "execution_count": null,
@@ -1440,7 +1440,7 @@
           "data": {
             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEKCAYAAAAcgp5RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ng8FgKHOMoTcYDIYy5/8BSx5BVGotCsEAAAAASUVORK5CYII=\n",
             "text/plain": [
-              "<matplotlib.figure.Figure at 0x17cfb7697b8>"
+              "<matplotlib.figure.Figure at 0x1dbe57beb70>"
             ]
           },
           "metadata": {},
@@ -1495,10 +1495,10 @@
               "  <th>Method:</th>             <td>css-mle</td>     <th>  S.D. of innovations</th>  <td>14.116</td>  \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Date:</th>          <td>Mon, 26 Jun 2017</td> <th>  AIC                </th> <td>2129.161</td> \n",
+              "  <th>Date:</th>          <td>Mon, 28 Aug 2017</td> <th>  AIC                </th> <td>2129.161</td> \n",
               "</tr>\n",
               "<tr>\n",
-              "  <th>Time:</th>              <td>23:02:05</td>     <th>  BIC                </th> <td>2161.173</td> \n",
+              "  <th>Time:</th>              <td>11:58:06</td>     <th>  BIC                </th> <td>2161.173</td> \n",
               "</tr>\n",
               "<tr>\n",
               "  <th>Sample:</th>                <td>1</td>        <th>  HQIC               </th> <td>2142.032</td> \n",
@@ -1572,8 +1572,8 @@
               "Dep. Variable:                    D.y   No. Observations:                  259\n",
               "Model:                 ARIMA(6, 1, 1)   Log Likelihood               -1055.581\n",
               "Method:                       css-mle   S.D. of innovations             14.116\n",
-              "Date:                Mon, 26 Jun 2017   AIC                           2129.161\n",
-              "Time:                        23:02:05   BIC                           2161.173\n",
+              "Date:                Mon, 28 Aug 2017   AIC                           2129.161\n",
+              "Time:                        11:58:06   BIC                           2161.173\n",
               "Sample:                             1   HQIC                          2142.032\n",
               "                                                                              \n",
               "==============================================================================\n",
@@ -1864,10 +1864,10 @@
             "text/plain": [
               "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
               "           max_features='auto', max_leaf_nodes=None,\n",
-              "           min_impurity_split=1e-07, min_samples_leaf=1,\n",
-              "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
