(this["webpackJsonpmy-portfolio"]=this["webpackJsonpmy-portfolio"]||[]).push([[0],{141:function(t){t.exports=JSON.parse('{"cells":[{"cell_type":"markdown","metadata":{"id":"wfFoTzio5NSv"},"source":["# Datamining on Stroke Dataset\\n"]},{"cell_type":"markdown","metadata":{"id":"0ZjCN0Ci6UyK"},"source":["This dataset contains information about 5110 people and their health in regards to their stroke history. There are 11 features (including the id) in the dataset which we will use to create models to predict whether or not someone is prone to stroke. These features include:\\n","\\n","\\n","*   Gender\\n","*   Age\\n","*   Hypertension (known as high blood pressure, or a blood pressure above 140/90)\\n","*   Heart Disease\\n","*   Marital History (if the subject is or was ever married)\\n","*   Type of Work (Government, Private, Self-Employed, Never Worked, or Children)\\n","*   Residency (Rural or Urban)\\n","*   Average Glucose Level\\n","*   BMI\\n","*   Smoking History (Smokes, Formally Smoked, Never Smoked or Unknown)"]},{"cell_type":"markdown","metadata":{"id":"3FzTzlzQ6Wjm"},"source":["We wanted to be able to accurately predict whether someone is prone to having a stroke in the future based on these features. If we were able to accurately predict signs of a stroke, it could become extremely valuable to the medical field. Modeling this data could help hospitals and doctors detect those prone to stroke and begin treating or warning their patients of possible strokes in the future. However, we must create a model first, and that requires us to preprocess our data. "]},{"cell_type":"code","execution_count":1,"metadata":{"id":"CvAO1keJ4-rp"},"outputs":[],"source":["# Standard imports\\n","import warnings\\n","import pandas as pd\\n","import numpy as np\\n","import sklearn as sk\\n","import matplotlib.pyplot as plt\\n","import time"]},{"cell_type":"markdown","metadata":{"id":"qYWusnxw76JJ"},"source":["## Data Preprocessing (Exploration and Cleaning)"]},{"cell_type":"code","execution_count":2,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"dV3GFepj8IyL","outputId":"c122875d-1bbe-41ab-8638-a127586a6d6a"},"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>id</th>\\n","      <th>gender</th>\\n","      <th>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>ever_married</th>\\n","      <th>work_type</th>\\n","      <th>Residence_type</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>smoking_status</th>\\n","      <th>stroke</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>9046</td>\\n","      <td>Male</td>\\n","      <td>67.0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>Yes</td>\\n","      <td>Private</td>\\n","      <td>Urban</td>\\n","      <td>228.69</td>\\n","      <td>36.6</td>\\n","      <td>formerly smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>51676</td>\\n","      <td>Female</td>\\n","      <td>61.0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>Yes</td>\\n","      <td>Self-employed</td>\\n","      <td>Rural</td>\\n","      <td>202.21</td>\\n","      <td>NaN</td>\\n","      <td>never smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>31112</td>\\n","      <td>Male</td>\\n","      <td>80.0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>Yes</td>\\n","      <td>Private</td>\\n","      <td>Rural</td>\\n","      <td>105.92</td>\\n","      <td>32.5</td>\\n","      <td>never smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>60182</td>\\n","      <td>Female</td>\\n","      <td>49.0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>Yes</td>\\n","      <td>Private</td>\\n","      <td>Urban</td>\\n","      <td>171.23</td>\\n","      <td>34.4</td>\\n","      <td>smokes</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>1665</td>\\n","      <td>Female</td>\\n","      <td>79.0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>Yes</td>\\n","      <td>Self-employed</td>\\n","      <td>Rural</td>\\n","      <td>174.12</td>\\n","      <td>24.0</td>\\n","      <td>never smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","</div>"],"text/plain":["      id  gender   age  hypertension  heart_disease ever_married  \\\\\\n","0   9046    Male  67.0             0              1          Yes   \\n","1  51676  Female  61.0             0              0          Yes   \\n","2  31112    Male  80.0             0              1          Yes   \\n","3  60182  Female  49.0             0              0          Yes   \\n","4   1665  Female  79.0             1              0          Yes   \\n","\\n","       work_type Residence_type  avg_glucose_level   bmi   smoking_status  \\\\\\n","0        Private          Urban             228.69  36.6  formerly smoked   \\n","1  Self-employed          Rural             202.21   NaN     never smoked   \\n","2        Private          Rural             105.92  32.5     never smoked   \\n","3        Private          Urban             171.23  34.4           smokes   \\n","4  Self-employed          Rural             174.12  24.0     never smoked   \\n","\\n","   stroke  \\n","0       1  \\n","1       1  \\n","2       1  \\n","3       1  \\n","4       1  "]},"execution_count":2,"metadata":{},"output_type":"execute_result"}],"source":["data = pd.read_csv(\'healthcare-dataset-stroke-data.csv\')\\n","data.head()"]},{"cell_type":"markdown","metadata":{"id":"5yDfYnJG8_Ls"},"source":["Here, we want to remove the id column of our data, since it provides us no use to the model we want to create."]},{"cell_type":"code","execution_count":3,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"7kSNR_AH8dWI","outputId":"3f632500-a49b-44dc-d995-f0d770d11684"},"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>gender</th>\\n","      <th>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>ever_married</th>\\n","      <th>work_type</th>\\n","      <th>Residence_type</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>smoking_status</th>\\n","      <th>stroke</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>Male</td>\\n","      <td>67.0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>Yes</td>\\n","      <td>Private</td>\\n","      <td>Urban</td>\\n","      <td>228.69</td>\\n","      <td>36.6</td>\\n","      <td>formerly smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>Female</td>\\n","      <td>61.0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>Yes</td>\\n","      <td>Self-employed</td>\\n","      <td>Rural</td>\\n","      <td>202.21</td>\\n","      <td>NaN</td>\\n","      <td>never smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>Male</td>\\n","      <td>80.0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>Yes</td>\\n","      <td>Private</td>\\n","      <td>Rural</td>\\n","      <td>105.92</td>\\n","      <td>32.5</td>\\n","      <td>never smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>Female</td>\\n","      <td>49.0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>Yes</td>\\n","      <td>Private</td>\\n","      <td>Urban</td>\\n","      <td>171.23</td>\\n","      <td>34.4</td>\\n","      <td>smokes</td>\\n","      <td>1</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>Female</td>\\n","      <td>79.0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>Yes</td>\\n","      <td>Self-employed</td>\\n","      <td>Rural</td>\\n","      <td>174.12</td>\\n","      <td>24.0</td>\\n","      <td>never smoked</td>\\n","      <td>1</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","</div>"],"text/plain":["   gender   age  hypertension  heart_disease ever_married      work_type  \\\\\\n","0    Male  67.0             0              1          Yes        Private   \\n","1  Female  61.0             0              0          Yes  Self-employed   \\n","2    Male  80.0             0              1          Yes        Private   \\n","3  Female  49.0             0              0          Yes        Private   \\n","4  Female  79.0             1              0          Yes  Self-employed   \\n","\\n","  Residence_type  avg_glucose_level   bmi   smoking_status  stroke  \\n","0          Urban             228.69  36.6  formerly smoked       1  \\n","1          Rural             202.21   NaN     never smoked       1  \\n","2          Rural             105.92  32.5     never smoked       1  \\n","3          Urban             171.23  34.4           smokes       1  \\n","4          Rural             174.12  24.0     never smoked       1  "]},"execution_count":3,"metadata":{},"output_type":"execute_result"}],"source":["data.drop(\'id\', axis=1, inplace=True)\\n","data.head()"]},{"cell_type":"markdown","metadata":{"id":"f0uH003u9ExY"},"source":["Now that the id column is removed, we want to one-hot encode the data so it can be used in the models."]},{"cell_type":"markdown","metadata":{"id":"KYQITudOOIG9"},"source":["**One-Hot Encoding**\\n","\\n","\\n","Since the non-binary categorical features are hard to extract data from, we use one-hot encoding to simplify this. These features include gender, marriage, work, residency, and smoking history."]},{"cell_type":"code","execution_count":4,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":261},"id":"b786Mh_kOh-0","outputId":"e98c2bf7-7aef-4f73-ee20-b0a064b72797"},"outputs":[{"name":"stdout","output_type":"stream","text":["(5110, 22)\\n"]},{"data":{"text/html":["<div>\\n","<style scoped>\\n","    .dataframe tbody tr th:only-of-type {\\n","        vertical-align: middle;\\n","    }\\n","\\n","    .dataframe tbody tr th {\\n","        vertical-align: top;\\n","    }\\n","\\n","    .dataframe thead th {\\n","        text-align: right;\\n","    }\\n","</style>\\n","<table border=\\"1\\" class=\\"dataframe\\">\\n","  <thead>\\n","    <tr style=\\"text-align: right;\\">\\n","      <th></th>\\n","      <th>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Female</th>\\n","      <th>gender_Other</th>\\n","      <th>ever_married_Yes</th>\\n","      <th>...</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_children</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>Residence_type_Rural</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>67.0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>228.69</td>\\n","      <td>36.6</td>\\n","      <td>1</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>...</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>61.0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>202.21</td>\\n","      <td>NaN</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>...</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>80.0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>105.92</td>\\n","      <td>32.5</td>\\n","      <td>1</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>...</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>49.0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>171.23</td>\\n","      <td>34.4</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>...</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>79.0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>174.12</td>\\n","      <td>24.0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>...</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>1</td>\\n","      <td>0</td>\\n","      <td>0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","<p>5 rows \xd7 22 columns</p>\\n","</div>"],"text/plain":["    age  hypertension  heart_disease  avg_glucose_level   bmi  stroke  \\\\\\n","0  67.0             0              1             228.69  36.6       1   \\n","1  61.0             0              0             202.21   NaN       1   \\n","2  80.0             0              1             105.92  32.5       1   \\n","3  49.0             0              0             171.23  34.4       1   \\n","4  79.0             1              0             174.12  24.0       1   \\n","\\n","   gender_Male  gender_Female  gender_Other  ever_married_Yes  ...  \\\\\\n","0            1              0             0                 1  ...   \\n","1            0              1             0                 1  ...   \\n","2            1              0             0                 1  ...   \\n","3            0              1             0                 1  ...   \\n","4            0              1             0                 1  ...   \\n","\\n","   work_type_Self-employed  work_type_Govt_job  work_type_children  \\\\\\n","0                        0                   0                   0   \\n","1                        1                   0                   0   \\n","2                        0                   0                   0   \\n","3                        0                   0                   0   \\n","4                        1                   0                   0   \\n","\\n","   work_type_Never_worked  Residence_type_Urban  Residence_type_Rural  \\\\\\n","0                       0                     1                     0   \\n","1                       0                     0                     1   \\n","2                       0                     0                     1   \\n","3                       0                     1                     0   \\n","4                       0                     0                     1   \\n","\\n","   smoking_status_formerly smoked  smoking_status_never smoked  \\\\\\n","0                               1                            0   \\n","1                               0                            1   \\n","2                               0                            1   \\n","3                               0                            0   \\n","4                               0                            1   \\n","\\n","   smoking_status_smokes  smoking_status_Unknown  \\n","0                      0                       0  \\n","1                      0                       0  \\n","2                      0                       0  \\n","3                      1                       0  \\n","4                      0                       0  \\n","\\n","[5 rows x 22 columns]"]},"execution_count":4,"metadata":{},"output_type":"execute_result"}],"source":["hot_data = data.copy()\\n","categorical_columns = [\\"gender\\", \\"ever_married\\", \\"work_type\\", \\"Residence_type\\", \\"smoking_status\\"]\\n","\\n","for c in categorical_columns:\\n","    vals = hot_data[c].unique()\\n","    for v in vals:\\n","        hot_data[f\\"{c}_{v}\\"] = hot_data[c].apply(lambda c_val: 1 if c_val == v else 0)\\n","    hot_data.drop(c, axis=1, inplace=True)\\n","print(hot_data.shape)\\n","hot_data.head()"]},{"cell_type":"markdown","metadata":{"id":"Ljo_ce-uDDs4"},"source":["**Data Cleaning (missing value detection)**\\n","\\n","\\n","Now we want to check for any missing values in the data."]},{"cell_type":"code","execution_count":5,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ZHcJ6EpT9C1g","outputId":"a9ad79df-f902-4528-91fb-4ed10b8e56ff"},"outputs":[{"name":"stdout","output_type":"stream","text":["# of missing gender values:  0\\n","# of missing age values:  0\\n","# of missing hypertension values:  0\\n","# of missing heart_disease values:  0\\n","# of missing ever_married values:  0\\n","# of missing work_type values:  0\\n","# of missing Residence_type values:  0\\n","# of missing avg_glucose_level values:  0\\n","# of missing bmi values:  201\\n","# of missing smoking_status values:  0\\n","# of missing stroke values:  0\\n"]}],"source":["# Check for null values in every column\\n","missing_gender = data[\'gender\'].isnull().sum()\\n","print(\\"# of missing gender values: \\", missing_gender)\\n","\\n","missing_age = data[\'age\'].isnull().sum()\\n","print(\\"# of missing age values: \\", missing_age)\\n","\\n","missing_hypertension = data[\'hypertension\'].isnull().sum()\\n","print(\\"# of missing hypertension values: \\", missing_hypertension)\\n","\\n","missing_heart_disease = data[\'heart_disease\'].isnull().sum()\\n","print(\\"# of missing heart_disease values: \\", missing_heart_disease)\\n","\\n","missing_ever_married = data[\'ever_married\'].isnull().sum()\\n","print(\\"# of missing ever_married values: \\", missing_ever_married)\\n","\\n","missing_work_type = data[\'work_type\'].isnull().sum()\\n","print(\\"# of missing work_type values: \\", missing_work_type)\\n","\\n","missing_Residence_type = data[\'Residence_type\'].isnull().sum()\\n","print(\\"# of missing Residence_type values: \\", missing_Residence_type)\\n","\\n","missing_avg_glucose_level = data[\'avg_glucose_level\'].isnull().sum()\\n","print(\\"# of missing avg_glucose_level values: \\", missing_avg_glucose_level)\\n","\\n","missing_bmi = data[\'bmi\'].isnull().sum()\\n","print(\\"# of missing bmi values: \\", missing_bmi)\\n","\\n","missing_smoking_status = data[\'smoking_status\'].isnull().sum()\\n","print(\\"# of missing smoking_status values: \\", missing_smoking_status)\\n","\\n","missing_stroke = data[\'stroke\'].isnull().sum()\\n","print(\\"# of missing stroke values: \\", missing_stroke)"]},{"cell_type":"markdown","metadata":{"id":"RGZBkMjL6uiE"},"source":["We see that there are only missing data values for BMI, so we decide to check how large the proportion of missing values to the positive case of strokes compared to the overall distribution."]},{"cell_type":"code","execution_count":6,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"XJmgOM9p6Gf-","outputId":"4c921ac3-e4f7-47d4-bc54-67b1e91cba4e"},"outputs":[{"name":"stdout","output_type":"stream","text":["Percent of patients that experienced stroke: 0.0487279843444227\\n","Percent of patients without a measured BMI that experienced stroke: 0.19900497512437812\\n"]}],"source":["stroke_dist = len(hot_data[hot_data[\'stroke\']==1].index)/len(hot_data.index)\\n","null_bmi = hot_data[hot_data[\'bmi\'].isnull()]\\n","null_stroke_dist = len(null_bmi[null_bmi[\'stroke\']==1].index)/len(null_bmi.index)\\n","print(\'Percent of patients that experienced stroke:\', stroke_dist)\\n","print(\'Percent of patients without a measured BMI that experienced stroke:\', null_stroke_dist)"]},{"cell_type":"markdown","metadata":{"id":"RChWSwOH6GgC"},"source":["We can see that there are NaN values for BMI, here we chose to use imputation to clean the data. We choose to use imputation over a deletion method because we believe this would introduce less bias. We see that the proportion of missing values is greater in the positive case (i.e., had a stroke) compared to the general distribution of data. This would make list wise deletion introduce bias and remove samples at a greater proportion for the positive case. One problem with using univariate imputation is that BMI is highly variable and the distribution of data could still introduce bias. Here we choose to use multivariate imputation to attempt to address this problem."]},{"cell_type":"code","execution_count":7,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":461},"id":"UbRfuLS3-lj9","outputId":"b02a29f4-5ea0-454c-c6bb-464cd68d9d71"},"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>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Female</th>\\n","      <th>gender_Other</th>\\n","      <th>ever_married_Yes</th>\\n","      <th>...</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_children</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>Residence_type_Rural</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>67.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>228.69</td>\\n","      <td>36.600000</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>61.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>202.21</td>\\n","      <td>31.171222</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>80.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>105.92</td>\\n","      <td>32.500000</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>49.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>171.23</td>\\n","      <td>34.400000</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>79.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>174.12</td>\\n","      <td>24.000000</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>...</th>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5105</th>\\n","      <td>80.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>83.75</td>\\n","      <td>32.263355</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5106</th>\\n","      <td>81.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>125.20</td>\\n","      <td>40.000000</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5107</th>\\n","      <td>35.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>82.99</td>\\n","      <td>30.600000</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5108</th>\\n","      <td>51.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>166.29</td>\\n","      <td>25.600000</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5109</th>\\n","      <td>44.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>85.28</td>\\n","      <td>26.200000</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","<p>5110 rows \xd7 22 columns</p>\\n","</div>"],"text/plain":["       age  hypertension  heart_disease  avg_glucose_level        bmi  stroke  \\\\\\n","0     67.0           0.0            1.0             228.69  36.600000     1.0   \\n","1     61.0           0.0            0.0             202.21  31.171222     1.0   \\n","2     80.0           0.0            1.0             105.92  32.500000     1.0   \\n","3     49.0           0.0            0.0             171.23  34.400000     1.0   \\n","4     79.0           1.0            0.0             174.12  24.000000     1.0   \\n","...    ...           ...            ...                ...        ...     ...   \\n","5105  80.0           1.0            0.0              83.75  32.263355     0.0   \\n","5106  81.0           0.0            0.0             125.20  40.000000     0.0   \\n","5107  35.0           0.0            0.0              82.99  30.600000     0.0   \\n","5108  51.0           0.0            0.0             166.29  25.600000     0.0   \\n","5109  44.0           0.0            0.0              85.28  26.200000     0.0   \\n","\\n","      gender_Male  gender_Female  gender_Other  ever_married_Yes  ...  \\\\\\n","0             1.0            0.0           0.0               1.0  ...   \\n","1             0.0            1.0           0.0               1.0  ...   \\n","2             1.0            0.0           0.0               1.0  ...   \\n","3             0.0            1.0           0.0               1.0  ...   \\n","4             0.0            1.0           0.0               1.0  ...   \\n","...           ...            ...           ...               ...  ...   \\n","5105          0.0            1.0           0.0               1.0  ...   \\n","5106          0.0            1.0           0.0               1.0  ...   \\n","5107          0.0            1.0           0.0               1.0  ...   \\n","5108          1.0            0.0           0.0               1.0  ...   \\n","5109          0.0            1.0           0.0               1.0  ...   \\n","\\n","      work_type_Self-employed  work_type_Govt_job  work_type_children  \\\\\\n","0                         0.0                 0.0                 0.0   \\n","1                         1.0                 0.0                 0.0   \\n","2                         0.0                 0.0                 0.0   \\n","3                         0.0                 0.0                 0.0   \\n","4                         1.0                 0.0                 0.0   \\n","...                       ...                 ...                 ...   \\n","5105                      0.0                 0.0                 0.0   \\n","5106                      1.0                 0.0                 0.0   \\n","5107                      1.0                 0.0                 0.0   \\n","5108                      0.0                 0.0                 0.0   \\n","5109                      0.0                 1.0                 0.0   \\n","\\n","      work_type_Never_worked  Residence_type_Urban  Residence_type_Rural  \\\\\\n","0                        0.0                   1.0                   0.0   \\n","1                        0.0                   0.0                   1.0   \\n","2                        0.0                   0.0                   1.0   \\n","3                        0.0                   1.0                   0.0   \\n","4                        0.0                   0.0                   1.0   \\n","...                      ...                   ...                   ...   \\n","5105                     0.0                   1.0                   0.0   \\n","5106                     0.0                   1.0                   0.0   \\n","5107                     0.0                   0.0                   1.0   \\n","5108                     0.0                   0.0                   1.0   \\n","5109                     0.0                   1.0                   0.0   \\n","\\n","      smoking_status_formerly smoked  smoking_status_never smoked  \\\\\\n","0                                1.0                          0.0   \\n","1                                0.0                          1.0   \\n","2                                0.0                          1.0   \\n","3                                0.0                          0.0   \\n","4                                0.0                          1.0   \\n","...                              ...                          ...   \\n","5105                             0.0                          1.0   \\n","5106                             0.0                          1.0   \\n","5107                             0.0                          1.0   \\n","5108                             1.0                          0.0   \\n","5109                             0.0                          0.0   \\n","\\n","      smoking_status_smokes  smoking_status_Unknown  \\n","0                       0.0                     0.0  \\n","1                       0.0                     0.0  \\n","2                       0.0                     0.0  \\n","3                       1.0                     0.0  \\n","4                       0.0                     0.0  \\n","...                     ...                     ...  \\n","5105                    0.0                     0.0  \\n","5106                    0.0                     0.0  \\n","5107                    0.0                     0.0  \\n","5108                    0.0                     0.0  \\n","5109                    0.0                     1.0  \\n","\\n","[5110 rows x 22 columns]"]},"execution_count":7,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.experimental import enable_iterative_imputer\\n","from sklearn.impute import IterativeImputer\\n","\\n","imp = IterativeImputer(max_iter=10, random_state=0)\\n","hot_data = pd.DataFrame(imp.fit_transform(hot_data), columns=list(hot_data.columns))\\n","hot_data"]},{"cell_type":"markdown","metadata":{"id":"cyOQ3bNN9E3D"},"source":["Now that we have filled in our missing values, we will now standardize all of the numerical data."]},{"cell_type":"code","execution_count":8,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":243},"id":"k5PCDn8L_4jJ","outputId":"506861db-1883-4c2a-bae7-d53e98c0f1ff"},"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>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Female</th>\\n","      <th>gender_Other</th>\\n","      <th>ever_married_Yes</th>\\n","      <th>...</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_children</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>Residence_type_Rural</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>1.051331</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>2.706111</td>\\n","      <td>0.990957</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>0.785993</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>2.121351</td>\\n","      <td>0.288746</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>1.626231</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>-0.005028</td>\\n","      <td>0.460623</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>0.255317</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.437217</td>\\n","      <td>0.706388</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>1.582008</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.501038</td>\\n","      <td>-0.638849</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","<p>5 rows \xd7 22 columns</p>\\n","</div>"],"text/plain":["        age  hypertension  heart_disease  avg_glucose_level       bmi  stroke  \\\\\\n","0  1.051331           0.0            1.0           2.706111  0.990957     1.0   \\n","1  0.785993           0.0            0.0           2.121351  0.288746     1.0   \\n","2  1.626231           0.0            1.0          -0.005028  0.460623     1.0   \\n","3  0.255317           0.0            0.0           1.437217  0.706388     1.0   \\n","4  1.582008           1.0            0.0           1.501038 -0.638849     1.0   \\n","\\n","   gender_Male  gender_Female  gender_Other  ever_married_Yes  ...  \\\\\\n","0          1.0            0.0           0.0               1.0  ...   \\n","1          0.0            1.0           0.0               1.0  ...   \\n","2          1.0            0.0           0.0               1.0  ...   \\n","3          0.0            1.0           0.0               1.0  ...   \\n","4          0.0            1.0           0.0               1.0  ...   \\n","\\n","   work_type_Self-employed  work_type_Govt_job  work_type_children  \\\\\\n","0                      0.0                 0.0                 0.0   \\n","1                      1.0                 0.0                 0.0   \\n","2                      0.0                 0.0                 0.0   \\n","3                      0.0                 0.0                 0.0   \\n","4                      1.0                 0.0                 0.0   \\n","\\n","   work_type_Never_worked  Residence_type_Urban  Residence_type_Rural  \\\\\\n","0                     0.0                   1.0                   0.0   \\n","1                     0.0                   0.0                   1.0   \\n","2                     0.0                   0.0                   1.0   \\n","3                     0.0                   1.0                   0.0   \\n","4                     0.0                   0.0                   1.0   \\n","\\n","   smoking_status_formerly smoked  smoking_status_never smoked  \\\\\\n","0                             1.0                          0.0   \\n","1                             0.0                          1.0   \\n","2                             0.0                          1.0   \\n","3                             0.0                          0.0   \\n","4                             0.0                          1.0   \\n","\\n","   smoking_status_smokes  smoking_status_Unknown  \\n","0                    0.0                     0.0  \\n","1                    0.0                     0.0  \\n","2                    0.0                     0.0  \\n","3                    1.0                     0.0  \\n","4                    0.0                     0.0  \\n","\\n","[5 rows x 22 columns]"]},"execution_count":8,"metadata":{},"output_type":"execute_result"}],"source":["# Standardize the age, avg_glucose_level, and bmi columns\\n","numerical_cols = [\\"age\\", \\"avg_glucose_level\\", \\"bmi\\"]\\n","for c in numerical_cols:\\n","    hot_data[c] = (hot_data[c] - hot_data[c].mean()) / hot_data[c].std()\\n","    \\n","hot_data.head()"]},{"cell_type":"markdown","metadata":{"id":"D-FvKrz1BVNS"},"source":["For all of our numerical data, we created boxplots to see the mean and quartile ranges for each feature. We also looked for any outliers to remove."]