diff --git a/.history/README_20250329013807.md b/.history/README_20250329013807.md
new file mode 100644
index 0000000000000000000000000000000000000000..55f9defd523f850d0616b9768ff9eaa88d528778
--- /dev/null
+++ b/.history/README_20250329013807.md
@@ -0,0 +1,44 @@
+## TOAR-classifier v2: A data-driven classification tool for global air quality stations
+This study develops a machine learning approach to classify 23,974 air quality monitoring stations in the TOAR database as urban, suburban, or rural using K-means clustering and an ensemble of supervised classifiers. The proposed method outperforms existing classifications, improving suburban accuracy and providing a more reliable foundation for air quality assessments.
+<img src="./figures/toar_classifier_v2.png" alt="My image" with="200">
+
+### Files
+- `data` is the folder containing all the data used in this work, including the predictions of station categories from the Machine Learning (ML) model.
+- `figures` contains all the figures.
+- `TOAR-classifier_v2.ipynb` is the notebook containing the code.
+- `requirements.txt` contains all the necessary packages.
+
+### Run the Code
+
+**Note:** This has been tested on Ubuntu 24.04.
+
+1. Install Python 3 if not already installed (most Linux systems have Python pre-installed).
+2. Install Jupyter Notebook:
+ - `pip install notebook` (for Jupyter Notebook) or
+ - `pip install jupyterlab` (for JupyterLab).
+3. clone the project by running the following command
+ - `git clone https://gitlab.jsc.fz-juelich.de/esde/toar-public/ml_toar_station_classification.git`
+4. Change directory to ml_toar_station_classification
+ - `cd ml_toar_station_classification`
+5. Creat virtual environment
+ - `python -m venv TOAR-classifier_v2` # feel free to change the virtual environment as convenient
+6. Activate the created venv
+ - `python -m ipykernel install --user --name=TOAR-classifier_v2 --display-name "Python (TOAR-classifier_v2)"`
+7. Install required package
+ - open jupyter notebook, `jupyter-notebook` and select kernel `TOAR-classifier_v2`
+ - Install all the required packages for the project by uncommenting the first cell in the notebook and running the cell
+8. Run the code cell by cell.
+
+
+### Citation
+
+If you use this please cite
+
+@article{Mache2025TOARClassifier,
+ author = {Ramiyou Karim Mache and Sabine Schröder and Michael Langguth and Ankit Patnala and Martin G. Schultz},
+ title = {TOAR-classifier v2: A data-driven classification tool for global air quality stations},
+ year = {2025},
+ note = {Correspondence: Ramiyou Karim Mache (k.mache@fz-juelich.de)},
+ url = {}
+}
+
diff --git a/README.md b/README.md
index 56e7c3acdf3c1e798abb1f7d48cc5d82e6591c91..55f9defd523f850d0616b9768ff9eaa88d528778 100644
--- a/README.md
+++ b/README.md
@@ -24,11 +24,10 @@ This study develops a machine learning approach to classify 23,974 air quality m
- `python -m venv TOAR-classifier_v2` # feel free to change the virtual environment as convenient
6. Activate the created venv
- `python -m ipykernel install --user --name=TOAR-classifier_v2 --display-name "Python (TOAR-classifier_v2)"`
-#### Install required package
-1. open jupyter notebook, `jupyter-notebook` and select kernel `TOAR-classifier_v2`
-2. Install all the required packages for the project by uncommenting the first cell in the notebook and running the cell
-
-Run the code cell by cell.
+7. Install required package
+ - open jupyter notebook, `jupyter-notebook` and select kernel `TOAR-classifier_v2`
+ - Install all the required packages for the project by uncommenting the first cell in the notebook and running the cell
+8. Run the code cell by cell.
### Citation
diff --git a/TOAR-classifier_v2.ipynb b/TOAR-classifier_v2.ipynb
index fc6b9bd626c556c6e4cb042e9bb3435925546760..199cc8473a23a87239cfca9f03b7cf64bde601bc 100755
--- a/TOAR-classifier_v2.ipynb
+++ b/TOAR-classifier_v2.ipynb
@@ -11,7 +11,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 1,
"id": "7fab0f12-d316-4be1-9825-e4e3e8265eee",
"metadata": {
"tags": []
@@ -24,7 +24,7 @@
},
{
"cell_type": "code",
- "execution_count": 58,
+ "execution_count": 7,
"id": "8666a040",
"metadata": {
"tags": []
@@ -87,21 +87,18 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 8,
"id": "a3d18ced-8c4d-472d-b31c-9329a8d29843",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
- "test_cord_urban = ['GB0682A', 'GB1072A', 'GB1095A', 'GB0960A', 'GB1035A', 'DEBE092', 'DEHH026', 'DENW376', 'DEBW118', 'FR04118', 'ES1422A',\n",
- " 'IT1016A', '36-081-0124', '13-121-0001', '06-037-4002','openaq_225445', 'jp23105510']\n",
- "\n",
+ "test_cord_urban = ['GB0682A', 'GB1072A', 'GB1095A', 'GB0960A', 'GB1035A', 'DEBE092', 'DEHH026', 'DENW376', 'DEBW118', 'FR04118', \n",
+ " 'ES1422A', 'IT1016A', '36-081-0124', '13-121-0001', '06-037-4002','openaq_225445', 'jp23105510']\n",
"test_cord_suburban = ['GB1092A', 'GB0885A', 'FR21031', '06-037-0019', 'jp20421030']\n",
- "\n",
- "test_cord_rural = ['GB1055R', 'GB0013R', 'GB0006R', 'IE0031R', 'NO0015R', 'FI00363', 'DENI051', 'DENW192', 'DERP015', 'FR23068','FR19020', \n",
- " 'ES1616A', 'IT1942A', 'MT0001R', '37-105-0002', '08-123-0013','06-111-0005', 'openaq_230819', 'openaq_226125', \n",
- " 'jp21601010']\n",
+ "test_cord_rural = ['GB1055R', 'GB0013R', 'GB0006R', 'IE0031R', 'NO0015R', 'FI00363', 'DENI051', 'DENW192', 'DERP015', 'FR23068','FR19020', 'ES1616A', \n",
+ " 'IT1942A', 'MT0001R', '37-105-0002', '08-123-0013','06-111-0005', 'openaq_230819', 'openaq_226125', 'jp21601010']\n",
"\n",
"test_cord = {'urban': test_cord_urban, 'suburban': test_cord_suburban, 'rural': test_cord_rural}\n",
"categories = ['urban', 'suburban', 'rural']"
@@ -117,7 +114,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 9,
"id": "7510e0d0-4c69-43ee-96f9-d4d391895eac",
"metadata": {
"tags": []
@@ -167,7 +164,7 @@
},
{
"cell_type": "code",
- "execution_count": 104,
+ "execution_count": 10,
"id": "40251495-25f5-44ec-9d82-1650efb88d5a",
"metadata": {},
"outputs": [],
@@ -266,7 +263,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 11,
"id": "d564a27a-c42c-4685-a8ff-1bafb3067c25",
"metadata": {},
"outputs": [],
@@ -437,7 +434,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 12,
"id": "997bc298-b007-4fa6-b638-0c96abb67f5e",
"metadata": {},
"outputs": [],
@@ -513,24 +510,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
- "id": "d58e8d30-7b75-4e89-8827-10cf1d420efc",
- "metadata": {
- "tags": []
- },
- "outputs": [],
- "source": [
- "#from scipy.special import inv_boxcox\n",
- "# y_boxcox_nox_2015, lmb_func_nox_2015 = stats.boxcox(dataset['mean_nox_emissions_10km_year2015'].values)\n",
- "# y_boxcox_nox_2000, lmb_func_nox_2000 = stats.boxcox(dataset['mean_nox_emissions_10km_year2000'].values)\n",
- "# #stats.boxcox(Y_train)\n",
- "# dataset['mean_nox_emissions_10km_year2015'] = y_boxcox_nox_2015\n",
- "# dataset['mean_nox_emissions_10km_year2000'] = y_boxcox_nox_2000"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 15,
+ "execution_count": 13,
"id": "31a37a57-8cb9-4ab7-9029-c91758d93698",
"metadata": {},
"outputs": [
@@ -892,7 +872,7 @@
"[5 rows x 22 columns]"
]
},
- "execution_count": 15,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
@@ -904,7 +884,7 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 14,
"id": "779730ee-dce3-4121-890a-1296ebb4f717",
"metadata": {
"collapsed": true,
@@ -912,902 +892,16 @@
"outputs_hidden": true
}
},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead tr th {\n",
- " text-align: left;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr>\n",
- " <th>lat</th>\n",
- " <th>-34.350000</th>\n",
- " <th>46.547500</th>\n",
- " <th>36.280000</th>\n",
- " <th>-40.683119</th>\n",
- " <th>-14.247000</th>\n",
- " <th>35.692200</th>\n",
- " <th>-54.848400</th>\n",
- " <th>-26.230594</th>\n",
- " <th>-25.722592</th>\n",
- " <th>19.325150</th>\n",
- " <th>...</th>\n",
- " <th>47.766666</th>\n",
- " <th>49.964994</th>\n",
- " <th>32.750000</th>\n",
- " <th>59.317200</th>\n",
- " <th>35.233400</th>\n",
- " <th>35.310600</th>\n",
- " <th>40.768800</th>\n",
- " <th>31.684200</th>\n",
- " <th>38.609769</th>\n",
- " <th>41.614685</th>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>lon</th>\n",
- " <th>18.480000</th>\n",
- " <th>7.985000</th>\n",
- " <th>100.900000</th>\n",
- " <th>144.689939</th>\n",
- " <th>-170.564000</th>\n",
- " <th>139.768900</th>\n",
- " <th>-68.310700</th>\n",
- " <th>28.020442</th>\n",
- " <th>28.420414</th>\n",
- " <th>-99.204100</th>\n",
- " <th>...</th>\n",
- " <th>16.766666</th>\n",
- " <th>8.565859</th>\n",
- " <th>128.680000</th>\n",
- " <th>18.048900</th>\n",
- " <th>129.010200</th>\n",
- " <th>128.987200</th>\n",
- " <th>114.903200</th>\n",
- " <th>120.288000</th>\n",
- " <th>-86.082006</th>\n",
- " <th>-87.124560</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>altitude</th>\n",
- " <td>230.0</td>\n",
- " <td>3578.0</td>\n",
- " <td>3810.0</td>\n",
- " <td>94.0</td>\n",
- " <td>42.0</td>\n",
- " <td>2.0</td>\n",
- " <td>18.0</td>\n",
- " <td>1773.0</td>\n",
- " <td>1754.0</td>\n",
- " <td>2326.0</td>\n",
- " <td>...</td>\n",
- " <td>117.0</td>\n",
- " <td>99.0</td>\n",
- " <td>500.0</td>\n",
- " <td>24.0</td>\n",
- " <td>12.0</td>\n",
- " <td>4.0</td>\n",
- " <td>728.0</td>\n",
- " <td>7.0</td>\n",
- " <td>226.1</td>\n",
- " <td>192.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_topography_srtm_alt_90m_year1994</th>\n",
- " <td>112.0</td>\n",
- " <td>3466.0</td>\n",
- " <td>3730.0</td>\n",
- " <td>53.0</td>\n",
- " <td>61.0</td>\n",
- " <td>21.0</td>\n",
- " <td>3.0</td>\n",
- " <td>1696.0</td>\n",
- " <td>1333.0</td>\n",
- " <td>2349.0</td>\n",
- " <td>...</td>\n",
- " <td>113.0</td>\n",
- " <td>110.0</td>\n",
- " <td>73.0</td>\n",
- " <td>43.0</td>\n",
- " <td>12.0</td>\n",
- " <td>4.0</td>\n",
- " <td>734.0</td>\n",
- " <td>7.0</td>\n",
- " <td>225.0</td>\n",
- " <td>196.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_topography_srtm_alt_1km_year1994</th>\n",
- " <td>84.784127</td>\n",
- " <td>3354.26945</td>\n",
- " <td>3697.235165</td>\n",
- " <td>54.430556</td>\n",
- " <td>31.087379</td>\n",
- " <td>20.059341</td>\n",
- " <td>20.369295</td>\n",
- " <td>1699.634568</td>\n",
- " <td>1332.74938</td>\n",
- " <td>2345.320413</td>\n",
- " <td>...</td>\n",
- " <td>113.502773</td>\n",
- " <td>104.123894</td>\n",
- " <td>65.820513</td>\n",
- " <td>25.151899</td>\n",
- " <td>21.232662</td>\n",
- " <td>34.868009</td>\n",
- " <td>734.194969</td>\n",
- " <td>5.645688</td>\n",
- " <td>238.473233</td>\n",
- " <td>195.474012</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max_topography_srtm_relative_alt_5km_year1994</th>\n",
- " <td>140.0</td>\n",
- " <td>603.0</td>\n",
- " <td>84.0</td>\n",
- " <td>57.0</td>\n",
- " <td>253.0</td>\n",
- " <td>53.0</td>\n",
- " <td>270.0</td>\n",
- " <td>102.0</td>\n",
- " <td>225.0</td>\n",
- " <td>350.0</td>\n",
- " <td>...</td>\n",
- " <td>10.0</td>\n",
- " <td>28.0</td>\n",
- " <td>315.0</td>\n",
- " <td>36.0</td>\n",
- " <td>611.0</td>\n",
- " <td>497.0</td>\n",
- " <td>398.0</td>\n",
- " <td>33.0</td>\n",
- " <td>73.0</td>\n",
- " <td>28.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>min_topography_srtm_relative_alt_5km_year1994</th>\n",
- " <td>-112.0</td>\n",
- " <td>-2341.0</td>\n",
- " <td>-752.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>-62.0</td>\n",
- " <td>-65.0</td>\n",
- " <td>-101.0</td>\n",
- " <td>...</td>\n",
- " <td>-8.0</td>\n",
- " <td>-23.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>-59.0</td>\n",
- " <td>0.0</td>\n",
- " <td>-23.0</td>\n",
- " <td>-26.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>stddev_topography_srtm_relative_alt_5km_year1994</th>\n",
- " <td>-68.713116</td>\n",
- " <td>-2873.715416</td>\n",
- " <td>-3560.781547</td>\n",
- " <td>-25.259737</td>\n",
- " <td>7.074067</td>\n",
- " <td>-9.352232</td>\n",
- " <td>31.423392</td>\n",
- " <td>-1668.003299</td>\n",
- " <td>-1281.138149</td>\n",
- " <td>-2262.805486</td>\n",
- " <td>...</td>\n",
- " <td>-110.112831</td>\n",
- " <td>-102.475289</td>\n",
- " <td>-0.688234</td>\n",
- " <td>-28.694404</td>\n",
- " <td>149.07072</td>\n",
- " <td>106.34765</td>\n",
- " <td>-654.514377</td>\n",
- " <td>-4.160356</td>\n",
- " <td>-211.419405</td>\n",
- " <td>-187.203666</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>climatic_zone_year2016</th>\n",
- " <td>6.0</td>\n",
- " <td>7.0</td>\n",
- " <td>8.0</td>\n",
- " <td>5.0</td>\n",
- " <td>2.0</td>\n",
- " <td>5.0</td>\n",
- " <td>8.0</td>\n",
- " <td>6.0</td>\n",
- " <td>6.0</td>\n",
- " <td>6.0</td>\n",
- " <td>...</td>\n",
- " <td>6.0</td>\n",
- " <td>6.0</td>\n",
- " <td>5.0</td>\n",
- " <td>8.0</td>\n",
- " <td>5.0</td>\n",
- " <td>5.0</td>\n",
- " <td>8.0</td>\n",
- " <td>5.0</td>\n",
- " <td>5.0</td>\n",
- " <td>6.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>distance_to_major_road_year2020</th>\n",
- " <td>24.235018</td>\n",
- " <td>6197.879799</td>\n",
- " <td>3595.14835</td>\n",
- " <td>5230.684142</td>\n",
- " <td>511.406318</td>\n",
- " <td>34.213688</td>\n",
- " <td>1043.890433</td>\n",
- " <td>96.586932</td>\n",
- " <td>85.449787</td>\n",
- " <td>238.581076</td>\n",
- " <td>...</td>\n",
- " <td>197.163512</td>\n",
- " <td>338.208048</td>\n",
- " <td>505.19027</td>\n",
- " <td>0.008772</td>\n",
- " <td>37.155775</td>\n",
- " <td>23.089269</td>\n",
- " <td>34.57441</td>\n",
- " <td>95.008468</td>\n",
- " <td>634.926265</td>\n",
- " <td>106.808314</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_stable_nightlights_1km_year2013</th>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>8.0</td>\n",
- " <td>63.0</td>\n",
- " <td>14.0</td>\n",
- " <td>63.0</td>\n",
- " <td>56.0</td>\n",
- " <td>63.0</td>\n",
- " <td>...</td>\n",
- " <td>8.0</td>\n",
- " <td>31.0</td>\n",
- " <td>6.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>61.0</td>\n",
- " <td>63.0</td>\n",
- " <td>62.0</td>\n",
- " <td>43.0</td>\n",
- " <td>59.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_stable_nightlights_5km_year2013</th>\n",
- " <td>0.0</td>\n",
- " <td>2.082707</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>4.041237</td>\n",
- " <td>63.0</td>\n",
- " <td>18.677019</td>\n",
- " <td>62.951456</td>\n",
- " <td>43.861386</td>\n",
- " <td>62.979381</td>\n",
- " <td>...</td>\n",
- " <td>5.402878</td>\n",
- " <td>33.524476</td>\n",
- " <td>2.198198</td>\n",
- " <td>63.0</td>\n",
- " <td>61.486957</td>\n",
- " <td>52.678261</td>\n",
- " <td>54.017094</td>\n",
- " <td>60.288288</td>\n",
- " <td>21.913043</td>\n",
- " <td>54.239669</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max_stable_nightlights_25km_year2013</th>\n",
- " <td>50.0</td>\n",
- " <td>52.0</td>\n",
- " <td>54.0</td>\n",
- " <td>0.0</td>\n",
- " <td>47.0</td>\n",
- " <td>63.0</td>\n",
- " <td>62.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>...</td>\n",
- " <td>57.0</td>\n",
- " <td>63.0</td>\n",
- " <td>41.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>54.0</td>\n",
- " <td>63.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max_stable_nightlights_25km_year1992</th>\n",
- " <td>40.0</td>\n",
- " <td>42.0</td>\n",
- " <td>23.0</td>\n",
- " <td>0.0</td>\n",
- " <td>32.0</td>\n",
- " <td>63.0</td>\n",
- " <td>61.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>63.0</td>\n",
- " <td>...</td>\n",
- " <td>41.0</td>\n",
- " <td>63.0</td>\n",
- " <td>32.0</td>\n",
- " <td>63.0</td>\n",
- " <td>61.0</td>\n",
- " <td>61.0</td>\n",
- " <td>62.0</td>\n",
- " <td>59.0</td>\n",
- " <td>53.0</td>\n",
- " <td>63.0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_population_density_250m_year2015</th>\n",
- " <td>1.339095</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>4910.184053</td>\n",
- " <td>0.0</td>\n",
- " <td>683.74621</td>\n",
- " <td>23970.364846</td>\n",
- " <td>2455.663829</td>\n",
- " <td>...</td>\n",
- " <td>86.317243</td>\n",
- " <td>991.754621</td>\n",
- " <td>20.728297</td>\n",
- " <td>92283.564398</td>\n",
- " <td>278516.909439</td>\n",
- " <td>27525.825639</td>\n",
- " <td>78374.101361</td>\n",
- " <td>53906.583742</td>\n",
- " <td>402.390048</td>\n",
- " <td>1728.186332</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_population_density_5km_year2015</th>\n",
- " <td>0.652963</td>\n",
- " <td>1.64468</td>\n",
- " <td>0.0</td>\n",
- " <td>0.39804</td>\n",
- " <td>259.638586</td>\n",
- " <td>13827.48708</td>\n",
- " <td>1230.179145</td>\n",
- " <td>3772.176769</td>\n",
- " <td>3844.87098</td>\n",
- " <td>9133.435742</td>\n",
- " <td>...</td>\n",
- " <td>31.180866</td>\n",
- " <td>494.029262</td>\n",
- " <td>32.934671</td>\n",
- " <td>6697.156876</td>\n",
- " <td>4361.633324</td>\n",
- " <td>1192.047095</td>\n",
- " <td>3462.646015</td>\n",
- " <td>2220.836806</td>\n",
- " <td>102.342802</td>\n",
- " <td>244.667348</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max_population_density_25km_year2015</th>\n",
- " <td>5356.886333</td>\n",
- " <td>3845.294377</td>\n",
- " <td>4960.909885</td>\n",
- " <td>230.173008</td>\n",
- " <td>2648.221272</td>\n",
- " <td>22738.63592</td>\n",
- " <td>13383.638228</td>\n",
- " <td>59504.144856</td>\n",
- " <td>27545.669643</td>\n",
- " <td>37332.160881</td>\n",
- " <td>...</td>\n",
- " <td>5965.60968</td>\n",
- " <td>6787.482117</td>\n",
- " <td>2050.823651</td>\n",
- " <td>23707.527229</td>\n",
- " <td>30320.882859</td>\n",
- " <td>30349.793048</td>\n",
- " <td>43562.550513</td>\n",
- " <td>90470.580623</td>\n",
- " <td>971.682298</td>\n",
- " <td>2829.102999</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_population_density_250m_year1990</th>\n",
- " <td>297.652617</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>0.0</td>\n",
- " <td>4062.595509</td>\n",
- " <td>0.0</td>\n",
- " <td>295.306869</td>\n",
- " <td>10017.486937</td>\n",
- " <td>1740.817741</td>\n",
- " <td>...</td>\n",
- " <td>3.767978</td>\n",
- " <td>747.184087</td>\n",
- " <td>18.049066</td>\n",
- " <td>60448.397686</td>\n",
- " <td>260331.747506</td>\n",
- " <td>21824.192189</td>\n",
- " <td>74094.474606</td>\n",
- " <td>72672.567509</td>\n",
- " <td>472.818732</td>\n",
- " <td>1225.107014</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_population_density_5km_year1990</th>\n",
- " <td>-1.744407</td>\n",
- " <td>1.979643</td>\n",
- " <td>0.0</td>\n",
- " <td>0.343217</td>\n",
- " <td>122.390694</td>\n",
- " <td>10911.150329</td>\n",
- " <td>661.777481</td>\n",
- " <td>1628.943119</td>\n",
- " <td>1701.044226</td>\n",
- " <td>8097.352945</td>\n",
- " <td>...</td>\n",
- " <td>36.922714</td>\n",
- " <td>452.74777</td>\n",
- " <td>37.191118</td>\n",
- " <td>4467.052098</td>\n",
- " <td>3816.279499</td>\n",
- " <td>556.038768</td>\n",
- " <td>3156.993358</td>\n",
- " <td>1236.589719</td>\n",
- " <td>92.550635</td>\n",
- " <td>196.181747</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max_population_density_25km_year1990</th>\n",
- " <td>2700.758359</td>\n",
- " <td>3047.447108</td>\n",
- " <td>5311.749164</td>\n",
- " <td>260.081728</td>\n",
- " <td>4216.752589</td>\n",
- " <td>17796.726371</td>\n",
- " <td>7554.773868</td>\n",
- " <td>25683.872557</td>\n",
- " <td>12208.812374</td>\n",
- " <td>40316.710058</td>\n",
- " <td>...</td>\n",
- " <td>4922.215553</td>\n",
- " <td>7162.790053</td>\n",
- " <td>2733.137455</td>\n",
- " <td>16324.082429</td>\n",
- " <td>30455.515258</td>\n",
- " <td>30484.553816</td>\n",
- " <td>39192.710089</td>\n",
- " <td>61069.93486</td>\n",
- " <td>887.06403</td>\n",
- " <td>2336.722565</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_nox_emissions_10km_year2015</th>\n",
- " <td>148.964264</td>\n",
- " <td>72.823387</td>\n",
- " <td>192.781525</td>\n",
- " <td>0.084431</td>\n",
- " <td>34.131149</td>\n",
- " <td>132864.84375</td>\n",
- " <td>632.638062</td>\n",
- " <td>14969.367188</td>\n",
- " <td>1623.259521</td>\n",
- " <td>18833.537109</td>\n",
- " <td>...</td>\n",
- " <td>158.663528</td>\n",
- " <td>2697.373291</td>\n",
- " <td>194.632599</td>\n",
- " <td>27311.535156</td>\n",
- " <td>27003.199219</td>\n",
- " <td>5525.850098</td>\n",
- " <td>75536.96875</td>\n",
- " <td>42011.578125</td>\n",
- " <td>748.71698</td>\n",
- " <td>2249.27417</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean_nox_emissions_10km_year2000</th>\n",
- " <td>101.446434</td>\n",
- " <td>114.493874</td>\n",
- " <td>107.847565</td>\n",
- " <td>0.13728</td>\n",
- " <td>28.190823</td>\n",
- " <td>253501.109375</td>\n",
- " <td>622.21759</td>\n",
- " <td>18234.947266</td>\n",
- " <td>1864.907349</td>\n",
- " <td>17115.994141</td>\n",
- " <td>...</td>\n",
- " <td>192.829391</td>\n",
- " <td>3962.57959</td>\n",
- " <td>369.921936</td>\n",
- " <td>40483.15625</td>\n",
- " <td>34869.03125</td>\n",
- " <td>6456.148438</td>\n",
- " <td>37354.527344</td>\n",
- " <td>21093.828125</td>\n",
- " <td>1135.177734</td>\n",
- " <td>2862.125488</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>area_code</th>\n",
- " <td>CPT134S00</td>\n",
- " <td>CH0001G</td>\n",
- " <td>WLG</td>\n",
- " <td>CGO540S00</td>\n",
- " <td>SMO514S00</td>\n",
- " <td>jp13101010</td>\n",
- " <td>AR0002G</td>\n",
- " <td>RSA024</td>\n",
- " <td>RSA005</td>\n",
- " <td>MX_PED</td>\n",
- " <td>...</td>\n",
- " <td>AT0002R</td>\n",
- " <td>DEHE160</td>\n",
- " <td>JP0004C</td>\n",
- " <td>SE0003A</td>\n",
- " <td>KOR221183</td>\n",
- " <td>KOR238363</td>\n",
- " <td>1060A</td>\n",
- " <td>1195A</td>\n",
- " <td>18-175-0001</td>\n",
- " <td>18-127-0903</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>type_of_area</th>\n",
- " <td>rural</td>\n",
- " <td>unknown</td>\n",
- " <td>rural</td>\n",
- " <td>rural</td>\n",
- " <td>rural</td>\n",
- " <td>urban</td>\n",
- " <td>rural</td>\n",
- " <td>urban</td>\n",
- " <td>suburban</td>\n",
- " <td>urban</td>\n",
- " <td>...</td>\n",
- " <td>unknown</td>\n",
- " <td>urban</td>\n",
- " <td>rural</td>\n",
- " <td>unknown</td>\n",
- " <td>urban</td>\n",
- " <td>urban</td>\n",
- " <td>unknown</td>\n",
- " <td>unknown</td>\n",
- " <td>suburban</td>\n",
- " <td>suburban</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>22 rows × 22378 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- "lat -34.350000 46.547500 \\\n",
- "lon 18.480000 7.985000 \n",
- "altitude 230.0 3578.0 \n",
- "mean_topography_srtm_alt_90m_year1994 112.0 3466.0 \n",
- "mean_topography_srtm_alt_1km_year1994 84.784127 3354.26945 \n",
- "max_topography_srtm_relative_alt_5km_year1994 140.0 603.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 -112.0 -2341.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -68.713116 -2873.715416 \n",
- "climatic_zone_year2016 6.0 7.0 \n",
- "distance_to_major_road_year2020 24.235018 6197.879799 \n",
- "mean_stable_nightlights_1km_year2013 0.0 0.0 \n",
- "mean_stable_nightlights_5km_year2013 0.0 2.082707 \n",
- "max_stable_nightlights_25km_year2013 50.0 52.0 \n",
- "max_stable_nightlights_25km_year1992 40.0 42.0 \n",
- "mean_population_density_250m_year2015 1.339095 0.0 \n",
- "mean_population_density_5km_year2015 0.652963 1.64468 \n",
- "max_population_density_25km_year2015 5356.886333 3845.294377 \n",
- "mean_population_density_250m_year1990 297.652617 0.0 \n",
- "mean_population_density_5km_year1990 -1.744407 1.979643 \n",
- "max_population_density_25km_year1990 2700.758359 3047.447108 \n",
- "mean_nox_emissions_10km_year2015 148.964264 72.823387 \n",
- "mean_nox_emissions_10km_year2000 101.446434 114.493874 \n",
- "area_code CPT134S00 CH0001G \n",
- "type_of_area rural unknown \n",
- "\n",
- "lat 36.280000 -40.683119 \\\n",
- "lon 100.900000 144.689939 \n",
- "altitude 3810.0 94.0 \n",
- "mean_topography_srtm_alt_90m_year1994 3730.0 53.0 \n",
- "mean_topography_srtm_alt_1km_year1994 3697.235165 54.430556 \n",
- "max_topography_srtm_relative_alt_5km_year1994 84.0 57.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 -752.0 0.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -3560.781547 -25.259737 \n",
- "climatic_zone_year2016 8.0 5.0 \n",
- "distance_to_major_road_year2020 3595.14835 5230.684142 \n",
- "mean_stable_nightlights_1km_year2013 0.0 0.0 \n",
- "mean_stable_nightlights_5km_year2013 0.0 0.0 \n",
- "max_stable_nightlights_25km_year2013 54.0 0.0 \n",
- "max_stable_nightlights_25km_year1992 23.0 0.0 \n",
- "mean_population_density_250m_year2015 0.0 0.0 \n",
- "mean_population_density_5km_year2015 0.0 0.39804 \n",
- "max_population_density_25km_year2015 4960.909885 230.173008 \n",
- "mean_population_density_250m_year1990 0.0 0.0 \n",
- "mean_population_density_5km_year1990 0.0 0.343217 \n",
- "max_population_density_25km_year1990 5311.749164 260.081728 \n",
- "mean_nox_emissions_10km_year2015 192.781525 0.084431 \n",
- "mean_nox_emissions_10km_year2000 107.847565 0.13728 \n",
- "area_code WLG CGO540S00 \n",
- "type_of_area rural rural \n",
- "\n",
- "lat -14.247000 35.692200 \\\n",
- "lon -170.564000 139.768900 \n",
- "altitude 42.0 2.0 \n",
- "mean_topography_srtm_alt_90m_year1994 61.0 21.0 \n",
- "mean_topography_srtm_alt_1km_year1994 31.087379 20.059341 \n",
- "max_topography_srtm_relative_alt_5km_year1994 253.0 53.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 0.0 0.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 7.074067 -9.352232 \n",
- "climatic_zone_year2016 2.0 5.0 \n",
- "distance_to_major_road_year2020 511.406318 34.213688 \n",
- "mean_stable_nightlights_1km_year2013 8.0 63.0 \n",
- "mean_stable_nightlights_5km_year2013 4.041237 63.0 \n",
- "max_stable_nightlights_25km_year2013 47.0 63.0 \n",
- "max_stable_nightlights_25km_year1992 32.0 63.0 \n",
- "mean_population_density_250m_year2015 0.0 4910.184053 \n",
- "mean_population_density_5km_year2015 259.638586 13827.48708 \n",
- "max_population_density_25km_year2015 2648.221272 22738.63592 \n",
- "mean_population_density_250m_year1990 0.0 4062.595509 \n",
- "mean_population_density_5km_year1990 122.390694 10911.150329 \n",
- "max_population_density_25km_year1990 4216.752589 17796.726371 \n",
- "mean_nox_emissions_10km_year2015 34.131149 132864.84375 \n",
- "mean_nox_emissions_10km_year2000 28.190823 253501.109375 \n",
- "area_code SMO514S00 jp13101010 \n",
- "type_of_area rural urban \n",
- "\n",
- "lat -54.848400 -26.230594 \\\n",
- "lon -68.310700 28.020442 \n",
- "altitude 18.0 1773.0 \n",
- "mean_topography_srtm_alt_90m_year1994 3.0 1696.0 \n",
- "mean_topography_srtm_alt_1km_year1994 20.369295 1699.634568 \n",
- "max_topography_srtm_relative_alt_5km_year1994 270.0 102.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 0.0 -62.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 31.423392 -1668.003299 \n",
- "climatic_zone_year2016 8.0 6.0 \n",
- "distance_to_major_road_year2020 1043.890433 96.586932 \n",
- "mean_stable_nightlights_1km_year2013 14.0 63.0 \n",
- "mean_stable_nightlights_5km_year2013 18.677019 62.951456 \n",
- "max_stable_nightlights_25km_year2013 62.0 63.0 \n",
- "max_stable_nightlights_25km_year1992 61.0 63.0 \n",
- "mean_population_density_250m_year2015 0.0 683.74621 \n",
- "mean_population_density_5km_year2015 1230.179145 3772.176769 \n",
- "max_population_density_25km_year2015 13383.638228 59504.144856 \n",
- "mean_population_density_250m_year1990 0.0 295.306869 \n",
- "mean_population_density_5km_year1990 661.777481 1628.943119 \n",
- "max_population_density_25km_year1990 7554.773868 25683.872557 \n",
- "mean_nox_emissions_10km_year2015 632.638062 14969.367188 \n",
- "mean_nox_emissions_10km_year2000 622.21759 18234.947266 \n",
- "area_code AR0002G RSA024 \n",
- "type_of_area rural urban \n",
- "\n",
- "lat -25.722592 19.325150 \\\n",
- "lon 28.420414 -99.204100 \n",
- "altitude 1754.0 2326.0 \n",
- "mean_topography_srtm_alt_90m_year1994 1333.0 2349.0 \n",
- "mean_topography_srtm_alt_1km_year1994 1332.74938 2345.320413 \n",
- "max_topography_srtm_relative_alt_5km_year1994 225.0 350.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 -65.0 -101.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -1281.138149 -2262.805486 \n",
- "climatic_zone_year2016 6.0 6.0 \n",
- "distance_to_major_road_year2020 85.449787 238.581076 \n",
- "mean_stable_nightlights_1km_year2013 56.0 63.0 \n",
- "mean_stable_nightlights_5km_year2013 43.861386 62.979381 \n",
- "max_stable_nightlights_25km_year2013 63.0 63.0 \n",
- "max_stable_nightlights_25km_year1992 63.0 63.0 \n",
- "mean_population_density_250m_year2015 23970.364846 2455.663829 \n",
- "mean_population_density_5km_year2015 3844.87098 9133.435742 \n",
- "max_population_density_25km_year2015 27545.669643 37332.160881 \n",
- "mean_population_density_250m_year1990 10017.486937 1740.817741 \n",
- "mean_population_density_5km_year1990 1701.044226 8097.352945 \n",
- "max_population_density_25km_year1990 12208.812374 40316.710058 \n",
- "mean_nox_emissions_10km_year2015 1623.259521 18833.537109 \n",
- "mean_nox_emissions_10km_year2000 1864.907349 17115.994141 \n",
- "area_code RSA005 MX_PED \n",
- "type_of_area suburban urban \n",
- "\n",
- "lat ... 47.766666 \\\n",
- "lon ... 16.766666 \n",
- "altitude ... 117.0 \n",
- "mean_topography_srtm_alt_90m_year1994 ... 113.0 \n",
- "mean_topography_srtm_alt_1km_year1994 ... 113.502773 \n",
- "max_topography_srtm_relative_alt_5km_year1994 ... 10.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 ... -8.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 ... -110.112831 \n",
- "climatic_zone_year2016 ... 6.0 \n",
- "distance_to_major_road_year2020 ... 197.163512 \n",
- "mean_stable_nightlights_1km_year2013 ... 8.0 \n",
- "mean_stable_nightlights_5km_year2013 ... 5.402878 \n",
- "max_stable_nightlights_25km_year2013 ... 57.0 \n",
- "max_stable_nightlights_25km_year1992 ... 41.0 \n",
- "mean_population_density_250m_year2015 ... 86.317243 \n",
- "mean_population_density_5km_year2015 ... 31.180866 \n",
- "max_population_density_25km_year2015 ... 5965.60968 \n",
- "mean_population_density_250m_year1990 ... 3.767978 \n",
- "mean_population_density_5km_year1990 ... 36.922714 \n",
- "max_population_density_25km_year1990 ... 4922.215553 \n",
- "mean_nox_emissions_10km_year2015 ... 158.663528 \n",
- "mean_nox_emissions_10km_year2000 ... 192.829391 \n",
- "area_code ... AT0002R \n",
- "type_of_area ... unknown \n",
- "\n",
- "lat 49.964994 32.750000 \\\n",
- "lon 8.565859 128.680000 \n",
- "altitude 99.0 500.0 \n",
- "mean_topography_srtm_alt_90m_year1994 110.0 73.0 \n",
- "mean_topography_srtm_alt_1km_year1994 104.123894 65.820513 \n",
- "max_topography_srtm_relative_alt_5km_year1994 28.0 315.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 -23.0 0.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -102.475289 -0.688234 \n",