-              "           n_estimators=10, n_jobs=1, oob_score=False, random_state=None,\n",
-              "           verbose=0, warm_start=False)"
+              "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
+              "           min_samples_leaf=1, min_samples_split=2,\n",
+              "           min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n",
+              "           oob_score=False, random_state=None, verbose=0, warm_start=False)"
             ]
           },
           "execution_count": null,
@@ -1889,7 +1889,7 @@
         {
           "data": {
             "text/plain": [
-              "0.50216101860132234"
+              "0.39863343986737942"
             ]
           },
           "execution_count": null,
@@ -1911,7 +1911,7 @@
         {
           "data": {
             "text/plain": [
-              "<matplotlib.collections.PathCollection at 0x17cfcc52b70>"
+              "<matplotlib.collections.PathCollection at 0x1dbe9ce11d0>"
             ]
           },
           "execution_count": null,
@@ -1920,9 +1920,9 @@
         },
         {
           "data": {
-            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAGgxJREFUeJzt3X+MXXd55/H3J5NJGafdjE28XnsSr4OIHAEpNh0lQUFV\nsAHTUhILoTRZWlnd7PqfqgWK3Dq7iIQKCaNUhUgrdWVBW2vLmoTU2FHQErJO8kcj1ewYOzi/rIT8\nHpx4SjNBlCm1nWf/uOfG15P745y5595z7jmflzSae8+9M/PcmTvP/d7nPN/vVxGBmZmNvvOKDsDM\nzPLhhG5mVhFO6GZmFeGEbmZWEU7oZmYV4YRuZlYRTuhmZhXhhG5mVhFO6GZmFXH+MH/YxRdfHOvW\nrRvmjzQzG3mHDx/+p4hY2et+Q03o69atY2ZmZpg/0sxs5El6Ic39XHIxM6sIJ3Qzs4pwQjczqwgn\ndDOzinBCNzOriFRdLpI+C/wXIIBjwB8Aq4FvAW8HDgO/HxH/NqA4zaym9h+Z5Y77j/OT+QXWTE6w\nY8t6tm6cKjqsUuo5Qpc0BfwxMB0R7wHGgJuArwBfjYh3Aq8BtwwyUDOrn/1HZrl13zFm5xcIYHZ+\ngVv3HWP/kdmiQyultCWX84EJSecDy4ATwCbgnuT2PcDW/MMzszq74/7jLJw6c86xhVNnuOP+4wVF\nVG49E3pEzAJ/AbxII5G/TqPEMh8Rp5O7vQz4PZCZ5eon8wuZjtddmpLLcuAG4DJgDXAh8NG0P0DS\ndkkzkmbm5uaWHKiZ1c+ayYlMx+suTcnlQ8BzETEXEaeAfcC1wGRSggG4BGhb1IqI3RExHRHTK1f2\nXIrAzOxNO7asZ2J87JxjE+Nj7NiyvqCIyi1NQn8RuEbSMkkCNgNPAA8Bn0zusw04MJgQzayutm6c\n4sufuJKpyQkETE1O8OVPXOkulw4UEb3vJH0R+F3gNHCERgvjFI22xRXJsd+LiF92+z7T09PhxbnM\nzLKRdDgipnvdL1UfekTcBty26PCzwFVLiM3MzAbAM0XNzCrCCd3MrCKc0M3MKsIJ3cysIpzQzcwq\nwgndzKwinNDNzCrCCd3MrCKc0M3MKsIJ3cysIpzQzcwqwgndzKwinNDNzCrCCd3MrCKc0M3MKsIJ\n3cysIpzQzcwqomdCl7Re0tGWj59J+oykFZIekPR08nn5MAI2M7P2eib0iDgeERsiYgPwG8AvgO8A\nO4GDEXE5cDC5bmZmBclactkM/DgiXgBuAPYkx/cAW/MMzMzMskm1SXSLm4C9yeVVEXEiufwKsCq3\nqMwsk/1HZrnj/uP8ZH6BNZMT7Niynq0bp4oOy4Ys9Qhd0gXA9cC3F98WEQFEh6/bLmlG0szc3NyS\nAzWz9vYfmeXWfceYnV8ggNn5BW7dd4z9R2aLDs2GLEvJ5beAH0bEq8n1VyWtBkg+n2z3RRGxOyKm\nI2J65cqV/UVrZm9xx/3HWTh15pxjC6fOcMf9xwuKyIqSpeRyM2fLLQD3AtuAXcnnAznGZTZQVSpR\n/GR+IdNxq65UI3RJFwIfBva1