},{"cell_type":"code","execution_count":9,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":282},"id":"ilxy1b1zAnaU","outputId":"8baa7642-745a-41b9-bee8-dfc57542d0cc"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":9,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXwAAAD4CAYAAADvsV2wAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAOlklEQVR4nO3dbYilZ33H8e/PPIBNU62sTh6bzYstdfugyGE1+KJHjDZZpdtYLcmLam1hUAy+KrgloC2FktJXLYmu0xKM0BotZZOls91NYj2NFlKTSJJmE2OXNJLphqYxsmZMQLf++2KOZZg9sztn7nPO7O71/cBh74dr7v81cPGbe69zP6SqkCSd+16z1R2QJM2GgS9JjTDwJakRBr4kNcLAl6RGnL/VHTiVbdu21fbt27e6G9JJfvjDH3LRRRdtdTekkzzyyCMvVtUbR+07owN/+/btPPzww1vdDekkg8GAfr+/1d2QTpLku+vtc0pHkhph4EtSIwx8SWqEgS9JjTDwJakRBr4kNcLAl6RGGPiS1Igz+sYraRaSzKyW75/QVvIMX82rqrE/V33qHzf1c9JWMvAlqREGviQ1wsCXpEYY+JLUCANfkhph4EtSIwx8SWrERAI/yR1JXkjyxDr7+0mOJ3l0+Pn0JOpKkjZuUnfafgG4DfjiKdp8vareP6F6kqQxTeQMv6oeAF6axLEkSdMxy2fpXJPkMeAY8IdVdWRUoyTzwDzA3Nwcg8Fgdj2UxuDY1NlmVoH/LeCqqlpOshu4G9gxqmFVLQALAL1er/r9/oy6KI3h0CKOTZ1tZnKVTlX9oKqWh8sHgQuSbJtFbUnSipkEfpJLMnwGbZJdw7rfm0VtSdKKiUzpJPkS0Ae2JVkCPgNcAFBV+4APAh9PcgJ4FbixfFasJM3URAK/qm46zf7bWLlsU5K0RbzTVpIaYeBLUiMMfElqhIEvSY0w8CWpEQa+JDXCwJekRhj4ktQIA1+SGmHgS1IjDHxJaoSBL0mNMPAlqREGviQ1wsCXpEYY+JLUCANfkhph4EtSIyYS+EnuSPJCkifW2Z8kf5XkaJLHk7xtEnUlSRs3qTP8LwDXnWL/9cCO4Wce+NyE6kqSNmgigV9VDwAvnaLJHuCLteJB4PVJLp1EbUnSxpw/ozqXA8+tWl8abnt+bcMk86z8L4C5uTkGg8Es+ieNzbGps82sAj8jttWohlW1ACwA9Hq96vf7U+yWtEmHFnFs6mwzq6t0loArV61fARybUW1JErML/APAh4dX67wDOF5VJ03nSJKmZyJTOkm+BPSBbUmWgM8AFwBU1T7gILAbOAq8Anx0EnUlSRs3kcCvqptOs7+AT0yiliRpc7zTVpIaYeBLUiMMfElqhIEvSY0w8CWpEQa+JDViVo9WkGbmLX9yL8df/fHU62zfuzjV47/utRfw2GfeO9UaaouBr3PO8Vd/zLO3vm+qNQaDwdSfpTPtPyhqj1M6ktQIA1+SGmHgS1IjDHxJaoSBL0mNMPAlqREGviQ1wsCXpEYY+JLUCANfkhoxkcBPcl2Sp5McTbJ3xP5+kuNJHh1+Pj2JupKkjev8LJ0k5wG3A+8BloCHkhyoqifXNP16Vb2/az1J0uZM4gx/F3C0qp6pqh8BdwF7JnBcSdIETSLwLweeW7W+NNy21jVJHkvyT0l+eQJ1JUljmMTjkTNiW61Z/xZwVVUtJ9kN3A3sGHmwZB6YB5ibm2MwGEygi2rNtMfN8vLyTMam41+TNInAXwKuXLV+BXBsdYOq+sGq5YNJPptkW1W9uPZgVbUALAD0er2a9jPHdQ46tDj1Z9XP4nn4s/g91JZJTOk8BOxIcnWSC4EbgQOrGyS5JEmGy7uGdb83gdqSpA3qfIZfVSeS3AwcBs4D7qiqI0k+Nty/D/gg8PEkJ4BXgRurau20jyRpiibyisOqOggcXLNt36rl24DbJlFLkrQ53mkrSY0w8CWpEQa+JDXCwJekRhj4ktQIA1+SGmHgS1IjDHxJaoSBL0mNMPAlqREGviQ1wsCXpEYY+JLUCANfkhph4EtSIwx8SWqEgS9JjTDwJakRE3nFoXQmufjNe/nVO/dOv9Cd0z38xW8GeN90i6gpEwn8JNcBf8nKS8z/pqpuXbM/w/27gVeA36uqb02itrTWy0/dyrO3TjcoB4MB/X5/qjW2712c6vHVns5TOknOA24Hrgd2Ajcl2bmm2fXAjuFnHvhc17qSpPFMYg5/F3C0qp6pqh8BdwF71rTZA3yxVjwIvD7JpROoLUnaoElM6VwOPLdqfQl4+wbaXA48v/ZgSeZZ+V8Ac3NzDAaDCXRRrZn2uFleXp7J2HT8a5ImEfgZsa020WZlY9UCsADQ6/Vq2vOkOgcdWpz6/Pos5vBn8XuoLZOY0lkCrly1fgVwbBNtJElTNInAfwjYkeTqJBcCNwIH1rQ5AHw4K94BHK+qk6ZzJEnT03lKp6pOJLkZOMzKZZl3VNWRJB8b7t8HHGTlksyjrFyW+dGudSVJ45nIdfhVdZCVUF+9bd+q5QI+MYlakqTN8dEKktQIA1+SGmHgS1IjDHxJaoSBL0mNMPAlqREGviQ1wsCXpEYY+JLUCANfkhph4EtSIwx8SWqEgS9JjTDwJakRBr4kNcLAl6RGGPiS1AgDX5Ia0ekVh0neAHwZ2A48C/xOVX1/RLtngZeB/wVOVFWvS11J0vi6nuHvBb5aVTuArw7X1/OuqnqrYS9JW6Nr4O8B7hwu3wn8VsfjSZKmpNOUDjBXVc8DVNXzSd60TrsC7k1SwOeramG9AyaZB+YB5ubmGAwGHbuoFk173CwvL89kbDr+NUmnDfwk9wOXjNh1yxh13llVx4Z/EO5L8u2qemBUw+EfgwWAXq9X/X5/jDIScGiRaY+bwWAw9Rqz+D3UltMGflVdu96+JP+d5NLh2f2lwAvrHOPY8N8XkuwHdgEjA1+SNB1d5/APAB8ZLn8EuGdtgyQXJbn4p8vAe4EnOtaVJI2pa+DfCrwnyX8A7xmuk+SyJAeHbeaAbyR5DPgmsFhVhzrWlSSNqdOXtlX1PeDdI7YfA3YPl58B3tKljiSpO++0laRGGPiS1AgDX5IaYeBLUiMMfElqhIEvSY0w8CWpEQa+JDXCwJekRhj4ktQIA1+SGtH1BSjSGWn73sXpFzk03Rqve+0FUz2+2mPg65zz7K3vm3qN7XsXZ1JHmiSndCSpEQa+JDXCwJekRhj4ktQIA1+SGtEp8JN8KMmRJD9J0jtFu+uSPJ3kaJK9XWpKkjan6xn+E8AHgAfWa5DkPOB24HpgJ3BTkp0d60qSxtT1JeZPASQ5VbNdwNHhy8xJchewB3iyS21J0nhmcePV5cBzq9aXgLev1zjJPDAPMDc3x2AwmGrnpM1ybOpsc9rAT3I/cMmIXbdU1T0bqDHq9L/Wa1xVC8ACQK/Xq36/v4ES0owdWsSxqbPNaQO/qq7tWGMJuHLV+hXAsY7HlCSNaRaXZT4E7EhydZILgRuBAzOoK0lapetlmTckWQKuARaTHB5uvyzJQYCqOgHcDBwGngK+UlVHunVbkjSurlfp7Af2j9h+DNi9av0gcLBLLUlSN95pK0mNMPAlqREGviQ1wsCXpEYY+JLUCANfkhph4EtSIwx8SWqEgS9JjTDwJakRBr4kNcLAl6RGGPiS1AgDX5IaYeBLUiMMfElqhIEvSY0w8CWpEV3fafuhJEeS/CRJ7xTtnk3y70keTfJwl5qSpM3p9E5b4AngA8DnN9D2XVX1Ysd6kqRN6voS86cAkkymN5KkqZnVHH4B9yZ5JMn8jGpKklY57Rl+kvuBS0bsuqWq7tlgnXdW1bEkbwLuS/LtqnpgnXrzwDzA3Nwcg8FggyWk2XJs6mxz2sCvqmu7FqmqY8N/X0iyH9gFjAz8qloAFgB6vV71+/2u5aXJO7SIY1Nnm6lP6SS5KMnFP10G3svKl72SpBnqelnmDUmWgGuAxSSHh9svS3Jw2GwO+EaSx4BvAotVdahLXUnS+LpepbMf2D9i+zFg93D5GeAtXepIkrrzTltJaoSBL0mNMPAlqREGviQ1wsCXpEYY+JLUCANfkhph4EtSIwx8SWqEgS9JjTDwJakRBr4kNcLAl6RGGPiS1AgDX5IaYeBLUiMMfElqRKc3XknngiSb+7k/H/9nqmpTtaRJ8AxfzauqsT9f+9rXNvVz0lbq+hLzv0jy7SSPJ9mf5PXrtLsuydNJjibZ26WmJGlzup7h3wf8SlX9GvAd4I/WNkhyHnA7cD2wE7gpyc6OdSVJY+oU+FV1b1WdGK4+CFwxotku4GhVPVNVPwLuAvZ0qStJGt8kv7T9feDLI7ZfDjy3an0JePt6B0kyD8wDzM3NMRgMJthFaTKWl5cdmzrrnDbwk9wPXDJi1y1Vdc+wzS3ACeBvRx1ixLZ1v72qqgVgAaDX61W/3z9dF6WZGwwGODZ1tjlt4FfVtafan+QjwPuBd9foyxCWgCtXrV8BHBunk5Kk7rpepXMd8CngN6vqlXWaPQTsSHJ1kguBG4EDXepKksbX9Sqd24CLgfuSPJpkH0CSy5IcBBh+qXszcBh4CvhKVR3pWFeSNKacyTeDJPkf4Ltb3Q9phG3Ai1vdCWmEq6rqjaN2nNGBL52pkjxcVb2t7oc0Dh+tIEmNMPAlqREGvrQ5C1vdAWlczuFLUiM8w5ekRhj4ktQIA1+SGmHgS1IjDHxphCR3J3kkyZHhI7tJ8gdJvpNkkOSvk9w23P7GJP+Q5KHh551b23tpNK/SkUZI8oaqeinJa1l5AOBvAP8KvA14Gfhn4LGqujnJ3wGfrapvJPkF4HBVvXnLOi+tY5IvQJHOJZ9McsNw+Urgd4F/qaqXAJL8PfCLw/3XAjuT/3/1w88lubiqXp5lh6XTMfClNZL0WQnxa6rqlSQD4GlgvbP21wzbvjqTDkqb5By+dLLXAd8fhv0vAe8Afgb49SQ/n+R84LdXtb+XlUeAA5DkrbPsrLRRBr50skPA+UkeB/4UeBD4L+DPgH8D7geeBI4P238S6CV5PMmTwMdm32Xp9PzSVtqgJD9bVcvDM/z9wB1VtX+r+yVtlGf40sb9cZJHgSeA/wTu3tLeSGPyDF+SGuEZviQ1wsCXpEYY+JLUCANfkhph4EtSI/4PZI4yKSH7LuQAAAAASUVORK5CYII=\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Boxplot for age\\n","hot_data.boxplot(column=[\'age\'])"]},{"cell_type":"code","execution_count":10,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":283},"id":"0AOfJ_lDBvfi","outputId":"bfe0525b-c9ae-4270-939c-533632493306"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":10,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Boxplot for glucode level\\n","hot_data.boxplot(column=[\'avg_glucose_level\'])"]},{"cell_type":"code","execution_count":11,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":282},"id":"8o07dtAfBvo6","outputId":"d06a7c2b-5e48-42cb-ad61-f072f1a50101"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":11,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Boxplot for BMI\\n","hot_data.boxplot(column=[\'bmi\'])"]},{"cell_type":"markdown","metadata":{"id":"QWQ1qh63BZic"},"source":["Create bar charts for all categorical data"]},{"cell_type":"code","execution_count":12,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":308},"id":"Hz-DQxgXDI0E","outputId":"e40f46c8-de94-46e3-f324-51f8aa81e0f2"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":12,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for gender counts\\n","data[\\"gender\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":13,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":279},"id":"-X8HPziZD4Gc","outputId":"4d4977ad-bee1-4fca-d01c-4661f0ddd4e5"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":13,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for hypertension counts\\n","data[\\"hypertension\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":14,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":279},"id":"pHhFUsUAEAAg","outputId":"a402a9c7-582d-4465-8714-4d241b8d774e"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":14,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for heart_disease counts\\n","data[\\"heart_disease\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":15,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":291},"id":"-8AF4qSIEGCt","outputId":"6e218e12-b102-4ed3-8f26-10f9a94d371a"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":15,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for ever_married counts\\n","data[\\"ever_married\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":16,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":344},"id":"SKKadf_0EK2U","outputId":"73f5fbce-5e04-4a17-ff04-5d71404b4944"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":16,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for work_type counts\\n","data[\\"work_type\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":17,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":303},"id":"4iumIQiGESeM","outputId":"39007d8a-84ca-4644-de35-7fd1a3141160"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":17,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAX0AAAENCAYAAADjW7WQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAQGUlEQVR4nO3df6zddX3H8eeLVpEIGEgvDNtiGemihWUoN00zlkxlG51uFpeQlSzSP0xqSI26uT/AfzRZmrgfasIyWGp0lIWNdVHTJsI2JGSGDMUL6VZK7ewsg2ubtqKbnS5o63t/nO+NZ+3pvbe3t+fA+Twfycn5nvf5fL/nfXJvXvd7P9/v95xUFZKkNlww6gYkScNj6EtSQwx9SWqIoS9JDTH0Jakhhr4kNWTpqBuYy7Jly2rVqlWjbkOSXlWefvrp71bVxKn1V3zor1q1iqmpqVG3IUmvKkn+c1Dd6R1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQ17xF2e9Wqy668ujbmFsPP/Jd4+6BWlsuacvSQ0x9CWpIYa+JDVkztBPsjLJ40n2Jdmb5MNd/RNJvpNkd3d7V986dyc5kGR/klv66jcm2dM9d0+SnJ+3JUkaZD4Hck8AH62qZ5JcAjyd5NHuuc9U1Z/1D06yBtgIXAe8EfhKkl+oqpPAfcBm4GvAw8B64JHFeSuSpLnMuadfVYer6plu+TiwD1g+yyobgIeq6uWqOggcANYmuQq4tKqerKoCHgBuPdc3IEmav7M6ZTPJKuCtwNeBm4APJrkDmKL338D36f1B+FrfatNd7Sfd8ql1SeeRpxMvrlf7KcXzPpCb5GLgC8BHquoH9KZqrgVuAA4Dn5oZOmD1mqU+6LU2J5lKMnXs2LH5tihJmsO8Qj/Ja+gF/oNV9UWAqjpSVSer6qfAZ4G13fBpYGXf6iuAQ119xYD6aapqW1VNVtXkxMRp3/YlSVqg+Zy9E+BzwL6q+nRf/aq+Ye8Fnu2WdwEbk1yY5BpgNfBUVR0GjidZ123zDmDnIr0PSdI8zGdO/ybgfcCeJLu72seA25PcQG+K5nngAwBVtTfJDuA5emf+bOnO3AG4E7gfuIjeWTueuSNJQzRn6FfVEwyej394lnW2AlsH1KeA68+mQUnS4vGKXElqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGjJn6CdZmeTxJPuS7E3y4a5+eZJHk3yru7+sb527kxxIsj/JLX31G5Ps6Z67J0nOz9uSJA0ynz39E8BHq+otwDpgS5I1wF3AY1W1Gnise0z33EbgOmA9cG+SJd227gM2A6u72/pFfC+SpDnMGfpVdbiqnumWjwP7gOXABmB7N2w7cGu3vAF4qKperqqDwAFgbZKrgEur6smqKuCBvnUkSUNwVnP6SVYBbwW+DlxZVYeh94cBuKIbthx4sW+16a62vFs+tS5JGpJ5h36Si4EvAB+pqh/MNnRArWapD3qtzUmmkkwdO3Zsvi1KkuYwr9BP8hp6gf9gVX2xKx/ppmzo7o929WlgZd/qK4BDXX3FgPppqmpbVU1W1eTExMR834skaQ7zOXsnwOeAfVX16b6ndgGbuuVNwM6++sYkFya5ht4B26e6KaDjSdZ127yjbx1J0hAsnceYm4D3AXuS7O5qHwM+CexI8n7gBeA2gKram2QH8By9M3+2VNXJbr07gfuBi4BHupskaUjmDP2qeoLB8/EAN59hna3A1gH1KeD6s2lQkrR4vCJXkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGzBn6ST6f5GiSZ/tqn0jynSS7u9u7+p67O8mBJPuT3NJXvzHJnu65e5Jk8d+OJGk289nTvx9YP6D+maq6obs9DJBkDbARuK5b594kS7rx9wGbgdXdbdA2JUnn0ZyhX1VfBb43z+1tAB6qqper6iBwAFib5Crg0qp6sqoKeAC4dYE9S5IW6Fzm9D+Y5N+66Z/Lutpy4MW+MdNdbXm3fGpdkjRECw39+4BrgRuAw8CnuvqgefqapT5Qks1JppJMHTt2bIEtSpJOtaDQr6ojVXWyqn4KfBZY2z01DazsG7oCONTVVwyon2n726pqsqomJyYmFtKiJGmABYV+N0c/473AzJk9u4CNSS5Mcg29A7ZPVdVh4HiSdd1ZO3cAO8+hb0nSAiyda0CSvwXeDixLMg18HHh7khvoTdE8D3wAoKr2JtkBPAecALZU1cluU3fSOxPoIuCR7iZJGqI5Q7+qbh9Q/tws47cCWwfUp4Drz6o7SdKi8opcSWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaMmfoJ/l8kqNJnu2rXZ7k0STf6u4v63vu7iQHkuxPcktf/cYke7rn7kmSxX87kqTZzGdP/35g/Sm1u4DHqmo18Fj3mCRrgI3Add069yZZ0q1zH7AZWN3dTt2mJOk8mzP0q+qrwPdOKW8AtnfL24Fb++oPVdXLVXUQOACsTXIVcGlVPVlVBTzQt44kaUgWOqd/ZVUdBujur+jqy4EX+8ZNd7Xl3fKpdUnSEC32gdxB8/Q1S33wRpLNSaaSTB07dmzRmpOk1i009I90UzZ090e7+jSwsm/cCuBQV18xoD5QVW2rqsmqmpyYmFhgi5KkUy009HcBm7rlTcDOvvrGJBcmuYbeAdunuimg40nWdWft3NG3jiRpSJbONSDJ3wJvB5YlmQY+DnwS2JHk/cALwG0AVbU3yQ7gOeAEsKWqTnabupPemUAXAY90N0nSEM0Z+lV1+xmeuvkM47cCWwfUp4Drz6o7SdKi8opcSWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5Iack6hn+T5JHuS7E4y1dUuT/Jokm9195f1jb87yYEk+5Pccq7NS5LOzmLs6b+jqm6oqsnu8V3AY1W1Gnise0ySNcBG4DpgPXBvkiWL8PqSpHk6H9M7G4Dt3fJ24Na++kNV9XJVHQQOAGvPw+tLks7gXEO/gH9K8nSSzV3tyqo6DNDdX9HVlwMv9q073dUkSUOy9BzXv6mqDiW5Ang0yTdnGZsBtRo4sPcHZDPA1VdffY4tSpJmnNOeflUd6u6PAl+iN11zJMlVAN390W74NLCyb/UVwKEzbHdbVU1W1eTExMS5tChJ6rPg0E/y+iSXzCwDvwE8C+wCNnXDNgE7u+VdwMYkFya5BlgNPLXQ15cknb1zmd65EvhSkpnt/E1V/UOSbwA7krwfeAG4DaCq9ibZATwHnAC2VNXJc+peknRWFhz6VfVt4JcG1F8Cbj7DOluBrQt9TUnSufGKXElqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGjL00E+yPsn+JAeS3DXs15eklg019JMsAf4C+E1gDXB7kjXD7EGSWjbsPf21wIGq+nZV/Rh4CNgw5B4kqVnDDv3lwIt9j6e7miRpCJYO+fUyoFanDUo2A5u7h/+TZP957aody4DvjrqJueSPR92BRsTfz8X1pkHFYYf+NLCy7/EK4NCpg6pqG7BtWE21IslUVU2Oug9pEH8/h2PY0zvfAFYnuSbJa4GNwK4h9yBJzRrqnn5VnUjyQeAfgSXA56tq7zB7kKSWDXt6h6p6GHh42K8rwCkzvbL5+zkEqTrtOKokaUz5MQyS1BBDX5IaYuhLUkOGfiBXw5fkl4FV9P28q+qBkTUkaWQM/TGX5K+Ba4HdwMmuXIChr5FJcpwBV+PTu2q/qurSIbfUDM/eGXNJ9gFryh+0JNzTb8GzwM8Bh0fdiHQmSa4AXjfzuKpeGGE7Y83QH3/LgOeSPAW8PFOsqveMriWpJ8l7gE8BbwSO0vuQsH3AdaPsa5wZ+uPvE6NuQJrFHwHrgK9U1VuTvAO4fcQ9jTVDf8xV1T+PugdpFj+pqpeSXJDkgqp6PHkVfXjxq5ChP+aSrAP+HHgL8Fp6H3T3Q8+O0CvEfyW5GPgq8GCSo8CJEfc01jx7Z8wlmaL3EdZ/D0wCdwCrq+pjI21MApK8HvhfeheK/h7wBuDBqnpppI2NMff0G1BVB5IsqaqTwF8l+ZdR9yQlWQLsrKpfA34KbB9xS00w9Mffj7ovrNmd5E/onbr5+hH3JFFVJ5P8KMkbquq/R91PK5zeGXNJ3gQcoTef//v0/n2+t6oOjLQxCUiyg97ZO48CP5ypV9WHRtbUmDP0G9Dt6b+Z3mXv+6vqxyNuSQIgyaZB9apyquc8MfTHXJJ3A38J/Ae9zzW5BvhAVT0y0sYkjYShP+aSfBP4rZnpnCTXAl+uqjePtjMJkhxkwAevVdXPj6CdJnggd/wdPWX+/tv0LneXXgkm+5ZfB9wGXD6iXprgnv6YSvI73eKv0/s8kx309qhuozev/9FR9SbNJskTVfUro+5jXLmnP75+u2/5CPCr3fIx4LLhtyOdLsnb+h5eQG/P/5IRtdME9/THWHfxy4eq6jOj7kUaJMnjfQ9PAM8Df1pV/z6ajsafoT/mkjxeVe8YdR/SfCRZCvxuVT046l7GlaE/5pJspXdB1t/x/y9+eWZkTal5SS4FtgDLgZ3AV7rHfwj8a1VtGGF7Y83QH3N9/z7P/KBnvoP0nSNqSSLJTuD7wJPAzfSOM70W+HBV7R5ha2PP0B9TSf5gZrG7L3oHcZ+oqoOj6UrqSbKnqn6xW14CfBe4uqqOj7az8XfBqBvQeXNJd7u4u11C78yIR5JsHGVjEvCTmYXu018PGvjD4Z5+Y5JcTu+r6d4252DpPElykp8dYwpwEfAjfjb96Jf8nCeep9+Yqvpeksw9Ujp/qmrJqHtoldM7jUnyTnoH0CQ1yD39MZVkD6d/kNXlwCF6X5koqUHO6Y+p7stT+hXwUlX9cNB4SW0w9CWpIc7pS1JDDH1JaoihL0kNMfQlqSGGviQ15P8AZfowVRUn2QwAAAAASUVORK5CYII=\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for Residence_type counts\\n","data[\\"Residence_type\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":18,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":356},"id":"VC5Hnhy7EVTn","outputId":"9b231c74-c3d5-4ea2-8219-c58a59f9eedb"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":18,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for smoking_status counts\\n","data[\\"smoking_status\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"code","execution_count":19,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":279},"id":"5b8FRsO0GOtQ","outputId":"a8492334-62cc-46f2-8c75-94c28679162a"},"outputs":[{"data":{"text/plain":["<AxesSubplot:>"]},"execution_count":19,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Bar chart for stroke counts\\n","data[\\"stroke\\"].value_counts().plot(kind=\'bar\')"]},{"cell_type":"markdown","metadata":{"id":"kQem21jQ7Ns2"},"source":["Since we only detected outliers in the Average Glucose Levels and BMI, we promptly remove those data points from our data."]},{"cell_type":"code","execution_count":20,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":261},"id":"Ns-j7EkfEa9A","outputId":"49d159e6-c47b-49d4-fc32-646a79a0f2b4"},"outputs":[{"name":"stdout","output_type":"stream","text":["New hot_dataset shape: (5002, 22)\\n"]},{"data":{"text/html":["<div>\\n","<style scoped>\\n","    .dataframe tbody tr th:only-of-type {\\n","        vertical-align: middle;\\n","    }\\n","\\n","    .dataframe tbody tr th {\\n","        vertical-align: top;\\n","    }\\n","\\n","    .dataframe thead th {\\n","        text-align: right;\\n","    }\\n","</style>\\n","<table border=\\"1\\" class=\\"dataframe\\">\\n","  <thead>\\n","    <tr style=\\"text-align: right;\\">\\n","      <th></th>\\n","      <th>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Female</th>\\n","      <th>gender_Other</th>\\n","      <th>ever_married_Yes</th>\\n","      <th>...</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_children</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>Residence_type_Rural</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>1.051331</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>2.706111</td>\\n","      <td>0.990957</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>0.785993</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>2.121351</td>\\n","      <td>0.288746</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>1.626231</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>-0.005028</td>\\n","      <td>0.460623</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>0.255317</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.437217</td>\\n","      <td>0.706388</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>1.582008</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.501038</td>\\n","      <td>-0.638849</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","<p>5 rows \xd7 22 columns</p>\\n","</div>"],"text/plain":["        age  hypertension  heart_disease  avg_glucose_level       bmi  stroke  \\\\\\n","0  1.051331           0.0            1.0           2.706111  0.990957     1.0   \\n","1  0.785993           0.0            0.0           2.121351  0.288746     1.0   \\n","2  1.626231           0.0            1.0          -0.005028  0.460623     1.0   \\n","3  0.255317           0.0            0.0           1.437217  0.706388     1.0   \\n","4  1.582008           1.0            0.0           1.501038 -0.638849     1.0   \\n","\\n","   gender_Male  gender_Female  gender_Other  ever_married_Yes  ...  \\\\\\n","0          1.0            0.0           0.0               1.0  ...   \\n","1          0.0            1.0           0.0               1.0  ...   \\n","2          1.0            0.0           0.0               1.0  ...   \\n","3          0.0            1.0           0.0               1.0  ...   \\n","4          0.0            1.0           0.0               1.0  ...   \\n","\\n","   work_type_Self-employed  work_type_Govt_job  work_type_children  \\\\\\n","0                      0.0                 0.0                 0.0   \\n","1                      1.0                 0.0                 0.0   \\n","2                      0.0                 0.0                 0.0   \\n","3                      0.0                 0.0                 0.0   \\n","4                      1.0                 0.0                 0.0   \\n","\\n","   work_type_Never_worked  Residence_type_Urban  Residence_type_Rural  \\\\\\n","0                     0.0                   1.0                   0.0   \\n","1                     0.0                   0.0                   1.0   \\n","2                     0.0                   0.0                   1.0   \\n","3                     0.0                   1.0                   0.0   \\n","4                     0.0                   0.0                   1.0   \\n","\\n","   smoking_status_formerly smoked  smoking_status_never smoked  \\\\\\n","0                             1.0                          0.0   \\n","1                             0.0                          1.0   \\n","2                             0.0                          1.0   \\n","3                             0.0                          0.0   \\n","4                             0.0                          1.0   \\n","\\n","   smoking_status_smokes  smoking_status_Unknown  \\n","0                    0.0                     0.0  \\n","1                    0.0                     0.0  \\n","2                    0.0                     0.0  \\n","3                    1.0                     0.0  \\n","4                    0.0                     0.0  \\n","\\n","[5 rows x 22 columns]"]},"execution_count":20,"metadata":{},"output_type":"execute_result"}],"source":["from scipy import stats\\n","# Remove some outliers\\n","\\n","# From average glucose level\\n","hot_data = hot_data[(np.abs(stats.zscore(hot_data[\\"avg_glucose_level\\"])) < 3)]\\n","\\n","# From BMI\\n","hot_data = hot_data[(np.abs(stats.zscore(hot_data[\\"bmi\\"])) < 3)]\\n","\\n","print(f\\"New hot_dataset shape: {hot_data.shape}\\")\\n","hot_data.head()"]},{"cell_type":"markdown","metadata":{"id":"vWxBvjdJ6Gga"},"source":["## Feature Engineering"]},{"cell_type":"markdown","metadata":{"id":"MUYznMzi7hoC"},"source":["Now that we have cleaned the data, we perform feature engineering to select useful features to use to build our models. In the histograms shown below, the orange displays the data that has had a stroke, while the blue shows the data that has not."]},{"cell_type":"code","execution_count":21,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"MrZEE4IU6Ggb","outputId":"15365443-1f8e-4361-dee2-43c56219ff9f"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["stroke = data[data[\'stroke\'] == 1]\\n","no_stroke = data[data[\'stroke\']==0]\\n","\\n","plt.hist(no_stroke[\'age\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'age\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":22,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"D8AKvRBj6Ggc","outputId":"272759e8-0f4e-4c72-ec71-d464ecc578e6"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'hypertension\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'hypertension\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":23,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":269},"id":"Vpt6rzDi6Ggd","outputId":"a5cd8399-0155-4ffd-ba19-b28a661e6128"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'heart_disease\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'heart_disease\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":24,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"rosleKuc6Ggf","outputId":"5e000670-1367-4334-bc2a-2608e8005e52"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'avg_glucose_level\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'avg_glucose_level\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":25,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"xukHb2XB6Ggg","outputId":"4aa31f51-4504-401b-94e1-6d10112fd5a7"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'bmi\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'bmi\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":26,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"AuL0f_rP6Ggh","outputId":"b6e5fb7d-2abf-4527-8717-0854fbb8c3ed"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'gender\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'gender\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":27,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":266},"id":"fTNgXlx96Ggi","outputId":"678f3999-1775-411e-a170-7c5552e98100"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'work_type\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'work_type\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":28,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"5krU2o446Ggj","outputId":"0248accf-b7fb-4edd-e477-64d998448b65"},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAWoAAAD4CAYAAADFAawfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAKUUlEQVR4nO3dYYylZ1nH4f9Ntw1EiLXZCTGUOIpVg0SoTGq0xkpBUxQ1JppAFDUxWTWo1UiMJobIFz4ZYmI0cRWCiRUElYSSoFShxSoFZ0uB1raCTY0NjTulQVgiaNvbD3OWLtuZnbPtnJk73etKJnvmnPe85/702yfPvu/Z6u4AMNczDnsAAM5NqAGGE2qA4YQaYDihBhjuyCpOevTo0V5fX1/FqQGelk6cOPFQd6/t9NpKQr2+vp7Nzc1VnBrgaamq/mO312x9AAwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwy3kjsTAQ7VH19zOJ/7C7es5LRW1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwy31JcyVdX9Sb6Q5NEkj3T3xiqHAngqPnXy1KF87hUrOu/5fHvey7r7oRXNAcAubH0ADLdsqDvJ+6vqRFUd2+mAqjpWVZtVtbm1tbV/EwJc4JYN9dXd/Z1JXpnkdVX1fWcf0N3Hu3ujuzfW1tb2dUiAC9lSe9Td/ZnFnyer6t1JrkryoVUM9CN/cOsqTrunG3/lew/lcwH2sueKuqq+pqqec/pxkh9McueqBwNg2zIr6ucmeXdVnT7+L7r7b1c6FQBfsWeou/u+JC8+gFkA2IHL8wCGE2qA4YQaYDihBhhOqAGGE2qA4YQaYDihBhhOqAGGE2qA4c7nf3g5EG/+3PWH9MknDulzAc7NihpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGG7pUFfVRVX1sap67yoHAuCrnc+K+vokd69qEAB2tlSoq+ryJD+c5E9XOw4AZ1t2Rf37SX4zyWO7HVBVx6pqs6o2t7a29mM2ALJEqKvqVUlOdveJcx3X3ce7e6O7N9bW1vZtQIAL3TIr6quT/GhV3Z/kHUmurao/X+lUAHzFnqHu7t/u7su7ez3Jq5N8oLt/euWTAZDEddQA4x05n4O7++YkN69kEgB2ZEUNMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADDCTXAcEINMNyeoa6qZ1bVR6vq41V1V1W98SAGA2DbkSWO+XKSa7v7VFVdnOTWqnpfd9+24tkAyBKh7u5Ocmrx68WLn17lUAA8bqk96qq6qKruSHIyyU3d/ZEdjjlWVZtVtbm1tbXPYwJcuJYKdXc/2t0vSXJ5kquq6kU7HHO8uze6e2NtbW2fxwS4cJ3XVR/d/bkkNye5bhXDAPBEy1z1sVZVly4ePyvJK5Lcs+K5AFhY5qqPr0/yZ1V1UbbD/s7ufu9qxwLgtGWu+vhEkisPYBYAduDORIDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDhhBpgOKEGGE6oAYYTaoDh9gx1VT2/qj5YVXdX1V1Vdf1BDAbAtiNLHPNIkt/o7tur6jlJTlTVTd39ryueDYAssaLu7ge7+/bF4y8kuTvJ81Y9GADbzmuPuqrWk1yZ5CM7vHasqjaranNra2ufxgNg6VBX1bOT/HWSX+vuz5/9encf7+6N7t5YW1vbzxkBLmhLhbqqLs52pG/o7r9Z7UgAnGmZqz4qyVuS3N3db179SACcaZkV9dVJXpvk2qq6Y/HzQyueC4CFPS/P6+5bk9QBzALADtyZCDCcUAMMJ9QAwwk1wHBCDTCcUAMMJ9QAwwk1wHBCDTCcUAMMJ9QAwwk1wHBCDTCcUAMMJ9QAwwk1wHBCDTCcUAMMJ9QAwwk1wHBCDTCcUAMMJ9QAwwk1wHBCDTCcUAMMJ9QAwwk1wHBCDTDcnqGuqrdW1cmquvMgBgLgqy2zon5bkutWPAcAu9gz1N39oSQPH8AsAOzAHjXAcPsW6qo6VlWbVbW5tbW1X6cFuODtW6i7+3h3b3T3xtra2n6dFuCCZ+sDYLhlLs97e5IPJ/nWqnqgqn5+9WMBcNqRvQ7o7tccxCAA7MzWB8BwQg0wnFADDCfUAMMJNcBwQg0wnFADDCfUAMMJNcBwQg0wnFADDCfUAMMJNcBwQg0wnFADDCfUAMMJNcBwQg0wnFADDCfUAMMJNcBwQg0wnFADDCfUAMMJNcBwQg0wnFADDCfUAMMJNcBwQg0w3FKhrqrrqureqvp0Vf3WqocC4HF7hrqqLkryh0lemeSFSV5TVS9c9WAAbFtmRX1Vkk93933d/b9J3pHkx1Y7FgCnHVnimOcl+c8zfn8gyXedfVBVHUtybPHrqaq690nOdDTJQ0/yvU/eG+rAPxJ4mnlDPZV+fcNuLywT6p0K1k94ovt4kuPnMdTOH1a12d0bT/U8AAdtVf1aZuvjgSTPP+P3y5N8Zr8HAWBny4T6X5JcUVXfWFWXJHl1kvesdiwATttz66O7H6mqX07yd0kuSvLW7r5rhTM95e0TgEOykn5V9xO2mwEYxJ2JAMMJNcBwKw91VT1aVXdU1Z1VdWNVXbqP5z61X+cCOK2q1qvqzrOe+92qev0Ox76tqn5ilfMcxIr6f7r7Jd39oiQPJ3ndsm+sqmWu8wY4FAfVqIPe+vhwtu90TFXdXFUbi8dHq+r+xeOfq6p3VdWNSd5fVc+uqn+oqtur6pNV5fZ14NAs2vWmqrolyfWLp19RVf9YVf9WVa9aHLe+eO72xc/3LJ7//sU5/qqq7qmqG6rqnLdGH9iKdfHlTi9P8pYlDv/uJN/R3Q8v/sb68e7+fFUdTXJbVb2nXa4CHJ5Lu/uaZHvrI8l6kmuSvCDJB6vqm5OcTPID3f2lqroiyduTnL5r8cok357tmwf/KcnVSW7d7cMOYkX9rKq6I8lnk1yW5KYl3nNTdz+8eFxJ3lRVn0jy99lekT93FYMCLOy2EDz9/F+e9fw7u/ux7v5UkvuSfFuSi5P8SVV9Msm7sv3to6d9tLsf6O7HktyR7dDv6sD2qLP9hSOX5PE96kfO+PxnnvWeL57x+KeSrCV56eI8/7XD8QD76bNJvu6s5y7L41+49MWzXjs77J3k17PdqxdneyV9yRmvf/mMx49mj92NA9uj7u7/TvKrSV5fVRcnuT/JSxcvn+tfTL82ycnu/r+qelnO8Q1TAPuhu08lebCqXp4kVXVZkuuy+/bET1bVM6rqBUm+Kcm92W7Xg4tV82uzfWf3k3Kg/5jY3R9L8vFsf1/I7yX5par652x/telubkiyUVWb2V5d37PyQQGSn0nyO4ut2w8keWN3//sux96b5JYk70vyi939pSR/lORnq+q2JN+SJ67Cl+YWcoDh3JkIMJxQAwwn1ADDCTXAcEINMJxQAwwn1ADD/T96Lf3jTkyDQQAAAABJRU5ErkJggg==\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'Residence_type\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'Residence_type\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"code","execution_count":29,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"id":"p50Gle7-6Ggk","outputId":"ed7a1b6c-0be0-408b-831a-37d929268c68"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(no_stroke[\'smoking_status\'], density=True, alpha=0.8)\\n","plt.hist(stroke[\'smoking_status\'], density=True, alpha=0.8)\\n","\\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"hTZ35I5Q6Ggl"},"source":["We can also visualize the correlations between features and the stroke class using a correlation matrix. We want to ensure that features are not highly correlated with each other because this means that they are redundant."]},{"cell_type":"code","execution_count":30,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":970},"id":"z3LAGmQRIREf","outputId":"bbb725ba-4cf1-4d1c-d076-8f3ef4660a02"},"outputs":[{"data":{"image/png":"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\\n","text/plain":["<Figure size 1152x1008 with 2 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["#Using Pearson Correlation\\n","import seaborn as sns\\n","\\n","plt.figure(figsize=(16, 14))\\n","cor = hot_data.corr()\\n","cor = abs(cor)\\n","sns.heatmap(cor, annot=True, cmap=plt.cm.Blues)\\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"o0E-uwm2O4CK"},"source":["This tells us that among these features, age is highly correlated with ever_married_No and with ever_married_Yes. work_type_children is similarly correlated. In addition, ever_married_No and ever_married_Yes are highly correlated. Therefore, we can remove all of ever_married_No, ever_married_Yes, and work_type_children because they are redundant when we have age. Lastly, Residence_type_Urban and Residence_type_Rural are highly correlated, as well as gender_Male and gender_Female so we will only keep one of each. This results in the following correlation matrix for the features to select."]