- "climatic_zone_year2016 6.0 5.0 \n",
- "distance_to_major_road_year2020 338.208048 505.19027 \n",
- "mean_stable_nightlights_1km_year2013 31.0 6.0 \n",
- "mean_stable_nightlights_5km_year2013 33.524476 2.198198 \n",
- "max_stable_nightlights_25km_year2013 63.0 41.0 \n",
- "max_stable_nightlights_25km_year1992 63.0 32.0 \n",
- "mean_population_density_250m_year2015 991.754621 20.728297 \n",
- "mean_population_density_5km_year2015 494.029262 32.934671 \n",
- "max_population_density_25km_year2015 6787.482117 2050.823651 \n",
- "mean_population_density_250m_year1990 747.184087 18.049066 \n",
- "mean_population_density_5km_year1990 452.74777 37.191118 \n",
- "max_population_density_25km_year1990 7162.790053 2733.137455 \n",
- "mean_nox_emissions_10km_year2015 2697.373291 194.632599 \n",
- "mean_nox_emissions_10km_year2000 3962.57959 369.921936 \n",
- "area_code DEHE160 JP0004C \n",
- "type_of_area urban rural \n",
- "\n",
- "lat 59.317200 35.233400 \\\n",
- "lon 18.048900 129.010200 \n",
- "altitude 24.0 12.0 \n",
- "mean_topography_srtm_alt_90m_year1994 43.0 12.0 \n",
- "mean_topography_srtm_alt_1km_year1994 25.151899 21.232662 \n",
- "max_topography_srtm_relative_alt_5km_year1994 36.0 611.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 0.0 0.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -28.694404 149.07072 \n",
- "climatic_zone_year2016 8.0 5.0 \n",
- "distance_to_major_road_year2020 0.008772 37.155775 \n",
- "mean_stable_nightlights_1km_year2013 63.0 63.0 \n",
- "mean_stable_nightlights_5km_year2013 63.0 61.486957 \n",
- "max_stable_nightlights_25km_year2013 63.0 63.0 \n",
- "max_stable_nightlights_25km_year1992 63.0 61.0 \n",
- "mean_population_density_250m_year2015 92283.564398 278516.909439 \n",
- "mean_population_density_5km_year2015 6697.156876 4361.633324 \n",
- "max_population_density_25km_year2015 23707.527229 30320.882859 \n",
- "mean_population_density_250m_year1990 60448.397686 260331.747506 \n",
- "mean_population_density_5km_year1990 4467.052098 3816.279499 \n",
- "max_population_density_25km_year1990 16324.082429 30455.515258 \n",
- "mean_nox_emissions_10km_year2015 27311.535156 27003.199219 \n",
- "mean_nox_emissions_10km_year2000 40483.15625 34869.03125 \n",
- "area_code SE0003A KOR221183 \n",
- "type_of_area unknown urban \n",
- "\n",
- "lat 35.310600 40.768800 \\\n",
- "lon 128.987200 114.903200 \n",
- "altitude 4.0 728.0 \n",
- "mean_topography_srtm_alt_90m_year1994 4.0 734.0 \n",
- "mean_topography_srtm_alt_1km_year1994 34.868009 734.194969 \n",
- "max_topography_srtm_relative_alt_5km_year1994 497.0 398.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 0.0 -59.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 106.34765 -654.514377 \n",
- "climatic_zone_year2016 5.0 8.0 \n",
- "distance_to_major_road_year2020 23.089269 34.57441 \n",
- "mean_stable_nightlights_1km_year2013 61.0 63.0 \n",
- "mean_stable_nightlights_5km_year2013 52.678261 54.017094 \n",
- "max_stable_nightlights_25km_year2013 63.0 63.0 \n",
- "max_stable_nightlights_25km_year1992 61.0 62.0 \n",
- "mean_population_density_250m_year2015 27525.825639 78374.101361 \n",
- "mean_population_density_5km_year2015 1192.047095 3462.646015 \n",
- "max_population_density_25km_year2015 30349.793048 43562.550513 \n",
- "mean_population_density_250m_year1990 21824.192189 74094.474606 \n",
- "mean_population_density_5km_year1990 556.038768 3156.993358 \n",
- "max_population_density_25km_year1990 30484.553816 39192.710089 \n",
- "mean_nox_emissions_10km_year2015 5525.850098 75536.96875 \n",
- "mean_nox_emissions_10km_year2000 6456.148438 37354.527344 \n",
- "area_code KOR238363 1060A \n",
- "type_of_area urban unknown \n",
- "\n",
- "lat 31.684200 38.609769 \\\n",
- "lon 120.288000 -86.082006 \n",
- "altitude 7.0 226.1 \n",
- "mean_topography_srtm_alt_90m_year1994 7.0 225.0 \n",
- "mean_topography_srtm_alt_1km_year1994 5.645688 238.473233 \n",
- "max_topography_srtm_relative_alt_5km_year1994 33.0 73.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 0.0 -23.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -4.160356 -211.419405 \n",
- "climatic_zone_year2016 5.0 5.0 \n",
- "distance_to_major_road_year2020 95.008468 634.926265 \n",
- "mean_stable_nightlights_1km_year2013 62.0 43.0 \n",
- "mean_stable_nightlights_5km_year2013 60.288288 21.913043 \n",
- "max_stable_nightlights_25km_year2013 63.0 54.0 \n",
- "max_stable_nightlights_25km_year1992 59.0 53.0 \n",
- "mean_population_density_250m_year2015 53906.583742 402.390048 \n",
- "mean_population_density_5km_year2015 2220.836806 102.342802 \n",
- "max_population_density_25km_year2015 90470.580623 971.682298 \n",
- "mean_population_density_250m_year1990 72672.567509 472.818732 \n",
- "mean_population_density_5km_year1990 1236.589719 92.550635 \n",
- "max_population_density_25km_year1990 61069.93486 887.06403 \n",
- "mean_nox_emissions_10km_year2015 42011.578125 748.71698 \n",
- "mean_nox_emissions_10km_year2000 21093.828125 1135.177734 \n",
- "area_code 1195A 18-175-0001 \n",
- "type_of_area unknown suburban \n",
- "\n",
- "lat 41.614685 \n",
- "lon -87.124560 \n",
- "altitude 192.0 \n",
- "mean_topography_srtm_alt_90m_year1994 196.0 \n",
- "mean_topography_srtm_alt_1km_year1994 195.474012 \n",
- "max_topography_srtm_relative_alt_5km_year1994 28.0 \n",
- "min_topography_srtm_relative_alt_5km_year1994 -26.0 \n",
- "stddev_topography_srtm_relative_alt_5km_year1994 -187.203666 \n",
- "climatic_zone_year2016 6.0 \n",
- "distance_to_major_road_year2020 106.808314 \n",
- "mean_stable_nightlights_1km_year2013 59.0 \n",
- "mean_stable_nightlights_5km_year2013 54.239669 \n",
- "max_stable_nightlights_25km_year2013 63.0 \n",
- "max_stable_nightlights_25km_year1992 63.0 \n",
- "mean_population_density_250m_year2015 1728.186332 \n",
- "mean_population_density_5km_year2015 244.667348 \n",
- "max_population_density_25km_year2015 2829.102999 \n",
- "mean_population_density_250m_year1990 1225.107014 \n",
- "mean_population_density_5km_year1990 196.181747 \n",
- "max_population_density_25km_year1990 2336.722565 \n",
- "mean_nox_emissions_10km_year2015 2249.27417 \n",
- "mean_nox_emissions_10km_year2000 2862.125488 \n",
- "area_code 18-127-0903 \n",
- "type_of_area suburban \n",
- "\n",
- "[22 rows x 22378 columns]"
- ]
- },
- "execution_count": 16,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "dataset.T"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
- "id": "9d0305e1-6e18-49a1-ac9f-5386c0913de8",
- "metadata": {},
+ "outputs": [],
+ "source": [
+ "#dataset.T"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "id": "9d0305e1-6e18-49a1-ac9f-5386c0913de8",
+ "metadata": {},
"outputs": [
{
"data": {
@@ -1840,7 +934,7 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 16,
"id": "5aa52c01-bae2-40f5-a03e-eb1d9d7588de",
"metadata": {
"tags": []
@@ -1850,779 +944,166 @@
"def train_clustering_model(df, num_clusters:int=3, use_pca:bool=False, model='kmeans'):\n",
" train_model={'pca':None}\n",
" if use_pca:\n",
- " pca_ = PCA(n_components=0.99, random_state=42)\n",
+ " pca_ = PCA(n_components=0.95, random_state=42)\n",
" print(f\"Original shape: {df.shape}\")\n",
" pca = pca_.fit(df)\n",
" df_pca = pca.transform(df)\n",
" print(f\"Shape after PCA: {df_pca.shape}\")\n",
" df=df_pca\n",
" train_model['pca']=pca\n",
- " model=model.lower()\n",
- " if model=='kmeans':\n",
- " clustering_model= KMeans(n_clusters=num_clusters, n_init=300, random_state=42)\n",
- " else:\n",
- " raise ValueError('invalid model, Implement another method if needed')\n",
- " clustering_model.fit(df)\n",
- " train_model['model']=clustering_model\n",
- " return train_model\n",
- "\n",
- "def kmeans_predict(df, kmeans):\n",
- " kmean=kmeans['kmean']\n",
- " pca = kmeans['pca']\n",
- " if pca is not None:\n",
- " df = pca.transform(df)\n",
- " y_pred = kmean.predict(df)\n",
- " return y_pred\n",
- "\n",
- "def visualized(df, labels_pred, labels_truth=None, centroid=None, num_clusters=3):\n",
- " print(\"Visualization...\")\n",
- " cluster_labels = ['cluster_' + str(i) for i in range(num_clusters)]\n",
- " tsne = TSNE(n_components=2,perplexity=50,n_iter=2000, init='pca', learning_rate=200, random_state=42)\n",
- " df_tsne = tsne.fit_transform(df)\n",
- " fig = plt.figure(figsize=(15,6))\n",
- " ax1 = fig.add_subplot(121)\n",
- " scatter = ax1.scatter(df_tsne[:,0], df_tsne[:,1], c=labels_pred, cmap='Set1')\n",
- " ax1.legend(handles=scatter.legend_elements()[0], labels=cluster_labels, loc=\"best\", title=\"Clusters\")\n",
- " ax1.set_title(f'2D visualization of different Kmeans clusters: {num_clusters} clusters.')\n",
- " if labels_truth is not None:\n",
- " ax2 = fig.add_subplot(122)\n",
- " scatter = ax2.scatter(df_tsne[:,0],df_tsne[:,1], c=labels_truth, cmap='Set1')\n",
- " ax2.legend(handles=scatter.legend_elements()[0], labels=cluster_labels, loc=\"best\", title=\"Clusters\")\n",
- " ax2.set_title(f'2D visualization of truth clusters')\n",
- " plt.subplots_adjust(wspace=0.3)\n",
- " plt.show()\n",
- " return\n",
- "\n",
- "def ari_nmi_clustering(y_pred, y_truth):\n",
- " ari = adjusted_mutual_info_score(y_pred, y_truth)\n",
- " nmi = normalized_mutual_info_score(y_pred, y_truth)\n",
- " return ari, nmi"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 19,
- "id": "18d71fc1-7a34-4ddc-a646-661b235b91f0",
- "metadata": {
- "tags": []
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "number of data point: (22378, 22)\n",
- "number of known station: (12408, 22)\n",
- "number of unknown station: (9970, 22)\n",
- "test shape: (1000, 22)\n"
- ]
- }
- ],
- "source": [
- "labelled_data = dataset[~(dataset['type_of_area']=='unknown')]\n",
- "unlabelled_data = dataset[(dataset['type_of_area']=='unknown')]\n",
- "df_train, df_test = train_test_split(labelled_data, test_size=1000, shuffle=False, random_state=42)\n",
- "df_test_idx = list(df_test.index)\n",
- "df_train_idx = list(df_train.index)\n",
- "un_labeled_idx = list(unlabelled_data.index)\n",
- "print('number of data point: ', dataset.shape)\n",
- "print(\"number of known station: \", labelled_data.shape)\n",
- "print(\"number of unknown station: \", unlabelled_data.shape)\n",
- "print('test shape: ', df_test.shape)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 20,
- "id": "ee9ab82e-e32a-4b70-bf34-c189dabbec4d",
- "metadata": {
- "tags": []
- },
- "outputs": [],
- "source": [
- "test_data_path = os.path.abspath(os.path.join(\"data\", \"hand_labeled_test_data.csv\"))\n",
- "test_data = pd.read_csv(test_data_path)\n",
- "test_data.set_index(['lat', 'lon'], inplace=True)\n",
- "test_data = test_data[~(test_data['type_of_are_toar']=='unknown')]\n",
- "test_indeces = list(test_data.index)\n",
- "test_indeces = [idx for idx in test_indeces if idx in list(dataset.index)]\n",
- "test_data = test_data.loc[test_indeces]"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "54d9c49a-b81b-4382-8567-6bd1c6871a7e",
- "metadata": {},
- "source": [
- "#### Prepare train and test data and train clustering algorithm"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 21,
- "id": "bfed2136-c5af-4612-9820-e94ccde98518",
- "metadata": {
- "tags": []
- },
- "outputs": [],
- "source": [
- "# we first selecte the most recent infos and analyse the correlation between variable\n",
- "selected_colunms_0 = [\n",
- " 'altitude', \n",
- " 'mean_topography_srtm_alt_90m_year1994', \n",
- " 'mean_topography_srtm_alt_1km_year1994', \n",
- " 'max_topography_srtm_relative_alt_5km_year1994', \n",
- " 'min_topography_srtm_relative_alt_5km_year1994',\n",
- " 'stddev_topography_srtm_relative_alt_5km_year1994', \n",
- " 'climatic_zone_year2016', \n",
- " 'distance_to_major_road_year2020', \n",
- " 'mean_stable_nightlights_1km_year2013', \n",
- " 'mean_stable_nightlights_5km_year2013', \n",
- " 'max_stable_nightlights_25km_year2013', \n",
- " 'max_stable_nightlights_25km_year1992', \n",
- " 'mean_population_density_250m_year2015', \n",
- " 'mean_population_density_5km_year2015', \n",
- " 'max_population_density_25km_year2015',\n",
- " 'mean_population_density_250m_year1990', \n",
- " 'mean_population_density_5km_year1990', \n",
- " 'max_population_density_25km_year1990', \n",
- " 'mean_nox_emissions_10km_year2015', \n",
- " 'mean_nox_emissions_10km_year2000',\n",
- "]\n",
- "dataset_0 = feature_engineering_selection(dataset, selected_columns=selected_colunms_0, scaling='minmax')\n",
- "df_train_0 = dataset_0[~(dataset_0.index.isin(df_test_idx))]\n",
- "df_test_0 = dataset_0[(dataset_0.index.isin(df_test_idx))]\n",
- "test_data_0 = dataset_0.loc[test_indeces]\n",
- "#display(dataset_0.head())\n",
- "#plot_correlation(df_train_0)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 22,
- "id": "47ab7414-d151-4384-a7f1-fe9e6f5de4b3",
- "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></th>\n",
- " <th>altitude</th>\n",
- " <th>mean_topography_srtm_alt_90m_year1994</th>\n",
- " <th>mean_topography_srtm_alt_1km_year1994</th>\n",
- " <th>max_topography_srtm_relative_alt_5km_year1994</th>\n",
- " <th>min_topography_srtm_relative_alt_5km_year1994</th>\n",
- " <th>stddev_topography_srtm_relative_alt_5km_year1994</th>\n",
- " <th>climatic_zone_year2016</th>\n",
- " <th>distance_to_major_road_year2020</th>\n",
- " <th>mean_stable_nightlights_1km_year2013</th>\n",
- " <th>mean_stable_nightlights_5km_year2013</th>\n",
- " <th>max_stable_nightlights_25km_year2013</th>\n",
- " <th>max_stable_nightlights_25km_year1992</th>\n",
- " <th>mean_population_density_250m_year2015</th>\n",
- " <th>mean_population_density_5km_year2015</th>\n",
- " <th>max_population_density_25km_year2015</th>\n",
- " <th>mean_population_density_250m_year1990</th>\n",
- " <th>mean_population_density_5km_year1990</th>\n",
- " <th>max_population_density_25km_year1990</th>\n",
- " <th>mean_nox_emissions_10km_year2015</th>\n",
- " <th>mean_nox_emissions_10km_year2000</th>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>lat</th>\n",
- " <th>lon</th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " <th></th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>-34.350000</th>\n",
- " <th>18.480000</th>\n",
- " <td>0.081592</td>\n",
- " <td>0.096158</td>\n",
- " <td>0.087507</td>\n",
- " <td>0.052006</td>\n",
- " <td>0.952157</td>\n",
- " <td>0.799935</td>\n",
- " <td>0.500000</td>\n",
- " <td>9.811367e-04</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.793651</td>\n",
- " <td>0.634921</td>\n",
- " <td>0.000005</td>\n",
- " <td>0.000012</td>\n",
- " <td>0.018819</td>\n",
- " <td>0.001143</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.011706</td>\n",
- " <td>7.704299e-05</td>\n",
- " <td>9.664181e-05</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>46.547500</th>\n",
- " <th>7.985000</th>\n",
- " <td>0.521482</td>\n",
- " <td>0.737335</td>\n",
- " <td>0.721275</td>\n",
- " <td>0.223997</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.323021</td>\n",
- " <td>0.583333</td>\n",
- " <td>2.509335e-01</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.033059</td>\n",
- " <td>0.825397</td>\n",
- " <td>0.666667</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000030</td>\n",
- " <td>0.013509</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000071</td>\n",
- " <td>0.013209</td>\n",
- " <td>3.766361e-05</td>\n",
- " <td>1.090713e-04</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>36.280000</th>\n",
- " <th>100.900000</th>\n",
- " <td>0.551964</td>\n",
- " <td>0.787803</td>\n",
- " <td>0.787757</td>\n",
- " <td>0.031204</td>\n",
- " <td>0.678770</td>\n",
- " <td>0.206204</td>\n",
- " <td>0.666667</td>\n",
- " <td>1.455567e-01</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.857143</td>\n",
- " <td>0.365079</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.017428</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000033</td>\n",
- " <td>0.023024</td>\n",
- " <td>9.970488e-05</td>\n",
- " <td>1.027398e-04</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>-40.683119</th>\n",
- " <th>144.689939</th>\n",
- " <td>0.063724</td>\n",
- " <td>0.084879</td>\n",
- " <td>0.081624</td>\n",
- " <td>0.021174</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.807323</td>\n",
- " <td>0.416667</td>\n",
- " <td>2.117746e-01</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000007</td>\n",
- " <td>0.000809</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000040</td>\n",
- " <td>0.001127</td>\n",
- " <td>4.366696e-08</td>\n",
- " <td>1.307782e-07</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>-14.247000</th>\n",
- " <th>-170.564000</th>\n",
- " <td>0.056891</td>\n",
- " <td>0.086408</td>\n",
- " <td>0.077099</td>\n",
- " <td>0.093982</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.812821</td>\n",
- " <td>0.166667</td>\n",
- " <td>2.070524e-02</td>\n",
- " <td>0.126984</td>\n",
- " <td>0.064147</td>\n",
- " <td>0.746032</td>\n",
- " <td>0.507937</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.004726</td>\n",
- " <td>0.009303</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.002356</td>\n",
- " <td>0.018277</td>\n",
- " <td>1.765233e-05</td>\n",
- " <td>2.685567e-05</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>...</th>\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",
- " </tr>\n",
- " <tr>\n",
- " <th>-34.353480</th>\n",
- " <th>18.489680</th>\n",
- " <td>0.081592</td>\n",
- " <td>0.091761</td>\n",
- " <td>0.085071</td>\n",
- " <td>0.060550</td>\n",
- " <td>0.961982</td>\n",
- " <td>0.804237</td>\n",
- " <td>0.500000</td>\n",
- " <td>2.604173e-02</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.777778</td>\n",
- " <td>0.634921</td>\n",
- " <td>0.000005</td>\n",
- " <td>0.000012</td>\n",
- " <td>0.015572</td>\n",
- " <td>0.000002</td>\n",
- " <td>0.000039</td>\n",
- " <td>0.009673</td>\n",
- " <td>7.704299e-05</td>\n",
- " <td>9.664181e-05</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>47.766666</th>\n",
- " <th>16.766666</th>\n",
- " <td>0.066745</td>\n",
- " <td>0.096349</td>\n",
- " <td>0.093074</td>\n",
- " <td>0.003715</td>\n",
- " <td>0.996583</td>\n",
- " <td>0.792897</td>\n",
- " <td>0.500000</td>\n",
- " <td>7.982492e-03</td>\n",
- " <td>0.126984</td>\n",
- " <td>0.085760</td>\n",
- " <td>0.904762</td>\n",
- " <td>0.650794</td>\n",
- " <td>0.000310</td>\n",
- " <td>0.000568</td>\n",
- " <td>0.020957</td>\n",
- " <td>0.000014</td>\n",
- " <td>0.000734</td>\n",
- " <td>0.021335</td>\n",
- " <td>8.205936e-05</td>\n",
- " <td>1.836968e-04</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>59.317200</th>\n",
- " <th>18.048900</th>\n",
- " <td>0.054526</td>\n",
- " <td>0.082967</td>\n",
- " <td>0.075948</td>\n",
- " <td>0.013373</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.806740</td>\n",
- " <td>0.666667</td>\n",
- " <td>2.887682e-07</td>\n",
- " <td>1.000000</td>\n",
- " <td>1.000000</td>\n",
- " <td>1.000000</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.331339</td>\n",
- " <td>0.121900</td>\n",
- " <td>0.083285</td>\n",
- " <td>0.232198</td>\n",
- " <td>0.084799</td>\n",
- " <td>0.070756</td>\n",
- " <td>1.412528e-02</td>\n",
- " <td>3.856583e-02</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>40.768800</th>\n",
- " <th>114.903200</th>\n",
- " <td>0.147024</td>\n",
- " <td>0.215064</td>\n",
- " <td>0.213391</td>\n",
- " <td>0.147845</td>\n",
- " <td>0.974797</td>\n",
- " <td>0.700336</td>\n",
- " <td>0.666667</td>\n",
- " <td>1.399748e-03</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.857414</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.984127</td>\n",
- " <td>0.281398</td>\n",
- " <td>0.063026</td>\n",
- " <td>0.153036</td>\n",
- " <td>0.284616</td>\n",
- " <td>0.059940</td>\n",
- " <td>0.169880</td>\n",
- " <td>3.906705e-02</td>\n",
- " <td>3.558537e-02</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>31.684200</th>\n",
- " <th>120.288000</th>\n",
- " <td>0.052293</td>\n",
- " <td>0.076085</td>\n",
- " <td>0.072167</td>\n",
- " <td>0.012259</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.810911</td>\n",
- " <td>0.416667</td>\n",
- " <td>3.846541e-03</td>\n",
- " <td>0.984127</td>\n",
- " <td>0.956957</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.936508</td>\n",
- " <td>0.193549</td>\n",
- " <td>0.040423</td>\n",
- " <td>0.317825</td>\n",
- " <td>0.279154</td>\n",
- " <td>0.023498</td>\n",
- " <td>0.264706</td>\n",
- " <td>2.172801e-02</td>\n",
- " <td>2.009480e-02</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>21378 rows × 20 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- " altitude mean_topography_srtm_alt_90m_year1994 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.081592 0.096158 \n",
- " 46.547500 7.985000 0.521482 0.737335 \n",
- " 36.280000 100.900000 0.551964 0.787803 \n",
- "-40.683119 144.689939 0.063724 0.084879 \n",
- "-14.247000 -170.564000 0.056891 0.086408 \n",
- "... ... ... \n",
- "-34.353480 18.489680 0.081592 0.091761 \n",
- " 47.766666 16.766666 0.066745 0.096349 \n",
- " 59.317200 18.048900 0.054526 0.082967 \n",
- " 40.768800 114.903200 0.147024 0.215064 \n",
- " 31.684200 120.288000 0.052293 0.076085 \n",
- "\n",
- " mean_topography_srtm_alt_1km_year1994 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.087507 \n",
- " 46.547500 7.985000 0.721275 \n",
- " 36.280000 100.900000 0.787757 \n",
- "-40.683119 144.689939 0.081624 \n",
- "-14.247000 -170.564000 0.077099 \n",
- "... ... \n",
- "-34.353480 18.489680 0.085071 \n",
- " 47.766666 16.766666 0.093074 \n",
- " 59.317200 18.048900 0.075948 \n",
- " 40.768800 114.903200 0.213391 \n",
- " 31.684200 120.288000 0.072167 \n",
- "\n",
- " max_topography_srtm_relative_alt_5km_year1994 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.052006 \n",
- " 46.547500 7.985000 0.223997 \n",
- " 36.280000 100.900000 0.031204 \n",
- "-40.683119 144.689939 0.021174 \n",
- "-14.247000 -170.564000 0.093982 \n",
- "... ... \n",
- "-34.353480 18.489680 0.060550 \n",
- " 47.766666 16.766666 0.003715 \n",
- " 59.317200 18.048900 0.013373 \n",
- " 40.768800 114.903200 0.147845 \n",
- " 31.684200 120.288000 0.012259 \n",
- "\n",
- " min_topography_srtm_relative_alt_5km_year1994 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.952157 \n",
- " 46.547500 7.985000 0.000000 \n",
- " 36.280000 100.900000 0.678770 \n",
- "-40.683119 144.689939 1.000000 \n",
- "-14.247000 -170.564000 1.000000 \n",
- "... ... \n",
- "-34.353480 18.489680 0.961982 \n",
- " 47.766666 16.766666 0.996583 \n",
- " 59.317200 18.048900 1.000000 \n",
- " 40.768800 114.903200 0.974797 \n",
- " 31.684200 120.288000 1.000000 \n",
- "\n",
- " stddev_topography_srtm_relative_alt_5km_year1994 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.799935 \n",
- " 46.547500 7.985000 0.323021 \n",
- " 36.280000 100.900000 0.206204 \n",
- "-40.683119 144.689939 0.807323 \n",
- "-14.247000 -170.564000 0.812821 \n",
- "... ... \n",
- "-34.353480 18.489680 0.804237 \n",
- " 47.766666 16.766666 0.792897 \n",
- " 59.317200 18.048900 0.806740 \n",
- " 40.768800 114.903200 0.700336 \n",
- " 31.684200 120.288000 0.810911 \n",
- "\n",
- " climatic_zone_year2016 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.500000 \n",
- " 46.547500 7.985000 0.583333 \n",
- " 36.280000 100.900000 0.666667 \n",
- "-40.683119 144.689939 0.416667 \n",
- "-14.247000 -170.564000 0.166667 \n",
- "... ... \n",
- "-34.353480 18.489680 0.500000 \n",
- " 47.766666 16.766666 0.500000 \n",
- " 59.317200 18.048900 0.666667 \n",
- " 40.768800 114.903200 0.666667 \n",
- " 31.684200 120.288000 0.416667 \n",
- "\n",
- " distance_to_major_road_year2020 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 9.811367e-04 \n",
- " 46.547500 7.985000 2.509335e-01 \n",
- " 36.280000 100.900000 1.455567e-01 \n",
- "-40.683119 144.689939 2.117746e-01 \n",
- "-14.247000 -170.564000 2.070524e-02 \n",
- "... ... \n",
- "-34.353480 18.489680 2.604173e-02 \n",
- " 47.766666 16.766666 7.982492e-03 \n",
- " 59.317200 18.048900 2.887682e-07 \n",
- " 40.768800 114.903200 1.399748e-03 \n",
- " 31.684200 120.288000 3.846541e-03 \n",
- "\n",
- " mean_stable_nightlights_1km_year2013 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.000000 \n",
- " 46.547500 7.985000 0.000000 \n",
- " 36.280000 100.900000 0.000000 \n",
- "-40.683119 144.689939 0.000000 \n",
- "-14.247000 -170.564000 0.126984 \n",
- "... ... \n",
- "-34.353480 18.489680 0.000000 \n",
- " 47.766666 16.766666 0.126984 \n",
- " 59.317200 18.048900 1.000000 \n",
- " 40.768800 114.903200 1.000000 \n",
- " 31.684200 120.288000 0.984127 \n",
- "\n",
- " mean_stable_nightlights_5km_year2013 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.000000 \n",
- " 46.547500 7.985000 0.033059 \n",
- " 36.280000 100.900000 0.000000 \n",
- "-40.683119 144.689939 0.000000 \n",
- "-14.247000 -170.564000 0.064147 \n",
- "... ... \n",
- "-34.353480 18.489680 0.000000 \n",
- " 47.766666 16.766666 0.085760 \n",
- " 59.317200 18.048900 1.000000 \n",
- " 40.768800 114.903200 0.857414 \n",
- " 31.684200 120.288000 0.956957 \n",
- "\n",
- " max_stable_nightlights_25km_year2013 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.793651 \n",
- " 46.547500 7.985000 0.825397 \n",
- " 36.280000 100.900000 0.857143 \n",
- "-40.683119 144.689939 0.000000 \n",
- "-14.247000 -170.564000 0.746032 \n",
- "... ... \n",
- "-34.353480 18.489680 0.777778 \n",
- " 47.766666 16.766666 0.904762 \n",
- " 59.317200 18.048900 1.000000 \n",
- " 40.768800 114.903200 1.000000 \n",
- " 31.684200 120.288000 1.000000 \n",
- "\n",
- " max_stable_nightlights_25km_year1992 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.634921 \n",
- " 46.547500 7.985000 0.666667 \n",
- " 36.280000 100.900000 0.365079 \n",
- "-40.683119 144.689939 0.000000 \n",
- "-14.247000 -170.564000 0.507937 \n",
- "... ... \n",
- "-34.353480 18.489680 0.634921 \n",
- " 47.766666 16.766666 0.650794 \n",
- " 59.317200 18.048900 1.000000 \n",
- " 40.768800 114.903200 0.984127 \n",
- " 31.684200 120.288000 0.936508 \n",
- "\n",
- " mean_population_density_250m_year2015 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.000005 \n",
- " 46.547500 7.985000 0.000000 \n",
- " 36.280000 100.900000 0.000000 \n",
- "-40.683119 144.689939 0.000000 \n",
- "-14.247000 -170.564000 0.000000 \n",
- "... ... \n",
- "-34.353480 18.489680 0.000005 \n",
- " 47.766666 16.766666 0.000310 \n",
- " 59.317200 18.048900 0.331339 \n",
- " 40.768800 114.903200 0.281398 \n",
- " 31.684200 120.288000 0.193549 \n",
- "\n",
- " mean_population_density_5km_year2015 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.000012 \n",
- " 46.547500 7.985000 0.000030 \n",
- " 36.280000 100.900000 0.000000 \n",
- "-40.683119 144.689939 0.000007 \n",
- "-14.247000 -170.564000 0.004726 \n",
- "... ... \n",
- "-34.353480 18.489680 0.000012 \n",
- " 47.766666 16.766666 0.000568 \n",
- " 59.317200 18.048900 0.121900 \n",
- " 40.768800 114.903200 0.063026 \n",
- " 31.684200 120.288000 0.040423 \n",
- "\n",
- " max_population_density_25km_year2015 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.018819 \n",
- " 46.547500 7.985000 0.013509 \n",
- " 36.280000 100.900000 0.017428 \n",
- "-40.683119 144.689939 0.000809 \n",
- "-14.247000 -170.564000 0.009303 \n",
- "... ... \n",
- "-34.353480 18.489680 0.015572 \n",
- " 47.766666 16.766666 0.020957 \n",
- " 59.317200 18.048900 0.083285 \n",
- " 40.768800 114.903200 0.153036 \n",
- " 31.684200 120.288000 0.317825 \n",
- "\n",
- " mean_population_density_250m_year1990 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.001143 \n",
- " 46.547500 7.985000 0.000000 \n",
- " 36.280000 100.900000 0.000000 \n",
- "-40.683119 144.689939 0.000000 \n",
- "-14.247000 -170.564000 0.000000 \n",
- "... ... \n",
- "-34.353480 18.489680 0.000002 \n",
- " 47.766666 16.766666 0.000014 \n",
- " 59.317200 18.048900 0.232198 \n",
- " 40.768800 114.903200 0.284616 \n",
- " 31.684200 120.288000 0.279154 \n",
- "\n",
- " mean_population_density_5km_year1990 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.000000 \n",
- " 46.547500 7.985000 0.000071 \n",
- " 36.280000 100.900000 0.000033 \n",
- "-40.683119 144.689939 0.000040 \n",
- "-14.247000 -170.564000 0.002356 \n",
- "... ... \n",
- "-34.353480 18.489680 0.000039 \n",
- " 47.766666 16.766666 0.000734 \n",
- " 59.317200 18.048900 0.084799 \n",
- " 40.768800 114.903200 0.059940 \n",
- " 31.684200 120.288000 0.023498 \n",
- "\n",
- " max_population_density_25km_year1990 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 0.011706 \n",
- " 46.547500 7.985000 0.013209 \n",
- " 36.280000 100.900000 0.023024 \n",
- "-40.683119 144.689939 0.001127 \n",
- "-14.247000 -170.564000 0.018277 \n",
- "... ... \n",
- "-34.353480 18.489680 0.009673 \n",
- " 47.766666 16.766666 0.021335 \n",
- " 59.317200 18.048900 0.070756 \n",
- " 40.768800 114.903200 0.169880 \n",
- " 31.684200 120.288000 0.264706 \n",
- "\n",
- " mean_nox_emissions_10km_year2015 \\\n",
- "lat lon \n",
- "-34.350000 18.480000 7.704299e-05 \n",
- " 46.547500 7.985000 3.766361e-05 \n",
- " 36.280000 100.900000 9.970488e-05 \n",
- "-40.683119 144.689939 4.366696e-08 \n",
- "-14.247000 -170.564000 1.765233e-05 \n",
- "... ... \n",
- "-34.353480 18.489680 7.704299e-05 \n",
- " 47.766666 16.766666 8.205936e-05 \n",
- " 59.317200 18.048900 1.412528e-02 \n",
- " 40.768800 114.903200 3.906705e-02 \n",
- " 31.684200 120.288000 2.172801e-02 \n",
- "\n",
- " mean_nox_emissions_10km_year2000 \n",
- "lat lon \n",
- "-34.350000 18.480000 9.664181e-05 \n",
- " 46.547500 7.985000 1.090713e-04 \n",
- " 36.280000 100.900000 1.027398e-04 \n",
- "-40.683119 144.689939 1.307782e-07 \n",
- "-14.247000 -170.564000 2.685567e-05 \n",
- "... ... \n",
- "-34.353480 18.489680 9.664181e-05 \n",
- " 47.766666 16.766666 1.836968e-04 \n",
- " 59.317200 18.048900 3.856583e-02 \n",
- " 40.768800 114.903200 3.558537e-02 \n",
- " 31.684200 120.288000 2.009480e-02 \n",
- "\n",
- "[21378 rows x 20 columns]"
- ]
- },
- "execution_count": 22,
- "metadata": {},
- "output_type": "execute_result"
+ " model=model.lower()\n",
+ " if model=='kmeans':\n",
+ " clustering_model= KMeans(n_clusters=num_clusters, n_init=300, random_state=42)\n",
+ " else:\n",
+ " raise ValueError('invalid model, Implement another method if needed')\n",
+ " clustering_model.fit(df)\n",
+ " train_model['model']=clustering_model\n",
+ " return train_model\n",
+ "\n",
+ "def kmeans_predict(df, kmeans):\n",
+ " kmean=kmeans['kmean']\n",
+ " pca = kmeans['pca']\n",
+ " if pca is not None:\n",
+ " df = pca.transform(df)\n",
+ " y_pred = kmean.predict(df)\n",
+ " return y_pred\n",
+ "\n",
+ "def visualized(df, labels_pred, labels_truth=None, centroid=None, num_clusters=3):\n",