HN4FfFjS08CHkutmpVe1EsWayYlMx626UiX0iPiXiHh7RLzecuyn\nEbE5Ii6PiA9FxD8PLkyz/FStRLFjy3omxsfOOTYxPsaOLesLisiKkrXLxWzkVa1E0SwVVaWEZEvn\nhG61s2Zygtk2yXuUSxRbN045gZvXcrH6cYnCqsojdKsdlyisqpzQrZZcorAqcsnFzKwinNDNzCrC\nCd3MrCKc0M3MKsIJ3cysItzlYmaVVqWF2HpxQjezymouxNZcu6e5EBtQyaTukouZVVbVFmLrxQnd\nzCqragux9eKEbmaVVbe14p3Qzayy6rYQm0+Kmlll1W0htlQJXdIk8HXgPUAA/xk4DtwFrAOeB26M\niNcGEqWZ2RLVaSG2tCWXO4HvRcQVwHuBJ4GdwMGIuBw4mFw3M7OC9Ezoki4CfhP4BkBE/FtEzAM3\nAHuSu+0Btg4qSDMz6y3NCP0yYA74G0lHJH1d0oXAqog4kdznFWDVoII0M7Pe0tTQzwfeB/xRRByS\ndCeLyisREZKi3RdL2g5sB1i7dm2f4ZrVQ52mq1t+0ozQXwZejohDyfV7aCT4VyWtBkg+n2z3xRGx\nOyKmI2J65cqVecRsVmnN6eqz8wsEZ6er7z8yW3RoVnI9E3pEvAK8JKnZuLkZeAK4F9iWHNsGHBhI\nhGY1U7fp6paftH3ofwR8U9IFwLPAH9B4Mbhb0i3AC8CNgwnRrF7qNl3d8pMqoUfEUWC6zU2b8w3H\nzNZMTjDbJnlXdbq65cdT/81Kpm7T1S0/nvpvVjJ1m65u+XFCNyuhOk1Xt/y45GJmVhFO6GZmFeGE\nbmZWEU7oZmYV4YRuZlYRTuhmZhXhhG5mVhFO6GZmFeGEbmZWEU7oZmYV4YRuZlYRTuhmZhXhxbnM\nrNLqtD+rE7qZVVZzf9bmln7N/VmBSib1VCUXSc9LOibpqKSZ5NgKSQ9Iejr5vHywoZqZZVO3/Vmz\n1NA/GBEbIqK5Fd1O4GBEXA4cTK6bmZVG3fZn7eek6A3AnuTyHmBr/+GYmeWn0z6sVd2fNW1CD+D7\nkg5L2p4cWxURJ5LLrwCrco/OzKwPddufNe1J0Q9ExKykfw88IOmp1hsjIiRFuy9MXgC2A6xdu7av\nYM3Msqjb/qyKaJuHO3+BdDvwc+C/AtdFxAlJq4GHI6Lry9709HTMzMwsNVYzs1qSdLjl/GVHPUsu\nki6U9GvNy8BHgMeAe4Ftyd22AQeWHq6ZmfUrTcllFfAdSc37/++I+J6k/wfcLekW4AXgxsGFaWZm\nvfRM6BHxLPDeNsd/CmweRFBmZpad13IxM6sIJ3Qzs4pwQjczqwgvzmVWAXVaUdA6c0I3G3F1W1HQ\nOnPJxWzE1W1FQevMCd1sxNVtRUHrzAndbMTVbUVB68wJ3WzE1W1FQevMJ0XNRlzdVhS0zpzQzSpg\n68YpJ3BzycXMrCqc0M3MKsIJ3cysIpzQzcwqwidFK8TreZjVmxN6RXg9DzNLXXKRNCbpiKT7kuuX\nSTok6RlJd0m6YHBhWi9ez8PMstTQPw082XL9K8BXI+KdwGvALXkGZtl4PQ8zS5XQJV0CfAz4enJd\nwCbgnuQue4CtgwjQ0vF6HmaWdoT+NeBPgTeS628H5iPidHL9ZcCF2gJ5PQ8z65nQJf0OcDIiDi/l\nB0jaLmlG0szc3NxSvoWlsHXjFF/+xJVMTU4gYGpygi9/4kqfEDWrkTRdLtcC10v6beBtwL8D7gQm\nJZ2fjNIvAWbbfXFE7AZ2A0xPT0cuUVtbg17PY5TaIrPGOkqPzayTniP0iLg1Ii6JiHXATcCDEfEp\n4CHgk8ndtgEHBhalFa7ZFjk7v0Bwti1y/5G2r+OFyhrrKD02s276mSn6Z8CfSHqGRk39G/mEZGU0\nSm2RWWMdpcdm1k2miUUR8TDwcHL5WeCq/EOyMhqltsissY7SYzPrxmu5WCqj1BaZNdZRemxm3Tih\nWyqj1BaZNdZRemxm3XgtF0tllLY5yxrrKD02s24UMbxOwunp6ZiZmRnazzMzqwJJhyNiutf9XHIx\nM6sIl1xqzJNpzKrFCb2mvH66WfU4oddUt8k0eSX0UXoHMEqxmnXihF5Tg55MM0rvAEYpVrNufFK0\npgY9mWaUptOPUqxm3Tih11SayTT7j8xy7a4HuWznd7l214OZFqsapen0oxSrWTcuudRUr8k0/ZYh\n1kxOMNsmIQ5rOn2WmnjRsZrlxQm9QEWfiOu2fnq/J013bFl/zgsCDG86/f4js+z49qOceqMxaW52\nfoEd334UaP9iVGSsZnlyyaUgZV+Du92ItdvxxYrcQen2ex9/M5k3nXojuP3ex0sXq1mePEIvyDDa\nBvsxJnGmzbIQY1Lq7zHoHZQ6mV84lek4FBerWZ6c0AsyiBNxvUo4WUo87ZJ58/hlO7/rXm2zEnLJ\npSB5tw32KuFkLfFMdYmjjCWiVsuXjWc6blYVPRO6pLdJ+oGkRyU9LumLyfHLJB2S9IykuyRdMPhw\nqyPvNbh79VJn7bVuF99iZe3Vvu3j72Z87NzS0PiYuO3j7y4oov5aQM3SSlNy+SWwKSJ+Lmkc+AdJ\n/wf4E+CrEfEtSf8TuAX4qwHGWil5r8Hdq4STtcSzOL5Oiyzn2audV9dP2dY390