},{"cell_type":"code","execution_count":31,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":970},"id":"lZkjWfp2Palr","outputId":"5d6534be-9486-4a85-a1ab-651049bdb6cd"},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAA/AAAAO5CAYAAAC+CM6qAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdd1hT1x/H8fdhKFacEIIDt3XbZVut27oQleHeWq2jzqp114l7VK22jrrr1rr3Blx1tG6t9qe4ylJbt0Jyf38kQgKoICCk/b6ex0dIzk0+nHGTe8/JjdI0DSGEEEIIIYQQQqRtdqkdQAghhBBCCCGEEK8nB/BCCCGEEEIIIYQNkAN4IYQQQgghhBDCBsgBvBBCCCGEEEIIYQPkAF4IIYQQQgghhLABcgAvhBBCCCGEEELYADmAF0IIIYQQQgghkpFSar5SKkwpdfYl9yul1HSl1BWl1Gml1IcJeVw5gBdCCCGEEEIIIZLXQqD2K+73BAqb/3UEfkzIg8oBvBBCCCGEEEIIkYw0TQsA7r6iiDewWDM5AmRVSuV43ePKAbwQQgghhBBCCPF25QJuWPx+03zbKzmkWBwhUkmGD7ppqZ0hIbauGJHaERLEQdnGeb6M6exTO0KCnIn4J7UjJEiRbJlSO0KClMiVObUjJIi9vUrtCAkS9s+z1I6QIL/d/ju1IyTIezmypnaEBMmY3jb2n5mcbONt65kb91M7QoLos6ZP7QgJ8o6NvL5vuvBXakdIsPaf5LGNFyWztPre/unvMzthWvr+whxN0+Yk4iHia4fX/q22sScUQgghhBBCCCHSCPPBemIO2GO7CXhY/J4buP26jWxjak0IIYQQQgghhPj32Ai0Nl+Nvizwj6Zpr13KITPwQgghhBBCCCHSJhv5OGdsSqnlQBXAVSl1ExgGOAJomjYL2ArUAa4Aj4F2CXlcOYAXQgghhBBCCCGSkaZpzV5zvwZ0Tezj2ubpDCGEEEIIIYQQ4j9GZuCFEEIIIYQQQqRNyqYump/iZAZeCCGEEEIIIYSwAXIAL4QQQgghhBBC2ABZQi+EEEIIIYQQIm2y0avQpxSpDSGEEEIIIYQQwgbIAbwQQgghhBBCCGEDZAm9EEIIIYQQQoi0Sa5Cb0Vm4IUQQgghhBBCCBsgB/BCCCGEEEIIIYQNkCX0QgghhBBCCCHSJrkKvRWpDSGEEEIIIYQQwgbIAbwQQgghhBBCCGEDZAm9EK8wa1gLPCuVJPzuA8o0GpOqWc6dOMKqn6ZiNBgoX7MetRu2trr/6P4d7Fz7MwDpM2SgeZdvyJ2/MAC7N6zg4M5NKAU58xakTc/BOKZLnyI5z544zPK532E0GqlYoz51GlnnPLJ/O9vWLgHAyekdWn7VDw9zzscPH7Do+zHcCv4fKGjXcwgFi5ZKkZy/HzvE4lmTMRqMVPX0xrtJW6v7g/ZuY+OqxeacGWjffQB5C77LnbAQfpg4nL/v3UEpxed1fPH0bZYiGQH+PPUru5b8gGY08l4VTz6rb/1cfxw/yIE1C1HKDjt7e2q06oJHkVLcuX2Ddd/7R5f7O+wvKjVswyeeDVIk5+njh1k6ewpGo5HKtepTt3Ebq/sP7dvOltXmds+QgTZd+5GnwLvR9xsNBob1bEs2Fx29R0xJ1myHDgYyafwYjEYjPr4Nadv+S6v7NU1j0vgxHAwKwMnJieGjxlC0WAlCQv5i2OAB3LkTgZ1S+DZsTLMWpv78x6WLjPUfzuPHj8mZMxejxk7E2dk5STkPBgUycdxojAYjPg0a8kWHjnFyThg7moOBppwjRo+lWPESAAwfMoiAgP1kz+7CmvWborfZtWM7s36YwdX//cmS5asoUTJ5x9OxIweZNXU8BqMRz3q+NGnV3ur+68FXmTJ6KFf+uECbjt1p1NzUL8JCQ5g4ajD37prGUR3vhvg2bpGs2Sxd+u0omxZ8j2Y08vHnXlTxtX6uc8eC2LViXvQ4qte2G/mKlSby+TNmD+1BVFQkRoOBUmUrU6PJFymW8/iRg8yaNh6j0Ujtur40jlWfN4KvMmWMuT6/7E5Dc30+f/aMb7q1I/J5JAZDFBWq1qBV+6+SNdvRQ0FMnzwOo9GAl3cDWrbtYHW/pmlMnzyWIwcDSe/kxMBhoylStDjXr11l+KC+0eVu377JFx270bh5K678cZHJ40bx+PFjcuTIybejxpMxOcbRePM48nvJOBpnMY78LcbRtxbjaF3MOJr1w/f8snY12bJlB6Bbj6+pWKlyknJaOn38MD/Pnmzef3pTL979p+n1KH2GDLTt2j/O/nNozzZkc9HRZ8R3yZYrtujxbjCP99axxvs1i/HeKZ7xfucOyk5Rp35DfJsk73g/eiiIaZNM/bOuT/z9c9qkmP45aHhM/xxm2T9v3aR9J1P/nDltEocCDuDg6ECu3B4MHOZPpkyZky3z/04fY4/59b10FU/K1mtqdf/lE4cIWrsQpRTK3p7PW3xF7iIlATi2bS2nD2xDoXD1yEedL7/BIV26ZMuW5shV6K3IAbwQr7Bk0xFmrTzAT6Nav75wCjIaDCyfPYmeI6eRzcWNsX3aU/qTiuTMkz+6jKs+J73HziSjc2bOnjjMzzPHM2DST9y7E86+TasZNnMZ6dKnZ874IRwL3M1nn3ulSM6lsybRe9R0srm44d+7He9/Gjdnv7E/ktE5M2eOH2LxjLEMnjwfgOVzv6PEh2XpMnAsUZGRPH/2NNkzvsi5YOYEBo2dgYurnsHd2/BR2UrkzlsguoybPidDJ87GOVNmfj92kLnTxuA/fSF29g607NiL/IWL8uTxIwZ1a02pDz+12jbZchoN7Fj4Pc0Gjidzdh0Lvu1K4Q8/Q5c7b3SZfCU/pPBHn6GUIuz6//hl+ig6T1qAS04POoydHf0433drSpEyFZI9I5jqc/EPE+k3+nuyu7oxvFdbPihbkVx5YupEp8/JoPE/kjFTZk4dO8SC6eMYNnV+9P07N6wkp0c+njx+lKzZDAYD48eMYubseej1elo3b0ylKlUpULBQdJmDQQHcuB7Muk3bOXvmFGP9R7Jo6Uoc7O35um8/ihYrwaNHj2jVtAGflv2MAgUL4T/iW3r2/oaPynzChnVrWbJwHl269UxSznH+I/lx7nz07npaNGlE5arVKGiRMygwgOvXg9mwdQdnTp9izKgRLFm+CoB6Pr40ad6CbwcNsHrcgoUKM3nqdPxHDHvjbK/KPHPyGMZOnY2rm57uHZpTtkIV8uYvGF0mc+bMdPm6P4cC9llta29vT8fufSlcpBiPHz2iW/umfPhxWattk4vRYGDDvKm0/3YyWbLrmDGwE8XKlEfvkS+6TKGSH1J8UnmUUvwV/CfLpgynz7QlODim48th35E+wzsYoqKY9W03inzwKXneLZHsOQ0GAzOnjGHMd6b67NmhOZ/Gqs9MmTPTuVd/DseqT8d06Rg37ScyvPMOUVGR9O3SljKfVqBYydLJlu27Cf5MmTEXnd6djm2aUKFSVfIViMl25FAgN69fZ9kvWzl/9jRTxo1i9sLl5MmXn/nL1kY/ToM61ahU9XMAJvgP46uefXn/o4/ZsvEXli9ZQIcu3ZOUc9zokfw4xzyOmr5kHAUHs2GLeRz5j2DJMvM48valSbMWfDt4QJzHbtmqDa3bto9ze1KZ9p8T6Dd6Btld3RjWqw0fxrv/nBW9/5w/fSzDpy6Ivn/HhhUpsv+0ZDAYmDlpDGOnmcd7++aUrfiG4/2Lpnz4SfKNd4PBwJTx/nw309Q/v2zdhPKVqpLfsn8eDOTmjessX2fqn5PHjmLOIlP/XGDRP/0s+ufHn5ajU9deODg48OP0Kfy84Ce69OidLJmNRgO7F31P4/7jyZTdlcVDu1How3K45op5fc9b4gMKfVgu+vV94wx/OkyYz4O7EZzcuZ4vxv+EY7r0bPh+FBeO7KNUpVrJkk2kfbKEXohXOHjyT+7+8zi1Y3Dt8nnccuRG554LB0dHPq5YndNHA63KFCxWiozOpjPD+YuU4F5EWPR9RqOByOfPMBiiiHz2lKzZXVMk59VYOT+pVIPfjwZYlSlUrHR0zgJFS3IvIhyAJ48fcfnsb1SsWR8AB0dH3nHOlCI5r1w6h3tOD/Q5cuPg6Ei5KjU4fviAVZl3S7yHs/lMe6Gipbhrrs9sLq7kL1wUgAzvZCSXRz7umv+G5Hb7z0tk0+ckm1tO7B0cKV62CpdPHLQqk84pA8p8Zvr5s6fRP1u6dvY3srnlJItOnyI5//fHefQ5c+OWw9Tun1aqwcnD1u1euHhpMkbXZ0nu3onpn3cjQjl17CCVa3kne7ZzZ0/j4ZGH3Lk9cHRMR83adTiwf69VmQP79lKnnjdKKUqVfp8HD+4TER6Gq86NosVMB2oZM2YkX4GChIWFAhB87SoffvQxAJ+W+4y9e3YlKefZM6fxyJOH3B6mnLU867B/755YOfdQt74pZ+n3TDnDw031+FGZj8mSJUucxy1QsCD58if/ySWASxfOkjO3Bzly5cbR0ZEqn9fmcOB+qzJZs7lQpFhJHBys5wtcXHUULlIMgHcyZsQjbwEiwsNICTeuXMDFPRcu+pw4ODryXvlqnD8eZFUmfYZ3YsbR0ydgHkZKKdJneAcAgyEKgyEqxWaC/ohVn5Wr1+ZI0H6rMi+rT6UUGd4x5YyKiiLKEJWsMS+cO0MujzzkzO2Bo6Mjn9fwJOiA9TgKOrCPWl71UUpRotR7PHzwgIhY+8YTx46QM7cH7jlyAnD9+jXe+7AMAGU+KceBfSkwjvYlfRylpD//OIebxf6zbKWar91/3otn/1klBfafli6djzXeq8cz3rO7UKT42x/vcfpnzfj7Z+06CeifuWL65ydly0f/LSVKlSbcvP9PDn/9eYms+pxkdcuBvYMjxcpW4cqJQ1ZlLF/fI589jd4vgel9XdTzZxgNpvd3ztlcki2bSPvkAF68dUqp9UqpE0qpc0qpjubb2iul/lBK7VdKzVVKzTDfrlNKrVVKHTP/K5+66VPHvTvhZHONOfjK6qrj3p2XHzQe3LWZkh+VAyCbi47qPs0Y1N6X/m3q45TRmeIffJqCOd2if8/m4vbKnEE7N1Hyo7IAhIfcwjlLNhZMHcWInq1ZOH00z54+SbGcLhYHsy6u+ugTCfHZv30D73/8WZzbw0Nuc+3PSxQqmvyzcQAP7kaQ2SWmPjNl1/Hg3p045S4dC2JW33asmjgYr45949x//sg+in9WNUUyAty7E0Z2i/6Z3fXV7X5g50ZKm/snwNLZ39H4i24ou+Q/MAoLC0Pv7h79u5ubnrBQ6zdh4WGhuOtjyuj17oSFWb+5vH3rFpcuXqBkqfcA08z2ixMBu3fuIDTkryTmDEXvnsMqQ+w3i2GhobjHKhP7b3mb7oSHoXOLqTdXNzciwhOfJ+SvW/x5+SJFS6TMx2Xu340gi8U4ypJdx/07EXHKnT0awOSerVg4dgANu/SPvt1oMDCtb3v82/tQuHQZ8hQuniI5I2LXp86NO4moT4PBQNe2jWlWryoflClL0RLJM/v+IpubxRjR6fXRB70xZUKty7jpiYjVh/fu3MbntepE/56/QCGCzLO1+/fsJCw0JEk54x1HscZIWFg84ygBB2Yrli+lsV99hn87iPv//JOknJbu3QnHJYn7zyZfdEfZpexb+jvhYej01v0zrYz38LCwePqedf8MDw/Fzd26D8fun3t2bKO6Rf+0tGXjOj79LPlWsT28F0Gm7Lro3zNld+XBvbj7pT+OB/FTvy9YO3kInh36Rpf9uE5DZvVqwczuTUifISP5S5VJtmxpkrJLm/9SiRzAi9TwhaZpHwFlgB5KqVzAt0BZoAZQ1KLsNOA7TdM+BhoAP8X3gEqpjkqp40qp41ER51I2fWrQ4t4U30wrwKXTJzi0axO+bUyff3z08D6njwbiP3cN4xdu5PnTJxzdtz2FcsYN+rJZoIunTxC4ayMN23YDTG+Sr/95iSp1/Bg2bTHpnTKwbc3iFIoZb4XGW/bc78fZt2Mjzdp3s7r96ZPHfDeqP6079+adjEn7zObLxZcz7k1FPq5A50kLaPj1CAJWL7C6zxAVyeUThyn6afJ9XjO2+Ksz/vq8cOo4ATs30eQLU33+fjSIzFmzk79wsbcWLnY2LZ56tizz+PEj+vXpQZ9vBkR/zn3oiNGsXrGMlk0b8PjxIxwdHZOYM57bYudMRD2/DfGNo8TmefL4MaMG96Fzj2/ImELjKL72jW8clfy0En2mLaFVv9HsWhnz8Q47e3t6TprHwNmruXHlAiHX/5ciOV/SwAne3N7enpkLV7Hkl538ceEs1/53ORmjJWAcvaZMZGQkBwP2U/XzmtG3DRg6inWrl9OhVePUHUfxdQgLjRo3Y9PWXaxYsx5XnY4pk8YnIWQs8QeK1/lTxzmwcyONzfvP344GkilrtpTbf1p43X4yIZ48fsyoQX3o3DO5x/vrx06C+2f1mnHKLZ43G3t7e2p61k161ATmeeHdMhXoMGE+vr2GE7R2IQBPHz3gyonDdJqyhK+mryDy2VPOHdydbNlE2iefgRepoYdSytf8swfQCjigadpdAKXUauDF1VmqA8UtdmqZlVKZNE17YPmAmqbNAeYAZPigW3wv4TYtm6uOexExZ4r/jgiPdxn8zatXWDJjLN2HTcE5s2kZ4MXfj+Oiz0mmLNkA+KBcFf68eIZPq9ZOgZxuVkv3790JI6vFGeYXbly9zKLvx9Bz+HfRObO5upHNVUcB8wVaPipfLcUO4LO7Ws9s3YkIJZtL3PoM/t9l5kz1Z4D/NDJlzhp9e1RUFN+N6k/5arX5pEK1FMkIphn3+xZLJR/cDSdT1pcvk8tTrDT3Zv/F4wf/8E4mU73++fuvuOcrjLO5/VNCdlc37lr0z7sRYfH2z+tXLzNv2hj6jpwa3e5/nD/Fb0cCOH3sEJGRz3jy+BGzJg6j8zcjkiWbm15PaEjMrF5YWCg6NzfrMm7uhFjM/IWGhqDTmfptVGQk/Xr3pHadelSzeGOXL38BZs6eB5iW0wcFWH8E481yxszimzJY59S76wmJXSbW3/I2ubrpCQ+LqbeIsDBcXBOeJyoqklGDe1OtZh0qVKmeEhEB04z7Pxbj6J+74WR+xceIChR/j9Uht3h0/28yWoz7DBkzUaDEB/zx+6+4W3w+ObnEqc/wxNXnC86ZMlP6g485fuQQ+QoUTpZsOje91ex4eGgorq66WGXcrcuEheJi0YePHAqkcNFiZLfY1+bNV4ApM+YCcCP4GoeDrJeOJ1a84yjWGNHrEz+OXFxjMvs1aESPbl2SlNNSNlc37sTaf2aL53Xz+tXLzJ82mj4jp0a/Hl0+f5rfjgTG2n8OpfM3I5Mt3wuuOj3hoW/eP6OiIhk1KGXGe5z+GRaKq866Dt3c3AkLse7DVv3zYCDvxuqfANs2b+BQUABTf/wpWU+YZsqu48HdmJUWD+5G4PyK13ePoqX5O9T0+n79/O9k0bnzjrkfvPtxBW5dPk+J8im3HxVpi8zAi7dKKVUF00F5OU3T3gN+Ay69YhM7c9n3zf9yxT54/y/IW7gYYbdvEhFym6jISI4F7qb0p9ZLue6GhzB77EDafT0Mfa480bdn1+m5eukcz589RdM0Lp46Tg6Lizclp3yFixF6+wbh5py/BuzivU8qWpW5ExbCD2MH0r73MNwtcmbJ5kJ2Vz0hN4MBuHDqGDk98pMSChYpTsit64SF3CIqMpLD+3fxUdlKVmUiwkL4bmQ/un4zghwWF43TNI05U0aR0yMfXg1S7qrZADkLFOFeyC3+DvsLQ1Qk54/sp/BH1kv574bcij6TH3L1MoaoSDI4x1wl99zhlF0+D5D/Xet2Pxqwiw9i1eedsBC+9x9Ap77Dcc8d0+6N23Vl6pLNTF64ni79/SlWukyyHbwDFC9RihvXg7l18yaRkc/ZuX0rlSpb10flKlXZumkDmqZx5vTvODtnwlXnhqZpjBw+hPwFCtCydVurbe7eMX2UwWg0Mm/uLBo0apKknCVKluK6Rc4d27ZSpar1yaHKVaqxeaMp5+lTppyxD/LfpiJFS3Dr5nVCbt8kMjKS/Xu2U7ZCwlZ6aJrGlLHD8chbgAZNU/YiobkLFeXOXze5G/oXUZGRnDq4l+JlrD+NFfHXzehxdOt/f2CIiuKdTFl4+M/fPHlkesmJfPaMK6ePo7PYbyWnd4uW4PaNmPo8sHs7ZcsnrD7/vneXhw/uA/Ds2VN+O34Ej7z5ki1b0eIluXn9OrdvmbLt2bWN8pWsx1GFSlXYsWUjmqZx7swpMjo7Wx3k79mxleo1rZcn37sbM44Wz5+Nd4PGScpZomQprgfHGkdVYo2jqokfR5YfF9i7ZzcFCyXPiRGAAu8WN+8/Ta9HRwJ28kFZ69fNiLAQpvv3p1Nf69ejxu26Mm3JZqYs3MBX/Ueb95/Jf/AOUKRYrPG+O5HjfcxwPPIVoEGz5B/vRYuX5OYNi/65cxsVYvXP8pWrsH1rTP90jtU/d+/YavXxDjBd2X7ponmMnfI9Tk4ZkjVzjliv7xeO7KfQh+WsytwLtXh9v3YZg8H0+p7ZxY3bf14g0vy+Lvjcb7jkTJn9UpqhVNr8l0pkBl68bVmAe5qmPVZKFcW0bH4uUFkplQ14gGmp/Blz+Z1AN2AigFLqfU3Tfn9bYReNbUvFjwrjmtWZK9tHMWrWVhatP/y2nj6avb0DTTr1ZvrwrzEaDXxWvS458xQgYNs6ACp5+rJlxQIePbjP8lmTANOyz0FT5pO/SAk+LF+V0b3aYm9vj0eBd6mQQhe7sbd3oHnnvkwd1hOj0Uj56nXJlbcA+7f9AkAVTz82rZjHo/v/sPTHidE5v/1uIQDNOvVh7uRhREVFotPnol2vISmWs23Xfowd1AOj0UCVmvXxyFeQXZtNV6KtUbcBvyz9iYcP/mH+jPHmnA6MmbGYS+dOEbhnKx75CzGgS3MAmrTrygefJP/lGezs7anZtjsrxg/AaDTyXuXa6HLn4+Ru09cbfVi9HpeOBXImcBd29g44pkuHb/chVhe9uXb2BJ7teyV7Nkv29g606tKXiUN6YDQaqVSzHrnzFmDvFlO7V/PyY/2yeTx88A+Lf5hg+tvs7BkxfVGK5gJwcHDgm4FD6N6lAwajkfo+fhQsVJg1q1YA0LBxU8pXrMzBoAB86tbCycmJYSNNXxl56reTbN28kUKF36V5Y9Oioa+696JCxcrs2L6F1SuWAVD18xrU9/FLcs7+g77lq07tMRqMePs2oGChwqxeacrZqElTKlSqTFBgAPU9a+KUwfR1dy8M+KY3J44d4++/71Hr88p0/qo7vg0asnf3LsaP9efe3bv0+KozRYoW5Yc585KU9QV7Bwe6fj2QQb27YDQYqVnXh3wFCrF5nemK3nV9G3P3TgTd2zfj8aNHKDs71q/6mTlL13H1yh/s2b6Z/AUL06WN6aCtXafufPJZxVc95ZvltHegfvtezB/dF6PRSJmqddB75OfIzg0AlK3pzdmjAZw8sAN78zhq/vUwlFI8+PsOq2aMQTMa0TSNUuWqUOyjuNfDSJacDg506T2QIb27YDAaqenlQ94Chdiy3lSfXj6m+uzRwVSfdnZ2rF/9M7N/Xse9OxFMGj0Eo9GIZjRSsVpNPk3gwX9CODg40KvfIPr26ITRYKBOfV/yFyzEhrUrAfBu0ISy5Stx+GAgzXw9Se+UgYFDR0Vv//TpE47/epi+g6y/DWH3jq2sW2Pq45WqVKdOPV+SInocdY41jszjvVHjplSoWJmggADq16lp+tpIf4tx1C/WOOraHV+/hkybMolLFy+glCJHrlwMGZp8Jxnt7R1o3eUbJgzpgRa9/yzI3i2m16NqXg3YsMz0erToB/PrkZ09I6enzAq1l+Z0cKBr74EM+vo14/0Li/G+8mfmLEv58e7g4MDX3wyiT3dT//Qy98/1a0z906dhE8qVr8SRg4E09fHEySkDA4fF7Z/fDLbun99NGE1k5HN6dzV99WiJkqXj9OE3ZWdvT/XW3Vg9cSCa0UipSrVwzZ2P3/aYXt8/+LwefxwL5GzQbuzt7XFIl576XU2v7zkLFaPIxxVZ9O1X2NnZ45avIO9Vjf+z++LfScX7WVAhUohSKj2wHsiFaeZdBwzHtGS+L3AbuADc1TRtsFLKFZgJFMN0wilA07TOr3oOW1lCv3VF8r0BSEkOqXiRjsTImM4+tSMkyJmI5Lv4UUoqki1lvgEguZXIlXzfyZuS7O1t4ztsw/55ltoREuS323+ndoQEeS9H1tSOkCAZ09vG/jOTk23MO525cT+1IySIPmv61I6QIO/YyOv7pgtJu5jp29T+kzy28aJklqHcgDT53v7J4XGpUo+2sScU/xqapj0DPGPfrpQ6rmnaHKWUA7AO08w7mqZFAElbmyqEEEIIIYSwTTYymfS2SG2ItGK4Uup34CxwFdMsvRBCCCGEEEIIM5mBF2mCpmlxv7xaCCGEEEIIIUQ0OYAXQgghhBBCCJE2peIV39MiWUIvhBBCCCGEEELYADmAF0IIIYQQQgghbIAsoRdCCCGEEEIIkTbJVeitSG0IIYQQQgghhBA2QA7ghRBCCCGEEEIIGyBL6IUQQgghhBBCpE1yFXorMgMvhBBCCCGEEELYADmAF0IIIYQQQgghbIAsoRdCCCGEEEIIkTbJVeitSG0IIYQQQgghhBA2QA7ghRBCCCGEEEIIGyBL6IUQQgghhBBCpE2yhN6K1IYQQgghhBBCCGEDZAZe/OtsXTEitSMkSJ2mw1I7QoJc3js5tSMkyKNnhtSOkCBlc7ukdoQESe9oG+d37exs47th7WzkO2z1WZxSO0KCfKiypXaEBNFlSpfaERLEqKV2goSxlfFe2N05tSMkiEGzjYZP72Abr0dV8ulSO4L4j5ADeCGEEEIIIYQQaZONnLx7W2zjlJYQQgghhBBCCPEfJwfwQgghhBBCCCGEDZAl9EIIIYQQQggh0ia5Cr0VqQ0hhBBCCCGEEMIGyAG8EEIIIYQQQghhA2QJvRBCCCGEEEKItMlGvor1bZEZeCGEEEIIIYQQwgbIAbwQQgghhBBCCGEDZAm9EEIIIYQQQoi0Sa5Cb0VqQwghhBBCCCGEsAFyAC+EEEIIIYQQQtgAWUIvhBBCCCGEECJtkqvQW5EZeCGEEEIIIYQQwgbIAbwQQgghhBBCCGEDZAm9EEIIIYQQQoi0Sa5Cb0VqQwghhBBCCCGEsAFyAC/+886dOMKwLk35tmMjtq9ZHOf+o/t3MKp7K0Z1b8WEfh25efVy9H27N6xgRNcWjOzWgp8mDiXy+bO3GT3arGEtCN4zluOrB6XK87/Mr4eDaNO4Hq0aerF88bw491+/dpVuHVpSu+JHrFq68K1mO3H0IF1a+tCxeX3WLJ0f5/6bwVf5pktr/Kp/wroV1v3i4YMHjBvaly6tfPmqlR8Xz576z+c8duQg7ZvWp22juqx8SVv3+rIVdSuXYfWyRdG3h4WG8E239nRo5sOXLXxZt3Jpsmc7GBSIb73a1K9TkwU/zYlzv6ZpTBjrT/06NWnsV58L589F3zf820F8XvkzGvnWs9qmf9+vadrQh6YNffCqVY2mDX0SnyswgPpetahbuwbz5safa9wYf+rWrkFD33pWuV637aIF83ivRBHu3bsLwK1bN/nkw9I09vOmsZ83o0YMTVzWoAC869ainmcN5r+kDseP8aeeZw0axc76km1/nPk9NapVpHEDbxo38CYw4AAAf/99jw7tWlHu4w8YO3pkonK+zLEjQXzRtB5tG3mx4iX9s+eXLfGq/BGrly2Mvv1F/2zfzNvcP39OljyWDgYF4mPuny+t21f0z2qVP6NhrP4JsHzpEnzq1aaBT12mTpmY5JyHggLxq1cbb6+aLJj3knE0zh9vr5o0aRCTMyTkLzq2b00D7zo08q3Lsp9j9lO7dm6nkW9dyrxXjPPnzrxxtpQYS//8/TedOrSjnmdNOnVox/1//gEg8vlzvh08kAY+9WjkW59jvx59o8yHDwbSxLcODevXYvGCufFmnjJhNA3r16JlYx8uXTgffZ+vV3VaNPamdVNf2rVoFH37T7NmUK9WFVo39aV1U18OBR14o2yWjhwKpJmfF028a7PkJTmnThhDE+/atGnia5XzwYP7DOnXi+Z+dWnRoB5nT/8OwLzZM/GpXZW2zfxo28yPw0EBSc556GAgDep74lu3FgvnxZ9z0rjR+NatRbOG3ly8YOoDz549o03zxjRv5ENj37rM/uH76G1+nDGNZg29ad7Yl26d2hMeFpbknJaOHz3Il829ad+0Hqt+jvv6fiP4Kr07t6Z+tY9Zu3xRnPsNBgPdvmjCsH7dkzWXSPtkCb2NUkrlAzZrmlbyLT3f+0BOTdO2JvPjjgQCNE3bnZyPm1BGg4HlsyfRc+Q0srm4MbZPe0p/UpGcefJHl3HV56T32JlkdM7M2ROH+XnmeAZM+ol7d8LZt2k1w2YuI1369MwZP4Rjgbv57HOvt/53LNl0hFkrD/DTqNZv/blfxmAwMH3SGCZMn4POTc9X7ZpRrmIV8uUvGF0mU+bMdOs9gIMH9r71bLOnjmPk5B9x0enp06kFn5SvTJ58MdmcM2ehY4/+HAnaF2f7ud9P4MNPPmPAyElERkby7OnT/3zOmZPGMHbabFzd9HRv35yyFauQ16KtM2fOTJev+3MowDqnvb09Hbv3pXCRYjx+9IhuXzTlw0/KWm2b1GzjR4/khznz0bvradm0EZWrVqNAwULRZQ4GBnA9OJgNW3Zw5vQpxvqPYPGyVQDU8/alSbMWDB08wOpxx0/6LvrnKRPH4eycKdG5xoweyey5C9Dr9TRv0pAqVatRsFBMrqDAAK4HX2PTtp2cOX0K/5HDWbpi9Wu3DfnrLw4fOkSOHDmtnjO3Rx5W/bIhUTlfZB3rP5JZcxegd9fToklDKletRsGCsbJev8bGraaso0cN5+flq1+7bctWbWnTrr3V86VPl56u3Xty5fJlrly5TFIZDAZmTBrDuGlzzP3TtC/KG2tf9NXXAzgUYL0vMvXPPhQuUpzHjx7R9YumfPhJuWTtn+NGj+RHc/9sYe6fcerWon+O8R/Bklj989tY/fPYr0fYv28vq9ZuJF26dNy9cyfpOceYx5FeT6tmjahcJdY4CgrgRnAw6zfv4KzFOLK3t+frPv0pVrwEjx49pGXTBpQt9xkFChaiUKHCTJwynTGjhiUpW0qMpfk/zeGTT8vR/suOzJs7h3k/zeHrPt+wds1qANau38SdO3fo2vlLlq1cg51dwufEDAYDk8f7M+2Hn3DT6/miZRMqVq5K/gIxmQ8fDODG9WBWb9jOuTOnmTB2BPMWr4y+f+bshWTNli3OYzdt0ZoWrb94g5qMP+eUcaP57oe5uOn1dGjVhAqxch45GMiNG8GsWL+Nc2dPM2nsSOYuXgHAtIlj+bRcBfwnTCUy8jlPLV6DGjdvTfPW7ZIt54Qxo5gxex56vZ42zRtTqUpVq/55KCiA69eD+WXTds6eOcU4/5EsXLqSdOnS8eNPC3jnnYxERUbSoW1LPqtQkVKl36dV2/Z06dYTgBVLl/DT7B8Y+O3wZMv8w5SxjP5uFq46Pb2+bEHZ8pXJY7VfykLnnv04HBj39R1gw+pleOTNz+NHj5IlU5omV6G3IjPw4rWUUg7A+0Cd5H5sTdOGptbBO8C1y+dxy5EbnXsuHBwd+bhidU4fDbQqU7BYKTI6ZwYgf5ES3IuIOQNrNBqIfP4MgyGKyGdPyZrd9a3mf+HgyT+5+8/jVHnul7l4/iy5cuchZ67cODo6UrVG7TgHb9myu1C0eEkcHN7uucTLF86SI5cH7jlN2SpWq8XRoP1WZbJmy07hYiWwj5Xt8aOHnDt1khpevgA4OjrinClxB2//tpyXzp8lZ24Pcpjbukr12hwOjJUzuwtF4mlrF1cdhYsUA+CdjBnxyFuAiPDkm+U4e+Y0ufPkIbeHB46O6ajlWYf9+/ZYldm/bw9163ujlKL0e+/z4MF9ws0ZPirzMVmyZHnp42uaxq4d26ldJ3En7s6eOY2HR15TrnTpqF3HK06ufXv3UK++T5xcr9t24vixfN3nG1QyveE5e+Y0HnnyWtShF/v3xleHL8n6mm1jy/DOO3zwYRnSpU+fLPlN/TNPdP+sXL02hwLj7ouKFC8ZZxyZ+mdxwNQ/8+TNn+z90+M1/fPAG/TP1StX0K79l6RLlw6A7C4uScp57qw5Z25Tzpq148/pVc+Us9R77/PQnFOnc6NY8RIAZMzoTP78BQkLCwUgf4GC5MtfIEnZUmos7du3h/o+PgDU9/Fh317TW5X//XmFT8uWBcDFxYVMmTJx7uzZRGU+f/YMuXPnIZe5PqvX8iRgv/XJo4D9e/Gsa6rPkqXf4+GDB0SEh79JFb2xC+fOkNvDIyZnzToE7bceO4EH9lLbq74pZ6n3ePjQlPPRw4ec+u0EdX0aAODomI5MmTKnSM5zZ0/j4RHTP2vUrsOBWPV5YN/emP5Z2tQHIsLDUErxzjsZAYiKiiIqKhKFad/p7Owcvf2Tp0+S9RjyjwtnyZnLgxzm1/dKn9ficDyv7+8Wi7tfAogIC+XY4UBq1fVLvlDCZsgBvG2zV0rNVUqdU0rtVEqVUEqdfHGnUqqwUuqE+edrSqnxSqlfzf8KmW/XKaXWKqWOmf+VN98+XCk1Rym1E1gMjASaKKV+V0o1UUplVErNN2/zm1LK27xdW6XUL0qp7Uqpy0qpCebb7ZVSC5VSZ5VSZ5RSX5tvX6iUamj++XPzY50xP3Z6i+wjlFInzfcVTa4KvHcnnGyu+ujfs7rquHfn5S+QB3dtpuRH5QDI5qKjuk8zBrX3pX+b+jhldKb4B58mVzSbFxEeis4tpm51bvpkfeObFHciwnC1yOaq03MnImFvjEJu3yJL1mxMGzeMnu2b8v2EETx98uS/nTM8DJ3e3SKnGxHhoYl+nJC/bvHn5YsULVEq2bKFh4Xi7p4j+nc3vTthodbZwsJC0ccqEx6WsPwnTxwnu4sLefLmS1SusNBQ3HPE1JmbXk9ovLliyujN2V+17f69e3DTu1GkaNzd5K1bN2ncwIcv2rTk5InjCc8aFoq7VQ599AGY1d8TX9bXbLti+VIa+dZj2JCB0UuUk1tEeCg6vcW+SKfnzhvsi0L+usWVZO6fsfueXu9OeDz9wD1Wmdj1H1tw8DV+O3mcVs0b075tS86dffPl6WBqX70+Vs7YfSAB4+j2rZtcvHiBkqXeS1Ke2NlSYizdvXMHnc4NAJ3Ojbt3TR9HebdIUfbv3UNUVBQ3b97gwvlzhIb8lajM4eGhuFnkcXNzj7M8OzwsDL3FflXnpifcvF9VStGzawfaNm/I+rWrrLZbs3IZLRv74D98MPfvJ21MhYeF4mbR7jp9TIYXIsLCcNNb/i16IsJDuX3rBlmzZWPM8MG0a96AcSOH8uRJzETDL6uW0aaJL2NGDEmGnGHW7eumjzOOwsNCrerTTe9OmLnODQYDzRv7UrNqBT4t+xklS8f0zx++n4pXzaps37KJTl/1SFJOS3fCw3B1s3zd1HMnIuH7pdnTJ/LFV72ws5OZ6f8iOYC3bYWBmZqmlQD+Bj4A/jEvdwdoByy0KH9f07RPgBnAVPNt04DvNE37GGgA/GRR/iPAW9O05sBQYKWmae9rmrYSGAzsNW9XFZiolMpo3u59oAlQCtNBv4f5tlyappXUNK0UsMDyD1FKOZmzNjHf7wB0sSgSoWnah8CPQN+EV9FraHFvetms1aXTJzi0axO+bb4C4NHD+5w+Goj/3DWMX7iR50+fcHTf9mSLZvPiq1vSxguNFm+2hDEYovjz8kU8vRsxbd4KnJwysGZZ3M+uJQebyRlPYyd29vfJ48eMGtSHzj2/IWNG59dvkEDx1mHsbPGUSWhN79i2JdGz76anTECdxRNeKfXSbZ88ecLcObP4yrzk05JO58aO3ftYtXY9ffsNYEC/Pjx8+DBhWV+SIyFlXrVt4ybN2LxtFyvXbsBV58bkieMSlCc5vEn/HDmoN1169kvW/hlv34tTt/EUeU3/NBgM3L9/n8VLV/J1n37069sr3rZIWswE5LQo8/jxI77p3YO+/QZazWwmVUqMpVfx8WuAXu9O88YNmDhuDO+9/wH2DvaJyxxvnjil4pYxt/vsBUtZtGwtU2bMZu2q5fxmPiHn16gpazbuYPGKX3B11TF9yoRE5YqbM+5tCRn7KIXBYOCPixfwadiUBcvW4pQhAz8vML3F9G3YhJUbtrNg+VpcXHXM+C5p12hI0D7qFW1tb2/PslXr2LJzH+fOnuHK5T+iy3zVvRdbdu6jtlc9Vq1Ivmu0xJsnga87Rw8GkDVbtujVQf8Jyi5t/kslcgBv265qmva7+ecTQD5MB+DtlFL2mA6il1mUX27xfznzz9WBGUqp34GNQGal1It1ths1TXvZlF1NYIB5u/2AE5DHfN8eTdP+0TTtKXAeyAv8DyiglPpeKVUbuB/r8YqY/54Xe81FQCWL+3+J9XdaUUp1VEodV0od37wy7oU+Xiabq457ETFnaf+OCI93GfzNq1dYMmMsXQaPxzmzabnixd+P46LPSaYs2bB3cOCDclX482LSZjn+TVzd9FazL+FhobjodKmYKIarzo0Ii2wR4aFkd01YNledHledG0WKm2bhPqtcnf/9cfE/nlNPeGiIRc4wXFzdErx9VFQkowb1plrNOlSoUj1Zs7np9YRYzI6FhYagc3OLUyb0NWXiExUVxd7du6hZK/GfLtLr3Qn5K6bOwkJDcYuTy53QkJgyoeZcL9v25o3rpll2P288a1QjNDSEpg39iAgPJ126dGTNavq8bPESJfHwyEPwtasJz2qVIzR6ZjK6jHvsMhZZX7Kti6sr9vb22NnZ4dewEWeTOEv8Mqb+abEvSsQ4AlP/HDmoN9VqeqVI/7Tse6Hx9D19rD4cX5nY9Ho9n1evYV7WXBo7Zce9e/feOKderyc01DqDa+w+EM84elEmMjKSb3r3wNOrHtWq13zjHPFnS/6xBKaPHbz4qEJ4eBjZs2cHwMHBgW8GDGLVLxuYNuNHHjx4QJ48+RKV2c3NnTCLPGFhcetT56Yn1GK/Gh4WGl3mxRjKnt2FylU/5/y50+bMMWPK268RF5JwYUAw9c8wi3YPDw3FNda+XafXExZq+beYyujc9Ojc9JQoVRqAqtVr8sfFC3Fy1vdtmCw5rdo3LBTX2H3Azd2qPsNCQ9DFek+SKXNmPvr4Ew4fCorzHLU9vdi7e2eSclpy1emJCLN83Uz4fun8md85cvAAbRt5Mn74AE6fPMbEkWnrIsYiZckBvG2zvOS5AdOs9VrAE6gLnNA0zfLKNVo8P9sB5cwz6+9rmpZL07QH5vtedVUMBTSw2C6PpmkXXpZL07R7wHuYDva7Yj3T/+LxXuXFY774O61omjZH07QymqaVqdukzWseKkbewsUIu32TiJDbREVGcixwN6U/rWBV5m54CLPHDqTd18PQ58oTfXt2nZ6rl87x/NlTNE3j4qnj5PDIl+Dn/rcrWqwEt24E89ftm0RGRrJv13Y+q1gltWMBULhoCW7fvE7IX7eIjIwkcO8OPi1fJUHbZnNxxVXnzs3r1wA4dfJXPPIl7TOctp6zSLES3Lp5nRBzW+/fvZ2yFSonaFtN05gyZjge+QrQoFnyX4SxRMlS3AgO5tbNm0RGPmfHtq1UrlLNqkzlqtXYvHEDmqZx+tTvODtninOAGp+jRw6TL39+q6Wbicl1/fo1bt68QeTz52zfuoXKVa1zValajU0b18fJ9bJtC79bhP2Bh9m2ay/bdu1Fr3dnxZpfcNXpuHv3LgaDAYCbN24QHHyN3Lk9EpX11s0b5jqMm7VylWpsfkXW+LYNt1jGvnfPbgoVKpzoekwIU/+M2Rcd2L2dchWqJGhbU/8cRp58+WmYQv3zeqz+WSUZ+meVatX59ajp6ujB164SGRlJtngueJZQxUtYj6Od2+OOo0pVqrFlkynnmVO/45zJlFPTNEYNG0L+/AVpmUwXLbOUEmPpxTYb168HYOP69VSt+jkAT5484fFj01Lww4cOYm9vb3XBvIQoVqIkN24Ec/uWqT5379hGxcpVrcpUrFyNbZtN9Xn29CkyOmfCVafjyZPHPDJftOzJk8ccPXKIAgVNY8fyM/L79+6Ovv1NFS1ekhs3rsfk3LmV8rFyVqhUle1bNppynjmFs7MzrjodLq463PTuXDefKDz+6xHyFSgYJ2fAvqTnLF6iFNevx/TPXdu3UilWzkpVqsb0z9OmPuCqc+Pe3bs8uG+aU3r69Cm/HjlMvnymCxlfD74Wk3P/viRfr8HSuy9e32+bXt8D9uxI8Otmu849WPLLThau3kb/4eMo/eHHfDN0TLJlE2mfXIX+X0bTtKdKqR2Ylpq3j3V3E2Cc+f/D5tt2At2AiWC62rzFrL6lB4DlFbB2AN2VUt01TdOUUh9omvbby3IppVyB55qmrVVK/Yn10n6Ai0A+pVQhTdOuAK2ApH//yWvY2zvQpFNvpg//GqPRwGfV65IzTwECtq0DoJKnL1tWLODRg/ssnzUJADt7ewZNmU/+IiX4sHxVRvdqi729PR4F3qVCLe+UjhyvRWPbUvGjwrhmdebK9lGMmrWVResPv37DFGTv4ED3voPo37MLRqMBz7o+5CtQiE2/mK+e7NeYu3ci6NK2KY8fPULZ2bF2xc/MX7E+eZeoviRbp179Gd73K4xGI9XreJMnf0G2bTBdXdjTuxH37kTQu1MLHj96hJ2dYuOapcxctJZ3MjrTsWd/pvgPIjIyCvecueg5YMR/PmfX3gMZ9HUXjAYjNc1tvXmdqa3r+prauvsXzaLbev3Kn5mzbB1Xr/zBnu2byV+wMF3aNAagXafufPJZxWTJ5uDgQP9B39K1c3uMBiP1fRtQsFBh1qwyXSW5YeOmVKhYmaCAALzr1MTJyYnh/jFvhAb2682JY8f4++971P68Mp27dsfHryEAO7dtoXadum+ca+DgoXTp2AGj0YCPbwMKFSrMqpWmhVKNmzSjYqXKBAUcoK5nDZycMjDSnOtl277KyePHmDljOg729tjZ2zNk6AiyZM2a4KwDBg2lS6cOGA0GvM3Pt9qctdGLrIEHqOdZA6cMGRgxaswrtwWYOnkily5dRAE5c+ViyLCYr4zzrFmNRw8fmk7+7d3Nj3PmW12ZPTHsHRzo1nuQuX8aqPWS/tnti5h90bqVPzN32XquXvmD3eb+2bmN6eu6vujUI9n751fm/ult7p+rzf2zkUX/rB9P/xxg0T9rmfunr19DfHz9GP7tYBr61sPR0ZGRo8cl6aKGDg4O9Bv0Ld26tMdgMOLtE/84OhgYgLeXOae5D/z+20m2bN5AocLv0qyRDwBde3xNhYqV2btnFxPH+nPv3l16du3Mu0WLMnNW3K/5e122lBhLX3ToyDe9e7H+lzW458jBpCnTALh79w5dOrbHzs4ONzc9o8clfpm6g4MDffoPplfXLzEajdSt70uBgoX5ZY2pPv0aNuWzCpU4FBRAI+/apHdyYsjw0abnv3OHAX1Mn8U2GKKoWduLcuVN/XHmtEn88cdFFIocOXPRf/DwRGeLnbN3v8H07tYRo8GIl7cvBQoWYv0a09XwfRo2oVyFShw+GEATb0+cnJwYNNw/evuv+w1ixJD+REVGkjNXbgaa7/tx+mQuX7qIUgr3nDn5ZlDSc/YbOIQeXTpgMBqp7+NHwUKFWWvunw0aN6V8xcocDArAt24tnJycGDrS1AciIsIZPmQgRqPB9Dpbs3b0yZQZ06YQfO0qdnZ2uOfIycAhSctpyd7BgS5fD2BIny4YjUZqenmTN38htqw3vb57+TTi7p0Ien7ZPPr1ff3qpcxe8gvvpPB7pDRJrkJvRSXlM1Ei9cT+GjmlVF/AWdO04Uqpsphm4vNommYw338N0+fO62CadW+madoV84H1TKAYphM6AZqmdVZKDQceapo2ybx9dkwH7Y7AWEzL7acCn2GaPb+maVpdpVRboIymad3M220GJgH3zM//YtXHQE3TtimlFpr/jjVKqc/NZR2AY0AXTdOembOX0TQtQilVBpikaVqVl9XNvkt3bKJT12n65l+b8zZd3js5tSMkyKNnhtSO8K+S3tE2FmjpMiXPlcpTmr2NXGjIVt4ShN1/9vpCaYAuU7rUjpAgRhtpdwd72xhHj23k9chgIwM+vYNtvB6F28h+CaCgWwbbGExmGTy/S5Od9cm2r1OlHmUG3kZpmnYNKGnx+ySLuysA818cvFuYqWma1RScpmkRmGbkYz/+8Fi/3wU+jlWsUzzbLcRidl3TNMvpqQ/jKd/W4uc9mC7EF7tMPoufjwNVYpcRQgghhBBC/Aul4gXj0iI5gP+XUUqtAwoC1V5XVgghhBBCCCGE7ZAD+H8ZTdN8X3J7vrccRQghhBBCCCFEMpIDeCGEEEIIIYQQaZNcxM6KfKBACCGEEEIIIYSwAXIAL4QQQgghhBBC2ABZQi+EEEIIIYQQIm2Sq9BbkdoQQgghhBBCCCFsgBzACyGEEEIIIYQQNkCW0AshhBBCCCGESJtkCb0VqQ0hhBBCCCGEEMIGyAG8EEIIIYQQQghhA2QJvRBCCCGEEEKItEmp1E6QpsgMvBBCCCGEEEIIYQPkAF4IIYQQQgghhLABsoReCCGEEEIIIUTaJFehtyK1IYQQQgghhBBC2AA5gBdCCCGEEEIIIWyALKEX/zoONrLM5vLeyakdIUEKV+uT2hES5PdtE1I7QoK4ZkqX2hES5P6TqNSOkCAPbCRn1oyOqR0hQWzlQr/pHWxjP69spELt0FI7wr+KQbON+szgaJ/aERIkymhM7QgJkvkd29jP2yQb2Ze+LbbxCiiEEEIIIYQQQvzHyQG8EEIIIYQQQghhA2QJvRBCCCGEEEKItMlGPh77tkhtCCGEEEIIIYQQNkAO4IUQQgghhBBCCBsgS+iFEEIIIYQQQqRNchV6KzIDL4QQQgghhBBC2AA5gBdCCCGEEEIIIWyALKEXQgghhBBCCJEmKVlCb0Vm4IUQQgghhBBCCBsgB/BCCCGEEEIIIYQNkCX0QgghhBBCCCHSJFlCb01m4IUQQgghhBBCCBsgB/BCCCGEEEIIIYQNkCX0QgghhBBCCCHSJllBb0Vm4IUQQgghhBBCCBsgB/BCCCGEEEIIIYQNkAN48Z939sRhBnduzMCODdm6enGc+4/s386w7i0Y1r0FY7/5khtXL0ff9/jhA34cO5AhnZswpEsT/rx45q1k/vVwEG0a16NVQy+WL54X5/7r167SrUNLalf8iFVLF76VTAkxa1gLgveM5fjqQakdhRNHD9KlpQ8dm9dnzdL5ce6/GXyVb7q0xq/6J6xbYd0vHj54wLihfenSypevWvlx8eypZM125GAgTX29aFS/NosXzI1zv6ZpTJkwhkb1a9OqsS+XLpyPvs/PqwYtG/vQpqkfX7RobLXd6hVLaerrRYuG9Zk5dVKyZj5+5CDtm9anXeO6rFwSt0/eCL5Kr46tqFelDGuWLYq+/fmzZ/To0JwubRrRsYUvS376IVlzARw9HETLhnVp7ufJ0kU/xblf0zSmTRpDcz9P2jX35Y+LMfW5atli2jTxpm1TH0YM+YZnz55Zbbvi5wVU/qQkf/99L9G5DgYGUN+rFnVr12De3Dnx5ho3xp+6tWvQ0LceF86fe+22M6ZPpaFvPRr7edPpyy8ICwu1esy/bt+mbJkPWLQgbhu97aw7d2zDt74X75csyrmzMfvOW7du8smHpWns501jP29GjRiaqKwvpES7Dx/Uh/YtGtC+RQOaeNekfYsGb5TtYFAA3nVrUc+zBvN/ir8+x4/xp55nDRrFrs+XbPvjzO+pUa0ijRt407iBN4EBBwD4++97dGjXinIff8DY0SPfKK/peQPxqVeb+nVqvjzzWH/q16lJY7/6VpmHfzuIapU/o6FvvTjbLV+6BJ96tWngU5epUya+Wba3OJYOHzpI00Z+NPCpR9NGfhw9cviNMh85FEgzPy+aeNdmyUv281MnjKGJd23aNLHezz94cJ8h/XrR3K8uLRrU4+zp36PvW7NiKc38vGjZqD4/TEv6fv5QUCB+9Wrj7VWTBfPir9sJ4/zx9qpJkwYx7R4S8hcd27emgXcdGvnWZdnPMa+ju3Zup5FvXcq8V4zz55LnfdPhg4E08q5Dg3q1WDQ//vqcPH40DerVokUjHy5a1CeAwWCgVRM/enfvEn3brJnTadHIh5aNfeneuQPhYWFJzmkr7Z7WKKXS5L/UIp+BF/9pRoOBpbMm0XvUdLK5uOHfux3vf1qRnHnyR5dx1eek39gfyeicmTPHD7F4xlgGTzYd8C2f+x0lPixLl4FjiYqM5Pmzpyme2WAwMH3SGCZMn4POTc9X7ZpRrmIV8uUvGF0mU+bMdOs9gIMH9qZ4nsRYsukIs1Ye4KdRrVM1h8FgYPbUcYyc/CMuOj19OrXgk/KVyZMvpg6dM2ehY4/+HAnaF2f7ud9P4MNPPmPAyElERkby7GnytbvBYGDS+NFM+2Eubno97Vs2oWLlquQvUCi6zOGDgdy8HsyqDds4d+Y0E8eO5KfFK6LvnzF7AVmzZbN63BPHjhK4fy+LV64jXbp03L17J1kzz5w8hjFTZ+PqpqdHh+aUrVCFvLH6ZJev+3M4wLo+HdOlY/z0n8jwzjtERUXSp0tbypStQLGSpZMt29QJ/kyeMRedmzud2jShfMWq5CsQk+3ooUBu3rjO0rVbOX/2NFPGj2LWguWEh4WyduVSFq/cQHonJ4YN7MPeXdvwrOsDQFjoXxw/ehi9e443yjVm9Ehmz12AXq+neZOGVKlajYKFYto5KDCA68HX2LRtJ2dOn8J/5HCWrlj9ym3bftGBbj16AbD058XM/nEm3w6LOWibOH4sFSpWTBNZCxV6l++mfc+oEcPiPGdujzys+mVDonLGzpwS7T58zOTo7WdOnUhGZ+c3yjbWfySz5i5A766nRZOGVK5ajYIFY9Xn9Wts3Gqqz9GjhvPz8tWv3bZlq7a0adfe6vnSp0tP1+49uXL5MleuXOZNGAwGxo0eyY9z5puet2mj+DMHB7Nhyw7OnD7FGP8RLFm2CoB63r40adaCbwcPsHrcY78eYf++vaxau9G0X7qT+P3S2x5LWbNlY/rMH3Fz03P58h906die3fsCE515yrjRfGfez3do1YQKsfbzRw4GcuNGMCvWb+Pc2dNMGjuSueb9/LSJY/m0XAX8J0wlMvI5T82vQSePHSXwwF4WrTDt5+8lcT9vMBgYN2YkP8yZj16vp1WzRlSuUo0CFu1+MCiAG8HBrN+8g7OnTzHWfwSLl63C3t6er/v0p1jxEjx69JCWTRtQttxnFChYiEKFCjNxynTGjIo79t8058Sx/nw/6yfc9HratjC9blrmPBQUwI3rwazZuJ2zZ04zYfQI5v+8Mvr+lcuWkC9/QR49ehh9W8s2X9C5a4/o++fN+YEBQ4YnKacttLtI+2QG3gYopfIppc6mwOO+r5Sqk8htrimlXM0/H0ruTG/b1cvnccuRG517LhwcHfmkUg1+PxpgVaZQsdJkdM4MQIGiJbkXEQ7Ak8ePuHz2NyrWrA+Ag6Mj7zhnSvHMF8+fJVfuPOTMlRtHR0eq1qjNoVgHRdmyu1C0eEkcHNLWObqDJ//k7j+PUzsGly+cJUcuD9xzmuqwYrVaHA3ab1Uma7bsFC5WAvtYdfj40UPOnTpJDS9fABwdHXHOlHztfv7sGXLn9iBXbg8cHdNRvVYdAvdbt2/g/r3UrlsfpRQlS7/HwwcPiAgPf+XjrluzklbtOpAuXToAsmd3SbbMly6cJUduD3KY+2Tlz2tzOHC/VZms2VwoUqxknPpUSpHhnXcAiIqKIioqiuQ8qX3h3BnzePHA0dGRajU9CQqwPrEVFLCPWnVM9VmilKk+75jHucEQxbNnz4iKiuLZ0ye4uuqit5vx3QQ6d+/9Rmfhz545jYdHXnJ7eOCYLh2163ixf98eqzL79u6hXn0flFKUfu99Hjy4T3h42Cu3dbY4oHz65IlVtr17dpPbIzcFCxVOE1kLFCxIvvwFEpUloVKy3cE0S7Zv93aq10zUSyhgrs885jpxTEctTy/277Wuz/379lD3ZfX5mm1jy/DOO3zwYRnSpU+f6KzWmfNYPG+dOH3gwL491K3vHSczwEdlPiZLlixxHnf1yhW0a/9lzH7JJfH7pbc9looVK46bmx6AQoUK8/zZc54/f56ozBfOnSG3h8V+vmYdgmLv5w/spbaXeT9f6j0ePjTt5x89fMip305Q18e0+sPRMR2ZMpneo6xbs5KWbWP289mSuJ8/d9bc7uacNWvH3+5e9UztXuq993lorludzo1ixUsAkDGjM/nzF4xexZC/QPKO/fNnz5DbI090fdao5UnAfuvxHrB/L551zTlLv8cDi9fN0NAQDgYewNvPekWNZR94Emt/+iZspd1F2icH8P9RSikH4H0g8e8+zDRN+yzZAqWSe3fCyebqFv17Nhc37t15+YFQ0M5NlPyoLADhIbdwzpKNBVNHMaJnaxZOH82zp09SPHNEeCg685sHAJ2bnojwpC/r+i+5ExGGq0Uduur00W/cXyfk9i2yZM3GtHHD6Nm+Kd9PGMHTJ8nX7uHhoVYzujo3PeGxlkGHh4Wh17tblwk3lVFK0avrl7Rr3oj1a1dFl7kRfI1TJ0/QoXVTvurQJtmWLQLcCQ9D5xaTx9XNjTvhoa/YwprBYOCrNo1pWrcqH35clqIlkmf2HSAiPAy3WHUVe7xEhIXGKRMeZhpnTVu2pXH96vjVqUpG50x8XLY8AAcD9uGqc6PQu0XfKFdYaCjuOWKe002vJzTUus7CwkLRu8eU0evdCQsNfe2230/7jpqfV2bL5k181a0nAI8fP2bBvLl07tItTWV9mVu3btK4gQ9ftGnJyRPHE505pdr9hdO/nSB7dhdy58mb6GxhYaG4W9WVPs5HHcJCY5cx1+drtl2xfCmNfOsxbMhA7v/zT6KzvSqz5X5Jr3cnPJ4+4B6rTOy/K7bg4Gv8dvI4rZo3pn3bllYfpUhwtrc8lizt3rmDosWKRR84JVR4WChueov9vD5mH/5CRJh1H3Zz0xMRHsrtWzfImi0bY4YPpl3zBowbOZQnT0wnxm9cv8bp307wZeumdPuyDReSuJ8PCw1Fr4/V7rH7aqy+4RZPmdu3bnLx4gVKlnovSXlemjNW+5oyWI/38LCwWGViXlu/mziObr36olTcw6Ifv59KvVrV2LF1Mx27dE9STltp97QotZfKp7Ul9HIAbzvslVJzlVLnlFI7lVIZlFIFlVLblVInlFKBSqmiAEqpekqpo0qp35RSu5VSevPtw5VSc5RSO4HFwEigiVLqd6VUk/ieVCnlYn6+35RSs7H4Igel1EPz/zmUUgHmxzmrlKpovr2mUuqwUuqkUmq1UsrZfPtQpdQxc9k5yjwClFI9lFLnlVKnlVIrzLdlVErNN5f/TSnlnay1qmnx/M3xF714+gSBuzbSsK3pDbDRYOD6n5eoUsePYdMWk94pA9vWxP0MfbKLGxkl36+RKPE0e4Jr0GCI4s/LF/H0bsS0eStwcsrAmmVxP0P/5uHiyRarU2rxFHrRB2Yt+JmFy9YwecYsflm1nN/MBz9RBgP3H9xn7qLldOvVh2/790GLryLeJHK84yjhfdLe3p4fFq3i53U7uXT+LNf+92bLfBOaLXZrx1ufSvHg/j8EHdjHivU7+GXrXp4+ecLObZt4+vQJSxbM4YtOiT8Yft1zxgofb5nXbdu959fs3HMAr7r1WLHsZ8D0+eiWrdvwTsaMaSprfHQ6N3bs3seqtevp228AA/r14eHDh6/cJk7mFGh3S7t3buXzWm92/jsh4+VlZV61beMmzdi8bRcr127AVefG5Inj3ihfvOKtztiZ4ynymj2rwWDg/v37LF66kq/79KNf316J3i+97bH0wpUrl5n63SSrj6gkOHNC9vPxF8JgMPDHxQv4NGzKgmVrccqQgZ8XmK7xYDAYeHD/PnMWLeernn0YOiBp+/n4mz0B7W5R5vHjR3zTuwd9+w20mtFOVgl4L/ey+gwK2E/2bNmjVwvE1qV7Lzbt2EutOnVZvWJpcsdMk+0u0j45gLcdhYGZmqaVAP4GGgBzgO6apn0E9AVeXP0pCCiradoHwAqgn8XjfAR4a5rWHBgKrNQ07X1N01YSv2FAkPmxNgJ54inTHNihadr7wHvA7+Zl9kOA6pqmfQgcB3qby8/QNO1jTdNKAhmAuubbBwAfaJpWGuhsvm0wsFfTtI+BqsBEpVScd6BKqY5KqeNKqeMbVy58yZ8SVzZXN+5FxJylvXcnjKzZdXHK3bh6mUXfj6HbkIk4Z84SvW02Vx0FipQE4KPy1Qj+81KCn/tNucaakQ0PC8VFFzezeDlXnRsRFnUYER5KdteE1aGrTo+rzo0ixUsB8Fnl6vzvj4vJlk3npic05K/o38PDQnHVuVmVcXPTExoaEm8Znfn/7NldqFS1evSZeDc3PVWqVUcpRfGSpVF2dm904bX4mPpkTJ6IsDCyu7q9Yov4OWfKTOkPP+b4keT7dI7OTU9YnLrSxSrjHk8ZN47/eoQcOXORNVt2HBwcqVj1c86e/p1bN2/w1+1b0RcyCw8L5ctWjbgTEZHgXHq9OyF/xTxnWGgobm6x2lnvTmhITJnQ0BB0bm4J2hbA06suu3ftBODM6VNMnTwJzxrVWLpkET/Nmc3ypT/H2Sa1slpKly4dWbOaruFQvERJPDzyEHztaoKyvpAS7f5CVFQUgft3U7V67URlekGvdyfEqq5Co8dtdBn32GUs6vMl27q4umJvb4+dnR1+DRtx9g1ms1/GTW+9X3qRx/rv0hPymjKx6fV6Pq9ew7xcuDR2yo579xK3X3rbYwkgNCSEr3t0w3/MeDzyxPe26NXc9HrCQi3286GhuMbaZ+r01n04LMxURuemR+emp0Qp00qlqtVr8sfFC6Zt3PRUstzPq6Tt5/V6PaGh1m0a+/VIH6tvhFmUiYyM5JvePfD0qke16jXfOMfrxG7fsHhymvqwdVvrdG6c+v0kAQf24eNZnSED+nD82FGGDepHbLU8vdi3Z1cSc9pGu4u0Tw7gbcdVTdN+N/98AsgHfAasVkr9DswGXqzLyQ3sUEqdAb4BLE8rbtQ0LTHrfSsBPwNomrYFiG+PcAxop5QaDpTSNO0BUBYoDhw052sDvFhrWNW8QuAMUM0i32lgqVKqJRBlvq0mMMD8GPsBJ+I5iaBp2hxN08pomlamfpO2Cf7j8hUuRujtG4SH3CYqMpJfA3bx3ifWF3i6ExbCD2MH0r73MNxzxTx1lmwuZHfVE3IzGIALp46R0yM/Ka1osRLcuhHMX7dvEhkZyb5d2/msYpUUf95/k8JFS3D75nVC/rpFZGQkgXt38Gn5KgnaNpuLK646d25evwbAqZO/4pEv+T7LV6xESW7euM7tWzeJjHzO7h1bqVC5qlWZCpWrsn3zRjRN4+zpU2R0dsZVp+PJk8c8evQIgCdPHvPrkUPRF/GpVPVzThw7CsD14GtERUZGHyQlVZEX9Wnukwf2bKdshcoJ2vbve3d5+OA+AM+ePeW3Y0fwyJsvWXIBFC1uqs+/bpmy7d25jfIVreuzfMUq7Nhqqs9zZ0z16eKqQ++eg/NnT/P06RM0TePksaPkzVeAgoXeZcOOAFZu2MnKDTvRuemZu2Q1Lq6uCc5VomQprl+/xs2bN4h8/pztW7dQuWo1qzJVqlZj08b1aJrG6VO/4+ycCZ3O7ZXbBgdfi95+/7695Dd/znThkmVs27WXbbv20qJVGzp07ESzFi1TNevL3L17F4PBAMDNGzcIDr5G7tweCcr6Qkq0+wsnjh0hT94CVstcE+NFndy6eYPIyOfs2Ba3TipXqcbmV9RnfNuGW3xEYO+e3RRK5LUOXps5OJhbN2+an3crVarEyly1Gps3boiT+VWqVKvOr0dN+6Xga1eJjIwkW7bE7Zfe9li6f/8+3bp0pGev3nzw4UeJyvpC0eIluWG5n9+5lfKx9/OVqrJ9i3k/f+YUzub9vIurDje9O9fNJ7WO/3ok+uKMlap8zknL/XxU0vbzxUuU4oZFu+/cvpXKsdq9UpVqbNlkavczp37HOZOpbjVNY9SwIeTPX5CWrdu9cYaEKFaiJDeuB0fX564d26gUqz4rVq7Gts3mnKdP4eycCVedjq49erN55z7Wb9uN/7jJlPn4U0aMmQCY6vCFwAP7yJvEz+3bSrunRam9VP5l/1JL2rrClXgVy+8uMgB64G/zrHds3wNTNE3bqJSqAgy3uO/RGzz3K9fhaJoWoJSqBHgBS5RSEzEd6O/SNK2ZZVmllBOmlQJlNE27YT7odzLf7YXphEF94FulVAlMax4baJqWIlPb9vYONO/cl6nDemI0GilfvS658hZg/7ZfAKji6cemFfN4dP8flv5o+nobO3t7vv1uIQDNOvVh7uRhREVFotPnol2vISkR0zqzgwPd+w6if88uGI0GPOv6kK9AITb9Yr7ar19j7t6JoEvbpjx+9AhlZ8faFT8zf8V6MmZMoeVrCbRobFsqflQY16zOXNk+ilGztrJo/Zt9BU9S2Ds40KlXf4b3/Qqj0Uj1Ot7kyV+QbRtWA+Dp3Yh7dyLo3akFjx89ws5OsXHNUmYuWss7GZ3p2LM/U/wHERkZhXvOXPQcMCLZsjk4ONC7/2C+7toRg9FI3fq+FChYiHVrTItkfBs24bMKlTgcFEAjb0+cnJwYPNwfgLt37jCwj+mKuQaDgRq1vShb3nRCqq63L6OHf0uLRt44OjoyZMToZHvxsXdw4KuvBzK4dxeMBiM1zX1yyzpTn/TyNfXJHu2bRffJ9at+ZvbSddy9E8Fk/yEYjEY0o5FK1WryafmEHfwnhIODA72+GUTfHp0wGg3UqedL/oKF2LDWVJ/eDZpQtnwljhwKpLmfJ+mdMjDg21EAFC9Zmsqf1+DLVo2xt7enUJGi1PNtlGy5Bg4eSpeOHTAaDfj4NqBQocKsWrkcMC2HrlipMkEBB6jrWQMnpwyM9B/zym0Bpk2ZzLVrV7GzU+TIkYshw5LeN1Mq657duxg3ZhT37t6l21edKFKkGLPmzuPk8WPMnDEdB3t77OztGTJ0BFmyZk105pRq9707t/F5Tc8k1eeAQUPp0qkDRoMBb3OdrDbXZ6MX9Rl4gHqeNXDKkIERo8a8cluAqZMncunSRRSQM1cuhlgs7fasWY1HDx+aTvru3c2Pc+ZbXUE+IZn7D/qWrzq3x2gw4u3bgIKFCrN6lenq2I0aN6VCxcoEBQRQv05NnJycGG7uAwAD+vXmxLFj/P33PWp9XpnOXbvj69cQH18/hn87mIa+9XB0dGTk6HGJ3i+97bG0YtnPXL9xnTmzfmDOLNPCxx/nzsclERfgc3BwoHe/wfTu1hGjwYiXt2k/v968n/dp2IRyFSpx+GAATcz7+UHm/TzA1/0GMWJIf6IiI8mZKzcDzfd5efsydsS3tGrsjaODI4OHJ20/7+DgQL9B39KtS3sMBiPePqZ2X2Nu94bmdj8YGIC3l7ndzX31999OsmXzBgoVfpdmjXwA6NrjaypUrMzePbuYONafe/fu0rNrZ94tWpSZsxL31Zaxc/YdMJgeXb7EaDRSz9uXAoUK88tqU06/Rk0pX7ESh4ICaFCvNk5OTnw7YvRrH3fm9O+4fu0qdnZ2uOfISf/BSbtqvq20u0j7lHxGIu1TSuUDNpuXnKOU6gs4Y5qd/k7TtNXmz5GX1jTtlFLqN6CDpmknlFILgPyaplUxHyw/1DRtkvlxGgD1NU1r84rnng6EaZrmr5TyBLYCOk3TIpRSDzVNc1ZK5QVuaZoWpZTqhWl1wGhMKwWqaZp2RSn1DqaVAWHAJXMZe+AIsAbT5/HzaJp2TSnlCNwEimBa/p8Z00cFNKXUB5qm/faq+gr8455NdOr8undSO0KCFK7WJ7UjJMjv2yakdoQEcc2UuIsdpZb7T6JeXygNyOBon9oREiRrRsfUjvCv8vejyNSOkCBZ3rGNdreV94J2drZxUPLgqew/k1OU0ZjaERIk0mAb4whA5+xgG4PJLHPTxWmycu+vaJ0q9ShL6G1bC6C9UuoUcA54cYG34ZiW1gcCr/pQ5j6guHrFReyAEUAlpdRJTCcMrsdTpgqmz73/humz+dM0TQsH2gLLlVKnMR2oF9U07W9gLnAGWI9p+T2YDuZ/Ni+r/w3TiYm/gVGAI3Bamb5Kb9Qr/h4hhBBCCCHEv0hqL5VPa0voZQZe/OvIDHzykhn45CUz8MnLVmaQZAY+eckMfPKylfeCMgOfvGxl/ykz8MnP1mbgszRbkiYr95/lrWQGXgghhBBCCCGEEPGTi9gJAJRS7YCesW4+qGla19TII4QQQgghhBDY1HqBlCcH8AIATdMWAAtSO4cQQgghhBBCiPjJEnohhBBCCCGEEMIGyAy8EEIIIYQQQog0Sb7X3prMwAshhBBCCCGEEDZADuCFEEIIIYQQQggbIEvohRBCCCGEEEKkSbKE3prMwAshhBBCCCGEEDZADuCFEEIIIYQQQggbIEvohRBCCCGEEEKkSbKE3prMwAshhBBCCCGEEDZADuCFEEIIIYQQQggbIEvohRBCCCGEEEKkSbKE3prMwAshhBBCCCGEEDZAZuDFv07GdPapHSFBHj0zpHaEBPl924TUjpAg73v2S+0ICXJs07jUjpAgzhls4+UhnYNtnIc2GrXUjpAgNhKT9I620e5CpGUatjHg0zvYxvu6p88jUzuC+I+wjXdoQgghhBBCCCH+e2QFvRU5hS2EEEIIIYQQQtgAOYAXQgghhBBCCCFsgCyhF0IIIYQQQgiRJslV6K3JDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAbKEXgghhBBCCCFEmiRL6K3JDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAbKEXgghhBBCCCFEmiRL6K3JDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAbKEXgghhBBCCCFE2iQr6K3IDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAXIAL4QQQgghhBAiTVJKpcl/CcxeWyl1SSl1RSk1IJ77syilNimlTimlziml2r3uMeUAXgghhBBCCCGESEZKKXtgJuAJFAeaKaWKxyrWFTivadp7QBVgslIq3aseVw7ghRBCCCGEEEKI5PUJcEXTtP9pmvYcWAF4xyqjAZmUaUrfGbgLRL3qQeUAXvzn/X7sEL3bN6BXW182rFwY5/6gvdvo17kZ/To3Y2ivLwj+8w8A7oSFMOqbzvTp0Ii+XzZm27rlKZrzxNGDdGnpQ8fm9VmzdH6c+28GX+WbLq3xq/4J61Ystrrv4YMHjBvaly6tfPmqlR8Xz576z+d8lVnDWhC8ZyzHVw9Klee39Nuvh+jexo+urbz5ZfmCOPffvH6Vgd3a0qR2WTassq7PTWuW0vOLRvRq35gp/oN4/vxZiuU8duQg7ZvWp22juqxcPC/O/devXaXXl62oW7kMq5ctir49LDSEb7q1p0MzH75s4cu6lUuTPdvhg4E08a1Dw/q1WLxgbpz7NU1jyoTRNKxfi5aNfbh04Xz0fb5e1WnR2JvWTX1p16JRnG2XLp5PuQ+L8/e9e0nOeTAoEJ96talfpybzf5oTb87xY/2pX6cmjf3qc+H8uej7hn87iGqVP6Ohbz2rbWZ+P43GfvVp0tCHLh2/ICwsNMk5DwUF4levNt5eNVkwL/6cE8b54+1VkyYNYnKGhPxFx/ataeBdh0a+dVn28+I42y5eOI+PShflXjLU5+GDgTT2Mbf7/PjbffJ4U7u3aOzDRYt296lTnRaNvGnVxJe2zWPafc+u7TRrUI9yH5bgwrmzb5ztYFAA3nVrUc+zxsvbeow/9Txr0Mi3nlVbv2zbfn160biBN40beONZsxqNG5jeI0ZGRjJkUH8a+tbDt54n8+bOfsPMttE/DwYGUN+rFnVr12De3PhzjhvjT93aNWgYu25fs+2iBfN4r0QR7t27m+ScRw4F0szPiybetVnykv3S1AljaOJdmzZNfK32Sw8e3GdIv14096tLiwb1OHv6dwDmzZ6JT+2qtG3mR9tmfhwOCkhyTtN498THqxYL58Wfc+K40fh41aJpA28uWoz3Tu3b0NDbi8a+dVluMd5n/zADz+qVad7Il+aNfAkKPPBG2d7mODpz5nT07Y396rN39643ynz0cBAtG9aluZ8nSxf9FG/maZPG0NzPk3bNffnjYky7r1q2mDZNvGnb1IcRQ77h2TPTa/qCOTNp4FWN9i0a0L5FA44cTHq7pzWpvVT+Ff86KqWOW/zrGCt6LuCGxe83zbdZmgEUA24DZ4CemqYZX1Uf8jVy4j/NaDCwYOYEBo2dgYurnsHd2/BR2UrkzlsguoybPidDJ87GOVNmfj92kLnTxuA/fSF29g607NiL/IWL8uTxIwZ1a02pDz+12ja5GAwGZk8dx8jJP+Ki09OnUws+KV+ZPPkKRpdxzpyFjj36cyRoX5zt534/gQ8/+YwBIycRGRnJs6dPkz2jLeV8nSWbjjBr5QF+GtU6VZ7/BYPBwNzp4xg64QdcdHr6f9WKj8tVxiNfTB/LlCkL7bt9w9GD+622vRMextZ1K5g6fzXp0zsxaWR/gvbuoFrt+imSc+akMYydNhtXNz3d2zenbMUq5M0f0+6ZM2emy9f9ORRg3e729vZ07N6XwkWK8fjRI7p90ZQPPylrtW1Ss00e78+0H37CTa/ni5ZNqFi5KvkLFIouc/hgADeuB7N6w3bOnTnNhLEjmLd4ZfT9M2cvJGu2bHEeOzTkL44dOYy7e45kyTlu9Eh+nDMfvbueFk0bUblqNQoWjMkZFBjA9eBgNmzZwZnTpxjjP4Ily1YBUM/blybNWvDtYOuP17Vp156u3XsCsGzpYubM+oEhQ0ckLeeYkfwwZz56vZ5WzRpRuUo1CljkPBgUwI3gYNZv3sHZ06cY6z+CxctWYW9vz9d9+lOseAkePXpIy6YNKFvus+htQ0L+4uiRQ7jnyPnG+SxzThrnz/QfTe3eroW53S1yHg6K1e5jRjB/iUW7z4nb7gUKFmbc5OmM8x+epGxj/Ucya+4CU1s3aRh/W1+/xsatOzlz+hSjRw3n5+WrX7nthMlTo7efPHEczs7OAOzauZ3I589Zs24TT548wc/bi9p1vMiVK3eiMttK/xwzeiSz5y5Ar9fTvElDqlStRsFCsXNeY9M2U936jxzO0hWrX7ttyF9/cfjQIXIkU/+cMm403/0wFze9ng6tmlAh1n7pyMFAbtwIZsX6bZw7e5pJY0cyd/EKAKZNHMun5SrgP2EqkZHPeWrxWtm4eWuat37tR2gTnHP8mFHMnDMPvV5P62aNqVSlarzjfd3m7ebxPpJFy1biYG/P1336UbR4CR49ekSrpg341GK8N2/ZhlZtv0hStrc5jgoVKsyylWtxcHAgPDyMxg28qVSlKg4OCT+UMhgMTJ3gz+QZc9G5udOpTRPKV6xKvgIxr3dHDwVy88Z1lq7dyvmzp5kyfhSzFiwnPCyUtSuXsnjlBtI7OTFsYB/27tqGZ10fABo1a0XTlsnT7iLhNE2bA8Q9exQjvg/Ka7F+rwX8DlQDCgK7lFKBmqbdf9mDygz8v4BSqopSanNq57CklMqnlHrz6Ym39JhXLp3DPacH+hy5cXB0pFyVGhw/bH0m+N0S7+GcKTMAhYqW4m5EGADZXFzJX7goABneyUguj3zcjQhPznjRLl84S45cHrjnzI2joyMVq9XiaNB+qzJZs2WncLES2Md6MXn86CHnTp2khpcvAI6OjjhnyvSfzvk6B0/+yd1/HqfKc1u6cvEc7hb1WaFqTY4d2m9VJku27BQqWiLeNxEGg4Hnz55hMETx/OlTsrvqUiTnpfNnyZnbgxy5TDmrVK/N4UDrnFmzu1CkeMk4OV1cdRQuUgyAdzJmxCNvASLCw5It2/mzZ8idOw+5cnvg6JiO6rU8Cdi/16pMwP69eNb1RilFydLv8fDBAyLCXz+Wp00eT9defSCBF7J5lbNnTuORJw+5PUw5a3nWYf++PVZlDuzbQ936ppyl33ufBw/uE26uq4/KfEyWLFniPO6LN58AT548SfBFd17m3FlzTnN91qwdf06veqacpd57n4fmnDqdG8WKlwAgY0Zn8ucvaDXjOmXCWHp+/U1yVKep3T1i2r1GfO1+YC91Etnu+QsUJG++/EnKZmrrvBZt7cX+vdZ1uH/fHurW94nT1gnZVtM0dm7fRu06dQHTzNWTJ0+Iiori2bOnpn2rRb9IeOa03z/PnjmNh4e5ftKlo3Ydrzg59+3dQ72X1e0rtp04fixf9/kmyRkBLpw7Q24Pj5j9Us06BO23PrkZeGAvtb3qm/pnqfd4+NDUPx89fMip305Q16cBAI6O6chkfo+S3OIb7wf2WY+jA/v2UsdivD94cJ+I8DBcdW4UjR7vGckXa7wn1dseRxkyZIh+/Xr+7BnqDb6Y/MK5M+TKnYecuTxwdHSkWk1PggKs6zMoYB+16pjavUQp037pjvm9pcEQxbNnz0xj+ekTXFPoNV0kq5uAh8XvuTHNtFtqB/yimVwBrgJFX/WgcgAv/tPu3QnHRaeP/t3FVc+9VxyE79++gfc//izO7eEht7n25yUKFS2RIjnvRITh6haT01Wnj96hv07I7VtkyZqNaeOG0bN9U76fMIKnT578p3PairsRYbha9M/siahPF50b9Ru1pHMzLzo0qsU7zs68X6ZciuS8Ex6GTu8e/burzo2I8MS/UQv56xZ/Xr5I0RKlki1beHgobu4x2dzc3AkPsz5BEB4Wht4iv85NT7g5v1KKnl070LZ5Q9avXRVdJvDAXnRubhR+95WvsQkWFhaK3mImX693Jzw0NE4Z91hlEvKGeMb076hdvQrbtmymS9ceScsZGopeHytnWNycln+LWzxlbt+6ycWLFyhZ6j3AdBCgc9PzbpHkqc/wsFDcLNrUTe8efTAZUybMum/o9dE5lVL0+KoDbWK1e3IwtWPM8+r1+jjtGBYau4w7YaGhCdr25InjuLi4kDdvPgCq16hFhgwZqFG1ArVrVKV12y/IkiVrojPbSv90z2HdpqHx5NTHV7ev2Hb/3j246d0oUjQ5+2dMXen0MfucFyLCwqz7sJueiPBQbt+6QdZs2RgzfDDtmjdg3MihPHkSc8L5l1XLaNPElzEjhnD//j9JyhkWar1vdIunv4XH6ZPuhMXax96+dYtLFuMdYNWKpTRt4M2IoYPfKOfbHkcAZ06fws/bi4a+9RkydESiZt8BIsKt21Tnpo9zwjoi1r5L52baL+nc9DRt2ZbG9avjV6cqGZ0z8XHZ8tHl1q1eTrvmvowbNYQHSWz3tCgNLJV/06vQHwMKK6Xymy9M1xTYGKvMdeBz89+pB4oA/3vVg8oBfApQSq1XSp0wfxVAR6VUF6XUBIv72yqlvjf//K1S6qJSapdSarlSqu8rHvdjpdRppdRhpdTE+GajlVLDLR9DKXVWKZXP/HNr8/anlFJLzLflVUrtMd++RymVx3x7I/O2p5RSAebb7M3Pe8xcvlMC6yPe7ZRSK5VSdSzKLVRKNXiT51EWn0H5ZVnczwm/jKbFXsXCS2fTzv1+nH07NtKsfTer258+ecx3o/rTunNv3smYuFmNhIo3ZgK3NRii+PPyRTy9GzFt3gqcnDKwZlncz6YnB1vJaSu0OKusSPDsz8MH9zl26AA/LN3E3FXbefrkCQd2bU3uiEDScr7w5PFjRg3qQ+ee35AxGcdRfGM8brR4yph77uwFS1m0bC1TZsxm7arl/HbiOE+fPGHhvNl82bl7suWMJ0KcoPGPr9fXc7ceX7N99348veqycvnPbxjQnCGe22K3dfy71Zgyjx8/4pvePejbbyDOzs48efKEeXNn0TmJB2/WOeNLGqtMvH3DlHPOgqUsXr6W72bMZs1KU7snW7ZXPO/ryiRk2+1bN0fPGoJpptLO3o6dewPZun0PSxbN5+aNG7Ef5jWh47ktTfbPBOyLXla3L9n2yZMnzJ0zi6+69UxSttdESFAfQCkMBgN/XLyAT8OmLFi2FqcMGfh5gemz1L4Nm7Byw3YWLF+Li6uOGd9NTGrSN8oZe7z3692DPv0GRK+4aNikKeu37GTZ6nW4uur4btKEOI/x2mRveRwBlCr9Hr9s2MLSFWuY99Ps6M+gJyVz7HdJL+uHD+7/Q9CBfaxYv4Nftu7l6ZMn7Ny2CQDvBk1Y9ss25v28FhcXHTOnJbXdRXLRNC0K6AbsAC4AqzRNO6eU6qyU6mwuNgr4TCl1BtgD9Nc0LeJVjysH8CnjC03TPgLKAD2AXwA/i/ubACuVUmWABsAH5vvLvOZxFwCdNU0rBxgSE0gpVQIYDFQzf03Bi1eiGcBiTdNKA0uB6ebbhwK1zGVffGi2PfCPpmkfAx8DXyqlErKW8GXbrcBUF5jPSn0ObH2T59E0bY6maWU0TSvj1zzhnwHK7urGHYuz3nciQsnm4hqnXPD/LjNnqj99h08iU+as0bdHRUXx3aj+lK9Wm08qVEvw8yaWq86NCIuzwxHhoQleDu2q0+Oqc6NIcdOs5meVq/O/Py7+p3PaChdXvdVM9t3wULLH0z/jc/rkUdzcc5ElazYcHBwpW7Eal86nzEUBXXV6wkNDon+PCA/DxdUtwdtHRUUyalBvqtWsQ4Uq1ZM1m5ubO2EhMdnCwkJw1Vln07npCbXIHx4WGl1GZ/4/e3YXKlf9nPPnTnPz5g3+unWLVk198fWqTnhYKG1bNEjw6oh4c+r1hIb8Ff17aGgIOjfrnHq9npDXlHkVzzp12fOGF16yzBAaap0hdn3qY/0tYRZlIiMj+aZ3Dzy96lGtek0Abt64zu1bN2nWyJu6tasRFhpKiyZ+RCSlPt3cCbNo07DQkOi2jC6j11v3jVCLdnezaPdqpnZPLnq9OyEWzxsaGhonm949dhlTW79u26ioKPbs3kWt2tHnxtm2dTPly1fE0dGR7C4uvP/+h5w7dyZRmW2nf7oT8pd1m7q5xW53d0JfVrfxbHvzxnVu3bpJYz9vPGtUIzQ0hKYN/RL0MZuXcdPrCbMYR+GhobjG2mfq9HrrPhxmKqNz06Nz01OiVGkAqlavyR8XLwCQ3cUVe3t77OzsqO/bkAuJbOf4coaGWtdJ3HEUT1/VmV73oyIj6de7J7UtxjuAi0VO3waNOHcm8ePrbY8jSwUKFiRDhgxcufxHojLr3Kzb1PRao4tVxj2eMm4c//UIOXLmImu27Dg4OFKx6ufRFy+0bPe6Pg25mIQLbIrkp2naVk3T3tU0raCmaaPNt83SNG2W+efbmqbV1DStlKZpJTVNe+2ZTDmATxk9lFKngCOYPveQH/ifUqqsUsoF09KIg0AFYIOmaU80TXsAbHrZAyqlsgKZNE07ZL5pWSIzVQPWvDijo2nai0uolrN4rCXmTJjzLVRKfQnYm2+rCbRWSv0OHAVcgMIJeO6XbbcNqKaUSo/p+xEDNE17koTnSbSCRYoTcus6YSG3iIqM5PD+XXxUtpJVmYiwEL4b2Y+u34wgR+680bdrmsacKaPI6ZEPrwYtUiJetMJFS3D75nVC/rpFZGQkgXt38Gn5KgnaNpuLK646d25evwbAqZO/Wl0E7b+Y01YUKlqcv27dINRcn0H7dlLms8oJ2tbVzZ0/Lpzh2dMnaJrGmZO/kjtP0j67+zJFipXg1s3rhNy+SWRkJPt3b6dshYTl1DSNKWOG45GvAA2aJf9FA4uVKMmNG8HcvnWTyMjn7N6xjYqVq1qVqVi5Gts2b0DTNM6ePkVG50y46nQ8efKYR48eAfDkyWOOHjlEgYKFKVT4XbbuCWLdlt2s27IbnZuehUtNM15vqkTJUlwPDubWTVPOHdu2UqWK9UnBylWrsXmjKefpU7/j7JwpzhvW2IKDr0X/fGDfXvLlT1ofKF6iFDcscu7cvpXKsXJWqlKNLZtMOc+c+h3nTKacmqYxatgQ8ucvSEuLi2wVfrcIuw8cYvP2vWzevhc3vZ6lK39J0uc7i5UoyY3rMe2+a8c2KlaJ2+5bLdrd+SXt/uthU7snlxIlS3H9+jVu3bxhbustVK4aq62rVGPzxvVx2vp12x49coj8BQpYLRHPkSMHv/56FE3TePL4MWdOnyJ//sTtW22lf76on5s3bxD5/Dnbt8at2ypVq7HpFXUbe9vC7xZhf+Bhtu3ay7Zde9Hr3Vmx5pc4B16JUbR4SW6YT1xFRj5n986tlI+1X6pQqSrbt2w09c8zp3B2dsZVp8PFVYeb3p3r164CcPzXI9EXQbM8qRCwb3eS+218471SrHFUuUpVtsYa767m8T5y2BDy5y9Ay9ZtrbaxXDa+b+8uChZOfM63PY5u3bxBVJTpm71u375F8LWr5MwV+2Lir1a0eElu3rjOX7dMr5V7d26jfEXr+ixfsQo7tpra/dyZU2R0dsbFVYfePQfnz57mqfk1/eSxo+Q1v0eyPHkcuH+P1cU6/y3e9tL4hP5LLXIV+mSmlKoCVAfKaZr2WCm1H3ACVgKNgYvAOk3TNJW4lk9o2SisT8w4WWz/+jWF5jKapnVWSn0KeAG/K6XeNz9Gd03TdiQwywsv3c5cP7UwzcQvf1X5Fx8FSE729g607dqPsYN6YDQaqFKzPh75CrJr81oAatRtwC9Lf+Lhg3+YP2M8AHb2DoyZsZhL504RuGcrHvkLMaBLcwCatOvKB5+Uf+nzvXFOBwc69erP8L5fYTQaqV7Hmzz5C7Jtw2oAPL0bce9OBL07teDxo0fY2Sk2rlnKzEVreSejMx179meK/yAiI6Nwz5mLngPe/Cq//4acr7NobFsqflQY16zOXNk+ilGztrJo/eG3nsPe3oEO3fsxqn83jEYD1Ty9yZOvIDs2rQGgVr2G3LsbQb8urXjy+BFKKTavXc60+at5t1gpylX6nL6dW2Bv70D+QkWo4eX3mmd8w5wODnTtPZBBX3fBaDBSs64P+QoUYvM602eH6/o25u6dCLp/0YzHjx6h7OxYv/Jn5ixbx9Urf7Bn+2byFyxMlzaNAWjXqTuffFYxWbI5ODjQp/9genX9EqPRSN36vhQoWJhf1piu5uzXsCmfVajEoaAAGnnXJr2TE0OGjwbg7p07DOhjWtZtMERRs7YX5conT674cvYf9C1fdW6P0WDE27cBBQsVZvUqU85GjZtSoWJlggICqF+nJk5OTgz3HxO9/YB+vTlx7Bh//32PWp9XpnPX7vj6NWT61MkEX7uGnVLkyJmTwd8mbUw5ODjQb9C3dOvSHoPBiLePKecac86G5pwHAwPw9jLnHGXK+ftvJ9myeQOFCr9Ls0Y+AHTt8TUVKibsZE9ic/btP5ieX5nb3dvc7qvN7d4opt0b1q+NU6x279/bot09Y9p9/97dTB4/mr/v3aV3jy68W6Qo036I+9Var8s2YNBQunTqgNFgwNu3AYUKFWb1StNLYKMmzahYqTJBgQeo51kDpwwZGGGuw5dt+8L2bVup7ell9XxNmrVg6JCBNPCpC5pGfR+/RF9rwJb658DBQ+nSsQNGowEfc/2sMtdt4xd1G3CAup41cHLKwEj/Ma/cNiU4ODjQu99genfriNFgxMvblwIFC7F+jelbEHwaNqFchUocPhhAE29PnJycGDTcP3r7r/sNYsSQ/kRFRpIzV24Gmu/7cfpkLl+6iFIK95w5+WbQ8CTn/GbQELp36YDBYKS+j1+c8V7ePN59vGrh5OTEMHNfPfXbSbZu3kihwu/SvJHp4rRf9ehFhYqVmfbdJP64aMqZI2cuBg9NfM63PY5+O3mC+fPm4uDggJ2dHQOHDCdbtuyJztzrm0H07dEJo9FAnXq+5C9YiA1rTe3u3aAJZctX4sihQJr7eZLeKQMDvh0FQPGSpan8eQ2+bNUYe3t7ChUpSj1f01dc/vj9ZK78cQmlwD1HLvoOHJbo+hS2RcX/eQzxppRS3kAHTdPqKaWKYvpagNrAKeAEEIzpsw2/KqU+BmYDn2E6mXICmKtp2qSXPPZZ82MfUUqNAeprmlbSfNKgr6ZpdZVSLYG6mqY1VUp9iOniCQWBjMA6TCcW7iilsmuadlcptRFYrWnaEqVUW8Bb0zRfpVRBTdP+ND/vb5iukPgJUAdopGlapFLqXeCWpmmP4smaD9hsztfxZdsppbyADpg+PlBQ07TnLysP6F485qva4OS1+zbRqTM6yfmz5PS+Z7/UjpAgxzaNS+0ICeKcwTb6Z2Ynx9SOkCBOjrax4M1oE3tPiDS88ity0wwnR/vXF0oDbOW9oJ1d6s14JcaDp1GpHSFBbGW/5GBnGzn/eRyZ2hESzD2Lo20MJrOcnX5Jkzup27P9UqUebeMdmm3ZDnRWSp0GLmFaRo+mafeUUueB4pqm/Wq+7Zj5APoUpgP748CrLh3ZHpirlHoE7H9J2bXELD8/Bvxhfq5zSqnRwAGllAH4DWiL6TP685VS3wDhmA7UASYqpQpjmg3fY854GsgHnDSvHggHfBJQJz+9YrudwGJgo6ZpzxNQXgghhBBCCPFfYVOnG1KezMCnMqWUs6ZpD5VS7wABQEdN006+qqz55wFADk3Tku+yqP8SMgP/3yQz8MlLZuCTl63MdMkMfPKSGfjkJTPwyctW9ksyA5/8bG4GvnManYGfJTPw/1VzlFLFMX1WfdHLDt7NvJRSAzG1WzCmGXQhhBBCCCGEEP8BcgCfyjRNax77NqXUTCD2ldCmaZq2ANPF8NIMpVQpTFevt/RM07RPUyOPEEIIIYQQ4t8jNa/4nhbJAXwapGla19TOkFCapp0B3k/tHEIIIYQQQgjxb2cbHyoRQgghhBBCCCH+42QGXgghhBBCCCFEmiRL6K3JDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAbKEXgghhBBCCCFEmiRL6K3JDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAbKEXgghhBBCCCFE2iQr6K3IDLwQQgghhBBCCGED5ABeCCGEEEIIIYSwAbKEXgghhBBCCCFEmiRXobcmM/BCCCGEEEIIIYQNkBl48a9zJuKf1I6QIGVzu6R2hARxzZQutSMkyLFN41I7QoJ8XG9AakdIkNsHp6V2hARJ52Aj56FtZPLgj78epHaEBMmZNUNqR0iQp5GG1I6QIM7pbePtoK3Mwj16FpXaERLkyXPbqE8724jJ2dv3UztCgtXOokvtCCIJbGOPLYQQQgghhBDiP8dWTt69LTYydSGEEEIIIYQQQvy3yQy8EEIIIYQQQog0SWbgrckMvBBCCCGEEEIIYQPkAF4IIYQQQgghhLABsoReCCGEEEIIIUSaJEvorckMvBBCCCGEEEIIYQPkAF4IIYQQQgghhLABsoReCCGEEEIIIUTaJCvorcgMvBBCCCGEEEIIYQPkAF4IIYQQQgghhLABsoReCCGEEEIIIUSaJFehtyYz8EIIIYQQQgghhA2QA3ghhBBCCCGEEMIGyBJ6IYQQQgghhBBpkiyhtyYz8EIIIYQQQgghhA2QA3ghhBBCCCGEEMIGyBJ6IYQQQgghhBBpkqygtyYH8OI/789Tv7JryQ9oRiPvVfHks/rNrO7/4/hBDqxZiFJ22NnbU6NVFzyKlOLO7Rus+94/utzfYX9RqWEbPvFskCI5Txw9yE/fT8RgNFLTy4eGLb6wuv9m8FWmjRvGn5cv0qpDN3ybto6+7+GDB8yYOILgq3+iUPToP4yiJd9LtmxHDgYyddI4DAYD9Xwb0Lrdl1b3a5rGdxPHcjgoACenDAwZMZoixYoD4OdVg3cyZsTezg57ewfmL10Vvd3qFUtZu3IZ9vb2fFahEl179U22zL/9eoj5MydhNBr4vI4Pfs3aWd1/8/pVZk4Ywf+uXKT5F1/h3TimPjetWcruretRSpEnfyG69RtGunTpky1bQs0a1gLPSiUJv/uAMo3GvPXnP3wwkO8mjsVoNFDfpyGtv4jb7lMmjOHwwQDSO2Xg2xFjKGpudwCDwUC7Fo3QuemZPP1HAAb37831a1cBePDgAZkyZWLJynVJynkwKJBJ40djMBjx9WtIuw4d4+ScOG40QYEBODk5McJ/LMWKlwBg+LeDCAzYT/bsLqxetyl6mz8uXWT0yGE8efyYHLlyMXrcJJydnZOcc+K40RgNRnwaNOSLeHJOGDuagy9yjrbIOWQQAeaca9bH5Pxu0gQCDuzD0cGR3B55GOE/hkyZMycpp6Xfjx1i8Y+TMRqNVK3tjXfTtlb3B+3ZxsZViwFwypCB9t0HkLfgu9wJC+GHicP5++4dlJ3i8zq+ePo2i+cZ3tzRw0F8P3kcRqMBL+8GtGjTwep+TdOYPnksRw8Fkt7JiYFDR/NuUVP/XLVsMVs2rEUpRf5ChRnwrT/p05vG+NqVS1m3ejn29vaULV+JLj36JDnnjMnjMRgNeHn7xZvz+8njOHIoECcnJwYM9Y/OuXrZYrZs+AWUokChwvT/dhTp06fnyh+XmDJuJE+ePMY9Ry6GjBxHxiT2z0MHA5k0fgxGoxEf34a0bR93vE8aP4aDQab+OXzUGIoWK0FIyF8MGzyAO3cisFMK34aNadbCtD/949JFxvoP5/Hjx+TMmYtRYye+0Tg6GBTABPPY8W3Q6KVjJyjwAE5OTowcPS567Lxs2x9nfs8va1eRLVt2ALr37E3FSpWjH/Ovv27jV9+Lzl91o0279onObOnYkYPMmjoeg8GIZz1fmrS2frzr164yZfRQrvxxgTadutOoeRsAwkJDmDhqMPfumMZRnfoN8W3SIklZXpnzcBA/TB2P0WDEs74fTePJOWn0t1y5dIF2nbrTqEXb6Psm+Q/l6KEDZM2WnblLk7ZPT4xfDwcx87vxGI1G6tT3o1k8mSf4mzJ/0bk7jS0yv00XTh7hl/nTMBqNlK1elxp+razuP/NrIFuW/4SdUtjZ2+P7RQ8KFku+93HCtsgSevFWKaXyKaXOvuG2OZVSa5Izj9FoYMfC72nSbwwdJ8zj/OF9hN8MtiqTr+SHdBg7hw5jZ1O3Y1+2zJ0CgEtODzqMnU2HsbP5YvQPOKZPT5EyFZIzXjSDwcDsqeMYNmEGMxetJWDPdq5f+9OqjHPmLHTs0R/fJq3jbD/3+wl8+Mln/LhkHdPmryR33gLJmm3S+NFM/n4Wy9ZuZPf2rVz93xWrMocPBnLzejCrNmyj/5DhTBw70ur+GbMXsGjFL1YH7yeOHSVw/14Wr1zH0jUbadba+gA7qZnnTh/H4LHTmTp/DUF7d3Dj2v+symTKlIX23b6hfiPrF9E74WFsXbeCCT8uYeq8VRiNBoL27ki2bImxZNMRvLvOTJXnNhgMTBrnz3czZrN87SZ2bt/K1T9jtXtQADeuB7N6w3YGDhnBhDEjrO5fuWwJ+fIXtLpt9PgpLFm5jiUr11H18xpUqVYjyTnHjx7J9z/MZe2GzWzftoX/xcp5MDCA68HBbNiygyHDRjLWPyZnPW9fZvw4N87jjhw2hB69+rBq3Saqfl6DxQvmJTnnOP+RzPhxLms3bmb71i38GStnUGAA168Hs2HrDoYMH8mYURY5fXyZOStuzrLlPmP1uk2sWreRvPnyMf+nOUnKacloMLBgxgT6j57GpLmrOLR/JzeDrceRm3tOhk6azYTZy/Fr3p65U00nmuzsHWjZsReT561m1LQF7Ny4Js62SWEwGJg6wZ8J035k0cqN7NmxlWv/s95nHj0UyM0b11m6dit9Bw5nyvhRAISHhbJ25VLmLFrJwhXrMRqM7N21DYCTx3/lYMA+5i/7hUUrN9C0Zdsk55w2YTTjp/3AopUb2Ltj20tyBrN07Rb6DBzGd+P9LXIuY/aiFSxcsQ6jwRCdc+LoYXTs1osFy9dRscrnrPh5QZJzjh8ziuk/zGH1uk3s2B7PODKP93WbtjN46AjG+pv28w729nzdtx9r1m9hwc8rWb1iWfS2/iO+pVvP3qxcu5Eq1aqzZGHix5HBYGCs/0hm/vgTv2zcwvatm18ydq6xcetOvh0+itGjhido25at2rJq7QZWrd1gdfAOMGn8WMpXrJjovPHlnzlpDP6Tf2DusnXs272d4KvWfSBz5sx0+bo/DZq1sbrd3t6ejt378tPy9Uyb8zObflkRZ9vkYjAY+H7yGMZM+ZGflq9n365tcZ4rU+bMdP16AA2bt4mzfU2v+oz57scUyfYyBoOB6ZPGMPa7H5m/fD17d27jWjyZu/UeEH1SJDUYDQZWz51CpyGTGDjtZ04G7ibkxlWrMu+W+oj+UxbSb8pCmnUdyIofxqdSWpEWyAG8sBmapt3WNK1hcj7m7T8vkU2fk2xuObF3cKR42SpcPnHQqkw6pwzRV798/uxpvFfCvHb2N7K55SSLTp+c8aJdvnCWHLk8cM+ZG0dHRypWq8XRoP1WZbJmy07hYiWwd7BeWPP40UPOnTpJDS9fABwdHXHOlCnZsp0/e4bcuT3IldsDR8d0VK9Vh8D9+6zKBO7fS+269VFKUbL0ezx88ICI8PBXPu66NStp1a4D6dKlAyB7dpdky3zl4jncLeqzQtWaHDu036pMlmzZKVS0BA4OcRcqGQwGnj97hsEQxfOnT8nuqku2bIlx8OSf3P3ncao89/mzZ8jtkSe63WvU8iRg/16rMgEH9lKnrne87R4WGsKhoAPU941/xYqmaezZtYMateskKefZM6fJnScPuT1MOWt51mH/vj1WZfbv20Pd+qacpd97nwcP7hMeHgbAR2U+JkuWLHEeN/jaVT4s8zFgOkjes3tnknN6xM651zrngTfIWa58heg+XKr0e4SGhiQpp6Url87hntMDfY7cODg6Uq5yDY4fOmBV5t0S7+GcyTTjX6hYKe5GmPJmc3Elf+GiAGR4JyO58uTjbsSr9wmJceHcGXLlzkPOXB44OjpSraYnQQHW/TMoYB+16pj2SyVKmfrnHXMGgyGKZ8+eERUVxbOnT3A1j/ENa1fSvE376P1StiTuly7Gk/NggPX+82ACcz59+hRXVzcAbly/xnsflAGgzKflCNi3O0k5z509jYdHHnKbx3vN2nU4EGu8H9i3lzr1TP2zVGlT/4wID8NV50bRYqbZ7owZM5KvQEHCwkIB8zj6yDSOPi33GXv37Ep0NtPYyWsxdrzijB3TGPeJM3YSsm189u7ZTa7cuSlYsHCi88Z26fxZcub2IEcu0+tRleq1ORy436pM1uwuFCleMs7rkYurjsJFigHwTsaMeOQtQIR5n5DcTDnzWOU8FKuvZntJToDSH5QhU+a4+6iUdPH8WfP4MmWuWiP+zEVfkvltCb5yAV2O3P9n777Dojj+OI6/h6KomFjgDmussZueGBt2Reki9sREY9cYe+/d2FKN3Ri7xt4VFLDElthN+0UsUcAWGyoc+/vjTuS4A6nCJd/X8/gk3M3efdidnd3ZmV1wcSuCg6Mjb9ZswOkjYWZlcubKbX4uyn9rTrlSKlv+yyrSgRdZwUEptUQpdUoptVYplVspdVEpNVEpdUgpdUwp9aZSaqdS6k+lVFdI3+h9Uu7dusFLBXXxP+ct4Mq92zctyv16NIw5/T9i9bRhNOtsOY373OFgKlavm5HRzNy8EYmL7tnFARdXffwJ3PNc//sqL+fLz+zJo/i0Yyu+nDqGR9HRGZYtKioCvVuh+J9ddXqiTCdn8WUiI9Hr3czLRBnLKKXo0+MTPmrTgg3rno3AXw6/yMkTx+n0QSu6d/qQc2dPZ1jmWzcicUlwsaVAKtZnQVcd3i3a0bV1Mzq1aExuZ2def/v9DMtmK6IiI9Al2KY6vVt8Z/JZmUh0bgnLPKsbM6dNpuen/VF21g9Dv5w4ToECBSn+Sol053RLUD91ejciI8zrZ2SkeR3W6d0s6nBipcuUZX+wsQOzZ+cOIq5fS1fOxBn0VjJERpj/Lnorv0tyNq5fR42atdOVM6HbN6IomGA/Kuiq5/bNpPejfTs28vo71S1ej7r+Nxf/+JUy5StlWLYbUZFm9dNVp7fo2NxIVIeftl2uOj2t2nUg0LsB/k3rksc5L+9UqwHAlUsXOfXLcbp+1JreXTpw/lz62qWoqEhck2gb48tEWikTGYmrTk/Ldh0I9G5I86b1cHZ25p1qxvVbslSZ+AsB+/bsJDKdF24iIyPRJ9yXdXqLuhcVGYFbgpx6vRuRkebr/O+rV/n1wnkqVzFO/S1dpmz8hYA9u3amaT+KjIzAzS3h9+rjLxDEl4lIXMa47zxv2ZUrltHCz4tRw4dw959/AIh++JDFC+fRtXvPVGe15maiOuDiquNGVMr366euX7vKn79foHylKhmSK7EbUcZ94ykXK/tUdpM4s7V2IDv452YU+RKci+Yr6Mo/tyzb0pOH9zOhVxvmThhA655DXmREkc1IB15khXLAXE3TqgJ3ge6m1y9rmvY+EAosBgKAasBYax+SkFKqs6njf2zfj8tSEUWz8mFWAr9Tk66fLyLgszGErDGfimiIjeH344co/5675YIZREtZTKsMhlj+/P0CHj4tmL1gJU5OuVi7fGEGhrOSLdFVSc1KoadXj+cs+oHFy9cy/as5/Lh6BT8fPwZArMHA3Xt3mbdkBT379GPEoH5o1lZEmiJbyZPCK6n3793l6MH9fLNsM/NW7+BRdDT7d2/LkFy2xNo6tChjZXsppQgL2Uf+AgUoXzHpDtuuHVvTPfpuzGD5msW2tvqrJF8fRo2dyOqVy2gT6M+Dhw9wdHRMc8YkMyTej1LyuyRh/ndzsLd3oKmnVxrCWWe1DiSR5+wvxwjesYnWncw7PY+iHzJz7CA+6NaX3HnSd4+2WTarbUUK2iWluHf3H8L2B7Nyw05+3BbEo+hodm03PlfAYDBw7+5dvl24nG69+zF6SP/0tUvW9hGLumctJ9y7+w8H9gezcsMO1m3bS3SCnANHjGXD2pV0/iCQhw8f4uiQ3vr5/Dbzee3qw4cPGNivN/0GDI6/z33kmAmsWbmcdq2a8zCN+1FS7UxKyiS3bGDL1mzZvptV6zbi4qpj+rTJgPHe+LbtPyR37jypzmpNeo5HT0U/fMi4of3o+ukA8mTgfpRQetqfLGP13Cn7ZbZ+PLXM+Vo1d4Z9uZyOgyaxbYXlLVPiv0MeYieywmVN057OU/8B6G36/02m/54GnDVNuwfcU0o9UkrlS+4DNU2bC8wFWHLscorPpvIWcOXuzWdXY+/diiJvvqSnRBavUJXb313j4b1/yJ3XOBXsz1+O4FaiLM4v50/p16aai6uOGwlGBW5ERaR42raLqx4XVx3lKhqvyld3b8C65em7HzIhV53ebNQkKjICF1edWRmdTm82dTdhGVfTfwsUKEjtug04f/Y0b7z1Njqdnjr1GqCUomLlqig7O+7cuR3/QKH0KOiiNxvhuBUVQYGCLila9tSJn9C5FeHlfMbtXa1WPX49dxL3hunvbNoSnc7NbFQvMuJ6/LaML6PXE3k9YRnjdg/as5PQ/cEcDAvhyZPHPHjwgFHDBjJmwlQAYmNj2Re0hyXL16Q/p17P9QT1MzLiOq46y5wRzymTWMlSpfhmrvFCWPjFvwgL2Z9s+ZTkTJghwsr61LuZ/y4RKcgJsGnjekJCgvlu/uIMPeEu4KLjZoL96GZUBPkLWO5H4f/7nbkzxzN4wmzyvpQv/vXY2Fhmjh1EjXpNeLdmvQzLBcZ2KdKizXFNVMbNShkdx44cplDhIuQztTW16tbnzKlfaOThhatOT+26xnapQqUq2Nkp/rlzO75sWnJGJdE2Pq/M8UQ5a9dtwNlTJ2nk4cUrJUrx+ZfG5x1cDr/I4QMhacr3lLF+JtiXIyMs9yOdG9cT5DTWYeM6j42JYWDfT2nS1It6DRrFlylRshRff2e87z2t+5Fe78b16wm/N8LKvpO4jHHfiYmJSXLZgi7P6rJ/QAt69+gKwOnTJ9m9eyezZnzOvXt3sVN25MyZk1Zt2qU6OxiP0Qm3742oSAq6PH+/fio2NoZxQ/tSr1FTatZpkKYMKZF4dt2NyAgKZtHtYynlkihzVGQEBV2zX+Z8BXXcSXAueudmFC9baUufKlPpdZZ9+Tf3797BOUGb+m+W3a8VvWgyAi+yQuIO9tOfH5v+G5fg/5/+nCkXmwqXKsft61e5E3kNQ2wM5w7vo+xb5lM8b12/Gn+V/vpfv2OIjSGX87OnOJ89lLnT5wHKlq/E31cucf3aVWJiYggN2sl7NeqkaNn8BV1wcXXjyqWLAJw8cYRiJTLuIXYVKlXmyuVL/H31CjExT9izcxs13c3XR033uuzYsglN0zhz6iR5nJ1xcXUlOvohDx48ACA6+iFHDh+kVOkyANSuW5/jR38C4FL4RWJjYsiXL2MukpQpX5FrVy8TYVqfYcG7eLt6ymZQuOjc+O38aR4/ikbTNE6fOELR4iUzJJctqVCpMpcvhcdv9907t1Orjvl2r+Vej21bNsZvd2fnvLi4utK9d1827wxmw7Y9jJs8nbffeS++8w5w9KdDlChR0mx6c1pVqlyFy+HhXL1izLlz+zbc65h3Ft3r1mPLJmPOUyd/wdk5r0UHILFbN4232sTFxTF/7hyaB7ZKd85Ll8xz1qmbKGed1Oc8EBbK4gXzmfXlt+TKlStdGRMrXa4i169eIvLaVWJjYji0fzdvvW8+Rf9G5HVmjh1Ij4FjKFT0lfjXNU1j7oxxFC5egmYBGf/U7PIVje3StatXiImJIWjXdmrUMq+fNWrVYec2Y7t09rSxXSro4orerRDnzpzikWkfP3H0J14xtZk13etx4tgRwNgxjomJib+YlxblKlbmyuVws5zVa9UxK1O9Vl2rOXVWcxrbotu3ntXPpQvn4u0fmOaMABUrVeFygvq5a8c2aidq593r1GXbZmP9PH3qF9P+rkPTNMaOHk7JUqVo90EHs2US7kcL5s2heYuWqc5m3HcucvXKZdO+sxV3q/vOBot9J7llE94SFLR3D2XKGO93X/T9crbvCmL7riDatvuQjp90SXPnHaBchUpcvXKJ638b68C+PTuoVjNlxyNN05gxcTTFSpSieWvLB9hmpHIVKnH1cjjXEuR8P1FdzW7KJ8ocvHuHxf6VHRQvU56oa5e5GfE3sTExnAjbQ+V3apiVibp2Jf5c9PKfv2KIjSFP3hf7TAGRfcgIvMgKxZVS72uadghoDYQBb2RFEDt7exp16MXKKYOJi4vjNfcmuBYtwYk9xmmIbzbw4tejoZwO3Y2dvQOOOXLg12t4/ChWzONHXDxzHI+OfTI1p72DA136DGJ0/+7ExcXRoKkPxUuWZvtG4wilh08Lbt+8Qd8ubXn44AF2dopNa5fx9ZJ15M7jTOdPBzFj/FBiYmJxK1yETwePec43ppyDgwN9Bw3jsx6dMcTF4entR6nSZVi/dhUAfgEtqV6zNofCQmjh44GTkxPDRhufonzr5k2G9DNOwDAYDDRs0oxqNYxP9fX08WPC6BG0beGDo6Mjw8dMyLDRQ3t7Bzr1Gsi4QT2JizNQz8OH4iVKs3Oz8Y8cNPYK4PatGwzs1p7ohw9QSrFl3QpmL1zDqxWq8H7t+vTv2hZ7ewdKlilHw2b+GZIrtZZM6kCtt8riks+ZP3aMY9ycbSzZcOiFfLeDgwP9Bw3j0+6fEBcXh6ePH6VKl+XHNSsB8G/Riuo1a3MwLIQA7yY4OTkxfPSEFH327p3bM2T6/NOcg4aOoEfXjsQZ4vD2a07pMmVZu9qYMyCwFTVruRMWEoJP00bGP381/tmf5BsysC/Hjx7lzp3bNKnvTtcevfD1D2DH9q2sXmm8Xade/Ub4+KavDjzN2b2LMaePKeeaVcacLVq2omZtd8JCQ/D2aIRTLuOf6Xpq8IBnORvXd6dr9174NQ9gyoRxPHnyhG6fGP/sZJWqrzF8VMbs//b2DnToOZBJQ3sTF2egTmNvipUoze4t6wBo6NmcH3+Yz/27/7DwS+MTk+3sHZj49ff8evYkoXu2UaxkGQZ3bQNAy4978Ma7NZL8vtRwcHCgz4Ch9O/dhbg4A029/ChZugwb1xnbJZ/mLalWozaHD4bSxt+DnE65GDzC+BT6ipWr4l6/IZ+0D8Te3p4y5crj5dcCgKbe/kwZN5wOrXxxcHRk6KiJ6WqXHBwc+HTAUAb07kpcnAGP+JyrTTkDqVajFj8dDKGtf1NyOjkxaMR4KzkdKFuuPJ6mnHt3bWeDaV+sVbc+Hl6+ac74NOeAIcPp1a0Thrg4vH39LfajGrXcORAWgq9nY5ycnBg11lg/T/58gm1bNlGm7Ku0CTQ+TLV7rz7UrOXOzh1bWbNyOQB16zfEOw37kYODA4OHjqRbl07EGQz4+DWnTJmyrFm1AoAWLVtTq7Y7YaH78fJoiFOuXIwx7TtJLQswa/o0fv31AgooXKQIw0c9926+NLF3cKBH3yEM/awbcYY4Gnn6UqJUGbasN9YBT79Abt28Qa+PW/PwwQOUnR0bVv3A3OXr+euP39i7YwslS5el24fGizQfdenFu9XT/3R8azl79hvKkD7diIsz0NiUc/OPxpxe/sacPT5qFZ/zx1U/MH/FBvLkcWbCyIGcOnGMf+7cobV3Az7o1B0P78w9dto7ONCr/1AGfWrM7JFE5m4dnmVet/IHFq7ckGm3IljNae9A8059+XZsX+OfkavfjELFSxG2cwMANRv7cvLQPo7u34G9vQOOOXLyYb8x2f8WBpFpVEbdUypESiilSgDbgBCgOvA70B44B7ytadoNpVQH0//3NC1zEXgbcAa2aJpWObnvSM0U+qxUrWjGPVU9M7nkzZHVEVLk2u1HWR0hRd7xGpzVEVLk7wOzszpCiuRwsI2JZLZynnXh73tZHSFFCufL2NkEmSUlz4rIDpxz2sZ4joO9bezvEXdt43hkZyMNk51txOTM33ezOkKKNankaiNr1ajcoJ3ZsjH9dUrjLFmPttFii38NTdMuAhWtvFUiQZnFGB9i9/Tnp+/dAJLtvAshhBBCCCHEv5VtXMoUQgghhBBCCCH+42QEXgghhBBCCCFEtmQjd3u8MDICL4QQQgghhBBC2ADpwAshhBBCCCGEEDZAptALIYQQQgghhMiW7GzlTxG8IDICL4QQQgghhBBC2ADpwAshhBBCCCGEEDZAptALIYQQQgghhMiW5Cn05mQEXgghhBBCCCGEsAHSgRdCCCGEEEIIIWyATKEXQgghhBBCCJEtKZlDb0ZG4IUQQgghhBBCCBsgHXghhBBCCCGEEMIGyBR6IYQQQgghhBDZksygNycj8EIIIYQQQgghhA2QDrwQQgghhBBCCGEDZAq9+Ncplz9vVkdIkZyOtnH97G50bFZHSBHnXLbRnP19YHZWR0iRwjU+zeoIKXL76FdZHeFfpULhl7I6Qorcf2wb7VJeJ8esjpAicXFaVkdIkTjNNnLmyWkbx6O8TraR00Y2O++WKJDVEf615Cn05myjByGEEEIIIYQQQvzHSQdeCCGEEEIIIYSwAbYxd0YIIYQQQgghxH+OTKE3JyPwQgghhBBCCCGEDZAOvBBCCCGEEEIIYQNkCr0QQgghhBBCiGxJZtCbkxF4IYQQQgghhBDCBkgHXgghhBBCCCGEsAEyhV4IIYQQQgghRLYkT6E3JyPwQgghhBBCCCGEDZAOvBBCCCGEEEIIYQNkCr0QQgghhBBCiGxJZtCbkxF4IYQQQgghhBDCBkgHXgghhBBCCCGEsAEyhV4IIYQQQgghRLYkT6E3JyPwQgghhBBCCCGEDZAOvBBCCCGEEEIIYQOkAy/+804dO8SgT1owoGNztqxeYvH+weAdDOvelmHd2zKuXycu/e83s/fjDAZG9GzPjFF9MzXn0cMH6NjKmw4tPFn1/QKL9y9d/Is+n7TH0/1t1ix/9ntERlxnQM+OdGrtyydt/Vi/almm5jxmyvlRoCerllrmvBz+F306t8erztusTZDzyePH9O7Uhm4ftqBzWz+Wzv8mU3Nm5/V56EAogb5NCfBuzPcL51m8r2ka06dMIMC7MW0Dfblw/pzZ+waDgQ9a+dOvd7f414YN6kv7ln60b+mHb9MGtG/pl+G5kzJnVFvC907i2JqhL+T7DoSG4N2sMZ5NGrJg3lyL9zVNY/LE8Xg2aUiAnxfnz5197rJffTGLAD8vAv196PLJx0RGRph95rW//6ba22+wZJFlXcoOOQ8dPECrFv409/WiVQt/fjp8KMU5AQ6EheDj2Rgvj4YsnG8965SJ4/HyaEiLxFmTWPbChfO0bxNIYHMf2gT6c/r0KQC2btlEYHOf+H9vVCnPhQvnU5UX4PDBUFr7N6OlTxOWLrK+H82aOpGWPk34sKUfvybYj+7du8vwgX1o4+9J2+ZenDn1S/x7a1cuo7V/M9q18Oab2Z+nOhe82G0fExPD8CGDaO7rha+XBwvmfZe2zGGh+Ho1wbtpo6TrwKTxeDdtRKC/t1nm0SOGUs+9OgF+XmbLzPnmSxrVr03LAF9aBvgSGrI/TdkS5/Qz5VyURM6pyeSs716dFoly/nrhPB+0bUmrAF/atmzOGVNdTY/Mqp8Ay79fRM23KnHn9u00ZcuM+vnUkkULeK1SOW7fvhX/2m+/XqB9m5b4eTejua8Xjx8/TlnOTGiXAFYsW4qPZ2P8fZoxc/pUAO7cuU2nj9rz/jtvMGnC2BTls+bQgVBa+pmO70ls9xlTjcf3doG+Ztvdr1kD2gb68EErPz5q2yL+9e+++YJ2gb580MqPT7t3IioqMs35siulsue/rCL3wIsMp5TqA8zVNO1hKpYpAWzRNK1yZuWyJs5g4PtvpjFwwpcUcNExuk8H3qhWiyLFS8WXcdUXZuiUb8mT9yVOHj3Ioi8mM2rWwvj3d21cReFiJYh++CDTchoMBr7+fCKTZn+Hi05Pr45tqFarDq+ULB1f5qWXXqLbZ4M4GBJstqy9vT2de/WnbLkKPHzwgJ4ft+LNd6uZLZuhOadPZOIsY87endpQraZ5zrymnIcS5XTMkYMpX8wnV+7cxMbG0K9bB96uVpMKlatmTs5suj4NBgOfTx7PF9/OR6fX81HbltRyr0vJ0mXiyxwKC+HypXDWbNzB2dOnmDpxDAuXrop/f9XypZQoWZoHD+7HvzZhyoz4/589fQrOznkzJG9KLN18mDmr9jN/3AeZ/l0Gg4GJE8by3bxF6PV62rQMoE7depQu82z9hYWGcCn8Ipu37+L0qZOMHzuaZSvXJLtsh4870bN3HwCW/fA93337NSNGPTuJmzZlEjVr1cq2OfPlz88XX3+LTqfn999/o1vnjuwJDk1x1knjxzJn3iL0bnratgzAvW49SpdOlPXSRTZtM2adMG40P6xYk+yys6ZPo0u3HtSs5U5oyH5mTZ/GgsVLaebpTTNPbwB+/+1X+vTuTvnyFVK8bp9mnjF5AjO/mYdOr6dT+5bUdK9LyVLPMh8+EMrly+Gs3LCds2dO8fmkscz7fiUAs6dN4r33azJ+6ixiYp7w6NEjAE4c/YnQ/UEsWbmeHDlycPvWzVTleprtRW773Tt38CTmCes2bCY6Ohp/72Y0adqMIkWKpirz5Alj+XbuQuN2bNXCeh0ID2fj1p2cPnWSiePHsHT5agC8fPxo2botI4YNtvjsdu0/5IMOHVO9HpPKOWXCWL4x5WxnylkqQc4DiXJOGj+G7xPlHJko5+wZ0+jStQc1atUmLGQ/s2dMY96ipenKmRn1EyDi+jWO/XQQvVuhNGfLjPoJcP3aNQ4dPEihQoXjPys2NpahgwcwYdI0ypUvz507t3FweH73JLPapaNHDrMveC9rftxMjhw5uHXTuI/nzJGTHr0+5Y/ff+ePP35P87qdPmU8s78xHt8/bmc6vifY7ocOJDq+TxrDgu+fHd+//m4x+fLnN/vcdh98TJfuvQFYvWIpC+d+w6Bho9OUUdgGGYEXmaEPkNvaG0op+xcbJXn/++0c+sJF0RUqgoOjI+/VbsiJQyFmZcpWrEqevC8BUKZ8ZW7dfHZl89aNCE4ePYB7Y59MzfnruTMULlqMQkWK4ujoSJ0GTTgUus+sTL4CBSlXsbLFga+giytlyxlPfnPnyUOxV0pxI5Ouzv56/gyFEuR0r28lZ/6ClKtQGftEOZVS5MptrDaxsbHExsZm2tXN7Lw+z505TdFixSlStBiOjjlo2NiDkH1BZmVC9gfR1NMHpRSVq77G/Xv3uBEVBRhnCBwM24+3X3Orn69pGnt376Rhk6YZlvl5Dpz4k1v/pPh6XrqcOX2KYsVeoWixYjjmyEGTps3YF7zXrExw0F68vH1RSlH1tde5d+8uUVGRyS7r7Owcv/yj6GizB+oE7d1D0WJFKV2mbLbNWaFCRXQ6PQBlypTlyeMnPHnyJOVZi5u+zzEHjT2asS/IPOu+4L14JpU1iWWVUjy4b7zwef/+PVx1Oovv3r5tK008PFOUM6HzZ09TtFix+P2oQaOmhO0zvxgXuj+IJs28jftRlde4f9+4Hz24f5+TPx/H09e4Dzk65iCv6Riwfu0q2nXoRI4cOQDIX6BgqrO96G2vlCL6YTSxsbE8fvwIB0dHnPM4kxrG7Vg8wXZsapF5f/BePL19LDIDvPX2O7z88supXlepdeb0KYo+J+e+tORUivumC6L379/D1dWyrqZGZtVPgC9nTKHbp/3S/NCvzKqfYLzQ+Vm/AWbZDh08QNlXy1GufHkA8uXLj739808VM6tdWr1qBR917By/jxcoaNzHc+XOzRtvvk2OnDnTsFaNzp05TdGiz47vDawd3/cF4ZHE8T0peRLs+9GJjk/i30k68CJdlFJ5lFJblVInlVJnlFKjgMJAsFIq2FTmvlJqrFLqJ+B9pVRfU9kzptH6xJ9ZSin1s1LqHaVUaaXUDqXUcaVUqFKqfEbmv30zkgIu+vifC7jouH0z6YZy/65NVH3r/fifl303k8CPe6LsMrexvBkViaveLf5nF1cdN6IiklnCuuvXrvLn7xcoX6lKRsaLdzMqElddgpw6HTdTkdNgMND9w0BaedblzXeqUb5Sxo++Q/Zen1GREegSZNPp3Symw0VFRqJzS1hGT5RpuuzMaZPp+Wl/lJ315v2XE8cpUKAgxV8pkWGZs5PIiAjcCpmvm4gI820bGRmBPsH60+vdiIyIeO6yX86eSaP67mzdspnuPT8F4OHDhyxaMI+u3Xpm65wJ7dm1k/IVKsSfoD43a2QEbmY59Ba3EERGJC5jyprMsgMGDWXm9Kk0ru/OjM+n0LuP5W1Iu3Zsw6NpsxTlTMi4Hz0bgXTV64lKtI/fiIw039d0em5ERfD31cvky5+fiaOH8VGb5kweO5LoaOMFqMuXLnLq5+N88kEren7yIefPnk51the97Rs0akyu3LloUKcmjRvU5cMOH/NyvnypyxwZYTaiq9e7EWUls1uiMonriTUrVywj0N+b0SOGcveff1KVK7GoRBl0pvWWOKc+UZmo5+TsP2gos6dPw6NBHWZOn0pPK3U1tTkzo36G7Q/CxVVP2VfTfqqUWfVzX9BedHpdfEf9qfCLf6GUousnHWkZ4MeiBZbTyq3mzKR2KfziRU4cP0a71i3o2KFdhtwu8VRUVIT5sVvnRlSk5fFdn2C7u+qe1Q2lFJ/26ESHNgFsWLfabLk5X83Cx6Meu7Zv4ZNuvTIsc3ahlMqW/7KKdOBFejUB/tY07TXT9PdZwN9AXU3T6prK5AHOaJr2HhANfAS8B1QDPlFKvfH0w5RS5YB1wEeaph0F5gK9NE17C+gPZOiN0Zpm+VpSO+T5k8cI2bWZlh8bT9R/+SmMl/IVoGTZ1E3tTAsNy6CpbTiiHz5k3NB+dP10AHlSOfqSUpqVFZqanPb29nyzZDU/rN/Fr+fOcPF/aZum9jzZeX1ay2ZRJon1HBayj/wFClC+YqUkl921Y+sLHX1/0VK0bZNYf89bttenn7Fr736aeXqxcvkPAHz79Ze0++BDcufJk61zPvXHH78za+bnZtP/n5s1Bft1UmWSW3bNqhX0HzSEnXv303/gEMaMHGZW7vSpkzjlykWZsq+mOOuzPJavpSQzSmEwGPjtwnl8A1qxaPk6nHLl4odF8wHjRcZ7d+8yd8kKun/aj5GD+1n/nOSyveBtf+b0Kezt7NgdHMq2nXv5fslCrly+nKrMVpsli/VppQjJt6stAluzedtuVq7dgIurKzM+n5K6XImk6JhudXMln3PtqhX0GziY7Xv20W/AEMaOHJ7mjJA59fNRdDRLFsylU9fUXUy0yJYJ9TM6Opp5c+dYvaBoMBj4+cRxJk2dxuKlywnauydFz+jIrHbp6T6+dPlq+vQbyMD+fVK9j6cus0UpyzKm+vndomUsWb6OGV99x7rVK/j5+LH4Ml179mHj9iAaeXiydmXmPutIZD3pwIv0Og00UEpNUUrV0jTN2uVzA8ZOOUBNYL2maQ80TbsP/Ag8vXHUFdgItNM07RellDNQHVijlPoF+A6welOXUqqzUuqYUurYhpWLUxy+gIuOWzeeXbG9dSOSfAVcLMpd+ut3FsyeSJ8R03B+yTi97rdzJ/n5cAj9Ovjy7ZThnD91jDnTRqX4u1PDxVVPVMT1+J9vREVS0CXlU/hiY2MYN7Qv9Ro1pWadBpkREQAXnZ6oyAQ5IyMpkIqcTznnfYmqb77DscMHMzJevOy8PnU6NyITZIuMuG4xXVOn1xN5PWGZCFxcdZz65QSh+4PxbdqAEYP7cezoT4waNjBB7lj2Be2hYWOPDM2cnej1bly/Zr5udLrE68+NiATrLyLiOq46XYqWBfBo5sme3bsAYydz1vTP8WhYj2VLlzB/7nesWPaDxTJZnRMg4vp1Puvdk/ETp1CsePHnZjTLapYjwqJO6t0Sl0mQNYllN29aT/0GjQBo1NjDYqRrx/atNPFI/eg7mPaRiGvxP0dFROCSaB931evN97VIYxlXnR5XnZ5KVYwzgOo2aMRvpofouer01K7XAKUUFStXRSk77txJ3YPCXvS23751C9Vr1sLR0ZGCBQvy+htvcjaVMwd0ej0R15+tz6d5zH8vPdefUyaxgi4u2NvbY2dnh3/zFpw5k/oZDYlzJswQaSVD4t/FWpnEtmzaQD1TXW3YuAlnz6RvVDYz6ufVK5e59vdVOrT2J8CzIVGREXzcNoCbN5Kffp1YZtTPK5cvcfXqFQL9ffBoWI+IiOu0CvDnRlQUOr0bb7/9LvnzFyBXrlzUrFXb7GFzyebMhHZJr9dTr0FDlFJUqVIVO2XH7TQ+DDAxnc7N/NgdeR2XRJlddXoiEmz3qMiI+DJPMxYoUBD3uvU5d9ayHjZq0ox9QbszJK/IvqQDL9JF07TfgLcwduQnKaVGWin2SNM0g+n/k7vM/Q9wGahh+tkOuKNp2usJ/lkd7tY0ba6maW9rmva2b6sOKc5f8tUKRPx9majrfxMbE8NPIbt5o1ptszI3I6/z5fjBdOk/Greiz056Az/qwaylW5i+eAPdBo2nQtW36TpgTIq/OzXKVajE1SuXuP73FWJiYti3ZwfVarqnaFlN05gxcTTFSpSieevMfYhYufKV+DtBzv17U57zzu1b3L93F4DHjx/x89HDFMukad7ZeX1WqFSZy5fC+fvqFWJinrB753Zq1alrVqaWez22bdmIpmmcOXUSZ+e8uLi60r13XzbvDGbDtj2Mmzydt995jzETpsYvd/SnQ5QoUdJsWua/TaXKVbh06SJXrlwm5skTdmzbinvdemZl6tStx+ZNG9A0jVMnf8HZOS+urrpklw0Pvxi//L7gIEqWND7ocvHS5WzfHcT23UG0bf8hnTp3oXXbdtku5927d+nZrTOf9unLG2++laZ1evXKZWJinrBzu2VW9zr12JJMVmvLurrqOHb0CABHfjpsdltHXFwcu3ftSHMHvnzFyly+fCl+P9qzaxs13M33o5q167Jj6ybjfnT6JM7Ozri4ulLQxRWd3o1LF/8C4NiRw5QoZXxIZe069Tlx9CcALoVfJDY2hnz5zB8o9Twvetu7FSrEkZ9+QtM0Hj58yOmTJ+PfS1Xm8HCuXrli2o7bqFMnUR2oW48tmzZaZE5OwtuDgvbuSdVzJJLKeTlRTvcMyOniquP4sWd1tVjxV9KVMzPqZ+myr7JlTyhrt+xm7ZbduOr0LFy2loIurqnKlhn1s+yr5dgXeii+rdTr3Vi59kdcXF2pUaMmv/32K9HRxuc0HD921Oyhg8/LmdHtUt16DTh65DBgnN4fExND/vyp28eTUqFSZS5ffnZ837NzO7XcLY/v2xMc3/OYju/R0Q958MD4zJDo6If8dPggpUob95fLly7GLx8WEswrJVK3f9uCrHjCfEr+ZRV5Cr1IF6VUYeCWpmk/KKXuAx2Ae0Be4IaVRUKAxUqpyRg7835Ae9N7TwBfYKdS6r6macuVUn8ppVpomrZGGec3VdU07WRG5be3d6B9t/5MG96buLg4ajfyougrpQja+iMA9Zr5s2H5Au7f+4fvvzF2hOzs7BnzxZKMipCynA4O9Og7hKGfdSPOEEcjT19KlCrDlvXGe6A8/QK5dfMGvT5uzcMHD1B2dmxY9QNzl6/nrz9+Y++OLZQsXZZuHwYC8FGXXrxbPeVPzE5Nzu6fDWFYX/OcW005m5ly9u6YIOfqH/hu2Xpu3bzB9PHDMcTFocXFUbteI96rkbJOdVpyZtf16eDgQP9Bw/i0+yfExcXh6eNHqdJl+XGN8enD/i1aUb1mbQ6GhRDg3QQnJyeGj56Qos/evXN7lkyfXzKpA7XeKotLPmf+2DGOcXO2sWRD6v6MWUo5ODgwZNhIunXuRFycAV+/5pQpU5bVq1YAENiyNbVquxMWsh9Pj4Y4OeVi7PiJyS4LMHvGdC5e/As7O0WhQkUYPip9F+tedM6Vy3/g0uVLzJ3zDXPnGO9E+nbeQgoWfP5D2BwcHBg8dCTdunQizmDAx/R9a0xZWzzNGrofL4+GOOXKxZhxE5NdFmDkmHFMnTwRQ2wsOXLmNJvWf/zYUfR6N4oWK5bm9dt34DD69uxMnCGOZj5+lCpdhg1rjU9z9g1oyfs1a3PoQAgtfTxwcnJi6Ojx8ct/NnAoY4YPIjYmhsJFijLE9F4zHz8mjRlB+0AfHB0cGTZ6Qqpvv3nR275V67aMHD4Efx9P0DR8/Px5tVzq7pF2cHBg0NARdO/akThDHD5+zSldpixrVhvbpRaBrahZy52wkBC8mzbCycmJ0abMAIMH9uX40aPcuXObxvXd6dqjF37+Acye8Tm/XjiPUopCRYowfGT696tBQ0fQw5TT25RzrSlnQIKcPlZyDkmQs4kpp69/ACNGj2Pa5AkYDAZy5szJ8FTcgpJUzsyonxkhs+pnUl56+WXaf9iBNi0DUEpRq1ZtarvXSVHOzGiXfP2bM2r4UJr7euLo6Mi4iZPj93GPRvV4cP8+MTExBAft4du5C82eep+SzP0GDaNPD9Px3dt0fF9rOr4HPDu+t/BpQs4Ex/dbN28yuJ/xSfMGQyyNmjTj/RrG845vvpjJpfC/UMoOt0KFGTgsc2aDiuxDZdR9HeK/SSnVGJgGxAExQDfgfaAHcE3TtLqmzrhzgmX6Ah+bfpyvadosleDPyCml8gG7gfHAKeBbjFPnHYGVmqYle+Q8/Ocdm6jUbvmcsjpCithKE2ErD119OZdjVkdIkcI1LO9VzI5uH/0qqyP8q9jK/n7/cWxWR0iRvE62MU4SF2cbG942UsLDJ4bnF8oGbKV+2kq7FG0j2x2gQB57GzlrMnp34r5sWQuODK2TJevRNvZckW1pmrYT2Jno5WPAlwnKOCdaZgYwI9FrF4HKpv+/A7yT4O0mGRZYCCGEEEIIYTPkT+OZk3vghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSzKA3JyPwQgghhBBCCCGEDZAOvBBCCCGEEEIIYQNkCr0QQgghhBBCiGxJnkJvTkbghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSzKA3JyPwQgghhBBCCCGEDZAOvBBCCCGEEEIIYQNkCr0QQgghhBBCiGxJnkJvTkbghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSzKA3JyPwQgghhBBCCCGEDZAOvBBCCCGEEEIIYQNkCr3416lU5KWsjpAidna2MR/oXnRsVkdIkRwOtnE90lZy3j76VVZHSJH87/TM6ggpcvPIl1kdIUUMBi2rI6SIk43sR3FxtrE+sY3DEfY2Mo/W0d42ct649ySrI6RIAWfHrI6QInGajezvNkieQm/ONo6AQgghhBBCCCHEf5x04IUQQgghhBBCCBsgU+iFEEIIIYQQQmRLMoXenIzACyGEEEIIIYQQNkA68EIIIYQQQgghhA2QKfRCCCGEEEIIIbIlmUFvTkbghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSPIXenIzACyGEEEIIIYQQNkA68EIIIYQQQgghhA2QKfRCCCGEEEIIIbIlmUFvTkbghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSPIXenIzACyGEEEIIIYQQNkA68EIIIYQQQgghhA2QDrz4Tzp4IBR/bw98PRuzeME8i/c1TWPa5An4ejamVYAPF86fBeD69Wt06fghAb7NCPTzZMWy7+OX+e3XC3zUvhUtm3vzWa9u3L9/P905D4SF4ufVBO+mjVg0f67VnFMnjce7aSMC/b05f+5s/HujRwylvnt1Wvh5mS0zqP9ntArwpVWAL80a16NVgG+6c/50KIx2AZ608fdg2ZL5VnPO/nwibfw9+KiNH79dOBf/3url3/NhSx86tPJlzPABPH782GzZlT8swv3dyty5czvdOQ8dCKWlX1MCvBvz/SLr233G1AkEeDemXaAvv55/ltOvWQPaBvrwQSs/PmrbwmLZZd8v5P03K3LndvpzZsZ2/+3XC3zYtiWBfl582rNrmurngdAQvJs1xrNJQxbMs55r8sTxeDZpSICfl1mupJb96otZBPh5EejvQ5dPPiYyMsLsM6/9/TfV3n6DJYsWpDpvas0Z1ZbwvZM4tmZopn+XNQfCQvH1bIK3RyMWJrHdp0wcj7dHIwL9Em334UOpV7s6Ab7m233m51Px8/Ig0M+bvr17cu/u3XTnTGv7+fjxYz5oE0jrFr4E+nny3Tdfxi8ze8Y0mvs0pVWAD/37ZG3OrGjnfU37e5LbPZn9vZ57dQIS7e8AK5YtxderCc19PZk1Y1rG5Mzg+rl75w6a+3jyZpUKnD1zOu3ZMqFtemrJogW8Vqkct2/fAuDOndt07NCeam+/wcTxY9Oc+dCBUFr4NKW5V2OWLLReP6dPmUBzr8a0beHLhQTHIwCDwUD7lv707dUt/rU5X39B2xa+tAv0o1fXTkRFRqY53/McORTGh4FetA9oyorvLY/7ly7+j56d2tKk1pusXrY403I8ZSvtZ0KHD4bSyr8ZgT5NWJrEOcnMqRMJ9GnCBy39zM5J7t27y7CBfWjt70mb5l6cOfVLhmbLbpTKnv+yinTgxX+OwWBgysRxfPHNXNas38zOHVv5359/mJU5EBbC5UvhrN+8g2EjxzDJdJB2sLfns/4DWbthK4t+WMWalcvjlx0/ZgQ9P+3LqnWbqFOvAUsXp6/DYTAYmDJhLF9+M491G7ewY7uVnKEhXAoPZ+PWnQwfNZZJ48fEv+fl48dX31oeEKZ8PpOVazewcu0G6jdoRL36DdOdc9bU8Uyd/S1LVm1i785tXPzfn2ZlfjoYypXLl1i2bhv9h4xmxpRxAERFRrBu1TLmLlnF4pUbiDPEEbR7e/xykRHXOPbTIfRuhdKV8WnO6VPGM+PL71ixbjO7d2zjr/+Zr89DB4zbfc3GHQwePoapk8aYvf/1d4v5fuV6Fi1bY/Z6xPVrHD18CLcMypkZ233sqOH07tOP1es3U7d+Q75PZYfYYDAwccJYvpkzn/WbtrJj2xb+/MM8V1hoCJfCL7J5+y5Gjh7H+LGjn7tsh487sXb9Zlb/uJHa7nX47tuvzT5z2pRJ1KxVK1VZ02rp5sP49Pj6+QUzgcFgYPL4sXz17TzWbdrCjm1b+fNPK+v3Ujgbt+1k+OixTByXYLv7+vH1HMvtXu396qxZv5nV6zfxSokSVk9sU5szre1njhw5mDN/ESvWbGD56vUcPBDGadNJ53vVqrNq3SZWrt1I8VdKsGhB1uV80e385Alj+SrB/m51uyfY3ycm2t+/trK/Hz1ymH3BQaxet4l1G7bwwYcfpz9nJtTP0mXKMn3WF7z51tvpypYZbRPA9WvXOHTwIIUKFY5/LUeOnPTo9Sl9BwxMV+Zpk8Yz6+vvWPnjZnbt2GZRPw+a6ufaTTsYPGIMUyeYH49WLV9KiZKlzV5r9+HHLFuzgR9Wr6dmbXcWzP0mzRmfl/+LzycwaeY3LFyxkaBd27n4l/lxP+9LL9Oz7xBatOmQKRkS57GF9jNx5umTJzD9izksW7uJPTutnZOEcuVyOKs2bGfg8NF8PunZBaNZ0ybx3vs1WfHjFpasXMcrJUtlWDaR/UkHXlhQSi1WSgVkwOeMVkppSqkyCV77zPRaskdrpdS+55VJq7NnTlGsWHGKFi2Go2MOGjVpyv59QWZl9gcH0dTLB6UUVaq+zr17d7kRFYmLq47yFSoBkCdPHkqUKh0/Yhh+8S/efOsdAN57vzpBe3enK+eZ06coWrw4RYsZczb2aMq+4L1mZfYF78XT25iz6mvGnFFRxivub739Di+//HKSn69pGrt37qBJ02bpynn+7GmKFC1O4SLFcHR0pF4jD8JCzNdnWEgwjZt6o5SiUpXXuH/vHjdvRAFgMMTy+PFjYmNjefwoGhcX1/jlvpo5la69+mbIw0vOnTlN0aLFKWLa7g0aexCSaLuH7AvCw9O4PitXNea8ERX13M+ePX0KPfr0y5DLsZm13cMv/sWbbxvrZ7X3q7N3z65U5ypW7BVjrhw5aNK0mUWu4KC9eHn7WuRKbllnZ+f45R9FR5tt66C9eyharCily5RNVda0OnDiT2798/CFfFdiZ06folji7R5kvn73p2G7v1+jJg4OxufVVqn6GhER19OVMz3tp1KK3LnzABAbG0tsbAwK4/auVr2GWc7EMzFeZM4X3c5bbPfg9G/3NatW8lHHT8iRIwcABQoWzPicGVA/S5UuTYl0djwyq20C4wXEz/oNMGuXcufOzZtvvU3OHDnTnPncmdMULfbseNTwOcejKlVf416C41FExHUOhO7Hx7+52TIJ29PoRO1pRrpwzvy4X7ehBwdDgs3K5C9QkPIVK8fv15nJVtrPhM6fPU3RYsXi60D9Rk0J3We+DsP2B9GkmfHcqXKV17h331gHHty/z8mfj