+ " print(\"Visualization...\")\n",
+ " cluster_labels = ['cluster_' + str(i) for i in range(num_clusters)]\n",
+ " tsne = TSNE(n_components=2,perplexity=50,n_iter=2000, init='pca', learning_rate=200, random_state=42)\n",
+ " df_tsne = tsne.fit_transform(df)\n",
+ " fig = plt.figure(figsize=(15,6))\n",
+ " ax1 = fig.add_subplot(121)\n",
+ " scatter = ax1.scatter(df_tsne[:,0], df_tsne[:,1], c=labels_pred, cmap='Set1')\n",
+ " ax1.legend(handles=scatter.legend_elements()[0], labels=cluster_labels, loc=\"best\", title=\"Clusters\")\n",
+ " ax1.set_title(f'2D visualization of different Kmeans clusters: {num_clusters} clusters.')\n",
+ " if labels_truth is not None:\n",
+ " ax2 = fig.add_subplot(122)\n",
+ " scatter = ax2.scatter(df_tsne[:,0],df_tsne[:,1], c=labels_truth, cmap='Set1')\n",
+ " ax2.legend(handles=scatter.legend_elements()[0], labels=cluster_labels, loc=\"best\", title=\"Clusters\")\n",
+ " ax2.set_title(f'2D visualization of truth clusters')\n",
+ " plt.subplots_adjust(wspace=0.3)\n",
+ " plt.show()\n",
+ " return\n",
+ "\n",
+ "def ari_nmi_clustering(y_pred, y_truth):\n",
+ " ari = adjusted_mutual_info_score(y_pred, y_truth)\n",
+ " nmi = normalized_mutual_info_score(y_pred, y_truth)\n",
+ " return ari, nmi"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "id": "18d71fc1-7a34-4ddc-a646-661b235b91f0",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "number of data point: (22378, 22)\n",
+ "number of known station: (12408, 22)\n",
+ "number of unknown station: (9970, 22)\n",
+ "test shape: (1000, 22)\n"
+ ]
}
],
"source": [
- "df_train_0"
+ "labelled_data = dataset[~(dataset['type_of_area']=='unknown')]\n",
+ "unlabelled_data = dataset[(dataset['type_of_area']=='unknown')]\n",
+ "df_train, df_test = train_test_split(labelled_data, test_size=1000, shuffle=False, random_state=42)\n",
+ "df_test_idx = list(df_test.index)\n",
+ "df_train_idx = list(df_train.index)\n",
+ "un_labeled_idx = list(unlabelled_data.index)\n",
+ "print('number of data point: ', dataset.shape)\n",
+ "print(\"number of known station: \", labelled_data.shape)\n",
+ "print(\"number of unknown station: \", unlabelled_data.shape)\n",
+ "print('test shape: ', df_test.shape)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "id": "ee9ab82e-e32a-4b70-bf34-c189dabbec4d",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "test_data_path = os.path.abspath(os.path.join(\"data\", \"hand_labeled_test_data.csv\"))\n",
+ "test_data = pd.read_csv(test_data_path)\n",
+ "test_data.set_index(['lat', 'lon'], inplace=True)\n",
+ "test_data = test_data[~(test_data['type_of_are_toar']=='unknown')]\n",
+ "test_indeces = list(test_data.index)\n",
+ "test_indeces = [idx for idx in test_indeces if idx in list(dataset.index)]\n",
+ "test_data = test_data.loc[test_indeces]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "54d9c49a-b81b-4382-8567-6bd1c6871a7e",
+ "metadata": {},
+ "source": [
+ "#### Prepare train and test data and train clustering algorithm"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "id": "bfed2136-c5af-4612-9820-e94ccde98518",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "# we first selecte the most recent infos and analyse the correlation between variable\n",
+ "selected_colunms_0 = [\n",
+ " 'altitude', \n",
+ " 'mean_topography_srtm_alt_90m_year1994', \n",
+ " 'mean_topography_srtm_alt_1km_year1994', \n",
+ " 'max_topography_srtm_relative_alt_5km_year1994', \n",
+ " 'min_topography_srtm_relative_alt_5km_year1994',\n",
+ " 'stddev_topography_srtm_relative_alt_5km_year1994', \n",
+ " 'climatic_zone_year2016', \n",
+ " 'distance_to_major_road_year2020', \n",
+ " 'mean_stable_nightlights_1km_year2013', \n",
+ " 'mean_stable_nightlights_5km_year2013', \n",
+ " 'max_stable_nightlights_25km_year2013', \n",
+ " 'max_stable_nightlights_25km_year1992', \n",
+ " 'mean_population_density_250m_year2015', \n",
+ " 'mean_population_density_5km_year2015', \n",
+ " 'max_population_density_25km_year2015',\n",
+ " 'mean_population_density_250m_year1990', \n",
+ " 'mean_population_density_5km_year1990', \n",
+ " 'max_population_density_25km_year1990', \n",
+ " 'mean_nox_emissions_10km_year2015', \n",
+ " 'mean_nox_emissions_10km_year2000',\n",
+ "]\n",
+ "dataset_0 = feature_engineering_selection(dataset, selected_columns=selected_colunms_0, scaling='minmax')\n",
+ "df_train_0 = dataset_0[~(dataset_0.index.isin(df_test_idx))]\n",
+ "df_test_0 = dataset_0[(dataset_0.index.isin(df_test_idx))]\n",
+ "test_data_0 = dataset_0.loc[test_indeces]\n",
+ "#display(dataset_0.head())\n",
+ "#plot_correlation(df_train_0)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "47ab7414-d151-4384-a7f1-fe9e6f5de4b3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#df_train_0"
]
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": 24,
"id": "a9d5427e-5485-4a75-a967-5cee7d2618ea",
"metadata": {
"tags": []
@@ -2637,12 +1118,12 @@
"id": "ba1f605a-f8f0-45dd-a09d-9d5a9b4de17d",
"metadata": {},
"source": [
- "## Testing 1:"
+ "#### Testing 1:"
]
},
{
"cell_type": "code",
- "execution_count": 24,
+ "execution_count": 36,
"id": "c8e396aa-46f5-4f77-80d3-9432af0a6c60",
"metadata": {
"tags": []
@@ -2666,7 +1147,7 @@
},
{
"cell_type": "code",
- "execution_count": 25,
+ "execution_count": 28,
"id": "d416394f-4792-43a5-8595-b56d290e43df",
"metadata": {
"tags": []
@@ -2708,7 +1189,7 @@
},
{
"cell_type": "code",
- "execution_count": 26,
+ "execution_count": 29,
"id": "9cc1c033-e9fd-4c49-a001-fe11eae49255",
"metadata": {
"collapsed": true,
@@ -2750,7 +1231,7 @@
},
{
"cell_type": "code",
- "execution_count": 27,
+ "execution_count": 30,
"id": "bad6a85f-b1fe-47d4-933f-99257d9afca6",
"metadata": {
"collapsed": true,
@@ -2805,7 +1286,7 @@
},
{
"cell_type": "code",
- "execution_count": 28,
+ "execution_count": 31,
"id": "d0f5a5c9-dba4-4e57-819b-17bc18e1bc69",
"metadata": {
"collapsed": true,
@@ -2817,7 +1298,7 @@
"outputs": [
{
"data": {
- "image/png": "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",
+ "image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
@@ -2848,7 +1329,7 @@
},
{
"cell_type": "code",
- "execution_count": 29,
+ "execution_count": 33,
"id": "c7d0e67a-beab-432e-9b13-eb5cb834add8",
"metadata": {
"tags": []
@@ -2887,7 +1368,7 @@
},
{
"cell_type": "code",
- "execution_count": 30,
+ "execution_count": 34,
"id": "f0506280-ce8b-492e-8807-9e873bd033cb",
"metadata": {},
"outputs": [],
@@ -2911,7 +1392,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 35,
"id": "c3003969-5442-4992-8839-c4483e9fd62d",
"metadata": {},
"outputs": [],
@@ -3020,7 +1501,7 @@
},
{
"cell_type": "code",
- "execution_count": 32,
+ "execution_count": 54,
"id": "2909f641-7ba5-48aa-af36-13b8f9d36951",
"metadata": {},
"outputs": [],
@@ -3064,7 +1545,7 @@
},
{
"cell_type": "code",
- "execution_count": 33,
+ "execution_count": 40,
"id": "d69bb783-8849-43da-be63-54773dbc7733",
"metadata": {
"tags": []
@@ -3088,37 +1569,7 @@
},
{
"cell_type": "code",
- "execution_count": 34,
- "id": "1148cc1f-f9ea-4888-8362-c54ed142317a",
- "metadata": {
- "tags": []
- },
- "outputs": [],
- "source": [
- "#from scipy.special import inv_boxcox\n",
- "# y_boxcox_nox_2015, lmb_func_nox_2015 = stats.boxcox(data['mean_nox_emissions_10km_year2015'].values)\n",
- "# y_boxcox_nox_2000, lmb_func_nox_2000 = stats.boxcox(data['mean_nox_emissions_10km_year2000'].values)\n",
- "# #stats.boxcox(Y_train)\n",
- "# data['mean_nox_emissions_10km_year2015'] = y_boxcox_nox_2015\n",
- "# data['mean_nox_emissions_10km_year2000'] = y_boxcox_nox_2000"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 35,
- "id": "98b25a8a-1764-453b-a2ed-a02264e74890",
- "metadata": {
- "tags": []
- },
- "outputs": [],
- "source": [
- "# df_train = labelled_data[~(labelled_data.index.isin(df_pm_test_idx))]\n",
- "# df_test = labelled_data[(labelled_data.index.isin(df_pm_test_idx))]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 36,
+ "execution_count": 42,
"id": "293de0f8-8b16-4375-bc48-d9d43f41cea4",
"metadata": {
"tags": []
@@ -3153,7 +1604,7 @@
},
{
"cell_type": "code",
- "execution_count": 37,
+ "execution_count": 43,
"id": "4df7f2e2-e5ab-473b-8745-185032e2a9f6",
"metadata": {
"tags": []
@@ -3169,7 +1620,7 @@
},
{
"cell_type": "code",
- "execution_count": 38,
+ "execution_count": 44,
"id": "d528dc81-b8b5-4ea1-8ef8-3ea04f4c2271",
"metadata": {
"tags": []
@@ -3230,7 +1681,7 @@
},
{
"cell_type": "code",
- "execution_count": 39,
+ "execution_count": 45,
"id": "d9776c21-9ea6-489f-b2cd-678da3010d1f",
"metadata": {
"tags": []
@@ -3249,7 +1700,7 @@
},
{
"cell_type": "code",
- "execution_count": 40,
+ "execution_count": 46,
"id": "38eb85e7-1fc2-474a-b415-4e220a3bac68",
"metadata": {
"tags": []
@@ -3267,21 +1718,7 @@
},
{
"cell_type": "code",
- "execution_count": 41,
- "id": "c5c5659c-171c-4ad4-8f2a-d7b4e422f20d",
- "metadata": {},
- "outputs": [],
- "source": [
- "techniques = [\n",
- " (SMOTE(random_state=42), \"SMOTE\"),\n",
- " (SMOTETomek(random_state=42), \"SMOTE+Tomek\"),\n",
- " (SMOTEENN(random_state=42), \"SMOTE+ENN\")\n",
- "]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 42,
+ "execution_count": 48,
"id": "061a82c0-3944-414b-9c61-595bf9437afb",
"metadata": {},
"outputs": [],
@@ -3293,7 +1730,7 @@
},
{
"cell_type": "code",
- "execution_count": 43,
+ "execution_count": 49,
"id": "a5f15c86-2ab0-42ac-a9a2-05c92a6a0ed3",
"metadata": {},
"outputs": [],
@@ -3305,13 +1742,13 @@
},
{
"cell_type": "code",
- "execution_count": 44,
+ "execution_count": 50,
"id": "22c7764c-2b29-4922-bc9b-0a8ddcc50445",
"metadata": {},
"outputs": [
{
"data": {
- "image/png": "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",
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqEAAAHwCAYAAACWmLaRAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABo3ElEQVR4nO3deVwU9eM/8Ncu9yGHJCJ4oBQgokYoHokGWKBlXh/TUrMU7yMT/WTfSss+fczUNDXBIzOPUktBKwUVjDwQAQ0PFAw1EEVR7nuB/f3hb+fDsAtyjrC+no9Hj9z3zHvmPcvs7Gvf854ZmVKpVIKIiIiISELyJ90AIiIiInr6MIQSERERkeQYQomIiIhIcgyhRERERCQ5hlAiIiIikhxDKBERERFJjiGUiIiIiCTHEEpEREREkmMIJSIiIiLJMYTSU2H9+vVwcnIS/lu8ePGTbhIRSktLsXnzZowcORJubm6ifXT79u1PunkEIC8vD6tXr8arr76K559/XvQ3On78+JNuHlGLpvukG0Dabdq0aYiMjBRee3t7IzAwsMY6AQEB+O2334TXHh4e2LlzZ5O1saWKiYlBWFgYzp8/j/T0dOTm5kJPTw9t2rSBs7MzXnzxRQwZMgRmZmZPuqmNbvv27cjLyxNejxw5Eu3bt3+CLaqfGTNm4PTp0w1axtWrV0VhyM7ODqNGjWpo0whASUkJ3nrrLSQlJT3ppmi1rKws7N27F5GRkbhx4wby8/NhYGAAc3NzWFhY4LnnnkPXrl3Rr18/ODs7i+oeOHAAH374odoyv/jiC/zrX//SuL6oqCi88847auVz5szB3Llzq21nVFQUQkNDcf78edy/fx/5+fkwNTWFtbU13Nzc4Ofnh/79+6vVW79+PTZs2PCYd6F64eHhLfL4VhsModSkRowYIQqhJ0+eRFZWFiwtLTXOX1BQgPDwcLVl0P/cunULixcvxoULF9SmKRQK/PPPP/jnn38QFhaGdevWNTjkNEc7duxAWlqa8NrDw6PFHaSjo6PV/jZ6enowNzcHABgbG9dqOVevXhV9wXl4eDCENpLff/9dLYAaGhrC1NQUAKCvr/8kmqVV4uPjMWPGDGRmZorKy8rKUFBQgDt37iAhIQEHDx7EyJEj8eWXX9Zqubt27ao2hO7YsaNObUxNTcWiRYs0HnOzs7ORnZ2NpKQk7N27F25ubli5ciU6dOhQp3U8rRhCqUkNHjwYZmZmyM3NBfAoJB0+fBjjx4/XOP+xY8dQVFQkvDYyMoKvr68kbW0JYmNjMW3aNBQUFKhNMzY2hlwuR35+vlBWUlIiZfOoDi5fvix67eLigp9++gmGhoZPqEVUVdW/kZeXFzZs2ABdXX51Nob8/HzMnj1bLYDq6enBxMQEBQUFUCgU9Vr21atXERsbi169eonKU1NT8ccff9R6OdevX8f48eORk5MjKpfL5WjVqhXy8vJQUVEhlF+4cAGjR4/Gjz/+iGeffRbAo2PzM888o7bszMxMUV1jY2ONPz51dHRq3d6Whp8kalL6+voYMmQI9u7dK5T9+uuv1YbQQ4cOiV6//PLLQq/D0+7u3buYPXu2KIDq6elhypQpGDt2LGxtbQE8GsMWHR2NX375BXFxcU+qufQYlX9sAYCTkxMDaDNT9W/UrVs3BtBGFBoaioyMDOF1586dsXLlSnTv3h0AUFFRgZSUFMTExODo0aOQy+t2GcuuXbvUQuju3btFwa8mxcXFmDlzpiiAGhkZ4d///jeGDx8uBOWQkBCsXLlS2F9ycnIwc+ZM/PrrrzA0NMSUKVMwZcoUteV7e3uLzuhMnjy5xuEA2oifJmpyI0eOFIXQCxcuICUlBR07dhTNl5GRgbNnz6rVBR4dDPbv348rV67g2rVryMzMRE5ODhQKBUxNTWFvb48XX3wRY8eOhbW1dZ3bWHXMjqbTPhMnTsS5c+eE18uXL9d42vP27dvYs2cPoqKikJqaisLCQrRq1QouLi4YOnQohg8fXq8vsnXr1iE7O1t4LZPJsG7dOnh7e4vma9WqFQYPHozBgwfj4sWLGpd169Yt/PTTT4iOjkZaWprQRgcHB3h6emLs2LEah0zU5j1wcnISva46nqnqgXfHjh2ws7NDUFAQTp48iYcPH8LS0hIvvfQS3nvvPVEPQtW6Km+//bbodeV2PXjwADt37sTp06eRkpKCgoICGBoawtLSEra2tujevTvc3d3V3sfaUCqVCA8Px6+//opLly7h4cOHUCqVsLKyQvfu3fHaa69h8ODBoi/P6saHBQcHIzg4uNr3rarqxsKdO3dO9Dews7PDTz/9BG9vb5SVlQEAzMzMcPLkSbXQm5ubi/79+wu9T0ZGRjh16hRMTU3V1ufh4YEdO3Zg7969+OWXX5CcnAy5XA4XFxdMmjQJgwcPrrbtTfUZ0UShUOD3339HaGgoEhISkJWVBR0dHbRp0wY9e/bE6NGj0a9fP1GdxYsXi/4WKhs2bBD97RITE+vcnoYeyzQdq5YuXYqtW7fi8OHDSEtLg76+PmJjY0X1Tp48iZCQEMTHx+PBgweoqKiAtbU1evfujYkTJ8LFxUVje/ft2ye0NSMjA9nZ2SgpKYGJiQk6dOgADw8PjBs3Dp06darze3Ht2jXR6zfffFMIoMCj3kZ7e3vY29tjzJgxKC0tfewyjYyMhDB47Ngx3Lt3D23btgUAFBYWYv/+/Rrn1WTPnj1ITU0Vla1evRo+Pj7CaxMTE4wfPx5t27bF7NmzhfKUlBTs2bNH49hT+h+GUGpybm5usLe3x61bt4SyQ4cOYc6cOaL5fv/9d5SXlwuv27Zti759+wJ4FCSWLVumcflZWVnIysrChQsXsGPHDgQGBqr9+pXKrl278OWXX6qdQsrMzMSpU6dw6tQp7NmzB99++22dwnJBQQF+/fVXUdmIESMeG5x69OihVrZlyxasXbtWCCQqWVlZiI2NRWxsLL7//nt89dVXGDRoUK3bWF8nT57E7t27UVhYKJTdv38f+/btQ1RUFIKDg9GqVat6LfvmzZt466231E735efnIz8/H6mpqYiOjsavv/5a5xCamZmJ999/X+2HEwDcuXMHd+7cQVhYGDw8PLB27VpYWVnVaxsaQ9u2bTF48GCEhoYCeBQ2Q0ND1cZbHz9+XLTv+vr6VnsmQqFQYPbs2WpjuM+dO4dz585hxowZeP/999XqNdVnRJPU1FTMmTNHLewAj0JCSkoKfv31V/j6+mLFihUwMjJq0Ppqo7GPZXl5eXjzzTdx9epVoazyWNXc3FwsXLhQNDZfJTU1FampqQgODsaMGTMwf/58tXk++eQTjevNyclBTk4OLl++jN27d+Orr76Cn59fjW2tqmqoPHHiBIYPHw4LCwuN89dmDK6rqysePHiAmzdvoqysDD/99JOwXSEhIcLQMHNzc/Tr10/4TGhS9YdI7969RQG0ssGDB8Pd3V109ikkJIQh9DF4iyaSxPDhw0WvqwYqQP1U/PDhw6s9/WJkZARLS0uNPTnz589/ImMhf/nlF3z++eeiL1eZTKb2JX7x4kXMmDGjVr/qVWJiYtS+tOtzwdauXbuwatUqtQBadRxSdnY25syZU21PamPasmULCgsLoaenp9b7lZqaim3btgmvLS0t8cwzz6jtF+bm5njmmWeE/1T7xbp169QCqKmpaa0v+qlOWVkZZs6cqRZA9fT0oKenJyo7d+4cZs6cKbznqvFhVdtgYGAg2obHjQMzNDTEM888o7Z/6enpiZaj6tGeOHGiaL5ffvlFbZlVv5BrusDpwoULQgDV9H4GBQWp3cKoKT8jVeXl5WHy5MlqAdTAwEDtvQ0LC8O///1v4bWpqSmeeeYZGBgYiOZT/e1U/zWGhh7Ljh8/LgRQU1NT0f5XUVGB2bNnqwVQfX190bYplUoEBgbiu+++q3FdBgYGsLS0VPt7l5SU4IMPPhCdWq+NLl26iF5HRUXB09MTEyZMwFdffYXDhw/j/v37dVqmTCbDhAkThNf79u0T9qNdu3YJ5aNHj67xR0dOTo5aT7eXl1eN6676Q/batWtC6CXNGEJJEiNGjIBMJhNe37p1SxRwkpOTceXKFbU6KhYWFvj6669x7NgxXLlyBX/99RfOnj2L+Ph4REREwMPDQ5g3IyOjTgPPG0NeXh6++uorUdm0adMQGxuLuLg4HD9+XHSa6cqVK9izZ0+tl6/pFHTXrl3r1MacnBysWbNGVDZo0CCcOnUKFy5cQEhIiOiUWmlpaa2vRG0ImUyGRYsWIS4uDufOncNLL70kmn7ixAnh3/v378fp06fRrl070Tzr16/H6dOnhf+GDh0KAKLeISsrKxw+fBhxcXG4cOECzp07h/379+PDDz/ECy+8UKc2h4SE4K+//hJe6+joYMmSJTh//jzOnz+PTz/9VBR04uPjhV6VKVOm4PTp05g8ebJomUOHDhVtQ9VtrEo1/0cffSQqd3NzEy1HdfqxV69eotvbxMTEiM5O5Obm4syZM8Jr1anWmri5uSE8PBwXLlzAb7/9BgcHB9H0devWCf9u6s9IVdu2bUNKSorw2tjYGN988w3Onz+PuLg40alTADh69CiioqIAAB9//LFoP1KZPHmy6L2tj6Y4lj333HMIDg5GXFwcLl68iC1btgB49MO+8vAZCwsLbN26FRcvXkR8fDzWr18v6l1ct26dWpBcsWIFfvvtN1y+fBkXL17E2bNnceHCBbX3p7i4WGPnQk1ef/11tbMcpaWliImJwXfffYf3338fnp6eGDNmDA4fPlzr5Y4YMUL4YfPw4UMcPnwYp0+fRnJyMoBHp/nfeuutGpeRnp4OpVIpKrO3t6+xTtXpSqUS9+7dq3W7n0YMoSQJW1tbtS+0gwcPCv+u2gvao0cP0ReaqakpfHx8EB0djYULF2L48OHw9vbGgAED8MYbb6hdxVr1dVP7448/RIPX3d3dERAQIBwIO3TooDZ+T9OYs+po+jVtYmJS5zZWvnLe2NgYK1euRJs2bQA8CrVLly4V1YmLi2vyg+hLL70Ef39/GBgYwMTEBFOnThVN/+eff+q97Mo9Npp6T11dXfHOO+9g7dq1dVpu5fvYAsCQIUMwfvx46OvrQ19fH2+++SaGDBkimqcuX6JNpeoFgZV7Q8PDw0U9lCNHjhT9cKxKLpfjyy+/FMatPvfcc1iyZIlonsTERGFMXVN/Rqqq+jd6++234efnB11dXRgZGWHevHlwc3MTzfP777/Xe3211RTHspUrVwpjOuVyubBdVY+rCxcuhKenJ2QyGWQyGV555RXRWari4mIcOXJEVOe1115DUlIS/u///g+jRo0S2jp8+HD8+eefdW5rZa1bt0ZgYCBat25d43wXL17E+++/j8WLF6sFQ01MTU2F6wmARz2gle817eXl9dhbKFU+Vqo87gyKpmNy5fsZkzqOCSXJjBw5EtHR0cLrw4cP48MPP4SOjo7G8Y6V3bhxA/7+/hp7BDWpfAGPFBISEkSv4+Li1C7Qqerq1asoLS2t1TgnTTecz8/Pr3bslCZVT0v26NFDuCelSr9+/aCrqys6XX/t2jVhYH9TqHoBS9UvpMpjRevqlVdeEXrYMzIyMHToUFhYWMDOzg6dO3dG9+7dMWjQIHTu3LlOy636Xnp6eqrN4+npKQpCmsYlSm3YsGFYtWqVEAZDQkIwf/586OrqisKHXC4XfYlr0r59e7WeHw8PD+jp6YnC7N9//40OHTo0+WekssLCQlEvKKD5bzRgwADRvR+l+Bs19rHMxcWl2rMiVd/zjz/+GB9//HGNy4uPjxf+/eDBA0yZMqXW70t9jru9e/fG0aNHsXfvXhw7dgyXL19WGy6kEhwcDG9vb7zyyiuPXe6ECROwa9cuKJVKXLp0STSt6tAUTTSNhX7csUjTrfPqO579acGeUJLMK6+8IvolqboQIS4uTnRA1tPTw6uvviqqu3jx4loftAHU+95yKpp+bde0zPr82lUqlcjKyqrVvKrbL1VW+VRzbVRto6YLZeRyudpV8TWNaar6PtXnfbexsRG9rjqmsiH8/f0xceJE0TKzs7Nx5coV/Pbbb1i+fDn8/PzwwQcf1KnttXkvq5Y1hx4RIyMjjB49WnitOt1b9VR8v379NO5zlWnqvZLL5Wo/jFTb3dSfEU3rrExTe6uO65Tib9TYx7Ka7qBQn+2p/H5/8cUXdQrm9T3utmrVCv7+/ti7dy/i4uKwe/duvP/++3B0dFSbt7aPSrW3t9f4w+PZZ59VuxuCJjY2NmpnAm7evFljnarTZTJZk/6A1wYMoSQZExMTvPzyy6KyQ4cOqZ0y8vLyEn2R3blzR/TrHADmzZuHU6dO4erVq0hMTMS0adMa1LaqBxtNB9OaTktX/bVrZGQkuoChuv9qe7+63r17q4Wzup6qrNrGqhfsAI8uZKjam1HTYz+rvk/p6el1ahOgHjprOgVcV7q6uvj4449x8uRJrFmzBtOnT4efnx8cHR1F6wkJCRHdRuxxavNePnz4sMY6T8pbb70lGprw888/q52Kr80Tl2q7/6i2u6k/I5rW+bj2Sv03aopjWU3Dcqpuj4WFxWPfb1VHQWlpqVrge+uttxAREYGEhAQkJibis88+q1Nba8PQ0BC9evXCjBkzcPDgQbVhXA8ePKj1sipfoFRTmSbm5uZqPfWVx6drEhERIXrt7OyslY9Nbkw8HU+SGjlypGgsaHh4uNoVqFVPxVcNNubm5moXFTT0puxVr0ytOjg/ISEBd+7cqbZ+1dNh3bt3f+zz7isqKmp982VTU1O8+uqrCAkJEcoOHToEPz+/Gm8tFB8fj549ewKA2jOX4+PjkZubKzpInj17Vi1YVj4QV72atOoXQtXb9TSVqkH1cUHF0tISQ4cOFV1IERkZKfrCP3PmTK2/oJydnYWLWADg1KlTaneAOHXqlFqdplB1H6p8mzNNOnTogEGDBglfqCdPnhT9Hc3MzNR+LGpy+/Zttfv9arqLg+qpMU39GanM2NgYHTt2FJ2SP3XqlNrtjqT6G6lIcSyrrGvXrqIe7sWLFz92mIXq7EZWVpba3Qkqj+FtjLaeO3cOZmZm1b7vcrkcTk5Ooour6vLwkoEDB4puD2hmZqb2Oa3JiBEjRBdnxsbGIjw8XONtmo4fP47z58+r1aeasSeUJNWnTx/RVb/FxcWiixVat26NgQMHiupUHbeYm5sr/OIsLCzEihUrGnwwrHqj5QsXLgiD7lXPaq/JSy+9JOp1OHfuHJYtW4a7d+8KZWVlZUhOTsa+ffswbdo0BAUF1amN8+bNEwVGpVKJefPmYe3ataL15OXlITw8HNOnT4e/v7+ojZUP4IWFhfj3v/8tBJBr166p9Wy4u7uLTpdXfcDAgQMHkJqaioqKCpw6dQobN26s0zbVV9XehZiYGI3zffHFF1i5ciWioqJE+1lpaanaRRTVjUPTpOpwkcOHD2PPnj0oLS1FaWkp9u3bp3aBR9UrrRtL1d6uGzduqPXwVVX5AqXy8nLRe/Hqq6+q/TDUpKKiAh988IFwajk5OVnt/peOjo7CBSBSfEYqq/o3+uGHH3D06FGUlZWhuLgYGzZsUAsNVes0NimOZZUNGzZM9Pq///0vfvvtN9Ftn/Lz8xEbG4v169fDz89P+Hu2atVK7QfAwYMHoVQqUVpaiu3bt9f5aviq/vrrLwwfPhwTJ07E3r17cePGDSEEl5eXIzw8XNRpATy6D2htyWQyzJ49G/369UO/fv0wa9asOt2ebdy4cWrDHRYuXIiffvpJGB9aWFiIH3/8EQsXLhTN1759e4wbN67W63pasSeUJCWXyzF8+PBqv1xee+01tdOzXbp0QYcOHYSrbJVKJWbOnAkTExMUFRWhoqLisU++eJx+/frB2NhYOLAoFApMnTq11ss1MzPDv//9b9GNnXfv3o3du3fDyMgIBgYGyMvLE/VSVb4dTW3Y2dnh22+/xfTp00XtDAwMRGBgIExMTCCXy0XjwCp/6Zubm+P999/H559/LpSdOHECL774omjbVfT19dXCt4+Pj6j3Ki0tDYMHD4ahoSGKi4vrtD0N4ezsLLro4ttvv8X3338vfMGcOHEC+vr6uHfvHsLCwrB161YAj3rIVH+LqqGz6pXSNRk5ciR+/vln4dRqWVkZli5div/85z8A1Icp9OjR47E9UPVVtRcpKysLnp6eMDc3h1wux4QJEzBz5kzRPAMGDFB7gIRK5TGjNZHL5Th//jy8vb017j/Aox9OKlJ8RiqbMmUKfv/9d6E3tLCwEHPnzoWBgQHKysrUeoxfeeWVWo0VbAgpjmWVDR8+HPv37xeenpSbm4uAgADIZDKYmZmhrKxM48U0wKPPSq9evUS9kMuWLcPKlStRVlYGhULRaG1VPeAAeDSExtTUVG1fAB4NPajNUJHKXn/9dbz++uv1apeRkRE2btyICRMmCGPjCwsL8emnn+Lzzz9Hq1atkJubq3YmxszMDIGBgXwMby2wJ5QkV9MpCk1f1DKZDJ9++qlaOC0oKEBFRQXc3NyqfRZ9bZmammLBggVq5aoDrKen52O/EN944w188sknalfyFhUVITs7W+MBta48PDzwyy+/CKfYKysoKFC7EKFqj9aECRMQEBCgdrPuqgHCwsICGzZsUHviUr9+/eDr66u2blUAnTVrVu03pgHefPNNjdvw4MEDPHjwoNrbuBQWFiIrK0stgD7//PN1erKJrq4ugoKC0KdPH1G5QqFQC6C9e/dGUFBQkz1z3M7OTm1IRnl5OTIzM/HgwQONIUMmk2n8zDg6OtY6+L3wwgtCT5umADp16lS10/pSfEZUWrVqhW3btqmN6yspKVFbj6+vr9o9TJuCFMeyynR0dBAYGKh2k3WlUomcnBy1fUNfX1/Uto8++kjt9HdRUREUCgXs7e0b/JxzTRchlpWVadwXzMzMsH79+kZ7SEBtOTk5Yf/+/WrH3PLycmRnZ6sF0B49emD//v0aL6oidewJJcl17twZzz//vOhm38CjL8Dqnl88YMAA/Pjjj9iwYQPi4uKgUCjQvn17vPrqq5g6dSo2bdrU4HZNnDgRVlZW+OGHH3Dt2jXI5XI8++yzGDNmDP71r39h0qRJj13GhAkT4OPjIzxy8ubNm8jPz4euri6srKzw7LPPolevXvD29hbGytWVg4MD9u3bh+joaBw9ehTnz5/H3bt3kZeXBz09PVhbW8PZ2Rmenp4aH6M3bdo0DB48GHv27EF0dDRu376N4uJimJqaokuXLhg4cCDGjh1b7b37Vq9eja5du+LgwYO4ffs2jI2N0bNnT0yePBn9+vWT5JR8jx498P3332Pz5s24fPkycnJyNAbPBQsWoE+fPoiLi8P169eF53TLZDJYWVnB0dERL7/8MkaMGFHnq/Jbt26N7du3IyIiAocOHRKeHa+apnp2/Msvv1yvcY11sXr1agQGBuLo0aNIS0ur1VXKo0aNwpo1a0QBsi69THK5HKtWrULfvn2xd+9e/P333wCAbt26YdKkSdWOK5XiM6LSoUMH/PLLLzh8+DBCQ0Nx5coV0bPjn3/+eYwaNQr9+/dv0HrqQopjWWVmZmYICgpCVFQUDh48iL/++gv37t1DSUkJjI2NYWtrCxcXF/Tt2xc+Pj6isyfOzs74+eefsW7dOkRFRaGgoAA2NjZ4+eWXMWvWLBw7dqxBbXv33XfRr18/REVF4eLFi0hOTsa9e/eEcGxmZoYuXbqgf//+eOONNyQPoCodO3bEvn37cPr0aYSFheH8+fO4f/8+CgoKYGJiAmtra7i5ucHX1xcDBgx4Im1sqWTK2tz5lYiItM706dOFJ/Lo6enhzz//rPbHx4EDB0Q3k/fw8HjshUVERDXh6XgioqdQWloazp49K7z29vZ+7JNriIgaE0/HExE9Ja5fv47169ejsLAQFy5cEF1MVvlOCs3NvXv3an3hm66uLuzs7Jq4RY+kpaXV+q4KhoaGvHE5URUMoURET4nMzEyEhYWplU+ePFntIrTmZOHChaKrtGtiZ2endtPwpjJx4sRaP/2IwxeI1DGEEhE9hUxMTGBvb4+JEyfyptpE9ETwwiQiIiIikhwvTCIiIiIiyfF0fAuhVCpRUcFOayIiImq+5HIZZDJZreZlCG0hKiqUyMzU/Hg1IiIiouagdWsT6OjULoTydDwRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpKc7pNuABEREdWdXC6DXC570s0gLVFRoURFhVLSdTKEEhERtTByuQwWFsbQ0eEJTWoc5eUVyM4ulDSIMoQSERG1MHK5DDo6cnz702mk3c950s2hFs7O2hyz33wRcrmMIZSIiIgeL+1+Dm6lZT3pZhDVC/vxiYiIiEhyLaIndPHixQgODq52+tixY7Fs2TK18tTUVKxbtw5RUVHIycmBjY0N/Pz8MHPmTBgbG1e7vCNHjmDHjh1ITEwEADg7O2PSpEnw9fWttk5BQQGCgoIQGhqK9PR0mJubo3///pg7dy46dOhQh60lIiIi0n4tIoSqDBgwAG3atFErd3NzUytLSEjAhAkTUFBQgG7duqFXr164ePEiNm/ejMjISPz4448wNTVVq7d27VoEBgZCX18fL774IgDg9OnTmDdvHubOnYs5c+ao1cnLy8Nbb72FpKQk2NnZwcfHBykpKTh48CDCw8Oxe/duODs7N8I7QERERKQdWlQInTZtGvr06fPY+crLyxEQEICCggIEBARg2rRpAIDS0lLMmzcPJ06cwMqVK/HZZ5+J6sXFxSEwMBBmZmbYs2cPHBwcAADJyckYN24c1q9fD09PT/Ts2VNUb8WKFUhKSoKXlxfWrVsHfX19AMCmTZvw9ddfY+HChTh06BDkco5+ICIiIgK0dExoREQEbty4AUdHR0ydOlUo19fXx7Jly6Crq4v9+/cjK0s8mHvr1q0AgBkzZggBFAAcHBwwffp0AMCWLVtEdTIzMxEcHAxdXV0sW7ZMCKDAo9Ds6OiI69ev48SJE42+nUREREQtldaGUADw9fWFTCa+ka+1tTXc3d2hUCgQGRkplJeWluL06dMAgCFDhqgtc+jQoQCAkydPorS0VCiPjIxEWVkZ3N3dYW1tLaojk8mEcaTh4eGNsGVERERE2qFFnY4/duwYjh07htLSUrRr1w4vvvgievTooTbftWvXAACurq4al9OtWzdER0cLFx4BwI0bN1BSUgJLS0vY2tqq1bG1tYWFhQWys7Nx69YtODo6itbVrVu3atdVeb7mgE/ZoMb0JJ6yQURELV+LCqE7d+4UvV67di0GDRqEr776ChYWFkL5nTt3AAA2NjYal9O2bVsAQFpaWq3rqKZlZ2cjLS1NCKGPq6cqr7yu+tLVbXjHtUwmQ6tWhnzKBjWa8vIK5OUVQ6lkECWSCo/h1BSk3q9aRAh1dnbG0qVL0bdvX7Rr1w6ZmZk4d+4cvv76a0RGRmLGjBn48ccfhQt/CgoKAABGRkYal2diYiKarzZ1AAi3ddJUr7pbPmmqUx9yuQyWliYNWkZlfMoGNQbVUzYsLKq/5RkREbUMZmbVZ6Cm0CJC6DvvvCN6bWdnh5EjR6J///54/fXXceHCBYSGhgrjNlWqjgdV0dRjoyqrrs7j6tWlTn1UVCiRm1vY4OXo6MhhZmbEp2xQo8rNLUJ5ecWTbgYA9vZT02huPf6qYzlRY2qMY7mZmVGtj78tIoRWp23bthg1ahS2bduGkydPCiHU2NgYOTk5KCzUHNqKiooA/K9HtPK/q6sDAMXFxXWup6lOfZWVNY8veaKqyssrms3+qasr5zO1qVGpevyVSmWz2c+JmoLUx/IWHUIBwN7eHgBw//59oczW1hY5OTlIT0/XeJP49PR0YT4VOzs70TRNNNVT/bu6eqpy1fKJSBrs7Sciat5a/PmqnJxHPR2Vx3J27doVAHDlyhWNdRISEgBAFFA7d+4MAwMDZGVl4e7du2p17t69i6ysLBgaGqJz585q61ItsypVG5ycnGq9TURERETarkWHUKVSiaNHjwIQ347Jy8sLABAWFqZWJyMjA3FxcdDV1cXAgQOFcgMDA/Tv3x8AEBoaqlbvyJEjAB49OrTyDekHDhwIHR0dxMXFISMjQ619qmX5+PjUaxuJiIiItFGzD6EJCQn49ddfRTeIB4D8/Hx8/PHHuHTpEoyNjTF69Ghhmre3N+zt7ZGYmCh6wpFCocCSJUugUCgwevRotG7dWrTMKVOmAAACAwORnJwslCcnJyMoKAgA4O/vL6pjZWWFESNGQKFQYOnSpVAoFMK0LVu2ICkpCQ4ODkIwJiIiIqIWMCb0zp07WLhwIT7//HO4urrC0tISDx48wNWrV5GTkwNjY2OsXbsWbdq0Eero6upi9erVmDhxIlatWoXQ0FB07NgR8fHxwj0+Fy1apLau3r17Y/r06di0aZNw9T0AnDlzBiUlJZg1axbc3NzU6i1evBjx8fEIDw+Hr68vevbsiZSUFFy+fBkmJiZYvXo1dHR0mu5NIiIiImphmn0IdXJywsSJE3Hp0iUkJSUhOzsbenp6sLOzw4gRI/D222+jffv2avVcXV0REhKC9evXIyoqComJibCxsYG/vz9mzZpV7dXqCxYsgLOzM3bs2IHo6GgAgIuLCyZNmqTxcZ4AYGZmhr179yIwMBBhYWE4duwYzM3NMWzYMMybNw8dO3ZsvDeEiIiISAs0+xDaoUMHfPzxx/Wq26lTJ6xatarO9YYOHap2z9HHMTU1xaJFizT2sBIRERGRWLMfE0pERERE2ochlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOd0n3YD6UCqVmDRpEqKjowEAhw8fhoODg9p8qampWLduHaKiopCTkwMbGxv4+flh5syZMDY2rnb5R44cwY4dO5CYmAgAcHZ2xqRJk+Dr61ttnYKCAgQFBSE0NBTp6ekwNzdH//79MXfuXHTo0KGBW0xERESkXVpkT+jevXsRHR0NmUxW7TwJCQkYPnw4Dh06BGtra/j4+KC8vBybN2/GuHHjkJ+fr7He2rVrMX/+fFy+fBkeHh7w8PDApUuXMG/ePGzYsEFjnby8PIwbNw6bN29GeXk5fHx8YG1tjYMHD2LEiBG4du1ao2w3ERERkbZocSE0PT0dK1euhKenJ2xtbTXOU15ejoCAABQUFCAgIAAHDhzA2rVrERoaCi8vLyQmJmLlypVq9eLi4hAYGAgzMzOEhIQgKCgIQUFBCAkJgZmZGdavX4/4+Hi1eitWrEBSUhK8vLwQGhqKtWvX4sCBA1iwYAHy8/OxcOFCVFRUNPp7QURERNRStbgQumTJElRUVODTTz+tdp6IiAjcuHEDjo6OmDp1qlCur6+PZcuWQVdXF/v370dWVpao3tatWwEAM2bMEJ3ed3BwwPTp0wEAW7ZsEdXJzMxEcHAwdHV1sWzZMujr6wvTpk2bBkdHR1y/fh0nTpyo9zYTERERaZsWFUJDQkIQGRmJ9957D+3bt692voiICACAr6+v2il7a2truLu7Q6FQIDIyUigvLS3F6dOnAQBDhgxRW+bQoUMBACdPnkRpaalQHhkZibKyMri7u8Pa2lpURyaTCeNIw8PD67KpRERERFqtxYTQBw8eYPny5XB1dcXEiRNrnFc1BtPV1VXj9G7dugGAcOERANy4cQMlJSWwtLTUeJrf1tYWFhYWKC4uxq1bt9TWpVpmdeviuFAiIiKi/2kxV8cvW7YM+fn5+OKLL6Cjo1PjvHfu3AEA2NjYaJzetm1bAEBaWlqt66imZWdnIy0tDY6OjrWqpyqvvK760tVt+G8GHZ0W87uDWpDmtF81p7aQdmlO+1ZzagtpD6n3qxYRQo8ePYqwsDBMmzYNzs7Oj52/oKAAAGBkZKRxuomJiWi+2tQBINzWSVO96m75pKlOfcjlMlhamjRoGURNxcys+s8Nkbbgfk7aTup9vNmH0OzsbHz22Wfo2LEjZs+eXae61d3CSalUVltW022faqpXlzr1UVGhRG5uYYOXo6Mj54GUGl1ubhHKy5vHHSC4j1NT4X5O2q4x9nEzM6Na96g2+xC6fPlyPHjwANu3b4ehoWGt6hgbGyMnJweFhZpDW1FREYD/9YhW/nd1dQCguLi4zvU01amvsrLmcfAjqqq8vIL7J2k97uek7aTex5t9CA0PD4eBgQE2btyIjRs3iqZlZGQAAD744AMYGRlh/Pjx8PPzg62tLXJycpCenq7x9H16ejoAiC5AsrOzE03TRFM91b+rq6cqVy2fiIiIiFpACAWAkpISnDt3rtrply5dAgD4+PgAALp27YqrV6/iypUreOmll9TmT0hIAABRQO3cuTMMDAyQlZWFu3fvol27dqI6d+/eRVZWFgwNDdG5c2ehvGvXrqJlVnXlyhUAgJOT0+M2k4iIiOip0ewvr4uNjUViYqLG/1S9i4cPH0ZiYiLeeecdAICXlxcAICwsTG15GRkZiIuLg66uLgYOHCiUGxgYoH///gCA0NBQtXpHjhwBAAwYMEB0Q/qBAwdCR0cHcXFxQs+silKpFJalCshERERE1AJCaH14e3vD3t4eiYmJoiccKRQKLFmyBAqFAqNHj0br1q1F9aZMmQIACAwMRHJyslCenJyMoKAgAIC/v7+ojpWVFUaMGAGFQoGlS5dCoVAI07Zs2YKkpCQ4ODgIwZiIiIiIWsjp+LrS1dXF6tWrMXHiRKxatQqhoaHo2LEj4uPjhXt8Llq0SK1e7969MX36dGzatAkjR44UekbPnDmDkpISzJo1C25ubmr1Fi9ejPj4eISHh8PX1xc9e/ZESkoKLl++DBMTE6xevfqx9zYlIiIieppoZU8o8OhpSSEhIRg2bBjS09Nx7NgxyOVy+Pv7Y8+ePWjVqpXGegsWLMCaNWvg4uKC6OhoREdHw8XFBWvXrsV7772nsY6ZmRn27t0Lf39/yOVyHDt2DOnp6Rg2bBhCQkKEcaNERERE9EiL7glVPSO+Op06dcKqVavqvNyhQ4cKz4qvLVNTUyxatEhjDysRERERiWltTygRERERNV8MoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyek21YKLi4tx+vRp3L9/H66urujevXtTrYqIiIiIWpgGhdDjx4/j+++/x6hRozB69Gih/M6dO5g8eTL++ecfoWzSpElYvHhxQ1ZHRERERFqiQafjw8LCcP78eXTt2lVU/uWXX+LWrVswMTGBk5MT5HI5fvjhB0RGRjaosURERESkHRoUQq9cuQJTU1O4uLgIZdnZ2YiIiECrVq3w+++/IyQkBMuXL4dSqcS+ffsa3GAiIiIiavkaFEIfPnyItm3bisrOnTuHsrIy+Pn5CdNef/11tG7dGpcuXWrI6oiIiIhISzQohBYUFMDAwEBUdvHiRchkMvTr109U3q5dO2RmZjZkdURERESkJRoUQs3MzJCeni4qO3v2LADA3d1dVF5RUQEjI6OGrI6IiIiItESDQqiLiwsyMzNx9OhRAI96QS9fvoxOnTqpnaZPTU2FlZVVQ1ZHRERERFqiQbdoGjduHE6dOoUFCxbA0dERN2/ehEwmw5gxY0TzJSQkID8/H/3796/Xevbu3YuoqCgkJibi4cOHKCgogLm5Obp3745x48bBy8tLY73U1FSsW7cOUVFRyMnJgY2NDfz8/DBz5kwYGxtXu74jR45gx44dSExMBAA4Oztj0qRJ8PX1rbZOQUEBgoKCEBoaivT0dJibm6N///6YO3cuOnToUK/tJiIiItJWDeoJHTx4MGbPng3gUdAsKirCa6+9hkmTJonmCwkJAQC1caK19f333+PYsWMwNDTECy+8gJdffhnt2rXDH3/8gRkzZmDFihVqdRISEjB8+HAcOnQI1tbW8PHxQXl5OTZv3oxx48YhPz9f47rWrl2L+fPn4/Lly/Dw8ICHhwcuXbqEefPmYcOGDRrr5OXlYdy4cdi8eTPKy8vh4+MDa2trHDx4ECNGjMC1a9fqtd1ERERE2qrBT0yaO3cuJkyYgJSUFLRr1w7W1tZq83h5ecHDwwO9evWq1zqWL18OR0dHmJiYiMpjY2MxdepUbNu2DX5+fujZsycAoLy8HAEBASgoKEBAQACmTZsGACgtLcW8efNw4sQJrFy5Ep999ploeXFxcQgMDISZmRn27NkDBwcHAEBycjLGjRuH9evXw9PTU1iPyooVK5CUlAQvLy+sW7cO+vr6AIBNmzbh66+/xsKFC3Ho0CHI5XxKKhERERHQSM+Ot7S0RM+ePTUGUOBRD+jgwYNhYWFRr+W7ubmpBVAA6NWrF4YMGQIAiIqKEsojIiJw48YNODo6YurUqUK5vr4+li1bBl1dXezfvx9ZWVmi5W3duhUAMGPGDCGAAoCDgwOmT58OANiyZYuoTmZmJoKDg6Grq4tly5YJARQApk2bBkdHR1y/fh0nTpyo17YTERERaaMGhVAfHx+8//77tZp3wYIFGDx4cENWp5Gu7qPO3MrhLyIiAgDg6+sLmUwmmt/a2hru7u5QKBSiJziVlpbi9OnTACAE28qGDh0KADh58iRKS0uF8sjISJSVlcHd3V0thMtkMmEcaXh4eL23kYiIiEjbNCiEpqWl4f79+7WaNyMjA2lpaQ1ZnZqrV6/iyJEj0NHRgaenp1CuGoPp6uqqsV63bt0AQLjwCABu3LiBkpISWFpawtbWVq2Ora0tLCwsUFxcjFu3bqmtS7XM6tbFcaFERERE/9PgMaG1VVZW1uAxkfv370dMTAwUCgXS0tLw119/QVdXF59++imee+45Yb47d+4AAGxsbDQuR3X7qMqh+HF1VNOys7ORlpYGR0fHWtVTlTdGANfVbfjoCR0djkulxtec9qvm1BbSLs1p32pObSHtIfV+JUkILSsrQ0pKCszNzRu0nPPnzyM4OFh4bWRkhI8++gijR48WzVdQUCBM10Q1vlQ1X23qABBu66SpXnW3fNJUpz7kchksLdXHxRI1B2ZmfBAFaT/u56TtpN7H6xRCY2JiEB0dLSq7e/dutbcuAoCSkhLExcUhMzMTAwcOrF8r/78vvvgCX3zxBQoLC/HPP/9g586d+Pjjj3H8+HGsW7dO7RGiVceDqiiVymrLqqvzuHp1qVMfFRVK5OYWNng5OjpyHkip0eXmFqG8vOJJNwMA93FqOtzPSds1xj5uZmZU6x7VOoXQ6OhobNiwQRTU7t69i2+//bbGekqlEkZGRpgxY0ZdVlctY2NjdO3aFf/9738hl8vx888/Y9u2bZg5c6YwPScnB4WFmkNbUVERAIiuuFf9u7o6AFBcXFzneprq1FdZWfM4+BFVVV5ewf2TtB73c9J2Uu/jdQqhzs7OGDlypPA6ODgYVlZWoouCqjIyMkLHjh3h5+dX43jL+ho+fDh+/vlnhIeHCyHU1tYWOTk5SE9Ph7Ozs1od1fPuK1+AZGdnJ5qmiaZ6qn9XV09Vrlo+EREREdUxhA4ePFh0m6Xg4GB06tQJy5cvb/SG1Vbr1q0BPLpfp0rXrl1x9epVXLlyBS+99JJanYSEBAAQBdTOnTvDwMAAWVlZuHv3Ltq1ayeqc/fuXWRlZcHQ0BCdO3cWravyMqu6cuUKAMDJyakeW0dERESknRp0GVR4eDi++eabxmpLvZw9exYA0KlTJ6FM9Sz5sLAwtfkzMjIQFxcHXV1d0RhVAwMD4dn2oaGhavWOHDkCABgwYIDonqQDBw6Ejo4O4uLikJGRIaqjVCqFZfn4+NRr+4iIiIi0UYNCqJ2dHZ555pnGaotGly5dQnBwsOgG8SonTpzA2rVrAQBjxowRyr29vWFvb4/ExETRE44UCgWWLFkChUKB0aNHC72oKlOmTAEABAYGIjk5WShPTk5GUFAQAMDf319Ux8rKCiNGjIBCocDSpUuhUCiEaVu2bEFSUhIcHByEYExEREREjXiLpvz8fKSmpqKgoKDGK8J79+5dp+Xeu3cPixcvxhdffAFXV1dYWVkhLy8PN2/eREpKCgBg8uTJwhONgEdPUVq9ejUmTpyIVatWITQ0FB07dkR8fLxwj89FixZpbNv06dOxadMmjBw5UugZPXPmDEpKSjBr1iy4ubmp1Vu8eDHi4+MRHh4OX19f9OzZEykpKbh8+TJMTEywevVq6Ojo1Gm7iYiIiLRZg0Po5cuXsWLFCsTFxT32dkQymazasZPV6d69O2bPno2YmBjcvHkTcXFxkMvlsLa2xuuvv46xY8eiV69eavVcXV0REhKC9evXIyoqComJibCxsYG/vz9mzZpV7dXqCxYsgLOzM3bs2CHcjsrFxQWTJk3S+DhPADAzM8PevXsRGBiIsLAwHDt2DObm5hg2bBjmzZuHjh071mmbiYiIiLRdg0LolStXMHHiRBQXF0OpVEJfXx9WVlY13muzrtq2bYt58+bVq26nTp2watWqOtcbOnSoqGe1NkxNTbFo0SKNPaxEREREJNagELp+/XoUFRXh+eefx8cff1zts9qJiIiIiCprUAg9f/48DAwMEBgYCEtLy8ZqExERERFpuQZdHV9SUoIuXbowgBIRERFRnTQohHbq1KnGx1wSEREREWnSoBA6atQopKSk4OrVq43VHiIiIiJ6CjQohL799tvo378/5s6di/PnzzdWm4iIiIhIyzXowqSPPvoIVlZWiI6Oxvjx4+Hk5AR7e3sYGRlpnF8mk+G///1vQ1ZJRERERFqgQSE0ODgYMplMuEn9tWvXcO3atWrnZwglIiIiIqCBIXTOnDmN1Q4iIiIieoowhBIRERGR5Bp0YRIRERERUX0whBIRERGR5Bp0Oj4mJqbOdXr37t2QVRIRERGRFmhQCJ04cSJkMlmt55fJZEhISGjIKomIiIhICzQohNra2lY7raioCFlZWQAAPT09tGnTpiGrIiIiIiIt0qAQGhERUeP03Nxc7N69G5s3b8aYMWMwc+bMhqyOiIiIiLREg0Lo45iZmWHmzJno1KkTAgIC4OjoCB8fn6ZcJRERERG1AJJcHT906FBYWVnh+++/l2J1RERERNTMSXaLJhsbmxof6UlERERETw9JQmhFRQVSUlJQXl4uxeqIiIiIqJlr8hCqUCiwfPly5ObmwtHRsalXR0REREQtQIMuTPrwww9rnP7gwQNcvXoVDx8+hEwmw6RJkxqyOiIiIiLSEg0KocHBwZDJZFAqlTXOZ2RkhICAAAwdOrQhqyMiIiIiLdGgEDpnzpwapxsbG6NTp07o27cvTExMGrIqIiIiItIiTRpCiYiIiIg0kewWTUREREREKo36xKSKigrcuXMHBQUFMDExga2tLeRy5lwiIiIiEmuUEHrlyhUEBQXh1KlTKC4uFsoNDQ3h6emJ6dOno1u3bo2xKiIiIiLSAg3upvzll18wduxYHD9+HEVFRVAqlcJ/RUVFOHr0KMaOHYv9+/c3RnuJiIiISAs0qCf06tWrWLp0KcrLy9GrVy9MnjwZjo6OsLa2xv3793H9+nVs27YNMTExWLp0KVxcXNC1a9fGajsRERERtVAN6gn97rvvUF5ejnfffRe7du2Ct7c32rdvD319fbRv3x5eXl7YuXMnJk+ejLKyMmzbtq2x2k1ERERELViDQmhMTAzMzMywYMGCGuebP38+WrVqhXPnzjVkdURERESkJRoUQh8+fIhOnTpBT0+vxvn09fVhb2+PzMzMhqyOiIiIiLREg0KoiYkJHjx4UKt5Hz58CGNj44asjoiIiIi0RINCqIuLC9LT0xEeHl7jfBEREbhz5w5cXFwasjoiIiIi0hINCqGjR4+GUqnEwoULsX37dtE9QgGgpKQEP/zwAxYtWgSZTIZ//etfDWosEREREWmHBt2i6bXXXsPRo0dx9OhRrFixAt988w06dOiANm3aICMjA6mpqSguLoZSqYSvry9effXVxmo3EREREbVgDX5i0po1axAYGIjt27cjPz8fSUlJSEpKEqabmprinXfewcyZMxu6KiIiIiLSEg0OoTo6OpgzZw6mTJmC2NhY3Lx5U3h2fJcuXeDu7g4jI6PGaCsRERERaYlGeXY8ABgZGcHT0xOenp6NtUgiIiIi0lJ1vjBp9uzZ8PDwwJYtW2o1/+bNm+Hh4YH33nuvzo0jIiIiIu1UpxB66dIlhIeHo127dvD3969VHX9/f7Rr1w5Hjx7F5cuX69VIIiIiItIudQqhv/32G2QyGWbMmAGZTFa7FcjlmDVrFpRKJQ4dOlSvRhIRERGRdqlTCI2NjYW+vj68vLzqtJJBgwZBX18fsbGxdapHRERERNqpTiE0NTUVdnZ2MDQ0rNNKDA0N0aFDB6SmptapHhERERFppzqF0MLCQpiamtZrRSYmJigqKqpXXSIiIiLSLnUKoWZmZsjOzq7XirKzs9GqVat61SUiIiIi7VKnEGpnZ4fbt28jMzOzTivJzMwUTuUTEREREdUphHp4eECpVGLPnj11WslPP/0EpVKJvn371qkeEREREWmnOoXQsWPHQi6XY9OmTbW+0j02NhabNm2Crq4uxowZU69GEhEREZF2qVMI7dixI95++22UlJTg3XffxYYNG5CVlaVx3qysLGzYsAGTJ0+GQqHA+PHj0alTp0ZpNBERERG1bHV+dvyiRYuQmpqK48eP49tvv0VQUBCeffZZdOjQAcbGxigsLERqair+/vtvlJeXQ6lUwsfHBx988EFTtJ+IiIiIWqA6h1C5XI4NGzbgu+++w5YtW5CdnY1r167h2rVrkMlkUCqVwrzm5uaYOnVqrR/xSURERERPhzqHUJUpU6bgrbfeQmRkJOLi4nDv3j3k5+fDxMQEbdu2Ra9evTBw4EAYGxs3ZnuJiIiISAvUO4QCgJGREfz8/ODn59dY7SEiIiKip0CDQqgUFAoFoqOj8ccff+D8+fNIS0tDYWEhbGxsMGDAAPj7+1d7/9HU1FSsW7cOUVFRyMnJgY2NDfz8/DBz5swae2iPHDmCHTt2IDExEQDg7OyMSZMmwdfXt9o6BQUFCAoKQmhoKNLT02Fubo7+/ftj7ty56NChQ8PeBCIiIiItU6er45+EmJgYTJkyBTt37kR2drZwmr+4uBg//vgjXn/9dVy4cEGtXkJCAoYPH45Dhw7B2toaPj4+KC8vx+bNmzFu3Djk5+drXN/atWsxf/58XL58GR4eHvDw8MClS5cwb948bNiwQWOdvLw8jBs3Dps3b0Z5eTl8fHxgbW2NgwcPYsSIEbh27VqjvidERERELV2z7wmVyWQYOnQo3n33XfTo0UMoLykpwaeffooDBw4gICAAYWFh0NPTAwCUl5cjICAABQUFCAgIwLRp0wAApaWlmDdvHk6cOIGVK1fis88+E60rLi4OgYGBMDMzw549e+Dg4AAASE5Oxrhx47B+/Xp4enqiZ8+eonorVqxAUlISvLy8sG7dOujr6wMANm3ahK+//hoLFy7EoUOHIJc3+8xPREREJIlmn4r69euHNWvWiAIoABgYGODTTz9Fq1atkJaWJuoNjYiIwI0bN+Do6IipU6cK5fr6+li2bBl0dXWxf/9+tXucbt26FQAwY8YMIYACgIODA6ZPnw4A2LJli6hOZmYmgoODoauri2XLlgkBFACmTZsGR0dHXL9+HSdOnGjgO0FERESkPZp9CK2JgYEB7O3tAQD3798XyiMiIgAAvr6+kMlkojrW1tZwd3eHQqFAZGSkUF5aWorTp08DAIYMGaK2rqFDhwIATp48idLSUqE8MjISZWVlcHd3h7W1taiOTCYTxpGGh4fXdzOJiIiItE6LDqHl5eVIS0sDADzzzDNCuWoMpqurq8Z63bp1AwDhwiMAuHHjBkpKSmBpaQlbW1u1Ora2trCwsEBxcTFu3bqlti7VMqtbF8eFEhEREf1Psx8TWpODBw8iMzMTVlZWeOGFF4TyO3fuAABsbGw01mvbti0ACAG2NnVU07Kzs5GWlgZHR8da1VOVV15XfenqNvw3g45Oi/7dQc1Uc9qvmlNbSLs0p32rObWFtIfU+1WLDaG3b9/GihUrAADvv/++aCxmQUEBgEf3MdXExMRENF9t6gAQbuukqV51t3zSVKc+5HIZLC1NGrQMoqZiZlb954ZIW3A/J20n9T7eIkNoXl4eZs2ahezsbAwZMgRjxozROF/V8aAqlR8tWrWsujqPq1eXOvVRUaFEbm5hg5ejoyPngZQaXW5uEcrLK550MwBwH6emw/2ctF1j7ONmZka17lFtcSG0uLgYM2fORGJiIvr27YuvvvpKbR5jY2Pk5OSgsFBzaCsqKgLwvx7Ryv+uro5q3XWtp6lOfZWVNY+DH1FV5eUV3D9J63E/J20n9T7eogaVlJaWYu7cuYiJiYGbmxs2btwoOg2vorqwKD09XeNyVOWVL0BSPXWpujrV1avtuqp7qhMRERHR06jFhNDy8nIsXLgQf/75J7p27YrNmzdX27vYtWtXAMCVK1c0Tk9ISADw6HGcKp07d4aBgQGysrJw9+5dtTp3795FVlYWDA0N0blzZ7V1qZZZlaoNTk5Oj9tEIiIioqdGiwihSqUSH330EcLCwuDg4IBt27bBzMys2vm9vLwAAGFhYWrTMjIyEBcXB11dXQwcOFAoNzAwQP/+/QEAoaGhavWOHDkCABgwYICo93XgwIHQ0dFBXFwcMjIy1NqtWpaPj09tN5eIiIhI67WIEPrf//4XwcHB6NChA77//nu0bt26xvm9vb1hb2+PxMRE0ROOFAoFlixZAoVCgdGjR6stZ8qUKQCAwMBAJCcnC+XJyckICgoCAPj7+4vqWFlZYcSIEVAoFFi6dCkUCoUwbcuWLUhKSoKDg4MQjImIiIioBVyYdPz4cezYsQPAo/GXa9as0Tjf4MGDMXjwYACArq4uVq9ejYkTJ2LVqlUIDQ1Fx44dER8fL9zjc9GiRWrL6N27N6ZPn45NmzZh5MiRQs/omTNnUFJSglmzZsHNzU2t3uLFixEfH4/w8HD4+vqiZ8+eSElJweXLl2FiYoLVq1dDR0ensd4SIiIiohav2YfQ3Nxc4d/R0dHVzmdnZyeEUODR05JCQkKwfv16REVFITExETY2NvD398esWbOqHU+6YMECODs7Y8eOHcL6XFxcMGnSJI2P8wQAMzMz7N27F4GBgQgLC8OxY8dgbm6OYcOGYd68eejYsWN9Np2IiIhIazX7EDpq1CiMGjWqXnU7deqEVatW1bne0KFDhWfF15apqSkWLVqksYeViIiIiMRaxJhQIiIiItIuDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESS033SDaiNK1eu4MyZM7h06RIuX76MtLQ0AEB4eDjat29fbb3U1FSsW7cOUVFRyMnJgY2NDfz8/DBz5kwYGxtXW+/IkSPYsWMHEhMTAQDOzs6YNGkSfH19q61TUFCAoKAghIaGIj09Hebm5ujfvz/mzp2LDh061HPLiYiIiLRTiwih3377LcLDw+tUJyEhARMmTEBBQQG6deuGXr164eLFi9i8eTMiIyPx448/wtTUVK3e2rVrERgYCH19fbz44osAgNOnT2PevHmYO3cu5syZo1YnLy8Pb731FpKSkmBnZwcfHx+kpKTg4MGDCA8Px+7du+Hs7Fy/jSciIiLSQi0ihD7//PNwdHSEq6srunfvjlGjRuHBgwfVzl9eXo6AgAAUFBQgICAA06ZNAwCUlpZi3rx5OHHiBFauXInPPvtMVC8uLg6BgYEwMzPDnj174ODgAABITk7GuHHjsH79enh6eqJnz56ieitWrEBSUhK8vLywbt066OvrAwA2bdqEr7/+GgsXLsShQ4cgl3P0AxERERHQQsaETps2DfPnz8fgwYPRtm3bx84fERGBGzduwNHREVOnThXK9fX1sWzZMujq6mL//v3IysoS1du6dSsAYMaMGUIABQAHBwdMnz4dALBlyxZRnczMTAQHB0NXVxfLli0TAqiq3Y6Ojrh+/TpOnDhR9w0nIiIi0lItIoTWVUREBADA19cXMplMNM3a2hru7u5QKBSIjIwUyktLS3H69GkAwJAhQ9SWOXToUADAyZMnUVpaKpRHRkairKwM7u7usLa2FtWRyWTCONK6DicgIiIi0mZaGUKvXbsGAHB1ddU4vVu3bgAgXHgEADdu3EBJSQksLS1ha2urVsfW1hYWFhYoLi7GrVu31NalWmZ161LNR0REREQtZExoXd25cwcAYGNjo3G66pS+6ir72tRRTcvOzkZaWhocHR1rVU9VXnld9aWr2/DfDDo6Wvm7g56w5rRfNae2kHZpTvtWc2oLaQ+p9yutDKEFBQUAACMjI43TTUxMRPPVpg4A4bZOmupVd8snTXXqQy6XwdLSpEHLIGoqZmbVf26ItAX3c9J2Uu/jWhlCVaqOB1VRKpXVllVX53H16lKnPioqlMjNLWzwcnR05DyQUqPLzS1CeXnFk24GAO7j1HS4n5O2a4x93MzMqNY9qloZQo2NjZGTk4PCQs2hraioCMD/ekQr/7u6OgBQXFxc53qa6tRXWVnzOPgRVVVeXsH9k7Qe93PSdlLv41o5qER1YVF6errG6aryyhcg2dnZ1Vinunq1XZdq+URERESkpSG0a9euAB497lOThIQEABA9xahz584wMDBAVlYW7t69q1bn7t27yMrKgqGhITp37qy2LtUyq1K1wcnJqR5bQkRERKSdtDKEenl5AQDCwsLUpmVkZCAuLg66uroYOHCgUG5gYID+/fsDAEJDQ9XqHTlyBAAwYMAA0Q3pBw4cCB0dHcTFxSEjI0NUR6lUCsvy8fFp4FYRERERaQ+tDKHe3t6wt7dHYmKi6AlHCoUCS5YsgUKhwOjRo9G6dWtRvSlTpgAAAgMDkZycLJQnJycjKCgIAODv7y+qY2VlhREjRkChUGDp0qVQKBTCtC1btiApKQkODg5CMCYiIiKiFnJh0h9//IGNGzcKr3NycgAAc+bMEXolBw0ahNmzZwMAdHV1sXr1akycOBGrVq1CaGgoOnbsiPj4eOEen4sWLVJbT+/evTF9+nRs2rQJI0eOFHpGz5w5g5KSEsyaNQtubm5q9RYvXoz4+HiEh4fD19cXPXv2REpKCi5fvgwTExOsXr0aOjo6jf6+EBEREbVULSKEZmZmIj4+Xq386tWrwr+7dOkimubq6oqQkBCsX78eUVFRSExMhI2NDfz9/TFr1qxqr1ZfsGABnJ2dsWPHDkRHRwMAXFxcMGnSJI2P8wQAMzMz7N27F4GBgQgLC8OxY8dgbm6OYcOGYd68eejYsWN9N52IiIhIK7WIEDpq1CiMGjWqzvU6deqEVatW1bne0KFDhWfF15apqSkWLVqksYeViIiIiMS0ckwoERERETVvDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIsnpPukGaBOFQoHvv/8eBw8eRGpqKoyNjdGrVy/MmjULLi4uT7p5RERERM0Ge0IbiUKhgL+/P1avXo2srCx4eXmhS5cuOHbsGN544w2cPn36STeRiIiIqNlgT2gj2bp1K86ePYvu3btj+/btMDU1BQD89ttvCAgIwMKFC3H8+HGYmJg84ZYSERERPXnsCW0EZWVl2L59OwBg6dKlQgAFgNdeew2DBg1CZmYm9u/f/4RaSERERNS8MIQ2gvPnzyM7Oxvt27dH9+7d1aYPHToUABAeHi5104iIiIiaJYbQRnDt2jUAQLdu3TROV12UpJqPiIiI6GnHMaGN4M6dOwAAGxsbjdNV5dnZ2SgoKKjXuFC5XIbWrRs+nlQme/T/D6Z4o7y8osHLo6ebjs6j37Hm5kZQKp9wY/4/7uPU2Lifk7ZrzH1cLpfVel6G0EZQUFAAADAyMtI43djYWDRvfUKoTCaDjk7t/7CPY25q2GjLIpLLm99JFe7j1Ni4n5O2k3ofb36fqBZI+f9/NshkjRcSiYiIiLQZQ2gjUPVsFhYWapxeVFSkNi8RERHR04whtBHY2toCANLT0zVOV5VbWFgwhBIRERGBIbRRdO3aFQCQkJCgcfqVK1cAAE5OTpK1iYiIiKg5YwhtBC+88AIsLCyQmpqKy5cvq00PDQ0FAPj4+EjdNCIiIqJmiSG0Eejq6uLtt98GACxbtgz5+fnCtMOHD+PEiROwtLTE6NGjn1QTiYiIiJoVmVLZXO561rKVlpZiypQpOHfuHKysrNC7d288ePAAsbGx0NPTw8aNGzFw4MAn3UwiIiKiZoEhtBGVlpZi27ZtOHToEFJTU2FsbAx3d3fMnj272qcpERERET2NGEKJiIiISHIcE0pEREREkmMIJSIiIiLJMYQSERERkeQYQomIiIhIcgyh9NRav349nJyccODAgSfdFCI10dHRcHJywuLFi590UwAAt2/fhpOTEyZOnPikm0L0xPH7o3EwhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUGr2HjcWTdPYuQMHDsDJyQnr169HSkoKFi5ciAEDBqBr167Yvn272jKuXLmCGTNmoE+fPnj++efxxhtv4PDhwxrXFxMTg//85z8YPnw4+vTpA1dXV3h7e+OTTz7B7du3NdaZOHEinJyccPv2bURGRuKtt96Cm5sbXnjhBUyZMgWXLl2q+xtDT1xycjI+/PBDvPLKK+jRowd69+6NIUOG4MMPPxT9TZ2cnODt7a1xGbUZa5mZmYklS5bA09MT3bt3h6+vL4KCglBaWqo2b+V9TRNvb284OTmplavaqFAoEBQUhFdffRU9evTA8OHD1eYtLCzEihUr4O3tje7du8Pb2xtfffUV8vPz1ea9d+8etmzZgrfffhsvvfQSXF1d0adPH7z77ruIiIjQ2MbKn9979+7hww8/xIsvvoju3btjyJAh2LFjR7XvFbVMj9v/HjdGuvI+U1nlsZtXrlzBrFmz0K9fPzg7O+P48eMAgFu3bmHDhg0YN24cBgwYAFdXV/Tv3x8zZ85EbGxs0274U073STeAqCndunULo0ePhomJCXr16oWioiIYGRmJ5vnrr7+wdOlS2Nra4sUXX0RGRgZiY2Px/vvvIyUlBTNmzBDNv3z5ciQlJcHJyQm9evWCTCbD9evXsW/fPhw9ehQ//fQTunTporE9e/fuxZYtW+Ds7AxPT08kJyfj1KlTiIuLw/79++Hg4NBk7wU1roSEBLz55psoLi6Go6MjvLy8UFZWhrt37+LgwYPo0KEDunfv3uD1ZGdnY8yYMcjPz0efPn1QWlqK6OhorFmzBrGxsdi0aRN0dHQaYYuAiooKzJkzB1FRUejduzeee+45KBQK0TwKhQKTJk3C33//jb59+6Jbt26Ijo7Gd999h7Nnz2Lnzp0wMTER5j927BhWrVqFTp06oUuXLnBzc8O9e/cQHR2NM2fOYNGiRfD399fYnjt37mD06NEAgO7du6OoqAhxcXH44osvkJeXh9mzZzfKdlPzUJv9r77i4uKwZMkS2NnZoV+/fsjKyoKu7qMItHfvXmzbtg3PPvssunbtChMTE9y+fRsRERH4448/sGrVKrz66quN0g4SYwglrfbbb79hzJgxWLp0KfT09DTOs3fvXrzzzjv44IMPIJc/OjkQHR2NqVOn4ptvvsHAgQPh4uIizD9v3jy4ubnB3NxcKFMqldi3bx+WLFmCL774At99953Gdf3www8IDAyEl5eXUO/TTz/Fnj17sHXrVixfvryxNp2a2I4dO1BcXIzFixfj3XffFU3LyMhAdnZ2o6znxIkT6N27N4KCgmBqagoASE9Px9tvv42TJ09i9+7dePvttxtlXXfv3oVcLsfhw4fRvn17jfNcuHABXbp0QVhYGKytrQEAubm5mDp1Kv766y+sW7cOH374oTB/r169cPDgQTg7O4uWc+vWLbzzzjtYs2YNXn31VbRr105tXQcOHMDo0aOxdOlSGBgYAHgUJiZMmICtW7fi3XffhbGxcaNsOz15tdn/6uuXX37B3LlzMXv2bMhkMtG0l19+GW+++SY6duwoKr948SImT56Mzz77DN7e3modGNRwPB1PWs3CwgIffvhhtQEUAGxsbBAQECAEUADo06cPxowZg4qKCuzatUs0/0svvSQKoAAgk8kwduxYuLm54cyZMxpPSwLA22+/LQRQVb333nsPwKPgSy1HZmYmAKBv375q09q0aYPnnnuuUdYjk8mwdOlSIYAC/9tnATT6qemAgIDHBoDFixcLARQAzMzM8MknnwAA9u3bh+LiYmGas7OzWgAFAHt7e8yaNQtlZWXVnpa3tbXFp59+KgRQAHB3d4enpycKCws5jEUL1Wb/qw8HBwfMmjVLLYACwAsvvKAWQAGgR48eGD9+PHJycnh8biLsCSWt9uKLL4pODWri6+sLfX19tfLXX38du3bt0jgm6OHDh4iIiEBycjLy8vJQXl4ulFdUVCAlJUXUe6oycOBAtbLWrVvDwsIC9+/fr+1mUTPQrVs3REZGYsmSJXjvvffg4eGhcT9qKGdnZ42B9pVXXoGRkRFSU1Nx7949tG3btlHWN3jw4Bqnm5ubY9CgQWrlrq6u6NKlC27cuIErV67A3d1dmKZQKHDmzBnEx8fjwYMHUCgUUCqVyMjIAADcvHlT47r69u2r8T3t0qULIiMjhfqkPR63/9WXj4+PqKOhquLiYvz555+4fPkysrKyhGEA//zzD4BHPffU+BhCSatpOsVXlZ2dXY3l6enpovLdu3djxYoVKCkpqXaZ1fWEVtceExOTRjt9S9Lw9/dHfHw8Tp8+jSlTpkBfXx89evRA//79MWrUqFrte7VR3f4pk8nQrl073LhxA+np6Y0SQq2srES9jprY2tpWO83Ozg43btzAvXv3hLLk5GTMmjWrxi/xgoICjeU2NjYay1U/LDVdmEUtV232v/qq6fMYFxeH+fPn19gRUN0xnRqGIZRavIqKimqnGRoaPra+ptMzwKPxmlWnX7x4EZ9//jmMjY3xySefoG/fvmjTpo2wngULFuD3338X6tZ2XdTymJiYYNu2bYiPj8cff/yBmJgYxMfHCxcLrVmzBj4+Po9dTnX7ikpN+4ymffRxmurzUrk9lb333nu4desW/vWvf+HNN99Ep06dYGJiArlcjpMnT8Lf37/a96CmnivSPrXZ/6pT035d07ILCwsxd+5cPHz4ENOmTcNrr70GOzs7GBsbQy6XY8+ePVi6dOljP6dUPwyh1OypxnNW11ty9+7dBi0/LS1NY/mdO3cAQNTDdPToUSiVSixYsABjxoxRq6M6dUNPj549e6Jnz54AHu2jW7ZsQWBgIJYsWSKEUD09vWr3X9V+Vp3q9k+lUin00lfeR2v6vJSVlTX4FHZ17QHUPzPJycm4fv06unXrhi+++EJt/pSUlAa1hZ4eTfU9EBsbi4cPH8LX11cYZ10Zj+lNiz8zqdmztLSEnp4eUlNTNd6u49SpUw1aflhYmMbl/vrrrwAeXd2rkpOTA0DzacK///4b165da1BbqGUzMTHB/PnzYWRkhAcPHggXL1lbWyM7O1t4Xdnj9t+rV6/i77//Vis/fvw4ioqK0KFDB1EIVV0wpGmcZVRUFMrKyuq0TVXl5OTgzz//VCtPSEjAjRs3YGxsjG7dugnzAtWfCv3tt98a1BZ6etS0XwPA6dOn67Xcmo7ppaWlOHbsWL2WS7XDEErNnr6+Ptzc3JCbm4vvv/9eNO2XX36p9qbytXX37l18/fXXotM5MTEx+PnnnyGXyzF+/HihXHX/z3379onGoz18+BCLFy9u8Bc8tRw//vijxl6Ss2fPoqioCCYmJmjVqhWAR3dbAIB169aJTutFRkbihx9+qHE9SqUSn332mWhM2r1797Bq1SoAULvJvWpdW7duFdW5ceMG/vOf/9RlE6u1YsUKUY9qXl4eli1bBgAYM2aMcOrT3t4ecrkcUVFRoiBdUVGBDRs24Pz5843SHtJ+7du3h52dHa5fv47ff/9dKFftSxcuXKjXclXH9LCwMNGY0NLSUnz++edITU1tWMOpRjwdTy3C3Llz8e6772L16tUICwuDnZ0d/v77b/zzzz949913sW3btnove+zYsdi5cyciIiLg6uqK+/fvIzY2FhUVFXj//feFXh0AGDVqFLZv347IyEi8/PLL6NmzJ0pKSnDu3Dm0bdsWgwcPFp7CQdpt7969+Oyzz2Bvb4/nnnsOBgYGSEtLQ3x8PIBH44NVpxCnTp2K0NBQ/PTTT4iJiYGDgwNSU1Nx9epVTJ06FZs3b652PV5eXkhKSsLgwYPh4eEBhUKBs2fPorCwEJ6enpgwYYJo/ldffRXbtm3DpUuX4OfnBzc3N2RlZeHixYt4+eWXoVAoajyl/jjPP/88ysvL4evri759+0JXVxfR0dHIzs6Gi4uLcMsx4NGdH8aNG4cff/wRI0aMQJ8+fWBmZoZLly7hzp07mDx5coM+u/R0mT9/PhYtWoSAgAD8+OOPsLS0xNWrV5GZmYmJEydi586ddV5mt27dMGjQIERGRsLPzw8eHh4wMDDA+fPnkZeXV+/lUu2wJ5RaBA8PD2zduhXu7u7CU4asra2xY8cOvPTSSw1a9vPPP4+ffvoJHTt2xJ9//omLFy/C1dUVa9asUXtakrm5OX755ReMHj0aOjo6OHHiBK5fv44xY8Zg3759Qs8Xab/33nsPb7zxBvT19RETE4Njx47hwYMHePnll7F7925ROOzSpQt27dqFAQMG4O7du/jzzz9hYGCAjRs3YuzYsTWux8LCAvv27YO3tzfOnz+PkydPwtraGvPnz8fGjRvVnpakr6+P7du3Y9SoUaioqMCJEyfw4MEDvPfee1i5cmWDt1tfXx8//PADxowZg6tXryIiIgLGxsaYPHmy2tOSAOCTTz7BkiVL0KVLF8TFxeHMmTNwcHDA7t27G/zZpafL66+/jjVr1sDFxQUXL14UHuX5888/a7wlXm1t2LAB8+bNg42NDc6cOYOYmBi4u7tj//79DVouPZ5MyUu+iIiIiEhi7AklIiIiIskxhBIRERGR5BhCiYiIiEhyDKFEREREJDmGUCIiIiKSHEMoEREREUmOIZSIiIiIJMcQSkRERESSYwglIiIiIskxhBIRaSnVYw29vb2fdFOIiNQwhBIRSWDx4sVwcnJ67LPiKxs+fDicnJzwn//8pwlbRkT0ZDCEEhFJYPTo0QCAv/76C8nJyY+dPyEhAdeuXQMAjBo1qknbRkT0JDCEEhFJoHfv3ujUqRMAIDg4+LHzHzhwAADQtWtXuLi4NGnbiIieBIZQIiKJjBw5EgBw8OBBlJeXVztfaWkpfv31VwDsBSUi7aX7pBtARPS0GDVqFNatW4f79+/j9OnTGDhwoMb5Tpw4gezsbOjp6WHYsGHIzMzE8ePH8ccffyA5ORn37t0DALRv3x6DBg3ClClT0Lp16zq1xcnJCQAQHh6O9u3bq00/cOAAPvzwQ3h4eGDnzp1q0ysqKvDrr7/i4MGDSEhIQH5+PiwtLdG7d2/4+/tr7L2tqKhASEgIgoODkZSUhPz8fLRq1Qpt2rRBjx49MGzYMPTt27dO20FELRdDKBGRRNq2bYsXX3wRJ0+eRHBwcLUhVHUq3sfHB5aWlti1axc+//xz6OnpoU2bNnBwcEB+fj5u3bqF69ev47fffsPu3bs1hsmmkJ+fj7lz5+LMmTMAgDZt2uC5555DSkoKfv/9d4SFhWH58uV4/fXXRfX+7//+TxiK0LZtW7Rv3x75+fm4ffs2kpKSUFJSwhBK9BRhCCUiktCoUaNw8uRJhIeHIzc3F2ZmZqLpGRkZOHnypDAvAPTo0QObN29Gv379oK+vL8ybmZmJNWvWYN++fVi2bBk2b94syTZ8/PHHOHPmDFxcXLBs2TJ0794dwKOezp07d+LLL7/ERx99BFdXV3Tp0gUAcO3aNQQHB8PU1BQbN25Enz59hOUplUrExsYiIyNDkvYTUfPAMaFERBIaPHgwLCwsUFJSgt9//11tumq8aNu2bTFgwAAAj0LooEGDRAEUAFq3bo3PP/8cbdu2xZ9//okHDx40efsvXryII0eOwMLCAkFBQUIABQC5XI5JkyZh/PjxKC0txfbt24VpqjsC9O3bVxRAAUAmk6F3794YOnRok7efiJoP9oQSEUlIX18fw4YNw86dO3HgwAG8+eaboumq09UjR46Ejo6OUF5SUoKjR48iJiYGaWlpKCoqglKpBAAUFBRAqVQiISGh2lP8jeXIkSMAAC8vL7Rt21bjPK+88gp27tyJ6Ohooaxdu3YAHt2i6tatW7C3t2/SdhJR88cQSkQksdGjR2Pnzp24ePEikpOT4eDgAACIj4/H33//DeB/V9IDj3oRp02bhtu3b9e43Ozs7CZrs0piYiIA4Ny5c2oBWqWkpAQAkJ6eLpQ9//zz6NWrF2JjYzFkyBC4u7ujd+/e6NmzJ3r16gVTU9MmbzsRNS8MoUREEuvatSu6du2Kq1ev4sCBA1i0aBGA/12Q1KtXL6GnsKKiAnPnzsXt27fRrVs3zJkzB926dYOlpaVwen78+PGIjY1FWVlZk7c9JycHAJCWloa0tLQa5y0uLhb+LZfLsWnTJmzevBkhISGIiYlBTEwMAMDAwACvvfYaFi1aBEtLy6ZrPBE1KwyhRERPwOjRo/Gf//wHhw4dwoIFC1BWVobDhw8L01RUvaWGhobYunWrxlsxNaQHVHVKv6qioiKN5cbGxgAeXek+adKkOq3L1NQUCxYswIIFC/DPP/8gLi4Op0+fxrFjx7B//37cunULu3btglzOyxWIngb8pBMRPQHDhg2Dvr4+7t+/j1OnTuHYsWPIzc2FsbEx/Pz8hPlUp+AdHBw0BtCsrCzcvHmzzutXhcmHDx9qnF7dMlX3Fz1//nyd11lZp06dMGrUKKxevRr79u2DTCZDXFyccLqfiLQfQygR0RNgYWEBb29vAI8uRlKdih8yZIgQEAHA0NAQwKNbN2nqtdy+fXuNT1+qjuoRon/99ZfatLy8PI1X7gMQAnJ4eDiSkpLqvF5NnJ2d0apVKwDA/fv3G2WZRNT8MYQSET0hqtPu4eHhiIqKEpWpuLm5QU9PD/fv38eaNWuEwKm6J+fmzZthYGBQ53WrAvDWrVtx7do1oTwjIwMLFixAXl6exnq9evWCn58fFAoFpkyZgoiICLVwfPv2bWzduhU///yzUHbw4EF88803wq2aVBQKBbZu3Yrc3Fzo6urC2dm5zttCRC0Tx4QSET0hAwYMgI2NjXAVub29Pdzd3UXzWFlZYerUqdi4cSM2bdqEffv2wc7ODmlpacjKysKYMWPwzz//4Ny5c3Va97vvvotDhw4hNTUVI0eORKdOnWBgYIC///4b1tbWmD17NtauXaux7pdffonS0lJERERg5syZsLCwQIcOHVBRUYF79+4J9yudM2eOUCcrKwsbN27Exo0bYWFhATs7OyiVSty+fRu5ubkAgEWLFlV72yci0j4MoURET4hcLseIESMQFBQEQL0XVOW9996Dra0tdu/ejeTkZNy6dQvPPvssAgICMGbMGEycOLHO627VqhV++uknfPPNN/jjjz9w+/ZtPPPMMxg7dizmzp2LEydOVFvXyMgIGzduxIkTJ3DgwAHEx8fj2rVrMDExgbW1Nfr27Qtvb28MGjRIqOPr64uKigpER0fj77//xs2bN6FQKPDMM89gwIABGD9+PHr16lXn7SCilkumrO7SSCIiIiKiJsIxoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJJjCCUiIiIiyTGEEhEREZHkGEKJiIiISHIMoUREREQkOYZQIiIiIpIcQygRERERSY4hlIiIiIgkxxBKRERERJL7f/IRpviPqqjKAAAAAElFTkSuQmCC",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
@@ -3330,9 +1767,17 @@
"plt.show()"
]
},
+ {
+ "cell_type": "markdown",
+ "id": "2d171422",
+ "metadata": {},
+ "source": [
+ "#### Train machine learning (ML) models"
+ ]
+ },
{
"cell_type": "code",
- "execution_count": 45,
+ "execution_count": 51,
"id": "b01d0d23-cf4c-43d6-8072-e204ddd8f8fe",
"metadata": {
"tags": []
@@ -3751,7 +2196,7 @@
"LGBMClassifier(n_estimators=500, verbose=0)"
]
},
- "execution_count": 45,
+ "execution_count": 51,
"metadata": {},
"output_type": "execute_result"
}
@@ -3759,16 +2204,8 @@
"source": [
"rf_clf.fit(X_train_0, Y_train_0)\n",
"cboost_clf.fit(X_train_0, Y_train_0)\n",
- "lgbm_clf.fit(X_train_0, Y_train_0)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 34,
- "id": "127d1265-fbfe-4678-852b-56324a3300c2",
- "metadata": {},
- "outputs": [],
- "source": [
+ "lgbm_clf.fit(X_train_0, Y_train_0)\n",
+ "\n",
"# rf_clf.fit(X_train_0_smot, Y_train_0_smot)\n",
"# cboost_clf.fit(X_train_0_smot, Y_train_0_smot)\n",
"# lgbm_clf.fit(X_train_0_smot, Y_train_0_smot)\n",
@@ -3784,12 +2221,20 @@
},
{
"cell_type": "code",
- "execution_count": 72,
+ "execution_count": null,
"id": "cf5330cd-0e27-4e4a-b20e-ba52f35bffd9",
"metadata": {
"tags": []
},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "0.781\n"
+ ]
+ }
+ ],
"source": [
"#print('training lgbm')\n",
"#selected_feature_lgbm = train_with_important_feature({'model_name': 'lgbm', 'model': lgbm_clf}, X_train_0, Y_train_0, X_test_0, Y_test_0, init_num_feature=15)\n",
@@ -3801,7 +2246,7 @@
},
{
"cell_type": "code",
- "execution_count": 73,
+ "execution_count": null,
"id": "2152f09f-b5a8-44ee-976c-f61551bdbff2",
"metadata": {
"tags": []
@@ -4242,33 +2687,21 @@
},
{
"cell_type": "code",
- "execution_count": 54,
- "id": "03e95f14-9f4e-4ae7-bd24-9a52a5c73c83",
- "metadata": {
- "tags": []
- },
- "outputs": [],
- "source": [
- "df_test_misclass = df_test_result[~(df_test_result['type_of_area']==df_test_result['type_of_area_pred_rf'])]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 158,
+ "execution_count": 1,
"id": "ab20cb39-9d15-48a4-8fdb-7d3f18495044",
"metadata": {
"tags": []
},
"outputs": [
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "number of data point: (22352, 22)\n",
- "number of known station: (12455, 22)\n",
- "number of unlabelled station: (9897, 22)\n",
- "training shape: (11455, 22)\n",
- "test shape: (1000, 22)\n"
+ "ename": "NameError",
+ "evalue": "name 'data' is not defined",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[1], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m labelled_data \u001b[38;5;241m=\u001b[39m \u001b[43mdata\u001b[49m[\u001b[38;5;241m~\u001b[39m(data[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtype_of_area\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;241m==\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124munknown\u001b[39m\u001b[38;5;124m'\u001b[39m)]\n\u001b[1;32m 2\u001b[0m unlabelled_data \u001b[38;5;241m=\u001b[39m data[(data[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtype_of_area\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;241m==\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124munknown\u001b[39m\u001b[38;5;124m'\u001b[39m)]\n\u001b[1;32m 3\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnumber of data point: \u001b[39m\u001b[38;5;124m'\u001b[39m, data\u001b[38;5;241m.\u001b[39mshape)\n",
+ "\u001b[0;31mNameError\u001b[0m: name 'data' is not defined"
]
}
],
@@ -4289,7 +2722,7 @@
},
{
"cell_type": "code",
- "execution_count": 57,
+ "execution_count": null,
"id": "e08c059d-d44f-41f9-b9fe-be925a3d034b",
"metadata": {
"collapsed": true,
@@ -4524,20 +2957,7 @@
}
],
"source": [
- "# we first selecte the most recent infos and analyse the correlation between variable\n",
- "selected_colunms_1 = ['mean_topography_srtm_alt_90m_year1994', \n",
- " 'max_topography_srtm_relative_alt_5km_year1994', \n",
- " 'stddev_topography_srtm_relative_alt_5km_year1994', \n",
- " 'min_topography_srtm_relative_alt_5km_year1994', \n",
- " 'climatic_zone_year2016', \n",
- " 'distance_to_major_road_year2020', \n",
- " 'mean_stable_nightlights_1km_year2013',\n",
- " 'mean_stable_nightlights_5km_year2013', \n",
- " 'mean_population_density_250m_year2015',\n",
- " 'mean_population_density_5km_year2015', \n",
- " 'max_population_density_25km_year2015',\n",
- " 'mean_nox_emissions_10km_year2015', \n",
- " 'type_of_area']\n",
+ "\n",
"colunms_1 = [\n",
" 'mean_topography_srtm_alt_90m_year1994', \n",
" 'max_topography_srtm_relative_alt_5km_year1994', \n",
@@ -6251,12 +4671,24 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 1,
"id": "7f50ad6f-3996-4d54-8c45-66ea7b9ec195",
"metadata": {
"tags": []
},
- "outputs": [],
+ "outputs": [
+ {
+ "ename": "NameError",
+ "evalue": "name 'df_train' is not defined",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[1], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m lat_train \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(\u001b[43mdf_train\u001b[49m\u001b[38;5;241m.\u001b[39mindex\u001b[38;5;241m.\u001b[39mget_level_values(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mlat\u001b[39m\u001b[38;5;124m'\u001b[39m))\n\u001b[1;32m 2\u001b[0m lon_train \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(df_train\u001b[38;5;241m.\u001b[39mindex\u001b[38;5;241m.\u001b[39mget_level_values(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mlon\u001b[39m\u001b[38;5;124m'\u001b[39m))\n\u001b[1;32m 4\u001b[0m lat_test \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(df_test\u001b[38;5;241m.\u001b[39mindex\u001b[38;5;241m.\u001b[39mget_level_values(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mlat\u001b[39m\u001b[38;5;124m'\u001b[39m))\n",
+ "\u001b[0;31mNameError\u001b[0m: name 'df_train' is not defined"
+ ]
+ }
+ ],
"source": [
"lat_train = list(df_train.index.get_level_values('lat'))\n",
"lon_train = list(df_train.index.get_level_values('lon'))\n",
@@ -6268,8 +4700,8 @@
"lat_no2 = list(df_no2.index.get_level_values('lat'))\n",
"lon_no2 = list(df_no2.index.get_level_values('lon'))\n",
"\n",
- "lat_pm = list(df_pm_test.index.get_level_values('lat'))\n",
- "lon_pm = list(df_pm_test.index.get_level_values('lon'))"
+ "lat_pm = list(df_pm.index.get_level_values('lat'))\n",
+ "lon_pm = list(df_pm.index.get_level_values('lon'))"
]
},
{
diff --git a/catboost_info/catboost_training.json b/catboost_info/catboost_training.json
index 522cc3fe583f68b38596086d8f53f0b02247a260..1e6ebdf9bdd13dc4cb026cb70bb7ce9566c55883 100644
--- a/catboost_info/catboost_training.json
+++ b/catboost_info/catboost_training.json
@@ -1,504 +1,504 @@
{
"meta":{"test_sets":[],"test_metrics":[],"learn_metrics":[{"best_value":"Min","name":"MultiClass"}],"launch_mode":"Train","parameters":"","iteration_count":500,"learn_sets":["learn"],"name":"experiment"},
"iterations":[
-{"learn":[1.068921125],"iteration":0,"passed_time":0.09501817241,"remaining_time":47.41406803},
-{"learn":[1.042920351],"iteration":1,"passed_time":0.1272402288,"remaining_time":31.68281696},
-{"learn":[1.01801279],"iteration":2,"passed_time":0.1583119472,"remaining_time":26.22701258},
-{"learn":[0.9955724455],"iteration":3,"passed_time":0.1895189122,"remaining_time":23.50034511},
-{"learn":[0.9736690809],"iteration":4,"passed_time":0.2214313789,"remaining_time":21.92170651},
-{"learn":[0.9539294313],"iteration":5,"passed_time":0.254059332,"remaining_time":20.91755167},
-{"learn":[0.9356639615],"iteration":6,"passed_time":0.2871532434,"remaining_time":20.22379272},
-{"learn":[0.9191377944],"iteration":7,"passed_time":0.3205513684,"remaining_time":19.71390916},
-{"learn":[0.9031989432],"iteration":8,"passed_time":0.3551142354,"remaining_time":19.3734544},
-{"learn":[0.8878384884],"iteration":9,"passed_time":0.3895673295,"remaining_time":19.08879915},
-{"learn":[0.8738034802],"iteration":10,"passed_time":0.4243528134,"remaining_time":18.86441143},
-{"learn":[0.8609447068],"iteration":11,"passed_time":0.4585863567,"remaining_time":18.6491785},
-{"learn":[0.8494076364],"iteration":12,"passed_time":0.4927037581,"remaining_time":18.45744078},
-{"learn":[0.8376116526],"iteration":13,"passed_time":0.5261370976,"remaining_time":18.26447353},
-{"learn":[0.8262527645],"iteration":14,"passed_time":0.5604266206,"remaining_time":18.12046073},
-{"learn":[0.8156703929],"iteration":15,"passed_time":0.5943689281,"remaining_time":17.97966008},
-{"learn":[0.8061405245],"iteration":16,"passed_time":0.6288167044,"remaining_time":17.86579225},
-{"learn":[0.7962708289],"iteration":17,"passed_time":0.6650916834,"remaining_time":17.8096773},
-{"learn":[0.787517053],"iteration":18,"passed_time":0.7002040567,"remaining_time":17.72621849},
-{"learn":[0.7790596225],"iteration":19,"passed_time":0.7360828232,"remaining_time":17.66598776},
-{"learn":[0.7707419955],"iteration":20,"passed_time":0.7746310381,"remaining_time":17.66896511},
-{"learn":[0.7628618298],"iteration":21,"passed_time":0.809716054,"remaining_time":17.59292154},
-{"learn":[0.7568458468],"iteration":22,"passed_time":0.8467397937,"remaining_time":17.56064703},
-{"learn":[0.7496544902],"iteration":23,"passed_time":0.8824789368,"remaining_time":17.50249891},
-{"learn":[0.7429806121],"iteration":24,"passed_time":0.920280358,"remaining_time":17.4853268},
-{"learn":[0.7365733331],"iteration":25,"passed_time":0.9572626826,"remaining_time":17.45163506},
-{"learn":[0.7301107638],"iteration":26,"passed_time":0.9940774976,"remaining_time":17.41476505},
-{"learn":[0.7244057584],"iteration":27,"passed_time":1.031475359,"remaining_time":17.38772748},
-{"learn":[0.7183963567],"iteration":28,"passed_time":1.068413034,"remaining_time":17.35250134},
-{"learn":[0.7133137141],"iteration":29,"passed_time":1.106583337,"remaining_time":17.33647228},
-{"learn":[0.7081582063],"iteration":30,"passed_time":1.146441298,"remaining_time":17.34454738},
-{"learn":[0.7029195396],"iteration":31,"passed_time":1.184023579,"remaining_time":17.31634484},
-{"learn":[0.6981650667],"iteration":32,"passed_time":1.220392211,"remaining_time":17.27039887},
-{"learn":[0.694229036],"iteration":33,"passed_time":1.259242513,"remaining_time":17.25902974},
-{"learn":[0.6898580387],"iteration":34,"passed_time":1.296335758,"remaining_time":17.2227465},
-{"learn":[0.6858643858],"iteration":35,"passed_time":1.334293853,"remaining_time":17.19756522},
-{"learn":[0.6818898314],"iteration":36,"passed_time":1.374870445,"remaining_time":17.2044599},
-{"learn":[0.6782096663],"iteration":37,"passed_time":1.413393171,"remaining_time":17.18388539},
-{"learn":[0.6749976546],"iteration":38,"passed_time":1.455058458,"remaining_time":17.19953716},
-{"learn":[0.6717289324],"iteration":39,"passed_time":1.494085983,"remaining_time":17.1819888},
-{"learn":[0.6685944214],"iteration":40,"passed_time":1.532498059,"remaining_time":17.15650266},
-{"learn":[0.6656078326],"iteration":41,"passed_time":1.572755708,"remaining_time":17.15052653},
-{"learn":[0.6627272627],"iteration":42,"passed_time":1.610193023,"remaining_time":17.11298166},
-{"learn":[0.6596591536],"iteration":43,"passed_time":1.648810355,"remaining_time":17.08767095},
-{"learn":[0.6567951356],"iteration":44,"passed_time":1.688017614,"remaining_time":17.06773366},
-{"learn":[0.6539799136],"iteration":45,"passed_time":1.726591807,"remaining_time":17.04071044},
-{"learn":[0.6512287405],"iteration":46,"passed_time":1.766039823,"remaining_time":17.02161787},
-{"learn":[0.6483621371],"iteration":47,"passed_time":1.80382067,"remaining_time":16.98597798},
-{"learn":[0.646045433],"iteration":48,"passed_time":1.841594554,"remaining_time":16.95018661},
-{"learn":[0.6436106576],"iteration":49,"passed_time":1.880428518,"remaining_time":16.92385666},
-{"learn":[0.6412037332],"iteration":50,"passed_time":1.918243538,"remaining_time":16.88806566},
-{"learn":[0.6385693614],"iteration":51,"passed_time":1.958382904,"remaining_time":16.87222194},
-{"learn":[0.6359669929],"iteration":52,"passed_time":1.999263969,"remaining_time":16.86171687},
-{"learn":[0.6336380251],"iteration":53,"passed_time":2.038427085,"remaining_time":16.83589777},
-{"learn":[0.6317596332],"iteration":54,"passed_time":2.076199805,"remaining_time":16.79834388},
-{"learn":[0.6300162524],"iteration":55,"passed_time":2.11449159,"remaining_time":16.76489761},
-{"learn":[0.6279497237],"iteration":56,"passed_time":2.154021886,"remaining_time":16.74090694},
-{"learn":[0.626202365],"iteration":57,"passed_time":2.192757561,"remaining_time":16.71032486},
-{"learn":[0.6245478766],"iteration":58,"passed_time":2.231588619,"remaining_time":16.68017934},
-{"learn":[0.6232750103],"iteration":59,"passed_time":2.270541085,"remaining_time":16.65063462},
-{"learn":[0.6216488728],"iteration":60,"passed_time":2.308857332,"remaining_time":16.61620277},
-{"learn":[0.6199963519],"iteration":61,"passed_time":2.346679755,"remaining_time":16.57815698},
-{"learn":[0.6183914027],"iteration":62,"passed_time":2.385484171,"remaining_time":16.54692989},
-{"learn":[0.6167257552],"iteration":63,"passed_time":2.424395246,"remaining_time":16.51619262},
-{"learn":[0.6151270752],"iteration":64,"passed_time":2.465687931,"remaining_time":16.50114231},
-{"learn":[0.6136229865],"iteration":65,"passed_time":2.50377904,"remaining_time":16.46424399},
-{"learn":[0.6124914371],"iteration":66,"passed_time":2.542541156,"remaining_time":16.43164657},
-{"learn":[0.6112142134],"iteration":67,"passed_time":2.582221249,"remaining_time":16.4046997},
-{"learn":[0.6100471536],"iteration":68,"passed_time":2.619579145,"remaining_time":16.36287843},
-{"learn":[0.6088553938],"iteration":69,"passed_time":2.658224894,"remaining_time":16.32909578},
-{"learn":[0.6077217705],"iteration":70,"passed_time":2.696442646,"remaining_time":16.29259007},
-{"learn":[0.6065233622],"iteration":71,"passed_time":2.736162971,"remaining_time":16.26496877},
-{"learn":[0.6055522725],"iteration":72,"passed_time":2.774878856,"remaining_time":16.23114071},
-{"learn":[0.6042869014],"iteration":73,"passed_time":2.812255125,"remaining_time":16.18946869},
-{"learn":[0.6029862228],"iteration":74,"passed_time":2.850278432,"remaining_time":16.15157778},
-{"learn":[0.6020286548],"iteration":75,"passed_time":2.889392126,"remaining_time":16.1197666},
-{"learn":[0.6007111945],"iteration":76,"passed_time":2.929351152,"remaining_time":16.09240958},
-{"learn":[0.5995650312],"iteration":77,"passed_time":2.967938484,"remaining_time":16.05730821},
-{"learn":[0.598267565],"iteration":78,"passed_time":3.008289513,"remaining_time":16.03151753},
-{"learn":[0.5973018607],"iteration":79,"passed_time":3.052917179,"remaining_time":16.02781519},
-{"learn":[0.5964098279],"iteration":80,"passed_time":3.093401134,"remaining_time":16.00166759},
-{"learn":[0.5952645015],"iteration":81,"passed_time":3.13322147,"remaining_time":15.97178749},
-{"learn":[0.5939932919],"iteration":82,"passed_time":3.174064855,"remaining_time":15.94680776},
-{"learn":[0.5927819224],"iteration":83,"passed_time":3.214882243,"remaining_time":15.92132158},
-{"learn":[0.5916101899],"iteration":84,"passed_time":3.253718647,"remaining_time":15.88580281},
-{"learn":[0.5904890406],"iteration":85,"passed_time":3.292805903,"remaining_time":15.85141446},
-{"learn":[0.5898393665],"iteration":86,"passed_time":3.330855123,"remaining_time":15.81199041},
-{"learn":[0.5889665637],"iteration":87,"passed_time":3.368324585,"remaining_time":15.76988328},
-{"learn":[0.5881368208],"iteration":88,"passed_time":3.407668654,"remaining_time":15.73653727},
-{"learn":[0.5871073965],"iteration":89,"passed_time":3.445511484,"remaining_time":15.69621898},
-{"learn":[0.5861402271],"iteration":90,"passed_time":3.485315383,"remaining_time":15.66476914},
-{"learn":[0.5852869207],"iteration":91,"passed_time":3.523609074,"remaining_time":15.62644024},
-{"learn":[0.584532599],"iteration":92,"passed_time":3.561032227,"remaining_time":15.58430233},
-{"learn":[0.5839068284],"iteration":93,"passed_time":3.600364317,"remaining_time":15.55050971},
-{"learn":[0.5831785979],"iteration":94,"passed_time":3.637974234,"remaining_time":15.50925858},
-{"learn":[0.582371746],"iteration":95,"passed_time":3.678029053,"remaining_time":15.47837227},
-{"learn":[0.581512497],"iteration":96,"passed_time":3.717286794,"remaining_time":15.44398534},
-{"learn":[0.5808661149],"iteration":97,"passed_time":3.755034074,"remaining_time":15.40330304},
-{"learn":[0.5802827802],"iteration":98,"passed_time":3.793199995,"remaining_time":15.36437574},
-{"learn":[0.5798054416],"iteration":99,"passed_time":3.830154309,"remaining_time":15.32061724},
-{"learn":[0.5788760699],"iteration":100,"passed_time":3.869319495,"remaining_time":15.28572751},
-{"learn":[0.5780138828],"iteration":101,"passed_time":3.908805488,"remaining_time":15.25200573},
-{"learn":[0.5771851598],"iteration":102,"passed_time":3.947823625,"remaining_time":15.21636873},
-{"learn":[0.5764314408],"iteration":103,"passed_time":3.987482606,"remaining_time":15.18310685},
-{"learn":[0.5756151644],"iteration":104,"passed_time":4.026864711,"remaining_time":15.14868153},
-{"learn":[0.5747869309],"iteration":105,"passed_time":4.066166925,"remaining_time":15.11386574},
-{"learn":[0.5739388675],"iteration":106,"passed_time":4.104491279,"remaining_time":15.07537451},
-{"learn":[0.5732935814],"iteration":107,"passed_time":4.142449868,"remaining_time":15.03555878},
-{"learn":[0.5724740546],"iteration":108,"passed_time":4.181446529,"remaining_time":14.99950085},