xUG5aeCT0aC6b/\nPLk6nnwEsAn4T8nxPcDtOKFnkueJuF691EvptW6N79pdDw60VzvvpFemk5xlPwFu1ZGqhi5pTNJR\n4CTwAPBjYD4iTid3eRnwM7NAvUo4/ZZ4Br1NW5Wn33smqg1Lqi6XiDgDbJA0CXwHuCLtD5C0HdgO\nsHbt2qXEaCn0KjP0W4YYdBmjyknPM1FtWDJvQSfpC8AC8GfAf4iI05LeD9weEVu6fa23oCu3Imeu\ndirpTE1O8MjOTUOJYVAWl5Og8e7Gk5csrdy2oJO0MhmZI2kC+DDwJPAQ8MnkbtuAA0sPtzpGtZuh\n6Jmrgy7pFMkzUW1Yeo7QJf06jZOeYzReAO6OiD+X9A7gW8AK4AjwexHxy27fq+oj9FEeiZVhhFz0\n2jb9xFKm2K160o7Q03S5/AjY2Ob4s8BVSwuvmobdzZBnEilDDTvPzpR+fjdZO27clmhl4an/ORpm\nUsy6omCn79FMeud1WLul9cTdqIxC+02wWV+Y3ZZoZeGp/zma7DC1vNPxfmRdUXCxxTXzdsm8tYZd\ndI09i35bILO+MJfh3Y0ZeISeq39dlER6Hc+qdYTc6cxHtxUFW7VLetBYTfGNiLeMwDslyc/d/Sif\nvetoqUbs/SbYrG2Gbku0svAIPUcLp97IdDyLxSPkfnVKbm9E8Nyuj/HIzk3nJOdO9z8TUboRe78L\nn2XtuKlyh46Nlton9FFpM+w0ol6qrEkvTTLsVdYY1u+63wSbtc3QbYlWFrUuueTdnbB82Tiv/eKt\nJY88lm1NWy5Iu/1E1m3X2t0/S5zD7ATJY1Zr1o6bMq0dY/VV64Sed3fCbR9/NzvueZRTZ84WRVqX\nbe2nS6RTnXaxT12TbnmFrElv8f3TdMW0GnYniBOslcGwO8NqndDz7k7oliT7HaF+8IqV/N0/vtjx\n9jGJa96xnIeemuu4o1C7J1eWSUOtSbLTJKpOI/y8f9ej0kJp9VXE/IRaJ/RBdCd0Ghn22yXy3R+d\naHt8+bJxjnzhIz2fPGmeXJ/ff4y9h17iTARjEjdffSlf2nplx8fZfFxpkmqev2tP5LFRUMT8hFon\n9Kx15DQWjxw/eMVKHnpqrmO5pFm26JWU2tXmW4/3evL0uv3z+4+d8w7gTMSb17sl9bRPzDx/157I\nY6OgiPkJtU7oeZw8a03gk8vG+fm/nj5n9ma3Msli/SSlfncs2nvopba37z30UseEnkWey+96Io+N\ngiLmJ9Q6oUN/J88Wv/XvNIrOojUptb5YCNr2n09ONDpo+t2xqN0Jzm7HlyKvE5WeyGOjYBAVgF5q\n34fej7x7w+FsUlo8kahdWh0/T9x+faODJs2ORe02Tm7ePqb2DY+djjdjLKKH3xN5bBQUMT+h9iP0\nfuT9Fr81KWWdmp+qpLH4VaHl+s1XX9q2PHTz1Ze2jbXIE5Nl2wTarJNht89m3rGoH1VbD33jn3+/\n7zJLa4JunkDttlaLgOd2fSzzz0mz3nmWLpcyrJ9uVhe5rYdunXV6LWwWKSaXjfP6L07RbSWX5top\n7fq627loYmmzTtOcSPzS1itTnwD1iUmz8nFC78PrXVY2bI6imyc2O7UtNmvmaevxXUraXeV9ItEn\nJs3KJ82eopdKekjSE5Iel/Tp5PgKSQ9Iejr5vHzw4ZZLmoWstm6c4pGdm/ja725g/LxFJyXPO3tS\nMu3Idn6JJZ68TyT6xKRZ+aTpcjkNfC4i3gVcA/yhpHcBO4GDEXE5cDC5XiuZk9ri0XXL9bQj26WO\ngPM+4+4VBs3KJ/NJUUkHgP+RfFwXESckrQYejoiuw7OqnRSF9GuK9DqJmKaGPiobTptZvgZyUlTS\nOhobRh8CVkVEc4GRV4BVGWOshLRtSb1OIrZrxWvtenFrnpn1kjqhS/pV4O+Bz0TEz9Rydi4iQlLb\nob6k7cB2gLVr0y3tWkUXTYy33R6utWul14tDcyKPE7yZtZMqoUsap5HMvxkR+5LDr0pa3VJyOdnu\nayNiN7AbGiWXHGLuqqzLqnbqTknbteIVBs2slzRdLgK+ATwZEX/ZctO9wLbk8jbgQP7hZVOGnek7\nTYfv1J2Stmul353szaz60ozQrwV+Hzgm6Why7L8Bu4C7Jd0CvADcOJgQ0yt6WdVuo+h++7Y9kcfM\neumZ0CPiH+i8VeXmfMPpT9FJr9sLSr8rr3kij5n1UqnVFvvZsT4PnWaDzs4v9N237Yk8ZtZLpab+\nF7H+cKuxDhsnN5eg7WflNa8waGa9VCqhF530Br1JhHeyN7NuKpXQodikN9Whzj3lOreZDUGlauhF\nc53bzIpUuRF6kQZd8inrpCkzKwcn9JwNquTjmaJm1otLLiPCM0XNrBcn9BFR9KQpMys/J/QRUfSk\nKTMrPyf0EeEOGjPrxSdFR0TRk6bMrPxKn9Dr3KrX7rE/snNT0WGZWUmVOqHXuVWviMde5xdPsyoo\ndQ29zq16w37sZdgcxMz6U+qEXudWvWE/9jq/eJpVRakTep1b9Yb92Ov84mlWFWn2FP1rSSclPdZy\nbIWkByQ9nXxePojg6tyqN+zHXucXT7OqSDNC/1vgo4uO7QQORsTlwMHkeu763eVnlA37sdf5xdOs\nKhQpNl+QtA64LyLek1w