+Pla9z+jo45yJv3pQzLlh3ZKZUt/2XZ+siybxb/Gkqp5Frn00CrBD8HAOeSKPtCREZGondzi/9Zp9MTGWF+shgVGYGb/lkZvd6NyERT0f6+epVfL5yncpXXAONIwtMTxD27dhJx/Vq6ckZFRpiN6Or0bhY5IyMjzEandXo3olJ44nvi+DEKFCxI8VdKpCvnjahIdAnWlatOz40o83V1IzLCokxUZASuOj2t2nUg0LsB/k3rksc5L+9UqwHAgZBgXFx1lHm1fLryPRUVFYHObLu7WUwvjIqMRJ84Z5RxfSql+LRHJzq0CWDDutXxZUL3B+Gq01E2o3Jm0nYvXaYs+4NN9XPnjlTXz8iICNwKJVh/ej0RVnMl2m8iIp677JezZ9Kovjtbt2yme89PAXj48CGLFsyja7eeqcppqxJvU72VbRoZYV439FbqRnI2rl9HjZq105kzfe2nwWCgTaAfDevW5L1q1alc9TWL79i04Ueq10jfrAtbaeetbncr+5XFdn/O/h4efpGfTxyjfZtAOnZol67p6UnmzOD6meZsmdQ27Qvai06vo1z5jGnbk8tjbMOtHI/czLM9Xeczp02mZ5/+KGV5Gv/tl7PwalyPndu20LlbrwzPDsbjvqsu8XE/87d1Umyl/UwoKjICnT7hcfzZ+cazMubnVzrTOcnVq5fJlz8/E0YPo0Ob5kwaO5Lo6Ky5+CyyhnTgbZxSaoRS6oJSardSaoVSqr9SqrRSaodS6rhSKlQpVd5UdrFS6gul1EGl1P+ejrIro6+UUueUUlsBXYLPf0sptd/0WTuVUoVMr+9TSk1USu0HPk0m4gbAx7RMKeAfIH5IUyn1rVLqmFLqrFJqjLUPUEo1UkodUkqdUEqtUUo5WyuXYppm7TvMi5B8mYcPHzCwX2/6DRgcf8V75JgJrFm5nHatmvPw4QMcHR0zOqbl1XQrZSBlVwR3bt+a7tF3MI7kPy9DUuvz3t1/CNsfzMoNO/lxWxCPoqPZtX0zjx5Fs3TRXD7uknGdN2s5LS+eWilj+l2+W7SMJcvXMeOr71i3egU/Hz/Go+hoFi/4jk+6ZtxJUmZt91FjJ7J65TLaBPrzIA3183n7hLGQ9TLPW7bXp5+xa+9+mnl6sXL5DwB8+/WXtPvgQ3LnyZOqnDbL6m6UaD9KSd1Iwvzv5mBv70BTT8t7pVMlne2nvb09y1evZ9uuYM6eOc0fv/9mVm7BvDnY29vj0Sxrc8KLaefTvN2fs78bDAbu3r3L98tW8Vm/gQzs3yeJtjqTc76AEarMaJuio6OZN3dO/AXFDJeC45HV7aUUYSH7KJC/ABUqVrL60d169WHzziAaN/VkzcplGZHWkrX8KTz3yBS20n6mMo/18xaFwWDgtwvn8QtoxeLl68iVKxdLF1k+h0D8e8nfgbdhpinmzYE3MG7LE8BxYC7QVdO035VS7wHfAPVMixUCagLlgU3AWsAPKAdUAfQYR8gXKqUcgS8BH03TopRSLYEJwNOb6fJpmub+nJh3gctKqcoYO/KrgI8SvD9M07RbSil7YK9SqqqmaacS/I4uwHCggaZpD5RSg4C+gNmTY5RSnYHOALO/+paPOnZOMpBOryfi+rNpUJGREbjqdOZldG5cTzBVKiLiOq6uxqndsTExDOz7KU2aelGvQaP4MiVKluLr74z3Q4Zf/IuwkP3PWTXJ0+n1XE8wuhMZcd0yp15vNgJkrYw1sbGxBO3ZzbJV69KVEYxX3iMTrKuoyAhcXF0TlXGzUkbHsSOHKVS4CPnyFwCgVt36nDn1C6XLluPa31fp2LZ5fPlP2rdgzqKVFHRxSVNOnc6NSLPtfh0XV/N15arTm02Re5oTwNX03wIFCuJetz7nzp4i70svce3qVdq38osv36FtcxZ8v4qCLubrIMU5M2m7lyxVim/mLgTSVj/1ejeuX0uw/iIi0FnkcjPbtyJMuWJiYp67LIBHM096dutC9569OX3qJHt27WTW9M+5d+8uStmRI0dOWrdtl6rctiLxNjW2OebrSO9mXjciUri/b9q4npCQYL6bvzjdHar0tp9P5X3pJd56510OHQyjTNlXAdiyaQNhIfv4du6iLM/5Itt5i+2eKKden/rtrtfrqd+goWnqbVXslB23b9+mQIECGZczg+pnemVG23Tl8iWuXr1CoL9PfPlWAf4sW7nG4viWFonzREZYHo8s6nBEBK6uOoL27CRkfzAHw0J4/OQxDx48YNTQgYyZONVs+cYezejbqxudu2f8KLyLTk9UpPmxsqBr5m/rpNhK+5k4c2REwuN4BC4ulnUg4blTZKSxjFLG85VKVaoCUKdBI374l3fgs/KBcdmRjMDbtprARk3TojVNuwdsBpyA6sAapdQvwHcYO+1PbdA0LU7TtHMYO+sAtYEVmqYZNE37G3h6I1Y5oDKw2/RZw4GiCT5rVQpzrsQ4jd4XWJ/ovUCl1AngZ6ASUDHR+9VMrx0wZfgQeCXxF2iaNlfTtLc1TXs7uc47QMVKVbh8KZyrV64QE/OEXTu2Udu9rlkZ9zp12bZ5I5qmcfrULzg758XFVYemaYwdPZySpUrR7oMOZsvcunkTgLi4OBbMm0PzFi2TzfE8lSpX4XL4s5w7t2/DvU49szLudeuxZZMx56mTxpyJD1rW/HT4ECVKljSbnpdW5StW5srlS1y7eoWYmBiCdm2nRi3z9VmjVh12btuEpmmcPX2SPM7OFHRxRe9WiHNnTvHoUTSapnHi6E+8UqIUpcu8ysadIazauItVG3fhqtMzb+maNHfeASpUqszly+H8fdW4Pvfs3E6tRNu9lns9tm8xrs8zp06SxzkvLq6uREc/5MGDBwBERz/kp8MHKVW6LGXKvsq2vWGs37qH9Vv34KrTs3jZujR33iHztnvC+jl/7hyaB7ZKtry1XJcuXeTKlcvEPHnCjm1bca9rnqtO3Xps3rTBIldyy4aHX4xffl9wECVN98MuXrqc7buD2L47iLbtP6RT5y7/2s47PF2/5tu9TqL1614n9dv9QFgoixfMZ9aX35IrV65050xP+3n71q34pzg/evSII4cPUaJEScD4xPgli+YzY/Y3OGVxzhfdzl9KtL/XyYD9vU69Bhz56SfAeKEhJiaG/Pnzpy9nJtTPjJAZbVPZV8uxL/RQfBuk17uxcu2PGdJ5B9Px6NKz49Hundst6mfC49HpUydN9dOVHr37smVXMBu272H85Om8/c578Z33Swna09D9wZn2YLPyFSpz9XI41/42HveDd2+neq06mfJdKWEr7WdCT8+dntaBvbu2UTNRHahZuy47thrPnc6cPomzszMurq4UdHFFp3cj/OJfABw/cpgSpUpb+xrxLyUj8LbN2vUoO+COpmmvJ7FMwr/RlXD5pOZBn9U07f0kPuvBcxMabQamAcc0Tbv79AqmUqok0B94R9O020qpxRgvQCTOsFvTtNYp/K7ncnBwYMCQ4fTq1glDXBzevv6ULlOWtatXAhAQ2Ioatdw5EBaCr2djnJycGDV2IgAnfz7Bti2bKFP2VdoEGkddu/fqQ81a7uzcsZU1K5cDULd+Q7x9/dOdc9DQEfTo2pE4Qxzefs0tctas5U5YSAg+TRvh5OTE6PET45cfMrAvx48e5c6d2zSp707XHr3w9Tc+m3DX9q00aeqZrnwJc/YZMJT+vbsQF2egqZcfJUuXYeM64/Udn+YtqVajNocPhtLG34OcTrkYPML4FPqKlaviXr8hn7QPxN7enjLlyuPlZ/kn2jIqZ79Bw+jT4xPi4uLw9PajVOmy/LjWuD79A1pRvWZtDoaF0MKnCTmdnBg+egJgPGkf3K83YHzoXqMmzXg/nffoJpczM7b7ju1bWW2aTlmvfiN8Ulk/HRwcGDJsJN06dyIuzoCvX3PKlCnL6lUrAAhs2Zpatd0JC9mPp0dDnJxyMdaUK6llAWbPmM7Fi39hZ6coVKgIw0dZvZPmhVgyqQO13iqLSz5n/tgxjnFztrFkw6EX8t1Pt3v3Lsbt7mPa7mtWGbd7i5atqFnbnbDQELw9GuGUy4nR455t98EDnm33xvXd6dq9F37NA5gyYRxPnjyh2yfGiVNVqr6WrnWcnvbzxo0oRg0fQlycgbi4OBo2ahJ/EW3qpPHEPHlCj64dAahc5TWGjhidJTmzop3v3jXRdjflbJFgf/e2sr8PHphou/fohZ9/AL5+/oweMYwAPy8cHR0ZO2FyukYPM6t+Bu3ZzZRJ47l96xa9u3elXPnyfDM3dU/2z6y2KTkeDetx//59Y+c1aA9z5i6kdJkyz10uYeb+g4fRu5vxeOTl40epMmX5cY3peNSiFTVqGY9Hzb2a4OTkxIgxE577uV9/MZNLF//Czs4Ot0KFGTRsVIozpYa9gwO9+g9l0KddiYsz4OHpR4lSZdj8o/H5MF7+gdy6eYNuHVry8MEDlJ0d61YuZeHKjeTJk767IK2xlfYzcebPBg6jb8/OGAxxePr4Uap0GdavNZ47+QW05P2atTl0IIRAHw+cnJwYOnp8/PKfDRzKmOGDiI2JoXCRombviX8/la57okSWUkq9g3GEvTrGizHHgXmAPzBT07Q1ynjErqpp2klTB3mLpmlrTcvf1zTNWSnlD3QBmmK8//0c8AnGKfbngPaaph0yTal/VdO0s0qpfUB/TdOOJZNvNHBf07TPlVKtgN80TTvxdFkgBvge4y0ArsApYJCmaYsTlAk3/V71NE37QymVGyiqadpvFl9ocu9RnE1Uajs725gPdC86NqsjpEgOB9uYUJTT0TZy2ttI/cz/jm085O7mkS+fXygbMBhsovm0GbayH2Xl7cupkZVPfU6NRzGGrI6QIvcf2UbOAs7pfNbEC/LwsW2sTwAXZwfb2JlMGn/zU7Y8OO3s/l6WrEcZgbdhmqYdVUptAk5i7Ogew/iQuLbAt0qp4YAjxinsJ5P5qPUY75E/DfwG7Dd9/hPTg+6+UEq9jLG+zALOpiHrSiuvnVRK/Wz6vP8BB6yUiVJKdQBWKKWe/s2W4aacQgghhBBCCPGfISPwNk4p5axp2n3TyHQI0FnTtBNZnSsryQh8xpIR+IwlI/AZS0bgM5aMwGcsW9mPZAQ+Y8kIfMaSEfiMJyPwGUNG4EVazVVKVcR47/iS/3rnXQghhBBCCPHvYSvXQl8U6cDbOE3T2mR1BqXUMCDxk8fWaJr2/CeuCCGEEEIIIYRIEenAi3QzddSlsy6EEEIIIYQQmUg68EIIIYQQQgghsqX0/BnMfyPbeJqSEEIIIYQQQghhQ5RSTZRSvyql/lBKDU6iTB2l1C9KqbNKqf3P+0wZgRdCCCGEEEIIITKQUsoe+BpoCFwBjiqlNmmadi5BmXzAN0ATTdMuKaV0z/tc6cALIYQQQgghhMiWbHgG/bvAH5qm/Q9AKbUS8AHOJSjTBvhR07RLAJqmRT7vQ2UKvRBCCCGEEEIIkQpKqc5KqWMJ/nVOVKQIcDnBz1dMryX0KpBfKbVPKXVcKfXB875XRuCFEEIIIYQQQohU0DRtLjA3mSLW5g5oiX52AN4C6gO5gENKqcOapv2W1IdKB14IIYQQQgghRLakrPaDbcIVoFiCn4sCf1spc0PTtAfAA6VUCPAakGQHXqbQCyGEEEIIIYQQGesoUFYpVVIplQNoBWxKVGYjUEsp5aCUyg28B5xP7kNlBF4IIYQQQgghhMhAmqbFKqV6AjsBe2ChpmlnlVJdTe/P0TTtvFJqB3AKiAPma5p2JrnPlQ68EEIIIYQQQohsyc5mZ9CDpmnbgG2JXpuT6OdpwLSUfqZMoRdCCCGEEEIIIWyAdOCFEEIIIYQQQggbIFPohRBCCCGEEEJkS0rZ8Bz6TCAdePGvY29vGzu5nY00RvnyOGZ1hBSJi0v8ZzWzKdvY7Dbj5pEvszpCihR8t1dWR0gRW1mfNvwnhbIlGzkc2Uw7b28jK7SAs20c321lf8+d0z6rI4j/CJlCL4QQQgghhBBC2AAZgRdCCCGEEEIIkS3ZyKSWF0ZG4IUQQgghhBBCCBsgHXghhBBCCCGEEMIGyBR6IYQQQgghhBDZkq08+PlFkRF4IYQQQgghhBDCBkgHXgghhBBCCCGEsAEyhV4IIYQQQgghRLYkM+jNyQi8EEIIIYQQQghhA6QDL4QQQgghhBBC2ACZQi+EEEIIIYQQIltSMofejIzACyGEEEIIIYQQNkA68EIIIYQQQgghhA2QKfRCCCGEEEIIIbIlmUFvTkbghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSncyhNyMj8OI/6UBYKL6eTfD2aMTC+XMt3tc0jSkTx+Pt0YhAP2/Onzsb/97o4UOpV7s6Ab5eZsvs3rmD5j6evFmlAmfPnE5brtAQvJs1xrNJQxbMs55r8sTxeDZpSICfl1mu5y27ZNECXqtUjtu3bwFw9eoV3n2zKoH+PgT6+zBuzMgszfnVF7MI8PMi0N+HLp98TGRkhNlnXvv7b6q9/QZLFi1IcU6L3GGh+Ho1wbtpMtt90ni8mzYi0D/Rdh8xlHru1QnwM9/uX385m0B/b1oG+NKts2XuNOfM4Po58/Op+Hl5EOjnTd/ePbl3927qc73A7X7o4AFatfCnua8XrVr489PhQ6nOG//d2XR9ptWcUW0J3zuJY2uGvrDvTCg7r88DYSH4eDbGy6Nhstm8PBrSInEdTWbZFcuW4uPZGH+fZsycPhUw1tHWgf4E+HnROtCfIz+lvI5mRs5vv/6ShvVqEdjch8DmPoSG7Df7zGvX/ub9d5JvQzNjH//nzh26dPoIL49GdOn0EXf/+QeAmCdPGDFsCM19vWjh583RIz9ZfF/vHl3x9/FMMq/V3yET2vmnvl+8gDeqlOf27dupymTNwQOh+Ht74OvZmMUL5lnNOW3yBHw9G9MqwIcL5405Hz9+zAdtAmndwpdAP0++++bL+GX27NpBoJ8n77xekXNnz6Q7I2Tf8yXLnC9u3099trSvw6SW/fXCBT5o25IWfl582qMr9+/fN/vMa9f+pvo7b/J9Os6ZRPYlHXjxn2MwGJg8fixffTuPdZu2sGPbVv788w+zMmGhIVy6FM7GbTsZPnosE8eNiX/Py9ePr+dYHmxLlynL9Flf8OZbb6c518QJY/lmznzWb9rKjm1b+PMPK7nCL7J5+y5Gjh7H+LGjU7Ts9WvXOHTwIIUKFTb7vKLFirP6x42s/nEjI0aNzdKcHT7uxNr1m1n940Zqu9fhu2+/NvvMaVMmUbNWrRRlTCr35Alj+eqbeazbuIUd25PY7uHhbNy6k+GjxjJxfILt7uPH199abvcPP+rI6h83sWrtBmq512HunG/SnDE+ZybUz2rvV2fN+s2sXr+JV0qUsHoS8bxcL3K758ufny++/pZ1GzYzbuJkhg0ZmKq8CXNnx/WZHks3H8anx9fPL5gJsvP6NBgMTBo/lq+/nc+PT+uZ1WwX2bRtFyNGj2PCuNHPXfbokcPsC97Lmh838+PGrXzYoSMA+fPnZ/ZX37J2/WbGTUh5Hc2snADt2ndg9bqNrF63kVq13c0+8/Mpk6iRTBuaWfv4wvlzefe999m8fRfvvvc+C0zbdt3aNcb/btjMnPmLmD5tCnFxcfHftWf3LnLnzpOidZrwd8iMdh7g+vVrHD50ELdEx9G0MBgMTJk4ji++mcua9ZvZuWMr/0uU80BYCJcvhbN+8w6GjRzDpPHGY3SOHDmYM38RK9ZsYPnq9Rw8EMbpU78AxvOQqTO/5I00nodYy5kdz5es5XyR+35qs6V1HSa37NhRw+ndpx9r1m+mbv2GFhfmnre/C9smHfj/OKXUYqVUQAZ9Vmel1AXTvyNKqZoJ3uujlMqd4Of71j8l8505fYpixYtTtFgxHB1z0NijKfuC9pqV2R+8F09vH5RSVH3tde7du0tUVCQAb739Di+//LLF55YqXZoSJUulL1exV4y5cuSgSdNm7As2zxUctBcvb1+LXM9bdtqUSXzWbwAqA6YgZVZOZ2fn+OUfRUebZQ3au4eixYpSukzZ9OVOvN2D07/dE+aOTpQ7w3JmQP18v0ZNHByMd01VqfoaERHXU5/rBW73ChUqotPpAShTpixPHj/hyZMnqcocnzsbrs/0OHDiT2798/CFfV9C2Xl9GrO9kiBbM4ts+4L34plUHU1i2dWrVvBRx87kyJEDgAIFCwJQPkEdLZ2KOppZOZMTtHcPRYoWpXTppNvQzNrHg4P34u3rC4C3ry/BQXsA+N+ff/BetWoAFCxYkLx583L2jHHU+OGDByxdsohPunR77u9m8TtkQjsP8PnUSXzad0CGPA377JlTFCtWnKJFjTkbNWnK/n1BiXIG0dTLmLNKVWPOG1GRKKXiL2zExsYSGxuDwhiqZKnSlChRMv0BTbLr+ZL1nC9u3099trStw+SWDb/4F2+9/Q5gvAC6d/eu+M8L3ruHokWLUbp0mVTnza5UNv2XVaQDL1JFKWX1uQlKKU+gC1BT07TyQFdguVLKzVSkD5Db2rIZlSGlIiMj0LsViv9Zr3cjKtG058iICNwSlYmMSP/U6GRzRUTgVsgt/medXk9Eou80Zn9W5mmu5JbdF7QXnV5HufLlLb7z6tUrBDb35eMP23Hi+LEszQnw5eyZNKrvztYtm+ne81MAHj58yKIF8+jarWeK8iWZ29p2t5LbYrunYEr8V1/MpEmDOmzfuoVuPXpnfM4Mrp8b16+jRs3aqcv1grd7Qnt27aR8hQrxJ1Gpyp1N16etys7r07j/Jqx/eov915jNSh1NZtnwixc5cfwY7Vq3oGOHdpw5fcriu/fsTnkdzaycACtXLKOFnxejhg+Jn6oe/fAhixfOo2v35NvQzNrHb928iaurDgBXVx23bhlv43q1XHn2Be0lNjaWK1cuc/7cWSKuXwOMtyZ90OFjnHI5JZvZ4nfIpHZ+X3AQOp2ecuUsj6NpERkZabYedTq9xT4SFRmBmz7Ruo40dowNBgNtAv1oWLcm71WrTuWqr2VILsuc2fN8KbGs3PdTki2t6zC5ZUuXKcu+YONFn927dsTvO9EPH7Jo4Ty6dO+R6qzCdkgH3oYopUaYRrd3K6VWKKX6K6VKK6V2KKWOK6VClVLlTWUXK6W+UEodVEr97+kouzL6Sil1Tim1FdAl+Py3lFL7TZ+1UylVyPT6PqXURKXUfsDy7NpoEDBA07QbAJqmnQCWAD2UUr2BwkCwUio4wfdNUEqdVEodVkrpTa+5KqXWKaWOmv7VML0+Wik1Vym1C/jeyrrprJQ6ppQ69typl5rVlWtexEqZjBi9To5mJZjFd1oJppRKctno6GjmzZ1jtVPk6qpj555gVq/bQP+Bgxk8sJ/FPVQvKudTvT79jF1799PM04uVy38AjPd1tvvgQ3LnSd1USivBLaVku6fgGmvP3p+xY88+PJp5smrFD2kM+DSEldcysH7O/24O9vYONPW0fo9n0rFe7HZ/6o8/fmfWzM9TfIuHleCWssH6tFnZeH1qSdS/lJRJblmDwcC9u3dZunw1ffoNZGD/Pmbl//jjd2bP+JzhI1NWRzMrZ2DL1mzZvptV6zbi4qpj+rTJgLENbdv+w+dOR8/MfdwaX//m6PVutAlszrTJE3nt9Tewd7DnwvnzXLp0ifoNGia7vFWZ0M5HR0ezYN6cdF+cfV4IizqQzDq1t7dn+er1bNsVzNkzp/nj998yLpt5CEvZ4Hwpsaza91MWzsprKV2HySw7etxEVq9YRptAfx4+eICjoyNgOmdq3yHVt58I2yJPobcRSqm3gebAGxi32wngODAX6Kpp2u9KqfeAb4B6psUKATWB8sAmYC3gB5QDqgB64BywUCnlCHwJ+GiaFqWUaglMAD42fVY+TdPMb6gzV8mUJ6FjwIeapvkrpfoCdZ928IE8wGFN04YppaYCnwDjgdnATE3TwpRSxYGdQAXTMm9hHOGPTvzlmqbNNa0LHsYk37rq9Pr4K5UAERHX40cHntK76bmeuIzOvExG0+vduH7t2bTRyIgIdIm+U6d3I+L6szJPc8XExFhd9srlS8ZRdn+f+PKtAvxZtnINLq6u8aNFFStVplix4oRf/ItKlau88JyJeTTzpGe3LnTv2ZvTp06yZ9dOZk3/nHv37qKUHTly5KR123bJ5kzM6nZP9N16ffq2u0dTT3r36JquE73MrJ+bNq4nJCSY7+YvTvUJ1ove7gAR16/zWe+ejJ84hWLFi6cq77NM2XN92qrsvD71ejeum9W/CCvZEpdJUEeTWFav11OvQUPjVOYqVbFTdty+fZsCBQoQcf06fT/tybhU1NHMylnQxSX+df+AFvTu0RWA06dPsnv3TmbNMLahdsqOnDlz0qqNeRuaWft4gYIFiYqKxNVVR1RUJAUKFADAwcGBAYOfPYjxg7atKF68BMePHeH8uTN4NKxHrCGWWzdv0bFDexYsXprsejXmy/h2/ulxtGWAT/zv1ibQn6UrVuPi4vrcTEnnTLC+IiMsMuh0blyPSLSuXc2/L+9LL/HWO+9y6GAYZcq+mqYsz8+Z/c6XEsuKfT+l0rMOY2Jikly2ZKlSfDtvIWCcTv/0oZVnTp9iz+6dzJoxjXv37mGn7MhhZX+3Nf+VY2xKyQi87agJbNQ0LVrTtHvAZsAJqA6sUUr9AnyHsdP+1AZN0+I0TTuHsbMOUBtYoWmaQdO0v4GnN12VAyoDu02fNRwomuCzVqUhs8L69UOAJ8AW0/8fB0qY/r8B8JUpwybgJaVUXtN7m6x13lOrUuUqXLoUztUrV4iJecLO7duoU7eeWRn3OvXYsmkjmqZx6uQvODvntWhwM5ox10WuXLlMzJMn7Ni2FfdEuerUrcfmTRssciW1bNlXy7Ev9BDbdwexfXcQer0bK9f+iIurK7du3cJgMABw5fJlwsMvUrRosSzJCRAefjF++X3BQZQ03R+3eOny+Pxt239Ip85dUt15j88dnmi710m03eumfrsnzL0/OIgSJdN3/2Fm1c8DYaEsXjCfWV9+S65cudKY68Vt97t379KzW2c+7dOXN958K9V5zXNnv/Vpq7Lz+nxaz65euWzKZllHjdmSrqPWlq1brwFHjxwGjCfKMTEx5M+fn7t379Kre2d6p7KOZlbOp/cdg/Ge9zKmZ4Ys+n4523cFsX1XEG3bfUjHT7pYPZnPrH28Tt16bNqwAYBNGzZQt259wDiy/fCh8VkOhw4ewN7entJlyhDYqg179oWxfXcQi5cu55USJVLUeY//HTK4nS/7ajmC9h9k284gtu0MQqfXs3z1j2nuvANUrFSFywn2o107tlHbva55zjp12bbZmPP0KWNOF1cdt2/div8rDY8ePeLI4UMZet97Qtn1fMl6zhe376c+W9rWYXLL3rp5E4C4uDjmfTeHgMBWACz8fhnbdgWxbVcQbdt9QMdPOtt8511YkhF422Ht0pMdcEfTtNeTWOZxEstbnZQDnNU07f0kPuvBc/KdwzhCnvApLG+aXrcmRns2D8nAs7poB7yfuKNuuvL2vAwp4uDgwKChI+jepSNxhjh8/JpTukxZ1qxaCUCLlq2oWdudsNAQvD0a4ZTLidHjJsYvP3hAX44fPcqdO7dpXN+drt174dc8gKA9u5kyaTy3b92id/eulCtfnm/mpvzPdzg4ODBk2Ei6de5EXJwBX7/mlClTltWrVgDG6ZG1arsTFrIfT4+GODnlYuz4ickum5wTx47y9Vdf4GBvj529PcNHjuHlfPmyLOfsGdO5ePEv7OwUhQoVYfioMUlmSIv47d410XZfbdruga2oWcudsJAQvJs2wsnJidHjE2z3gYm2e49e+PkH8MWs6YRfvIidUhQqXJhhI9KXO7Pq55QJ43jy5AndPjFOqqlS9bVUreMXvd1XLv+BS5cvMXfON/FP9v923kIKpvIhQtl1fabHkkkdqPVWWVzyOfPHjnGMm7ONJRvS/mf2UiM7r08HBwcGDx1Jty6diDMY8DHVszWmOtriaR0N3Y+XR0OccuVizLiJyS4Lxqneo4YPpbmvJ46OjoybOBmlFKtWWNbROXMXPvdBV5mVc9b0afz66wUUULhIEYan8raTzNrHP+7UmQF9+7Dhx7W4FSrE5zNmA3Dr1k26de6InZ0dOp2eCZPT9ie6Ev8OmdHOZzQHBwcGDBlOr26dMMTF4e3rT+kyZVlryhkQ2Ioatdw5EBaCr2djnJycGDXWmPPGjShGDR9CXJyBuLg4GjZqQi1T5z94726mTZ7A7du36NOzK6+WK89Xc+anK2d2PF+ylvNF7vsvah0mtSzAjm1bWbVyGQD1GjTCx88/zetP2B6V6ns5RJZQSr2DcYS9OsbO7nFgHuCPccr5GmVsVapqmnZSKbUY2KJp2lrT8vc1TXNWSvljfNhcU4z3v5/DOH19k+n/22uadsg0pf5VTdPOKqX2Af01TUvyKWdKKW9gBNBE07SbSqnXTZ/5nqZp15RSpwFvTdP+SpjH9P8BgKemaR2UUsuBnzVNm2Z673VN035RSo0G7mua9vnz1tXzptBnF3YyHShDxcXZxGbP2seWpoKt1M8429jdKfhur6yOkCI3j3z5/ELZQEqeTSFSzkZ2d5tp5w02ktPe3jY2vK3s79aeW5Bd5Xa0lb3eqO3SX7Llyl3W/vUsWY8yAm8jNE07qpTaBJwEwjHeX/4P0Bb4Vik1HHAEVprKJGU9xnvkTwO/AftNn//E1JH+Qin1Msa6MQs4m8J8m5RSRYCDSikNuAe00zTt6c07c4HtSqlrmqbVTfKDoDfwtVLqlClDCMYn2gshhBBCCCHEf5qMwNsQpZSzpmn3lfHvqYcAnU1PexcJyAj8f5OtjMzYyECCzdRPGYHPWDIC/99kI7u7zbTzMgKfsWxlf5cR+MwjI/DmZATetsxVSlXE+PC6JdJ5F0IIIYQQQvybyVPozUkH3oZomtYmqzMopYYBLRK9vEbTtAlZkUcIIYQQQggh/iukAy9SxdRRl866EEIIIYQQQrxg0oEXQgghhBBCCJEtyQx6c3ZZHUAIIYQQQgghhBDPJx14IYQQQgghhBDCBsgUeiGEEEIIIYQQ2ZI8hd6cjMALIYQQQgghhBA2QDrwQgghhBBCCCGEDZAp9EIIIYQQQgghsiU7mUFvRkbghRBCCCGEEEIIGyAdeCGEEEIIIYQQwgbIFHohhBBCCCGEENmSPIXenIzACyGEEEIIIYQQNkA68EIIIYQQQgghhA2QKfTiXyfyn8dZHSFF9C87ZXWEFLGVWUtxWlYnSJnfrt3L6ggpUqHwS1kdIUUMBtvY8DePfJnVEVKk4Lu9sjpCinw8skdWR0gRt7yOWR0hRTq/+0pWR0gRZycbOW21kePmzXtPsjpCitjK9OnIfx5ldYQUe7OEbRzjn7KNGvDiyAi8EEIIIYQQQghhA6QDL4QQQgghhBBC2AAbmYskhBBCCCGEEOK/xs5GbqN4UWQEXgghhBBCCCGEsAHSgRdCCCGEEEIIIWyATKEXQgghhBBCCJEtyQx6czICL4QQQgghhBBC2ADpwAshhBBCCCGEEDZAptALIYQQQgghhMiWlMyhNyMj8EIIIYQQQgghhA2QDrwQQgghhBBCCGEDZAq9EEIIIYQQQohsSWbQm5MReCGEEEIIIYQQwgZIB14IIYQQQgghhLABMoVeCCGEEEIIIUS2ZCdz6M3ICLwQQgghhBBCCGEDpAMvhBBCCCGEEELYAJlCL4QQQgghhBAiW5IZ9OZkBF785x09fICOrbzpEOjJqqULLN6/FP4XfTq3x7PO26xZviT+9ciI6wzo2ZFObXz5pK0f61cvS3eWA2Eh+Hg2xsujIQvnz7V4X9M0pkwcj5dHQ1r4eXH+3NnnLvvt11/SsF4tApv7ENjch9CQ/QDcuXObTh+15/133mDShLGpyxkagnezxng2aciCedZzTp44Hs8mDQlInDOJZXft3I6fdzNer1yes2dOx79+9eoV3n2zKoH+PgT6+zBuzMhUZU3oYFgo/l5N8GnWiEULrOeeOnk8Ps0a0bK5d3zu69ev0bnjBzT3aUoLP0+W//C9xbLfL17AW1XLc/v27TTns+aXowfp+3Fz+nTwY+PKxRbvh+3dzsAurRnYpTUj+3xM+J+/AXAz8jrjBnSlX8cW9P8kkO3rV6Q7S2bUzwsXztO+TSCBzX1oE+jP6dOnANi6ZVN8nQ1s7sMbVcpz4cL5NOU+eCAUf28PfD0bs3jBPKu5p02egK9nY1oF+HDhvDH348eP+aBNIK1b+BLo58l333wZv8zsGdNo7tOUVgE+9O/Tk3t376YpW0IHwkLx9WyCt0ejZNevt0cjAv28zdbv6OFDqVe7OgG+XmbLzPx8Kn5eHgT6edO3d8bkTKk5o9oSvncSx9YMfWHfmRIVdXkY2aAUoxuWpuGrBS3eL+uSm889X2VI3ZIMqVsSj3IuLyzb32ePsXFMZzaM6sSZXauTLHcj/DeW9fQi/ESY2etxcQa2TupF8LejMzzbTwfDaNvck9Z+HvyweL7F+5qmMfvzibT286BDaz9+vXAOgEsX/+LjNs3j/zWp8x6rly8FYNSQfvGvB3o34uM2zTM084GwUPy8muDdtBGLktinpk4aj3fTRgT6J2rzP/4Af++mBPhab/PTy3g88sC32XPapWaNadXchwsJsnXp+CEBPs0I9PNkRYJsQwZ8RpsWfrRp4YdXk/q0aeGXoZmPHAqjQ0svPghoxorvrZwvXfyLXp+0w6P2W6xettjsvWnjRxLQ1J1ObTM20/McORTGh4FetA9oyorvLevtpYv/o2entjSp9aZF5sz2y9GD9O1oOr6vsvzusKDtDOzamoFdkzi+d8q447uwLTICL/7TDAYDX0+fyKRZ3+Gi09OrUxuq1azDKyVLx5d56aWX6PbZIA6GBJsta29vT+de/SlbrgIPHzygZ8dWvPlONbNlU5tl0vixzJm3CL2bnrYtA3CvW4/SpcvElwkLDeHSpYts2raL06dOMmHcaH5Ysea5y7Zr34EPP+po9n05c+SkR69P+eP33/njj99TlXPihLF8N28Rer2eNi0DqFO3HqXLJMoZfpHN2405x48dzbKVa5JdtkyZV5k5+0vGjRll8Z1FixVn9Y8bU7lGLXNPnjiWb+YuRK/X0751C9zr1KNUgvV7ICyEy+HhbNiykzOnTjJp/Bi+X74ae3t7Pus3iAoVK/HgwX3atWpOtferxy97/fo1fjp8ELdChdOVMbE4g4FFX01l6OSvKOiiZ1ivD3nr/doUfaVUfBmdW2FGfv4dznlf4pcjB5g3ayLjv1yMnb0D7Tr3oWTZ8kQ/fMDQHh9Q5c33zJZNjcyqn7OmT6NLtx7UrOVOaMh+Zk2fxoLFS2nm6U0zT28Afv/tV/r07k758hXSlHvKxHF8/d0C9Ho9H7QJpHadupbb/VI46zfv4Mzpk0waP5Yly1aRI0cO5sxfRO7ceYiNiaFjh3ZUr1mLKlVf571q1enR+zMcHBz4YubnLFowl96f9U/Tun2ac/L4sXw7b6FpHbVIYv2Gs3HbTk6fOsnEcWNYusLYyfPy9aNlm7aMGDrY7HOrvV+dXn364uDgwOwZn7Nw/lw+7Zv2nKmxdPNh5qzaz/xxH7yQ70sJBQS+5saXBy5xJzqGgXVLcvraPa7fe2JW7o+bD5lz6MoLzRYXZ+DI6m+p32s8ufO5sH3qZxStUo18hYpblPt5wyIKVXjT4jMuBG/iZbdixDx6mKHZDAYDM6eOZ8ZX83DVu9H5w5bUrF2XEqWeHe8OHwzlyqVLLP9xG+fOnGLG5HF8t3gFxUuUZOHydfGf07xpPWrXrQ/AmEnT45f/auY0nJ2dMzTzlAmmNt9NT7tWxn3KbN8PDeFSeDgbtxr3KbM2v/+zNr9tS/M2P0OyTRzH13NN7VLrJNql8HDWb9lhOh6NZcnyVTjY2/NZv4GUr1iJBw8e0L5Vc94zZZs0bWb88jM/n5Lh6/PL6ROZMnsurjo9PT5uTfVa5udLeV96iR6fDeZgSJDF8o2beePbohVTxg7LsEwpyfzF5xOY+sVcXHVudP+oFe/XqksJs8wv07PvEA7st8ycmeIMBhZ9PZWhkxIc36slOr7rCzNymun4fvQA82ZPZPwXVo7vPdN3fBe2R0bgsyml1EWl1HMv+yul8imlur+ITKbvu6iUOq2UOqmU2qWUckui3DalVL40fP4L/X1+PX+GwkWLUahIURwdHalTvwmHQveZlcmXvyDlKlTGwcH8eldBF1fKljN2KHLnyUOxV0pxIyoyzVnOnD5FseKvULRYMRwdc9DYoxn7gvaaldkXvBdPb1+UUlR97XXu3btLVFRkipZNLFfu3Lzx5tvkyJkz9TmLmb4rRw6aNG3GvmDz7woO2otXUjmTWLZU6dKUKJl5B5+zZ05RrHhxihY1rqNGTZpa5N4fvJdmXj4opajy2uvcN+V2ddVRoWIlAPLkcaZkydJERkbELzdj6iQ+/WxAhk/x+uPXs7gVLoa+UFEcHB15370hxw7uNyvzaqXXcM77EgBlKlTh1g1jHcxf0IWSZcsDkCt3HooUL8GtG1FpzpJZ9VMpxYP7DwC4f/8erjqdxXdv37aVJh6eacp99swpihUz3+7795mfqO0PDqLp0+1e1Zj7RlQkSily584DQGxsLLGxMSiMG7la9RrxbUKVqq+Z1Ye0MK6j4gnWUVOL9bs/eC+e3j4W6xfgrbff4eWXX7b43Pdr1DTLGRFxPV05U+PAiT+59U/GdiTTq0SBXEQ9eMLNhzEYNDh+5S5VC+XN6lgA3Lz4G3ldC5PXpRD2Do6UeKs2V04dtij3677NFH+9Bk55zbf3g9s3+PvMUcpUb5zh2c6fPU2RYsUpXLQYjo6O1G/oQViiDk/Y/mAaN/NGKUWlKq9x/949biRqc44fPUzhosUsLnZqmkbwnh3Ub9w0wzKfOX2Koon3qWBrbZblPmW1zY9I3z6ekLXj0f7gZNql1561Sy6uOsrHZ8tDiUTHIzCuzz07d9DYo1mGZf713BkKFy1O4afnSw2acCDRwEb+AgUpX7Ey9g6W44NV33ibvC9ZtlGZ6cK50xQpWpzCRYz1tm5DD4vBmKeZE5/jZTaL43udhhw7lMzxvXwyx/di6Tu+2wKlVLb8l1WkA58NKaXsU1E8H/DCOrwmdTVNew04BpjNjVRGdpqmNdU07U4aPjsfL/D3uRkViavu2TUIF52OG1GpP0hfv3aVP3+/QPlKVdKcJTIyAje3Z1n0er3FQTkyInEZNyIjIp677MoVy2jh58Wo4UO4+88/ac4Yn6HQs+/S6fVEJDqxiYyMQG8tZwqWtebq1SsENvfl4w/bceL4sTTn1usLmWWKSrx+IyPQuz0ro7NS5u+rV7hw4TyVq7wGGE+yXHV6Xi1XPk25knP7RhQFXfXxPxd01XP7ZtIH6X07NvL6O9UtXo+6/jcX//iVMuUrpTlLZtXPAYOGMnP6VBrXd2fG51Po3aevxXfv2rENj6ZpOxGNjIw0q4s6nd7iRDwqMgI3faLckcYTJYPBQJtAPxrWrcl71apTueprFt+xacOPVK9RK035nuU0r3tW62dEBG6JyqSmU7Fx/Tpq1Kydrpy2Lp+TA7ejY+N/vhMdQz4nyxP3kgVyMaReSbq/X4xCeXO8kGwP79wkd/5n1+1z53Ph4Z2bicrc4PLJQ5St5WGx/PG1c3nD76NMuVn0RlQkugT7iKteH3/x6FmZCPMyOj03EtXhoF3brXbST/58nAIFC1Ks+CsZljkq0nx/0VnZX1La5v964bzVfT+tIiMi0evNj4WJ29Moi3bzWbv0LNtVY7Yq5tl+Pn6MAgULUvyVEhmW+UZUBDrds+ORq07PzXQMWrwINxKd47nq9Gk6x8sMt28mOr676LmdTCc82eP7n+k7vgvbIx34DKaUGqiU6m36/5lKqSDT/9dXSv2glGptGsE+o5SakmC5+0qpsUqpn4D3E7yeSym1Qyn1SRJfORkorZT6RSk1TSm1VCnlk2D5ZUopb6VUB6XURtNn/aqUGpWgTDul1BHTZ3yXigsIIUAZpVQJpdR5pdQ3wAmg2NMZBEqpKQlH1JVSo5VS/ZRSzkqpvUqpE6b18TSz2e9jWmaAUuqoUuqUUmpMEuu9s1LqmFLq2HIr92UlRdM0a5+V4uUBoh8+ZNywfnTtPYA8edI+XS0lWZIqk9yygS1bs2X7blat24iLq47p0yanOSOARgrWWVI5U7JsIq6uOnbuCWb1ug30HziYwQP7cf/+/dSFBivfbG39Jl/m4cMHDOjbm/4Dh+Ds7Ex0dDQL5s2ha4/eqc6TEtbWV1In52d/OUbwjk207tTT7PVH0Q+ZOXYQH3TrS+5sWD/XrFpB/0FD2Ll3P/0HDmHMSPPpladPncQpVy7KlH01rcGfnzuZemlvb8/y1evZtiuYs2dO88fvv5mVWzBvDvb29ng087L4jNTltPJaKutncuZ/Nwd7eweaeqYz579Q4tV6+c4jRu74g0lBf7H/f7foXK1YFiWx3N2PrZ3LG74fYWdnfpi+cvoITnlfpmDxspmTLB37/1MxMTEcCNlH3fqNLMrt3bWN+o0ybvTdmMfyNctjlbUlzdv8/p/1pt+gIRk6Hd36tk7d+nz48AED+/am38DBFtl2bt+aoaPvxjxWXszuTxaztg7JHpmtbd9kj+87N9G6o5Xj+7hBfNA1fcd3YXvkHviMFwL0A74A3gZyKqUcgZrA78AU4C3gNrBLKeWradoGIA9wRtO0kRDfSDsDK4HvNU1L6gkqg4HKmqa9blrOHfgM2KiUehmoDnwItAPeBSoDD4GjSqmtwAOgJVBD07QYUye8LZCSJ7Z4Ak+fNlYO+EjTtO4J8mPKPwv4xvRzINAEeAT4aZp2VxlvFTislNpk5fdpBJQ1ZVfAJqVUbU3TQhIG0TRtLjAX4OKNR1YPyda46PRERT6bUnojMpKCLpZTeJMSGxvDuGF9qdeoKTXrNEjxctbo9W5cv/4sS0REBK6u5ln0bonLXMdVpyMmJibJZQu6PBvR8Q9oQe8eXdOf89qz74qMiECXaNqzTu9GRFI5n7NsYjly5CBHDuMIWMVKlSlWrDjhF/+iUuXUzXbQ6/VERFwzy+SSeP3q9URcf1YmMkGZmJgYBvTtjUczL+o1MJ6AXrl8ib+vXqF1C5/436dtS3++X74aFxfXVOWzpoCLjpsJRgtuRkWQv4DlnTXh//uduTPHM3jCbPK+lC/+9djYWGaOHUSNek14t2a9dGXJrPq5edN6Bg4xdtobNfZg7KjhZp+5Y/tWmqTjRFSn15vVxcjICItp+jqdG9cjEuV2Nd9+eV96ibfeeZdDB8PiLyZs2bSBsJB9fDt3Ubqn0ukS1T1jhsTrV8/1xGWes/8AbNq4npCQYL6bvzhLp/xlB3cexZI/17NTn3y5HPnnUaxZmUexcfH/fzbiAS0V5Mlhz4MnhkzNljufCw9v34j/+eGdG+R62fwhezcv/UHYQuO1/8f373L17DHs7O25cfFXrpz+iatnj2GIeULMo2jCFk+jZocBGZLNVacnMsE+EhURYdHGuerczMtERlAwQR0+fDCUsuUrUKCgeRsWGxtLSPAe5n2f9EP70kKnN99fIq3sL4n3u4RlYmJi6P9Zb5o286J+A8uLDunNlvB2lkgr7anOos191i7FxsQwsO+nNElwPHoqNjaW4L17WLpybYZmdtWZzxKIioygYAYc5zJT4nO8xHUyK1kc329EkL9gEsf3WeMZPN7K8X1cxhzfbYGMOJuT9ZHxjgNvKaXyAo+BQxg78rWAO8A+TdOiNE2LBZYBT+czGoB1iT5rI7Aomc67BU3T9mMcFdcBrYF1pu8C2K1p2k1N06KBHzFeVKiP8YLCUaXUL6afn3cjcrCp7EvAJNNr4ZqmWdysp2naz4BOKVVYKfUacFvTtEsYO+MTlVKngD1AEUCfeHmgkenfzxhH98tj7NBniHLlK3H1yiWu/32FmJgY9u3dQbWa7ilaVtM0ZkwaTbFXStG8Vfof0lSpchUuXbrI1SuXiYl5ws7tW3Gva94ou9epx5ZNG9A0jVMnf8HZOS+urrpkl004zTFo7x7KlEnf6nv6XVeuXCbmyRN2bLPMWaduPTYnkzO5ZRO7desWBoPxxPnK5cuEh1+kaNHUj4hVrFSFy+HhXL1yhZiYJ+zasQ33OubfXbtOPbZu3oimaZw++QvOeY25NU1j3KjhlCxZmnYffBRfvuyr5diz/yBbdgSxZUcQOr2eZat+zJDOO0DpchW5fvUSkdeuEhsTw6H9u3nrffMp0DcirzNz7EB6DBxDoaLPpp9qmsbcGeMoXLwEzQLapjtLZtVPV1cdx44eAeDIT4fNpnzGxcWxe9eOdHXgK1aqwuVL5tu9tnvdRLnrsu3pdj9lzO3iquP2rVvxT21/9OgRRw4fokSJkoDxyfZLFs1nxuxvcMqVK835njKuo2c5d27fRh2r63ejxfpNzoGwUBYvmM+sL78lVwbktHXht6PROeegYG5H7BW8VfQlTl+7Z1bmpZzPRrdfye9kfE5DJnfeAQq+8ir3Iq9y/8Z1DLExXDweQtEq75mV8Ru7EL9xi/Abt4jib9Tg3ZbdKfba+7zh0wH/Cd/jN24RNT8ehFu5qhnWeQcoX7EyVy4ZL1jGxMSwd/d2atQ2349q1q7Dzq2b0DSNs6dPksfZ2awt3LtzGw2sjLIfP3KY4q+UMpt+nxEqVTZv83dut2zz3eta36c0TWPsqOGULFWadh9+lMQ3pJ2141HtOsm0S6bjkUvCbCVL0e6DDhaffeTwIUqULGl261BGKFehElcvh3Pt6fnSnh1Ur1UnQ78jo5WvUNksc/Du7dkmc/zx/brp+L5vN29VS+L4PiCJ43uxEjRrnv7ju7A9MgKfwUyj2BeBj4CDwCmgLlAauISxs2zNI03TEp8hHAA8lFLLNatzbZK0FOMoeivg44TxEsfF2JFeomnakFR8fl1N0+KHCZTxYXUPkim/FggA3DCOyGPK5wq8lWCdOVlZVgGTNE37LhX5UszewYEenw1haN9uxBniaOTpS4lSZdiy3jgS4OkXyK2bN+jVsTUPHzxA2dmxYfUPzF22nr/++I29O7ZQsnRZun0YCMBHXXrxbvW03Qvr4ODA4KEj6dalE3EGAz5+zSlTpixrVhn/PEiLlq2pVdudsND9eHk0xClXLsaMm5jssgCzpk/j118voIDCRYowfNSzPxnn0ageD+7fNx7Ygvbw7dyFZk+9TirnkGEj6da5E3FxBnxN37XalDPwac6Q/Xh6NMTJKRdjx09MdlmAvXt2M3niOG7fukXP7l0oV64Cc+Yt4MSxo3z91Rc42NtjZ2/P8JFjeDlfvjSt34FDR9CzW0cMhjh8fJtTukxZ1q42VsmAwFbUrOXOgdAQfJo1wsnJidGm9fvLzyfYumUjZcq+SusWvgD06P0ZNWul7GJPWtnbO9Ch50AmDe1NXJyBOo29KVaiNLu3GK/1NfRszo8/zOf+3X9Y+KVxVM7O3oGJX3/Pr2dPErpnG8VKlmFw1zYAtPy4B2+8WyNNWTKrfo4cM46pkydiiI0lR86cjEhQP48fO4pe70bRYmmfwuzg4MCAIcPp1a0Thrg4vH39LbZ7jVruHAgLwdezMU5OTowaa8x940YUo4YPIS7OQFxcHA0bNaGWqfM/ddJ4Yp48oUdX4193qFzlNYaOGJ2unIOGjqB7l47EGeLw8TPWzzWrjDlbtGxFzdruhIWG4O3RCKdcz+onwOABfTl+9Ch37tymcX13unbvhV/zAKZMGMeTJ0/o9onxUFCl6msMH2X1TqQMt2RSB2q9VRaXfM78sWMc4+ZsY8mGQy/ku5MSp8Hqk9fpUaMYdigOhd/h2r0n1CyRD4Cwi3d4o8hL1CqZH4OmEWPQWHj06gvJZmdvzzuB3dj79Qi0uDhKv9+QfIVf4bfQbQC8Witjp5inhoODA30GDqV/7y7EGQw09fajZOkybFy3CgCf5i2pVqM2hw6E0trPg5xOuRgyclz88o8eRXPsyCH6D7X8KyN7d22nQWPLe/ozIvOgoSPo0dW4T3n7WW/zw0JC8GlqavPHJ2jzNxvb/FYBvgD07P0ZNWtnTJvv4ODAgKGmdsmQTLsUGoJvM1O7ZNrfT/58gm1bNlGm7Kvxfyaue+8+8cejXTu20SiDp8+D8XypV7+hDO7Tjbg4A01M50ubfzT9JQx/4/lS949axZ8v/bjqBxas2ECePM5MGDmQkyeO8c+dO7TybsCHnbrj4e2f4TktMvcfyqBPuxIXZ8DD089q5m4dWsZnXrdyKQtXbkzXLZEpymbvQIceCY7vjawc35fN5/69f1j4VYLj+1em4/te0/G9m+n4/lHaj+/C9qjU9QtFSiilRmPsOH+McYr5UYwj892BwzybQr8T+FLTtI1Kqfuapjkn+IyLGEfuRwA5NE3rlsR3FQROaJr2SoLX9MAR4Lqmae+ZXusATMQ4hT4a+MmU7yHGkf4amqZFKqUKAHk1TQtP4vsuAm8n6sCXALZomlbZWjmlVCVgHuACuGuadk0p9SlQRtO0XkqpukAQUBK4l/D3MU2hHwfU1zTtvlKqCBCjaVqST05JzRT6rKR/2dr1iuzHVmbcxhpsYrPz2/V7zy+UDVQo/FJWR0iRWEPc8wtlA/b2trEjFXy3V1ZHSJGPR/bI6ggp4pbXMasjpEjndzPu4XGZydnKAwezozgbObe+8yAmqyOkiK3c+hP5z6OsjpBib5Z4yTZWqknvDRey5U71hW/5LFmPMoU+c4QChYBDmqZFYLzfO1TTtGvAECAYOImxo/q8P27dB3BSSk219qamaTeBA6aH4k0zvRYBnAcWJSoehnF0/heMU+uPaZp2DhiO8X78U8BuU/YMo2naWSAvcNW0DsB4+8DbSqljGEfjL1j7fTRN2wUsBw4ppU5jHM3PHn/zRwghhBBCCCFeINu4lGljNE3bCzgm+PnVBP+/HGOHNPEyzol+LpHgx2RvwNI0rU3Cn5VSuTHeJ74iUdFITdN6JnoNTdNWAauS+44kcj197SLGkf0ky2maViXRzzdI8LT9RO+1SfTzbGB2SvIJIYQQQgghxL+VdOD/ZZRSDYCFwAxN09L3B7+FEEIIIYQQIgvZ2dSE/8wnHXgbYbrXfa+Vt+qbpp0DoGnaHqB44kKapi0GFqfi+34CciZ6ub2maaetlRdCCCGEEEIIkbmkA28jTJ3011/g9733/FJCCCGEEEIIIV4U6cALIYQQQgghhMiWZAq9OXkKvRBCCCGEEEIIYQOkAy+EEEIIIYQQQtgAmUIvhBBCCCGEECJbUkrm0CckI/BCCCGEEEIIIYQNkA68EEIIIYQQQghhA2QKvRBCCCGEEEKIbEmeQm9ORuCFEEIIIYQQQggbICPwQgghhBBCCCGyJXmGnTkZgRdCCCGEEEIIIWyAdOCFEEIIIYQQQggbIFPohRBCCCGEEEJkS3Yyh96MdODFv87Pf9/J6ggp8qbKn9URUiSng21M1MnpaBs5C+fLldURUuT+49isjpAiTjZSPxW2cfLx8cgeWR0hRRaO/TqrI6TIiW1TsjpCiuTOYZ/VEVLEVs7h7z60jfYzX27HrI6QIg72ttHOg5bVAcR/hK3sEUIIIYQQQgghxH+ajMALIYQQQgghhMiWZMTZnKwPIYQQQgghhBDCBkgHXgghhBBCCCGEsAEyhV4IIYQQQgghRLZkKw+wfFFkBF4IIYQQQgghhLAB0oEXQgghhBBCCCFsgEyhF0IIIYQQQgiRLdnJHHozMgIvhBBCCCGEEELYAOnACyGEEEIIIYQQNkCm0AshhBBCCCGEyJZkBr05GYEXQgghhBBCCCFsgHTghRBCCCGEEEIIGyBT6IUQQgghhBBCZEt2MoXejIzACyGEEEIIIYQQNkA68EIIIYQQQgghhA2QKfRCCCGEEEIIIbIlO3kMvRnpwIv/vF9//onNi75Ei4vjnfrN3ZEwPwABAABJREFUqOPX1uz9s0fD2L1yAUrZYWdvj1eHnpSoUJWYJ4/5bmRvYmNjiDMYqFLNnYYtP860nEcPh/HtrCnEGeJo4uVPqw86mr1/6eJfTJ8wgj9+O0+HLr1o0aYDAJER15k2bhi3bt7Azs6Opt7N8WvZLkOz/XQojC+nTyYuzkAzn+a0/bCT2fuapvHF9En8dDCUnE5ODBk5gVfLVwRg9fLv2bpxHUopSpYpy+AR48mZMyejh/bjcvhFAO7fv4ezc14WLFuXrpyHDoQyc9ok4uIMePsG8MHHn1jknDF1IocOhJDTKRcjxkykfAVjTt+mDciTJw92dnbY2zuwePkaAPbu3sH8OV9z8a//sXDpKipUqpyujJA56xNg3aplrF+zAnt7e6rVqE233v3SlfPwwVBmfz6ZOIMBT9/mtP/Icn3OnjaJQwdCcHLKxdDREyhnWp/37t1lyriR/O+PP1BKMWTUOCpXfR2AtSuXsW71cuzt7aleszbdP+2frpwHD4Ty+ZSJxMXF4esXQIeOljk/nzKRA2EhODk5MXrcRMpXqMT169cYNWwwN2/ewE4p/AICad32AwB++/UCk8aP5uHDhxQuXIRxk6bh7OycqlwHwkKYOnkCcYY4/Jq34ONOnS1yTZ00gbDQ/Tg5OTF2wmQqVKz03GVXLFvKyhU/YG/vQK3a7nzWbyCHDh7gi1nTiYmJwdHRkc/6DeDd995Py+pMUkVdHgKq6rFTigPhd9j9202z98u65KZLtaLcfBADwC9/32P7rzcyNENazBnVFo/alYm6dY+3W0zM0iwnfjrA/K8+J85goGEzP5q3/cjs/Svhf/HllNH8+fsF2nXsgW8rY328euki08YMji8Xce0qrT/qincL8+NZemTX/SixA2GhTDPtG77NA5Lcrw6EGnOOmTApfr8aPXwoISH7KFCgIGs3bI5fZvfOHcz55iv++t+fLF2xmkqVq6QrI8CRQ2F8NWMKcXEGmnr708ZKO//VjMn8dDAUJycnBo4YH9/Or1v5A1s3rkPTNJr5NCegdfv45X5cvYwNa1bGt/NdevVNd9an0loHHj9+zCcftScm5gmG2FjqN2xMl+690p0nM9rQgf36cPHiXwDcu3ePvHnzsnrdRmJinjBuzCjOnT2DnVIMGDyMd959L92/w1NHDoXx9cwpxMXF0dTbn9ZWzvGmjh/BH7+e5+OuvQhs2yHDvlvYHunAi/+0OIOBjQtm0XHEdF4u4MpXQ7pQ4e0a6IuViC9TpvKbVPy8BkoproX/yfIZo+k3eykOjjn4ZNRMcubKjSE2ljkjelLujfco/mqlDM9pMBj46vOJTJ49Fxednl4dW/N+rTq8UrJ0fJm8L71E988GczAkyGxZe3t7OvfqR9lyFXn44AE9Pm7Fm+++b7ZserPNmjqe6V/Nw1XnRpcPW1KjVl1KlHr2+T8dDOXK5UssW7eNc2dOMWPKOOYsWkFUZATrVi3j+1UbyenkxKgh/QjavR0PT19GT5wev/zXs6aRJ50ndQaDgc8nj+eLb+ej0+v5qG1LarnXpWTpMvFlDoWFcPlSOGs27uDs6VNMnTiGhUtXPcsxdzH58uc3+9xSpcsyefoXTB4/Ol35EubMjPV54tgRDoQEs3D5j+TIkYPbt24mkyJlOWdMnsDMb+ah0+vp1L4lNd3rUrLUs/V5+EAoly+Hs3LDds6eOcXnk8Yy7/uVAMyeNon33q/J+KmziIl5wqNHjwA4cfQnQvcHsWTl+gzLOWXiOL7+bgF6vZ4P2gRSu05dSiXY7gdM23395h2cOX2SSePHsmTZKhzs7fms/0DKV6jEgwcPaN+qOe9Vq06p0mUYP2YEn/YdwFtvv8vG9etYungB3Xp+mqpck8aPZc68Rejd9LRtGYB73XqUTpArLDSES5cusmnbLk6fOsmEcaP5YcWaZJc9euQw+4L3subHzeTIkYNbN43rL3/+/Mz+6lt0Oj1//P4b3bp0ZHdQaLrWbUIKCHzNjS8PXOJOdAwD65bk9LV7XL/3xKzcHzcfMufQlQz73oywdPNh5qzaz/xxH2RpDoPBwHezpzDm828o6KpnQNd2vFvDnWIlSsWXcX7pZTr1HshPYcFmyxYpXoJZC1bGf07HgCZUq1U3Q7Nlx/3IWs7J48fy7byFpn2jRRL7VTgbt+3k9KmTTBw3hqUrVgPg5etHyzZtGTF0sNnnli5TlumzvmD8mFFpzpY45+xpE5j25VxcdW5069CK6lba+auXw1m6divnz5xi1tTxfLNwOX/9+TtbN67jm0XLcXRwZFCfrlSrUZuixV/h52NHOBgSzPxl6zKk/UycOa11IEeOHMyZv4jcufMQGxNDxw7tqF6zFlVMF23Tmicz2tCp02fFLz992uT4C0rr1hov2q9dv5lbN2/So9snLFu5Fju79N+NbDAY+OLziUz9Yi6uOj3dPzKe45VIdI7Xs+9gDuwPSuaTxH+F3AMPKKUuKqVcUlAun1Kq+4vIZPq+j5VSp5VSp5RSZ5RSPs8pP1op1d/0/+WVUr8opX5WSmVMT+35eVO0HtPx+fG/X0a5/Md5CroVoaC+MA6OjrxWox7njoWZlcmZKzfKNHXnyaNo45mqMQ85c+UGwGCIxWCIhUya4vPruTMULlqcQkWK4ujoiHuDJhwMNT+By1+gIOUqVsbewfy6XEEXV8qWM161z50nD8VfKcmNqMgMy3b+7GmKFC1O4SLFcHR0pF4jD8ISXUQICwmmcVNvlFJUqvIa9+/d4+aNKMC47h4/fkxsbCyPH0Xj4uJqtqymaQTv2UGDRk3TlfPcmdMULVacIkWL4eiYg4aNPQjZZ54zZH8QTT19UEpRuaox542oqGQ/t2Sp0rxSomS6siWUWetz47pVtPmwIzly5ACM9SW9OYsWKxa/Phs0akrYPvM6Gbo/iCbNjDkrV3mN+/eN6/PB/fuc/Pk4nr7NAXB0zEHevC8BsH7tKtp16JRhOc+eOUWxYsUpasrZqElT9ifa7vuDg2jqZdzuVaq+zr17d7kRFYmLq47yFYwX5PLkyUOJUqWJjIwAIPziX7z51jsAvPd+dYL27k5VrjOnT1Gs+CsULWbM1dijGfuC9pqV2Re8F09vX5RSVH3NmCsqKjLZZVevWsFHHTvHr78CBY3rr3yFiuh0esDYGXny+AlPnph3rtOjRIFcRD14ws2HMRg0OH7lLlUL5c2wz89MB078ya1/HmZ1DH6/cIZCRYriVtjYztes15ifDuwzK5MvfwHKlq+EvX3S4y+nThzBrUhRdG6FMyxbdt2PEjPuG8UT7BtNLfar/cF78fT2sdivAN56+x1efvlli88tVbo0JUqWsng9rS6cS9TON/TgYIh5+3kwJJiGHsb2s2KCdj784v+oWLkqTk65sHdw4LU33iZsv/F33PTjKlp/kHHtfELpqQNKKXLnzgNAbGwssbExKNJ3vpRZbehTmqaxa8d2mjT1BOB/f/7Be+9VA4ztat68eTl79ky6foenLpw7Y6oPxn2/bsMmFvUhf4GClK9YGQeH/+bYq1LZ819W+c934JVS9qkong94IR14pVRRYBhQU9O0qkA14FQqPsIX2Khp2huapv2ZCRH/Fe7eusHLBXXxP79cwJW7Ny2ndJ75KYTpn7Zn8aTBBHQbFP96nMHA7P4dGd/Rl7JV36Z42YqZkvNGVASuen38z66uem6moRN+/dpV/vj9AuUrpX/637Nskej0bs+y6fQWFwhuREZYlImKjMBVp6dVuw4EejfAv2ld8jjn5Z1qNcyWPfXzcQoUKEjR4q+kK2dUogw6vVv8SduzMpHo3BKWMeYE4wWb3t078WGbADasW52uLMnJrPV55dJFTv1ynK4ftaZ3lw6cP3c6XTmN67PQswx6PVFREYlymv8uOp2eG1ER/H31Mvny52fi6GF81KY5k8eOJDra2IG6fOkip34+zicftKLnJx9y/mz6ckZGRqJ3M88QGWGeMyoyArcEOfV6NyIjzdf531ev8uuF81Su8hpg7AQ/PXnd83/27juuqvqP4/jrC2iYWqbAxdyrLLWs7Fe5xYmDIYg7V0MttTT33pqm2dZKs9KcZeYeOHDP3FZWzmS4Fypcvr8/zgHuvQxBQLj1efq4D+He7znnfb9nn+/3HNasJiL8fDpzReDtbTtNS8JJTUKZCMcy3kRGRKQ67KmTJ9m3dw/tWregS8d2HD6UdLexbu1qyj/1VMJBfmYo4O7G5ejYhN+vRMdQwD3pgWapgnkY6FOK7q8Uo3D+zJv+v8GlqCg8PBPnayFPLy7dx3Z+S+hqavg0zMxoOXY9SpozAot34nbJYvFO2IYnlImIwNuhjON3yWqO20YPr2S2nynsC0qVLsfB/Xu5evUKt29Hs3NbGJER4QCcPX2KQ7/uo3vnNrzTtSPHj2bOCSZkfBmwWq20CQmkfp3qvPRyVSo+82wG82TNNjTevr17KFSoECVKlATgiSfLs2HDemJjYzl39gxHjx7J8PIa70KUsf+Ol9x+XwhbTn0Cr5Tqp5Tqaf48VSkVav5cVyn1vVKqtdmCfVgpNdFmuBtKqVFKqZ3AKzbv51FKrVJKvZ5kYoYJQBmzZXuSUuo721ZxpdQcpZSfUqqjUupnc1y/KaWG25Rpp5TaZY5jeioXELyA68ANAK31Da313+Y4ypjj3quUClNKlXeol8bAO8BrSqkNDuNFKdVAKbVdKbVPKbVQKZXPfP+kUmqc+dkepdTzSqnVSqk/lVJdzTK1lVKblVI/KaWOKqW+UEolWY6UUr3Nej+slHrHfG+0UqqXTZmxNvOvr1Jqt9nbYKRNmcFmHa4Dnkyhru6bRid9M5krahVfqkmfad/Rvt9Y1s6fmfC+i6srvSZ/zcDpCzlz4hjhp//K7IgpUum89Bd96xajBvWmW69+5M2bse7otrROpg4dKjG5elZKcf3aVbZs2sC8Jav5cUUot6OjWbPyF7ty69asoG7DjLW+p5QhSZlkvkt8Pc+YNYdvf1jM1E+ms2j+D+zfuyfDmdKaITPq02q1cv3aNT6fOZduPfswYuB7KUwrrTmTSemwTCY7fqWwWq38fvwYAcGtmDV3Me558vD9rK/scs6Y/QPde/Vh2IA+GcqZXNAkOVOoz3i3bt2kX5+e9Ok7IKE75bCRY1k4by7tWgVx69ZNcuXKlc5YaciVQpnUho2vv+/mLuCdPv3o9947duVPnPiDaVMmM2TYqHTlvR+OKc9cuc2wVScYH/o3m/66xBsvF8vyDM4k+f1R+rbzMTEx7Nq6mWq162dSKlMOXY+S5kzmvSTrVXJFHmxT2r3qClJY/4ESpUrT6tXO9O3xBv17daVMuSdxdTUOJa1WK9evX+PTr+fwZo8+jBqUse28Q6B7Z07le7m6ujJ3wU+sWLOBI4cPceKP3zMYJ2u2ofFWrViW0PoOEBAYhMXiTZuWQUyaOI5nKz+XUO8ZluyhqDy0TaTMqU/ggc1ADfPnKkA+pVQuoDrwBzAR8AEqAy8qpQLMsnmBw1rrl7TW8f2l8wG/AHO11l+mML0BwJ9a68pa677AV0AnAKXUo0BVYIVZ9n9AW3PaLZRSVZRSTwEtgWpa68qA1SyTnANABPC3UmqWUqqZzWczgB5a6xeA94DPbAfUWq8AvgCmaq3tboIzu7gPAepprZ8H9gC2Tzg5o7V+BQgDvgGCMVr/bY/2/gf0ASoBZYDmDtN4wayXl8xhX1dKPQd8DXQwy7gArYA5SqkGQDlzvJWBF5RSNc3xtAKeM6fxYgp1hVLqDfOiw541i75LqVgSjxb05OrFxKucVy9F8UjBlO8CKP30s1wMP8fNa1fs3s+TNz+lKzzH77/uSvO008PD00KUzZXuqKgICjp0NU9NbGwMowb1xqdBE6rXrpep2Ty9LAlX/8G4Au/h6elQxjuZMl7s2bWDwo8XocBjBXFzy0WNOnU5fPBXm9yxhG1cR516jTKc08shQ2REOJ6eXvZlLBYiw23LGDmN72D8X7BgIWr51OXokfR0iEm7rKpPTy8LNevUQynFUxUq4eKiuHrl8n3n9LJYiIxIbH2IiojAw8O+Pj0t9t8lMtIo4+llwdPLQoVKzwBQp14Dfj9+LDGnj5Hz6YrPoJQLVzKYMyLcPkP8vEwo4+VNuE3OiIhwPM06j42JoV/vXjRq3Ayfeg0SypQsVZpPp3/N9/MW07BRY4oULZ6uXBaLN+HhttOMSLI8Wrwdy4Tj6eWV6rAWiwWfevWNLqyVnsFFuXD5slF/EeHh9O71NqPHTaRY8fTlvZcrt2N5LE9ii3uBPLm4ejvWrszt2DjuWI0j1SMRN3FVkDd3Jh0A/wsU8vTiQlTifL0YFZmu7TwYD8Er/UR5CmRi12nIuetR8jkTt0sRyWznLd4Wwh3LOHyXrOa4nb8Qmcz2M5l9QSHzuzT2a86Mbxcwbfps8j/yKEWLlUgYpkbtxO28yuB23lZGl4F4+R95hBde/B/bt9nfrpheWbUNBePYY/26tTRslNh44ObmRt/+g1iw+Gc+/Phzrl+7TnGzdT6jPLwsdj1FjHmdvnX/385F5cxXttVH9k06U+zFONnLD9wBtmOcyNcArgAbtdZRWutYYA5Q0xzOCjg+zvpnYJbW+tu0TlxrvQkoq5TyAloDi81pAazVWl/UWkcDP2JcVKgLvADsVkr9av6e7E1VWmsr0AjjBPp3YKoy7gHPh3GhYKE5julA4eTGkYKXgaeBrebwHQDbvslLzf8PATu11te11lHAbaVUAfOzXVrrv8yMP5jfzVZ14Cet9U2t9Q3z+9fQWp8ELpon8w2A/Vrri+bPDYD9wD6gPMYJfQ1zPLe01tdssiWhtZ6hta6ita