-{"learn":[0.5718664243],"iteration":109,"passed_time":4.218689208,"remaining_time":14.95717083},
-{"learn":[0.5711293162],"iteration":110,"passed_time":4.256706802,"remaining_time":14.91764816},
-{"learn":[0.5705779324],"iteration":111,"passed_time":4.295867292,"remaining_time":14.88211169},
-{"learn":[0.569735314],"iteration":112,"passed_time":4.334200572,"remaining_time":14.84367806},
-{"learn":[0.5686404377],"iteration":113,"passed_time":4.374652044,"remaining_time":14.81241833},
-{"learn":[0.5680178459],"iteration":114,"passed_time":4.415820267,"remaining_time":14.78339829},
-{"learn":[0.5671996108],"iteration":115,"passed_time":4.454239792,"remaining_time":14.74506966},
-{"learn":[0.5665763407],"iteration":116,"passed_time":4.4944508,"remaining_time":14.7126039},
-{"learn":[0.5657710765],"iteration":117,"passed_time":4.533566862,"remaining_time":14.67646222},
-{"learn":[0.5653197317],"iteration":118,"passed_time":4.570969515,"remaining_time":14.63478475},
-{"learn":[0.5646326931],"iteration":119,"passed_time":4.611208534,"remaining_time":14.60216036},
-{"learn":[0.5641755543],"iteration":120,"passed_time":4.648077863,"remaining_time":14.55885546},
-{"learn":[0.5636121495],"iteration":121,"passed_time":4.688368714,"remaining_time":14.52625716},
-{"learn":[0.5629206857],"iteration":122,"passed_time":4.727892631,"remaining_time":14.4911831},
-{"learn":[0.5623847046],"iteration":123,"passed_time":4.764874453,"remaining_time":14.44832899},
-{"learn":[0.5618361757],"iteration":124,"passed_time":4.803544756,"remaining_time":14.41063427},
-{"learn":[0.5613260958],"iteration":125,"passed_time":4.841303571,"remaining_time":14.37021854},
-{"learn":[0.5605573167],"iteration":126,"passed_time":4.881905843,"remaining_time":14.3381959},
-{"learn":[0.559966854],"iteration":127,"passed_time":4.922134642,"remaining_time":14.3049538},
-{"learn":[0.5593610441],"iteration":128,"passed_time":4.960947735,"remaining_time":14.26753186},
-{"learn":[0.5587625708],"iteration":129,"passed_time":5.000178919,"remaining_time":14.23127846},
-{"learn":[0.5580857121],"iteration":130,"passed_time":5.03887804,"remaining_time":14.19348089},
-{"learn":[0.557514446],"iteration":131,"passed_time":5.077015205,"remaining_time":14.154103},
-{"learn":[0.556943387],"iteration":132,"passed_time":5.116521622,"remaining_time":14.11852207},
-{"learn":[0.5563473065],"iteration":133,"passed_time":5.154119811,"remaining_time":14.07767053},
-{"learn":[0.5559281273],"iteration":134,"passed_time":5.19482321,"remaining_time":14.04526275},
-{"learn":[0.5554834032],"iteration":135,"passed_time":5.234040526,"remaining_time":14.00875552},
-{"learn":[0.5547240917],"iteration":136,"passed_time":5.273970474,"remaining_time":13.97409695},
-{"learn":[0.554286871],"iteration":137,"passed_time":5.313866548,"remaining_time":13.93927312},
-{"learn":[0.5539844822],"iteration":138,"passed_time":5.351287441,"remaining_time":13.89794796},
-{"learn":[0.5531296918],"iteration":139,"passed_time":5.391356452,"remaining_time":13.86348802},
-{"learn":[0.5526438864],"iteration":140,"passed_time":5.430330092,"remaining_time":13.82615959},
-{"learn":[0.5521995723],"iteration":141,"passed_time":5.468684967,"remaining_time":13.78724802},
-{"learn":[0.5517475129],"iteration":142,"passed_time":5.50871578,"remaining_time":13.7525282},
-{"learn":[0.5512807855],"iteration":143,"passed_time":5.546574723,"remaining_time":13.71236529},
-{"learn":[0.5505009241],"iteration":144,"passed_time":5.585065202,"remaining_time":13.67378032},
-{"learn":[0.5499843323],"iteration":145,"passed_time":5.62374631,"remaining_time":13.63565886},
-{"learn":[0.5494553371],"iteration":146,"passed_time":5.662631747,"remaining_time":13.59802045},
-{"learn":[0.548825567],"iteration":147,"passed_time":5.702408286,"remaining_time":13.56248457},
-{"learn":[0.5483547038],"iteration":148,"passed_time":5.741567153,"remaining_time":13.52543672},
-{"learn":[0.5479018041],"iteration":149,"passed_time":5.779562381,"remaining_time":13.48564556},
-{"learn":[0.5474866017],"iteration":150,"passed_time":5.818536449,"remaining_time":13.44814053},
-{"learn":[0.5469354154],"iteration":151,"passed_time":5.858478343,"remaining_time":13.41283199},
-{"learn":[0.5464195173],"iteration":152,"passed_time":5.897112572,"remaining_time":13.37449714},
-{"learn":[0.5462957589],"iteration":153,"passed_time":5.901400021,"remaining_time":13.25898966},
-{"learn":[0.5457682748],"iteration":154,"passed_time":5.940702716,"remaining_time":13.22285443},
-{"learn":[0.5453126854],"iteration":155,"passed_time":5.978838694,"remaining_time":13.18410584},
-{"learn":[0.5449104012],"iteration":156,"passed_time":6.017232418,"remaining_time":13.14592815},
-{"learn":[0.5446294358],"iteration":157,"passed_time":6.054449983,"remaining_time":13.10520186},
-{"learn":[0.5441676844],"iteration":158,"passed_time":6.094316209,"remaining_time":13.07020017},
-{"learn":[0.5436853001],"iteration":159,"passed_time":6.132287978,"remaining_time":13.03111195},
-{"learn":[0.54326474],"iteration":160,"passed_time":6.170372318,"remaining_time":12.99227463},
-{"learn":[0.5428189052],"iteration":161,"passed_time":6.208499134,"remaining_time":12.95353523},
-{"learn":[0.5424821708],"iteration":162,"passed_time":6.247019876,"remaining_time":12.91561778},
-{"learn":[0.5420205298],"iteration":163,"passed_time":6.284517964,"remaining_time":12.87559778},
-{"learn":[0.5415372914],"iteration":164,"passed_time":6.323469889,"remaining_time":12.83856008},
-{"learn":[0.5411230188],"iteration":165,"passed_time":6.362178137,"remaining_time":12.80100902},
-{"learn":[0.5404682421],"iteration":166,"passed_time":6.40225323,"remaining_time":12.76616961},
-{"learn":[0.5401062328],"iteration":167,"passed_time":6.441454894,"remaining_time":12.72954181},
-{"learn":[0.5396218645],"iteration":168,"passed_time":6.479524368,"remaining_time":12.69066607},
-{"learn":[0.539240618],"iteration":169,"passed_time":6.516560485,"remaining_time":12.64979388},
-{"learn":[0.5388113906],"iteration":170,"passed_time":6.554250456,"remaining_time":12.61022456},
-{"learn":[0.538518505],"iteration":171,"passed_time":6.591474155,"remaining_time":12.56978792},
-{"learn":[0.5382252585],"iteration":172,"passed_time":6.629156516,"remaining_time":12.53025538},
-{"learn":[0.5376825766],"iteration":173,"passed_time":6.667433353,"remaining_time":12.49185789},
-{"learn":[0.5372177332],"iteration":174,"passed_time":6.705259898,"remaining_time":12.45262552},
-{"learn":[0.5368502891],"iteration":175,"passed_time":6.743864973,"remaining_time":12.41484234},
-{"learn":[0.536433271],"iteration":176,"passed_time":6.782429408,"remaining_time":12.3769757},
-{"learn":[0.5360177821],"iteration":177,"passed_time":6.821010938,"remaining_time":12.33913215},
-{"learn":[0.5355686621],"iteration":178,"passed_time":6.859551861,"remaining_time":12.30120753},
-{"learn":[0.5352466243],"iteration":179,"passed_time":6.896769905,"remaining_time":12.26092428},
-{"learn":[0.5349531898],"iteration":180,"passed_time":6.936151184,"remaining_time":12.22448745},
-{"learn":[0.5345933818],"iteration":181,"passed_time":6.974755544,"remaining_time":12.18666079},
-{"learn":[0.5340879627],"iteration":182,"passed_time":7.0144547,"remaining_time":12.15072208},
-{"learn":[0.5334066003],"iteration":183,"passed_time":7.052588109,"remaining_time":12.11205349},
-{"learn":[0.5331344614],"iteration":184,"passed_time":7.090148218,"remaining_time":12.07241453},
-{"learn":[0.5327729413],"iteration":185,"passed_time":7.131628306,"remaining_time":12.03941553},
-{"learn":[0.5322668217],"iteration":186,"passed_time":7.169345049,"remaining_time":12.00002674},
-{"learn":[0.5314967978],"iteration":187,"passed_time":7.208428675,"remaining_time":11.96292418},
-{"learn":[0.5310815884],"iteration":188,"passed_time":7.2452056,"remaining_time":11.92200498},
-{"learn":[0.5306990817],"iteration":189,"passed_time":7.283140728,"remaining_time":11.88301908},
-{"learn":[0.5303294313],"iteration":190,"passed_time":7.322017842,"remaining_time":11.84556813},
-{"learn":[0.5297662228],"iteration":191,"passed_time":7.359368397,"remaining_time":11.80565347},
-{"learn":[0.5294658769],"iteration":192,"passed_time":7.395985725,"remaining_time":11.76459905},
-{"learn":[0.5290541985],"iteration":193,"passed_time":7.435600898,"remaining_time":11.72831894},
-{"learn":[0.528513619],"iteration":194,"passed_time":7.473072652,"remaining_time":11.6886521},
-{"learn":[0.5280691496],"iteration":195,"passed_time":7.511647346,"remaining_time":11.65071833},
-{"learn":[0.5277436683],"iteration":196,"passed_time":7.549443887,"remaining_time":11.61158121},
-{"learn":[0.5273735004],"iteration":197,"passed_time":7.586779179,"remaining_time":11.5717541},
-{"learn":[0.5269597174],"iteration":198,"passed_time":7.624800579,"remaining_time":11.53298982},
-{"learn":[0.526580619],"iteration":199,"passed_time":7.66281373,"remaining_time":11.4942206},
-{"learn":[0.5261678093],"iteration":200,"passed_time":7.699821227,"remaining_time":11.45396292},
-{"learn":[0.5257991482],"iteration":201,"passed_time":7.740496848,"remaining_time":11.41914882},
-{"learn":[0.5255372878],"iteration":202,"passed_time":7.779164807,"remaining_time":11.38133964},
-{"learn":[0.5252407006],"iteration":203,"passed_time":7.817139621,"remaining_time":11.34251631},
-{"learn":[0.524997375],"iteration":204,"passed_time":7.854779772,"remaining_time":11.30321967},
-{"learn":[0.5247760811],"iteration":205,"passed_time":7.892038149,"remaining_time":11.26339425},
-{"learn":[0.523988662],"iteration":206,"passed_time":7.931937984,"remaining_time":11.22733251},
-{"learn":[0.5235269686],"iteration":207,"passed_time":7.970598282,"remaining_time":11.18949374},
-{"learn":[0.5231391961],"iteration":208,"passed_time":8.012937013,"remaining_time":11.15676876},
-{"learn":[0.522802282],"iteration":209,"passed_time":8.053347119,"remaining_time":11.12128888},
-{"learn":[0.5221873746],"iteration":210,"passed_time":8.091506053,"remaining_time":11.08267891},
-{"learn":[0.5216496981],"iteration":211,"passed_time":8.131254396,"remaining_time":11.04623239},
-{"learn":[0.5210661583],"iteration":212,"passed_time":8.169855916,"remaining_time":11.00820961},
-{"learn":[0.5207763789],"iteration":213,"passed_time":8.207653367,"remaining_time":10.96910684},
-{"learn":[0.5202717603],"iteration":214,"passed_time":8.247834674,"remaining_time":10.9331762},
-{"learn":[0.5197573737],"iteration":215,"passed_time":8.285540842,"remaining_time":10.89395185},
-{"learn":[0.5193722401],"iteration":216,"passed_time":8.325323585,"remaining_time":10.85744965},
-{"learn":[0.5190276841],"iteration":217,"passed_time":8.364824068,"remaining_time":10.82055223},
-{"learn":[0.5188233127],"iteration":218,"passed_time":8.402670554,"remaining_time":10.78150879},
-{"learn":[0.5183115318],"iteration":219,"passed_time":8.443789068,"remaining_time":10.74664063},
-{"learn":[0.5179261404],"iteration":220,"passed_time":8.484376809,"remaining_time":10.71104584},
-{"learn":[0.5175413522],"iteration":221,"passed_time":8.522902978,"remaining_time":10.67282445},
-{"learn":[0.5169007258],"iteration":222,"passed_time":8.562571073,"remaining_time":10.63601878},
-{"learn":[0.5165349089],"iteration":223,"passed_time":8.599874312,"remaining_time":10.59627371},
-{"learn":[0.5163301026],"iteration":224,"passed_time":8.637713605,"remaining_time":10.55720552},
-{"learn":[0.5160285647],"iteration":225,"passed_time":8.675104244,"remaining_time":10.51760426},
-{"learn":[0.5156218452],"iteration":226,"passed_time":8.713012865,"remaining_time":10.47864543},
-{"learn":[0.5152165155],"iteration":227,"passed_time":8.752713609,"remaining_time":10.44183378},
-{"learn":[0.5148155711],"iteration":228,"passed_time":8.791485706,"remaining_time":10.40389793},
-{"learn":[0.5143885804],"iteration":229,"passed_time":8.830168665,"remaining_time":10.36585017},
-{"learn":[0.5140794854],"iteration":230,"passed_time":8.867688006,"remaining_time":10.32644188},
-{"learn":[0.5136827366],"iteration":231,"passed_time":8.905032229,"remaining_time":10.28684757},
-{"learn":[0.5132724458],"iteration":232,"passed_time":8.943878628,"remaining_time":10.24899396},
-{"learn":[0.5129407929],"iteration":233,"passed_time":8.984868525,"remaining_time":10.21356849},
-{"learn":[0.5124816054],"iteration":234,"passed_time":9.027638024,"remaining_time":10.18010245},
-{"learn":[0.5121399791],"iteration":235,"passed_time":9.068527752,"remaining_time":10.14445477},
-{"learn":[0.5118334093],"iteration":236,"passed_time":9.105947931,"remaining_time":10.10491268},
-{"learn":[0.5114700787],"iteration":237,"passed_time":9.144634351,"remaining_time":10.06678235},
-{"learn":[0.5111381929],"iteration":238,"passed_time":9.183637391,"remaining_time":10.02899313},
-{"learn":[0.5108294424],"iteration":239,"passed_time":9.223131929,"remaining_time":9.991726256},
-{"learn":[0.5106367493],"iteration":240,"passed_time":9.262034356,"remaining_time":9.953804557},
-{"learn":[0.5101473498],"iteration":241,"passed_time":9.302644195,"remaining_time":9.917695051},
-{"learn":[0.5098278979],"iteration":242,"passed_time":9.340474545,"remaining_time":9.878608881},
-{"learn":[0.509406577],"iteration":243,"passed_time":9.378374796,"remaining_time":9.839606343},
-{"learn":[0.5087198045],"iteration":244,"passed_time":9.418265217,"remaining_time":9.802684205},
-{"learn":[0.5084669911],"iteration":245,"passed_time":9.45596502,"remaining_time":9.763476078},
-{"learn":[0.5082518498],"iteration":246,"passed_time":9.494581341,"remaining_time":9.725218945},
-{"learn":[0.5078444417],"iteration":247,"passed_time":9.533286008,"remaining_time":9.687048686},
-{"learn":[0.5074409544],"iteration":248,"passed_time":9.569680902,"remaining_time":9.646545809},
-{"learn":[0.5071251144],"iteration":249,"passed_time":9.606685872,"remaining_time":9.606685872},
-{"learn":[0.5067451669],"iteration":250,"passed_time":9.645321118,"remaining_time":9.56846597},
-{"learn":[0.5061366629],"iteration":251,"passed_time":9.683207982,"remaining_time":9.529506268},
-{"learn":[0.5058645418],"iteration":252,"passed_time":9.721027223,"remaining_time":9.490489028},
-{"learn":[0.5054086685],"iteration":253,"passed_time":9.760248756,"remaining_time":9.452839346},
-{"learn":[0.5048886353],"iteration":254,"passed_time":9.798211924,"remaining_time":9.413968319},
-{"learn":[0.5045785573],"iteration":255,"passed_time":9.836920903,"remaining_time":9.375815236},
-{"learn":[0.5040652562],"iteration":256,"passed_time":9.876010288,"remaining_time":9.33801751},
-{"learn":[0.5036517135],"iteration":257,"passed_time":9.915121143,"remaining_time":9.300229909},
-{"learn":[0.5031239291],"iteration":258,"passed_time":9.954543779,"remaining_time":9.262722204},
-{"learn":[0.5027971955],"iteration":259,"passed_time":9.991621704,"remaining_time":9.223035419},
-{"learn":[0.5022963486],"iteration":260,"passed_time":10.03360686,"remaining_time":9.187862216},
-{"learn":[0.5017921176],"iteration":261,"passed_time":10.0713065,"remaining_time":9.148744075},
-{"learn":[0.5013332178],"iteration":262,"passed_time":10.11092195,"remaining_time":9.111363127},
-{"learn":[0.5009457363],"iteration":263,"passed_time":10.15105638,"remaining_time":9.074429186},
-{"learn":[0.5007438477],"iteration":264,"passed_time":10.18831222,"remaining_time":9.034918384},
-{"learn":[0.5003634359],"iteration":265,"passed_time":10.22616737,"remaining_time":8.995951748},
-{"learn":[0.4997749466],"iteration":266,"passed_time":10.26438807,"remaining_time":8.957312436},
-{"learn":[0.4995323853],"iteration":267,"passed_time":10.30279905,"remaining_time":8.918840966},
-{"learn":[0.4992080672],"iteration":268,"passed_time":10.3411861,"remaining_time":8.880349398},
-{"learn":[0.4988797048],"iteration":269,"passed_time":10.37995545,"remaining_time":8.84218427},
-{"learn":[0.4984015217],"iteration":270,"passed_time":10.4192813,"remaining_time":8.804484938},
-{"learn":[0.4977663234],"iteration":271,"passed_time":10.45836938,"remaining_time":8.766574335},
-{"learn":[0.4973336106],"iteration":272,"passed_time":10.49583501,"remaining_time":8.727306034},
-{"learn":[0.4969323188],"iteration":273,"passed_time":10.53345445,"remaining_time":8.688177757},
-{"learn":[0.4966561773],"iteration":274,"passed_time":10.57140931,"remaining_time":8.649334888},
-{"learn":[0.496340795],"iteration":275,"passed_time":10.60995048,"remaining_time":8.610974299},
-{"learn":[0.4959069085],"iteration":276,"passed_time":10.64741822,"remaining_time":8.571748243},
-{"learn":[0.4954098916],"iteration":277,"passed_time":10.68748112,"remaining_time":8.534607227},
-{"learn":[0.4950271537],"iteration":278,"passed_time":10.72589538,"remaining_time":8.49613935},
-{"learn":[0.4946431444],"iteration":279,"passed_time":10.76519635,"remaining_time":8.458368557},
-{"learn":[0.4944029335],"iteration":280,"passed_time":10.80309636,"remaining_time":8.419495029},
-{"learn":[0.4942005513],"iteration":281,"passed_time":10.84228739,"remaining_time":8.381626426},
-{"learn":[0.4938728298],"iteration":282,"passed_time":10.88115595,"remaining_time":8.343501207},
-{"learn":[0.4935146349],"iteration":283,"passed_time":10.91931018,"remaining_time":8.304827461},
-{"learn":[0.493156001],"iteration":284,"passed_time":10.9571168,"remaining_time":8.26589513},
-{"learn":[0.4928198856],"iteration":285,"passed_time":10.99422939,"remaining_time":8.22645136},
-{"learn":[0.492561635],"iteration":286,"passed_time":11.03361959,"remaining_time":8.188714189},
-{"learn":[0.4922082132],"iteration":287,"passed_time":11.07156647,"remaining_time":8.149903095},
-{"learn":[0.4917932108],"iteration":288,"passed_time":11.10988959,"remaining_time":8.111372674},
-{"learn":[0.4915991369],"iteration":289,"passed_time":11.14760662,"remaining_time":8.072404793},
-{"learn":[0.4912496628],"iteration":290,"passed_time":11.18565405,"remaining_time":8.033682804},
-{"learn":[0.4909530477],"iteration":291,"passed_time":11.22309209,"remaining_time":7.99453135},
-{"learn":[0.4905833112],"iteration":292,"passed_time":11.26210462,"remaining_time":7.956503948},
-{"learn":[0.4902644174],"iteration":293,"passed_time":11.29932215,"remaining_time":7.917212121},
-{"learn":[0.4899068175],"iteration":294,"passed_time":11.33698086,"remaining_time":7.878240939},
-{"learn":[0.4895770055],"iteration":295,"passed_time":11.37485702,"remaining_time":7.839428484},
-{"learn":[0.4892402572],"iteration":296,"passed_time":11.4146948,"remaining_time":7.801963113},
-{"learn":[0.4889109272],"iteration":297,"passed_time":11.45253044,"remaining_time":7.763124661},
-{"learn":[0.4887388656],"iteration":298,"passed_time":11.49124797,"remaining_time":7.724885758},
-{"learn":[0.488525413],"iteration":299,"passed_time":11.53198209,"remaining_time":7.68798806},
-{"learn":[0.4883646535],"iteration":300,"passed_time":11.57026555,"remaining_time":7.649444663},
-{"learn":[0.4880313834],"iteration":301,"passed_time":11.60958483,"remaining_time":7.611582107},
-{"learn":[0.487600658],"iteration":302,"passed_time":11.6461879,"remaining_time":7.571943948},
-{"learn":[0.4872873471],"iteration":303,"passed_time":11.68525181,"remaining_time":7.533912352},
-{"learn":[0.4870871276],"iteration":304,"passed_time":11.7232872,"remaining_time":7.495216405},
-{"learn":[0.4864906459],"iteration":305,"passed_time":11.76182207,"remaining_time":7.456841446},
-{"learn":[0.486156593],"iteration":306,"passed_time":11.80000075,"remaining_time":7.418241512},
-{"learn":[0.4856904754],"iteration":307,"passed_time":11.83869766,"remaining_time":7.379967371},
-{"learn":[0.4852077101],"iteration":308,"passed_time":11.87865589,"remaining_time":7.342470145},
-{"learn":[0.4850039883],"iteration":309,"passed_time":11.91779947,"remaining_time":7.304457741},
-{"learn":[0.4847537811],"iteration":310,"passed_time":11.95624758,"remaining_time":7.26601541},
-{"learn":[0.4845627209],"iteration":311,"passed_time":11.99489635,"remaining_time":7.227693953},
-{"learn":[0.4845330932],"iteration":312,"passed_time":12.0323736,"remaining_time":7.18867049},
-{"learn":[0.4839737251],"iteration":313,"passed_time":12.06976343,"remaining_time":7.149605091},
-{"learn":[0.4835309302],"iteration":314,"passed_time":12.10672469,"remaining_time":7.110298626},
-{"learn":[0.4831540218],"iteration":315,"passed_time":12.14578878,"remaining_time":7.072231439},
-{"learn":[0.4828460852],"iteration":316,"passed_time":12.18338319,"remaining_time":7.033309541},
-{"learn":[0.4825408312],"iteration":317,"passed_time":12.2212377,"remaining_time":6.994544848},
-{"learn":[0.4820667862],"iteration":318,"passed_time":12.26034265,"remaining_time":6.956495359},
-{"learn":[0.4818882741],"iteration":319,"passed_time":12.29999108,"remaining_time":6.918744984},
-{"learn":[0.4814636366],"iteration":320,"passed_time":12.33873549,"remaining_time":6.880478671},
-{"learn":[0.481184684],"iteration":321,"passed_time":12.3753392,"remaining_time":6.841026019},
-{"learn":[0.4807144658],"iteration":322,"passed_time":12.41596062,"remaining_time":6.803792659},
-{"learn":[0.4804663756],"iteration":323,"passed_time":12.45410421,"remaining_time":6.765192412},
-{"learn":[0.4800794796],"iteration":324,"passed_time":12.4920808,"remaining_time":6.726505046},
-{"learn":[0.4798931718],"iteration":325,"passed_time":12.52893201,"remaining_time":6.687221382},
-{"learn":[0.4795692068],"iteration":326,"passed_time":12.56699338,"remaining_time":6.64859283},
-{"learn":[0.4791146234],"iteration":327,"passed_time":12.60744708,"remaining_time":6.611222247},
-{"learn":[0.4786749759],"iteration":328,"passed_time":12.64659707,"remaining_time":6.573155314},
-{"learn":[0.4784382168],"iteration":329,"passed_time":12.6844167,"remaining_time":6.534396483},
-{"learn":[0.4781887154],"iteration":330,"passed_time":12.72197219,"remaining_time":6.495508458},
-{"learn":[0.4780572928],"iteration":331,"passed_time":12.75899566,"remaining_time":6.456359248},
-{"learn":[0.4777415159],"iteration":332,"passed_time":12.79842041,"remaining_time":6.418427051},
-{"learn":[0.4774675269],"iteration":333,"passed_time":12.83639264,"remaining_time":6.379764008},
-{"learn":[0.4773121415],"iteration":334,"passed_time":12.87306359,"remaining_time":6.340464156},
-{"learn":[0.4769842799],"iteration":335,"passed_time":12.9120571,"remaining_time":6.302313586},
-{"learn":[0.4766920756],"iteration":336,"passed_time":12.95060821,"remaining_time":6.26394403},
-{"learn":[0.4764646891],"iteration":337,"passed_time":12.990201,"remaining_time":6.226072669},
-{"learn":[0.4760961631],"iteration":338,"passed_time":13.02966206,"remaining_time":6.188128588},
-{"learn":[0.4758210384],"iteration":339,"passed_time":13.06621011,"remaining_time":6.148804758},
-{"learn":[0.4754372506],"iteration":340,"passed_time":13.10432419,"remaining_time":6.110227407},
-{"learn":[0.4752409719],"iteration":341,"passed_time":13.14456266,"remaining_time":6.07263421},
-{"learn":[0.4750971587],"iteration":342,"passed_time":13.1818467,"remaining_time":6.033673271},
-{"learn":[0.4747375371],"iteration":343,"passed_time":13.22199566,"remaining_time":5.996021288},
-{"learn":[0.4745017945],"iteration":344,"passed_time":13.25926914,"remaining_time":5.957062948},
-{"learn":[0.4742736682],"iteration":345,"passed_time":13.29789632,"remaining_time":5.918716859},
-{"learn":[0.4739464879],"iteration":346,"passed_time":13.33632407,"remaining_time":5.880281217},
-{"learn":[0.4736754421],"iteration":347,"passed_time":13.37338327,"remaining_time":5.841247863},
-{"learn":[0.4734438174],"iteration":348,"passed_time":13.41429574,"remaining_time":5.803893},
-{"learn":[0.4732039158],"iteration":349,"passed_time":13.45216902,"remaining_time":5.765215296},
-{"learn":[0.4726483693],"iteration":350,"passed_time":13.49010811,"remaining_time":5.72657011},
-{"learn":[0.4723924666],"iteration":351,"passed_time":13.52795099,"remaining_time":5.687888486},
-{"learn":[0.472146367],"iteration":352,"passed_time":13.56481878,"remaining_time":5.648805554},
-{"learn":[0.4719615914],"iteration":353,"passed_time":13.60277992,"remaining_time":5.61018607},
-{"learn":[0.4718524672],"iteration":354,"passed_time":13.63985701,"remaining_time":5.5712092},
-{"learn":[0.4715708345],"iteration":355,"passed_time":13.67786976,"remaining_time":5.532621477},
-{"learn":[0.4712705574],"iteration":356,"passed_time":13.71664512,"remaining_time":5.494342442},
-{"learn":[0.4710028815],"iteration":357,"passed_time":13.7541049,"remaining_time":5.455538817},
-{"learn":[0.4707747208],"iteration":358,"passed_time":13.79141896,"remaining_time":5.41668544},
-{"learn":[0.470403996],"iteration":359,"passed_time":13.83055887,"remaining_time":5.37855067},
-{"learn":[0.4702629921],"iteration":360,"passed_time":13.86718889,"remaining_time":5.339443921},
-{"learn":[0.4699002207],"iteration":361,"passed_time":13.90529544,"remaining_time":5.30091373},
-{"learn":[0.4697222917],"iteration":362,"passed_time":13.9440997,"remaining_time":5.262649197},
-{"learn":[0.4692968498],"iteration":363,"passed_time":13.98363789,"remaining_time":5.224655915},
-{"learn":[0.4691166011],"iteration":364,"passed_time":14.02160627,"remaining_time":5.186073553},
-{"learn":[0.4688993415],"iteration":365,"passed_time":14.05857796,"remaining_time":5.147129637},
-{"learn":[0.4685782266],"iteration":366,"passed_time":14.09676224,"remaining_time":5.108635906},
-{"learn":[0.468261587],"iteration":367,"passed_time":14.13519477,"remaining_time":5.070232905},
-{"learn":[0.4680112053],"iteration":368,"passed_time":14.17223873,"remaining_time":5.031336784},
-{"learn":[0.4677463278],"iteration":369,"passed_time":14.21127893,"remaining_time":4.993152057},
-{"learn":[0.4673304386],"iteration":370,"passed_time":14.2508349,"remaining_time":4.955142056},
-{"learn":[0.4671175323],"iteration":371,"passed_time":14.28911375,"remaining_time":4.916684302},
-{"learn":[0.4668852125],"iteration":372,"passed_time":14.32902023,"remaining_time":4.878781688},
-{"learn":[0.4666989726],"iteration":373,"passed_time":14.36617411,"remaining_time":4.83994101},
-{"learn":[0.4663318389],"iteration":374,"passed_time":14.40471568,"remaining_time":4.801571892},
-{"learn":[0.4660052012],"iteration":375,"passed_time":14.44215681,"remaining_time":4.762838947},
-{"learn":[0.4656110221],"iteration":376,"passed_time":14.48014563,"remaining_time":4.724291546},
-{"learn":[0.4653915722],"iteration":377,"passed_time":14.52031956,"remaining_time":4.686452344},
-{"learn":[0.4651173142],"iteration":378,"passed_time":14.5563113,"remaining_time":4.647265612},
-{"learn":[0.4649721562],"iteration":379,"passed_time":14.59463567,"remaining_time":4.608832317},
-{"learn":[0.4647070469],"iteration":380,"passed_time":14.63424974,"remaining_time":4.570802413},
-{"learn":[0.4646002414],"iteration":381,"passed_time":14.67127577,"remaining_time":4.531964767},
-{"learn":[0.4643426168],"iteration":382,"passed_time":14.71374043,"remaining_time":4.494797989},
-{"learn":[0.4639029335],"iteration":383,"passed_time":14.75065633,"remaining_time":4.455927434},
-{"learn":[0.4635395322],"iteration":384,"passed_time":14.78804059,"remaining_time":4.417206931},
-{"learn":[0.4632807954],"iteration":385,"passed_time":14.82670475,"remaining_time":4.37887135},
-{"learn":[0.463002022],"iteration":386,"passed_time":14.86279143,"remaining_time":4.339781477},
-{"learn":[0.462628978],"iteration":387,"passed_time":14.90235451,"remaining_time":4.30171058},
-{"learn":[0.4620670856],"iteration":388,"passed_time":14.94397336,"remaining_time":4.264218619},
-{"learn":[0.4619002464],"iteration":389,"passed_time":14.98321463,"remaining_time":4.226034894},
-{"learn":[0.4613682405],"iteration":390,"passed_time":15.02659339,"remaining_time":4.188999179},
-{"learn":[0.4610835171],"iteration":391,"passed_time":15.06541142,"remaining_time":4.150674574},
-{"learn":[0.4606795599],"iteration":392,"passed_time":15.10357806,"remaining_time":4.112170107},
-{"learn":[0.4604993491],"iteration":393,"passed_time":15.14102223,"remaining_time":4.073472987},
-{"learn":[0.4602709248],"iteration":394,"passed_time":15.17814243,"remaining_time":4.034696088},
-{"learn":[0.460042721],"iteration":395,"passed_time":15.21569673,"remaining_time":3.996041565},
-{"learn":[0.4595936079],"iteration":396,"passed_time":15.25497673,"remaining_time":3.957840311},
-{"learn":[0.4592798285],"iteration":397,"passed_time":15.2935978,"remaining_time":3.919464764},
-{"learn":[0.458882499],"iteration":398,"passed_time":15.33392742,"remaining_time":3.881520475},
-{"learn":[0.4586580693],"iteration":399,"passed_time":15.37004742,"remaining_time":3.842511854},
-{"learn":[0.4581193803],"iteration":400,"passed_time":15.41014299,"remaining_time":3.804499141},
-{"learn":[0.4578728731],"iteration":401,"passed_time":15.44757332,"remaining_time":3.765826333},
-{"learn":[0.4576978232],"iteration":402,"passed_time":15.48491194,"remaining_time":3.727137613},
-{"learn":[0.457407343],"iteration":403,"passed_time":15.52334432,"remaining_time":3.688715482},
-{"learn":[0.4569395552],"iteration":404,"passed_time":15.56126312,"remaining_time":3.650172831},
-{"learn":[0.4563988559],"iteration":405,"passed_time":15.60082865,"remaining_time":3.612014516},
-{"learn":[0.4561760676],"iteration":406,"passed_time":15.63944003,"remaining_time":3.57363126},
-{"learn":[0.4558851807],"iteration":407,"passed_time":15.67590542,"remaining_time":3.534762988},
-{"learn":[0.4556379291],"iteration":408,"passed_time":15.71340746,"remaining_time":3.496137113},
-{"learn":[0.4553495443],"iteration":409,"passed_time":15.75215348,"remaining_time":3.457789789},
-{"learn":[0.4548500646],"iteration":410,"passed_time":15.79089495,"remaining_time":3.419439538},
-{"learn":[0.4547152754],"iteration":411,"passed_time":15.82687167,"remaining_time":3.380496861},
-{"learn":[0.454511923],"iteration":412,"passed_time":15.86509663,"remaining_time":3.342042147},
-{"learn":[0.4543781494],"iteration":413,"passed_time":15.90231846,"remaining_time":3.303380163},
-{"learn":[0.4542514987],"iteration":414,"passed_time":15.94238887,"remaining_time":3.265308563},
-{"learn":[0.4541160366],"iteration":415,"passed_time":15.98011173,"remaining_time":3.226753331},
-{"learn":[0.4539945058],"iteration":416,"passed_time":16.01805965,"remaining_time":3.188246886},
-{"learn":[0.4538843921],"iteration":417,"passed_time":16.05564968,"remaining_time":3.149672905},
-{"learn":[0.4537582704],"iteration":418,"passed_time":16.09360906,"remaining_time":3.111175021},
-{"learn":[0.4535486479],"iteration":419,"passed_time":16.134037,"remaining_time":3.073149904},
-{"learn":[0.453304122],"iteration":420,"passed_time":16.17143875,"remaining_time":3.034545514},