/DlwXESckrQYejoie//nT09MxMzPTX8Q2MO5yMSsnSYcjYrrX/Zbatrgq\nIk4kl18BVi3x+1iJeEcks9HW90nRaAzxOw7zJW2XNCNpZm5urt8fZ2ZmHSx1hP6qpNUtJZeTne4Y\nEbuB3dAouWT9QS4DmJmls9QR+r3AtuTyNuBAPuGcy5NdzMzS63lSVNJe4DrgYuBV4DZgP3A3sBZ4\nAbgxIv651w/LelL02l0Ptt10eUzijQiP2HvI+u7G74bMyim3k6IRcXOHmzZnjiqjTpNaziQvQnVa\n2yWrrGvB1HndHLOqGMmZoq08Pb29rFP5PfXfbPSVOqG3m+zSTruyTN1lncrvqf9mo6/UCX3xbMlO\nxtTt1nrKOpXfU//NRl+pEzo0kvojOzfx3K6PdbzPmRSzXesm61R+T/03G32l3uBisanJibbllSmP\nIt8i65Z13uLObPSlWsslL/2u5bK4EwMao8i6LNhlZvU06LVcCuFRpJlZZyOV0OGtC0h9fv8xPnf3\no5yJYEzi5qsv5UtbrywwQjOzYoxcQm/1+f3H+Lt/fPHN62ci3rzupG5mdVP6Lpdu9h56KdNxM7Mq\nG+mE3qld0W2MZlZHI53QO00o8kQjM6ujkU7oN199aabjZmZVNtInRZsnPvceesldLmZWeyM1scjM\nrI7STiwa6ZKLmZmd5YRuZlYRTuhmZhXhhG5mVhFO6GZmFTHULhdJc8ALQ/uB2VwM/FPRQXRQ5tig\n3PGVOTYod3xljg3KHV/esf3HiFjZ605DTehlJmkmTVtQEcocG5Q7vjLHBuWOr8yxQbnjKyo2l1zM\nzCrCCd3MrCKc0M/aXXQAXZQ5Nih3fGWODcodX5ljg3LHV0hsrqGbmVWER+hmZhVRy4Qu6a8lnZT0\nWMuxFZIekPR08nl5QbFdKukhSU9IelzSp8sSn6S3SfqBpEeT2L6YHL9M0iFJz0i6S9IFw45tUZxj\nko5Iuq9M8Ul6XtIxSUclzSTHCv+7tsQ3KekeSU9JelLS+8sQn6T1ye+s+fEzSZ8pQ2wtMX42+Z94\nTNLe5H9l6M+7WiZ04G+Bjy46thM4GBGXAweT60U4DXwuIt4FXAP8oaR3lSS+XwKbIuK9wAbgo5Ku\nAb4CfDUi3gm8BtxSQGytPg082XK9TPF9MCI2tLS0leHv2nQn8L2IuAJ4L43fYeHxRcTx5He2AfgN\n4BfAd8oQG4CkKeCPgemIeA8wBtxEEc+7iKjlB7AOeKzl+nFgdXJ5NXC86BiTWA4AHy5bfMAy4IfA\n1TQmUJyfHH8/cH+BcV1C4597E3AfoLLEBzwPXLzoWCn+rsBFwHMk59XKFl9LPB8BHilTbMAU8BKw\ngsYeE/cBW4p43tV1hN7Oqog4kVx+BVhVZDAAktYBG4FDlCS+pJxxFDgJPAD8GJiPiNPJXV6m8QQv\nyteAPwXeSK6/nfLEF8D3JR2WtD05Voq/K3AZMAf8TVKu+rqkC0sUX9NNwN7kcilii4hZ4C+AF4ET\nwOvAYQp43jmhtxGNl9RC238k/Srw98BnIuJnrbcVGV9EnInGW99LgKuAK4qIox1JvwOcjIjDRcfS\nwQci4n3Ab9Eopf1m640FP+/OB94H/FVEbAT+hUUljKL/L5Ia9PXAtxffVmRsSe3+BhovimuAC3lr\nSXconNDPelXSaoDk88miApE0TiOZfzMi9pUtPoCImAceovFWclJSczvDS4DZgsK6Frhe0vPAt2iU\nXe6kJPElIzki4iSNGvBVlOfv+jLwckQcSq7fQyPBlyU+aLwQ/jAiXk2ulyW2DwHPRcRcRJwC9tF4\nLg79eeeEfta9wLbk8jYateuhkyTgG8CTEfGXLTcVHp+klZImk8sTNGr7T9JI7J8sMjaAiLg1Ii6J\niHU03po/GBGfKkN8ki6U9GvNyzRqwY9Rgr8rQES8ArwkaX1yaDPwBCWJL3EzZ8stUJ7YXgSukbQs\n+f9t/u6G/7wr8gRHUR80nhQngFM0Ria30Ki1HgSeBv4vsKKg2D5A463jj4CjycdvlyE+4NeBI0ls\njwFfSI6/A/gB8AyNt8O/UoK/8XXAfWWJL4nh0eTjceC/J8cL/7u2xLgBmEn+vvuB5WWJj0YZ46fA\nRS3HShFbEssXgaeS/4v/BfxKEc87zxQ1M6sIl1zMzCrCCd3MrCKc0M3MKsIJ3cysIpzQzcwqwgnd\nzKwinNDNzCrCCd3MrCL+PzON/SVhiIemAAAAAElFTkSuQmCC\n",
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlcT/B9JhEaUvJ8kyAAAAAElFTkSuQmCC\n",
             "text/plain": [
-              "<matplotlib.figure.Figure at 0x17cf8efc9e8>"
+              "<matplotlib.figure.Figure at 0x1dbe9c6dac8>"
             ]
           },
           "metadata": {},
@@ -2144,7 +2144,9 @@
     {
       "cell_type": "code",
       "execution_count": null,