-{"learn":[0.4530209968],"iteration":421,"passed_time":16.21181979,"remaining_time":2.996497496},
-{"learn":[0.4529209824],"iteration":422,"passed_time":16.2485269,"remaining_time":2.957769672},
-{"learn":[0.4527216047],"iteration":423,"passed_time":16.28762247,"remaining_time":2.919479499},
-{"learn":[0.4523943482],"iteration":424,"passed_time":16.32667394,"remaining_time":2.881177755},
-{"learn":[0.452075209],"iteration":425,"passed_time":16.36541313,"remaining_time":2.842818244},
-{"learn":[0.4519256599],"iteration":426,"passed_time":16.40397536,"remaining_time":2.8044267},
-{"learn":[0.4517125164],"iteration":427,"passed_time":16.44304856,"remaining_time":2.766120318},
-{"learn":[0.4514656506],"iteration":428,"passed_time":16.48628672,"remaining_time":2.728499667},
-{"learn":[0.4511500975],"iteration":429,"passed_time":16.53140396,"remaining_time":2.691158783},
-{"learn":[0.4509773881],"iteration":430,"passed_time":16.57383981,"remaining_time":2.653352545},
-{"learn":[0.4507410304],"iteration":431,"passed_time":16.61756023,"remaining_time":2.615727073},
-{"learn":[0.4504520508],"iteration":432,"passed_time":16.66230157,"remaining_time":2.578231421},
-{"learn":[0.4502464623],"iteration":433,"passed_time":16.70317662,"remaining_time":2.540114417},
-{"learn":[0.4499163444],"iteration":434,"passed_time":16.74950391,"remaining_time":2.502799434},
-{"learn":[0.4495444335],"iteration":435,"passed_time":16.79807863,"remaining_time":2.46577301},
-{"learn":[0.4492822328],"iteration":436,"passed_time":16.84623355,"remaining_time":2.428633212},
-{"learn":[0.4490709995],"iteration":437,"passed_time":16.88556258,"remaining_time":2.39019379},
-{"learn":[0.4488742685],"iteration":438,"passed_time":16.92548394,"remaining_time":2.351832621},
-{"learn":[0.4488202023],"iteration":439,"passed_time":16.96400911,"remaining_time":2.313273969},
-{"learn":[0.4487087322],"iteration":440,"passed_time":17.01200894,"remaining_time":2.275983056},
-{"learn":[0.448424538],"iteration":441,"passed_time":17.05116688,"remaining_time":2.237483437},
-{"learn":[0.4481589914],"iteration":442,"passed_time":17.09036424,"remaining_time":2.198985918},
-{"learn":[0.4479120738],"iteration":443,"passed_time":17.12830625,"remaining_time":2.160326915},
-{"learn":[0.4478134597],"iteration":444,"passed_time":17.16549185,"remaining_time":2.121577644},
-{"learn":[0.4475742582],"iteration":445,"passed_time":17.20333132,"remaining_time":2.082914554},
-{"learn":[0.4472707527],"iteration":446,"passed_time":17.24082523,"remaining_time":2.044214177},
-{"learn":[0.4470743065],"iteration":447,"passed_time":17.27792118,"remaining_time":2.005472994},
-{"learn":[0.4467388758],"iteration":448,"passed_time":17.31607921,"remaining_time":1.966859776},
-{"learn":[0.4465189068],"iteration":449,"passed_time":17.35387227,"remaining_time":1.928208031},
-{"learn":[0.4462596129],"iteration":450,"passed_time":17.39134315,"remaining_time":1.889525088},
-{"learn":[0.4460411878],"iteration":451,"passed_time":17.42966633,"remaining_time":1.850938017},
-{"learn":[0.4459891391],"iteration":452,"passed_time":17.46697181,"remaining_time":1.812246523},
-{"learn":[0.4457786012],"iteration":453,"passed_time":17.50597339,"remaining_time":1.773732987},
-{"learn":[0.4456593077],"iteration":454,"passed_time":17.54432472,"remaining_time":1.735152995},
-{"learn":[0.4454243826],"iteration":455,"passed_time":17.58259955,"remaining_time":1.696566623},
-{"learn":[0.445302389],"iteration":456,"passed_time":17.61884307,"remaining_time":1.657790485},
-{"learn":[0.4451221153],"iteration":457,"passed_time":17.65748377,"remaining_time":1.619245237},
-{"learn":[0.4447792989],"iteration":458,"passed_time":17.69626321,"remaining_time":1.580711964},
-{"learn":[0.4447178247],"iteration":459,"passed_time":17.7330692,"remaining_time":1.542006017},
-{"learn":[0.444417161],"iteration":460,"passed_time":17.77215078,"remaining_time":1.503500825},
-{"learn":[0.4441799003],"iteration":461,"passed_time":17.8109063,"remaining_time":1.464966319},
-{"learn":[0.4437149583],"iteration":462,"passed_time":17.84932638,"remaining_time":1.426404052},
-{"learn":[0.4434962894],"iteration":463,"passed_time":17.88752391,"remaining_time":1.387825131},
-{"learn":[0.4433650256],"iteration":464,"passed_time":17.92560861,"remaining_time":1.349239357},
-{"learn":[0.4431129225],"iteration":465,"passed_time":17.96436755,"remaining_time":1.310704928},
-{"learn":[0.4430382611],"iteration":466,"passed_time":18.002004,"remaining_time":1.272090218},
-{"learn":[0.4427619817],"iteration":467,"passed_time":18.03861625,"remaining_time":1.233409658},
-{"learn":[0.4426803452],"iteration":468,"passed_time":18.07531374,"remaining_time":1.194743552},
-{"learn":[0.4423056868],"iteration":469,"passed_time":18.11367926,"remaining_time":1.156192293},
-{"learn":[0.4420721901],"iteration":470,"passed_time":18.15038173,"remaining_time":1.117539427},
-{"learn":[0.4417882376],"iteration":471,"passed_time":18.18803549,"remaining_time":1.078951258},
-{"learn":[0.4416311924],"iteration":472,"passed_time":18.22652394,"remaining_time":1.040414686},
-{"learn":[0.4413658561],"iteration":473,"passed_time":18.26783818,"remaining_time":1.002033318},
-{"learn":[0.4411799644],"iteration":474,"passed_time":18.31076382,"remaining_time":0.9637244116},
-{"learn":[0.4408430967],"iteration":475,"passed_time":18.35179758,"remaining_time":0.9253007182},
-{"learn":[0.4406326451],"iteration":476,"passed_time":18.3891758,"remaining_time":0.8866898183},
-{"learn":[0.4404920207],"iteration":477,"passed_time":18.42826989,"remaining_time":0.8481630495},
-{"learn":[0.4402559337],"iteration":478,"passed_time":18.46571189,"remaining_time":0.8095614818},
-{"learn":[0.4401127627],"iteration":479,"passed_time":18.50427938,"remaining_time":0.7710116408},
-{"learn":[0.4398614027],"iteration":480,"passed_time":18.54154164,"remaining_time":0.7324101689},
-{"learn":[0.4396763562],"iteration":481,"passed_time":18.58053674,"remaining_time":0.6938789655},
-{"learn":[0.4392673677],"iteration":482,"passed_time":18.620477,"remaining_time":0.6553791076},
-{"learn":[0.4388877804],"iteration":483,"passed_time":18.65971607,"remaining_time":0.616850118},
-{"learn":[0.438673879],"iteration":484,"passed_time":18.70041562,"remaining_time":0.5783633697},
-{"learn":[0.4383851937],"iteration":485,"passed_time":18.73878951,"remaining_time":0.539800521},
-{"learn":[0.4381989892],"iteration":486,"passed_time":18.77776306,"remaining_time":0.5012544555},
-{"learn":[0.4379762119],"iteration":487,"passed_time":18.81592874,"remaining_time":0.4626867724},
-{"learn":[0.437736575],"iteration":488,"passed_time":18.85684101,"remaining_time":0.4241825175},
-{"learn":[0.4372321637],"iteration":489,"passed_time":18.90304912,"remaining_time":0.3857765126},
-{"learn":[0.4369799398],"iteration":490,"passed_time":18.94758441,"remaining_time":0.3473080645},
-{"learn":[0.4365836973],"iteration":491,"passed_time":18.99265294,"remaining_time":0.308823625},
-{"learn":[0.4363145001],"iteration":492,"passed_time":19.03420231,"remaining_time":0.2702625075},
-{"learn":[0.4360217335],"iteration":493,"passed_time":19.07654972,"remaining_time":0.2316989845},
-{"learn":[0.4358808553],"iteration":494,"passed_time":19.12212822,"remaining_time":0.1931528103},
-{"learn":[0.4357672486],"iteration":495,"passed_time":19.162853,"remaining_time":0.1545391371},
-{"learn":[0.4354725607],"iteration":496,"passed_time":19.20445719,"remaining_time":0.1159222768},
-{"learn":[0.4353642518],"iteration":497,"passed_time":19.24609978,"remaining_time":0.07729357339},
-{"learn":[0.4350391341],"iteration":498,"passed_time":19.28873455,"remaining_time":0.03865477865},
-{"learn":[0.4347724016],"iteration":499,"passed_time":19.3325768,"remaining_time":0}
+{"learn":[1.068942757],"iteration":0,"passed_time":0.08279778233,"remaining_time":41.31609338},
+{"learn":[1.042941062],"iteration":1,"passed_time":0.1131763962,"remaining_time":28.18092265},
+{"learn":[1.017818973],"iteration":2,"passed_time":0.1433944312,"remaining_time":23.75567744},
+{"learn":[0.9954188838],"iteration":3,"passed_time":0.1745056364,"remaining_time":21.63869891},
+{"learn":[0.9735267885],"iteration":4,"passed_time":0.2063176909,"remaining_time":20.4254514},
+{"learn":[0.9537807289],"iteration":5,"passed_time":0.2381420839,"remaining_time":19.60703158},
+{"learn":[0.9354755366],"iteration":6,"passed_time":0.2719769961,"remaining_time":19.1549513},
+{"learn":[0.9187623571],"iteration":7,"passed_time":0.3050119893,"remaining_time":18.75823734},
+{"learn":[0.9023127175],"iteration":8,"passed_time":0.3392256541,"remaining_time":18.50664402},
+{"learn":[0.8868191679],"iteration":9,"passed_time":0.3733350948,"remaining_time":18.29341965},
+{"learn":[0.8730040311],"iteration":10,"passed_time":0.4084155332,"remaining_time":18.15592688},
+{"learn":[0.860627421],"iteration":11,"passed_time":0.4437198893,"remaining_time":18.04460883},
+{"learn":[0.8491307201],"iteration":12,"passed_time":0.4794973628,"remaining_time":17.9627089},
+{"learn":[0.8373120751],"iteration":13,"passed_time":0.5162526517,"remaining_time":17.92134205},
+{"learn":[0.825819299],"iteration":14,"passed_time":0.5516826554,"remaining_time":17.83773919},
+{"learn":[0.815364983],"iteration":15,"passed_time":0.5905684809,"remaining_time":17.86469655},
+{"learn":[0.8058413028],"iteration":16,"passed_time":0.6280071167,"remaining_time":17.84279043},
+{"learn":[0.7959650463],"iteration":17,"passed_time":0.6693215331,"remaining_time":17.92294328},
+{"learn":[0.7868600676],"iteration":18,"passed_time":0.7108126411,"remaining_time":17.99478318},
+{"learn":[0.7784883379],"iteration":19,"passed_time":0.7489838509,"remaining_time":17.97561242},
+{"learn":[0.7701307612],"iteration":20,"passed_time":0.7896698585,"remaining_time":18.01199344},
+{"learn":[0.7622975086],"iteration":21,"passed_time":0.8307501715,"remaining_time":18.04993554},
+{"learn":[0.7563031265],"iteration":22,"passed_time":0.8738670184,"remaining_time":18.12324208},
+{"learn":[0.7495887487],"iteration":23,"passed_time":0.9167154946,"remaining_time":18.18152398},
+{"learn":[0.7429723054],"iteration":24,"passed_time":0.9550352849,"remaining_time":18.14567041},
+{"learn":[0.7372392086],"iteration":25,"passed_time":0.9935027496,"remaining_time":18.11231936},
+{"learn":[0.7307956392],"iteration":26,"passed_time":1.032044367,"remaining_time":18.07988836},
+{"learn":[0.72524455],"iteration":27,"passed_time":1.071621181,"remaining_time":18.06447134},
+{"learn":[0.7192748437],"iteration":28,"passed_time":1.109982304,"remaining_time":18.02764362},
+{"learn":[0.7139143089],"iteration":29,"passed_time":1.147641418,"remaining_time":17.97971555},
+{"learn":[0.7087284348],"iteration":30,"passed_time":1.187638094,"remaining_time":17.96781504},
+{"learn":[0.7037127192],"iteration":31,"passed_time":1.229517366,"remaining_time":17.98169148},
+{"learn":[0.6988890737],"iteration":32,"passed_time":1.26810278,"remaining_time":17.9455757},
+{"learn":[0.6949220701],"iteration":33,"passed_time":1.308109207,"remaining_time":17.9287909},
+{"learn":[0.6905132299],"iteration":34,"passed_time":1.348107083,"remaining_time":17.91056553},
+{"learn":[0.6865428783],"iteration":35,"passed_time":1.387068428,"remaining_time":17.87777085},
+{"learn":[0.6825442625],"iteration":36,"passed_time":1.427890468,"remaining_time":17.86792666},
+{"learn":[0.678493613],"iteration":37,"passed_time":1.465300607,"remaining_time":17.81497054},
+{"learn":[0.6753394188],"iteration":38,"passed_time":1.504279007,"remaining_time":17.78134929},
+{"learn":[0.6720286547],"iteration":39,"passed_time":1.544592377,"remaining_time":17.76281233},
+{"learn":[0.6689512826],"iteration":40,"passed_time":1.583848876,"remaining_time":17.73138132},
+{"learn":[0.6659716333],"iteration":41,"passed_time":1.6233685,"remaining_time":17.70244698},
+{"learn":[0.6630764989],"iteration":42,"passed_time":1.663433486,"remaining_time":17.67881635},
+{"learn":[0.659979405],"iteration":43,"passed_time":1.700967777,"remaining_time":17.62821151},
+{"learn":[0.6571206305],"iteration":44,"passed_time":1.740605562,"remaining_time":17.59945624},
+{"learn":[0.6543860008],"iteration":45,"passed_time":1.778030683,"remaining_time":17.54838979},
+{"learn":[0.6516459562],"iteration":46,"passed_time":1.817541859,"remaining_time":17.51800983},
+{"learn":[0.6488159878],"iteration":47,"passed_time":1.85713373,"remaining_time":17.4880093},
+{"learn":[0.6465175253],"iteration":48,"passed_time":1.894888841,"remaining_time":17.44071158},
+{"learn":[0.6441297453],"iteration":49,"passed_time":1.936883193,"remaining_time":17.43194873},
+{"learn":[0.6416986449],"iteration":50,"passed_time":1.977790613,"remaining_time":17.41231343},
+{"learn":[0.6391230496],"iteration":51,"passed_time":2.020247316,"remaining_time":17.40520765},
+{"learn":[0.6366229183],"iteration":52,"passed_time":2.062408844,"remaining_time":17.39427837},
+{"learn":[0.6342664881],"iteration":53,"passed_time":2.104676555,"remaining_time":17.38306932},
+{"learn":[0.6323079472],"iteration":54,"passed_time":2.144704236,"remaining_time":17.352607},
+{"learn":[0.6305081074],"iteration":55,"passed_time":2.18672938,"remaining_time":17.33764009},
+{"learn":[0.62840159],"iteration":56,"passed_time":2.228523581,"remaining_time":17.31992888},
+{"learn":[0.6266940801],"iteration":57,"passed_time":2.267042881,"remaining_time":17.27643023},
+{"learn":[0.6248571874],"iteration":58,"passed_time":2.310638359,"remaining_time":17.27104265},
+{"learn":[0.6235244597],"iteration":59,"passed_time":2.354659629,"remaining_time":17.26750395},
+{"learn":[0.6219566147],"iteration":60,"passed_time":2.39551113,"remaining_time":17.239826},
+{"learn":[0.6204337402],"iteration":61,"passed_time":2.43655573,"remaining_time":17.21308726},
+{"learn":[0.6187130806],"iteration":62,"passed_time":2.475939703,"remaining_time":17.1743754},
+{"learn":[0.6170088968],"iteration":63,"passed_time":2.516637458,"remaining_time":17.14459268},
+{"learn":[0.615589758],"iteration":64,"passed_time":2.558334714,"remaining_time":17.12116308},
+{"learn":[0.6140922722],"iteration":65,"passed_time":2.599156649,"remaining_time":17.09142403},
+{"learn":[0.6129240173],"iteration":66,"passed_time":2.640820808,"remaining_time":17.06679716},
+{"learn":[0.6116130344],"iteration":67,"passed_time":2.681463865,"remaining_time":17.0351822},
+{"learn":[0.6104895206],"iteration":68,"passed_time":2.722620766,"remaining_time":17.00651522},
+{"learn":[0.6094134377],"iteration":69,"passed_time":2.766468263,"remaining_time":16.99401933},
+{"learn":[0.6080808677],"iteration":70,"passed_time":2.814897406,"remaining_time":17.00832376},
+{"learn":[0.6069087998],"iteration":71,"passed_time":2.861253252,"remaining_time":17.008561},
+{"learn":[0.6059376837],"iteration":72,"passed_time":2.904192323,"remaining_time":16.98753592},
+{"learn":[0.6046673652],"iteration":73,"passed_time":2.94932623,"remaining_time":16.9785537},
+{"learn":[0.6034098585],"iteration":74,"passed_time":2.99517606,"remaining_time":16.97266434},
+{"learn":[0.6023554521],"iteration":75,"passed_time":3.042245603,"remaining_time":16.9725281},
+{"learn":[0.6010549909],"iteration":76,"passed_time":3.08501937,"remaining_time":16.94757394},
+{"learn":[0.5999185725],"iteration":77,"passed_time":3.127500852,"remaining_time":16.92058153},
+{"learn":[0.5986988724],"iteration":78,"passed_time":3.167769793,"remaining_time":16.88140611},
+{"learn":[0.5976798218],"iteration":79,"passed_time":3.21258038,"remaining_time":16.866047},
+{"learn":[0.5967699009],"iteration":80,"passed_time":3.255356129,"remaining_time":16.83943479},
+{"learn":[0.5956767709],"iteration":81,"passed_time":3.295736532,"remaining_time":16.80021793},
+{"learn":[0.5945202655],"iteration":82,"passed_time":3.337328334,"remaining_time":16.76705922},
+{"learn":[0.5933468823],"iteration":83,"passed_time":3.379281941,"remaining_time":16.73549152},
+{"learn":[0.5923352617],"iteration":84,"passed_time":3.420565194,"remaining_time":16.70040654},
+{"learn":[0.5912329495],"iteration":85,"passed_time":3.46249877,"remaining_time":16.66830803},
+{"learn":[0.5906017033],"iteration":86,"passed_time":3.50344872,"remaining_time":16.63131404},
+{"learn":[0.5897119078],"iteration":87,"passed_time":3.545628746,"remaining_time":16.59998913},
+{"learn":[0.5889154258],"iteration":88,"passed_time":3.587549428,"remaining_time":16.56722264},
+{"learn":[0.5880849341],"iteration":89,"passed_time":3.62972093,"remaining_time":16.53539535},
+{"learn":[0.5871459274],"iteration":90,"passed_time":3.670646959,"remaining_time":16.49774292},
+{"learn":[0.5862610925],"iteration":91,"passed_time":3.71661749,"remaining_time":16.48239061},
+{"learn":[0.5854995429],"iteration":92,"passed_time":3.760147836,"remaining_time":16.45570074},
+{"learn":[0.5848049236],"iteration":93,"passed_time":3.800126576,"remaining_time":16.41331266},
+{"learn":[0.5842107749],"iteration":94,"passed_time":3.839823218,"remaining_time":16.36977266},
+{"learn":[0.583505085],"iteration":95,"passed_time":3.882318155,"remaining_time":16.3380889},
+{"learn":[0.5826180548],"iteration":96,"passed_time":3.924410172,"remaining_time":16.30450824},
+{"learn":[0.5818922325],"iteration":97,"passed_time":3.967770237,"remaining_time":16.27595546},
+{"learn":[0.5810033948],"iteration":98,"passed_time":4.013411652,"remaining_time":16.25634417},
+{"learn":[0.5805327081],"iteration":99,"passed_time":4.057082955,"remaining_time":16.22833182},
+{"learn":[0.5796118053],"iteration":100,"passed_time":4.098425454,"remaining_time":16.19080947},
+{"learn":[0.5787480749],"iteration":101,"passed_time":4.140077857,"remaining_time":16.15442144},
+{"learn":[0.5778552983],"iteration":102,"passed_time":4.181521281,"remaining_time":16.11712571},
+{"learn":[0.5771592524],"iteration":103,"passed_time":4.223501095,"remaining_time":16.08179263},
+{"learn":[0.5763560519],"iteration":104,"passed_time":4.26650729,"remaining_time":16.05019409},
+{"learn":[0.5755217776],"iteration":105,"passed_time":4.307396256,"remaining_time":16.01051061},
+{"learn":[0.574693159],"iteration":106,"passed_time":4.350898893,"remaining_time":15.98040435},
+{"learn":[0.5740325984],"iteration":107,"passed_time":4.389639107,"remaining_time":15.93276417},
+{"learn":[0.5733332696],"iteration":108,"passed_time":4.431671073,"remaining_time":15.89709532},
+{"learn":[0.5727932796],"iteration":109,"passed_time":4.471728988,"remaining_time":15.85431187},
+{"learn":[0.5721855741],"iteration":110,"passed_time":4.511135787,"remaining_time":15.80929569},
+{"learn":[0.5715813746],"iteration":111,"passed_time":4.551570859,"remaining_time":15.76794191},
+{"learn":[0.5707142227],"iteration":112,"passed_time":4.588736821,"remaining_time":15.7154084},
+{"learn":[0.5696757236],"iteration":113,"passed_time":4.628725501,"remaining_time":15.67270213},
+{"learn":[0.5691022951],"iteration":114,"passed_time":4.668145188,"remaining_time":15.62813824},
+{"learn":[0.5683067827],"iteration":115,"passed_time":4.706505122,"remaining_time":15.58015489},
+{"learn":[0.5675870432],"iteration":116,"passed_time":4.747501126,"remaining_time":15.54096522},
+{"learn":[0.5669024897],"iteration":117,"passed_time":4.786918318,"remaining_time":15.49663388},
+{"learn":[0.566486124],"iteration":118,"passed_time":4.825858459,"remaining_time":15.45085776},
+{"learn":[0.5658835071],"iteration":119,"passed_time":4.865544991,"remaining_time":15.40755914},
+{"learn":[0.5654351685],"iteration":120,"passed_time":4.904542439,"remaining_time":15.36216185},
+{"learn":[0.5649374002],"iteration":121,"passed_time":4.946800596,"remaining_time":15.32697234},
+{"learn":[0.5641840869],"iteration":122,"passed_time":4.988194545,"remaining_time":15.28901905},
+{"learn":[0.5636579743],"iteration":123,"passed_time":5.027717708,"remaining_time":15.24533757},
+{"learn":[0.5630732508],"iteration":124,"passed_time":5.06932412,"remaining_time":15.20797236},
+{"learn":[0.5626017666],"iteration":125,"passed_time":5.109755876,"remaining_time":15.16705316},
+{"learn":[0.5618269378],"iteration":126,"passed_time":5.150898207,"remaining_time":15.12822859},
+{"learn":[0.5612340032],"iteration":127,"passed_time":5.191128484,"remaining_time":15.08671716},
+{"learn":[0.5607264834],"iteration":128,"passed_time":5.229975383,"remaining_time":15.04124703},
+{"learn":[0.5602062292],"iteration":129,"passed_time":5.268836105,"remaining_time":14.99591815},
+{"learn":[0.5595741473],"iteration":130,"passed_time":5.307431574,"remaining_time":14.94994084},
+{"learn":[0.5594744956],"iteration":131,"passed_time":5.312126862,"remaining_time":14.8095658},
+{"learn":[0.5588128533],"iteration":132,"passed_time":5.352789802,"remaining_time":14.77048013},
+{"learn":[0.5583655792],"iteration":133,"passed_time":5.391555484,"remaining_time":14.72618886},
+{"learn":[0.5576841767],"iteration":134,"passed_time":5.432514326,"remaining_time":14.6879091},
+{"learn":[0.5570366592],"iteration":135,"passed_time":5.47382595,"remaining_time":14.65053416},
+{"learn":[0.5563692413],"iteration":136,"passed_time":5.51413485,"remaining_time":14.61044489},
+{"learn":[0.5558434603],"iteration":137,"passed_time":5.555750094,"remaining_time":14.57377923},
+{"learn":[0.5553565262],"iteration":138,"passed_time":5.596132446,"remaining_time":14.53384038},
+{"learn":[0.5549028143],"iteration":139,"passed_time":5.63594307,"remaining_time":14.49242504},
+{"learn":[0.5542420393],"iteration":140,"passed_time":5.677571504,"remaining_time":14.45566078},
+{"learn":[0.5537416549],"iteration":141,"passed_time":5.71600981,"remaining_time":14.4107853},
+{"learn":[0.5533645111],"iteration":142,"passed_time":5.757627458,"remaining_time":14.37393708},
+{"learn":[0.5525502359],"iteration":143,"passed_time":5.79708511,"remaining_time":14.33168263},
+{"learn":[0.551964004],"iteration":144,"passed_time":5.835042561,"remaining_time":14.28579386},
+{"learn":[0.5513711205],"iteration":145,"passed_time":5.876038701,"remaining_time":14.24738151},
+{"learn":[0.550670328],"iteration":146,"passed_time":5.916333051,"remaining_time":14.20724876},
+{"learn":[0.5500708666],"iteration":147,"passed_time":5.956233206,"remaining_time":14.16617627},
+{"learn":[0.5495710178],"iteration":148,"passed_time":5.995745902,"remaining_time":14.12420679},
+{"learn":[0.5491348099],"iteration":149,"passed_time":6.034787224,"remaining_time":14.08117019},
+{"learn":[0.5487764949],"iteration":150,"passed_time":6.075642632,"remaining_time":14.04237933},
+{"learn":[0.5482966498],"iteration":151,"passed_time":6.116062105,"remaining_time":14.00256324},
+{"learn":[0.5478112718],"iteration":152,"passed_time":6.15655175,"remaining_time":13.96289841},
+{"learn":[0.5471484806],"iteration":153,"passed_time":6.198392095,"remaining_time":13.92625756},
+{"learn":[0.5463325223],"iteration":154,"passed_time":6.238379952,"remaining_time":13.88542634},
+{"learn":[0.5459529093],"iteration":155,"passed_time":6.281674816,"remaining_time":13.85189831},
+{"learn":[0.5455842734],"iteration":156,"passed_time":6.322056059,"remaining_time":13.81188043},
+{"learn":[0.5452050181],"iteration":157,"passed_time":6.363638073,"remaining_time":13.77445709},
+{"learn":[0.5449451093],"iteration":158,"passed_time":6.400328173,"remaining_time":13.72648998},
+{"learn":[0.5445660503],"iteration":159,"passed_time":6.440461005,"remaining_time":13.68597964},
+{"learn":[0.5442224699],"iteration":160,"passed_time":6.481040465,"remaining_time":13.64641439},
+{"learn":[0.543859774],"iteration":161,"passed_time":6.52097236,"remaining_time":13.60548554},
+{"learn":[0.5434344075],"iteration":162,"passed_time":6.562003764,"remaining_time":13.56684214},
+{"learn":[0.5428666007],"iteration":163,"passed_time":6.603254029,"remaining_time":13.52861801},
+{"learn":[0.5423044133],"iteration":164,"passed_time":6.644894077,"remaining_time":13.49114858},
+{"learn":[0.5420020087],"iteration":165,"passed_time":6.684952619,"remaining_time":13.45044684},
+{"learn":[0.5417372893],"iteration":166,"passed_time":6.723337347,"remaining_time":13.40641519},
+{"learn":[0.5411407194],"iteration":167,"passed_time":6.762916945,"remaining_time":13.36481206},
+{"learn":[0.5404920135],"iteration":168,"passed_time":6.801918901,"remaining_time":13.32210152},
+{"learn":[0.5397785131],"iteration":169,"passed_time":6.842442162,"remaining_time":13.28238773},
+{"learn":[0.5393213651],"iteration":170,"passed_time":6.882946055,"remaining_time":13.2426272},
+{"learn":[0.5386490362],"iteration":171,"passed_time":6.921060437,"remaining_time":13.1983013},
+{"learn":[0.5382774466],"iteration":172,"passed_time":6.961217638,"remaining_time":13.15790848},
+{"learn":[0.5377320683],"iteration":173,"passed_time":7.000915958,"remaining_time":13.11665863},
+{"learn":[0.5373269499],"iteration":174,"passed_time":7.042183818,"remaining_time":13.07834138},
+{"learn":[0.5366982872],"iteration":175,"passed_time":7.086525078,"remaining_time":13.04564844},
+{"learn":[0.5362725514],"iteration":176,"passed_time":7.125810676,"remaining_time":13.00359801},
+{"learn":[0.5359097265],"iteration":177,"passed_time":7.167077309,"remaining_time":12.96516232},
+{"learn":[0.5353663714],"iteration":178,"passed_time":7.208184267,"remaining_time":12.92640866},
+{"learn":[0.5350468465],"iteration":179,"passed_time":7.248966356,"remaining_time":12.8870513},
+{"learn":[0.5347738379],"iteration":180,"passed_time":7.289007813,"remaining_time":12.84637289},
+{"learn":[0.5344498463],"iteration":181,"passed_time":7.329722865,"remaining_time":12.80687841},
+{"learn":[0.5339643331],"iteration":182,"passed_time":7.372016126,"remaining_time":12.77010444},
+{"learn":[0.5333646702],"iteration":183,"passed_time":7.412257385,"remaining_time":12.72974638},
+{"learn":[0.5329642669],"iteration":184,"passed_time":7.452784455,"remaining_time":12.68987623},
+{"learn":[0.5325614811],"iteration":185,"passed_time":7.493825815,"remaining_time":12.65086724},
+{"learn":[0.5320511457],"iteration":186,"passed_time":7.533890911,"remaining_time":12.61020243},
+{"learn":[0.5312388201],"iteration":187,"passed_time":7.576458049,"remaining_time":12.57369634},
+{"learn":[0.5309182557],"iteration":188,"passed_time":7.613762085,"remaining_time":12.52846565},
+{"learn":[0.5305264945],"iteration":189,"passed_time":7.65370127,"remaining_time":12.48761786},
+{"learn":[0.5301171286],"iteration":190,"passed_time":7.69439896,"remaining_time":12.44800669},
+{"learn":[0.529628972],"iteration":191,"passed_time":7.731203552,"remaining_time":12.40213903},
+{"learn":[0.5293146426],"iteration":192,"passed_time":7.771236683,"remaining_time":12.36150084},
+{"learn":[0.5289805915],"iteration":193,"passed_time":7.809701603,"remaining_time":12.31839531},
+{"learn":[0.5284659827],"iteration":194,"passed_time":7.850776734,"remaining_time":12.27942002},
+{"learn":[0.527862758],"iteration":195,"passed_time":7.893241222,"remaining_time":12.24257822},
+{"learn":[0.5274831458],"iteration":196,"passed_time":7.933892328,"remaining_time":12.20289023},
+{"learn":[0.5271515407],"iteration":197,"passed_time":7.975761096,"remaining_time":12.16504975},
+{"learn":[0.5266917025],"iteration":198,"passed_time":8.016492688,"remaining_time":12.12544874},
+{"learn":[0.5263565126],"iteration":199,"passed_time":8.055864892,"remaining_time":12.08379734},
+{"learn":[0.5260532248],"iteration":200,"passed_time":8.097974672,"remaining_time":12.04624093},
+{"learn":[0.5257055661],"iteration":201,"passed_time":8.138347032,"remaining_time":12.00607631},
+{"learn":[0.5254257524],"iteration":202,"passed_time":8.179338712,"remaining_time":11.96681575},
+{"learn":[0.524950575],"iteration":203,"passed_time":8.220757708,"remaining_time":11.92815824},
+{"learn":[0.5244807342],"iteration":204,"passed_time":8.260426309,"remaining_time":11.88695493},
+{"learn":[0.524249665],"iteration":205,"passed_time":8.300203303,"remaining_time":11.84592122},
+{"learn":[0.5234250272],"iteration":206,"passed_time":8.34502979,"remaining_time":11.812047},
+{"learn":[0.5229129359],"iteration":207,"passed_time":8.388605832,"remaining_time":11.77631203},
+{"learn":[0.5225299607],"iteration":208,"passed_time":8.427975443,"remaining_time":11.73464523},
+{"learn":[0.5222031189],"iteration":209,"passed_time":8.468006924,"remaining_time":11.69391432},
+{"learn":[0.5216340204],"iteration":210,"passed_time":8.507851709,"remaining_time":11.65293433},
+{"learn":[0.5210763366],"iteration":211,"passed_time":8.548940072,"remaining_time":11.61365444},
+{"learn":[0.520463507],"iteration":212,"passed_time":8.588498498,"remaining_time":11.5722961},
+{"learn":[0.520267466],"iteration":213,"passed_time":8.626902226,"remaining_time":11.52941139},
+{"learn":[0.5198975885],"iteration":214,"passed_time":8.668438911,"remaining_time":11.49072135},
+{"learn":[0.5194786412],"iteration":215,"passed_time":8.705984759,"remaining_time":11.44675774},
+{"learn":[0.5191099277],"iteration":216,"passed_time":8.746476441,"remaining_time":11.40669508},
+{"learn":[0.5187960194],"iteration":217,"passed_time":8.786473239,"remaining_time":11.36598832},
+{"learn":[0.5185900424],"iteration":218,"passed_time":8.823951088,"remaining_time":11.32205596},
+{"learn":[0.5180914041],"iteration":219,"passed_time":8.863084333,"remaining_time":11.28028915},
+{"learn":[0.5176345588],"iteration":220,"passed_time":8.900090083,"remaining_time":11.23586033},
+{"learn":[0.5172814099],"iteration":221,"passed_time":8.938826084,"remaining_time":11.1936651},
+{"learn":[0.5167825065],"iteration":222,"passed_time":8.979981819,"remaining_time":11.15450656},
+{"learn":[0.5163917592],"iteration":223,"passed_time":9.017901356,"remaining_time":11.11134274},
+{"learn":[0.5161350334],"iteration":224,"passed_time":9.055954875,"remaining_time":11.06838929},
+{"learn":[0.5157133787],"iteration":225,"passed_time":9.096262688,"remaining_time":11.02821229},
+{"learn":[0.5152784319],"iteration":226,"passed_time":9.135476858,"remaining_time":10.98671886},
+{"learn":[0.5148803192],"iteration":227,"passed_time":9.174872097,"remaining_time":10.94546145},
+{"learn":[0.5144992304],"iteration":228,"passed_time":9.213074139,"remaining_time":10.90280826},
+{"learn":[0.5142535669],"iteration":229,"passed_time":9.251266863,"remaining_time":10.86018284},
+{"learn":[0.5139649031],"iteration":230,"passed_time":9.291579795,"remaining_time":10.82006478},
+{"learn":[0.513576616],"iteration":231,"passed_time":9.329301206,"remaining_time":10.77695139},
+{"learn":[0.5132171213],"iteration":232,"passed_time":9.373462235,"remaining_time":10.74126359},
+{"learn":[0.512763067],"iteration":233,"passed_time":9.415620196,"remaining_time":10.70322638},
+{"learn":[0.5123182195],"iteration":234,"passed_time":9.457430926,"remaining_time":10.66476253},
+{"learn":[0.5120334231],"iteration":235,"passed_time":9.499128928,"remaining_time":10.62614422},
+{"learn":[0.5115620916],"iteration":236,"passed_time":9.541151467,"remaining_time":10.58786007},
+{"learn":[0.5112360631],"iteration":237,"passed_time":9.582135427,"remaining_time":10.54840118},
+{"learn":[0.5107944946],"iteration":238,"passed_time":9.62076548,"remaining_time":10.50635895},
+{"learn":[0.5104228562],"iteration":239,"passed_time":9.660447764,"remaining_time":10.46548508},
+{"learn":[0.5102575232],"iteration":240,"passed_time":9.697855872,"remaining_time":10.42217706},
+{"learn":[0.5098562657],"iteration":241,"passed_time":9.735858017,"remaining_time":10.37955111},
+{"learn":[0.5096847762],"iteration":242,"passed_time":9.775358832,"remaining_time":10.33854823},
+{"learn":[0.5091203404],"iteration":243,"passed_time":9.813098405,"remaining_time":10.2957098},
+{"learn":[0.5084373577],"iteration":244,"passed_time":9.853299234,"remaining_time":10.25547471},
+{"learn":[0.5081469859],"iteration":245,"passed_time":9.893884024,"remaining_time":10.21563635},
+{"learn":[0.5079091193],"iteration":246,"passed_time":9.934162937,"remaining_time":10.17547864},
+{"learn":[0.5074818428],"iteration":247,"passed_time":9.973696095,"remaining_time":10.13456216},
+{"learn":[0.5071478105],"iteration":248,"passed_time":10.010796,"remaining_time":10.091204},
+{"learn":[0.5068570743],"iteration":249,"passed_time":10.04969919,"remaining_time":10.04969919},
+{"learn":[0.5063689292],"iteration":250,"passed_time":10.09027869,"remaining_time":10.00987806},
+{"learn":[0.5058932147],"iteration":251,"passed_time":10.13010371,"remaining_time":9.969308414},
+{"learn":[0.505662824],"iteration":252,"passed_time":10.16980173,"remaining_time":9.928620657},
+{"learn":[0.5052695033],"iteration":253,"passed_time":10.20813501,"remaining_time":9.88661895},
+{"learn":[0.5048413537],"iteration":254,"passed_time":10.24814495,"remaining_time":9.84625691},
+{"learn":[0.5045419313],"iteration":255,"passed_time":10.29170851,"remaining_time":9.809284674},
+{"learn":[0.5042639141],"iteration":256,"passed_time":10.33129754,"remaining_time":9.768503119},
+{"learn":[0.5036848245],"iteration":257,"passed_time":10.37259299,"remaining_time":9.729331411},
+{"learn":[0.5030974571],"iteration":258,"passed_time":10.41138646,"remaining_time":9.687815201},
+{"learn":[0.5027814387],"iteration":259,"passed_time":10.44951972,"remaining_time":9.64571051},
+{"learn":[0.5022425808],"iteration":260,"passed_time":10.49100051,"remaining_time":9.606701613},
+{"learn":[0.5018966871],"iteration":261,"passed_time":10.52942375,"remaining_time":9.564896388},
+{"learn":[0.5014832647],"iteration":262,"passed_time":10.56972825,"remaining_time":9.524812152},
+{"learn":[0.5010494012],"iteration":263,"passed_time":10.611519,"remaining_time":9.486054866},
+{"learn":[0.5006483949],"iteration":264,"passed_time":10.64991775,"remaining_time":9.444266684},
+{"learn":[0.5003421562],"iteration":265,"passed_time":10.68900222,"remaining_time":9.403107212},
+{"learn":[0.4999624129],"iteration":266,"passed_time":10.72766925,"remaining_time":9.361599008},
+{"learn":[0.4996104834],"iteration":267,"passed_time":10.76711852,"remaining_time":9.320789166},
+{"learn":[0.499196608],"iteration":268,"passed_time":10.80845489,"remaining_time":9.281609958},
+{"learn":[0.4988927675],"iteration":269,"passed_time":10.84767712,"remaining_time":9.240613847},
+{"learn":[0.4981819706],"iteration":270,"passed_time":10.89057552,"remaining_time":9.202737248},
+{"learn":[0.4978315761],"iteration":271,"passed_time":10.93026095,"remaining_time":9.162130505},
+{"learn":[0.4973900121],"iteration":272,"passed_time":10.97579339,"remaining_time":9.126392306},
+{"learn":[0.4970109103],"iteration":273,"passed_time":11.0178793,"remaining_time":9.087739857},
+{"learn":[0.4965620122],"iteration":274,"passed_time":11.05856208,"remaining_time":9.047914428},
+{"learn":[0.4962427936],"iteration":275,"passed_time":11.09959313,"remaining_time":9.008365435},
+{"learn":[0.4958221039],"iteration":276,"passed_time":11.14002026,"remaining_time":8.968319562},
+{"learn":[0.4954809033],"iteration":277,"passed_time":11.17942675,"remaining_time":8.927455894},
+{"learn":[0.4948766247],"iteration":278,"passed_time":11.23952648,"remaining_time":8.902994097},
+{"learn":[0.4943739105],"iteration":279,"passed_time":11.33652049,"remaining_time":8.907266098},
+{"learn":[0.493975467],"iteration":280,"passed_time":11.38120835,"remaining_time":8.870052061},
+{"learn":[0.4935910408],"iteration":281,"passed_time":11.42517148,"remaining_time":8.832224762},
+{"learn":[0.493254366],"iteration":282,"passed_time":11.46537852,"remaining_time":8.791473991},
+{"learn":[0.4929261473],"iteration":283,"passed_time":11.5082262,"remaining_time":8.752735421},
+{"learn":[0.4925964609],"iteration":284,"passed_time":11.54808255,"remaining_time":8.7117114},
+{"learn":[0.4923808028],"iteration":285,"passed_time":11.58898243,"remaining_time":8.671476365},
+{"learn":[0.4921041946],"iteration":286,"passed_time":11.62824726,"remaining_time":8.630023224},
+{"learn":[0.4916744584],"iteration":287,"passed_time":11.6689865,"remaining_time":8.589670616},
+{"learn":[0.4913336962],"iteration":288,"passed_time":11.70892409,"remaining_time":8.548730044},
+{"learn":[0.4910190759],"iteration":289,"passed_time":11.74708172,"remaining_time":8.506507451},
+{"learn":[0.4906179297],"iteration":290,"passed_time":11.78772812,"remaining_time":8.466100266},
+{"learn":[0.4902202329],"iteration":291,"passed_time":11.8283,"remaining_time":8.425638353},
+{"learn":[0.489883628],"iteration":292,"passed_time":11.86837945,"remaining_time":8.384827802},
+{"learn":[0.4896095455],"iteration":293,"passed_time":11.90837363,"remaining_time":8.343962478},
+{"learn":[0.4893633368],"iteration":294,"passed_time":11.94617866,"remaining_time":8.301581783},
+{"learn":[0.4891310576],"iteration":295,"passed_time":11.98512649,"remaining_time":8.260019606},
+{"learn":[0.4887803286],"iteration":296,"passed_time":12.02466397,"remaining_time":8.218878064},
+{"learn":[0.4884834707],"iteration":297,"passed_time":12.06409256,"remaining_time":8.17767348},
+{"learn":[0.4880679615],"iteration":298,"passed_time":12.10509267,"remaining_time":8.13753721},
+{"learn":[0.4878523011],"iteration":299,"passed_time":12.14457842,"remaining_time":8.096385614},
+{"learn":[0.4873943721],"iteration":300,"passed_time":12.18463777,"remaining_time":8.055624306},
+{"learn":[0.4871038251],"iteration":301,"passed_time":12.22525302,"remaining_time":8.015232111},
+{"learn":[0.4867376985],"iteration":302,"passed_time":12.2660809,"remaining_time":7.974976691},
+{"learn":[0.4864321061],"iteration":303,"passed_time":12.3077537,"remaining_time":7.935262251},
+{"learn":[0.486210863],"iteration":304,"passed_time":12.34709934,"remaining_time":7.894047122},
+{"learn":[0.4858144853],"iteration":305,"passed_time":12.38841429,"remaining_time":7.854092721},
+{"learn":[0.4856499406],"iteration":306,"passed_time":12.42938119,"remaining_time":7.81391065},
+{"learn":[0.485347457],"iteration":307,"passed_time":12.46763986,"remaining_time":7.772035238},
+{"learn":[0.4848908087],"iteration":308,"passed_time":12.50889462,"remaining_time":7.732035186},
+{"learn":[0.4845522859],"iteration":309,"passed_time":12.54665956,"remaining_time":7.689888115},
+{"learn":[0.4840624415],"iteration":310,"passed_time":12.58525315,"remaining_time":7.648272816},
+{"learn":[0.4837871556],"iteration":311,"passed_time":12.62511093,"remaining_time":7.607438638},
+{"learn":[0.4834783178],"iteration":312,"passed_time":12.66382714,"remaining_time":7.565928675},
+{"learn":[0.4831740873],"iteration":313,"passed_time":12.70379925,"remaining_time":7.525180446},
+{"learn":[0.4829482632],"iteration":314,"passed_time":12.74199926,"remaining_time":7.483396393},
+{"learn":[0.4826739322],"iteration":315,"passed_time":12.78011649,"remaining_time":7.441586819},
+{"learn":[0.4823400925],"iteration":316,"passed_time":12.81872013,"remaining_time":7.400081337},
+{"learn":[0.4819943761],"iteration":317,"passed_time":12.85645718,"remaining_time":7.358098134},
+{"learn":[0.4817866029],"iteration":318,"passed_time":12.89481498,"remaining_time":7.316493766},
+{"learn":[0.4815804613],"iteration":319,"passed_time":12.93191893,"remaining_time":7.274204398},
+{"learn":[0.4809581003],"iteration":320,"passed_time":12.97193856,"remaining_time":7.233573216},
+{"learn":[0.4807254623],"iteration":321,"passed_time":13.00897046,"remaining_time":7.191294232},
+{"learn":[0.4802466482],"iteration":322,"passed_time":13.04820646,"remaining_time":7.150255554},
+{"learn":[0.4798784133],"iteration":323,"passed_time":13.08581465,"remaining_time":7.108343758},
+{"learn":[0.4797171462],"iteration":324,"passed_time":13.12251345,"remaining_time":7.065968782},
+{"learn":[0.4793949176],"iteration":325,"passed_time":13.16243026,"remaining_time":7.025346211},
+{"learn":[0.4788800195],"iteration":326,"passed_time":13.20162868,"remaining_time":6.984347897},
+{"learn":[0.4786136279],"iteration":327,"passed_time":13.23906222,"remaining_time":6.942435067},
+{"learn":[0.4782427334],"iteration":328,"passed_time":13.27842277,"remaining_time":6.901551046},
+{"learn":[0.4778863666],"iteration":329,"passed_time":13.3177195,"remaining_time":6.860643376},
+{"learn":[0.4775479356],"iteration":330,"passed_time":13.35590051,"remaining_time":6.819175789},
+{"learn":[0.4773015039],"iteration":331,"passed_time":13.39427948,"remaining_time":6.777828168},
+{"learn":[0.4770363259],"iteration":332,"passed_time":13.43297386,"remaining_time":6.736656562},
+{"learn":[0.4768298065],"iteration":333,"passed_time":13.47171123,"remaining_time":6.695521149},
+{"learn":[0.4765892571],"iteration":334,"passed_time":13.50934006,"remaining_time":6.65385406},
+{"learn":[0.4764456817],"iteration":335,"passed_time":13.54683855,"remaining_time":6.612147387},
+{"learn":[0.4763636442],"iteration":336,"passed_time":13.58543918,"remaining_time":6.570998773},
+{"learn":[0.4760378853],"iteration":337,"passed_time":13.62321996,"remaining_time":6.529472287},
+{"learn":[0.4758876647],"iteration":338,"passed_time":13.66171038,"remaining_time":6.488304927},
+{"learn":[0.4756349977],"iteration":339,"passed_time":13.69995533,"remaining_time":6.447037804},
+{"learn":[0.4752603973],"iteration":340,"passed_time":13.73959929,"remaining_time":6.406440724},
+{"learn":[0.4750204768],"iteration":341,"passed_time":13.77948293,"remaining_time":6.36595995},
+{"learn":[0.4748473115],"iteration":342,"passed_time":13.81901144,"remaining_time":6.325320107},
+{"learn":[0.4745999562],"iteration":343,"passed_time":13.85710386,"remaining_time":6.28403547},
+{"learn":[0.4743993595],"iteration":344,"passed_time":13.89464506,"remaining_time":6.242521695},
+{"learn":[0.4737070725],"iteration":345,"passed_time":13.93554055,"remaining_time":6.202523827},
+{"learn":[0.4734187079],"iteration":346,"passed_time":13.97598447,"remaining_time":6.162321681},
+{"learn":[0.4732100567],"iteration":347,"passed_time":14.01500159,"remaining_time":6.121494947},
+{"learn":[0.4730721577],"iteration":348,"passed_time":14.05227157,"remaining_time":6.079922658},
+{"learn":[0.4728648397],"iteration":349,"passed_time":14.09011358,"remaining_time":6.038620107},
+{"learn":[0.4726551113],"iteration":350,"passed_time":14.12743823,"remaining_time":5.997117654},
+{"learn":[0.4723837123],"iteration":351,"passed_time":14.16561214,"remaining_time":5.955996013},
+{"learn":[0.4722697334],"iteration":352,"passed_time":14.20317291,"remaining_time":5.914635742},
+{"learn":[0.4720394971],"iteration":353,"passed_time":14.23982658,"remaining_time":5.872922827},
+{"learn":[0.4717581939],"iteration":354,"passed_time":14.27959578,"remaining_time":5.832510954},
+{"learn":[0.4715443272],"iteration":355,"passed_time":14.3193661,"remaining_time":5.79210314},
+{"learn":[0.4713122025],"iteration":356,"passed_time":14.3568939,"remaining_time":5.750800636},
+{"learn":[0.4709434579],"iteration":357,"passed_time":14.39647857,"remaining_time":5.710335074},
+{"learn":[0.4707309341],"iteration":358,"passed_time":14.43679514,"remaining_time":5.670161881},
+{"learn":[0.4705579301],"iteration":359,"passed_time":14.47396182,"remaining_time":5.628762928},
+{"learn":[0.469916571],"iteration":360,"passed_time":14.51502942,"remaining_time":5.588889445},
+{"learn":[0.4695748498],"iteration":361,"passed_time":14.55229114,"remaining_time":5.547558501},
+{"learn":[0.4695389297],"iteration":362,"passed_time":14.59137289,"remaining_time":5.506936876},
+{"learn":[0.4693263375],"iteration":363,"passed_time":14.63077441,"remaining_time":5.466443185},
+{"learn":[0.4692589765],"iteration":364,"passed_time":14.66892758,"remaining_time":5.425493764},
+{"learn":[0.468858875],"iteration":365,"passed_time":14.70910911,"remaining_time":5.385302244},
+{"learn":[0.4685511528],"iteration":366,"passed_time":14.74699822,"remaining_time":5.344280007},
+{"learn":[0.4682883236],"iteration":367,"passed_time":14.78587938,"remaining_time":5.303630646},
+{"learn":[0.4680511877],"iteration":368,"passed_time":14.82443206,"remaining_time":5.262874255},
+{"learn":[0.467827367],"iteration":369,"passed_time":14.86298389,"remaining_time":5.222129476},
+{"learn":[0.4675352989],"iteration":370,"passed_time":14.9010351,"remaining_time":5.181222447},
+{"learn":[0.4673511276],"iteration":371,"passed_time":14.93934355,"remaining_time":5.140419287},
+{"learn":[0.467120547],"iteration":372,"passed_time":14.97746344,"remaining_time":5.099565298},
+{"learn":[0.4668196585],"iteration":373,"passed_time":15.01506871,"remaining_time":5.05855256},
+{"learn":[0.4664704386],"iteration":374,"passed_time":15.05342594,"remaining_time":5.017808645},
+{"learn":[0.466300754],"iteration":375,"passed_time":15.09126546,"remaining_time":4.976906693},
+{"learn":[0.4660529477],"iteration":376,"passed_time":15.12840164,"remaining_time":4.935791516},
+{"learn":[0.4659205615],"iteration":377,"passed_time":15.16662956,"remaining_time":4.895049753},
+{"learn":[0.4656226599],"iteration":378,"passed_time":15.20786954,"remaining_time":4.855282889},
+{"learn":[0.4651948256],"iteration":379,"passed_time":15.24568062,"remaining_time":4.81442546},
+{"learn":[0.4649232323],"iteration":380,"passed_time":15.2852586,"remaining_time":4.774135888},
+{"learn":[0.4645354476],"iteration":381,"passed_time":15.32520827,"remaining_time":4.733964858},
+{"learn":[0.4639030521],"iteration":382,"passed_time":15.36456206,"remaining_time":4.693612954},
+{"learn":[0.4635351896],"iteration":383,"passed_time":15.4029265,"remaining_time":4.652967379},
+{"learn":[0.4633385917],"iteration":384,"passed_time":15.44276281,"remaining_time":4.612773306},
+{"learn":[0.4631669568],"iteration":385,"passed_time":15.48167227,"remaining_time":4.572307353},
+{"learn":[0.4630072732],"iteration":386,"passed_time":15.52134492,"remaining_time":4.532072289},
+{"learn":[0.4627835133],"iteration":387,"passed_time":15.56077337,"remaining_time":4.491769633},
+{"learn":[0.4627417677],"iteration":388,"passed_time":15.59866463,"remaining_time":4.451032838},
+{"learn":[0.4627061504],"iteration":389,"passed_time":15.63803304,"remaining_time":4.410727267},
+{"learn":[0.4625064093],"iteration":390,"passed_time":15.67585136,"remaining_time":4.36999437},
+{"learn":[0.4619443772],"iteration":391,"passed_time":15.71630952,"remaining_time":4.330003643},
+{"learn":[0.4617952848],"iteration":392,"passed_time":15.75542667,"remaining_time":4.289645429},
+{"learn":[0.4614354597],"iteration":393,"passed_time":15.79521647,"remaining_time":4.249474482},
+{"learn":[0.4612137172],"iteration":394,"passed_time":15.83403142,"remaining_time":4.209046327},
+{"learn":[0.4609780065],"iteration":395,"passed_time":15.87373263,"remaining_time":4.168859075},
+{"learn":[0.460803717],"iteration":396,"passed_time":15.9138121,"remaining_time":4.128772409},
+{"learn":[0.4605055516],"iteration":397,"passed_time":15.9524316,"remaining_time":4.088311617},
+{"learn":[0.4604654562],"iteration":398,"passed_time":15.99223648,"remaining_time":4.048160112},
+{"learn":[0.4601521805],"iteration":399,"passed_time":16.03250432,"remaining_time":4.00812608},
+{"learn":[0.4601218957],"iteration":400,"passed_time":16.07042302,"remaining_time":3.967510921},
+{"learn":[0.4598802793],"iteration":401,"passed_time":16.1095588,"remaining_time":3.927205876},
+{"learn":[0.4596294913],"iteration":402,"passed_time":16.1486524,"remaining_time":3.886896484},
+{"learn":[0.459359852],"iteration":403,"passed_time":16.19024308,"remaining_time":3.847186474},
+{"learn":[0.4593234097],"iteration":404,"passed_time":16.2285362,"remaining_time":3.806693678},
+{"learn":[0.4590768781],"iteration":405,"passed_time":16.26882365,"remaining_time":3.766673457},
+{"learn":[0.458797219],"iteration":406,"passed_time":16.30805413,"remaining_time":3.726410404},
+{"learn":[0.4585924348],"iteration":407,"passed_time":16.34615651,"remaining_time":3.685898037},
+{"learn":[0.4585649874],"iteration":408,"passed_time":16.3867931,"remaining_time":3.6459613},
+{"learn":[0.4585363276],"iteration":409,"passed_time":16.4253083,"remaining_time":3.60555548},
+{"learn":[0.4579354313],"iteration":410,"passed_time":16.46443938,"remaining_time":3.565292225},
+{"learn":[0.4575848613],"iteration":411,"passed_time":16.50445659,"remaining_time":3.525223738},
+{"learn":[0.4574072665],"iteration":412,"passed_time":16.54247175,"remaining_time":3.484733759},
+{"learn":[0.4572256847],"iteration":413,"passed_time":16.58392085,"remaining_time":3.444969065},
+{"learn":[0.4569922307],"iteration":414,"passed_time":16.62180527,"remaining_time":3.404466139},
+{"learn":[0.4564843783],"iteration":415,"passed_time":16.66197238,"remaining_time":3.364436731},
+{"learn":[0.4562989952],"iteration":416,"passed_time":16.70174115,"remaining_time":3.324327376},
+{"learn":[0.4557830648],"iteration":417,"passed_time":16.74166908,"remaining_time":3.284250872},
+{"learn":[0.4552632395],"iteration":418,"passed_time":16.7829432,"remaining_time":3.24443532},
+{"learn":[0.45506771],"iteration":419,"passed_time":16.82025733,"remaining_time":3.20385854},
+{"learn":[0.4548371775],"iteration":420,"passed_time":16.86006607,"remaining_time":3.163765367},
+{"learn":[0.4543536083],"iteration":421,"passed_time":16.90189939,"remaining_time":3.124047755},
+{"learn":[0.4539135342],"iteration":422,"passed_time":16.94044689,"remaining_time":3.083722011},
+{"learn":[0.4536574805],"iteration":423,"passed_time":16.98011906,"remaining_time":3.043606247},
+{"learn":[0.4532930869],"iteration":424,"passed_time":17.01910529,"remaining_time":3.003371522},
+{"learn":[0.4530580282],"iteration":425,"passed_time":17.05548332,"remaining_time":2.96268959},
+{"learn":[0.4528470071],"iteration":426,"passed_time":17.09186302,"remaining_time":2.922028103},
+{"learn":[0.4525708397],"iteration":427,"passed_time":17.12905118,"remaining_time":2.881522628},
+{"learn":[0.4524322649],"iteration":428,"passed_time":17.16634356,"remaining_time":2.841049866},
+{"learn":[0.4521065026],"iteration":429,"passed_time":17.20502448,"remaining_time":2.800817939},
+{"learn":[0.4519013222],"iteration":430,"passed_time":17.24175028,"remaining_time":2.760280207},
+{"learn":[0.4518403387],"iteration":431,"passed_time":17.27868055,"remaining_time":2.719792308},
+{"learn":[0.4516184857],"iteration":432,"passed_time":17.31922401,"remaining_time":2.679879928},
+{"learn":[0.4513714288],"iteration":433,"passed_time":17.35983933,"remaining_time":2.639975567},
+{"learn":[0.451037779],"iteration":434,"passed_time":17.39995029,"remaining_time":2.599992572},
+{"learn":[0.4507709996],"iteration":435,"passed_time":17.43917189,"remaining_time":2.559878443},
+{"learn":[0.4505847307],"iteration":436,"passed_time":17.47957931,"remaining_time":2.519939351},
+{"learn":[0.4503895992],"iteration":437,"passed_time":17.51735354,"remaining_time":2.479625387},
+{"learn":[0.4499681477],"iteration":438,"passed_time":17.55697072,"remaining_time":2.439579075},
+{"learn":[0.4496796353],"iteration":439,"passed_time":17.59670634,"remaining_time":2.399550864},
+{"learn":[0.4494430951],"iteration":440,"passed_time":17.63396148,"remaining_time":2.359192126},
+{"learn":[0.4491757112],"iteration":441,"passed_time":17.6761703,"remaining_time":2.31949746},
+{"learn":[0.4489244193],"iteration":442,"passed_time":17.71672844,"remaining_time":2.279579054},
+{"learn":[0.4485276391],"iteration":443,"passed_time":17.75477825,"remaining_time":2.2393414},
+{"learn":[0.4481718053],"iteration":444,"passed_time":17.79171012,"remaining_time":2.198975408},
+{"learn":[0.4479506854],"iteration":445,"passed_time":17.82916459,"remaining_time":2.15868809},
+{"learn":[0.4476990522],"iteration":446,"passed_time":17.86739636,"remaining_time":2.118505608},
+{"learn":[0.4471948591],"iteration":447,"passed_time":17.90606749,"remaining_time":2.078382833},
+{"learn":[0.4470853092],"iteration":448,"passed_time":17.94235621,"remaining_time":2.037995917},
+{"learn":[0.446865776],"iteration":449,"passed_time":17.98269432,"remaining_time":1.998077146},
+{"learn":[0.4465824143],"iteration":450,"passed_time":18.02023885,"remaining_time":1.957853001},
+{"learn":[0.4461086564],"iteration":451,"passed_time":18.0593718,"remaining_time":1.917809395},
+{"learn":[0.4460486522],"iteration":452,"passed_time":18.09790554,"remaining_time":1.877707639},
+{"learn":[0.4458924182],"iteration":453,"passed_time":18.13555647,"remaining_time":1.837523342},
+{"learn":[0.4457369368],"iteration":454,"passed_time":18.17506831,"remaining_time":1.797534229},
+{"learn":[0.4457090735],"iteration":455,"passed_time":18.2140202,"remaining_time":1.757493177},
+{"learn":[0.4453923753],"iteration":456,"passed_time":18.25304478,"remaining_time":1.717463732},
+{"learn":[0.4453669142],"iteration":457,"passed_time":18.29142346,"remaining_time":1.677379444},
+{"learn":[0.4450574113],"iteration":458,"passed_time":18.33140201,"remaining_time":1.637445496},
+{"learn":[0.4448489841],"iteration":459,"passed_time":18.3714938,"remaining_time":1.5975212},
+{"learn":[0.4446117109],"iteration":460,"passed_time":18.41142161,"remaining_time":1.557582306},
+{"learn":[0.4444696636],"iteration":461,"passed_time":18.44769458,"remaining_time":1.517342844},
+{"learn":[0.444285799],"iteration":462,"passed_time":18.48687463,"remaining_time":1.477352832},
+{"learn":[0.4441396144],"iteration":463,"passed_time":18.52464426,"remaining_time":1.437256882},
+{"learn":[0.4437465091],"iteration":464,"passed_time":18.56461611,"remaining_time":1.397336697},
+{"learn":[0.4435487801],"iteration":465,"passed_time":18.60374158,"remaining_time":1.357354536},
+{"learn":[0.4433583675],"iteration":466,"passed_time":18.64031576,"remaining_time":1.317195761},
+{"learn":[0.4431235913],"iteration":467,"passed_time":18.68082829,"remaining_time":1.277321592},
+{"learn":[0.4428925062],"iteration":468,"passed_time":18.71932798,"remaining_time":1.237311657},
+{"learn":[0.4425307044],"iteration":469,"passed_time":18.7579164,"remaining_time":1.197313813},
+{"learn":[0.4424258336],"iteration":470,"passed_time":18.7979589,"remaining_time":1.157411482},
+{"learn":[0.4424000888],"iteration":471,"passed_time":18.83596343,"remaining_time":1.117387661},
+{"learn":[0.4422433144],"iteration":472,"passed_time":18.87643621,"remaining_time":1.077513272},
+{"learn":[0.4420168935],"iteration":473,"passed_time":18.91820045,"remaining_time":1.037707198},
+{"learn":[0.4417496517],"iteration":474,"passed_time":18.95977115,"remaining_time":0.9978826919},
+{"learn":[0.4414042041],"iteration":475,"passed_time":18.99903921,"remaining_time":0.9579347499},
+{"learn":[0.4411749674],"iteration":476,"passed_time":19.03723674,"remaining_time":0.9179380397},
+{"learn":[0.4407943905],"iteration":477,"passed_time":19.07654313,"remaining_time":0.8779998928},
+{"learn":[0.440671799],"iteration":478,"passed_time":19.11741207,"remaining_time":0.8381328882},
+{"learn":[0.4403708967],"iteration":479,"passed_time":19.15732612,"remaining_time":0.7982219217},
+{"learn":[0.4401620051],"iteration":480,"passed_time":19.19434474,"remaining_time":0.7581965698},
+{"learn":[0.4399914723],"iteration":481,"passed_time":19.23190252,"remaining_time":0.7182038287},
+{"learn":[0.4396355813],"iteration":482,"passed_time":19.27081271,"remaining_time":0.6782687703},
+{"learn":[0.4396078413],"iteration":483,"passed_time":19.30856357,"remaining_time":0.6382996223},
+{"learn":[0.4394233159],"iteration":484,"passed_time":19.34524654,"remaining_time":0.598306594},
+{"learn":[0.4392648597],"iteration":485,"passed_time":19.38319882,"remaining_time":0.5583637521},
+{"learn":[0.4390081285],"iteration":486,"passed_time":19.42192094,"remaining_time":0.5184496349},
+{"learn":[0.4389768991],"iteration":487,"passed_time":19.46118645,"remaining_time":0.4785537652},
+{"learn":[0.4386848902],"iteration":488,"passed_time":19.4996181,"remaining_time":0.438641716},
+{"learn":[0.4383633005],"iteration":489,"passed_time":19.53735017,"remaining_time":0.398721432},
+{"learn":[0.4382373497],"iteration":490,"passed_time":19.57670653,"remaining_time":0.3588398345},
+{"learn":[0.4379751734],"iteration":491,"passed_time":19.61602864,"remaining_time":0.3189598153},
+{"learn":[0.437740467],"iteration":492,"passed_time":19.65584935,"remaining_time":0.2790891388},
+{"learn":[0.4377072609],"iteration":493,"passed_time":19.69331298,"remaining_time":0.2391900362},
+{"learn":[0.437475001],"iteration":494,"passed_time":19.73367475,"remaining_time":0.199330048},
+{"learn":[0.4374511985],"iteration":495,"passed_time":19.77278216,"remaining_time":0.1594579207},
+{"learn":[0.4372524578],"iteration":496,"passed_time":19.81221839,"remaining_time":0.1195908554},
+{"learn":[0.4370094577],"iteration":497,"passed_time":19.84905419,"remaining_time":0.07971507707},
+{"learn":[0.4367382154],"iteration":498,"passed_time":19.88683169,"remaining_time":0.03985337013},
+{"learn":[0.4365629412],"iteration":499,"passed_time":19.92567299,"remaining_time":0}
]}
\ No newline at end of file
diff --git a/catboost_info/learn/events.out.tfevents b/catboost_info/learn/events.out.tfevents
index 506b9cc35d0a0483484eac2299b26a91f5423146..55e50976f56d196ab564e9212f8fe02d2c3c0ade 100644
Binary files a/catboost_info/learn/events.out.tfevents and b/catboost_info/learn/events.out.tfevents differ
diff --git a/catboost_info/learn_error.tsv b/catboost_info/learn_error.tsv
index 0f68162ed1c2a25dfea70acbc4cb2f659044d61b..3c696d0345d8ef6a073ae42ef57021ff0bfff2aa 100644
--- a/catboost_info/learn_error.tsv
+++ b/catboost_info/learn_error.tsv
@@ -1,501 +1,501 @@
iter MultiClass
-0 1.068921125
-1 1.042920351
-2 1.01801279
-3 0.9955724455
-4 0.9736690809
-5 0.9539294313
-6 0.9356639615
-7 0.9191377944
-8 0.9031989432
-9 0.8878384884
-10 0.8738034802
-11 0.8609447068
-12 0.8494076364
-13 0.8376116526
-14 0.8262527645
-15 0.8156703929
-16 0.8061405245
-17 0.7962708289
-18 0.787517053
-19 0.7790596225
-20 0.7707419955
-21 0.7628618298
-22 0.7568458468
-23 0.7496544902
-24 0.7429806121
-25 0.7365733331
-26 0.7301107638
-27 0.7244057584
-28 0.7183963567
-29 0.7133137141
-30 0.7081582063
-31 0.7029195396
-32 0.6981650667
-33 0.694229036
-34 0.6898580387
-35 0.6858643858
-36 0.6818898314
-37 0.6782096663
-38 0.6749976546
-39 0.6717289324
-40 0.6685944214
-41 0.6656078326
-42 0.6627272627
-43 0.6596591536
-44 0.6567951356
-45 0.6539799136
-46 0.6512287405
-47 0.6483621371
-48 0.646045433
-49 0.6436106576
-50 0.6412037332
-51 0.6385693614
-52 0.6359669929
-53 0.6336380251
-54 0.6317596332
-55 0.6300162524
-56 0.6279497237
-57 0.626202365
-58 0.6245478766
-59 0.6232750103
-60 0.6216488728
-61 0.6199963519
-62 0.6183914027
-63 0.6167257552
-64 0.6151270752
-65 0.6136229865
-66 0.6124914371
-67 0.6112142134
-68 0.6100471536
-69 0.6088553938
-70 0.6077217705
-71 0.6065233622
-72 0.6055522725
-73 0.6042869014
-74 0.6029862228
-75 0.6020286548
-76 0.6007111945
-77 0.5995650312
-78 0.598267565
-79 0.5973018607
-80 0.5964098279
-81 0.5952645015
-82 0.5939932919
-83 0.5927819224
-84 0.5916101899
-85 0.5904890406
-86 0.5898393665
-87 0.5889665637
-88 0.5881368208
-89 0.5871073965
-90 0.5861402271
-91 0.5852869207
-92 0.584532599
-93 0.5839068284
-94 0.5831785979
-95 0.582371746
-96 0.581512497
-97 0.5808661149
-98 0.5802827802
-99 0.5798054416
-100 0.5788760699
-101 0.5780138828
-102 0.5771851598
-103 0.5764314408
-104 0.5756151644
-105 0.5747869309
-106 0.5739388675
-107 0.5732935814
-108 0.5724740546
-109 0.5718664243
-110 0.5711293162
-111 0.5705779324
-112 0.569735314
-113 0.5686404377
-114 0.5680178459
-115 0.5671996108
-116 0.5665763407
-117 0.5657710765
-118 0.5653197317
-119 0.5646326931
-120 0.5641755543
-121 0.5636121495
-122 0.5629206857
-123 0.5623847046
-124 0.5618361757
-125 0.5613260958
-126 0.5605573167
-127 0.559966854
-128 0.5593610441
-129 0.5587625708
-130 0.5580857121
-131 0.557514446
-132 0.556943387
-133 0.5563473065
-134 0.5559281273
-135 0.5554834032
-136 0.5547240917
-137 0.554286871
-138 0.5539844822
-139 0.5531296918
-140 0.5526438864
-141 0.5521995723
-142 0.5517475129
-143 0.5512807855
-144 0.5505009241
-145 0.5499843323
-146 0.5494553371
-147 0.548825567
-148 0.5483547038
-149 0.5479018041
-150 0.5474866017
-151 0.5469354154
-152 0.5464195173
-153 0.5462957589
-154 0.5457682748
-155 0.5453126854
-156 0.5449104012
-157 0.5446294358
-158 0.5441676844
-159 0.5436853001
-160 0.54326474
-161 0.5428189052
-162 0.5424821708
-163 0.5420205298
-164 0.5415372914
-165 0.5411230188
-166 0.5404682421
-167 0.5401062328
-168 0.5396218645
-169 0.539240618
-170 0.5388113906
-171 0.538518505
-172 0.5382252585
-173 0.5376825766
-174 0.5372177332
-175 0.5368502891
-176 0.536433271
-177 0.5360177821
-178 0.5355686621
-179 0.5352466243
-180 0.5349531898
-181 0.5345933818
-182 0.5340879627
-183 0.5334066003
-184 0.5331344614
-185 0.5327729413
-186 0.5322668217
-187 0.5314967978
-188 0.5310815884
-189 0.5306990817
-190 0.5303294313
-191 0.5297662228
-192 0.5294658769
-193 0.5290541985
-194 0.528513619
-195 0.5280691496
-196 0.5277436683
-197 0.5273735004
-198 0.5269597174
-199 0.526580619
-200 0.5261678093
-201 0.5257991482
-202 0.5255372878
-203 0.5252407006
-204 0.524997375
-205 0.5247760811
-206 0.523988662
-207 0.5235269686
-208 0.5231391961
-209 0.522802282
-210 0.5221873746
-211 0.5216496981
-212 0.5210661583
-213 0.5207763789
-214 0.5202717603
-215 0.5197573737
-216 0.5193722401
-217 0.5190276841
-218 0.5188233127
-219 0.5183115318
-220 0.5179261404
-221 0.5175413522
-222 0.5169007258