-      "metadata": {},
+      "metadata": {
+        "collapsed": true
+      },
       "outputs": [],
       "source": [
         "keep = df.text.dropna().index"
@@ -2174,7 +2176,9 @@
     {
       "cell_type": "code",
       "execution_count": null,
-      "metadata": {},
+      "metadata": {
+        "collapsed": true
+      },
       "outputs": [],
       "source": [
         "from sklearn.feature_extraction.text import TfidfVectorizer\n",
@@ -2219,7 +2223,7 @@
       "outputs": [],
       "source": [
         "from sklearn.decomposition import LatentDirichletAllocation\n",
-        "lda = LatentDirichletAllocation(n_topics=10, max_iter=5,\n",
+        "lda = LatentDirichletAllocation(n_components=10, max_iter=5,\n",
         "                                learning_method='online',\n",
         "                                learning_offset=50.,\n",
         "                                random_state=0)"
@@ -2237,8 +2241,9 @@
               "             evaluate_every=-1, learning_decay=0.7,\n",
               "             learning_method='online', learning_offset=50.0,\n",
               "             max_doc_update_iter=100, max_iter=5, mean_change_tol=0.001,\n",
-              "             n_jobs=1, n_topics=10, perp_tol=0.1, random_state=0,\n",
-              "             topic_word_prior=None, total_samples=1000000.0, verbose=0)"
+              "             n_components=10, n_jobs=1, n_topics=None, perp_tol=0.1,\n",
+              "             random_state=0, topic_word_prior=None,\n",
+              "             total_samples=1000000.0, verbose=0)"
             ]
           },
           "execution_count": null,
@@ -2428,25 +2433,25 @@
           "output_type": "stream",
           "text": [
             "Topic #0:\n",
-            "serait sijetaispresident la le merde \u00e7a de et on ce\n",
+            "je sijetaispresident un pas en que serais ferais une ne\n",
             "Topic #1:\n",
-            "est macron le il de la pas hollande gauche qui\n",
+            "les sijetaispresident je et tous de pour seraient ballerines en\n",
             "Topic #2:\n",
-            "des sijetaispresident je de les et en le la pour\n",
+            "il macron de aurait qu est co https un pas\n",
             "Topic #3:\n",
-            "les sijetaispresident je et de tous pour ballerines seraient en\n",
+            "ministre premier nommerais de sijetaispresident je la serait mickey nommerai\n",
             "Topic #4:\n",
-            "vous seriez sijetaispresident je merde de pas la dans tous\n",
+            "macron est de hollande la le et pas qui sarko\n",
             "Topic #5:\n",
-            "https co macron de la le les et via sijetaispresident\n",
+            "des sijetaispresident je de et les en la pour le\n",
             "Topic #6:\n",
-            "je sijetaispresident le ferais serais pas un en que et\n",
+            "de sijetaispresident la je les et plus en au mois\n",
             "Topic #7:\n",
-            "ministre premier nommerais sijetaispresident de je la mickey serait 1er\n",
+            "la serait sijetaispresident merde dans je et france vous de\n",
             "Topic #8:\n",
-            "la de sijetaispresident je et france les en le r\u00e9publique\n",
+            "le sijetaispresident je monde et pour tout de serait les\n",
             "Topic #9:\n",
-            "sijetaispresident de le un pour plus une en et aurait\n",
+            "https co macron de la le les via et sijetaispresident\n",
             "\n"
           ]
         }
diff --git a/_doc/notebooks/sessions/tpot_boston_pipeline.py b/_doc/notebooks/sessions/tpot_boston_pipeline.py
new file mode 100644
index 0000000..b11aad2
--- /dev/null
+++ b/_doc/notebooks/sessions/tpot_boston_pipeline.py
@@ -0,0 +1,18 @@
+import numpy as np
+
+from sklearn.ensemble import GradientBoostingRegressor
+from sklearn.model_selection import train_test_split
+
+# NOTE: Make sure that the class is labeled 'class' in the data file
+tpot_data = np.recfromcsv(
+    'PATH/TO/DATA/FILE', delimiter='COLUMN_SEPARATOR', dtype=np.float64)
+features = np.delete(tpot_data.view(np.float64).reshape(
+    tpot_data.size, -1), tpot_data.dtype.names.index('class'), axis=1)
+training_features, testing_features, training_target, testing_target = \
+    train_test_split(features, tpot_data['class'], random_state=42)
+
+exported_pipeline = GradientBoostingRegressor(learning_rate=0.1, loss="huber", max_features=0.6500000000000001,
+                                              min_samples_leaf=1, min_samples_split=8, n_estimators=100, subsample=0.6000000000000001)
+
+exported_pipeline.fit(training_features, training_target)
+results = exported_pipeline.predict(testing_features)
diff --git a/_unittests/ut_documentation/test_notebook_sessions.py b/_unittests/ut_documentation/test_notebook_sessions.py
index 6116043..39a6c91 100644
--- a/_unittests/ut_documentation/test_notebook_sessions.py
+++ b/_unittests/ut_documentation/test_notebook_sessions.py
@@ -50,15 +50,12 @@ def setUp(self):
         add_missing_development_version(
             ["pyensae", "pymyinstall", "pymmails", "pyrsslocal", "mlstatpy",
              "jyquickhelper"], __file__, hide=True)
-
-    def test_notebook_session(self):
-        fLOG(
-            __file__,
-            self._testMethodName,
-            OutputPrint=__name__ == "__main__")