-223 0.5165349089
-224 0.5163301026
-225 0.5160285647
-226 0.5156218452
-227 0.5152165155
-228 0.5148155711
-229 0.5143885804
-230 0.5140794854
-231 0.5136827366
-232 0.5132724458
-233 0.5129407929
-234 0.5124816054
-235 0.5121399791
-236 0.5118334093
-237 0.5114700787
-238 0.5111381929
-239 0.5108294424
-240 0.5106367493
-241 0.5101473498
-242 0.5098278979
-243 0.509406577
-244 0.5087198045
-245 0.5084669911
-246 0.5082518498
-247 0.5078444417
-248 0.5074409544
-249 0.5071251144
-250 0.5067451669
-251 0.5061366629
-252 0.5058645418
-253 0.5054086685
-254 0.5048886353
-255 0.5045785573
-256 0.5040652562
-257 0.5036517135
-258 0.5031239291
-259 0.5027971955
-260 0.5022963486
-261 0.5017921176
-262 0.5013332178
-263 0.5009457363
-264 0.5007438477
-265 0.5003634359
-266 0.4997749466
-267 0.4995323853
-268 0.4992080672
-269 0.4988797048
-270 0.4984015217
-271 0.4977663234
-272 0.4973336106
-273 0.4969323188
-274 0.4966561773
-275 0.496340795
-276 0.4959069085
-277 0.4954098916
-278 0.4950271537
-279 0.4946431444
-280 0.4944029335
-281 0.4942005513
-282 0.4938728298
-283 0.4935146349
-284 0.493156001
-285 0.4928198856
-286 0.492561635
-287 0.4922082132
-288 0.4917932108
-289 0.4915991369
-290 0.4912496628
-291 0.4909530477
-292 0.4905833112
-293 0.4902644174
-294 0.4899068175
-295 0.4895770055
-296 0.4892402572
-297 0.4889109272
-298 0.4887388656
-299 0.488525413
-300 0.4883646535
-301 0.4880313834
-302 0.487600658
-303 0.4872873471
-304 0.4870871276
-305 0.4864906459
-306 0.486156593
-307 0.4856904754
-308 0.4852077101
-309 0.4850039883
-310 0.4847537811
-311 0.4845627209
-312 0.4845330932
-313 0.4839737251
-314 0.4835309302
-315 0.4831540218
-316 0.4828460852
-317 0.4825408312
-318 0.4820667862
-319 0.4818882741
-320 0.4814636366
-321 0.481184684
-322 0.4807144658
-323 0.4804663756
-324 0.4800794796
-325 0.4798931718
-326 0.4795692068
-327 0.4791146234
-328 0.4786749759
-329 0.4784382168
-330 0.4781887154
-331 0.4780572928
-332 0.4777415159
-333 0.4774675269
-334 0.4773121415
-335 0.4769842799
-336 0.4766920756
-337 0.4764646891
-338 0.4760961631
-339 0.4758210384
-340 0.4754372506
-341 0.4752409719
-342 0.4750971587
-343 0.4747375371
-344 0.4745017945
-345 0.4742736682
-346 0.4739464879
-347 0.4736754421
-348 0.4734438174
-349 0.4732039158
-350 0.4726483693
-351 0.4723924666
-352 0.472146367
-353 0.4719615914
-354 0.4718524672
-355 0.4715708345
-356 0.4712705574
-357 0.4710028815
-358 0.4707747208
-359 0.470403996
-360 0.4702629921
-361 0.4699002207
-362 0.4697222917
-363 0.4692968498
-364 0.4691166011
-365 0.4688993415
-366 0.4685782266
-367 0.468261587
-368 0.4680112053
-369 0.4677463278
-370 0.4673304386
-371 0.4671175323
-372 0.4668852125
-373 0.4666989726
-374 0.4663318389
-375 0.4660052012
-376 0.4656110221
-377 0.4653915722
-378 0.4651173142
-379 0.4649721562
-380 0.4647070469
-381 0.4646002414
-382 0.4643426168
-383 0.4639029335
-384 0.4635395322
-385 0.4632807954
-386 0.463002022
-387 0.462628978
-388 0.4620670856
-389 0.4619002464
-390 0.4613682405
-391 0.4610835171
-392 0.4606795599
-393 0.4604993491
-394 0.4602709248
-395 0.460042721
-396 0.4595936079
-397 0.4592798285
-398 0.458882499
-399 0.4586580693
-400 0.4581193803
-401 0.4578728731
-402 0.4576978232
-403 0.457407343
-404 0.4569395552
-405 0.4563988559
-406 0.4561760676
-407 0.4558851807
-408 0.4556379291
-409 0.4553495443
-410 0.4548500646
-411 0.4547152754
-412 0.454511923
-413 0.4543781494
-414 0.4542514987
-415 0.4541160366
-416 0.4539945058
-417 0.4538843921
-418 0.4537582704
-419 0.4535486479
-420 0.453304122
-421 0.4530209968
-422 0.4529209824
-423 0.4527216047
-424 0.4523943482
-425 0.452075209
-426 0.4519256599
-427 0.4517125164
-428 0.4514656506
-429 0.4511500975
-430 0.4509773881
-431 0.4507410304
-432 0.4504520508
-433 0.4502464623
-434 0.4499163444
-435 0.4495444335
-436 0.4492822328
-437 0.4490709995
-438 0.4488742685
-439 0.4488202023
-440 0.4487087322
-441 0.448424538
-442 0.4481589914
-443 0.4479120738
-444 0.4478134597
-445 0.4475742582
-446 0.4472707527
-447 0.4470743065
-448 0.4467388758
-449 0.4465189068
-450 0.4462596129
-451 0.4460411878
-452 0.4459891391
-453 0.4457786012
-454 0.4456593077
-455 0.4454243826
-456 0.445302389
-457 0.4451221153
-458 0.4447792989
-459 0.4447178247
-460 0.444417161
-461 0.4441799003
-462 0.4437149583
-463 0.4434962894
-464 0.4433650256
-465 0.4431129225
-466 0.4430382611
-467 0.4427619817
-468 0.4426803452
-469 0.4423056868
-470 0.4420721901
-471 0.4417882376
-472 0.4416311924
-473 0.4413658561
-474 0.4411799644
-475 0.4408430967
-476 0.4406326451
-477 0.4404920207
-478 0.4402559337
-479 0.4401127627
-480 0.4398614027
-481 0.4396763562
-482 0.4392673677
-483 0.4388877804
-484 0.438673879
-485 0.4383851937
-486 0.4381989892
-487 0.4379762119
-488 0.437736575
-489 0.4372321637
-490 0.4369799398
-491 0.4365836973
-492 0.4363145001
-493 0.4360217335
-494 0.4358808553
-495 0.4357672486
-496 0.4354725607
-497 0.4353642518
-498 0.4350391341
-499 0.4347724016
+0 1.068942757
+1 1.042941062
+2 1.017818973
+3 0.9954188838
+4 0.9735267885
+5 0.9537807289
+6 0.9354755366
+7 0.9187623571
+8 0.9023127175
+9 0.8868191679
+10 0.8730040311
+11 0.860627421
+12 0.8491307201
+13 0.8373120751
+14 0.825819299
+15 0.815364983
+16 0.8058413028
+17 0.7959650463
+18 0.7868600676
+19 0.7784883379
+20 0.7701307612
+21 0.7622975086
+22 0.7563031265
+23 0.7495887487
+24 0.7429723054
+25 0.7372392086
+26 0.7307956392
+27 0.72524455
+28 0.7192748437
+29 0.7139143089
+30 0.7087284348
+31 0.7037127192
+32 0.6988890737
+33 0.6949220701
+34 0.6905132299
+35 0.6865428783
+36 0.6825442625
+37 0.678493613
+38 0.6753394188
+39 0.6720286547
+40 0.6689512826
+41 0.6659716333
+42 0.6630764989
+43 0.659979405
+44 0.6571206305
+45 0.6543860008
+46 0.6516459562
+47 0.6488159878
+48 0.6465175253
+49 0.6441297453
+50 0.6416986449
+51 0.6391230496
+52 0.6366229183
+53 0.6342664881
+54 0.6323079472
+55 0.6305081074
+56 0.62840159
+57 0.6266940801
+58 0.6248571874
+59 0.6235244597
+60 0.6219566147
+61 0.6204337402
+62 0.6187130806
+63 0.6170088968
+64 0.615589758
+65 0.6140922722
+66 0.6129240173
+67 0.6116130344
+68 0.6104895206
+69 0.6094134377
+70 0.6080808677
+71 0.6069087998
+72 0.6059376837
+73 0.6046673652
+74 0.6034098585
+75 0.6023554521
+76 0.6010549909
+77 0.5999185725
+78 0.5986988724
+79 0.5976798218
+80 0.5967699009
+81 0.5956767709
+82 0.5945202655
+83 0.5933468823
+84 0.5923352617
+85 0.5912329495
+86 0.5906017033
+87 0.5897119078
+88 0.5889154258
+89 0.5880849341
+90 0.5871459274
+91 0.5862610925
+92 0.5854995429
+93 0.5848049236
+94 0.5842107749
+95 0.583505085
+96 0.5826180548
+97 0.5818922325
+98 0.5810033948
+99 0.5805327081
+100 0.5796118053
+101 0.5787480749
+102 0.5778552983
+103 0.5771592524
+104 0.5763560519
+105 0.5755217776
+106 0.574693159
+107 0.5740325984
+108 0.5733332696
+109 0.5727932796
+110 0.5721855741
+111 0.5715813746
+112 0.5707142227
+113 0.5696757236
+114 0.5691022951
+115 0.5683067827
+116 0.5675870432
+117 0.5669024897
+118 0.566486124
+119 0.5658835071
+120 0.5654351685
+121 0.5649374002
+122 0.5641840869
+123 0.5636579743
+124 0.5630732508
+125 0.5626017666
+126 0.5618269378
+127 0.5612340032
+128 0.5607264834
+129 0.5602062292
+130 0.5595741473
+131 0.5594744956
+132 0.5588128533
+133 0.5583655792
+134 0.5576841767
+135 0.5570366592
+136 0.5563692413
+137 0.5558434603
+138 0.5553565262
+139 0.5549028143
+140 0.5542420393
+141 0.5537416549
+142 0.5533645111
+143 0.5525502359
+144 0.551964004
+145 0.5513711205
+146 0.550670328
+147 0.5500708666
+148 0.5495710178
+149 0.5491348099
+150 0.5487764949
+151 0.5482966498
+152 0.5478112718
+153 0.5471484806
+154 0.5463325223
+155 0.5459529093
+156 0.5455842734
+157 0.5452050181
+158 0.5449451093
+159 0.5445660503
+160 0.5442224699
+161 0.543859774
+162 0.5434344075
+163 0.5428666007
+164 0.5423044133
+165 0.5420020087
+166 0.5417372893
+167 0.5411407194
+168 0.5404920135
+169 0.5397785131
+170 0.5393213651
+171 0.5386490362
+172 0.5382774466
+173 0.5377320683
+174 0.5373269499
+175 0.5366982872
+176 0.5362725514
+177 0.5359097265
+178 0.5353663714
+179 0.5350468465
+180 0.5347738379
+181 0.5344498463
+182 0.5339643331
+183 0.5333646702
+184 0.5329642669
+185 0.5325614811
+186 0.5320511457
+187 0.5312388201
+188 0.5309182557
+189 0.5305264945
+190 0.5301171286
+191 0.529628972
+192 0.5293146426
+193 0.5289805915
+194 0.5284659827
+195 0.527862758
+196 0.5274831458
+197 0.5271515407
+198 0.5266917025
+199 0.5263565126
+200 0.5260532248
+201 0.5257055661
+202 0.5254257524
+203 0.524950575
+204 0.5244807342
+205 0.524249665
+206 0.5234250272
+207 0.5229129359
+208 0.5225299607
+209 0.5222031189
+210 0.5216340204
+211 0.5210763366
+212 0.520463507
+213 0.520267466
+214 0.5198975885
+215 0.5194786412
+216 0.5191099277
+217 0.5187960194
+218 0.5185900424
+219 0.5180914041
+220 0.5176345588
+221 0.5172814099
+222 0.5167825065
+223 0.5163917592
+224 0.5161350334
+225 0.5157133787
+226 0.5152784319
+227 0.5148803192
+228 0.5144992304
+229 0.5142535669
+230 0.5139649031
+231 0.513576616
+232 0.5132171213
+233 0.512763067
+234 0.5123182195
+235 0.5120334231
+236 0.5115620916
+237 0.5112360631
+238 0.5107944946
+239 0.5104228562
+240 0.5102575232
+241 0.5098562657
+242 0.5096847762
+243 0.5091203404
+244 0.5084373577
+245 0.5081469859
+246 0.5079091193
+247 0.5074818428
+248 0.5071478105
+249 0.5068570743
+250 0.5063689292
+251 0.5058932147
+252 0.505662824
+253 0.5052695033
+254 0.5048413537
+255 0.5045419313
+256 0.5042639141
+257 0.5036848245
+258 0.5030974571
+259 0.5027814387
+260 0.5022425808
+261 0.5018966871
+262 0.5014832647
+263 0.5010494012
+264 0.5006483949
+265 0.5003421562
+266 0.4999624129
+267 0.4996104834
+268 0.499196608
+269 0.4988927675
+270 0.4981819706
+271 0.4978315761
+272 0.4973900121
+273 0.4970109103
+274 0.4965620122
+275 0.4962427936
+276 0.4958221039
+277 0.4954809033
+278 0.4948766247
+279 0.4943739105
+280 0.493975467
+281 0.4935910408
+282 0.493254366
+283 0.4929261473
+284 0.4925964609
+285 0.4923808028
+286 0.4921041946
+287 0.4916744584
+288 0.4913336962
+289 0.4910190759
+290 0.4906179297
+291 0.4902202329
+292 0.489883628
+293 0.4896095455
+294 0.4893633368
+295 0.4891310576
+296 0.4887803286
+297 0.4884834707
+298 0.4880679615
+299 0.4878523011
+300 0.4873943721
+301 0.4871038251
+302 0.4867376985
+303 0.4864321061
+304 0.486210863
+305 0.4858144853
+306 0.4856499406
+307 0.485347457
+308 0.4848908087
+309 0.4845522859
+310 0.4840624415
+311 0.4837871556
+312 0.4834783178
+313 0.4831740873
+314 0.4829482632
+315 0.4826739322
+316 0.4823400925
+317 0.4819943761
+318 0.4817866029
+319 0.4815804613
+320 0.4809581003
+321 0.4807254623
+322 0.4802466482
+323 0.4798784133
+324 0.4797171462
+325 0.4793949176
+326 0.4788800195
+327 0.4786136279
+328 0.4782427334
+329 0.4778863666
+330 0.4775479356
+331 0.4773015039
+332 0.4770363259
+333 0.4768298065
+334 0.4765892571
+335 0.4764456817
+336 0.4763636442
+337 0.4760378853
+338 0.4758876647
+339 0.4756349977
+340 0.4752603973
+341 0.4750204768
+342 0.4748473115
+343 0.4745999562
+344 0.4743993595
+345 0.4737070725
+346 0.4734187079
+347 0.4732100567
+348 0.4730721577
+349 0.4728648397
+350 0.4726551113
+351 0.4723837123
+352 0.4722697334
+353 0.4720394971
+354 0.4717581939
+355 0.4715443272
+356 0.4713122025
+357 0.4709434579
+358 0.4707309341
+359 0.4705579301
+360 0.469916571
+361 0.4695748498
+362 0.4695389297
+363 0.4693263375
+364 0.4692589765
+365 0.468858875
+366 0.4685511528
+367 0.4682883236
+368 0.4680511877
+369 0.467827367
+370 0.4675352989
+371 0.4673511276
+372 0.467120547
+373 0.4668196585
+374 0.4664704386
+375 0.466300754
+376 0.4660529477
+377 0.4659205615
+378 0.4656226599
+379 0.4651948256
+380 0.4649232323
+381 0.4645354476
+382 0.4639030521
+383 0.4635351896
+384 0.4633385917
+385 0.4631669568
+386 0.4630072732
+387 0.4627835133
+388 0.4627417677
+389 0.4627061504
+390 0.4625064093
+391 0.4619443772
+392 0.4617952848
+393 0.4614354597
+394 0.4612137172
+395 0.4609780065
+396 0.460803717
+397 0.4605055516
+398 0.4604654562
+399 0.4601521805
+400 0.4601218957
+401 0.4598802793
+402 0.4596294913
+403 0.459359852
+404 0.4593234097
+405 0.4590768781
+406 0.458797219
+407 0.4585924348
+408 0.4585649874
+409 0.4585363276
+410 0.4579354313
+411 0.4575848613
+412 0.4574072665
+413 0.4572256847
+414 0.4569922307
+415 0.4564843783
+416 0.4562989952
+417 0.4557830648
+418 0.4552632395
+419 0.45506771
+420 0.4548371775
+421 0.4543536083
+422 0.4539135342
+423 0.4536574805
+424 0.4532930869
+425 0.4530580282
+426 0.4528470071
+427 0.4525708397
+428 0.4524322649
+429 0.4521065026
+430 0.4519013222
+431 0.4518403387
+432 0.4516184857
+433 0.4513714288
+434 0.451037779
+435 0.4507709996
+436 0.4505847307
+437 0.4503895992
+438 0.4499681477
+439 0.4496796353
+440 0.4494430951
+441 0.4491757112
+442 0.4489244193
+443 0.4485276391
+444 0.4481718053
+445 0.4479506854
+446 0.4476990522
+447 0.4471948591
+448 0.4470853092
+449 0.446865776
+450 0.4465824143
+451 0.4461086564
+452 0.4460486522
+453 0.4458924182
+454 0.4457369368
+455 0.4457090735
+456 0.4453923753
+457 0.4453669142
+458 0.4450574113
+459 0.4448489841
+460 0.4446117109
+461 0.4444696636
+462 0.444285799
+463 0.4441396144
+464 0.4437465091
+465 0.4435487801
+466 0.4433583675
+467 0.4431235913
+468 0.4428925062
+469 0.4425307044
+470 0.4424258336
+471 0.4424000888
+472 0.4422433144
+473 0.4420168935
+474 0.4417496517
+475 0.4414042041
+476 0.4411749674
+477 0.4407943905
+478 0.440671799
+479 0.4403708967
+480 0.4401620051
+481 0.4399914723
+482 0.4396355813
+483 0.4396078413
+484 0.4394233159
+485 0.4392648597
+486 0.4390081285
+487 0.4389768991
+488 0.4386848902
+489 0.4383633005
+490 0.4382373497
+491 0.4379751734
+492 0.437740467
+493 0.4377072609
+494 0.437475001
+495 0.4374511985
+496 0.4372524578
+497 0.4370094577
+498 0.4367382154
+499 0.4365629412
diff --git a/catboost_info/time_left.tsv b/catboost_info/time_left.tsv
index ba52c770a85e2941f9fa7d22f3a3c4e8e0640bf0..d3999e32dc3cd8fd58018fd03ffb686a461934bd 100644
--- a/catboost_info/time_left.tsv
+++ b/catboost_info/time_left.tsv
@@ -1,501 +1,501 @@
iter Passed Remaining
-0 95 47414
-1 127 31682
-2 158 26227
-3 189 23500
-4 221 21921
-5 254 20917
-6 287 20223
-7 320 19713
-8 355 19373
-9 389 19088
-10 424 18864
-11 458 18649
-12 492 18457
-13 526 18264
-14 560 18120
-15 594 17979
-16 628 17865
-17 665 17809
-18 700 17726
-19 736 17665
-20 774 17668
-21 809 17592
-22 846 17560
-23 882 17502
-24 920 17485
-25 957 17451
-26 994 17414
-27 1031 17387
-28 1068 17352
-29 1106 17336
-30 1146 17344
-31 1184 17316
-32 1220 17270
-33 1259 17259
-34 1296 17222
-35 1334 17197
-36 1374 17204
-37 1413 17183
-38 1455 17199
-39 1494 17181
-40 1532 17156
-41 1572 17150
-42 1610 17112
-43 1648 17087
-44 1688 17067
-45 1726 17040
-46 1766 17021
-47 1803 16985
-48 1841 16950
-49 1880 16923
-50 1918 16888
-51 1958 16872
-52 1999 16861
-53 2038 16835
-54 2076 16798
-55 2114 16764
-56 2154 16740
-57 2192 16710
-58 2231 16680
-59 2270 16650
-60 2308 16616
-61 2346 16578
-62 2385 16546
-63 2424 16516
-64 2465 16501
-65 2503 16464
-66 2542 16431
-67 2582 16404
-68 2619 16362
-69 2658 16329
-70 2696 16292
-71 2736 16264
-72 2774 16231
-73 2812 16189
-74 2850 16151
-75 2889 16119
-76 2929 16092
-77 2967 16057
-78 3008 16031
-79 3052 16027
-80 3093 16001
-81 3133 15971
-82 3174 15946
-83 3214 15921
-84 3253 15885
-85 3292 15851
-86 3330 15811
-87 3368 15769
-88 3407 15736
-89 3445 15696
-90 3485 15664
-91 3523 15626
-92 3561 15584
-93 3600 15550
-94 3637 15509
-95 3678 15478
-96 3717 15443
-97 3755 15403
-98 3793 15364
-99 3830 15320
-100 3869 15285
-101 3908 15252
-102 3947 15216
-103 3987 15183
-104 4026 15148
-105 4066 15113
-106 4104 15075
-107 4142 15035
-108 4181 14999
-109 4218 14957
-110 4256 14917
-111 4295 14882
-112 4334 14843
-113 4374 14812
-114 4415 14783
-115 4454 14745
-116 4494 14712
-117 4533 14676
-118 4570 14634
-119 4611 14602
-120 4648 14558
-121 4688 14526
-122 4727 14491
-123 4764 14448
-124 4803 14410
-125 4841 14370
-126 4881 14338
-127 4922 14304
-128 4960 14267
-129 5000 14231
-130 5038 14193
-131 5077 14154
-132 5116 14118
-133 5154 14077
-134 5194 14045
-135 5234 14008
-136 5273 13974
-137 5313 13939
-138 5351 13897
-139 5391 13863
-140 5430 13826
-141 5468 13787
-142 5508 13752
-143 5546 13712
-144 5585 13673
-145 5623 13635
-146 5662 13598
-147 5702 13562
-148 5741 13525
-149 5779 13485
-150 5818 13448
-151 5858 13412
-152 5897 13374
-153 5901 13258
-154 5940 13222
-155 5978 13184
-156 6017 13145
-157 6054 13105
-158 6094 13070
-159 6132 13031
-160 6170 12992
-161 6208 12953
-162 6247 12915
-163 6284 12875
-164 6323 12838
-165 6362 12801
-166 6402 12766
-167 6441 12729
-168 6479 12690
-169 6516 12649
-170 6554 12610
-171 6591 12569
-172 6629 12530
-173 6667 12491
-174 6705 12452
-175 6743 12414
-176 6782 12376
-177 6821 12339
-178 6859 12301
-179 6896 12260
-180 6936 12224
-181 6974 12186
-182 7014 12150
-183 7052 12112
-184 7090 12072
-185 7131 12039
-186 7169 12000
-187 7208 11962
-188 7245 11922
-189 7283 11883
-190 7322 11845
-191 7359 11805
-192 7395 11764
-193 7435 11728
-194 7473 11688
-195 7511 11650
-196 7549 11611
-197 7586 11571
-198 7624 11532
-199 7662 11494
-200 7699 11453
-201 7740 11419
-202 7779 11381
-203 7817 11342
-204 7854 11303
-205 7892 11263
-206 7931 11227
-207 7970 11189
-208 8012 11156
-209 8053 11121
-210 8091 11082
-211 8131 11046
-212 8169 11008
-213 8207 10969
-214 8247 10933
-215 8285 10893
-216 8325 10857
-217 8364 10820
-218 8402 10781
-219 8443 10746
-220 8484 10711
-221 8522 10672
-222 8562 10636
-223 8599 10596
-224 8637 10557
-225 8675 10517
-226 8713 10478
-227 8752 10441
-228 8791 10403
-229 8830 10365
-230 8867 10326
-231 8905 10286
-232 8943 10248
-233 8984 10213
-234 9027 10180
-235 9068 10144
-236 9105 10104
-237 9144 10066
-238 9183 10028
-239 9223 9991
-240 9262 9953
-241 9302 9917
-242 9340 9878
-243 9378 9839
-244 9418 9802
-245 9455 9763
-246 9494 9725
-247 9533 9687
-248 9569 9646
-249 9606 9606
-250 9645 9568
-251 9683 9529
-252 9721 9490
-253 9760 9452
-254 9798 9413
-255 9836 9375
-256 9876 9338
-257 9915 9300
-258 9954 9262
-259 9991 9223
-260 10033 9187
-261 10071 9148
-262 10110 9111
-263 10151 9074
-264 10188 9034
-265 10226 8995
-266 10264 8957
-267 10302 8918
-268 10341 8880
-269 10379 8842
-270 10419 8804
-271 10458 8766
-272 10495 8727
-273 10533 8688
-274 10571 8649
-275 10609 8610
-276 10647 8571
-277 10687 8534
-278 10725 8496
-279 10765 8458
-280 10803 8419
-281 10842 8381
-282 10881 8343
-283 10919 8304
-284 10957 8265
-285 10994 8226
-286 11033 8188
-287 11071 8149
-288 11109 8111
-289 11147 8072
-290 11185 8033
-291 11223 7994
-292 11262 7956
-293 11299 7917
-294 11336 7878
-295 11374 7839
-296 11414 7801
-297 11452 7763
-298 11491 7724
-299 11531 7687
-300 11570 7649
-301 11609 7611
-302 11646 7571
-303 11685 7533
-304 11723 7495
-305 11761 7456
-306 11800 7418
-307 11838 7379
-308 11878 7342
-309 11917 7304
-310 11956 7266
-311 11994 7227
-312 12032 7188
-313 12069 7149
-314 12106 7110
-315 12145 7072
-316 12183 7033
-317 12221 6994
-318 12260 6956
-319 12299 6918
-320 12338 6880
-321 12375 6841
-322 12415 6803
-323 12454 6765
-324 12492 6726
-325 12528 6687
-326 12566 6648
-327 12607 6611
-328 12646 6573
-329 12684 6534
-330 12721 6495
-331 12758 6456
-332 12798 6418
-333 12836 6379
-334 12873 6340
-335 12912 6302
-336 12950 6263
-337 12990 6226
-338 13029 6188
-339 13066 6148
-340 13104 6110
-341 13144 6072
-342 13181 6033
-343 13221 5996
-344 13259 5957
-345 13297 5918
-346 13336 5880
-347 13373 5841
-348 13414 5803
-349 13452 5765
-350 13490 5726
-351 13527 5687
-352 13564 5648
-353 13602 5610
-354 13639 5571
-355 13677 5532
-356 13716 5494
-357 13754 5455
-358 13791 5416
-359 13830 5378
-360 13867 5339
-361 13905 5300
-362 13944 5262
-363 13983 5224
-364 14021 5186
-365 14058 5147
-366 14096 5108
-367 14135 5070
-368 14172 5031
-369 14211 4993
-370 14250 4955
-371 14289 4916
-372 14329 4878
-373 14366 4839
-374 14404 4801
-375 14442 4762
-376 14480 4724
-377 14520 4686
-378 14556 4647
-379 14594 4608
-380 14634 4570
-381 14671 4531
-382 14713 4494
-383 14750 4455
-384 14788 4417
-385 14826 4378
-386 14862 4339
-387 14902 4301
-388 14943 4264
-389 14983 4226
-390 15026 4188
-391 15065 4150
-392 15103 4112
-393 15141 4073
-394 15178 4034
-395 15215 3996
-396 15254 3957
-397 15293 3919
-398 15333 3881
-399 15370 3842
-400 15410 3804
-401 15447 3765
-402 15484 3727
-403 15523 3688
-404 15561 3650
-405 15600 3612
-406 15639 3573
-407 15675 3534
-408 15713 3496
-409 15752 3457
-410 15790 3419
-411 15826 3380
-412 15865 3342
-413 15902 3303
-414 15942 3265
-415 15980 3226
-416 16018 3188
-417 16055 3149
-418 16093 3111
-419 16134 3073
-420 16171 3034
-421 16211 2996
-422 16248 2957
-423 16287 2919
-424 16326 2881
-425 16365 2842
-426 16403 2804
-427 16443 2766
-428 16486 2728
-429 16531 2691
-430 16573 2653
-431 16617 2615
-432 16662 2578
-433 16703 2540
-434 16749 2502
-435 16798 2465
-436 16846 2428
-437 16885 2390
-438 16925 2351
-439 16964 2313
-440 17012 2275
-441 17051 2237
-442 17090 2198
-443 17128 2160
-444 17165 2121
-445 17203 2082
-446 17240 2044
-447 17277 2005
-448 17316 1966
-449 17353 1928
-450 17391 1889
-451 17429 1850
-452 17466 1812
-453 17505 1773
-454 17544 1735
-455 17582 1696
-456 17618 1657
-457 17657 1619
-458 17696 1580
-459 17733 1542
-460 17772 1503
-461 17810 1464
-462 17849 1426
-463 17887 1387
-464 17925 1349
-465 17964 1310
-466 18002 1272
-467 18038 1233
-468 18075 1194
-469 18113 1156
-470 18150 1117
-471 18188 1078
-472 18226 1040
-473 18267 1002
-474 18310 963
-475 18351 925
-476 18389 886
-477 18428 848
-478 18465 809
-479 18504 771
-480 18541 732
-481 18580 693
-482 18620 655
-483 18659 616
-484 18700 578
-485 18738 539
-486 18777 501
-487 18815 462
-488 18856 424
-489 18903 385
-490 18947 347
-491 18992 308
-492 19034 270
-493 19076 231
-494 19122 193
-495 19162 154
-496 19204 115
-497 19246 77
-498 19288 38
-499 19332 0
+0 82 41316
+1 113 28180
+2 143 23755
+3 174 21638
+4 206 20425
+5 238 19607
+6 271 19154
+7 305 18758
+8 339 18506
+9 373 18293
+10 408 18155
+11 443 18044
+12 479 17962
+13 516 17921
+14 551 17837
+15 590 17864
+16 628 17842
+17 669 17922
+18 710 17994
+19 748 17975
+20 789 18011
+21 830 18049
+22 873 18123
+23 916 18181
+24 955 18145
+25 993 18112
+26 1032 18079
+27 1071 18064
+28 1109 18027
+29 1147 17979
+30 1187 17967
+31 1229 17981
+32 1268 17945
+33 1308 17928
+34 1348 17910
+35 1387 17877
+36 1427 17867
+37 1465 17814
+38 1504 17781
+39 1544 17762
+40 1583 17731
+41 1623 17702
+42 1663 17678
+43 1700 17628
+44 1740 17599
+45 1778 17548
+46 1817 17518
+47 1857 17488
+48 1894 17440
+49 1936 17431
+50 1977 17412
+51 2020 17405
+52 2062 17394
+53 2104 17383
+54 2144 17352
+55 2186 17337
+56 2228 17319
+57 2267 17276
+58 2310 17271
+59 2354 17267
+60 2395 17239
+61 2436 17213
+62 2475 17174
+63 2516 17144
+64 2558 17121
+65 2599 17091
+66 2640 17066
+67 2681 17035
+68 2722 17006
+69 2766 16994
+70 2814 17008
+71 2861 17008
+72 2904 16987
+73 2949 16978
+74 2995 16972
+75 3042 16972
+76 3085 16947
+77 3127 16920
+78 3167 16881
+79 3212 16866
+80 3255 16839
+81 3295 16800
+82 3337 16767
+83 3379 16735
+84 3420 16700
+85 3462 16668
+86 3503 16631
+87 3545 16599
+88 3587 16567
+89 3629 16535
+90 3670 16497
+91 3716 16482
+92 3760 16455
+93 3800 16413
+94 3839 16369
+95 3882 16338
+96 3924 16304
+97 3967 16275
+98 4013 16256
+99 4057 16228
+100 4098 16190
+101 4140 16154
+102 4181 16117
+103 4223 16081
+104 4266 16050
+105 4307 16010
+106 4350 15980
+107 4389 15932
+108 4431 15897
+109 4471 15854
+110 4511 15809
+111 4551 15767
+112 4588 15715
+113 4628 15672
+114 4668 15628
+115 4706 15580
+116 4747 15540
+117 4786 15496
+118 4825 15450
+119 4865 15407
+120 4904 15362
+121 4946 15326
+122 4988 15289
+123 5027 15245
+124 5069 15207
+125 5109 15167
+126 5150 15128
+127 5191 15086
+128 5229 15041
+129 5268 14995
+130 5307 14949
+131 5312 14809
+132 5352 14770
+133 5391 14726
+134 5432 14687
+135 5473 14650
+136 5514 14610
+137 5555 14573
+138 5596 14533
+139 5635 14492
+140 5677 14455
+141 5716 14410
+142 5757 14373
+143 5797 14331
+144 5835 14285
+145 5876 14247
+146 5916 14207
+147 5956 14166
+148 5995 14124
+149 6034 14081
+150 6075 14042
+151 6116 14002
+152 6156 13962
+153 6198 13926
+154 6238 13885
+155 6281 13851
+156 6322 13811
+157 6363 13774
+158 6400 13726
+159 6440 13685
+160 6481 13646
+161 6520 13605
+162 6562 13566
+163 6603 13528
+164 6644 13491
+165 6684 13450
+166 6723 13406
+167 6762 13364
+168 6801 13322
+169 6842 13282
+170 6882 13242
+171 6921 13198
+172 6961 13157
+173 7000 13116
+174 7042 13078
+175 7086 13045
+176 7125 13003
+177 7167 12965
+178 7208 12926
+179 7248 12887
+180 7289 12846
+181 7329 12806
+182 7372 12770
+183 7412 12729
+184 7452 12689
+185 7493 12650
+186 7533 12610
+187 7576 12573
+188 7613 12528
+189 7653 12487
+190 7694 12448
+191 7731 12402
+192 7771 12361
+193 7809 12318
+194 7850 12279
+195 7893 12242
+196 7933 12202
+197 7975 12165
+198 8016 12125
+199 8055 12083
+200 8097 12046
+201 8138 12006
+202 8179 11966
+203 8220 11928
+204 8260 11886
+205 8300 11845
+206 8345 11812
+207 8388 11776
+208 8427 11734
+209 8468 11693
+210 8507 11652
+211 8548 11613
+212 8588 11572
+213 8626 11529
+214 8668 11490
+215 8705 11446
+216 8746 11406
+217 8786 11365
+218 8823 11322
+219 8863 11280
+220 8900 11235
+221 8938 11193
+222 8979 11154
+223 9017 11111
+224 9055 11068
+225 9096 11028
+226 9135 10986
+227 9174 10945
+228 9213 10902
+229 9251 10860
+230 9291 10820
+231 9329 10776
+232 9373 10741
+233 9415 10703
+234 9457 10664
+235 9499 10626
+236 9541 10587
+237 9582 10548
+238 9620 10506
+239 9660 10465
+240 9697 10422
+241 9735 10379
+242 9775 10338
+243 9813 10295
+244 9853 10255
+245 9893 10215
+246 9934 10175
+247 9973 10134
+248 10010 10091
+249 10049 10049
+250 10090 10009
+251 10130 9969
+252 10169 9928
+253 10208 9886
+254 10248 9846
+255 10291 9809
+256 10331 9768
+257 10372 9729
+258 10411 9687
+259 10449 9645
+260 10491 9606
+261 10529 9564
+262 10569 9524
+263 10611 9486
+264 10649 9444
+265 10689 9403
+266 10727 9361
+267 10767 9320
+268 10808 9281
+269 10847 9240
+270 10890 9202
+271 10930 9162
+272 10975 9126
+273 11017 9087
+274 11058 9047
+275 11099 9008
+276 11140 8968
+277 11179 8927
+278 11239 8902
+279 11336 8907
+280 11381 8870
+281 11425 8832
+282 11465 8791
+283 11508 8752
+284 11548 8711
+285 11588 8671
+286 11628 8630
+287 11668 8589
+288 11708 8548
+289 11747 8506
+290 11787 8466
+291 11828 8425
+292 11868 8384
+293 11908 8343
+294 11946 8301
+295 11985 8260
+296 12024 8218
+297 12064 8177
+298 12105 8137
+299 12144 8096
+300 12184 8055
+301 12225 8015
+302 12266 7974
+303 12307 7935
+304 12347 7894
+305 12388 7854
+306 12429 7813
+307 12467 7772
+308 12508 7732
+309 12546 7689
+310 12585 7648
+311 12625 7607
+312 12663 7565
+313 12703 7525
+314 12741 7483
+315 12780 7441
+316 12818 7400
+317 12856 7358
+318 12894 7316
+319 12931 7274
+320 12971 7233
+321 13008 7191
+322 13048 7150
+323 13085 7108
+324 13122 7065
+325 13162 7025
+326 13201 6984
+327 13239 6942
+328 13278 6901
+329 13317 6860
+330 13355 6819
+331 13394 6777
+332 13432 6736
+333 13471 6695
+334 13509 6653
+335 13546 6612
+336 13585 6570
+337 13623 6529
+338 13661 6488
+339 13699 6447
+340 13739 6406
+341 13779 6365
+342 13819 6325
+343 13857 6284
+344 13894 6242
+345 13935 6202
+346 13975 6162
+347 14015 6121
+348 14052 6079
+349 14090 6038
+350 14127 5997
+351 14165 5955
+352 14203 5914
+353 14239 5872
+354 14279 5832
+355 14319 5792
+356 14356 5750
+357 14396 5710
+358 14436 5670
+359 14473 5628
+360 14515 5588
+361 14552 5547
+362 14591 5506
+363 14630 5466
+364 14668 5425
+365 14709 5385
+366 14746 5344
+367 14785 5303
+368 14824 5262
+369 14862 5222
+370 14901 5181
+371 14939 5140
+372 14977 5099
+373 15015 5058
+374 15053 5017
+375 15091 4976
+376 15128 4935
+377 15166 4895
+378 15207 4855
+379 15245 4814
+380 15285 4774
+381 15325 4733
+382 15364 4693
+383 15402 4652
+384 15442 4612
+385 15481 4572
+386 15521 4532
+387 15560 4491
+388 15598 4451
+389 15638 4410
+390 15675 4369
+391 15716 4330
+392 15755 4289
+393 15795 4249
+394 15834 4209
+395 15873 4168
+396 15913 4128
+397 15952 4088
+398 15992 4048
+399 16032 4008
+400 16070 3967
+401 16109 3927
+402 16148 3886
+403 16190 3847
+404 16228 3806
+405 16268 3766
+406 16308 3726
+407 16346 3685
+408 16386 3645
+409 16425 3605
+410 16464 3565
+411 16504 3525
+412 16542 3484
+413 16583 3444
+414 16621 3404
+415 16661 3364
+416 16701 3324
+417 16741 3284
+418 16782 3244
+419 16820 3203
+420 16860 3163
+421 16901 3124
+422 16940 3083
+423 16980 3043
+424 17019 3003
+425 17055 2962
+426 17091 2922
+427 17129 2881
+428 17166 2841
+429 17205 2800
+430 17241 2760
+431 17278 2719
+432 17319 2679
+433 17359 2639
+434 17399 2599
+435 17439 2559
+436 17479 2519
+437 17517 2479
+438 17556 2439
+439 17596 2399
+440 17633 2359
+441 17676 2319
+442 17716 2279
+443 17754 2239
+444 17791 2198
+445 17829 2158
+446 17867 2118
+447 17906 2078
+448 17942 2037
+449 17982 1998
+450 18020 1957
+451 18059 1917
+452 18097 1877
+453 18135 1837
+454 18175 1797
+455 18214 1757
+456 18253 1717
+457 18291 1677
+458 18331 1637
+459 18371 1597
+460 18411 1557
+461 18447 1517
+462 18486 1477
+463 18524 1437
+464 18564 1397
+465 18603 1357
+466 18640 1317
+467 18680 1277
+468 18719 1237
+469 18757 1197
+470 18797 1157
+471 18835 1117
+472 18876 1077
+473 18918 1037
+474 18959 997
+475 18999 957
+476 19037 917
+477 19076 877
+478 19117 838
+479 19157 798
+480 19194 758
+481 19231 718
+482 19270 678
+483 19308 638
+484 19345 598
+485 19383 558
+486 19421 518
+487 19461 478
+488 19499 438
+489 19537 398
+490 19576 358
+491 19616 318
+492 19655 279
+493 19693 239
+494 19733 199
+495 19772 159
+496 19812 119
+497 19849 79
+498 19886 39
+499 19925 0
diff --git a/confision_matrix_kmeans_v2.png b/confision_matrix_kmeans_v2.png
new file mode 100644
index 0000000000000000000000000000000000000000..4b76d8182c0be2552b9335b39fc3924d42f7c9f0
Binary files /dev/null and b/confision_matrix_kmeans_v2.png differ
diff --git a/figures/kmeans_clusters_1.jpg b/figures/kmeans_clusters_1.jpg
index 5a0d98d5513283bdfe1dd06dcd37a4ae26ad2b5f..725b1188b50976b9389a2afcdd57c03042861841 100755
Binary files a/figures/kmeans_clusters_1.jpg and b/figures/kmeans_clusters_1.jpg differ
diff --git a/value_counts_smot_type_of_area.jpg b/value_counts_smot_type_of_area.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..39f77346a59affeefa80ed5d96133e4208520154
Binary files /dev/null and b/value_counts_smot_type_of_area.jpg differ