         fix_tkinter_issues_virtualenv()
+        self.fLOG = fLOG
 
+    def a_test_notebook_session(self, name):
         from src.actuariat_python.automation.notebook_test_helper import ls_notebooks, execute_notebooks, clean_function_notebook
+        fLOG = self.fLOG
         temp = get_temp_folder(__file__, "temp_sessions")
         keepnote = [_ for _ in ls_notebooks(
             "sessions") if "seance5_approche_fonctionnelle_enonce" not in _ and
@@ -67,7 +64,10 @@ def test_notebook_session(self):
         if is_travis_or_appveyor():
             keepnote = [
                 _ for _ in keepnote if "election_carte_electorale" not in _]
-        self.assertTrue(len(keepnote) > 0)
+        if name is not None:
+            keepnote = list(filter(lambda n: name in n, keepnote))
+        if len(keepnote) == 0:
+            return
         for k in keepnote:
             fLOG(k)
 
@@ -92,6 +92,92 @@ def test_notebook_session(self):
         execute_notebook_list_finalize_ut(
             res, fLOG=fLOG, dump=src.actuariat_python)
 
+    def test_notebook_session99(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session("2017_session6.ipynb")
+
+    def test_notebook_session999(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session("election_carte_electorale.ipynb")
+
+    def test_notebook_session3(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session("population_recuperation_donnees.ipynb")
+
+    def test_notebook_session4(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session(
+            "seance4_projection_population_correction.ipynb")
+
+    def test_notebook_session5(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session(
+            "seance5_approche_fonctionnelle_correction.ipynb")
+
+    def test_notebook_session6(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session(
+            "seance5_cube_multidimensionnel_correction.ipynb")
+
+    def test_notebook_session7(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session(
+            "seance5_cube_multidimensionnel_enonce.ipynb")
+
+    def test_notebook_session8(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session(
+            "seance5_sql_multidimensionnelle_correction.ipynb")
+
+    def test_notebook_session9(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session(
+            "seance5_sql_multidimensionnelle_enonce.ipynb")
+
+    def test_notebook_session10(self):
+        fLOG(
+            __file__,
+            self._testMethodName,
+            OutputPrint=__name__ == "__main__")
+
+        self.a_test_notebook_session("seance6_graphes_ml_correction.ipynb")
+
 
 if __name__ == "__main__":
     unittest.main()