7SILh9SsWSKFq2PBfPn+VSxHliY2I4sDWUp6vYd+G+cP5swtXac3/9jjU2lofzP8qNq1eIvnkdgJg7dzhxcA+eRTL3oDjek09V4NzZU5z/5ywxMTFsWreKV6rXTtOwWmumjBtO8ZKlCG6d+Q9pKv90Rc6eOc35c0a20DUrqebw8KRqNWqzesVStNYcOXSAvPnyUcjDE4t3YY4ePsjt29Fordm3eyclbB7YtHf3DoqXKG3XjfB+PVWhImdOn+Kfc2eJibnL2tUrqVHbPmeNWj6sWPYzWmsOHzxAvnz58fD0JDr6Fjdv3gQgOvoWu7Zvo3SZchnOlJysqs/qtXzYt8e4wHTm1EliYmJ4tMBjSaafnpxnzpxOqM91a1ZQrZZ9zuo167BquZHz8KED5MuXDw9PTwp5eOJl8ea0+aTfPbt2JDy8qWbtuuzbvROA06dOEhsbQ4EM5Hy6QiXOnD7FubNGzjWrVlDTIWet2nVY8Ysx3w8d/NWc715orRk1YgilSpem3asd7YaJfzhcXFwcX3/5BUEtWqYrV4WKlTh9+iTnzp4hJuYuq1cup1YdH4dcPixbugStNQcPGLk8Pb1SHbaOTz1279oBGPcXx8TE8Nhjj3Ht2jV6dH+Dnu/05rnnX0hX1rQ4dTkar3y5KfRwLlwVvFD0EQ6dv25X5pGHEk/WSzzmjlKKm3etmZ7FWZV7sgLnz54h4vw5YmJi2BK6mv9VrZWucYStX0XNupnbfR5y7nrkyFg3EnOuXrmC2smuVz8nWa8epPJPVeTcmcR9eujalbxSs7Zdmao16rB2pbH9PGqznQcSHk4XEX6esI3r8GngC0C1Wj7s32NsP8+cPklsBrfztjKyDFy+dInr164BcPv2bXbt2E7JDD47Jqu2oQA7d2yjVOnSdrcMREdHE33LuNVr+7atuLm52j0wLyPKP1XBbnnYsHYVVWvUzpRxi38np34SgtY6Ril1EqO1dxvGPeJ1MFqFT2OcLCfntnnyaWsr4KuUmqvT19/oO4xW9FaA7d8QcxyHxuj9NFtrPTAtIzZz7AJ2KaXWArOAKcAVswX/nswu+nvNX5cCuzEuLrROYZA75v9xNj/H/x6/vCT33ewmm0qkr4COgDcQ3xddAeO11tMdsr+TzLgzlaurG35d3mHm2PeIi4ujSp3GWIqVYseanwF4uYE/h3duZt+m1bi6upErd27avDvc6K585SILPhmHjotDa02lV2rz1AtVsyanmxtv9x7EoHe7EWe10rBpACVLl2XZT8a92E0DQ7h08QJvd27FrZs3US4u/DT/e76cu4S/T/zOulXLKFWmHF07tACg85s9+V/VGqlNMs3c3Nx4p+8g3uv5pvHncJoFUqpMWX5ebDy93T+oJS9Xq8mObWG0ae7LQ+55GDB0NABPV3yGWnXr83r7EFxdXSn7ZHmaBbZIGHfompXUNQ9MMiPne/0H06v768TFxdHUP5DSZcrx40Ljyc3NW7SiavWabNuymWC/Rri7uzNkxFjAOMjs37snYDwkroFvE16pZtTfxtB1fDBxLFcuX6J3z2488WR5pn2WUieetOXMivps7NeciaOH0LFVAG65cjFo+LgMdRt1c3Ojd7/B9H77DeKscTTxD6R0mbIsWWTkDAhuySvVa7J962Za+vvi7u7OoBFjEoZ/t98gRg7pT2xMDI8XKcpA87Mm/oGMHzmU9iH+5HLLxeARYzOcs+/AIfTo9hrWuDj8AppTpmw5Fi0w5ntwSCuq1ajF1i2bCWjaEHd3d4aPMv6U2IH9+1ixbCllyz1Bm5BAALr3eIfqNWqxetVyFs6bC0CduvXxC2iefIBUcg0YNIxub75GnNWKf2AQZcuWY+H8HwBo0bI1NWrWYkvYJpr51sc9Tx5Gjh6X6rAAAc2DGD5kEEEBTcmVKxejx01AKcX8H77n9JnTzPjiM2Z8YXTa+mLGzISH3GVUnIYFB8J5q1oxXFBsP3WF89fvUr1kAQC2nLzCc0UeoUapx7BqTYxVM3P3uUyZdkbNHt+RGi+Uw6NAPk6sGs3oL1Ywe8n2B57D1c2N13v1Z2Tft7DGxVHP14/ipcqw6udFADTyD+byxQu892Y7bt26iVKKXxbN5ePZi3g4bz7u3I7mwN6ddOszONOz5dT1KLmc/QcNpfubXYizxuEfGESZsuVYON/I2aJlK6rXrMWWsM34+TbAPY/xp87iDejbm727d3PlymUa1q1F1+49CAwKJnTdWiaOH8PlS5fo2b0rT5Yvz2czvr7vnK5ubvR4bxD9e3bFGmfFt1kgpUqXZemPxj7dr3kIL1Wrwc5tm2kX1Djhz8jFGzGgN9euXsHVzY1efQeT/xHjwXu+zQKZNGYonVsH4pYrF/2HZ2z7aSsjy8CFC1EMHzKQuDgrcXFx1G/QiBoOJ//3kycrtqEAq1auoJFvE7vpXbp0ke5vdsFFueBlsTBm/PsZym8rYXno1Y24OCu+5jHeL+by0Ky5cYzXrWPiMd7ied8zc96STL0lUjgPlWn3xmQTpdQIjBPnzhitxrsxTli7AzswTuIvA6uBj7XWPyulbmit89mM4yRGy/1QILfWulsK0yoE7NNal7B5z4Jxkh2utX7JfK8jMA6oCEQDO818tzBa+qtprSOVUgWB/FrrU8lM63HAW2u9z/z9NSBAa91UKbUNo3v8QmVsmZ/RWh8w6+KG1nqy7c8O4/U068dHa31CKfUwUFRr/Xt8PWitL5jfoYrW+m2HOqoIrMRoxT9l/jxDa73YpkxxjO73L2OcnO8E2mut9yulcpvzKRdQTmttNbvQjwbqaq1vKKWKADFAUXM8L2FcPNgHTHf8To5+OhjuFAv180Uy56p4VnvIzTk66jyUyzly3omJy+4IaeLq6hz337k7yfLp5uocOd9bdiy7I6TJzFGfZneENNm3YuK9C+UARR/Lk90R0sRZtkuXb8Zkd4Q0eSSZB03mRM6y/bx44869C+UQRR97yDlWJtO49X/myGP7QXXLZEs9OscakbowjC7k27XWEcBtIExrfR4YCGzAuJ98n9b653uM6x3AXSmV7GU1s7v3VvPBbJPM9yKAYxit47a2YLTO/4rRtX6P1vooxv3na5RSB4G1pNz9PRcwWSl13Ozq3hKIfwBcW6CLUuoAcARI9c/LOXyHKIwW8B/MDDswuqynx3aMB/odBv4GfnKYxj6ME+9dGCfvX2mt95uf3cWYJwvie0FordcAc4HtSqlDwCKMCxv7gPmYdYgxr4UQQgghhBDiP8npW+Czm9mCfQh4Xmt91XyvIzat1/8mSqnawHta66b3KJrS8C4YLekttNZ/ZGK0BNICn7mkBT5zSQt85pIW+MwlLfCZS1rgM5ezbJekBT5zOcv2U1rgs460wNtzjjUih1JK1QOOY3TNv5rdeXI6pdTTwAlgfVadvAshhBBCCCH+PbL7afM57Sn0znHp7QEz73Vfn8xHdc1u9ABorddh3O9tR2v9DUYX8rRObyfwkMPb7bXWh9I6jgdFa70R2Hifwx4lhafuCyGEEEIIIYRInZzAJ8M8Sa/8AKf30oOalhBCCCGEEEII5yQn8EIIIYQQQgghcqTs7K6eE8k98EIIIYQQQgghhBOQE3ghhBBCCCGEEMIJSBd6IYQQQgghhBA5klLSh96WtMALIYQQQgghhBBOQE7ghRBCCCGEEEIIJyBd6IUQQgghhBBC5EjyFHp70gIvhBBCCCGEEEI4ATmBF0IIIYQQQgghnIB0oRdCCCGEEEIIkSPJQ+jtSQu8EEIIIYQQQgjhBOQEXgghhBBCCCGEcALShV4IIYQQQgghRI7kIn3o7cgJvPjXebZwgeyOkCae+XNnd4Q0UbLRzFS3Y6zZHSFN8rvnyu4IaRIXp7M7wr+Kd37nmO/7VkzM7ghp8nzj/tkdIU0id3yU3RHSxklW90fyOMfhdazVOSpUE5fdEdJEjpfEgyJd6IUQQgghhBBCCCfgHJcIhRBCCCGEEEL857hI5wY70gIvhBBCCCGEEEI4ATmBF0IIIYQQQgghnIB0oRdCCCGEEEIIkSPJ8wHtSQu8EEIIIYQQQgiRyZRSjZRSvymlTiilBqRS7kWllFUpFXyvccoJvBBCCCGEEEIIkYmUUq7Ap4Av8DTQWin1dArlJgKr0zJe6UIvhBBCCCGEECJHcsFp+9D/Dzihtf4LQCk1D/AHjjqU6wEsBl5My0ilBV4IIYQQQgghhMhcRYAzNr+fNd9LoJQqAgQCX6R1pHICL4QQQgghhBBCpINS6g2l1B6b1xuORZIZTDv8/iHQX2ttTet0pQu9EEIIIYQQQogcKac+hV5rPQOYkUqRs0Axm9+LAv84lKkCzFPGl/QAGiulYrXWS1IaqZzACyGEEEIIIYQQmWs3UE4pVQo4B7QC2tgW0FqXiv9ZKfUNsCy1k3eQE3ghhBBCCCGEECJTaa1jlVJvYzxd3hWYqbU+opTqan6e5vvebckJvBBCCCGEEEKIHMklh3ahTwut9QpghcN7yZ64a607pmWc8hA7IYQQQgghhBDCCcgJvBBCCCGEEEII4QSkC70QQgghhBBCiBzJJac+hj6bSAu8+M/bs2Mrr7X2o3PLpiz47uskn5859TfvvtmeZnWqsGju7IT37965Q6/X29C9QwvebBfId19/lunZtm4JI6BZI/waN2DmV0n/SoXWmonjx+DXuAEhzf04dvRIwmcjhg7Cp1ZVggObJRnuhznfEdCsEUEBTflwyqT7yLUZ/6YNaeZbP+Vc48bQzLc+LQKb2eVKadjPP/2Y+j41CAnyJyTIn7DNmwC4cuUyr3VqzysvPsf4saOyPWe/Pu8kZPRt4ENIkD8AMTExDBnUn+DAZgQ28+XrL6enK6utndu30D64GW2aN2bO7K+Szf3R5PG0ad6Yzm2a8/vxowmfLZz7LR1bBtCxVSCjhvTjzp07AJz4/Te6d25Lp9aBDOz9Njdv3Eh3rq1hm/Fr0pCmjerz9ZfJ1+eEcWNo2qg+wY71mcKwn3z0IcGBzQhp7s+br3cmMjICMOtzYH+CApoRkMH6lPUo0fnz//DKi88xe1bSbd39+OfIHn4e+QZLhr/G4TULUix34dTvzHm7Gaf2bbF7Py7OyvLxPdjw+YhMyZOSfTu30r19IF3b+LF4zqwkn5899Tf9u3cguP5LLJn3bcL7506f5J0urRJerRvXYOnCOVmaNSVfDG/LqfXj2bNwULZMf9uWMJo38yWgSUO++frLJJ9rrZk0YSwBTRrSKsif4+byGh5+nje7dCDYvwkhgU354fvE+p3+2Sf41qtFmxaBtGkRyJawTUnGm145dX13lBX1+dvxY3Rs25I2LQJp3yqYw4cOZjjn9q1hhAQ0JtivId/OTD7nBxPHEuzXkLYhARw/dtTuc6vVyqutmtOnZ7eE97784hOaNahN+5aBtG8ZyLZMmO/btoYR5OdLYNOU63PyhLEENm1I62B/jh8z6vPOnTt0aBNCmxYBhAQ2ZfpnHycZ9rvZM3nx2ae4cvlyhnPu2r6FV1s0o11QY+amsH//+IPxtAtqzGtt7ffvi+d9T+fWgXRqFcCiH75LeH/U4Pd4vV0wr7cLpnVAQ15vF5zhnCJnkxZ48Z9mtVr5dMo4xk2djoeXhV6vteGl6rUpUapMQpn8jzxC13f6s33zBrthc+XOzYRpX5Hn4YeJjY3hvW4dqfJSdZ6q+EymZZswdhSfz5iJxdtC21YtqFXHhzJlyiaU2RK2mdOnTvHz8tUcOniAcWNG8t1c4yC6mX8gLVu3ZejgAXbj3b1rBxs3hLJg8VJy587NpYsX051r/JhRfPHlLCNXy+Dkc50+ydIVazh08ABjR4/g+x8W3nPYdu070qFTF7vpPZT7Id7q0YsTf/zBiRN/ZHvO9z/4MGH4DyZNIF++fACsXbOKmLt3WfTTL0RHR9PcvwmNGjehSJGi6a7fae+PZfInM/D08qZrh1ZUq1GHkqUTl8md28I4e+YUcxYv5+jhg0ydOIbPZ80lKjKCxfPnMnv+Eh5yd2fEwD6Erl2Jb9MAJo0dTrdefaj8/IusWPoT876fRZeuPdKVa9zYUUz/chYWi4U2LYOpXceHMmUdl8eT/LLSqM8xo0YwZ97CVIft2Pk13u75DgBzvv+W6Z9/ytDho1i7ehV3Y+6yeIlZn373X5+yHiWaPHE81WrUSFfWlMTFWdm14HPq9hjDwwU8WPn+uxSt9DIFChdPUm7/klkUfur5JOM4vmEpj3oXI+b2rUzJlByr1cr0aRMZOfkzCnla6Nu1Hf+rVotiJUsnlMn3yKO81rMfO7fYb+eLFC/Jh1/PSxhPl+BGvFyjTpZlTc13v+zgi/mb+Gr0qw982larlYnjRvPpjK+xWCy82jqEmrXrUNpmed26ZTNnTp3ip2WrOHzwAOPHjGL23Pm4ubrybp9+lH+6Ajdv3qR9qyBeeqVqwrBt2nWgfcfOmZYzJ67vyeXMivr8aOpkXu/6FtVq1GRL2CY+mjqZGTO/TSXJvXNOnjCGjz7/Ci+LhU5tW1KjVh1K2eTcvmUzZ06fYuHPqzhy6CDvjxvJzO/mJ3w+f+53lCxVhps37S8at2r3Km1fzbz5/v640Xwy3ajPDm2S1ue2LZs5ffoUP/6yisOHDjBhzCi+mTOf3Llz8/lXs3j44bzExsTwWsd2VK1eg0rPVAaMCya7tm/Du3DhTMk5bdJYJn1s7N+7dWxF1WT27+fOnOK7Rcs5dvggH74/hs9mzuXvP/9g+c+L+WzWXHK55aL/O115uVpNihYvwbCxkxOG/3zaJPLmzZfhrCJnkxb4bKCUOqmU8khDuQJKqe4PIpM5vXxKqc+VUn8qpfYrpfYqpV6/z3FVVko1vkcZP6XUgHuU+UYplWWXEn8/dpjHixajcJGi5MqVi1r1GrFjy0a7MgUeK8STT1XEzc3+epdSijwPPwxAbGwssdZYMrOHz+FDBylWvDhFixUjV67cNPRtzMYN6+3KbNqwnqZ+/iileObZyly/fo2oqEgAXqjyIo8++miS8S6cP49OXV4nd+7cABQsVOg+cpWwydWEjaH2uTZuWE9Tv4AkudIyrKM8Dz/Mc89XIfdDD+WonFpr1qxaSaPGTQFjeYiOjiY2NpY7d26TK1euhJP79Dh+5BBFihbn8SLFyJUrFz4NfNnqcPFo6+YNNGzsh1KKCpWe5cb161y8EAWA1RrLnTt3iI2N5fbt23h4eAFw5vRJnn2uCgBVXnqFzRvWpSvX4UMHKVbMrJPcuWnUuEmS5XFD6HqapVSfKQxrW0e3o6NR5kqklCL6VmJ9uuXKRb77OCiR9ShR6Pp1FClalDJlyqUra0ounvyd/J6Pk9+jMK5uuSj5Qk3OHtyRpNxvG3+heOVquOe3r8ebly/wz+HdlK3aMFPypOSP44cpXKQo3o8b2/nqPg3ZuXWjXZkCjxWkXPkKuLqm3K5xcN8uvIsUxcv78SzNm5Kt+/7k0tWsu9CRmiOHzfWoqLHMNWjUmE0bQu3KbNoQSuNmxnpUyVxeL0RF4uHpRfmnKwCQN29eSpYqk9DTJrPl1PXdUVbVp1Iq4UT5xvUbeHp6ZSjn0cOHKFqsOEXMnPUb+rJ5o33OzZtCadzUyFnxGWN/dCHK2B9FRoSzbcsm/AKDMpTjXo4cPkixYon1Wb9RYzZtTFqfTeLr85nE+lRK8fDDeQHzWC42BkXiwdzUSRPo8e57CfumjDh+1GH/Xt+XbQ77922bN1Df19i/P22zfz918i+ervgM7u55cHVz49nnqrBlU9Ljko3rVuPTINXDb6ekVM58ZRc5gX/AlFKu6SheAHhgJ/DAV8BloJzW+jmgEVDwPsdVGUh1C6K1Xqq1nnCf488UF6Ii8fTyTvjdw9OLi1FpP7CwWq281TGE1s3q8FyVlylfIXNa3wEiIyOweCde8bVYvImKiEhSxtuhzL0OjE6dOsn+fXto3yaELh3bceTwoXTn8vZOrDOLxZJkmpERjmW8iYyIuOew836YQ4vAZgwfMpBrV6+mK9eDzAmwb+8eChUqRIkSJQGoV78hefLkoX6d6jSqX4dXO3bm0UcLpDt3VFQknpbEaXt6WYhyWCajIpMpExmJp5eFlu06EuJXn6DGPuTLl48XX64KQKnSZRMuBGxct5rIiPB05YqMiMC7cOI0vSwWIpJZHi3J1ec9hv142lQa1K3F8mW/0P3tXgDUa9CQPA/noV7t6jSsV4cOHTvzaIEC6cqcmEnWo+hbt/hm5pd07f52unKm5taVizz8WOK16IcLeHDrykWHMhc4c2A75Wr4Jhl+76IZPBfYKcuPgi5FReHhmVhfhTy9uGSesKXHltDV1PDJ2osNOVVkRCQWi/067Li8RiVZLr2JjLSv53/OneO348eoWOnZhPcWzJtDqyB/Rg4bzLVrGd/u58T1PUnOLKrPPv0GMm3KZJrUr8O0Ke/zdq93M5QzKjICL7uc3gkXOxLLROLlbf9doszvMnXSBN7u9R7KJenpxsJ5c2kbEsCYERmf71GRkfb7Hi9LkvkeFRnhUOeJ9Wm1WmkTEkiDOtV56eWqVHzGqM9NG0Px9LLwxJPlM5Qv3oXISLv69Ehm/34hyr6Mp5eFC1GRlCpdjoP793L16hVu345m57awJPvxg7/u5bGChShavESm5BU5l5zAp4NSqp9Sqqf581SlVKj5c12l1PdKqdZKqUNKqcNKqYk2w91QSo1SSu0EXrF5P49SalUqrdwTgDJKqV+VUpOUUt8ppfxthp9jtmJ3VEr9bI7rN6XUcJsy7ZRSu8xxTE/pAoJSqgzwP2CI1joOQGsdpbWeaH6uzAyHze/Y0nx/vm1Lu9li3hIYBbQ0p9syhWl2VEp9Yv5cQim1Xil10Pzftg9mPaVUmFLqd6VU0xTq6v5onVywNA/u6urKp98s4Lsf1/D7scOc/CvtXbzvnS2Z9xyyJRuf1PNbrVauXbvGt3Pm826ffvR77x10ciNKKVYyZR2vTKdUJrVhQ1q2ZtnKtcxf/DMenl58MClj13ayKme8VSuWJbS+g9Hy4+LqwprQMFasWs93s2dy9syZ+wmedNpJ5mly+eD6tats3bSBeUtWsXjFeqKjo1mz8hcA+g0dxZJF83jj1RBu3bpFLrdc6YuV7DQdcqVUn/cYtkevd1mzfhNNmjZj3tzvAaM+XV1cWLshjBWr1/PtfddnMu/9B9ejzz/9mLbtOyS0NGWO5JdDW3sWzeC5gE64uNjves4e2oV7/kcpVDxzegOkJrnlL70XDWJiYti1dTPVatfPpFTO5v6X13i3bt2kX++e9Ok3IKHnTXDLVixZvoa5C3/Cw8OTqZPfz+yYOWJ9T0vQzKjPRQvm0bvvAJav3UDvvgMYPXxIBjKmsO44lkkh55bNG3msYMGE3gK2mrdoxeJfVvPdvB8p5OHJR1MyNt/TtD1Npc5dXV2Zu+Anlq/ZwJHDhzjxx+/cjo5m1pfT6do97bea3TPn/c53oESp0rR6tTN9e7xB/15dKVPuSVxd7beroWtW/itb30VScgKfPpuB+JsHqwD5lFK5gOrAH8BEwAej9flFpVSAWTYvcFhr/ZLWOv4JPvmAX4C5WuukT9swDAD+1FpX1lr3xWgh7wSglHoUqAqsMMv+D2hrTruFUqqKUuopoCVQTWtdGbCaZZJTATgQf/KejObmuJ8F6gGTlFKFgXnmNFBK5QbqAkuBYcB8M/v8ZMdo7xPgW631M8Ac4CObz0oCtYAmwBdKKXfHgZVSbyil9iil9vzwbdofzuThZSEqMvEK5oWoSAp5pL/LWb78j/DMcy+yZ8e2dA+bEi+LhYjw8wm/R0SE4+lln81isRB+jzKOLBYLdevVN7q7VXoGF+XC5XQ8mMVi8SY8PLHOIiIiknTTs3g7ljFypTZsIQ8PXF1dcXFxoXlwCw5nsIUjq3KC0c1u/bq1NGyUuKNcuWIZ1arVIFeuXBQsVIjKlZ/nyJH0fwdPLwtRNlfVoyIj8HDInVKZvbt2UPjxIhR4rCBubrmoWaceRw4eAKBEydJM/ngGM75dQN0GvjxetFi6clks3oSfT5xmZEQEXg7LmpfFm4iU6vMewwL4NmnKurVrAFi5fBlVqxv1WahQISo/d3/1KeuRUWeHDh3gwymT8W3gw5zvZ/P1l9MTLpbcr4cLeHDr8oWE329duUCeR+27Fl88fYItMyfy09BOnN6/lV3zP+PMge1E/XWUs4d28tPQTmyZOZHw3w6y5ZuMPxgsOYU8vbgQlVhfF6MiKejhma5x7Nu5ldJPlKdAwYx1nXZWRq8Z+3XYcXn1SrJchuPpadRzbEwM/Xr3olGTZvjUa5BQplChxOU1MKgFRzL40LWcur4nmzML6nPZ0iX41DMuMtVr0CjDPQW8vLztWnkjI8KTyWkhMtz+u3h4enHw132EbdpAQON6DB3Qhz27dzJ8cD/Afr77N2/B0YzmtFjs9z2REXg47p+8vB3qPLE+4+V/5BFeePF/bN+2hbNnz/DPubO0CQnAz7cukRERtGsVxAXzdrX74ellsavPC5ERCbe5pVQmKjKCQmadN/ZrzoxvFzBt+mzyP/IoRYsltrRbY2PZsmEdder9O3sJuSiVI1/ZVh/ZNmXntBd4QSmVH7gDbMc4ka8BXAE2mq3WsRgnoTXN4azAYodx/QzM0lqn+ekiWutNQFmllBfQGlhsTgtgrdb6otY6GvgR46JCXeAFYLdS6lfz99JJx5yUUmqw2Xr+j/lWdeAHrbVVax0BbAJeBFYCPkqphwBfYLOZIb1eAeaaP39nTi/eAq11nNb6D+AvIElfJq31DK11Fa11ldavJv/wpuQ8Ub4C/5w5Tfg/Z4mJiWHTulW8XK1Wmoa9cvkSN65fA+DOndvs37ODYmZ36sxQoWIlTp86xbmzZ4mJucvqlSuoXdvHrkytOj4sW/ozWmsOHviVfPny3/Oet9o+9di1cycAp07+TUxMDI899lj6cp0+ybmzZ8xcy6lVxyFXbR+WLV2SJFdqw9p2ywtdv46yZTPWMpdVOQF27thGqdKl7brsFS5cmF27dqK1JvrWLQ4dPECpUmla3ew8+XRFzp45xflzxjIZumYlVWvUtitTtUYdVq9YitaaI4cOkDdfPgp5eOLlXZijhw9y+3Y0Wmv27d5JiZKlALh8yejaHBcXx3czZ+DXPCRdueLr5OzZM8TcvcuqFUnrs3YdH35JpT6TG/bUqZMJw2/cEJpQZ96FC7Nrp1Gft27d4tCB+6tPWY+M9WjWt3NZuSaUlWtCaduuA11ef5NWbdqlOW9yCpV4guuR57hxIRxrbAwn926maKWX7MoEjppJ4OhZBI6eRfHnqvG/lt0p9uwrPOffkeZjvyVw9Cyqd+6P95PPUL1j3wzlSUm5Jytw/uwZIs6fIyYmhi2hq/lf1bRt5+OFrV9Fzbr/zgPjtHi6QiXO2KxHa1atoGZt+4f51apdhxW/GOvRoQO/ki9/fjw8vdBaM2r4EEqVKk27VzvaDXPBZnndELqWMuUyYbufA9d3R1lVn56eXuzdsxuA3Tt3UCyDXamfqlCRM6dP8c85I+fa1Sup4ZCzRi0fViwzch4+eIB8+fLj4elJ9569+WX1BpasWMfoCR9Q5cWXGDnWaGmPv0ceYFPoOkpn8LkcT1eoxOnTifW5dtUKatayz1mzdh2Wx9fnwV/NnF5cvnSJ69eMY7nbt2+za8d2SpYsRdlyT7Bm41aWrlzP0pXr8bJY+H7eYjzSefHPVvmnKnLuzCnOm8ecoWtX8krN2nZlqtaow9qVxv79qM3+HRL34xHh5wnbuA6fBom3Ju3dvYNiJUvZ3V4n/r3kKfTpoLWOUUqdxGgF3wYcBOoAZYDTGCfLybmttbY6vLcV8FVKzdXp64f1HUYreivA9vGdjuPQGL1uZmutB6ZhvEeBZ5VSLubJ8lhgrFIq/rGhyV5m0lrfVkptBBpitMT/kOZvkjqdws/J/X7fXN3c6NZ7IEN6d8MaF0eDJgGUKF2W5UuMJ9I2CQjh0sUL9HytNbdu3sTFxYUlC79n+vc/cfniBSaPHUJcXBw6Lo4aPg14KY0n/2nh5uZG/0FD6d61C3HWOPwDgyhTthwLFxhPRG4R0orqNWqxZfNm/Bo3wN3dnRFjxiUMP6Bfb/bu3s2VK5dpWLcWXd/qQWDzYAICmzNi6GCCA5uRK1cuRo2dkK6Hs7i5uTFg0DC6vfkacVYr/oFBlC1bjoXzjVnfomVratSsxZawTTTzrY97njyMHD0u1WEBPvxgEr/9dhwFPF6kCEOGJ/7JON8GPty8cYOYmBg2hK7j8xkz7Z4q/CBzAqxauYJGvk3spteydVuGDRlIUEBT0Bq/gOb3dd+cm5sbvfoOom/PrsTFWfFtFkipMmX5ebGxTPoHhfBytRrs3LaZts0b85C7O/2HjgHg6YrPUKtufV5vH4KrqxvlnixP08AWAKxfs5IlC41lp0aduvg2C0h3roGDh9HtjdeIi7MSYNbJArM+Q+Lrc/MmmvrWx909D6PGjEt1WIBpUz7g5Mm/cXFRFC5chCHDRwLQyqzP5v5GffoH3n99ynqUNVxcXXkxpBvrPx2KjoujzCv1KfB4CX4PMzqGPVEjZ3TldHVz4/Ve/RnZ9y2scXHU8/WjeKkyrPp5EQCN/IO5fPEC773Zjlu3bqKU4pdFc/l49iIezpuPO7ejObB3J936DM7W7zF7fEdqvFAOjwL5OLFqNKO/WMHsJdsfyLTd3NzoO2gIPbq9htUah19Ac8qULccicz0KDmlFtRq12Bq2mYAmDXF3d2e4ubwe2L+PFcuWUrbcE7RpEQhA957vUL1GLaZNnczvx4+jlKLw40UYPGxEhnPmxPU9uZxZUZ9Dho9i8sRxWK1Wcud+iMEZXP/d3Nx4r/9genV/nbi4OJr6B1K6TDl+NPclzVu0omr1mmzbsplgv0a4u7szZMTYe473k2mT+eO346CM7f6AISMynLPfwCH07PYa1rjE+lxs1mdQfH1u2UxgU6M+h40y6vPChShGDBlIXJyVuLg46jVoRA2Hk//M4urmRo/3BtG/Z1es8fv30mVZ+qOxf/drHsJL5v69XVBj3N3d6Wfu3wFGDOjNtatXcHVzo1ffweR/JPGBixvWSvf5/xKVsXt4/nuUUiMwTpw7A4eA3Rgt892BHRgn8ZeB1cDHWuuflVI3tNb5bMZxEqPlfiiQW2vdjWQopQoB+7TWJWzeswC7gHCt9Uvmex2BcUBFIBrYaea7hdHSX01rHamUKgjk11qfSmF6C4ATwFCttdXsqn5Ra51XKdUceBPjwXQFgT3AS1rrcKVUE+A18zuV0VrfVUoFAX5a6w6p1GVHoIrW+m2l1FJgodb6O/N9f611oFLqG8ALaAqUwmj5L6u1vp3SeP+Kuu0UC7X3o+l7qnp2ychBikjqyq272R0hTR7Lmzu7I6RJXJxTrO5Osx5N3nQiuyOkSXCFjP9Jpwfh+cb9sztCmkTu+OjehXIAVydZj6xOcmwda3WOnK4uzjHfr9+OvXehHKJIgdzOUammmbtP58iFtfOLxbOlHqULffqFAYWB7WZX8ttAmNb6PDAQ2AAcwDjx/vke43oHcFdKJfv0Dq31RWCr+eC4SeZ7EcAxYJZD8S0YrfO/YnSt36O1PgoMAdYopQ4Ca83sKXkNKAScUErtBdYB8UcfP2H0ODgAhAL9tNbxN+mswbhdYJ3WOv7sZAPwdGoPsYv/mub/PYFOZs72QC+bMr9hnLivBLqmdvIuhBBCCCGEEP9W0oU+nbTW64FcNr8/YfPzXBLv47YdJp/D7yVtfu10j+m1sf1dKfUwUI6kXdUjtdZJ/j6Q+QC5tDxEDq31NYxW9uQ+00Bf8+X4WQzGib/te5cw7pFPTSHgkln+JMYDAB3H3fHeyYUQQgghhBDi309O4J2IUqoeMBOYorXO2B/NzGZKqa5AR4yn2wshhBBCCCFEEtJl3J6cwOcA5r3u65P5qK7ZjR4ArfU6oLhjIa31N8A36ZjeTsDxBuz2WuuM/R2PlKfXCfsu8QBbtdaVsmJ6QgghhBBCCPFvJCfwOYB5kl75AU7vpXuXytTpzSLpPftCCCGEEEIIIdJBTuCFEEIIIYQQQuRIzvKXXB4UuaVACCGEEEIIIYRwAnICL4QQQgghhBBCOAHpQi+EEEIIIYQQIkeSDvT2pAVeCCGEEEIIIYRwAnICL4QQQgghhBBCOAHpQi+EEEIIIYQQIkdykafQ25EWeCGEEEIIIYQQwgnICbwQQgghhBBCCOEEpAu9EEIIIYQQQogcSTrQ25MWeCGEEEIIIYQQwgnICbwQQgghhBBCCOEEpAu9EEIIIYQQQogcSR5Cb09O4MW/Tt6HXLM7QprE6exOkDYuOElQJ5HvIefY7MY5ywLqJDt1Zzn4eON/JbI7Qpo8nNs5tvOROz7K7ghp4vVyz+yOkCZROz7O7ghp4ySbT2f501zOkRIK5s2d3RHEf4R0oRdCCCGEEEIIIZyAczQFCSGEEEIIIYT4z1FO0lvkQZEWeCGEEEIIIYQQwgnICbwQQgghhBBCCOEEpAu9EEIIIYQQQogcSVqc7Ul9CCGEEEIIIYQQTkBO4IUQQgghhBBCCCcgXeiFEEIIIYQQQuRI8hR6e9ICL4QQQgghhBBCOAE5gRdCCCGEEEIIIZyAdKEXQgghhBBCCJEjSQd6e9ICL4QQQgghhBBCOAE5gRdCCCGEEEIIIZyAdKEXQgghhBBCCJEjyVPo7UkLvBBCCCGEEEII4QSkBV4IIYQQQgghRI4kLc72pD6EEEIIIYQQQggnICfw4j9p57YttA1qSutAX77/5qskn2utmTZ5HK0DfenYOpDfjh8F4PTJv+ncJijh1aj2SyyY+x0AJ34/TrfObenQKpAB777FzRs3Mpxz25YwmjdrhH+TBsz6ekayOd+fMAb/Jg1oGeTHsaNHAAgPP88bXV4lyL8xLQKbMvf7bxOGWbtmFS0Cm1Ll2ac4euRQhjMCbN0SRkCzRvg1bsDMr5LPOXH8GPwaNyCkeWJOgBFDB+FTqyrBgc2SDPfDnO8IaNaIoICmfDhlUo7M+enH0whp7kfL4AC6vdGZyMiIDOfctjWM5n6+BDRtyDdff5lszkkTxhLQtCGtgv05fixxvr/ZpQPBAU0ICWzKD3MS5/vvvx2nU/tWtAzy490e3biRCctnVtTnF599TIO6NWkZHEDL4ADCNm/KnJxNG+Hnm0rOcWPw821ASKBDziGD8KlZleAA+5xrV68iyL8pz1d6iiOH074ebQ3bjF+ThjRtVJ+vv0w+y4RxY2jaqD7Bgc3ssqQ07NUrV3jztU40823Am6914trVqwDE3L3L0MEDCQpoRotAP3bv2plkej3f6kpz/6Zpzg9Zs/0cPrBPwvshfg3o3CYoXZmS4yzrkbGd9yWgyT1yNmlIqyB/jh91yOlv5rTZzk//7BN869WiTYtA2rQIZEtYxtejtPpieFtOrR/PnoWDHtg0bWXFfvOzT6bRMsiP1i0C6P5mZ6IyYTtvl/k+l9U7d+7wapsQWrcIICSwKdM/+zhTcwFs3xpGsL8vzZs1ZPbM5LNNnjiW5s0a0qaFfbaObUNoExJAy+ZNmWGTbVC/d2kbEkjbkED8fevSNiQwwzmzYn3/7fgxOrZrSZuQQNq3DubwoYP3lW3rls34N21IM9/6qe6DmvnWp4Xjdj+FYY8fP0b7NiGEBPnTJqQ5h8xs586d5aUXniEkyJ+QIH/GjBx2X5lFDqe1lpe8/lWv8Kt3dWqvc5eide06PnrfkT/1mQs3tW+TpnrH/mN2ZX5asU6379BZn79yR6/fslv7BwYlO56XX35F/3r8pA6/elf7+QfqVaFbdfjVu/rr7+bpMRM+SDXH9dtxqb6u3IzRdXzq6mMnTulL12/rJk2b6QNHfrcrs3LtBt2xUxd9Ldqqt+3cpwObB+vrt+P0X2fC9a59h/T123E6/OI1Xa9+/YRhDx79Qx86dkK3btNW79x74J45bt5J/XXtVoz28amrfztxSl+5YeQ8ePR3uzKrzJw3blv19l37dPOg4ITPNm/dqffsP6R9GzexG2Zj2Dbdrn0Hffn6bX3zTpw+80/UPbNkR86Ii9cSfv5y5jd64OChqeeItqb6unzjrq7jU1cf/eOkvngtWjdp2kz/evg3uzIr1oTqjp266Ku3YvXWnXt1YPNgfS3aqv86fV7v2ndQX4u26vMXrul69eonDBsQ2FxvCNuur0Vb9XdzF+iJk6akmiO76vODqdP059O/TPu8vZv661q0mfPPU/rKTTPnsd/tyqxat0F37NxF37hj1dt3mznNzzZv26n3/GrmtBnm0LE/9JHfjPVo174D98wRHaP1jdux2qduXf3HX6f1tZt3dNOmzfThY3/o6Bid8FqzfqPu1LmLvnU3Tu/cs183Dwq+57Djxk/Un3w2XUfHaP3JZ9P1uAnv6+gYrWfN/l6/12+Ajo7R+mz4Be0fEKhv3rEmTOuXFat1r3d6a9/GTRLey67tp+1ryIixesLkaanmcJb16Nrt1F+Xb5o5T5zUF6+bOY/8ZldmxVozZ7RNzttW/dcZM+dtqz5vbufjh500ZZr+9Isv7zn9+Jd75bcy7VW38xT9cqvx+vAf5zJ1vO6V38q2/eb5C9cShp/xtbGdTy3HvZbPzFpWr96K1eEXr+lr0VZ96dptHdg8WG/duTfN075yK/XXxetGtiO/n9RRV6N146bN9L5Dv9mVWb46VHfo1EVfvhmrw3YY2a7csurLN2P1Pxeu6Su3rPrCVSNb2I69SaYxYvQ4PXnqR6nmyK71vf2rHfXKtRsShm/Vpm2qOW7d1Ule16NjtY9PXf37n6f1VXPbfejYH3ZlVq8ztvs378TpHbuN7f69hn21Qye9Zv3GhOFbt2mnb93V+o+/z2jfxk2SzWL7yu5j9fS+fjxwXufEV3bVh7TAm5RSJ5VSHmkoV0Ap1f1BZDKnd1Iptdjm92Cl1DcPavpZQSl1300WaZ1PqTl25BBFihXn8aLFyJUrF3Xr+7JlU6hdmS2bNtCwiR9KKSpUepYb169z4UKUXZm9u3fweNFieBd+HIDTp0/y7PNVAKjyv1fYtGFtRmJy5PBBihUvTtGixciVKzcNGjVm44b1dmU2bVhPk2b+KKWo9Gxlbly/RlRUJJ6eXjz1dAUA8ubNR6lSZRJahkuVLkPJUqUzlM3W4UNmzmJGzoa+yeds6mfkfObZylw3cwK8UOVFHn300STjXTh/Hp26vE7u3LkBKFioUI7MmS9fvoSfo6OjM/yk1COHD1KsmP1837TRfvnctCGUxvHz/Rkj54WoSDw8vSj/VPx8z0vJ0onz/dTJv3n+hRcBeOmVqoSuz9jymVX1mdmSzRma8Zyly6R/PTp86CDFipUwsuTOTaPGTZLU2YbQ9TTzC0iSJbVhN2xYj19AAAB+AQFsCF0HwF9/nuCll18GoFChQuTPn58jhw8DcOvmTb6bPYvX3+yWru+QVdvPeFprNqxbRd2GjdOVy5GzrEfJbec3bUgl57MOOZ+2yWmznc9OW/f9yaWrt7Jl2lm133TczkPmPRE7I8uqUoqHH84LQGxsLLGxMahMzla0WHGKxGdr2JjNDtk2bwylcdM0ZnPYP2qtWbdmFQ0aNclwzqxY35VSCb0pb9y4gaenV7qzGfugEjb7oCZJ9kEbN6ynaUrb/RSGNbLdNLNdx9Mr/dmE85ITeEAp5ZqO4gWAB3YCb6qilKrwICaklMqyBxsqQ7YvcxeiIvGyeCf87mmxJBysJ5aJsC/jZeGCw4FR6JqVdgeZpUqXZcvmDQBsXL+GyIjwDOWMjIjAYimc8LvF4p2k215kZAQW78QyXsmU+efcWY4fP0bFSs9mKE+KOR0yWCzeREUkzentUOZeB5qnTp1k/749tG8TQpeO7dLVTflB5gT45KOpNKpXm5XLl9HtrZ4ZzBmJxTtx2fPyshDpkDMqMgJvm+XTyGm/DP9z7hy/2cz3MmXLJRzQrFuzmojw8xnMmXX1Oe+HOYQ092PE0EEJ3cEzNafjehSRTM6IzD8RioyIwLuwzby1WIhIps5s5398ltSGvXTxYsKBpaenF5cuXQLgiSfLszF0PbGxsZw9e4ZjR48kzPdPP57Gqx07457HPV3fIau2n/EO7N9LwUKFKFa8RLpyOXKa9SgiEovFfr46riNRkRF4Oy4T98gJsGDeHFoF+TNy2GCuXcvYeuQssnK/+elHU2lcvzarMmE7b58nY8uq1WqlTUgg9etU56WXq1Lxmczb10c5ZrNYUqhP2zL22dqGBNLQpzr/e7lqkuOQ/fv2ULBQIYqXKJmhnFm1vvfpN5BpUyfTpEEdpn3wPm/3fPc+sjmuv0nXcWMflMx2P5Vh+/YfxNQP3qdh3VpMmTyRnu/0Tih37txZWgYH0KVjO/bt3ZPuzCLny/aTqYxSSvVTSvU0f56qlAo1f66rlPpeKdVaKXVIKXVYKTXRZrgbSqlRSqmdwCs27+dRSq1SSr2ewiQnAGWUUr8qpSYppb5TSvnbDD9HKeWnlOqolPrZHNdvSqnhNmXaKaV2meOYnoYLCJOBJDeTKaXyKqVmKqV2K6X2x+dQSu20PeFXSm1USr2QSvmOSqmFSqlfgDUp1PNnSik/8+eflFIzzZ+7KKXGmD/3Nuv5sFLqHfO9kkqpY0qpz4B9QDGbcXoopbYrpZoopTyVUovNbLuVUtXMMoWUUmvMvNNJ4bK3UuoNpdQepdSe72YlvSfTltY6ueHTVSYmJoatmzdSp26DhPcGDBvNTwt/4LX2Idy6dZNcuXKlmuNekiZILmfqZW7duknf3j15r99AuxaETJV8UPsiyeW8RyuB1Wrl2rVrfDtnPu/26Ue/995Jdr5kd06At3u+y6p1G/Ft0pT5P3x/nwFTDpFkvifzZRzne78+PenTd0DCfB82ciwL582lXaugTFk+s6o+W4S05pcVa5m3aAkenp5MmTwx1fL3dL85s+Bvzt5rvqUURimVtmEdBDQPwmLxpk1IEJMmjOPZys/h6ubK8WPHOH36NHXr1U/fFyDrtp/x1q9ZQd0GGWt9N0PcO2dOWI/SMF/vVZ+3bt2kX++e9OmXmDO4ZSuWLF/D3IU/4eHhydTJ72cwp3PIyv3mWz3fZcXajTTKjO38PQKlZ1l1dXVl7oKfWLFmA0cOH+LEH79nYrRkK8uxUKrZ5iz4iWWrN3D08CH+PGGfbc2q5TTMYOv7vTIkFLmP9X3Rgnn07juA5Ws20LvvAEaPGHIf0e5/HU9t2IXzf+C9/gNZvX4T7/UbyMhhgwHjIu6qtRuYv2gJffoOYGC/PpnyrI7spnLoK7s4/Qk8sBmoYf5cBcinlMoFVAf+ACYCPkBl4EWlVIBZNi9wWGv9ktZ6i/lePuAXYK7WOukTMAwDgD+11pW11n2Br4BOAEqpR4GqwAqz7P+Atua0WyilqiilngJaAtW01pUBq1kmNQuA55VSZR3eHwyEaq1fBOoAk5RSeYF5QIiZqTDwuNZ6byrlwbiI0UFr7ZNCBtt6LgI8bf5cHQhTSr1g1sNLwMvA60qp58wyTwLfaq2f01qfMnNZgOXAMK31cmAaMNXMFoRRrwDDgS1a6+eApUDx5MJprWdoratorau07/RaCl/B4OllsWsdj4qIwMPD06GMt32ZyAgK2XSd2rEtjHLln6JgocTe/CVKlmbKJ1/y1XcLqNegMY8XKUZGWCwWIiISW3ciIsLxcOi+ZbFY7FqAIm3KxMTE0Ld3T3ybNMOnXtID5czi5ZAhIiI8SVcui8VC+D3KOLJYLNStVx+lFBUrPYOLcuHy5cs5Lqct38ZNWb8uY11qjZyJy15kZESSDF5e3oTbLJ8REeF4ehrLcGxMDP1696JRY/v5XrJUaT6d/jXfz1tMw0aNKVI02VUpnTkzvz4LeXjg6uqKi4sLzYNacDiDPS+Szem4HnlnbL6nlcXiTfh5m3kbEYGX47y1eNvN//gsqQ1bsFChhFbwqKhIChYsCICbmxt9BwxiwY8/M+2Tz7l+/TrFi5fk4IH9HDt6GN/6PnRs34ZTJ0/SpWP7NH2HrNp+gtHNdvOGdfjUb5SmLKlxqvUown6+Oi6fXhZvwh2XCcecDtv5QoUS16PAoBYcuc+HbzmbB7Hf9G3clNAMbudtZXRZjZf/kUd44cX/sX3bFjJLkmwpLJ/2ZZLP9nyV/7F9a2K22NhYNq5fR72GvpmfM5PW92W/LMGnrnGhs16DRvfVE9CSZP1NWocW72TW8fjtfgrD/rL0J+qaWRs09E14wF7u3LkpUOAxAJ6uUJGixYpz6uTf6c4tcrZ/wwn8XuAFpVR+4A6wHeNEvgZwBdiotY7SWscCc4Ca5nBWYLHDuH4GZmmtvyWNtNabgLJKKS+gNbDYnBbAWq31Ra11NPAjxsluXeAFYLdS6lfz93vdSGkFJgEDHd5vAAwwx7MRcMc4wV0AtDDLhAAL71E+PuulVDKEATWUUk8DR4EI8+LAK8A287v9pLW+qbW+YX7f+BP+U1rrHTbjygWsB/ppreP3gvWAT8xsS4FHzHlaE/gewDzRv/8zOFP5pyty9vRp/jl3lpiYGNavXUm1mnXsylSvWZvVy5eitebIoQPkzZfP7iB1/eoV1HNoJbp86SIAcXFxfDtzOv5BIRnK+XSFSpw5dYpzZ88SE3OXNatWUKu2/fWVmrV9WP7Lz2itOXTgV/Llz4+npxdaa0YPH0KpUmVo92qnDOW4lwoVK3HaJufqlSuo7ZCzVh0fli01ch488Cv58uW/571ktX3qsWun8dTsUyf/JiYmhsceeyzH5Tx16mTCz5s2hFKyVKn7zgjmfD9tP99r1rJfPmvVrsOK+Pl+0MjpYc73USOGUKp0adq92tFumEsXE5fPr7/8gqAWLTOUM6vq07Y7duj6dZQpWy7jOU875KzjkLN2+nPef5aTnD17hpi7d1m1Yjm1HLLUruPDL0uXJMmS2rC16/iwdMkSAJYuWUKdOnUB417dW7eMe5G3b9uKq6srZcqWJaRVG9Zt3MLKtaF8891cSpQsydfffJem75BV20+Avbt2ULxEabvu9/fLWdaj5LbzNWunktPczifkHD6EUqWS5rxgsx5tCF1LmXIZW4+cRVbtN0/bbuc3Znw7nyTzfS6rly9d4vq1awDcvn2bXTu2U7JkFmQ7Z2ZbvYIaDtlq1KrDimVpyLZzOyVs6m23+bslB6/vnp5e7N2z28i7a8d93doTv+0+d/aMuQ9Kut039kEpb/eTG9bT04s9u3cBsGvnjoTbEC5duoTVagXg7JkznD59kqLFMtagJHKeLLvf+UHRWscopU5itP5uAw5itC6XAU5jnCwn57bW2urw3lbAVyk1V6evr+53GK3orYDOtvEc42L0uJittXY8GU/LNAYCR2zeU0CQ1vo3x8JKqYtKqWcwWvvfTK28Uuol4GZqE9dan1NKPQY0wmiNL4hxceCG1vq6Sr0vp+O4YzEuvDQE4v+2jQvwinmxwzYbJN8r7r65ubnxTr9BvNfzTeKsVhr7BVKqTFl+XjwfAP+glrxcrSbbt4bROtCXh9zzMHDY6IThb9+OZs+u7bw3aLjdeNetXsFPi+YBULN2PRo3y9ifRXFzc6PfoKG83a0LVmsc/gFBlClbjkULjGkEh7Sieo1abA3bjH+TBri7uzNi9DgAft2/j+XLfqZsuSdo3SIAMLr/Va9Ri9D1a5k0fgyXL1+i11tdeaJ8eT794usM5ew/aCjdu3YhzhqHf6CRc6GZs4WZc8vmzfg1NnOOGZcw/IB+vdm7ezdXrlymYd1adH2rB4HNgwkIbM6IoYMJDmxGrly5GDV2Qoa6NWdVzo8+/IBTJ0/iohSFH3+cwUNH3nfG+Jx9Bw6hR7fXsMbF4RfQPMl8r1ajFlu3bCagaUPc3d0ZPsrIeWD/PlYsW0rZck/QxvyzPN17vEP1GrVYvWo5C+fNBaBO3fr4BTTPcM6sqM9pUybz2/FjKKUoXKQIQ4ZlvD77DxpK9zcdcs43c7ZsRfWatdgSthk/3wa450lcjwAG9HXI2b0HgUHBhK5by8TxY7h86RI9u3flyfLl+WxG6uuRm5sbAwcPo9sbrxEXZyUgMIiyZcuxYP4PAIS0bE2NmrXYsnkTTX3r4+6eh1FmnaU0LEDn196gb+93WPLjIrwLF2bylGkAXLp0kW5vdMHFxQUvLwtjJ2S8G3VWbT8B1q9ZmSmtcfE5nWU96jvIzGlNJWfYZgKamDlHJ5OzhZmzp5Fz2tTJ/H78uLEePV6EwcNGZChneswe35EaL5TDo0A+TqwazegvVjB7yfYHMu2s2m9+bG7nlYuicOHHGZTB7bxj5vtdVi9ciGL4kIHExVmJi4ujfoNGSU6wM5xtwBB6dnuNuLg4mvkb2RYvNLIFtTCybduymebNjGxDRyZmGzk0MVu9Bo2oYXOxb82qFRl+eJ1dzixY34cMG8Xk98dhtVrJnfshBg8bdV/ZBgwaRrc3XyPOasXf3HYvNLf7LeK3+2GbaOZbH/c8eRg5elyqwwIMGzma9yeMwxobS+6HHmLocCPbvr27+eyTj3BzdcXF1ZUhw0by6KMFMlS/OUEW3NXm1FSG7inNIZRSIzBOnDsDh4DdGCeI3YEdGCfxl4HVwMda65+VUje01vlsxnESo+V+KJBba53so3mVUoWAfVrrEjbvWYBdQLjW+iXzvY7AOKAiEA3sNPPdwmjpr6a1jlRKFQTyx3ctT2Z6J4EqWusLynj6/QCMbvAdlVLjgEeAHlprrZR6Tmu93xzuLYzW8ee01hXM95Itb2atorV++x71/A3G7Qg+QCFgEbBIa/2uUup54BuM7vPK/L7tzXpfprWuaDOeG8CjGD0DdmmtJyil5gL7tdaTzDKVtda/KqU+AiK11mOUUr4Ytyd4aq0vpJQz4lqMUyzUeR9yjutnLrLRzFTWOKdYPHF1lhnvJDFdnOTo4+qtmOyOkCYP507Ps2ezkXPMdrxezryHsmWlqB2Z/3fOs4KzHFs7ye7IaY5D3Fydp2NznlzOsnUy/HwoPEcurf6VvLOlHp1nSUtdGFAY2K61jgBuA2Fa6/MYrdYbgAMYJ94/32Nc7wDuSqlkmyq01heBreaD2iaZ70UAx4BZDsW3YLSc/4rRtX6P1vooMARYo5Q6CKw1s6fF19j3mhiN0R39oFLqsPl7vEUYPQIWpLF8WoQBblrrExgPpCtovofWeh/GCfwujJP3r+IvJiTH7P3QCqhjXpjoifG0/YNKqaNAV7PoSKCmUmofxi0Ap9OZWQghhBBCCCH+Ff4VLfDZTSn1MEbL//Na66vmex1JQ6u2yHzSAp+5nOXKt7OQFvhM5iQxpQU+c0kLfOaSFvjM5SzH1k6yO3Ka4xBpgc86vxyKyJFLa7NKFmmBd0ZKqXrAcYyu+VezO48QQgghhBBCiH8n52gCzAbmve7rk/mortmNHgCt9TqS+dNmWutvMLqUp3V6O4GHHN5ur7XO2N9QSielVCWMbv+27sTf2y+EEEIIIYQQInvICXwKzJP0yg9wejniBNm8YFA5u3MIIYQQQgghhJPchfbASBd6IYQQQgghhBDCCcgJvBBCCCGEEEII4QSkC70QQgghhBBCiBxJOddD87OctMALIYQQQgghhBBOQE7ghRBCCCGEEEIIJyBd6IUQQgghhBBC5EjyFHp70gIvhBBCCCGEEEI4ATmBF0IIIYQQQgghnIB0oRdCCCGEEEIIkSO5yFPo7UgLvBBCCCGEEEII4QTkBF4IIYQQQgghhHAC0oVeCCGEEEIIIUSOJE+htycn8OJfJ7+7cyzWLi6yNfovUk6yF4rTOrsjpImrs9RnnHPUZz4n2X46yWwH55jtRO34OLsjpInnyz2yO0KaXNr1SXZHSJPYuLjsjvCv4jTbJeH0pAu9EEIIIYQQQgjhBJzjUrsQQgghhBBCiP8c6d1gT1rghRBCCCGEEEIIJyAn8EIIIYQQQgghhBOQLvRCCCGEEEIIIXIkhfShtyUt8EIIIYQQQgghhBOQE3ghhBBCCCGEEMIJSBd6IYQQQgghhBA5kov0oLcjLfBCCCGEEEIIIYQTkBN4IYQQQgghhBDCCUgXeiGEEEIIIYQQOZI8hd6etMALIYQQQgghhBBOQE7ghRBCCCGEEEIIJyBd6IUQQgghhBBC5EhKetDbkRZ4IYQQQgghhBDCCcgJvBBCCCGEEEII4QSkC70QQgghhBBCiBxJnkJvT1rghRBCCCGEEEIIJyAn8OI/aeuWMAKaNcKvcQNmfjUjyedaayaOH4Nf4waENPfj2NEjCZ+NGDoIn1pVCQ5sZjfMF599TIO6NWkZHEDL4ADCNm9Kf66wzfg1aUjTRvX5+svkc00YN4amjeoTHNjMLldKw169coU3X+tEM98GvPlaJ65dvQpAzN27DB08kKCAZrQI9GP3rp3ZmvOTjz4kOLAZIc39efP1zkRGRgCwfdtWWrVoTlBAM1q1aM7OHdvTnPNB5I43e9bXPFvhSS5fvpT+XFs249+0Ic1866e8PI4bQzPf+rRwzJXCsJ9/+jH1fWoQEuRPSJB/kuXx/Pl/eOXF55g96+t0502cdhiB5no0K4Xc76eyHtWtVZUWDuvRb8eP8WrblrQKDqBtyyAOHzqY/lwPcD5fuXKZLh3b83KV5xg3ZlS6s9rlzoLtUrxvv/ma5yqV5/LlyxnKmFzm+1kGwsPP80bnV2nu15jggKbM/f7bTM0Vny2gaSP8fFOpz3Fj8PNtQEigQ30OGYRPzaoEB9jX59rVqwjyb8rzlZ7iyOFDmZczC+b7D3O+I6BZI4ICmvLhlEkZzrltSxjNmzXCv0kDZn2dwryeMAb/Jg1oGeQwr7u8SpB/Y1oE2s/rzz6ZRssgP1q3CKD7m52JMrf7D8IXw9tyav149iwc9MCmmRXb+uPHj9G+TQghQf60CWnOIYdtZka39cZ89yWgSUO++frLZDNPmjCWgCYNaRXkz3Gb+f5mlw4E+zchJLApP9jM9+mffYJvvVq0aRFImxaBbAlL//HSg8j52/FjdGzbkjYtAmnfKvi+9keQNfukKZMn4t+0EcGBzXin51tcu3YNyNxjJpGDaa3lJa9/1evmnTid2uvarRjt41NX/3bilL5y47Zu0rSZPnj0d7syq9Zu0B07ddE3blv19l37dPOg4ITPNm/dqffsP6R9GzexG+aDqdP059O/THXatq/oGG33unE7VvvUrav/+Ou0vnbzjm7atJk+fOwPuzJr1m/UnTp30bfuxumde/br5kHB9xx23PiJ+pPPpuvoGK0/+Wy6HjfhfR0do/Ws2d/r9/oN0NExWp8Nv6D9AwL1zTvWJLkeVM6oy9cThv9q1mw9aMhQHR2j9b6DR/Spc+E6Okbrg0d/09WqV79nxgeZOzpG679P/6M7dOysa9Wqrf+JvJhqjlt37V/Xo2O1j09d/fufp/VVc9yHjv1hV2b1OiPXzTtxesduI9e9hv1g6kf68+lfJZle/Ktb97d197d7pFjmxp24VF9XzfXo+IlT+rLNemRbJn49un7bqreZ61H8Z5u27tS7zfXIdphXO3TUq9dtTBi+dZu2qebI7vl88epNvXXHbj37+7l66PCRKc737Nou3bwTp/88dU536NhJ16xVW58Nv5hqjnvN98xaBv4+G6537z+kb9yJ0xGXrul69esnGTa11827qb+uRZv1+ecpfeWmme3Y73ZlVq3boDt27qJv3LHq7bvN+jQ/27xtp97zq1mfNsMcOvaHPvLbCd26TVu9a9+Be+bIrvm+MWybbte+g758/ba+eSdOn/knKtUc12+n/rpyM0bX8amrj504pS9dN3IeOPK7XZmVZs5r0Va9bec+Hdg8WF+/Haf/OhOud+07pK/fjtPhF415HT/s+QvXEoaf8fU3euDgoanmcK/8Vqa96naeol9uNV4f/uNcpo7XvfJbyW5Ts2pb/2qHTnrN+o0Jw7du0y7N2/prt62pvi7fvKvr+NTVR0+c1BevR+smTZvpX4/8ZldmxdpQ3bFTF301OlZv3blXBzYP1tduW/VfZ87rXfsO6mu3rfq8Od/jh500ZZr+9Isv7zn9tL6yKmf7Dh31yrUbEoZv1aZtqjke5LHH+o1h+np0jHGcN+H9hGO7tB4zZfexenpfm367qHPiK7vq4z/bAq+UsiqlflVKHVZK/aKUKnAf46iilPoohc9OKqU8Mhz0PimlHtglZcfvqpSqrZRalkLZjkqpTx5UtuQcPnSQYsWLU7RYMXLlyk1D38Zs3LDersymDetp6uePUopnnq3M9evXiIqKBOCFKi/y6KOPZk2uYiWMXLlz06hxkyS5NoSup5lfQJJcqQ27YcN6/AICAPALCGBD6DoA/vrzBC+9/DIAhQoVIn/+/Bw5fDjbcubLly9h+NvR0Sjzb4Y89dTTeHlZAChbthx379zl7t276a3eLMsNMGnieN7t0zchc7pzFS9hszw2YWOofa6NG9bTNKVc9xg2OaHr11GkaFHKlCmX7ry2uYveYz3aeD/rkVLcuHkDgBs3ruPp6ZXuXA9yPj/88MM8/0IVHsr9ULpyJps7i7ZLk98fT6/efTP9z/BkZBnw9PTiqacrAJA3bz5KlSpDZETmtb4mW5+hGa/P0mXKULJU6azNmQnzfeH8eXTq8jq5c+cGoGChQhnKeeSwmbOokbNBo+RzNmlm5Kz0bGVupDavzZZ22+1+dHQ0PMD7XLfu+5NLV289sOll1bZeKcXNGzcBc5vplbjNzOi2Prn5vmlDqF2ZTRtCaWwz369fv8aFqEg8PL0onzDf81LSZr5ntqzKqZTiZvz+6PqNdO+PIOv2SVWrVcfNzXiU2TPPViYyIhzIvGMmkbP9Z0/ggWitdWWtdUXgEvBWekegtd6jte6Z+dEyxYPrE5ZGSqkc8dDEyMgILN6FE363WLyJcjhwjIyMwNuhTFp2PPN+mENIcz9GDB2U0FU9zbkiIvAu7J3wu5fFQkQyuSzeiWUsFm8iIyJSHfbSxYsJOx1PTy8uXTK6/j7xZHk2hq4nNjaWs2fPcOzoESLCz2dbToCPp02lQd1aLF/2C93f7pVk2uvWrKb8U08lHJSmR1bl3hi6Hi+LF0+WL5/uTPHT9LabpiXJshYZ4VjGzHWPYef9MIcWgc0YPmRgwvIYfesW38z8kq7d376vvPGiHNYRLzOT43ezOJS5VxfZ9/oPYtoHk/CtV5upH7zP2+/0TleunDqf75k7i7ZLGzeE4uVl4cknMz93Zi0D/5w7y2/Hj1HxmWczLVuy9ZnseuVQn5l4ESEtsmq+nzp1kv379tC+TQhdOrbLcHf/yIgILJZ71Gca5/Xx48eoWClxXn/60VQa16/NquXL6PZWTj2kyris2tb37T+IqR+8T8O6tZgyeSI9zW1mZmzrIyMisVjst4mOmaOSZPMmMjLSrsw/584Z67jNfF8wbw6tgvwZOWww166l73jpQeXs028g06ZMpkn9Okyb8j5v93r3PrJl3TFTvCU/LqZajZpJ3s/IMZPI2f7LJ/C2tgNFAJRSZZRSq5RSe5VSYUqp8ub7LczW+gNKqc3mewktzUqpQkqpNUqp/Uqp6dhcRlZKtVNK7TJb/KcrpVzN928opcaa49yhlLKY71uUUj+Z7x9QSlVNbTyOlFITgDxmuTlKqdFKqV42n49VSvU08282p3VUKfWFUsrFLNNAKbVdKbVPKbVQKZUvuWndi1JqhFJqhlJqDRB/Y1Exs45/U0oNtym7xKz3I0qpN2zeT7aeHKbzhlJqj1JqT3L3ldnRyQa1L5JMmXs9AbNFSGt+WbGWeYuW4OHpyZTJE1PPkSRW0okmadFNJphSKm3DOghoHoTF4k2bkCAmTRjHs5Wfw9Ut2UXqgeXs0etd1qzfRJOmzZg393u7cidO/MGHUyczdPj93WecFbmjo6P5csYXyV5sSHOuFKaZljKpDRvSsjXLVq5l/uKf8fD04oNJEwDj3vi27Tvw8MN57zuzkSnpe0nrM7khU18uF83/gT79BrBy3Ub69B3IqGFD0pcrh87ne8qC7VJ0dDRff/lFlp0UZcYycOvWTd57tyd9+g+0a43NeLjkJpuG+szsbgr3kkX7I6vVyrVr1/h2znze7dOPfu+9k+z2ImMx01eft27dpG/vnrzXz35ev9XzXVas3UijJk2Z/8P3SUfyL5FV2/qF83/gvf4DWb1+E+/1G8jIYYOBzNrW33/meLdu3aRf75706TcgYb4Ht2zFkuVrmLvwJzw8PJk6+f0MZMy6nIsWzKN33wEsX7uB3n0HMHp4+vZHRrKsPbb7cvrnuLq50qSpn937GT1mymlUDv2XXf7zJ/DmSXBdYKn51gygh9b6BeA94DPz/WFAQ631s4BfkhHBcGCL1vo5c1zFzfE/BbQEqmmtKwNWoK05TF5ghznOzcDr5vsfAZvM958HjtxjPHa01gNI7GHQFvga6GDmcQFaAXPM4v8D+gCVgDJAc2V0hx8C1NNaPw/sAdLXDGbvBcBfa93GZpptgcpAC6VUFfP9zma9VwF6KqXi+/ylVE+233mG1rqK1rpK59fecPzYjpfFYtfSHBERbtflDIyr2+H3KOOokIcHrq6uuLi40DyoBYfT2eJhsXgTfj484ffIiAi8HKbpZfEmIjyxTHyu1IYtWKhQQnfLqKhIChYsCICbmxt9BwxiwY8/M+2Tz7l+/TrFi5fMtpy2fJs0Zd3aNYnDh4fzbs+3GTNuIsWKF79nxgeV++yZ05w7d5aQ5v741vchIiKcVsHNuRAVlb5cdtOMSNJNz+LtWMYmVwrD2i2PwYnL46FDB/hwymR8G/gw5/vZfP3l9CQXS9LCy2EdiUxmHXFc15Ir42jZ0iX41GsAQP2GjThyOH0PDcqp8/lesmK7FJ+7ZbA/jRv6EBkRQZuQ5ly4kDm5M7oMxMTE8N67PWncpBl1zXmeWZKtzyTrVfq385ktq/ZHFouFuvXqo5SiYqVncFEuGXqAocViISLCPoOHY30mM6/jy8TExNC3d098mzRLWL8d+TZuSui6tfedMafLqm39L0t/Slh/GjT0TXjQWmZs640WX/ttomNmryTZwvH09AQgNiaGfr170chhvhcqlLh/CgxqwZH7fDhcVuc09kf1AajXoNF99WTJymOmpUt+YvOmjYyfONnuxD4zjplEzvZfPoHPo5T6FbgIFATWmq3MVYGF5mfTgfj+YFuBb5RSrwPJNVPWBL4H0FovB+L3lHUxTmB3m+OsC8TfQHcXiL9XfC9Q0vzZB/jcHJdVa331HuNJldb6JHBRKfUc0ADYr7W+aH68S2v9l9baCvwAVAdeBp4GtprT6gCUSG0S93hvqdY62ub3tVrri+Z7P5rTBOOk/QCwAygGxN+0lVI93ZcKFStx+tQpzp09S0zMXVavXEHt2j52ZWrV8WHZ0p/RWnPwwK/ky5f/nvc+xZ8kg3HfWZmy6bvnrELFSpw+fZKzZ88Qc/cuq1Ysp1Yd+1y16/jwy9IlSXKlNmztOj4sXbIEgKVLllCnTl3AaJ27dcu4/2/7tq24urpSpmzZbMt56tTJhOE3bgillHmf6bVr13i72xv0eqc3zz3/QrrqNKtzl3viSTaGbWfl2lBWrg3FYvFm3qIf8TAPCtKT69zZM+bymDRXrdo+LEslV3LDOi6PZc3lcda3c1m5JpSVa0Jp264DXV5/k1Zt2t1XfZ5xWI9qZcJ65OHpxd49uwDYtXMHxYqntulJPldOnM9pyp3J26VyTzxJ6KZtrFgdyorVoXhZLMxd8CMeHpmTOyPLgNaaUcOHUKp0Gdp16JQpeRyznT7tUJ/JrlfpWz6zJGcW7I9q+9Rj107jL4ucOvk3MTExPPbYY/ed8+kK9vN6zaqk87pmbR+W/2LkPHTgV/LlT5zXo4cPoVSpMrR71X5en7bZ7m/aGErJUqXuO2NOl1Xbek9PL/bsTtxmFi9REsicbX1y871m7ToOmeuwwmG+e9iu46VK0+7VjnbDXLDZP20IXUuZcvf/PJaszOnp6cXePbsB2H0f+yPIumOmrWGbmfX1l0z75HPy5MmTMK7MOmYSOVuOuCc5m0RrrSsrpR7FODl8C/gGuGK2cNvRWndVSr0ENAF+VUolKUMKvcyA2Vrrgcl8FqMT+/RYSX1+pDaetPgK6Ah4AzNt3nfMrM1prdVat07juC8CjwEXzN8L2vwMcDOZadj9rpSqDdQDXtFa31JKbQTczc/TU0/35ObmRv9BQ+netQtx1jj8A4MoU7YcCxfMA6BFSCuq16jFls2b8WvcAHd3d0aMGZcw/IB+vdm7ezdXrlymYd1adH2rB4HNg5k2ZTK/HT+GUorCRYowZNjIdOcaOHgY3d54jbg4KwGBQZQtW44F838AjC7RNWrWYsvmTTT1rY+7ex5GmblSGhag82tv0Lf3Oyz5cRHehQszeco0AC5duki3N7rg4uKCl5eFsRPS1oUtq3JOm/IBJ0/+jYuLonDhIgwZbtTfvLnfc/rMaWZ88RkzvjA6xHz+5UwKpfOhTFmVO6Pc3NwYMGgY3d58jTirFX9z3AvNXC3ic4Vtoplvfdzz5GHk6HGpDgvw4QeT+O234yjg8SJFGJLJ3eji16O3zPXIz1yPFpnrUbDNeuSfzHo00GY9amSuRwHNgxk6YjSTJozFarXy0EMPpTt3dsxn3/o+3Lhxg5iYGDaEruOLGTPTdDHMMXdWbJeyUkaWgV/372P5Lz9TttwTtAoOAODtnu9SvWatTM3W/U2H+pxv1mfLVlSvWYstYZvx822Aex53Roy2qc++DvXZvQeBQcGErlvLxPFjuHzpEj27d+XJ8uX5bMb9/ynGrJrvAYHNGTF0MMGBzciVKxejxk7I0O0Bbm5u9Bs0lLe7dcFqjcM/IPl5vTVsM/5NzJyjbeb1MmNet24RABjd5qvXqMXHH37AqZMnUS6KwoUfZ9DQ9O03M2L2+I7UeKEcHgXycWLVaEZ/sYLZS7LuT25l1bZ+2MjRvD9hHNbYWHI/9FCmdpl2c3Oj76Ah9Oj2GlZrHH4BzZPM92rmfA9o0hB3d3eGm5kP7N/HimVLKVvuCdq0CASge893qF6jFtOmTub348eN46XHizB42IgcmXPI8FFMnjgOq9VK7twPMfg+6jar9knjx47mbsxdur5mXBSr9OyzDB0+KtOOmXKaB313U06nMnJPlDNTSt3QWuczf34O+BmjC/kmYKrWeqEy9nbPaK0PKKXKaK3/NMvvBzoBBYD3tNZNlfE0+kit9RillC+wAvAEvMxxV9NaRyqlCgL5tdanHDIEA0211h2VUvMwuox/aHbxzwsUTWk8KXy/y4CX1jrG/D03cAjIBZTTWlvNk+aVGK3tp8yfZ2B0U98L+GitTyilHgaKaq1/T2Fak4FbWuthZt6FwBKt9bdKqRHADa31ZLNsR2AcUBGIBnYCnTGeQfCa1rqZMp478CvQSGu9MaV6Smne3rrrHAu1i4tsjf6LnGPphDgnCerqJOtRXJxz1KdzpHSigzknqVAnWTzxfLlHdkdIk0u7svWP7aRZbFxcdkf4V8nl6jwdm93dsvEG7vuw5Y/LOXIrVb3cY9lSj86zpGUhrfV+4ADGveFtgS5mV+4jgL9ZbJJS6pBS6jDGCe4Bh9GMBGoqpfZhdFM/bY77KMb95GuUUgeBtSR2y09JL6COUuoQxol0hfsYzwzgoFJqjpnjLrABWGB2l4+3HZgAHAb+Bn7SWkdhtNb/YE5rB5DaI4xHA2XNOtsPnMC8nSAFW4DvME7SF2ut9wCrADdzeqPNaQohhBBCCCGEMP1nW+D/a8yH1+0DWmit/zDfq43ZgyAbo2U6aYEXOZlzLJ3SAp/ZpAU+c0kLfOZyksVTWuAzmbTAZy5pgc86W3NoC3w1aYEXWUUp9TRGq/j6+JN3IYQQQgghhBDO5b/8ELt/BaXUTuAhh7fba60T/taF2f0+yRPrtdYbgY3pmNZPgOMjYvtrrVendRxCCCGEEEIIIe6PnMA7Oa31Sw9wWoEPalpCCCGEEEII4eI09009GNKFXgghhBBCCCGEcAJyAi+EEEIIIYQQQjgB6UIvhBBCCCGEECJHkg709qQFXgghhBBCCCGEcAJyAi+EEEIIIYQQQjgB6UIvhBBCCCGEECJnkj70dqQFXgghhBBCCCGEcAJyAi+EEEIIIYQQQjgB6UIvhBBCCCGEECJHUtKH3o60wAshhBBCCCGEEE5ATuCFEEIIIYQQQggnIF3ohRBCCCGEEELkSEp60NuRE3jxr3PozLXsjpAm5bzzZXeENLFqnd0R/lVu3onN7ghpkvch59g95HJ1jr26q7McfThJzGu3nGM9eiSPc6xHOMlm/tKuT7I7QpoU/N/b2R0hTWbOHJjdEdKkgsej2R0hTc5eu5XdEdKscQWv7I4gMkC60AshhBBCCCGEEE7ASS4NCyGEEEIIIYT4r3GSzmEPjLTACyGEEEIIIYQQTkBO4IUQQgghhBBCCCcgXeiFEEIIIYQQQuRM0ofejrTACyGEEEIIIYQQTkBO4IUQQgghhBBCCCcgXeiFEEIIIYQQQuRISvrQ25EWeCGEEEIIIYQQwgnICbwQQgghhBBCCOEEpAu9EEIIIYQQQogcSUkPejvSAi+EEEIIIYQQQjgBOYEXQgghhBBCCCGcgHShF0IIIYQQQgiRI0kPenvSAi+EEEIIIYQQQjgBOYEXQgghhBBCCCGcgHShF0IIIYQQQgiRM0kfejvSAi+EEEIIIYQQQjgBaYEX/3kH92zn++kfEBcXR62G/jQL6WD3+bYNq1i+8FsAHsqTh45v9ad46ScSPo+zWhnWqwOPFfKkz8ipmZpt+9YwPpw8HqvVil9gMK92et3uc601UyeNY9uWzbi752HoyHE8+dTTAAQ2qcfDefPi6uKCq6sbs+YsBOCrLz7h558W8dhjjwHQ9e13qFq9VoZy7tgWxrTJE4izWmkaEET7ZHJOmzSe7VuNnINGjE3Ief36NSaOHsZfJ06glGLg8NFUfKYyAIvmzWHxgrm4urpStXpNuvd6L8fl/Hr6p/zy0yIKmPX55lvv8Er1mhnKaWv3jq188eFErNY4fJsF0vLVLnafnz75N1PGDuPE78fo8GYPWrQxlt/IiHAmjR7M5YsXUS6Kxn7BBLZsm2m5IOvmO8Dcb2fx2bTJLFu3JaFu79f2rWFMeX88cXHGetShc9KcU963WY9GjaO8mRPAarXSsU0LPL0sTPn4cwC++PQjwjaGopTisYKFGDZqHJ5eXhnKuW1rGJMnjiMuLo6AwGA6dkmac/LEcWzdshl3d3dGjB5H+acqcOfOHV7v1J6YmLtYY2OpW78hb3bvAcC6NauY8fkn/P33X8yes4CnK1TMUEaAbVtscjZPJWeYTc6nKxAefp7hgwdw8cIFXFwUgUEhtG73KgAD+77LqZMnAWPZyJ//EeYu/ClDOXdt38InUyYSF2elsV9z2nR4LUnOT6ZMYOe2MNzd3ek3dAxPlDfm++J537P858VorWniH0Rw6/YJw/24YA5LFs7D1dWVl6vV5M0evTOUMyvq87fjxxg/egR3797F1dWV/oOHUbHSMxnKaZc5C5bVjNi6ZTPvTxhLnDWOwKAWdH7tjSR53h8/li1hm3B3d2fU2Ak89XSFVIc9fvwYY0cN586dO7i5ujJw6Agq2dTh+fP/0NyvCV27v02HTvbb5cz0xfC2+NasSNSl61RpMS7LppMWf/y6i1WzPyEuLo7nfRpTw7+N3efH92wldMEslFK4uLrS6NW3KFG+EgBT327NQ3keRrm44OLqypvjvsiynPt3bWPWZ5OJi7NS1zeAwNad7D4/d/pvPp00kr9PHKd1p+74hbya8NnyH+eyfsUStNbUaxxIk6A2jqPPEsf27eSnmdPQcXG8VK8p9Zq3s/v80K4wVv7wFUoZ9RfYuSeln8q8dVo4lwd+Aq+Uqg28p7Vu6vC+H/C01npCFk8/APhda300M8qlc9o9gW7APq115h5Jpz3DCOCG1npydkzfzHASqKK1vnAfw27EWH72ZEaWOKuVbz97n35jP6GghxfD3+nA8y/XoEjx0gllPC2PM2jiF+TN/wgHdm9j5kfjGfHhrITPV/88j8eLlST61s3MiJTAarXywcQxTPvsK7wsFjq3a0mNWnUoVbpsQpntWzdz5vQpFv68iiOHDvL++JF8/e38hM8/nf5Nsic/rdq+SttXO2dazikTxjL1sy/xslh4rX1Lqjvk3LE1jDNnTjFvyUqOHD7I5PGj+PLbeQBMmzSel16pzpj3PyQm5i63b98GYN/unYRtCmX2vJ/InTs3ly9dzJE5AULavEqbVzslmWZGWa1WPp08jvHTpuPhZaFHlza8XKM2JUqVSSjzyCOP0O3d/mzbvMFuWFdXV97o8R7lnnyKWzdv8nbnVjz/v5fths1otqyqz4jw8+zZuQ2Ld+FMyTlp/Bg+/sJYjzq2Ndaj0mUSc27bYqxHi5au4vChg7w/diQzv09cj+bP/Y6Spcpw8+aNhPfadehM17d6Jnz+9YzPGDBkRIZyThw3mk+nf43FYuHVNiHUrG2fc6uZ86dfVnH40AHGjxnF7DnzyZ07N198NYuHH85LbEwMXTq2o2r1GlR6pjJlypbj/akfM2708PvOlmzOGWbO1inkPHWKn5at4vBBM+fc+bi5uvJun36Uf7oCN2/epH2rIF56pSqly5Rl/KTEi59TJ08kX758Gc45bdJYJn08A08vb7p1bEXVGnUoWTpx+d+5LYxzZ07x3aLlHDt8kA/fH8NnM+fy959/sPznxXw2ay653HLR/52uvFytJkWLl2D/nl1s27yBr+YszrTtUlbU50dTJ/N617eoVqMmW8I28dHUycyY+W2GsibJnMnLakbyjB8zii++nIXF20LblsHUquNDGZs8W8I2c/r0SZauWMOhgwcYO3oE3/+wMNVhP/xgEm92e4vqNWoRtnkTH34wia+/+S5hnJMnjqdajRr3nTutvvtlB1/M38RXo1+9d+EsFBdnZcXMabQfPIlHCnny5aBuPPlCVbyKlkwoU6ri83R7oSpKKcJP/cnCaaPoMWV2wucdhk4h7yOPZmlOq9XK1x9PYOjEzyjoaWHgW+2pUrUWxUokHtfly/8ond/qy65tG+2GPf33CdavWML4T2bjlisXYwf04PmXqlO4aPEszRxntbL4yyl0HT6VAoU8mdrvdSq+WA3vYqUSyjxR6QUqvlgdpRT/nDzB7A+GM/DjOVmaKydR0ofeTo7pQq+1XprVJ++mAODpexVKR7n06A40TuvJu1IqUy+wZPb4/g3+/P0IXo8XxatwEdxy5eLlmg3Yt32zXZlyTz9D3vyPAFC2fEUuX4xM+OzShQgO7N5K7Yb+mZ7t6OFDFC1anCJFi5ErV27qNfRl88ZQuzKbN4bi29QfpRQVn3mWG9evcyEqKtOzpObYkUMULVYsMWeDxmzZaH8yGbYplEZN/IyclZ7lxg0j580bNziwfy9NA4IAyJUrN/nNuv5p0XzadXyN3LlzA/BYwUI5MmdW+u3oYR4vWozCRYqSK1cuatdrxPawjXZlChQsxJNPV8TNzX71LuThSbknnwLg4bx5KVaiNBeiIsksWVmfH0+ZSLdefVAq4zvso4cPUbRY4npU/x7rUaVnnuW6zXoUERHO1rBN+DcPshvG9gQzOjo6w1mPHD5IsWLFKWrmbNCoMZsccm7aEErjZvE5K3P9+jUuREWilOLhh/MCEBsbS2xsTMLBTqnSZShZslSS6WUoZ3GHnBtSyflsYk4PTy/Km62eefPmpWSpMkRGRtgNq7Vm3epVNPRtkqGcx48eokjR4jxepBi5cuXCp75vkotc2zZvoL6vsXw+XcnYfl68EMWpk3/xdMVncHfPg6ubG88+V4Utm9YDsPTH+bR+tUumbZeyqj6VUgkXnG5cv4GnZ8Z6hyTJnAXL6v06fOggxYqXoGgxI09D3yZsDF1vV2bjhvU09QtAKcUzZh1GRUWmOqxSips3jAvzN25ct+thE7p+HUWKFqVMmXIZyp4WW/f9yaWrt7J8Ovdy7sRxCnoXoaDlcdzcclGxqg+/7dlmV+Yh9zwJ28KYO7ez5aTrxG9H8H68GJbHjf1mtdoN2LN1o12ZRx8rSNnyFXBztd9vnjv9N+WeqshD7nlwdXXj6WefZ9dW++1GVjh94hgehYvg4f04brly8Vz1uhzetcWuzEN5Hk6o27t3biM3hf+3pXoCr5TKq5RarpQ6oJQ6rJRqab5/Uik1Tim1XSm1Ryn1vFJqtVLqT6VUV7OMUkpNMoc7FD+sw/hfVErtV0qVVkp1VEp9Yr7/jVLqI6XUNqXUX0qpYPN9F6XUZ0qpI0qpZUqpFfGfpZB/glLqqFLqoFJqslKqKuAHTFJK/aqUKqOUel0ptdv8jouVUg+nUG6jUqqKOV4PsxUZpVQFpdQus9xBpVSyW3Ol1BdAaWCpUupdpVRBpdQSc5gdSqlnzHIjlFIzlFJrgG/N32crpdaY9d5cKfW+WaerlFK5zOFeUEptUkrtNedFYfP9jea82gT0sslTRim1z+b3ckqpvcnk7mlTh/NsMqYlU11z/h5SSs1USj3kMO48ZvnXzWVtpjkv9iul/G3KzDOnPx/Ik9L8vh+XL0ZRyMOS8HtBDy8uX0z5BHjTmqU888IrCb/PmT6Vlp17oFwy/1pYVFQEXt7eCb97eXkTFWl/AhYVGYnFkljG08tCVFTiAVyvt16jY5tglixeYDfcovlzaRcSwJgRg7l27WrGckZG4GVJbCn1tCRmiHchMhIvi+13sXAhKoJ/zp2hwGOPMW7EYDq1CWLCqGFERxsHKmdOn+Tg/r28/mor3n69A8eOHMqROQF+XDCXDi0DGTdySIbr09bFqEg8bfJ4eHpxwSFzWoSfP8effxynfIVKmZYtq+pzy6ZQPDwtlHuifKbkjIyMwGK7HllSWI/syliIMk+Epk6awNvvvIdSSdfxzz/+kGYNfVi9YhlvdMtYN+BIxwxeFiIj7OszKjICb5v6tFi8iTS/i9VqpU1IIPXrVOell6tS8ZlnM5QnxZwR9tscL4slyUl4VGQE3t7J54z3z7lz/Hb8GBUr2efcv3cPBQsVoniJkhnK6bjseXgls3xG2Zfx9LJwISqSUqXLcXD/Xq5evcLt29Hs3BZGZEQ4AGdPn+LQr/vo3rkN73TtyPGjhzOUM6vqs0+/gUybMpkm9eswbcr7vN3r3QzltMucw5bVyCT1k7QOIyOSqcOIiFSH7dt/EFM/eJ+GdWsxZfJEer5j3CoRfesW38z8kq7d385Qbmdz7dIFHimUeBHjkYIeXLuU9Hjp2K4wPu7dgTkTB+HftW/C+0opvhvXl+kD32TPumVZlvPShUgKedkc13lauJjKcZ2tYiXLcuzgfq5fvcKd29Hs27mVC5Hp3+em15WLURSwqdtHC3ly9VLSTqoHd2xmfI+2fDm2H63fHpDluUTOda+zjkbAP1rrZ7XWFYFVNp+d0Vq/AoQB3wDBwMvAKPPz5kBl4FmgHsbJcMLRnnmS/AXgr7X+K5lpFwaqA02B+Jb55kBJoBLwGvBKMsPFj78gEAhU0Fo/A4zRWm8DlgJ9tdaVtdZ/Aj9qrV/UWj8LHAO6pFAuJV2BaVrrykAV4GxyhbTWXYF/gDpa66nASGC/mW0QYNu37QWzXuJvvCkDNAH8ge+BDVrrSkA00MQ8Yf4YCNZavwDMBMbajK+A1rqW1voDmzx/AleVUpXNtzphzEdHA4DnzJxdbd6/VyZ3c3wtzffdMG4fiJcP+AWYq7X+EhgMhGqtXwTqYCwvec1hbpnTH2vWTRJKqTeUcTFpz5J5yX2NFGidzMiSL3r0wB42rVlKSGdjp71/Zxj5CzxGqXJPpX166aCTyZa0kS+ZMuYXmD5rDrPnLmbKJ9NZvOAH9u817jpo3qIVi5au5tt5P+Lh4clHU97PYM6k7ymHoMl9F5TCarXy+/FjBAS3YtbcxbjnycP3s74CjIO869euMWP2D3Tv1YdhA/okP55szhkY3JL5P69i1g+LKeThySdTJ913xiSZk5u/6Wzpjb51i9GD+tC1V1/y5s1Yt2RbWVGft6Ojmf31DF7rmokHxmlYj1LKuWXzRgo+VjDhXllH3Xq8wy+rQ2nYuCkL52WwK2OyOR3qM5XlwdXVlbkLfmLFmg0cOXyIE3/8nrE8KUpDznt8l1u3btKvd0/69BuQpKv86pXLM9z6bqS8z5xAiVKlafVqZ/r2eIP+vbpSptyTuLq6AuZ26fo1Pv16Dm/26MOoQe9laLuUVfW5aME8evcdwPK1G+jddwCjhw/JQEbHyDlrWb1X/aRWJrVhF87/gff6D2T1+k28128gI4cNBuDzTz+mbfsOCT0J/jvStj966n816DFlNq3eG0XogsTbDTuP/IiuE2bQdsAEdq9ZwsljB7IoZsrHRfdStEQp/Ft1YHT/7owd2IOSZZ5IWPdzgmdersnAj+fQuf84VvzwVXbHeaCUypmv7HKvE/hDQD2l1ESlVA2ttW3T0lKbMju11te11lHAbaVUAYyT7x+01latdQSwCXjRHOYpYAbQTGt9OoVpL9Fax5n3oMdfSqsOLDTfDwdS69dyDbgNfKWUag6k1P+oolIqTCl1CGgLJH+klrLtwCClVH+ghNY6Oo3DVQe+A9BahwKFlFLxNwYtdRjPSq11DEZdu5J4IeUQxgWNJ4GKwFql1K/AEKCozfDzSd5XQCellCvQEpibTJmDwBylVDsgNp2Z/tZax++ZZwO2T/b6GZiltY6/cNEAGGDm3wi4A8XNYb4H0FofNPMkobWeobWuorWuEtCqYwpfN6nHPLy4eCHx6uqlC5E8VtAzSbnTf//BzGljeWfoJPI/UgCAP44eZP+OMHp39OeziYM5dnAPX0waluZp34uXlzeR4eEJv0dGhuPh0A3S08tCRERimajIiIQy8V0mCxYsRK06dTl6xKi6goU8cHV1xcXFBf/mLTLcsu1lsRAZcT4xQ0QEHh4OOS2WhBYs47sYZTy9LHh6WahgPhioTr0G/H78WMJ3q+lTz+jeWvEZlHLhypXLOS6nbX36BQZnuD5teXhaiLLJcyEqkkIeae8KGxsbw+hBvfFp0JjqtetlWi7Imvo8d/YM5/85R8fWzQluWp+oyAg6tw3m4oX7vy3Ey+JNhO16FJF0PfKyWBzKRODp6cWBX/exedMGAnzrMWRAH/bs3snwQf2STKOhbxM2rF973xmTzRAZkeSheF5e3oTb1GdERDienvbbq/yPPMILL/6P7dvsu19mFi+L/TYnvq7sy3gTHp58ztiYGPr17kWjJs3wqdfAbrjY2Fg2rF9H/Ya+Gc7p6WW/7F2ITGb5dCgTFRlBIfO7NPZrzoxvFzBt+mzyP/IoRYuVSBimRm1ju/RUhUooF8XVDG6XsqI+ly1dgk+9+gDUa9CII4czb7uU05ZVS5L6SVqHFu9k6tDLK9Vhf1n6E3XNOm3Q0JfDh4x96KFDB/hwymR8G/gw5/vZfP3ldObN/T5D38EZPFLQk2s2txBeu3SB/I95pFi+5FPPcjniH26avdIeKWiUzffoY5R/sTrnThzPkpwFPS1ctGk1vxQVQcFCKed0VNc3gPe/mMuoqV+RL/8jFC5SLCti2ilQyJMrNnV79WIUjxZMOXOZCpW5GP4PN65dyfJsImdK9QTePPl6AeOkbLxSyvbs5I75f5zNz/G/u5H6zRnnMU6un0uljO04lcP/96S1jgX+ByzGuJ99VQpFvwHeNluJR2KcOCYnlsT6SiijtZ6L0d0+GlitlPJJY8Tkvkv8ZUPHp6HdMacVB8ToxEvGtnV9xOwtUFlrXUlrbXtklNLT1RYDvhi9HPZqrZN7Gk8T4FOM5WCvSryPPi2ZUrMV8FWJl28VEGTzHYprrY+Zn2WkeSNVpZ94moh/zhAVfo7YmBh2bF7Dcy/bP5TmQmQ4H43pz5vvjaRw0RIJ74d0eotp3y1jyjc/073/WJ56pgpd+45ynMR9e6pCRc6cOcU/584SE3OXdatXUqNWHbsyNWr5sHLZz2itOXzwAHnz5cfD05Po6FvcvGnM9ujoW+zcsY3S5r16tvfIbwxdl/D+/Sr/dEXOnDmdmHPNCqo55Kxesw6rli81ch46QL58+fDw9KSQhydeFm9On/wbgD27diQ8ZKpm7brs270TgNOnThIbG0OBAvf/NPKsymlbn5s3ZLw+bT35VAXOnT1N+D9niYmJYeO6Vbycxr8YoLVmyrgRFCtZmqDWmf/wo6yozzLlnmDZujAWLVvLomVr8fSyMHPOIgp5JL2ollZPVajImdOJ69Ha1Supmcp6dOjgAfKZ69FbPXuzbM0Glqxcx5gJH1DlxZcYOc7osXL61MmE4cM2baBEqdJkxNMVKnHm9CnOnTVyrlm1IknOWrXrsOKX+Jy/mjm9uHzpEtevXQPg9u3b7NqxPVPve0+S85RDztqp5DzwK/nyGzm11owaPoRSpUrT7tWOSca9a8d2SpYqZdc9+36Vf6oi586c4ry57oSuXckrNWvblalaow5rVxrL59FDB8ibL1/Cshb/cLqI8POEbVyHTwPjokK1Wj7s32Nsl86cPklsTAyPZmC7lFX16enpxd49uwHYvXMHxYqXILPktGW1QsVKnD59knNnzxATc5fVK5dTq479oVit2j4sW2o8WfzgASOPp6dXqsN6enqxZ/cuAHbt3JFwW8esb+eyck0oK9eE0rZdB7q8/iat2tg/Mfzf6PEy5bkYfo7LkeeJjY3h8LZQnnzBviPsxfBzCb0a/vn7d6yxMTyc/xHu3o7mjnmb1N3b0fx5cA9exbJmG1X2yac5f+4MEefPERMTw9aNa6hSNe1/aefq5UsAREWcZ+eWUKr5NMqSnLaKlS1P1PmzXIz4h9iYGPZvWU+FF6vblYk6fzahbs/8+RvW2Bjy5s/aBwKKzKGUaqSU+k0pdUIpleTeB6VUW/NW4YPKuH38nvcVpfpQM6XU48AlrfX3SqkbQMd05N0MvKmUmg0UxGhJ7QuUB64AXYA1SqmbWuuNaRznFqCDOU5PoDbJtxqjlMoHPKy1XqGU2gGcMD+6DuS3KZofOG92Q28LnEuh3EmMk9hdGLcLxE+nNPCX1voj8+dnAPunuSRvszm90cp4Mv8FrfW19HaPNf0GeCqlXtFabze/yxNa6yOpDaS1vq2UWg18jjE/7Cjjps9iWusNSqktQBuMru9pcRwoqZQqq7U+AbTH6IURbxgwFPgMo5v8aqCHUqqH1lorpZ7TWu8nsZ42KKUqYtRvpnF1dePVbn15f0hPdFwcNRs0o2iJMoQuXwyAT5Mgfp77FTeuX2X2ZxMBcHFxZdRHmfM039S4ubnRp/9g3nnrdeLi4mjqF0jpMuX4cZHxFO/mwa2oWr0m27ZspoV/Ix5yd2fICOPOiUsXLzKgj/GEbKs1lgaNmvBKNePCxKfTJvP778dRKAo/XoT+g0dkOGfvfoPp/fYbxFnjaOIfSOkyZVmyyOj4ERDckleq12T71s209PfF3d2dQSPGJAz/br9BjBzSn9iYGB4vUpSB5mdN/AMZP3Io7UP8yeWWi8Ejxqa7+/iDyPn5Rx/wx2/HUUrh/fjj9B004r4zOnJ1c+Ot3gMZ9G434qxxNGgaQMnSZVn2k/FMg6aBIVy6eIEenVtz6+ZNlIsLS+Z/z4y5P/H3id9Zv2oZpcqUo1uHEAA6vdmD/1XNnKcmZ1V9ZjY3NzfeGzCYnt2M9aiZfyCly5bjx4XmetSiFdVqGOtRULNGuLu7M3Tk2HuMFT79aCqnT/6Ni4sL3oUfp//gjD3l3c3Njb4Dh9Cj22tY4+LwC2hOmbLlWLTAyBkc0opqNWqxdctmApo2xN3dneGjjD8pdeFCFMOHDCQuzkpcXBz1GzRKuNi3Yf1aJk0Yy+XLl3jn7a488WR5Pvni/rteurm50XeQmdOaSs6wzQQ0MXOONnIe2L+PFcuWUrbcE7RpEQhA957vUL2GcXC9ZtUKGmRC93kw1p0e7w2if8+uWOOs+DYLpFTpsiz90Vh3/JqH8FK1Guzctpl2QY0T/oxcvBEDenPt6hVc3dzo1Xcw+c0nZ/s2C2TSmKF0bh2IW65c9B+e8e1SVtTnkOGjmDxxHFarldy5H2Lw8My7uJxVy2pG8gwYNIxub75GnNWKf2AQZcuWY+H8HwBo0bI1NWrWYkvYJpr51sc9Tx5GmnWY0rAAw0aO5v0J47DGxpL7oYcYmol1mB6zx3ekxgvl8CiQjxOrRjP6ixXMXrL9gedwdXWlcacefDeuPzrOynN1fPEqVorda40OuS/W9+PYzs0cCFuDi6sbuXI/RHCvYSiluHH1MvM/MNoA4+KsVKpWl3KV/5dFOd3o0qMfYwe8TVyclTqN/ClWsgxrflkEQINmwVy+dIEB3dsTfesmSimW//gDU79eyMN58zF5ZF+uX7uKm5sbr/UYQL4H8MBaV1c3gl57l+mj+hAXF8dLdZtQuHgptq5eAkC1hgEc3L6J3ZtW4WrW7at9RmZo2+NsnPWbmr2cPwXqY9xmvVsptVTb/5Wzv4FaWuvLSilfjF7qL6U63tTu3VJKNQQmYbSqxgDdtNZ7lM2fAVNKdTR/ftsc5iTGveAXgfcxWng1xj3o882T1fe01k2VUsWBlUBnjG71VbTWbyulvgGWaa0XmeO8obXOZ55QfvZ/9u47PIpqjeP49yWhNylJEAGpilQLFqQj0lvoAiqKDcUOCIiCdAR7BRQ7othFmtKbIBaaWFCpQhKaFCkp7/1jJmE32YRAEnbW+37ucx/Z3TO7v5ydnd0zpwzOyYDfgLzA06qaZuyiO9/+c5zecgEmqepbIlIPmIrTg9wFZ+j2IGAbzkiDwqraJ0C53MCHwBGcBnpvVS0vIkOA3m797AF6qur+dOrTt96KA28AFXCG99+hqusl1WXeAtw+oqqFUj/mzmV/HiiKc2LmWVWdKqkuuxbg+a7B6Ykvp6qJqfLmxpmmUNStw3dVdfwZZLoOmOTm+Q5n/zmRah+ZBsQBw4FngWvd19rq7iP53XqqBvwEVAbu0wwuI7f6j39yrMc+O1UplX3zkXNSYpbmd5rUjp5IOH0hDyiYNzQuWpE7LDS+1sNC5YdWiMQ8dCw0PkdF8ofG5yjnxrllr/Awz1w8KUPFrwqNBe6mTRsS7AiZUr1kaPQ07zwU/KsFZFbr6pEhcrR3rNt+2JNHqdrlCmdYjyJSFxihqi3c20MAVHVcOuWLARtV9YIMnzdri6+ceyJSSFWPiEgJnN7weu58eHMWRGQAUFRVHwt2luxiDfjsZQ347GUN+OxlDfhsFiIxrQGfzULkMG8N+OxlDfjsZQ34nOPVBvylFxa5E7jD564pqjol+YY4V0trqaq3ubdvBK5O7vhOzW2XVU0un54Q+WbxM8tdJC8PMMoa72dPRD7FWU0+s/P2jTHGGGOMMebc8ejpBrexPiWDIhmteeZfUKQJzpTm+oEe9xVyDXhVbZz6Prchmno1jEdUdd45CeWfpQSwIMBD16WzSFzQqGp0sDMYY4wxxhhjzH/QTsD3UgZlcC4r7kdEauFcHaxVZtqLIdeAD8RLDVG30i8Ndg5jjDHGGGOMMUHzHVBFRCrgLJTeA2dR8BTumnCfADf6XH47Q/+JBrwxxhhjjDHGmP8e8eoY+tNQ1QQR6Y9zta0wYJqqbhKRu9zHX8W5MlcJ4GX3ygIJqlono+e1BrwxxhhjjDHGGJPNVHU2MDvVfa/6/Ps2IMNF61ILjeU8jTHGGGOMMcaY/3PWA2+MMcYYY4wxxpNC5Uqs54r1wBtjjDHGGGOMMSHAeuCNMcYYY4wxxniSdcD7sx54Y4wxxhhjjDEmBFgD3hhjjDHGGGOMCQE2hN4YY4wxxhhjjDfZGHo/1gNvjDHGGGOMMcaEAGvAG2OMMcYYY4wxIcCG0BtjjDHGGGOM8SSxMfR+rAfeGGOMMcYYY4wJAdYDb/5zos7LG+wImZKoGuwImZI/d1iwI2SKEhr1eexkaJxFLpwvNL4e9h4+GewImVK8UO5gR8iUfSFSn+cVCI36TEgMjeNSLgmN41JCUlKwI2TKtGlDgh0hU269dVywI2TK9LeGBTtCpny9ZX+wI2Ra6+qRwY5gsiA0fqEZY4wxxhhjjPm/EyLnGM8ZG0JvjDHGGGOMMcaEAGvAG2OMMcYYY4wxIcCG0BtjjDHGGGOM8SQbQe/PeuCNMcYYY4wxxpgQYA14Y4wxxhhjjDEmBNgQemOMMcYYY4wx3mRj6P1YD7wxxhhjjDHGGBMCrAFvjDHGGGOMMcaEABtCb4wxxhhjjDHGk8TG0PuxHnhjjDHGGGOMMSYEWAPeGGOMMcYYY4wJATaE3hhjjDHGGGOMJ4mNoPdjPfDGGGOMMcYYY0wIsAa8McYYY4wxxhgTAmwIvTHGGGOMMcYYT7IR9P6sB94YY4wxxhhjjAkB1oA3//e++3YFfXu0p0/Xtnzw9utpHt++9S8euP1G2jaqw8zpb6XcHxuzh4H9+3LbDR25vVc0n37wXrZn+3blMm7o1IbuHVryzhtT0zyuqjz75Fi6d2jJzd2j+XXzzymPHT58iGGDHqBnp7b06tyOjet/AuD1yS/RsWUT+tzQiT43dGLV8qVZzrly+TI6tWtJhzbNeeP1KQFzPjl+NB3aNKd75/Zs/nkTAHv27OaOvjfRuUNruka3Zfq7b6ds8/X8uXSNbkud2pfw86YNWc54KmcrOrZpwZuvB67PiePH0LFNC3p07sAvPjnv7HszXTq0oVt0W973yTn55Rdp1awRPbtG07NrNMuXLcmWrMm+W7WcW7q34+YubZiRzv553+29ad3wCma+96bfY5NGP07X1o24vVd0tmRZsWwp7du0oG3L63l9auD3efzY0bRteT1dotulvM+Z2fatN16ndvWLOXBgf8p9v/36Czf27E50+zZ07tiOEydOZMvfkWzNquXc3K0dN3Zpzftvv5bm8e1b/6T/bb1o2eByPkxVtzlhxfJldGzbkvatmjPttcD1O2HsaNq3ak636PZ+9Tti2FCaNryWLh3b+W3z9by5dO7QlstrXsKmjdnzOfK1ZtVy+nRvx01d2vB+Ovvnvbf3plXDK9LU4cTRj9OldSNuy6b9MyMrVyyjU/tWdGx7ms9+2xb06NKBXzY7dXvixAlu6tmNG7p2pFt0Wya//EK2Z1u1YhndOramS/sWvD0tcLanJoyhS/sW9OrWkV98jvMAiYmJ3NSjEw/f1y/lvqmvvki75o25sXs0N3aPZmU2HJdWrVhGlw6t6NSuBW+lk3PShDF0ateCnl3967BPr2707NaR7p3aMsWnDocOepBe3aLp1S2aDq2uo1e3rO8LoXic//2nNbzw4E08d39vln0+Pc3jv6xdwcuDbuOVR25n8tC72PbLqc/yM/1v4OWBfVMeC5ZXh/di24JxrJ05NGgZkv3642om3debif17svjTtL/NNn23nGcfvoXnBvTlhUfuYOvm9QDEnzzBi4Pv5NkBt/L0gzfz9QfTcjRntaiCDL++EiOaV6b5RSXSPF6lZAGeancxQ5pWZEjTirSqWjLlsd6Xn8+E1hcx7LqKOZrReJMNoTf/1xITE3lp0ljGPTeZkpFR3Nu3J9c0aMyFFSqllClSpAj9HnyElUsX+W0bFhbGHfcOoMrFl/Dv0aP0v7UHl191jd+2Wc329PgxPPPyVCKjorjtxu7Ub9SEChUrp5T5dsUyduzYxozP5rBp43omjRvJ1LdnAPDcxHFcXbc+o598lvj4kxw/fjxlu249b6LnTbdkW87xY0fy8pRpREVFceMNXWnUuCkVK53KuWL5UnZs28Zns+axcf06xo1+grenf0hYWBgPPvwIl1SrztGjR+jdozPX1L2WipUqU7lyFSY+/TxjRw3PtpwTxo7ipSmvExUVxU03dKNh4yYBc346a66bcyRvTf+A8LAwHnx4EFWrVefo0aPc2KMzV7s5AXr2vpkb+9yaLTlTZ37hqbFMeG4KJSOj6H/rDdRNtX8WLlKEex4czIqlC9Ns37xNezp07cGTIx/Nlixjx4xk8tQ3iIqKomf3LjRu0pRKlU/V3/JlS9m+bStfzpnPhvXrGD1yBO/NmHnabffs3s2qlSs5//zSKc+VkJDA0MEDGTNuIhdXrcrBgwcID8++r6zExESenzSGJ5+fQkRkKe6+pQd1GzShvF/dFqX/Q0NYsSRt3Wa3xMRExo8eyStTpxFVKope3bvSqElTKlVKVb/bt/H57HlsWL+OsaOe4J33PwSgXcdouvfsxWNDB/s9b6XKVXjq2ecZ/UT2fI5SZ07ePyMio7jn1hu4Np39c2WA/bNFm/Z07NqDCdmwf54u54Sxo3hpsvvZ75nOZ3/7Nj79ci4bN7if/fc+IE+ePLz62hsUKFCQhPh4+vbpzbX1G1Cz1qXZlm3S+NE8/8prREZFcUuv7jRo1IQKPtlWudlmfj6XTRvW8+TYJ5j2zgcpj38w/R3KV6jE0aNH/J67R++b6HVT9hyXEhMTeXLcKF589XUio6K4uVc3GjTyr8OVbs6Pv3DqcMKYkbzxrlOHL089VYe339Kbum4djn3ymZTtn31qAoUKFcpyzlA7ziclJTJ72nPc+OhEipSIYOrQflx8xbVElimfUqZCjcvpd8W1iAh7tv3BzOdGcu/TpzoUbn7saQoWKZrt2c7EO19+y6sfLOG1UTcFNUdSYiKfv/4sfR97iqLFI3hxyJ1cUqceUWXLp5SpXONyqk2qh4iwe9sfTH96BA8/9w7hufNw+/BnyJu/AIkJCbz6WH8uvuxqyl1UPdtzCtC99vk8v3wbB4/F80iTiqzffZg9h0/6lduy919eWbUjzfbfbvuHJX8e4OYrSqd57D/JxtD7CZkeeBFpLCKzAtzfXkQGB9omm1+/o4hUy65y/89EZISIDDjLbfuIyIvZleXXnzdSukxZzr+gDLlz56Zxs5asWrbYr8x5xUtwcbUaaRoPJUpGUOXiSwAoULAgZS+syN642OyKxuZNGyhTtiwXlClL7tx5aNa8NcsX+59EWLZkIS3btEdEqFGzNkeOHGZvXBxHjxxh3Y/f07ZjZwBy585D4cJFsi2br00b11O2XDnKuDmbt2zN4kUL/MosWbSANu06ICLUrH0pRw4fIi4uloiISC6p5nwxFixYiAoVKhEbGwNAhYqVKF8h+84sB8q5ZJF/o2LJooW09sl5+PAh9sbFUjIikqopOQtS3idnTnL2z3J++2fqE0nF0tk/AWpdVofC2fSjbuOG9ZQteyFlypYld548tGzdJs37vGjhAtq174iIUMutv7i42NNuO3HCOB58eCDic52YVStXUOWii7m4alUAzjuvGGFhYdnytwD88vMGLihTjtIXlCV37tw0ub5VwLqtmk7dZreNG9z9s6yzf7Zo1ZrFC9N+jtq275CmfgGuqHMlRYumfa8rVsrez5Gv5P2ztM/+uSKdOgzL4f0zI5s2rqds2VSf/cUZfPZrnfrsiwgFChQEnJNKCQnxSDb+kvx54wbKlC2Xcpy/vkUrlqbKtnTJQlq3dbLVqFWbI4ed4zw4I8FWLl9C++jO2ZYpkE0b1/vlbN6iddqci0/lPG0dpromlKryzfy5NG/ZJss5Q+04v2vLLxQvdQHFo0oTHp6bGtc25de1K/3K5M2XP6XO4k8cz9Z9MLus+OEP9v/zb7BjsGPLZkqUuoASUaUJz52b2vWa8vPa5X5l8uYvkFKfJ48fS2kcigh58xcAIDExgcTEhBy7fln54vmJO3qSff/Gk6jw/c5/qH1+4Uxvv2Xfvxw9mZgj2Yz3hUwDPj2q+oWqjj8HL9URyEzDPLPlPElE/q9GZeyLiyUiqlTK7ZIRkeyNO/Mv7D27d/HH779QtXrNbMsWFxtDZNT5KbcjoqKIS5Vtb2wskT75IyOj2BsXw9+7dnBesWKMHfEot/TszPiRj3Ps2Kkv1k8+nM7N3aMZ+8QwDh36J0s5Y2NiiPLJGRVVirhUP3piY2OIKnWqTGSAMn/v2skvv2ymRs3aWcqTfs5YonzrKioqzY+zuNgYSpU6VSYqqhSxsf4nZf7etYtfU+X8cMZ79OjcgScefzTL9elrb1wMEZFRKbdLRkZl60miMxEbE0Op8/3rLyYm0Pucqv5iYjLcdvHCBURGRaY01JNt2/oXIsJdt/ele5do3ggwFDYr9sbFEhF5KlOE+9kJltSfkYCfo5gYSqUqExsTvMx742KI9Nk/IyKj2Bek/TMjsbGxfvtlZGRUmnqLi42hVFTgz35iYiI9u0VzfZP6XH3NtdSolX3HKOc47/vZKJVyUuZUmVgiS/l/fpL3jWcmjqf//QOQXGl/zs2cMZ1e3ToyekTWj0txqevQJ0Oy1J//yFR12KtbNC2a1ueqa65Nc5z/8Ye1FC9RgnIXls9SzlA8zh/av5ciJSJTbhcpXpJD++PSlNu8ZhkvPHQz700YSoe7BqbcLyK8M3Ygk4fcydpv0vRz/d85tH8vRX3qs2jxCA7t25um3MbVS3nq/ht5c9xguvR7JOX+pMREnhvQl9F9O1KlVh3KVcmZn/Tn5QvnwLH4lNsHjiVQNH/uNOUqFM/P0KYVuefacpxfOG+OZDGhJ0ca8CJSUES+EpF1IrJRRLq7928VkbEiskpE1orI5SIyT0T+EJG73DIiIhPd7TYkb5vq+a8UkR9FpKJvj6yIvCkiz4vIShH5U0S6uPfnEpGXRWSTiMwSkdnJj6WTf7yI/Cwi60VkkohcC7QHJorITyJSSURuF5Hv3L/xYxEpkE65xSJSx33ekiKy1f13dRFZ45ZbLyJV0slSXkQ2i8hUN/98EcnvPlZJROaKyPciskxEqopIUbeec7llCojIDhHJHai8T709LSKLgAmpXj9NTjfTLyLymvs+vScizURkhYj8LiJXudsWF5HP3O2+FZFaAf6+20VkjojkF5HePq81WUTC3DK3iMhvIrIEqJfe+3Y2FA1U52f0HMf+/ZdRQx/mrvsHUrBg1ob/+dK00QL2WgQoRGJiIr/9spmOXXrwxvSPyZc/P+++4czxje7SnQ8+n8sb739MiZIRvPjMxKzlDHBf2pwZl/n336MMfOg+BgwakuUhlOk7/XsdqD5T5xz00H08PGhwSs4u3Xvw2VfzmT7zU0qWjOCZSU9mX+JM7APnSqY+K+nUX3rbHjt2jKlTXuXu/veneTwxMZEff/iecU9O5M13prNwwTes/nbV2f8BqQXKGsxercAfJP8iHtofIHCenOqtypLTfK4h4/07LCyM6R9+yuz5i9i0cQNbfv8t+6IFfONTlUkn//KliylWvHhKr7GvTl178PGX83hnxieUKBnB809n7biU3ndNqkIBc4JTh+99+Cmz5i3i540b+GOLfx3On/sVLbLY++6GSDfDqZjeOs5nJjPAJVc14N6n36LHgJEs/PCNlPtvfeJ57ho/hV6Dx/Pd/M/YunldNmYLPQE/UwEOSzWubsjDz73DjYPG+M11zxUWxv2TXmfI5Jns2LKZPdv/zJmgmThU7jh4nMfm/s7YhX+y+I/93Fm3TM5kCQHi0f8FS071wLcE/lbV2qpaA5jr89gOVa0LLAPeBLoA1wAj3cc7AZcCtYFmOI3hlC4Ht5H8KtBBVQN9qs4H6gNtgeSe+U5AeaAmcBtQN73gIlIciAaqq2otYLSqrgS+AAaq6qWq+gfwiapeqaq1gc1A33TKpecu4DlVvRSoA+zMoGwV4CVVrQ4cBJLHyk0B7lXVK4ABwMuq+g+wDmjklmkHzFPV+EDlfV7jIqCZqj6cyZyVgeeAWkBVoCdOvQ8AklcweQL40a3HocCpFWEAEenv5uuI8/50B+q5r5UI9HLf+ydwGu7Xk87oBhG5Q5yTQmunv5V2IaX0lIyIIi5mT8rtvXGxlCgZmcEW/hIS4hk19CGaNm9N/cbNMr1dZkRGRREbszvldlxMDCVTZYuIiiLWJ39srFMmIjKKiMgoqtd0zpk0adac337ZDEDxEiUJCwsjV65ctI/uwuYsLhAXFRVFjE/OmJg9lIyITFtmz6kysT5l4uPjGfjQfbRq046mzZpnKUtGnF5fn7qKiSEiVc7IqFLs2XOqTEzMHiIiIgBIiI9n0EP30zJVzhI+9RnduSubNqzPtswRkf69XHtjYyhRMiLbnv9MREWVYs9u//qLjExbfzGp6y8yMt1td+7Yzq5dO+nWqQOtrm9KTMweenTpxN64OCKjSlGnzlUUK1ac/PnzU79BQ79F27KqZGQUcbGnMsXFxlAiIvOf/ewWmeoz4ux7qT5HpaLYk7pMZPAyR0T6927GBXH/zIhTt/7HydT1FhlZij0xgT/7yQoXKcIVV17FqpX+w3GzlC2ylP8xPMD7HhkVRewe/89PyYhI1v/0A8uWLKJj62Y8Nvhh1n63muGPDgL8j0sdOnXl5ywuYJimDtM5fvqXCVyHl9e5ilUrTtVhQkICixd8Q7MWrbKUMSVniB3nixSP4NC+UyMADu3fS+FiJdMtX/6S2hyI+Zuj7iiAIsWdsoWKFqPqlfXZteWXbMsWiooWj+Afn/r8Z39cSh0FUrFabfbt2cXRQwf97s9fsDAVq1/Gbz+tyZGcB48lUMynx71Y/nD+8emRBziekMSJROeExKaYI4SJUDBP9k0lM6ErpxrwG4BmIjJBRBq4jcpkX/iUWa2qh1U1DjguIufhNALfV9VEVY0BlgBXuttcgtMIbaeq29N57c9UNUlVfwaSx/bVB2a69+8BFqWzLcAh4Djwmoh0AtKb0FPD7cXeAPQCznSFi1XAUBF5BLhQVY9lUPYvVf3J/ff3QHkRKQRcC8wUkZ+AyTgnLwA+wGkMA/QAPjhNeXDqJ9BkmvRy/qWqG1Q1CdgELFDntPYGnMY4OPX+DoCqLgRKiEjyhMcbgVZAZ1U9AVwHXAF85+a7DqgIXA0sVtU4VT3p/m1pqOoUVa2jqnV63tw3UJGALr6kOrt2bmfP3zuJj49n8TdzuaZ+o9Nv6LwmT48dQdnyFel8Q/Yv2lK1Wg127NjO37t2Eh9/km/mz6ZeoyZ+Zeo3bMLcr75AVdm4YR2FChWiZEQEJUpGEBlViu1b/wJg7ZpvKV/RWVgqee4kwNJF31CxUsDBH5lWrXpNdmzbxq6dTs75c2fTqHFTvzINGzflqy8/R1XZsO4nChUuTEREJKrKqOHDqFChEr2zaVG9M8nZsLF/fTZq3ITZqXKWdHOOHD6MChUq0vumPn7b+A5pX7TwaypVyVp9+rr4kurs2rGN3T77Z90GjbPt+c9E9Ro12b59Kzt37iD+5Enmzv6KRk383+fGTZry5RefoaqsX/cThQo573N621a56GIWL1vFnK8XMufrhURFlWLGR59QMiKCevXq89tvv3Ls2DESEhL4fu13fgtRZVXVS2r41e2ir+dwbZDqFpLr99T+OW/ObBqnqt9GjZsy64vP09RvsATaP4NZh+mpVr0mO7an+uw3yuCzv96p25IRkRzYv5/Dhw4BcPz4cdZ8u4ry5StkW7ZLqtdgx/ZtKcf5r+fNoUGq41KDRk2ZPcvJtnH9OjdbBHff9xBfzlvEZ7O/YdT4p6hz5dU8McbpGfY9zi9ZmE3H+e3b2OXmnD9vNg0apc7ZJCVnhnW4ehUXVjhVh9+5t32HvmcpZ4gd50tXqsq+Pbs4ELubhIR4Nq5cyMVX+Pcz7duzK2XkwN9//UZiQjwFChfh5PFjnHCnx508fow/1q8lsmz27Z+hqEzlquzbvZP9MbtJiI9n3YqFVKvjP3hz7+6dKfW568/fSExIoEDhohz55yDHjh4GIP7ECbasX0vEBeVyJOe2A8eILJSHEgVyEyZwRZmirN/tvxBlkbynGusXFsuHiNi8dwPk0Cr0qvqbiFwBtAbGich8VU3uYU++DlCSz7+Tb4eT8aCS3UA+4DLg73TK+D6npPpvZrInuEPAr8Np/PYHmgYo+ibQUVXXiUgfoHE6T5nAqRMl+XxeZ7qIrAbaAPNE5Da3kRuI79+UCOR3n/Og21ud2hc49V4cp1G8ECiYQXmAo4HuDJQT+JO0753v+5q8XwWq9+SxTRtxRlqUAf5yy76lqkN8C4tIRwIPLs0WYeHh3PPQEIY+2I+kxCSat+1I+YqVmfWps7Jz2+hu7N+3l3tvvYF/jx5FcuXisw/eZcr0T/lry28smDuLCpWq0O/mbgDccue9XHVtg2zJFh4ezkODHuWh/neQlJhEmw7RVKxUmc8+cs5hdOzSnbr1G7JqxVK6d2hFvnz5GDpidMr2Dw4ayhPDHiEhPp7SF5RhiPvYK88/xe+//oKIUKp0aQYOHZHlnIOGPkb/fn1JTEyiQ8fOVKpchY8+dFbD79KtB/UbNGLFsqV0aNOcfPnyMWLUWAB++vEHvpr1OZWrXMQNXTsCcM99D1K/QSMWLviaieNGc+DAfu6/5y4uqlqVl17N/OiKQDkHDh3Gvf1uIzExifYdO6XJWc/N2bFNC/Lly8dwN+e6H39g9qwvqFzlInp2dS5zdPd9D1C/QSOee2YSv/3i1Of5pS/g0cdHnHXG1MLCw+n/8FCGPNCPpKREWrj755efuCuPd3L2z3tu6ZGyf37ywbu89v5nFCxYiDGPD2L9D2v55+BBbmjfjJtuu5tW7TudVZbw8HCGPPo4/e64jaSkRDpGd6Zy5Sp8+MH7AHTrfgMNGjZi+dIltG11Pfny5Wfk6LEZbpuRIkWLcuPNfejZvQsiQoMGDWnYqPFZZQ8kLDycewcM5ZH77yIpKZFWbaMD1m2/Pt1T6vbjGe8wbcbn2TpVJll4eDiPDH2Mu+/sS1JiEh2inc/RzA+c/bNr9x7Ub9iI5cuW0r5Vc/LlP/U5Ahg88CG+/+47Dh48QIvrGnHX3fcS3bkLC7/5mgnjRnNg/37uu/suLq5alZennP3nyFdYeDj3PjyUwe7+2TKd/fPuVPvn6z775zp3/+zRvhk3Z2H/zEh4eDgDh7if/aQMPvvLl9KxrfvZH+nU7d69cQwfNoSkpESSkpK4vnnLNA3XrGYb8Mij3H/37SQlJdG2QzQVK1Xhk5lOtk5de3Bt/YasXL6ULu1bki9fPoaNGHPa533xuUn8/usvIML551/A4GEjspxz4OBh3NfvNpKSkmjXwanDj92cnbs6dbhy+VI6tXPq8LEnTtXhE4+dqsNmzVvSoOGpOpw/d3aWF6/zyxlqx/mwMFrfci/vjH0ETUrksiatiCxbge++dvq7rry+PZtXL2XdsvnkCgsnd568dLn/cUSEI/8c4IOnHgec1exr1ruOKpdelW3ZzsRb4/rQ4IoqlDyvEFvmjmLUq7N567NsnPaUSWFh4bTv+wDTxgwgKSmJOk1aE1W2At/O/xyAa5p3YOPqpfywZB5hYeHkzpOHng8OR0Q4fHAfH744Fk1KQlWpWbcxl1xxbY7kTFL44Kc99K9XjlwirNp2kN2HT9CgQjEAlv11gMsuKEKDisVISoL4pCSmrTk1WPeWKy/googCFMoTzphWVfjq5zhWbjuYI1m9wIuzs4JJAs5ryuqTipQG9qvqcbcB1kdVO7rzv+uo6l630VtHVfu722zFGaLdELgTp/FfHFiL0wtbFWd4dl9gPnC/qi72fR4ReROYpaofuc95RFULiUhX4Gac+ekROEPe70gulyp7IaCAqsa6DeAtqlpcRF4AflDVN9xye3GGcx8AZgO7VLVPgHKvAd+r6isi8gDwgKqWF5GKOL3YKiLPAltV9dkAecq7f1MN9/YAoJCqjhCRlcAzqjpTnAlTtVR1nVtuJs5IgsOqerd7X8Dyqest1eunyQl8lipTyva+eUXkeSBOVUeJSGP3tS8TkRHAEeBb4BWgBXAe8DnOEPrkui8MnHTLXY4zOmIhsC55vwlk677jOdbgz04F84bGeoH5c4fGcK3MzCX1ggNH409fyAMii4TGYjl7U11yx6uKF0q7OJEX7QuR+jyvQGjUZ0JSaByXcoXIr+MAa/V50qyfd5++kAfceuu4YEfIlOlvDQt2hEz5esv+YEfItJc7VQuND71rS+wxTx5MK0fmD0o95tShsCawxh0K/SgwOuPifj4F1uPM414IDHKHvQPgDqtvB7wkIldn8jk/xpm7vRFn6PhqIL0lRAsDs0RkPc7w/Qfd+2cAA8VZPK8S8Jj7PF8DvhOOUpebBPRzG8++k3C6AxvdOqpKqvnhmdQL6Csi63CGsXfweewDoDf+Q84zKp+erOQcAdRx63I8zkmUFKq6HOekzFdALDAMmO+W/xo4X1V3u8+zCvgG+OEMXt8YY4wxxhhj/jNypAfei0SkkKoeEZESwBqcnt49p9vOhB7rgc9e1gOfvawHPntZD3z2sh747GU98NnLeuCzl/XAZy/rgc85f3i0B75SkHrgQ6MFkT1muYvk5QFGWePdGGOMMcYYY0wo+b9pwKtq49T3icinQOrlOh9R1XnnJJR/lhLAggAPXaeq+851HmOMMcYYY4wx3vJ/04APRFWjg50hmdtIvzTYOYwxxhhjjDHGM0JqwH/OC5HZRMYYY4wxxhhjzP83a8AbY4wxxhhjjDEh4P96CL0xxhhjjDHGGO8SG0Pvx3rgjTHGGGOMMcaYEGANeGOMMcYYY4wxJgTYEHpjjDHGGGOMMZ4kNoLej/XAG2OMMcYYY4wxIcAa8MYYY4wxxhhjTAiwIfTGGGOMMcYYYzzJRtD7sx54Y4wxxhhjjDEmBFgD3hhjjDHGGGOMCQE2hN4YY4wxxhhjjDfZGHo/1gNvjDHGGGOMMcaEAOuBN/85BfKEBTtCpuQND43zZwlJScGOkCl5w0Pjfc8VImeRVYOdIHOKF8od7AiZIiHSfSAhcrHd8LDQOH4qoXH8DI13PXRUL1k02BEyZfpbw4IdIVN63jw62BEy5av3nwh2BPN/whrwxhhjjDHGGGM8KVROgp8roXEK2xhjjDHGGGOM+T9nDXhjjDHGGGOMMSYE2BB6Y4wxxhhjjDGeFCLLs5wz1gNvjDHGGGOMMcaEAGvAG2OMMcYYY4wxIcCG0BtjjDHGGGOM8SQbQe/PeuCNMcYYY4wxxpgQYA14Y4wxxhhjjDEmBNgQemOMMcYYY4wxnmSr0PuzHnhjjDHGGGOMMSYEWAPeGGOMMcYYY4wJATaE3hhjjDHGGGOMR9kYel/WA2+MMcYYY4wxxoQAa8AbY4wxxhhjjDEhwIbQG2OMMcYYY4zxJFuF3p/1wBtjjDHGGGOMMSHAGvDm/9Lqlcvp2aktPTq24t03X0vzuKry7MSx9OjYipt7RPPrLz8DsH3rX9zSs3PK/1s0upoPp78DwEvPTaJX53bc3COaoQPu4/DhQ1nOuXLFMjq3b0V02xa8+frUgDknjR9DdNsW3NClA79s3gTAiRMnuLlnN3p27Ui36LZMfvmFlG1eefE5bujSgZ7doul/Z1/iYmOznHPVimV07dCazu1a8Na0wDmfmjCGzu1a0KtrR37Z/LPf44mJidzYvRMP3dsv5b5XX3qeXl070rtbNPfeddtZ5VyxfCkd2ragXavrmfbalIC5JowdTbtW19M1uh2bf9502m0HPfwA3Tp3oFvnDrRq3pRunTsAsGHD+pT7u3Vqz8Jvvj7jvKezZtVybu7Wjhu7tOH9t19P8/j2rX/R/7betGxwBR++92a2v35O1CfA+++9Q4e2LejUoQ3PPPUkAAcPHuC2W26k7pWXMW7MyCzmXkbHti1p36p5hrnbt2pOt+j2frlHDBtK04bX0qVjO79tnpn0JNHtWtEtuj0P3defw4ey/nk/l/WbE07tn615/+20x9XtW/+k/229aNng8hzZP+Hcfubj40/y+LAhdIluR7dO7fluzeqzypwTx/lk77w1jStrX8LBAwfOKlvqnJ3at6JjBjknjh9Dx7Yt6OGTc8+e3dzZ92a6dGxDt+i2vP/e2ynb/PrLZvr07k7PbtHceEMXNm5Yn/Wcy5fRqV0rOrY5Tc42LejRuQO//JwqZwc357upcvbqTs+u0dzYI3ty+vpxzUru69OJ/jd14NP330jz+K7tfzH03j7c0Ooavvjwbb/HvvpkOg/d1o0H+3blq4+nZ2uu1H79cTWT7uvNxP49Wfzpe2ke3/Tdcp59+BaeG9CXFx65g62bnXqKP3mCFwffybMDbuXpB2/m6w+m5WjOjLw6vBfbFoxj7cyhQcuQbNMP3zK8Xw8ev7Mr8z56O83jaxbPY/R9NzL6vhuZOOgOdv71e8pjCz6fwcj+vRh5by9en/Q48SdPnMvoJshsCL35v5OYmMjTE0bzzEtTiYgqxe03dadewyZUqFgppcy3K5axc8d23v90Nj9vXM9T40Yx5a33KVe+Am9M/zjleTq1bkrDJtcBcOXVdbnzngcIDw/nleef5t03XqPffQ9lKeeTY0fx4uTXiYqK4uae3WjYuAkVK1VOKbNy+VK2b9/GJ1/OZeOGdYwfPZI33/uAPHny8Mprb1CgQEES4uO5rU9vrq3fgJq1LuXGPn3p1/9+AGa89w6vTX6ZIY+NyFLOieNG88KrrxEZFUWfXt1p0Chtzh3bt/HRF3PZuGE9T455gmnvfpDy+AfT36F8hUocPXok5b7eN9/KXffcl/L461NeZvCwzOdMTExk3OiRvDr1DaJKRdGrexcaNWlKJZ9cy5ctZfv2rXwxez4b1q9jzKgRvPv+zAy3ffKpZ1O2f2rieAoVKgRA5cpVmP7Bx4SHhxMXF0u3zh1o2LgJ4eHZc5hNTEzk+UljefL5KURERnH3LTdQt0Fjylc4td8WLlKE/g8NZsWShdnymqlfPyfq87s137J40QJmfvIlefLkYf++fQDkzZOXe+69ny2//86WLb+nFytTucePHskrU6e5r901ndzb+Hz2PDasX8fYUU/wzvsfAtCuYzTde/bisaGD/Z73mrrXcu8DDxEeHs5zT09i2mtTuP+hAVnKeS7rN7s5++cYd/8sxd239KBugyap9s+i9H9oSI7sn8kZzuVn/uOPZgLw0adfsn/fPu7pdzvvzfiIXLky3zeSU8d5cBqka1atpNT552ehVk/lnDB2FC+5OW8KkHOFe5z/1M05bvRI3nrvA8LDwnhwwCCqXlKdo0ePcmOPzlx9zbVUrFSZ55+ZxO133UO9+g1ZvmwJzz87iSmvp23InHHOKW7OG9LJuW0bn86ay8b1bs7pbs6HB1G1mk/OuqlyNnBzPjOJKdPOPmfqzK+/MJ7HJrxM8YgohtxzI3WubUTZCyumlClUuCi33jOQNSsX+227/a8tLJj9GeNefIvw3LkZM/heLr+6PueXKZct2XwlJSby+evP0vexpyhaPIIXh9zJJXXqEVW2fEqZyjUup9qkeogIu7f9wfSnR/Dwc+8QnjsPtw9/hrz5C5CYkMCrj/Xn4suuptxF1bM95+m88+W3vPrBEl4bddM5f21fSYmJzJg8ifueeI5iJSIZP6Avta5qwPnlKqSUKRFVmgfHvkTBQkXY+P0q3ntpAo9Meo2D++JYNGsmj784nTx58zL1yWGsXfYNda9rE8S/KGfZCHp/1gOfSSLSWERmBbi/vYgMDrRNNr9+RxGpll3lgim9ujxXNm/awAVly1G6TFly587Ndc1bsTzVD8rlSxbRsnV7RITqNWtz5PBh9u6N8yvz/XffUvqCspQ6vzQAV11TL6WxVr1mLeJiY7KUc9PG9ZQtW44yZcqSO3cerm/ZmiWL/XMuWbSQNu06ICLUrHUphw8fYm9cLCJCgQIFAUhISCAhIR5xD3/JPz4Bjh0/luV5RT9v3ECZsuW4IDlni1YsTZVz6eKFtGqbnLM2hw8fZm+cU58xMXtYsWwJHTp19tvGL+exY8gZBt24YT1ly11ImbJOrhat2rB44QK/MosXLaBt+46ICLVqO/UXFxebqW1Vlflz59CydVsA8ufPn/L+nzxxIqW+s8svP2/kgjLlKH1BGXLnzk2T61uycukivzLFipegarUa2XbSwFdO1eeHH7zPLX3vIE+ePAAUL1ECgPwFCnDZ5XXIkzdvNuQu5/PardPkXrJoAW3bd0iTG+CKOldStGjRNM9bt179lHquWas2MTF7siHnuavf7PbLzxvc/bOsu3+2Oqf7J5z7z/yff2zh6quvAZx6LVy4MJs2bTyjzDl1nAd4ZuJ47n1wwBkfOzOTs3k6OVsHyFkyIpKqlzgNtYIFC1K+YiVi3e9HEeHoEefE7ZEjR4iIiMx6znKpci7KIGftVDmr+eSskCqne4L5yOGs5/S15ddNlCpdlqjSzrG9XuPmrF2x2K9M0WLFqVy1OuFh/p+dXdv/osolNcibLz9hYeFUq305a1b4f+6yy44tmylR6gJKRJUmPHduatdrys9rl/uVyZu/QMr+dvL4sZRWl4iQN38BABITE0hMTAjapOYVP/zB/n/+Dcpr+9r6+89ElCpDRKkLCM+dmzoNmrFuzTK/MpUuqUnBQkUAqHBxdQ7sOzUSMSkxkfiTJ0hMTODkieMULV7ynOY3wWUN+CxS1S9Udfw5eKmOQGYa5pkt938rLjaWyKhSKbcjIqPYm2p4dlxcDJGlfMpERbE3VYN8wbw5NGvROuBrfPXFp1x9bf0s54zyyRAVGUVcTEyqMjFE+fwtkVGliHX/lsTERHp2i6Z5k/pcfc211KhVO6Xcyy88S5vmTZj71Zfcefd9WcoZGxvjlzMyqlSa4e6p/5bIqKiUExzPTBxP/wcGIJL2cPTKC8/SrkVT5s2exR397j3jXKV86y8qKuXHWEqZmNRlShEbE5OpbX/4fi0lSpTgwgvLp9y3Yf06OnVoQ5fo9gx7/IlsbajsjYshIjIq5XZEZBR747I+/SGzcqo+t23dyg/fr6X3DV3p26d3tg9NdfbPUz2QUVGl0pxcc3L7l4mNyfwJuM8//Zh69RtmOWco1m+yvXGxRESmOq7GZe0k5pk615/5iy6uyqJFC0hISGDXzh38/PMmYvbsPqPMOXWcX7J4IRGRUVx0cdUzypOe2NTH8MioNJ+RuNgYSkWlqttU3wV/79rFr79spkZNJ+fDg4bw3DOTaNO8Cc899ST973swazljYlPVVdr3MS7Ne53JnE9Pos31TXju6Sfpf3/WcvravzeWEj7H9uIRUezbF5fBFqeULV+Zzet/5PA/Bzlx/Bg/rF6R5rdKdjm0fy9FS5w6cVG0eASH9u1NU27j6qU8df+NvDluMF36PZJyf1JiIs8N6Mvovh2pUqsO5ar8f/9UPbgvjmIlT73vxUpEcDCD933l17OofnldAM4rEUGz6Bt49LZoBvdpT/4Chah22dU5ntl4x3+2AS8iBUXkKxFZJyIbRaS7iGwVkbEiskpE1orI5SIyT0T+EJG73O1ERCa622wQke4BnvtKEflRRCqKSB8RedG9/00ReV5EVorInyLSxb0/l4i8LCKbRGSWiMxOfiyd7ONF5GcRWS8ik0TkWqA9MFFEfhKRSiJyu4h85/59H4tIgXTKLRaROu7zlhSRre6/q4vIGrfcehGpktl6dO/Pibq8QkSWiMj37nOd75a5z6c+ZmR6J0iXBvpD/Uto2jK+vRjx8fGsWLqYJs2apyn39uuTCQsLo3mrtllLeZoMABrgb0kuExYWxvQPP+Wr+YvYtHEDW37/LaXM3fc+wFfzF9GyTTs+nJF2HtsZBg2QIXWRwHW+fOliihcrziXVAg+j63fvA3w5byEtWrdl5hnmzFT9pVMmM9vOnT0rpScuWc1atfnk8694b8ZHvP7aZE6cyMY5aYGq8BwOKsup+kxMTOTwoUO8M/1DHnh4EIMGPBB4fzlbgZ4qTe5ARTJXt69NfpWwsHBat213+sIZCNn6PRUubYZzPOjxXH/mO0Z3JiqqFD27d2bihLHUvvQywsLCsj/zGR7njx87xhtTJ3PX3Wd20vM0QbOUE+Dff48y6OH7eHjg4JQRVh99OIOHBg7mq/mLeGjgYEaNGJbVoKfPeZq/5d9/jzLooft4eFCAnF+7OYdnNadfoLR5MvnZKXNhBTr0uJlRj9zNmCH3Ur7SRWe8D2ZWoPc3UMwaVzfk4efe4cZBY/zmuucKC+P+Sa8zZPJMdmzZzJ7tf+ZIzlAR+Ksp8Pv+6/rvWfnNl0TffDcAR48cYt3qZYya8hHj3/iCkyeOsXrx3BxMG3wi3vx/sPxnG/BAS+BvVa2tqjWA5D17h6rWBZYBbwJdgGuA5FWSOgGXArWBZjiN4ZSuGbeR/CrQQVUDHX3OB+oDbYHknvlOQHmgJnAbUDe90CJSHIgGqqtqLWC0qq4EvgAGquqlqvoH8ImqXqmqtYHNQN90yqXnLuA5Vb0UqAPsTKdcevUI2ViXwA7gBaCLql4BTAPGuEUHA5e59XFXOvV2h3siYe3bb6RdPMlXRGQUsT7DXeNiYygZEeFXJjKyFLF7fMrExFDCZ8jctyuWcVHVSyhewn/I0pxZn7Ny+VIeHz0hy8MWI6OiiPHJEBMbQ8lI/2F7kZGl/IbuxsbsISLV31K4SBGuuPIqVq30H+oG0LJVGxZ+Mz+LOUv55YyN2UPJVMMLU/8tsTExREREsu6nH1i6ZBEdWzVj2OCHWfvdaoYPHZTmNVq0asOiBWe2KFxUVCn2+Naf+5p+ZUqlLrOHiMjI026bkJDAgm++pkXLwCMwKlaqRP78+f1OmmRVycgov57juNgYSqR6r3NSTtVnVFQUTZtd7wxnrVmLXJKLA9mw4FYyZ9871SsaE7MnQO4o9qQuE3n6IbJffP4pS5cuYsyEiVn+vIdq/SZz9k//42qJbBxmnBnn+jMfHh7OwEeG8uHHn/PsC69w+NBhyvmMyMmMnDjO79y5g7937aRnt460b3UdsTEx9O7ROc00sKzkjI2NSfMZiYwsxZ6YVHXr5kyIj2fQQ/fTsnU7mvqc+J715Wc0ve56AJo1b8mmjRvOOmNKzpi03zX+ZQLsA6lztkmV84vPaNos+3L6Kh4RxT6fY/v+uJg0vy0ycl2rjjz56nRGPvMahQoX4fwLymZbNl9Fi0fwj88Q7n/2x1Ekg2HbFavVZt+eXRw9dNDv/vwFC1Ox+mX89tOaHMkZKoqViODA3lPv+4F9cQGHwe/cuoV3XxrHXUMnUKiIM53rl3VrKRlVmsJFixEWHs6l1zTmz1+yb5803vdfbsBvAJqJyAQRaaCq/7j3f+Hz+GpVPayqccBxETkPp/H9vqomqmoMsAS40t3mEmAK0E5Vt6fzup+papKq/gwkj42pD8x0798DZDRB6RBwHHhNRDoB6U3UqSEiy0RkA9ALONOVQFYBQ0XkEeBCVT2WTrn06hGyty4vBmoAX4vIT8AwoIxbdj3wnoj0BhIChVTVKapaR1Xr3HTLbRn+4VWr1WDnju38vWsn8fHxLJg/h/oNm/iVqdeoMXNnf4GqsmnDOgoVKkTJkqd+MH0zbzbXpRo+v3rlct5763XGPf0C+fLlzzBDZlSrXpPt27exa+dO4uNP8vXc2TRs5J+zYeMmfPXl56gqG9b/RKFChSkZEcmB/ftTVsU+fvw4a75dRfnyzsIo27dtTdl+6eJFlK9Qkay4pHoNdmzf5tbnSb6eNydNzgaNmjJnVnLOdW7OCO657yFmzV/EZ3O+YfT4p6hz5dU8MfbJNDmXLVnEhWeYs3qNmmzfvpVdO3cQH3+SeXO+olGTpn5lGjVuyqwvPkNVWb/Oqb+IiMjTbrv625VUqFjRb0jprp07SEhwds+//97Ftq1/UfqCC84oc0aqXlKdXTu2sftvZ79d9PVcrm3QONue/3Ryqj6bNG3Gd2u+BWDb1r+Ij4+nWLFi2Zz71Odo3pzZNA6Y+/M0uTOyYvky3nz9NZ594RXy58/65z1U6zdZ1UtqpNo/55zT/RPO/Wf+2LFjHPvX+YpetXIF4eFhfgvmZUZOHOcrV7mI+YtX8MWcBXwxZwGRUVG8O+Njv++wM1Wtek12+OScHyBno8ZNmB0gp6oycsQwKlSsSO+b+vhtExERyfdrvwPguzXfUrbchWedMSXntlQ5G2eQc91PFCrsk3P4MCpUOE3O1VnP6avyxdXYvWsHMbt3OaP7Fs+nzrWNMr39Pwf2AxAXs5vVyxdSr2nLbMvmq0zlquzbvZP9MbtJiI9n3YqFVKtTz6/M3t07U0Y47PrzNxITEihQuChH/jnIsaOHAYg/cYIt69cScUH2L7QXSi6scgmxu3eyN+ZvEuLjWbvsG2pd5T/1cn/cHqaMG0KfB4YT5VNfxUtG8devmzh54jiqyi/r11KqTPlz/BeYYPrPrkKvqr+JyBVAa2CciCR3MyaPaU3y+Xfy7XAyXuhwN5APuAz4O50yvs8pqf6bmdwJInIVcB3QA+gPNA1Q9E2go6quE5E+QON0njKBUydq8vm8znQRWQ20AeaJyG2qmmZp4ED1qKrJPezZWZcCbHJ79FNrAzTEmR7wmIhUV9WADfnMCA8P58GBQ3n43jtJSkykTftoKlSqzGcfOauid+zSnbr1GvLtimX06NiKfPnyM2T4qJTtjx8/xto1qxj46HC/533myTHEx5/koXtuB6B6jVoMGOpf5kxzDhoyjPv63UZiUhLtO3aiUuUqfPyhM4ugc7ce1GvQiBXLlxLdtgX58uXj8ZFjAdi7N44Rw4aQlJRIUlISzZq3pIH7Y+vF555m29a/yJUrF6XOL82QM1jZPb2cAwY/yn39bicpKYl2HaKpWLkKn8x0cnbq2oN6DRqycvlSOrdrSb58+XjsiTGneVZ46fln2O6T85FHz6wuw8PDGTz0cfrdeRtJiYl0iO5M5cpVmPnB+wB07X4DDRo2YvmyJbRrdT358ufniVFjM9w22dw5s2nZyn+11x9/+J5pr08lPDycXLlyMWTYCIoVK35GmTMSFh7OvQOG8sj9/UhKSqRV246Ur1iZLz9xV0vv1I39+/bSr08P/j16FMmVi49nvMu0GZ9RsGCh0zz76eVUfXbs1Jnhw4bSuWNbcufOzaix41N6s1s1b8rRI0ecBuHCb3hlyrQzbiCFh4fzyNDHuPvOviQlJtEhujOVKldh5gcz3Nw9qN+wEcuXLaV9q+bky5+PEW5ugMEDH+L7777j4MEDtLiuEXfdfS/RnbswYcwoTp48Sb/bbwWc6RPDhj8RUvWbnU7tn3e5+2d0Ovtnd5/98x2mzfg8W/ZPOPef+f3793H3nX3JJbmIjIpi9Lgzv0RfTh3ns1t4eDgDhwzj3lQ5P3JzdvHJ2dHNOdzNue7HH5g96wsqV7mInt2iAWcaV/0GjRj2+EgmPTmWxMRE8uTJy6OPZ+2SkeHh4Qwc6uZMzCDnsqV0bOPmHBUgZ1c3531uzuEjmTTBJ+fwrOX0FRYWTt97BzFmcH+SkhJp0rIDZctXYv6XHwHQvF0XDuzfy+C7b+TYv0cREb765H2eeX0mBQoWYtITAzl86B/Cw8O57d7BFCpcJNuypc7Zvu8DTBszgKSkJOo0aU1U2Qp8O/9zAK5p3oGNq5fyw5J5hIWFkztPHno+OBwR4fDBfXz44lg0KQlVpWbdxlxyxbU5kvN03hrXhwZXVKHkeYXYMncUo16dzVufrTrnOcLCwulxx0O8MOJBkpISufa6tpQuV5Glcz4FoGGraL6a8QZHDh9ixuRJAOTKFcaQp6dR4eLqXHZtE8Y+2IdcYWGUrXgR9Vt0OOd/w7l0rqdkeZ3kyFw4DxCR0sB+VT0uIh2BPjjDueuo6l630VtHVfu75bfiDCVvCNyJ02AtDqwFrgaqAgOAvsB84H5VXez7PCLyJjBLVT9yn/OIqhYSka7AzTgN0AicIe93JJdLlbsQUEBVY93h9FtUtbiIvAD8oKpvuOX24ixWdwCYDexS1T4Byr0GfK+qr4jIA8ADqlpeRCoCf6mqisizwFZVfTYz9aiqHZPrK7vqElgJ/AzcqKqrRCQ3cJFbV+VUdat7307gYlU9mOZNd8Uejg+JnTpf7pyZp5bdkkLkGJE3PDTqc9+R0LhWa4lCWVv9/VwJOC/Tg0Llx8e+IyeDHSFTShTKE+wImRKfmBTsCJkSGnsnIRP0r9jgr3KeGX8cOHL6Qh7Q8+bRwY6QKV+9f/Yncc+1plVLhMinybHnH2/+ti9VNHdQ6vE/2wOPM998oogkAfFAPyBNgzmAT3HmqK/DWWNikKruEZGqAKoaIyLtgDkicmsms3yM06O+EfgNWA38k07ZwsDnIpIP56sqeanTGcBUEbkPZ675Y+7zbMMZwl44nXKTgA9F5EbAt4e9O9BbROKBPZyat55aoHrMrEzXJXCrm/d5ESmKs28+i1Nf77r3CfBMRo13Y4wxxhhjjPmv+s/2wHuNiBRS1SMiUgJYA9Rz58ObbGY98NnLeuCzl/XAZy/rgc9e1gOfvawHPpuFSFDrgc9e1gOf/UKuB/6QN3/blypiPfD/dbPchd3yAKOs8W6MMcYYY4wx5kxYA/4cUdXGqe8TkU+BCqnufkRV552TUP5ZSgALAjx0naruO9d5jDHGGGOMMcb4swZ8EKlqdLAzJHMb6ZcGO4cxxhhjjDHGJAup8f7nwH/5OvDGGGOMMcYYY8x/hjXgjTHGGGOMMcaYEGBD6I0xxhhjjDHGeJLYGHo/1gNvjDHGGGOMMcaEAGvAG2OMMcYYY4wxIcCG0BtjjDHGGGOM8SSxdej9WA+8McYYY4wxxhgTAqwBb4wxxhhjjDHGhAAbQm+MMcYYY4wxxptsBL0f64E3xhhjjDHGGGNCgDXgjTHGGGOMMcaYEGBD6I0xxhhjjDHGeJKNoPdnDXjzn/Pl5t3BjpApjctHBDtCphQpkDvYETLl+Mn4YEfIlI1/Hwp2hEy5qnzxYEfIlCTVYEfIlAJ5w4IdIVNi/zke7AiZFBrvu0ho/OwsXjBPsCNkSohUJzsP/RvsCJny9Zb9wY6QKV+9/0SwI2RKmxuGBztCph378cVgRzBZYEPojTHGGGOMMcaYEGA98MYYY4wxxhhjPClURt+cK9YDb4wxxhhjjDHGhABrwBtjjDHGGGOMMSHAhtAbY4wxxhhjjPEksXXo/VgPvDHGGGOMMcYYEwKsAW+MMcYYY4wxxoQAG0JvjDHGGGOMMcaTbBV6f9YDb4wxxhhjjDHGhABrwBtjjDHGGGOMMSHAGvDGGGOMMcYYY0wIsAa8McYYY4wxxhgTAmwRO2OMMcYYY4wxnmSL2PmzHnhjjDHGGGOMMSYEWAPeGGOMMcYYY4wJATaE3hhjjDHGGGOMJwk2ht6X9cAbY4wxxhhjjDEhwBrwxhhjjDHGGGNMCLAh9Ob/3p/rv2PBOy+jSUnUatyKa9r18Hv89+9XsvzjNxERJCyM63rdTZmLawDw3ZyPWb9kDoJQsmx5Wt8+kPA8eXIk59rVK5j83JMkJSXRom003Xrf6vf4jm1/8cy44Wz5bTM3396fzjfc7Pd4YmIi99/ekxIlI3niyReyNdu3K5fx3KTxJCUm0rZjZ2685Xa/x1WV5yaOY9WKpeTLl5+hI8Zw8SXVADh8+BATRj3On1u2ICIMGT6KGrUuBeCjGe/x8YfTCQsL49r6Dbn7/gFZyrl61XJeeGo8SUmJtOnQmV4335Ym5/NPjWP1ymXkzZePIY+P4aKqTs4Pp7/NV59/jIhQoXIVBj82mrx58/LGlJeY9fnHnHdeMQBuv/t+rqnXMEs507P5h2/5ZNpzJCUlcU2ztlzf6Ua/xzesWcZX779GLhFyhYURfet9VLqkdo5kAVi1YhnPThpHYmIi7aO7cFOA9/2ZiWNZudx53x97YmzK+x7dphkFChYkLFcuwsLCeeO9mQBMfvl5li1eSK5cQrHiJRj2xFgiIiKzLfO3K5fxrLuvtktnX33WZ199NNW+Ot5nXx3qs6+ejRXLlzFx/BiSEpPo2LkLt952R5osT44bw4plS8mXLx9PjBnHJdWqZ7jtr7/8wphRwzn277+ULn0BYyZMolChQinPuXv333Ru35a77r6Hm27pe9bZk/303UrefvUpkhKTaNKqAx269/F7fPnCOXzx4dsA5MuXn773DubCShexL3YPL08cwcED+xARrmsdTavoG7KcJzPWrFrOS89MICkpidbtO3HDTf71sH3rXzw5+jG2/LqZW++6l269+gR+omzM8+LTE0hKSqR1+070DHBcevHp8axeuYx8+fIx6LHRKcelj2e8y1eff4yq0qZDZ7rc4BwTRj46gB3btgJw5MhhChUqzNR3PzrjbCuWL+VJdz+L7tw13X10+bIl5MuXj5Fjxvvso4G3/eWXzYwZOZwTJ04QHhbGkMdGULNmLXbt2kmn9q25sHwFAGrVqs2w4SMzl3PZUib4vFbf29PmnDBuDMuXLiFf/nyM8s2ZzrZPT5rAksWLyJ07N2XKlmPk6HEUKVKEVStX8NwzTxEfH0/u3Ll58OGBXH1N3TOu2/Rs/mE1n057Dk1K4upmbWnWqbff4xvWLGPO+68hkivlOF/xklrZ9voZqRZVkK61SiEirNx6gPm/7fN7vErJAtxVtyx7j8YD8NPfh5jzy14Ael9+PjVLFebwiQRGL/gzR3Nu+uFbPpz6LJqUSL3r29Giy01+j69ZPI/5n7wLQN58+bmh30DKVKgCwILPZ7Di6y9B4IILK3HTfY+SO0/eHM0byKvDe9GqYQ3i9h+mTtex5/z1vcZWofdnDXjzfy0pKZFv3nqBbo9MoHDxkrz9eH8qX16XkhdcmFLmwuqXUfnyuogIsdv/5IsXR3Pbk9M4vH8vP8z/jFsnvEbuPHn5/IVRbP52ETUbtsj2nImJibz89DjGPPMqJSOieOD2XlxTrxHlKlRKKVO4SFHuun8Qq5YtCvgcn8+cTtkLK/Dv0aPZnu3p8WN45uWpREZFcduN3anfqAkVKlZOKfPtimXs2LGNGZ/NYdPG9UwaN5Kpb88A4LmJ47i6bn1GP/ks8fEnOX78OAA/fLeaZUsW8taMT8mTJw8H9u8L+PpnkvPZJ0fz1ItTiYgsxZ03d6degyaUr3iqDlevXMbOHdt57+PZ/LxxPU9PGMWrb7xPXGwMH3/wHm9/8Dl58+Vj+JCHWfj1HFq17QhA1xtupEfvW7KU73SSEhOZOfVp7h7+DOeViOSpQbdR88r6lCpbIaXMRTWvoMaV9RERdm3dwptPPc6jL0zPkTyJiYk8NWE0z738GpFRUdzauzsNUr3vq1YsZcf2bcz8fC6bNqznyXFP8PrbH6Q8/tLkNzmvWDG/5+19063cefd9AHz4/jtMm/Iyjzw6Ivsyjx/Dsxnsq6tWLGPnjm18EGBffdbdV8ek2lfPNsv40SN5Zeo0okpF0at7Vxo1aUqlSqeyLF+2lO3bt/H57HlsWL+OsaOe4J33P8xw25HDh/HggEHUufIqPvvkY95643Xuuff+lOecNGEc9Ro0OOvcvpISE3njpScZOu5FSpSM4tF7b+aKaxpS5sKKKWUio0rz+MTJFCpchJ++W8HU58Yy+vk3yRUWTu87HqBClaoc+/coQ/vfRM3Lr/bbNickJiby/KSxPPn8FCIio7j7lhuo26Ax5f2OpUXo/9BgVixZmKNZkvM8N3EME1+YQkRkKfr16cG1AY5Lu3Zs452PvmLzxvU8++RoXp42nb/++J2vPv+Yl9+YTu7w3DzywF1cU68hZcpdyONjJqVs/8pzEylYsFCglz9ttnGjR/Lq1Dfc/axLOvvoVr6YPZ8N69cxZtQI3n1/ZobbPvvURO7sdw/1GzRi2dIlPPvURF5/8x0AypQtx4cff37GOceOGcnkqW8QFRVFz+5daNykKZUqp8q5bStfznFyjh45gvdmzMxw22vq1uO+Bx4mPDycZ56ayOtTJ/PgwwM5r1gxnn/pFSIjo/j999/od0dfvlm07IzrN5CkxEQ+nvo0dw1/hvNKRPDMoNupcWW9dI/zf2/dwltPDWfIC+9ly+tnRIDutc/n+eXbOHgsnkeaVGT97sPsOXzSr9yWvf/yyqodabb/dts/LPnzADdfUTpHcyYlJjJj8iTue+I5ipWIZPyAvtS6qgHnlztVhyWiSvPg2JcoWKgIG79fxXsvTeCRSa9xcF8ci2bN5PEXp5Mnb16mPjmMtcu+oe51bXI0cyDvfPktr36whNdG3XT6wub/jg2hd4lIYxGZFeD+9iIy+By8fkcRqZZd5c7gdcuLyMZU940QkQy7OkWkj4i8mF05gmX3H79yXlRpzos8n7Dw3FxyTWO2fL/Sr0yefPkR99Rf/Inj+K6jkZSUSMLJEyQlJhJ/8gSFipXIkZy/bd5I6QvKcn7pMuTOnZuG17Vg1fLFfmXOK1aciy6pQVh42vNye2Nj+G7VMlq07ZTt2TZv2kCZsmW5oExZcufOQ7PmrVm+2P8kwrIlC2nZpj0iQo2atTly5DB74+I4euQI6378nrYdOwOQO3ceChcuAsCnH31A7z63kccd0VCseNbqdvOmDVxQphylLyhL7ty5adq8FcuX+v9AX750ES1aOzmr16zNkcOH2bc3DoDExAROnDhBQkICJ44fo2TJiCzlOVPbtmwm4vwylCx1AeG5c3N5/WZsWLPcr0ze/AVS9tWTJ47n6KIvP2/cQJky5U697y1asXSxf30uXbyQVm07OO97Lac+98bFZfi8BX16i48dO5by92SH1Pvqdc1bsyzVvro81b56ONW+2i7Avno2Nm5YT9ly5ShT1snSolVrFi9c4FdmyaIFtG3v1F+t2pdy+PAh4uJiM9x229a/uKLOlQBcU/daFnw9P+X5Fi34hjJlyvo1wLJiy6+bKFW6LFHnlyE8d27qNr6etauW+JW5qHptCrn1VLlqTfbvjQWgWImSVKhSFYD8BQpyQdny7N+b8b6RHX75eaN7HHCOpU2ub8nKpf77QLHiJaharQbhAY6l2Z8n1XHp+lZp8qxcuojrWzn7ZDWf49K2rX9SrUYt8uXLT1h4OLUvq8PyJf77kKqy+Jt5NG3e+oyzOfvZhT77WZs0++jiRQto275jOvto4G1FhKNHnBPJR44cJiIyayNsNm5YT9my7mvlyUPL1m1YvMg/56KFC2iXXs50tr22Xv2UfaBW7UuJjdkDwCWXVCMyMgqAypWrcPLESU6e9G/Enq3tWzZT8vwLKFmqNOG5c3NZ/evYeJrjPOdoca/yxfMTd/Qk+/6NJ1Hh+53/UPv8wpnefsu+fzl6MjEHEzq2/v4zEaXKEOF+V9Zp0Ix1a/xPsFS6pCYFCznHpQoXV+fAvtiUx5J/zyUmJnDyxHGKFi+Z45kDWfHDH+z/59+gvLbxPmvAn4aqfqGq48/BS3UEMtMwz2w5kwlHDuylcPFTDbHCxUty+MDeNOV+W7uc1wbdysdPDaPVbQNSyl7ZuguvPtCLl+7tTt78BalQs06O5NwXF0vJyFIpt0tGRLFvb2wGW/ib/PxEbr37AXLlyv4v+rjYGCKjzk+5HREVRVxcjF+ZvbGxREadyh8ZGcXeuBj+3rWD84oVY+yIR7mlZ2fGj3ycY8ecL6wd27ey/sfvuf2mHvS//WY2b9qQpZx74/wzRERGsTfOvw73xsakKRMXG0NEZBQ9evehW/tmdGrdhIKFCnPlNfVSyn06831u6RnN+FHDOHzonyzlTM8/++I4r8SpH7rnlYjgn/1pGzzrvl3CmHt7MmXMQG7oPyRHsgDExcUQWcr3PS1FXKx/fcbFxhKVuj7dfUNEuP+e2+jTswufffyh33avvvgsHVo1Zf6cWdze797sy5xqX40MsK/GBdhX4+Ji2OXuq2NGPEqfnp0Z57Ovno3Y2BiiSp3KEhVVirhY/yyxMTGUSlUmNiYmw20rVa7C4kXOiZSv588lZs9uAI79+y9vTJvKnXffc9aZUzuwL44SEVEpt0uUjOJABo3wxXM/59Irr01zf9yev9n6x69Urlo927KlZ2+c83lOFug4cC6lPjaWjAxw/Ezn2FWhYhXW//g9//xzkOPHj7F65bKURmay9T99T7HiJShT7kLOVGxsDKV8PuNRUVHEBtxHfcuc2kfT23bgI0N55qknaXFdI56eNIH7HngopdyuXTvp3qUjffv05ofv12YuZ0wMpc73+cxGRRETkypnbAxRgXJmYluAzz75mHoN0k6N+mb+PKpecknKieasOpjqOF+0RAT/7E/7m2T9t0sZd28vpo4ZxA39c7yPCYDz8oVz4Fh8yu0DxxIomj93mnIViudnaNOK3HNtOc4vfO6Hnh/cF0exkqc+48VKRHBwX/rHpZVfz6L65c4UiPNKRNAs+gYevS2awX3ak79AIapddnWOZzanJx79f7CEbANeRAqKyFcisk5ENopIdxHZKiJjRWSViKwVkctFZJ6I/CEid7nbiYhMdLfZICLdAzz3lSLyo4hU9O1pFpE3ReR5EVkpIn+KSBf3/lwi8rKIbBKRWSIyO/mxdLKPF5GfRWS9iEwSkWuB9sBEEflJRCqJyO0i8p37930sIgXSKbdYROq4z1tSRLa6/64uImvccutFpMpZ1vNiEZngPtdvIpJm7KWItHHrvGQGdRSw3t16a+/++1MRmeb+u6+IjHZHCGwWkalu/c4XkfwBMtzhvudrl3ya+SHDqhrob05z30V16nPbk9OIfmAEyz9+E4DjRw+z5ftV3Pn0O9z9/AziTxxn04pvMv3aZ0IJkDOTh47VK5ZyXrFiVLk4Z877BKjCNHUYqJ4RITExkd9+2UzHLj14Y/rH5Mufn3ffeA1whkUePnSIKW+9z933P8zjgx8O/DyZzhlo21Q5A9WzCIcP/cPyJYuY8dk8Ppm9kOPHjjF/zpcAdOjcnemfzOH1dz+mRIkIXnpu4llnzEigbIG+Pmpf04hHX5hO30fGMfv9qTmSBdL77KQplbaMm3nyG+/x1vSPefrFyXz84fv86PNj/a7+D/D5nIU0b9WWj2Zk39DQs91XxWdfje7Sgzenf0z+/Pl5x91Xzy5MgPvSZEknbwbbjhg1lg/ff4+e3Trx79Gj5M7t/MB+5aUX6H1jHwoUKHj2mVNJ73MdyKaf1rJo3hfc0Le/3/3Hj/3LM6Me4aa7HqLAWQzzPmOBIgfxZ1h6xxy/MoH2SeDCChXpcdOtDLz3Dh65/y4qVbmYsLAwv3IL5885q973dF83k5+XjLad+cH7DHhkCPMWLGHAoCE88fijAERERDL360V88NFnPDxwMEMGPcyRI0dOnzMTdRjowyQimdp26uRXCAsPo03b9n73b9nyO88+M4nHMjlPPzvVuqYhQ154j1sfGcvs97NwHDoTmfiY7Dh4nMfm/s7YhX+y+I/93Fm3TM7nSiXw4TFw+F/Xf8/Kb74k+ua7ATh65BDrVi9j1JSPGP/GF5w8cYzVi+fmYFpjzk7INuCBlsDfqlpbVWsAyZ+wHapaF1gGvAl0Aa4Bko+wnYBLgdpAM5zGcEpXhttIfhXooKqBVtk4H6gPtAWSe+Y7AeWBmsBtQLqrmYhIcSAaqK6qtYDRqroS+AIYqKqXquofwCeqeqWq1gY2A33TKZeeu4DnVPVSoA6wM4OypxOuqlcBDwDDU/090cBgoLWqJp8mTq+OLiVtvS8Fkk8KXMCp0QX1cd5DgCrAS6paHTgIdE4dUFWnqGodVa3TKLpnpv+wwsUjOOzTi3l4/14KnZf+UO2yVWtxMGY3/x7+h60bf6BoRCkKFDmPsPBwLrqyPrt+/znTr30mSkZEsTf2VM/K3rgYimdyCPfPG37i2xVL6NO1FRNGDGb9D98xceTQbMsWGRVFbMzulNtxMTGULOk/JDIiKsqvZyg21ikTERlFRGQU1Ws6C/A0adac337Z7GwTGUXDps2cYaM1aiGSi4MHD5x1zohI/wxxsTGUjIhIVaZUgDKRrF3zLeeXvoDzihUnPDw3DZpcx8b1PwFQvERJwsLCyJUrF207duGXTX4zUrLNeSUiOegzzO/gvrgMh/ZVrn4pe/f8zZFDB3MkT2RkKWL3+L6neyiZarG5iMgoYgLUJ5CyMF3x4iVo1OQ6ft60Ps1rNG/ZhsULv86+zKn21dgA+2pkOvtqZKp9tbHPvnq2WZJ7xwFiYvakWawvqlQUe1KXiYzMcNsKFSvyytRpTP/wE1q2bkOZsuUAZ5jxs09PpHXzprz37tu8PnUKM6a/e9b5AYqXjGSfT2/xvr0xFCuRdp/c9ufvTHl2NANGTKJwkfNS7k9ISOCZUY9Qr2lLrqrfNEtZMqukO6omWVxsDCUizu10GF+pj0t7YwMcPwMcu0q473fr9p2Y8vaHPDf5LQoXKUqZsqd62hMTEli+6BuaNDu7dVmiokqxx+czHhMTE2AfTV3G2Ucz2vbLLz7lumbNAWjeohUbNzif/Tx58qQsBlqteg3KlC3Htq1/ZS7nbp/PbEwMkZGpP9eliEkvZwbbfvHZpyxdsphxEyb5NQJj9uzhwfv6M3rsBMqWK3fajJl1XokIv+P8P6c5zleqfin7cvA47+vgsQSK+fS4F8sfzj8+PfIAxxOSOJHoNKE3xRwhTISCefxPKuW0YiUiOLD31Gf8QDp1uHPrFt59aRx3DZ1AoSJFAfhl3VpKRpWmcNFihIWHc+k1jfnzl6yN/jMmJ4RyA34D0MztHW6gqsnjVr/weXy1qh5W1TjguIich9MwfF9VE1U1BlgCXOlucwkwBWinqtvTed3PVDVJVX8Gksfo1AdmuvfvAQKvIuY4BBwHXhORTkB6YzBriMgyEdkA9ALOdGzhKmCoiDwCXKiqx9Ipl16Xpu/9n7j//R7nREWyJsAjQBtV9W1ZpVdHgep9GdBAnHn9PwMxbsO+LpA8Gf0vVf0pnQxZcn7FizmwZxcHY3eTmBDP5m8XU/ly//MvB2J2pfQm7Nn6O4mJ8eQvVIQiJSL5+4/NxJ84jqqybdOPlCidfV/kvi6qWp2/d25nz9+7iI+PZ+mCeVxTv1Gmtr3lrvt455P5vDlzDo+MGE+ty69k4OPZt6Jp1Wo12LFjO3/v2kl8/Em+mT+beo2a+JWp37AJc7/6AlVl44Z1FCpUiJIREZQoGUFkVCm2uz/S1q75NmXxpoaNr+OH71YDsH3bVhIS4lN+3J1tzp07trN7107i4+NZOH8O9Rr456zXoDHzZjs5N21YR8FChShRMoKoUufz88b1HD9+DFXlh+9Wc2F5Z7GtfT5DhpctXkCFbJpfnFq5ylWJ272DfTF/kxAfzw/Lv6HGlfX8ysTt3pmyr+7441cSE+IpWLhojuS5pHoNduzYdup9nzeHBqne9waNmjJn1ufO+75+HQULFaZkRATHjv3LUXcxxWPH/mX1tyupWMkZJLRj+9aU7ZcvXZRSz9kheR9Izrxg/mzqn+G+mtyg+N5nXz0b1WvUZPv2beza6WSZN2c2jZv4N2IbNW7KrC+c+lu/7icKFSpMRERkhtvu3+cs9piUlMTUya/SpZtzVY1pb7/H7PkLmT1/Ib1630Tf2++gR0//1a3PVKWLq7Fn13Zi9+wiIT6eVYu/5opr/IcZ743dwzMjB3HPwCc4v8ypxqWqMuXpUZQuW542nXtlKceZqHpJdXbt2Mbuv53jwKKv53Jtg8bn7PXT5qnhl2fh13Oo29A/z7UNmvD1HGef/NnnuASkLO4Zs2c3yxZ/Q9PmrVK2+/67bylbvgIRPsPvz4Szn21l184d7n72FY0C7qOfpbOPBt42IiKStd+tAWDN6m8pd2F5APbv309iojNHeueOHWzfvpUyZctmOufOnTuIP3mSubPT5mzcpClfZpAz0LYrli3ljden8tyLr5A//6mBf4cOHaJ/vzu4/4GHuOzyK86qbtNTtnJV4nbvTDnO/7h8AdWvrO9X5lwe531tO3CMyEJ5KFEgN2ECV5Qpyvrd/iMkiuQ91Vi/sFg+Z72DczDv3deFVS4hdvdO9rp1uHbZN9S6yr8O98ftYcq4IfR5YDhRF5z63Va8ZBR//bqJk+7vul/Wr6VUmfLnNL9JR7DHyntsDH3IrkKvqr+JyBVAa2CciCSv1HPC/W+Sz7+Tb4eTcXXvBvIBlwF/p1PG9zkl1X8zkztBRK4CrgN6AP2BQF0PbwIdVXWdiPQBGqfzlAmcOhGTz+d1povIaqANME9EblPVQEvq7gNSt4qKA76nvZP/5kT895k/gYrARcDaAOXhNHWkqrtEpBjOiIql7mt3A46o6mERKZHq+RKBNEPoz1ausDCa3dSfmROHoElJ1GzYgpJlyvPjAmd49GXXteO375axcfk3hIWFEZ4nL+3vGYaIULryJVx8ZQPeeuxucuUKI7J8JWo3ObuhiqcTFh5OvwcHM+zhfiQlJdG8TQcurFCZrz5zLr3VpmNX9u/by/239+Tfo0fJlUv4bOZ7TH7nkxwflhoeHs5Dgx7lof53kJSYRJsO0VSsVJnPPnJWG+/YpTt16zdk1YqldO/Qinz58jF0xOiU7R8cNJQnhj1CQnw8pS8owxD3sTYdohn3xGPc2K0DucNz8+iIMVla0Cw8PJwHBg5lwH13OpdrahdNhUqV+fxjJ2eHzt25pl5Dvl25jJ6dWpE3X34GPzYKgGo1atHouuu5/cZuhIWFUfniqrSL7grAKy88xZbffkUESp1/AQOGDE83Q1aEhYXT+baHeGXkQ85l5K5rw/nlKrJ83mcA1G/RkXWrFvPdkrmEhYWTO09ebn74iWxdBM5XeHg4Dz/yKA/ccztJSUm0bR9NxUpV+OQjZ8X2Tl16cG39hqxcvpSuHVqSN18+ho0YAziNzMEPOyvNJyYm0LxlG+rWcwbivPz8M2zf9hciuSh1fmkGPZp99RkeHs6D7r6amJhEW3df/dTdV6N99tVumdhXfR87myyPDH2Mu+/sS1JiEh2iO1OpchVmfuDUX9fuPajfsBHLly2lfavm5MufjxGjxma4LcDc2V/xgTvtoGmz5nSIzv6FK5OFhYXT555BjBt6H0lJiTRu3p6y5Svx9ayPAbi+bWc+ee81jhz+h2kvTgAgV1g4Y198m183rWPZgtmUrVCZwf2cUVPdb7mHy66ql+7rZUvm8HDuHTCUR+7vR1JSIq3adqR8xcp8+YmzDkO7Tt3Yv28v/fr04N+jR5Fcufh4xrtMm/HZWa3knuk8991FYlIirdpFU6FiZb5w87Tv1I2r6zVg9cql9O7cOuUycslGDH6IQ/8cJCw8nPsHPkrhIqcacou+Pvvh8+DsZ4OHPk6/O28jKTGRDtGdqVy5CjM/eB+Art1voEHDRixftoR2ra4nX/78POGzjwbaFuDxJ0bx5PixJCYkkCdv3pQh6D98/x0vv/g84WFh5AoLY9jjT1C06HmZyjnk0cfpd8dtJCUl0tF9rQ/dnN2Scy5dQttW15MvX35Gjh6b4bYA48aM4mT8Se667RYAatauzWPDRzJj+rts37GdKa++zJRXXwbglanTKFEi64vYOsf5B5k88mGSkpK4+ro2nF+uAivc43y9Fh1Zv2qJ33H+phw8zvtKUvjgpz30r1eOXCKs2naQ3YdP0KCC8xNy2V8HuOyCIjSoWIykJIhPSmLamlODP2+58gIuiihAoTzhjGlVha9+jmPltoPZnjMsLJwedzzECyMeJCkpkWuva0vpchVZOudTABq2iuarGW9w5PAhZkx2rtaQK1cYQ56eRoWLq3PZtU0Y+2AfcoWFUbbiRdRv0SHbM2bGW+P60OCKKpQ8rxBb5o5i1KuzeeuzVUHJYrxHsjKnNJhEpDSwX1WPi0hHoA/OEO06qrrXbfTWUdX+bvmtOEPJGwJ34jT8i+M0PK8GqgIDgL7AfOB+VV3s+zwi8iYwS1U/cp/ziKoWEpGuwM0489MjcIa835FcLlXuQkABVY11h9NvUdXiIvIC8IOqvuGW24sznPwAMBvYpap9ApR7DfheVV8RkQeAB1S1vIhUxOm5VhF5Ftiqqs+mU5drgUdUdYGb6Vuglar+ISKLgQGqulZESgJr3efv49bnC8CnQFdV3ZRBHXUKVO+qusfdpqn7/xLAR8BHqvqgiJR3n6+G+3wDgEKqOiLQ3wLw+prtIbFTNy4fvGGbZ6JIgbSL1HhRYmJIvO38tPNgsCNkylXliwc7QqYkhch3WIG853YY6dn6ZdfhYEfIlMii535xrLNxLhpW2aF4wexZhC2nhUh1svDX4C2MeCZm/Zp2gTwv6lIt6vSFPKDNDTlzAj8nHPvxxRD5NDkOn0jy5Jd94bw5sDp0JoTyEPqawBoR+Ql4FMhsV8inwHpgHbAQGOQOewfAHd7dDnhJRDK79OTHOHPMNwKTgdXAP+mULQzMEpH1OMPIH3TvnwEMFGfxvErAY+7zfA384rN96nKTgH4ishLwneTTHdjo1k9V4O0M8t8EDHPLLgSeOM38+hSq+ivOEP+Zbp70ZFTvy3Dm2W8BfsBp4GfPRVWNMcYYY4wxIUs8+r+g1Ueo9sB7jYgUUtUj7pDvNUA93xMD5tyxHvjsZT3w2ct64LOX9cBnL+uBz17WA5+9QqQ6rQc+m1kPfPYLtR74Iye8+WVfKG9wjkohOwfeg2a5i+TlAUZZ490YY4wxxhhjTHayBnw2UdXGqe8TkU+BCqnufkRV552TUP5ZSgALAjx0naruO9d5jDHGGGOMMeZ0QmX0zbliDfgcpKrRwc6QzG2kXxrsHMYYY4wxxhhjzk4oL2JnjDHGGGOMMcb837AeeGOMMcYYY4wxnmQj6P1ZD7wxxhhjjDHGGBMCrAFvjDHGGGOMMcaEABtCb4wxxhhjjDHGm2wMvR/rgTfGGGOMMcYYY0KANeCNMcYYY4wxxpgQYEPojTHGGGOMMcZ4ktgYej/WA2+MMcYYY4wxxoQAa8AbY4wxxhhjjDEhwIbQG2OMMcYYY4zxJLER9H6sB94YY4wxxhhjjAkB1oA3xhhjjDHGGGNCgKhqsDMY43kicoeqTgl2jtOxnNnLcmYvy5m9LGf2spzZK1RyQuhktZzZy3KaUGU98MZkzh3BDpBJljN7Wc7sZTmzl+XMXpYze4VKTgidrJYze1lOE5KsAW+MMcYYY4wxxoQAa8AbY4wxxhhjjDEhwBrwxmROqMw9spzZy3JmL8uZvSxn9rKc2StUckLoZLWc2ctympBki9gZY4wxxhhjjDEhwHrgjTHGGGOMMcaYEGANeGOMMcYYY4wxJgSEBzuAMcYYY4wxxiEil2f0uKr+cK6yGGO8x+bAG3MaIlJQVY8GO8d/hYjkB8qp6q/BzmJMMhEpADyMs2/eLiJVgItVdVaQo6UQkTDgLVXtHewsmSEi9YEqqvqGiEQAhVT1r2DnClUikgeoCijwq6qeDHKkFCJSPKPHVXX/ucryXyAii9x/5gPqAOsAAWoBq1W1frCy+RKRF3D2x4BU9b5zGOeMiUgunOPSoWBnMeZMWA+8MekQkWuB14BCQDkRqQ3cqap3BzeZPxG5CBgIXIjPZ1pVmwYtVDpEpB0wCcgDVBCRS4GRqto+qMEAEdlA4B8iAqiq1jrHkQISkQ9VtVuAvJ7KmczdP18BolS1hojUAtqr6uggR0vtDeB7oK57eycwE/BMA15VE0UkQkTyeKnxFoiIDMdpeFyMU7e5gXeBesHMlUxEHsrocVV9+lxlyQwRaQO8CvyB81mvICJ3quqc4CZL8T3O8UiAcsAB99/nAduBCkFLlg73pNLtQHn8vztvDVYmnwxNAERkBnCHqm5wb9cABgQzWypr3f/WA6oBH7i3u+LsE54jItOBu4BEnIxFReRpVZ0Y3GT+RCQv0Jm0++fIYGUy3mENeGPS9wzQAvgCQFXXiUjD4EYKaCbOD7upOF9IXjYCuApYDKCqP4lI+SDm8dU22AEy6X73v6GSdyrOCabJAKq63v0B5bUGfCVV7S4iNwCo6jERkWCHCmArsEJEvgBSRgZ5rcEJRAOXAT8AqOrfIlI4uJH8JGe5GLgS9zgPtAOWBiVRxp4CmqjqFgARqQR8BXiiAa+qFQBE5FXgC1Wd7d5uBTQLZrYMfA4sA77Bu9+dVZMb7wCqutE98e0JqvoWgIj0wdk/493brwLzgxgtI9VU9ZCI9AJmA4/gNOQ91YDH2T//wcl2IshZjMdYA96YDKjqjlS/4b34JZ+gqq8EO0QmJajqP15sF6nqtuR/i8iFOEN/v3GH/HvmWKmqu93/bgMQkSJ4KF8ABVR1Tar3PCFYYTJw0n2vFVIaSF780fS3+/9cnGqEetFJVVURSa7PgsEO5EtVnwAQkfnA5ap62L09AuekqNfEJjfeXX8CscEKk4ErVfWu5BuqOkdERgUzUAYKqOojwQ5xGptF5DWc0SsK9AY2BzdSQKVxjkfJUyUKufd5UW4RyQ10BF5U1fjk45THlFHVlsEOYbzJyz/6jAm2He4wenXnHt6HN784vxSRu4FP8WlweHTO4UYR6QmEuXOM7wNWBjmTHxG5HbgDKA5UAsrgjHC4Lpi5UhORO4GRwDFODaVXoGLQQgW2120MJzfkugC7gxspoOHAXKCsiLyHMyS0T1ATBeDT8PT62hwfishk4Dz3M3UrzpQkrykH+E5HOIkzZNUTRKST+89NIjIb+BDns9QV+C5owdK3V0SG4d/g3BfcSOmaJSKtk0cLeNQtQD9OjbxaijMlyWvGAz/6zN1vhDPizosm44xkWgcsdU/Ye3EO/EoRqek7AsOYZLaInTHpEJGSwHM4w/8EZzjY/arqqR8jIhJoUShVVa815JIXCnsUaI5Tp/OAUap6PKjBfIjITzjD/Fer6mXufRtUtWZQg6UiIr8DdVV1b7CzZEREKgJTgGtx5sX+BfTyHfHgFSJSArgGZ9/81ot1KyJ1gddxFl7y7NocACJyPT6fdVX9OsiR0hCRR4FuOCdAFWfo/4eqOjaowVwi8kYGD6sX5mv7chezGw40xKnPpTjrnHjuhLKIHAYK4pz4jufUOiJFghoslVBZ+FVESgFXuzdXq+qeYOY5EyISrqqeGhkmIj8DlXG+M0/g0XVuTHBYA94YExTuitoFvbb6q4isVtWrReRHVb1MRMKBH7z2pSkic4FOqvpvsLNkRESuUNXv3SHUuVT1sIi0U9Uvg50NQu9yTSKyGuiCM884+QTTRlWtEdxk/kRkQurhyYHu8wJ3H2jg3lyqqj8GM89/gYgUUtUjwc4R6kSkPc7c7Dyq6qmFX32564X0Aiqq6kgRKQeUUtU1QY6WhohEAWOB0qraSkSq4ZwMfz3I0fy4IwPS8OLJb3Pu2RB6Y9IhIs8HuPsfYK2qfn6u86THncvVD6fHA5wF4iYnLybjJSGy+usSERkK5Hd7EO8GPNHYTGUIzhC71fhPnfDaZXumisjNPqso9wAexDt1+lQGjynguas5hMjaHNfjLA7lq1WA+7ygAHAo+XJ3IlLBa5e7E5EywAs4UzsUWI4zImxnUIOlIiFy9ZZkIlIMqIJzuTYAVNVLixgOx7sLv/p6GUjCOV6OBA4DH+MsEOk1b+JcGeNR9/ZvOKvne6oBD/TFWWRxpcenS5kgsAa8MenLh3PN3eQFjToDm4C+ItJEVR8IVrBUXsG5RNPL7u0b3ftuC1qi9IXC6q+Dcb44NwB34uT04tzdycBCnJxJQc6SkS7AR+57Xh+4CWdYtSckX64phHh6bQ4R6Ydz0quiiKz3eagwsCI4qdInHr/cnY83gOk4c9/BmVv+Bs6JEi8Jlau3ICK34cwtLwP8hDN9ZhXeOmnn2YVfU7laVS8XkR8BVPWAe3zyopKq+qGIDAFQ1QQR8eJJ0K3ADcDz7nSPZTgjhDzTgWSCxxrwxqSvMtA0eV6UiLyCMw/+epxGk1dcqaq1fW4vFJF1QUuTsVBY/bUD8LaqTg12kNNIUNUMr2XtBar6p9vr/hmwA2iuqseCmyowca6xXA3/3ri3g5cooLtw1ua4AOda9fOBe4KayN90nEubjcM5GZbssBfnQeP9y90li1BV3/nwb4rIA8EKk5EQGSECTuP9Spz1LpqISFXgiSBnSs3zC7+64t1pccmLlUbg3RPLR931TpKzXoMzutJTVHUaMM1dW6AbMABngV0vHp/MOWYNeGPSdwHOAjfJB/aCOHOmEkXES5eXShSRSqr6B6QsGubVH0yhsPpre+BZEVkKzMBZfMtTi9u4FonIHThD0T139QER2cCp1fHBWdU/DFgtInhwTYHhQGOcBvxsnOHeywFPNeDdhfV6BTtHelT1H5xj5g0AIhKJc0KkkDsvensw8wXg6cvd+dgrIr2B993bN+DN1d09PUIkleOqelxEEJG8qvqLiFwc7FCp3Isz1PsEzsmx+ThD1L3meZyFICNFZAzOyKthwY2UrodwRohUEpEVQAROXk8R5/KB1YAYnN73LrgnGo2xBrwx6XsS+ElEFuOs/tkQGOv+wPsmmMFSGYjTmPsTJ+eFOJee8RxVfR7niz7ZNhHx1BBmVb3FHSXQCugJvCwiX6uq16Yk9MRpIA9Odb9Xrj7QNtgBzlAXoDbwo7sPROHBqRNuz9btOJc6S/kO9+Bq5O2Ap3GuBR2Lc1zaDFQPZq4AAl3uzoujb24FXsQZoq44vbCees9dgUaIeHL+O7BTRM7DGfaoWLYAAEXzSURBVB30tYgcAP4OaqK0blDVRzk1XxsRGU/a436wfYQzHe46nN8hHXEanl60H+cydxfjZP0VuDSYgdJRAuek90GczHs92plggsBWoTcmAyJSGmdO+S84PfA7PbbADQAikpdTX0a/qKqXRgj4EZE2OD/ifYcpe65HwW3Et8Q5GdJAVSOCHMmPe2mhu3HmlSvOGfpXvTg83V3IKnmV72Wq6rkpHiKyRlWvEpHvgSY4izBtVFVPNThFZCXOe/09PiNtVPXjoIUKwJ3G0xT4xr2aQxOcxsgdQY6WhoTA5e5ChYi0UtU5qe67S1VfDVamzBCRRkBRYK6qngx2nmQiMgd4V1Xfc2+/BORT1b7BTeZPRL4COiYvnisi5wOzVPWK4CZLyz3Gt1fVXe7thsBLXrtUbDIRuQRnXYkHgTBVLRPkSMYDrAfemHR4fYEbEWmqqgtFpFOqhyq5Q5Q/CUqwDIjIqzgrPjfB6d3sAnjqMjMi0hLogZNxMU7ObsHMlI63cKYfJI9ouMG9z1NZReR+nB7j5P3xXRGZoqovBDFWIGvd3ripOI3jI3hs33QV8OKl2AKIV9V9IpJLRHKp6iIRmRDsUKmJSH/gPa822kVkkKo+KSIv4D8lBff2fpwG3h/nPl1Aj4nICVVdCE5+nGOpJxvw4lxCMPkk6AovNd5dnYAvRCQJZ1TYflX10poXyT4DZopIZ6AszhD1AUFNlL67gM/cUUKX41xSrnVwI6UlIm1xTnw3BIrhLFq7LKihjGdYD7wx6XDn8CYvcHNp8gI3qto9yNEAEJEnVHW4iLwR4GH12pBaABFZr6q1fP5bCPhEVT2zKrmIzMCZ+z7H4yMZ1qVavDDgfcHmrkReN/kyOO4UlFVemwPvS5zLNBVR1fWnK3uuichonMsKzQ52loyIyDc4w2jHASVxhtFfqarXBjNXam599sCZWzoNpwfeMz+MRKSdqn4pIjenU6QEcLNXPvciUhKYhTO1qyXOlVx6ePSypo/jrOqffHKxIzBTVUcHLZRLRIr73CyM00BeATwO3lnrxJeI3IPznpfHuXSgFxfbA0BE6uKsyXMcaKOqcUGOlIY72mIpzqg1r03tMEFmDXhj0iEi36nqlSLyE84lUk6IyE+qemmQo4UsEVmtqleLyLc4PQv7cIYpVwlyND/u4npVVPUbd6h6uKoeDnYuXyLyJs6Q+W/d21fj/JD31HzT5BNhqnrcvZ0P+M4rwxXdHrh0qaonFg0S5zJCijPMuyDOolbx7m1V1SJBjJeGe6LmGJALZ9G9ojg93Z5beE1EBGcI/S04l5T7EHjdQ73aGRKRu4DqqnpvsLNAysKF3+CMZLnVSydEfInIZuAyn2NTfuAHVb0kuMlARP7i1Oc9+b/JVFU9sdaJiPheCUVwphxuAJIvJ/d0MHIFIiJf4j+KpRqwGzgAoKrtg5ErI+5aLFe6N9eoamww8xjvsCH0xqQvFBa4SR6i/AbOnN2pOEPCBqvq/KAGC2yWW6cTcXq8FI8tFOYuZHUHzqrplXCmULyKszhP0Mmp1d1zAzeJyHb39oXAz8HMlo5pOCvPf+re7gi8Hrw4aawFNgHJPTB+P5TxyJQZVQ21SwfdgdObuRNnaodnuavQ7wH2AAk4w1U/chevHBTcdKenqq+KSFBPNPmcYEqWB2dBzS4i4rkTTK6tOGuxHHdv5wU8cdJGVSsEO0MmpT4ufZrO/V4wKdgBzoSIdMXJvBjne+kFERmoqh8FNZjxBOuBNyYTvLrADZwaNi0iLXCuB/0Y8IaqZtizGGzuwnv53MtOeYY74uIqYLWqXubet8FDPcYXZvS4qm47V1lOR0Ry4awdcRxnnqkAS1X1x6AG8yEiDwKdcS59NgP4VFWPBDdV+kRkgaped7r7gk2cy/J1w5mjPQP4SFU9tyq1iNwH3AzsxTmZ+Jmqxrv77u+qWimoATNJRH4I9jHfHclQVr13qUA/PusJlMPp3fzavX09sFxVewQxnh93MdV+OPOgwWnMTfbilAQAESmMc07Ms8dQCI2ebXch0OuTs4lzBZJvvDJdxgSX9cAbkwmquiTYGTKQ3GPYGqfhvs79IeU5IlIAeBgop6q3i0g5EWmgqrOCnc3HCVU9mVyFIhJO2sWjgsZLDfTTUdUkEXlKVevi0evXquozwDMiUgFnIcAFIrINGKuqPwU1nA936kFBoKSIFOPU574IzqXaPEVVnwCeEJFaQHdgiYjsVNVmQY6WWkmgU+rPlbvvhtqlEIPKHcnwKeC5lcdTWev+93tO9RiD0zj2mldwRlu97N6+0b3PU5c1FZEawDs4I9cQkb3ATaq6KajBAhCRbjijABfj7Z7tXKlOLOzDmZJkjDXgjfkP+F5E5gMVgCHuGfCkIGdKzxs4P5rqurd3AjNxFj3yiiUiMhTIL87lpe4GvgxyplA2312Z+BOvzoUFUNW/RORzID/Oj+SLcK4+4RV3Ag/gNNZ9T4YcAl4KRqBMisUZmr4PiAxyljRU9XERqe2uSA0+lzlU1c1BjHamvHLS9lsRuVJVvwt2kPSo6lsiEga8paq9g53nNK5M1eO60O2Z9ZopwEOqughARBrjTOnz1KKVrkdx6tWvZxvnWvZeMldE5gHvu7e7A55evNScOzaE3pgQ5w71vBT4U1UPikgJ4AKPrqC9VlXriMiPPsPTPbVyulufffG5LjTwmpcbn17mzo0tiDO3+DgeW3RNRCrirELeAdiBM9x7VvLCVl4jIveq9y7Bl4aI9MP5wRmB88P4A1X13BoN7hD6Ozi1Enk04LnLHIpIV1Wdmd59ItJHVd8MSjgfIvIzzsmvbcBRTn3ePXfVCbdx1M5r0+J8uWsbdE1eUNE9Xn0U7OkSqQX6Hvfad3uy1FPi3O/8dV6ZJufLPfldj1PTzz49zSbm/4Q14I35DxCRC3AWMUsZVaOqS4OXKDARWYmzGNwKVb1cRCoB76vqVUGOZv5PiXN95fXA5zi92X5fih5bRTkS6I+zerLiLFr4kkfnb44HZnhpGkIgEiKXOQw0x90L895TS2+NDi9O/RGRyTiLvn6Bc7IB8Nxn/jqckWt/4jTiLgRuSe7p9gp36sQPOMPoAXoDdVS1Y9BCpUNEJgK18O/Z3hAKC1Yak8yG0BsT4kRkAs4X0M9Aonu34lw/1GuGA3OBsiLyHs6Z5T5BTeTyWd09IK/9oA8VIbDo2khOve+FghkkIyJSD5gOvAm8jfNj/nJgjYj0UtUVQYyXQkSKqOoh4En3tu/1rL14/Wrh1HET999eGY6OiLTCWd/kAhF53uehIjijWjxFVbeJSG2ggXtXypQED/rb/X8uvLlqOqq6QESqABfj7Je/qOqJIMcK5FbgCZyRLILz++OWoCZKh6oOTNWzPcWLPdsi0gmYgDP1SPDY6DUTXNYDb0yIE5FfgVoe/VJPwx3ifw3Ol9G3qro3yJGA0FrdPRS4i64VABYBjfFfdG2OeuBay2dCRIao6rggvv63QL/UK/iLyKU4q1JfHZRgqYjILFVtK/7XsU6m6pHrVycT5zrWN3NqMbOOwJuq+mywMvlyG8OX4pxoetznocPAIlU9EIxc6RHnsqa34/EpCaHCnavfBiiP/wg7z4wSCFUiUgT/OvXUyUUR2YIzxSOU1uIw54g14I0JcSIyB2eOnKcv2wIpvYg/qepREemN04P4XCg1jkVklbuqusmA+0P+AZxF13bh9h7gNDymqKqXF15LI9jDlUXkZ1WtdqaPBUOoXE4smYhcjkcvc5jMXTn7c6+fqA2FKQki8iX+o60U5zKCi1T13eCkCkxEZuOsHbIBn8Vp3as8eIaI1AGGkvZEg2fe92QicifOCbFjOHWa3LPttZOLK1S1XrBzGG+yIfTGhL5/gZ9EZAGQ8uNOVe8LXqR0vQLUdnuVBgLTcIYDNwpqqjOTL9gBQoGqPgc8JyKPA8+q6iEReQznpM2q4KY7K8EeVi0iUix1j6s7RN1TlxYKhcuJpRrav9X9f8pjXuuNA1oBE0VkKc5Ci/NU1XND6PH4lATXpAD3FQd6i0gNVR18rgNloIwXG8EBvIfzne53osGjBgDVvTL6LwNrReQD4DP8f9t9ku4W5v+GNeCNCX1fuP8PBQnuj/sOwPOq+rqI3BzsUGfIhi2dmS6qOlJE6gPXA0/hnMjxxJDvMxDs9/0ZnEvyDeDUZeSuwJkj+UzQUqXP65cT+560Q/yTKeCp3jhVvUVEcuM05HsCL4vI16rqqeuB4yy4tto9gQPOlITXgxcnLVVdEuh+EfkCZ7/wUgN+jog0V9X5wQ5yGnGqGiq/Q/7A6fjwuiI4OZv73Kecmp5i/o/ZEHpj/gNEJD9QTlV/DXaWjIjIEpxF7G4BGgJxOEPqPXf5lvQEeyh1qEm+ZKCIjMNZ6Xe672UEQ4UXMotIW2AQUJ1Tq9BPVNUvg5krEPdyYhfj9Gx7+nJiocRtxLfEOYY2UNWIIEdKIxSmJKRHRH5S1UuDnSOZiEQD7+KMsonHowuZuavl3wCkHgnoucamiFyGe6IJD49aDDQSSEQqqOpfwcpkvMN64I0JcSLSDmdIYB6ggruo1UhVbR/UYIF1x+k96quqe0SkHDAxyJnOlNeGg3rdLvdyTc2ACSKSF48N+XYXirpPVTPqyZ6ZwWPnhKrOAmZlVCbYi+35aBXsAJnlrvZcH+ekyDJV/Sy4idISkZZAD6AJsBh4DegWzEyBiMhIYBnwevI8eK9JfWUEVzHgJmDTOY5zOk8BdXFOfnq5x+0WoCqQm1ND6L3aWzwZWIj3h/t/KSKt3Kt6ICKX4HwP1QhuLOMF1gNvTIgTke+BpsDi5B5CEdkQSr3aXuOuSF9FVb9xRzeEq+ph97EaqroxuAlDh4gUwOkx3KCqv4vI+UBNrw0JFZHFqto42DmyKtgjRNxr1Q8FKuP8QB6X/APUi0TkZZysvteE/kNV7wleqrREZAbO3Pc5Xl7ITkRuxTkZUhdnwcplOL3wnwc1mI8AV0hIXsRuMTDaS/uriMwDWqmqlxuaIfWbQ0RWquq1wc5xOiLSBmfEVRuc0UxvA71U9adg5jLeYA14Y0KciKxW1at9h/iKyHovDVUVkeWqWl9EDuM/l9hzwwFF5HbgDqC4qlZyr8H7qnrnuuUmB4jIGKAo8AHOkG8AVPWHdDfyoGAP9ReRuTjziJcCbYHCqtonWHlOR0Q2ATWSezdFJBfOyabqwU2WlohEAVe6N9eoamww82RERErhjBAYABRTVU9eZz0jInK9qn4d5Axv4qzHMAf/4d6euoyciEwFnlHVn4Od5XTcY/024Ev869RrC1ciIh1xGvGFgU6q+ntwExmvsCH0xoS+jSLSEwhzG5v3ASuDnMmPqtZ3/xsKP+LuAa7CmR+H22scGdxI5hxI7pEZ6XOf4oxuCSXBPitfSlUfdf89T0S8fgLkV6Aczg96gLLA+uDFCUxEuuJMlVqMc+LzBREZqKofBTVYKiLyGlANiMHpfe/CqUUXQ80EIKgNeOAv9/953P97VX3gZnd0wwm8veZFT/e/Q3zu88zClSLyAv7H8SLAn8C9IuK5ufomOKwBb0zouxd4FOdLczowDxgV1ESppDPnMIXHznyfUNWTzqWsQUTCCX6jyOQwVW0S7AzZJNhrNIiIFPPJEeZ722OfdYASwGYRWePevhJY5a5IjofWEhkGXJnc6y4iEcA3gKca8Dj1GQYcBPYDez16ubvMCPZnye967+7okEJeGuLvo2WwA2SWqlYIdobTWJvq9vdBSWE8zRrwxoS+Nm6PV3KvV3JvTdAX3fLhe8mmcsAB99/nAdsBL32hLhGRoUB+EbkeuBtnqJ35D3OHJ48FSqtqKxGpBtRVVc9cAitEFtsrivN59238JPfAeqaXy8fjwQ6QSblSDZnfh8cWgwRQ1WhIWXCrBbBIRMJUtUxwk52VoJ+4FZHpwF1AIs7nqqiIPK2qXlv8NRzYqaonRKQxUAtnzrbnuL+P5qrqYREZBlwOjPLK1RJU9a1gZzDeZ3PgjQlxgRatCvZCVukRkVeBL1R1tnu7FdBMVR8ObrJT3F6OvjjXXhWcEQ2veXwFYJNFIjIH59JCj6pqbXfkxY9eW5jpP7TYXnVV9cyK3yJSBJ9ODa+NFBCRiTiNIt/F9tar6iPBS5WWe6nDBjiXCS0GrMJZ2X9aUIOdBS98jyZf1k5EegFXAI8A33ttaLqI/ATUAcrjfGd+AVysqq2DGCug5DWCRKQ+MA5naspQVb06yNH8iEg9YARwIc6xKXlagtdOgpogsB54Y0KU2/htDVwgIs/7PFQE8OqQxStV9a7kG6o6R0Q8NdzfXe13KjDVHfpfxhrv/xdKquqHIjIEQFUTRCQx2KECWCEiLxLii+0B7+D0fAWViNyBM+XoGM4lpQQPjhRQ1YE+l7sTYIqqfhrkWIG0wlnA8DlV/TvYYbJoa7ADALlFJDfQEXhRVeNFxIvfR0nuMbMT8KyqviAinujRDiD5uN4GeEVVPxeREUHMk57XgQdxRl548bvIBJE14I0JXX/jzJVqj/8cqcM4B30v2usOWXsX50dyb5yhoJ4hIotx6jQc+AmIE5ElqvpQMHOZHHdURErgDpsVkWuAf4IbKaD/ymJ7QZ9f7BoIVFfVvcEOcjqq+gnevK52itNdfk9EVqlq3XOVJyPuJS4fBsqp6u3uIrAXq+osAFXtFNSAjsk4JxLWAUvdS5x6cQ58vIjcANwEtHPvyx3EPBnZJSKTgWbABBHJiwenowD/qOqcYIcw3mRD6I0JYe6c2LdVtVews2SG26M9HGd4peL01Iz00nDV5MtwichtQFlVHe61y/KZ7CcilwMvADWAjUAE0EVVPbci+X+BF4Ynuznm4lye6d9gZwlERPriXNJyont7J84oKwEGqeorwcx3poJ9mUNfIvIBzsnvm1S1hojkB1ap6qXBTZY+cVZXDfPawoDumiF34dTf+yJSAeiuquODHC0N98RNS5zLRf4uIucDNVV1vvt4MVU9ENSQTo7xOAtCfoL/5e5CbbSVyQHWgDcmxLk/QNur6slgZ8kqEXlBVe8NcoYNOPPf38KZD/2dNeD/P7jz3i/GaRz9qqrxQY6URigstpcZHmrAX4az9sFq/H8ke+JSTSLyHdBSVfe5t5NPMOYD5qtqw+AmPDNeed8BRGStqtbxPakgIutUtXaws5ng8co+KiKLAtytqhpqo61MDrAh9MaEvm0482K/wH9O7NPBi3TW6gU7AM7Q5HnACrfxXhH4PciZTA5x52wGcpF7zV2vDVl+E3exPff2bzjz4UOqAQ945YTjZGAhsAFnDrzX5EpuvLtmAqjqcbfH2Jy9k24dJk+bqYTPSRzzf8sT03v+Q5c2NTnAGvDGhL6/3f/nAgoHOUvIU9WZ+FyKS1X/BDoHL5HJYcnzNSNx5pcvdG83ARbjvTnHIbHYnjvUtxdQUVVHikg5oJSqrgFQ1WuCGvCUBI+vb1HU94aqjoWUq2WUCEqirPFE48g1HJgLlBWR93BOIPcJaiLjBUEdmiwiqY9HCuwFlqvqX0GIZDzIGvDGhDhVfQJARAqq6tHTlTcZE5EyOHOh6+F8cS4H7lfVnUENZnKEqt4CICKzgGqqutu9fT7wUjCzpSNUFtt7GadHuynOqJbDwMfAlcEMFcAidyX6L/EfQu+VdTnmi8hoVR2W6v6RwPxgBEqPuybLPFVtlkGxG89VntNR1a9F5AfgGpwTC/d7bTFDEVmLM+JmuhfmZQfivu/jVXVgsLP8RwTqiCkPPCoiI1R1xjnOYzzIGvDGhDgRqYszfLYQUE5EagN3qurdwU12VrzQO/MGMB3o6t7u7d53fdASmXOhfHLj3RUDXBSsMBl4COcay5VEZAXuYnvBjRTQ1ap6efKlpFT1gIjkCXaoAHq6/x3ic5+XLiM3EHhNRLbgrEQOUBvnCiS3BS1VAKqaKCL/ikhRVQ14UklVN57rXKfRCOfSfIqzavr/2rvzKEur+tzj36cREZBmMAheExRwIswgaitOKCpTKyIggVxEHHC4qFxxSNRoOwFKFIkgGkQEQWWQKcogMgg2IA0IKnpRwUSiIsoUGWR47h/7Pd2nTp2qahBq71P1fNaq1fW+Vb3Ws7prOL937/37tTaa77XAXsAP+4r5s1sabdr9v28uSS3l+itUfR3SW5QZ1DUB/i6QAj7SxC5i1Em6lPIC/rS+Rjw/tr1B3WTjSdq526I+9J6k19n+SpVwS/JcNdiFeNi9mFm62epPBY6nvJh/LfCL2k0VhxmRZnuXUo4k/LAr5FenFB5NdCAfNV0vjvW7y5/a/uXAx9e3/ZPpTzaWpG9SVrTPYWxPliaaAvaTdBjwFMr3PMCuwC+nGoVXQ3dkYnvgcMrOli8Dh7SyU0TSwZSfnycw9v+9tSNIvV4Hv7F9j6QXARtRpvnc2n18tVb+XQe1NMUh6soKfMQMYPu/ypHTxZo7E9t5P33nywfv1S7eOzdL2oMlL+p2o7FZ9fHws/32rqHd87tbX7TdzGrcCDbb+xxlNXMNSR+nPGQc3AZeXTdSaj/KLPA3Dc4Cb0XXi+NXk3zKMUD1ztnAf3Rvo+CFwAa9VWNJR1OaGTZF0kaUVfhtKcdQvkbZNfA9YJN6ycZYjfJ7sr9DummvhwiUf8NnSnoKZffiaZRdd9tCU8dnxpC0FdDkMYqYfingI0bff0l6LuBui+q+wLWVM40haRvKL8cnSvpc34fmAk3NswVeD/wb8BnKC5AfdPdihuuK4BZfcMKINduz/TVJi4CXdLdeZbupn0udoyizwJ/bXf+G8kCxqQJ+KbRw/AjbR3ed3dey/fPaeabwc2AtyiQXgL8Drq4XZ7zue+hWSqH5Ptu9Pg2XSmphaguwpJfIiHiga/65I/BZ24f2jvq0oBtlO7g9ejVKs+L/Pf2JokUp4CNG3z7AIcATgRspI9Ba2wL435Qzm/MpL5Z77gDeVSXRBGz/JyVnzCLdCveBlAJZ3Zttz60arDOCzfYAVgCWobwYbXXk2bq2d5W0G4DtuzSwnWlENHEeUtIOwKeBRwNrS9oEWGC7xZ+pjwOulXRZd70FsLAbyUojmXfudl+MY3uiXTnTTtLTKNv717C9QbdrYL7tj1WONsy93ff7nix5MLpsxTyDth+4NvDHwSbFklZttbFhPPJSwEeMuK5r7u61c0zG9o8k/Rh4me2ja+eZTLeN8h195+FWBQ62nVX4me0gYIdGV4n7jUSzPUkfojSCPInyMOQoSSc0+II+s8AfXh8GnkXZFYLtqyStXTPQJD5UO8BE+keJDXueZPtfpzXQ1L5Eabh4BIDtqyUdB7T2/Q7lOMI+wMdtX999fR5bOdNitn899WcBcC5tHJuJClLAR4y4rrnRIZTGQQYWAu+a6Kl9LV2n2sdJerTtv9TOM4mNesU7LO6enaYxM9/vR6B4Bzhf0lmMbbZ3Xt1IQ+0GbGr7bgBJBwBX0N4L+pkyC7yVn6n32b5toOhsYnfAEBsCX2t0FXPYKLGWrWD7soH/99aOxwFg+6eUo4a96+uBA+oleshGcadQPExSwEeMvuMoW2h37K5fS3lx/+xqiSb2a+Dibotif6fallYT5vRvTetGt+Rn5cx3uaRvAKcwdh54a2fLm2621+cG4DHA3d31csAvJ/zsSkZhFjhAt61/d2Ad2wskrQWsafsyANvPqRpwiR9L+gdgma4h4L6UPiItWpMynu0KSlf3s1oZg2b7I9189X1tf6Z2nqVwc7d7pbeT5TXAbyf/K3VIup4hD5VstzI6cmk18bUadWSMXMSIk3Sp7WcP3LukoRd0i0n6l2H3J5p7WoOk/03pjH9id2tnyla7Y+qlikeapKOG3HaOTjw0kk6hnCk+h/JCc2vgIuAmaHas2Idtf7h2jmEk9caHbWV7ve5oz9m2t6gcbYyuq/8/Ay/rbp0FfKy3E6M13YORl1G2VT8T+CZw5OCYvloknWf7xbVzTKXbCfhFSjPIW4Drgd0fxHbwaSPpcX2Xj6H8jl/NdrNHKoaRdIXtbKGfpVLAR4y4bmvqrcDXKS+Ud6Wsdn0e2h2J0jJJf08ZhyPg3G7LXUR1rTfb65G052Qfb7EXRssviHvZ+udAS/qR7Y1rZ+snaVPbzXT0XhqSNqYU8K+gHEd5DnCO7fdUDQZ0IxhXBr7B2F1rV1QLNYSkZbpjcisCc2zfUTvTgyHpIttb1s7xYGQm/OyWAj5ixHXbwSbilraFSVodeA+wPuXJNwC2t5rwL02zbmvqOF13+pihRqWLsqRfMALN9iRtD3zb9gO1swwjae3u7Gv/vWZfEEu6lLK6+cOukF+dsgLfVF5J5wFPoIzi+7rtn1SONCFJ+1I6kd8M/Dtwiu17Jc0BrrO9btWALP73HOSWfmcCSPpPSi+JbwDfa+UowjCS+h/SzaHsvHhLgw/D1gV+Y/seSS8CNgK+2tdgd7Us0MxeKeAjYtpIOpvyC/7dlC6wewJ/sP3eqsH6DMxgXR5YG/i57fXrpYpHmqQL6Loo961w/tj2BnWTjSXpYtvNzH+eiKRjgXmULvRHtfbAQdIi25tLOtf2S7p7cxp+4LA7ZXfV5sBXgNcAH7B9Qs1cw0haE9iFkncu8I3WHoQBSFpA2S4/bpu3pPVa+5ptWTfJYQdKD57NgDMoD3AuqhpsiIGHIvdRtvsfbPvnlSINJekqysOFJ1OOopwGPN32thVjRSNSwEeMOEmXUxrwHNffPb1FfS+ar7a9UXfvAtsvrJ1tIt3T+jfbfnPtLPHIkfRD21sMbFG+yvYmlaONIekQSvOtU2i42R6ApLmUbvR7UR6KHQUc38L2WklXUv4N3wCMaxLWWGNNACQ9A3hJd/m91gtMSRtSdlztavvRtfMMknSM7X+c6l5NktYAPgH8L9vbdMe75tk+snK0CXX9GQ6hnIFfpnaeQZLWGZzSM2xHTm19x2b2B+62fWjLu4Ries2pHSAi/mqvBZ5I6aL9dUkv17DBsW24t/vzt5K268az/W3NQFPpzho21SgqHhGj0kV5LnAnpfHWDt3b9lUTTcD27ZQV+K9TtlXvCFwh6f9UDVa8ltIh/1GUkV2Dby1aAViG8tpt+cpZhpK0nqQPS/oJ8G+UDvSt/owfs6uq6/q+eaUsE/kKZfX1f3XX/w94Z60wk5H0QkmHUcZFPoayC6NFJy7lvdrulbQbZafiGd29ZSvmiYZkBT5ihujO7W1POcf7AGVV/pCWzkh152K/D/wdcCilGPmI7dOqBusjab++yzmU7YCPs/3ySpFiGoxSF+VRIGk+ZeV9XeAY4GjbN3Vdyq+1/aSqATuStrH9ndo5piLpQ5Ru2SdRGhe+Cjihta3pki6hjDE9wfZ/184zjKT3A/9EeQhyZ+828BfKWMb318o2aIR2Bl0PXEXp4n+a7T9P/jemX7eDZX3gIMpxqZ65wP6tHZPrdlvsAyy0fbyktSm7WUZxZn08zFLAR8wAXcOt1wPbUJ7Wfw3YEvjH1n7Rt25g1N19lHnWJ7U6BikeHn0PbpanPLj5M3AbsMj2VbVyDRqhZntfBf7d9oVDPvYS2+dWiNWfYb/JPt7aFnpJ1wKb9n4OdWeOr7C9Xt1k43XZ1mrtTPEgSZ+crFiXtH7tJnySzgd2onTF30zSc4ADWzt2Jmlut+OmWZJeSXnwNZ9ynrznDsp5/R/UyBXxUDyqdoCI+OtIWkQZI/fvwHtt987FXiqpqWZXo1B8uKGZ9DGtntm9nUZZjdsd+CGwj6QTbB9UM1yfL9E12wOwfbWk44Bmvoc6vx0s3iUdaPu9tYv3Tqvb5CdyA2Vbcu9B4nJAE7PK+0naAfg08GhgbUmbAAtsz68abIilWGk/hrIDq6b9KD+T1pV0MbA6pYFha9aU9C3a/t1+KnCqpHm2F9bOM5VuV8O4VdaWJgtFPVmBjxhx3TarTYEn0fdQzvaCaqEm0HKnb0mnM+SXZU+LL0Dj4SPpLGAn2//TXT+Wci5yR8oq/N/XzNczQltqx81U729eGQ+OpFMovTjOofyc2hq4CLgJwPa+1cL16R4obwWc3/f1OZL/7600DJP0KODplAeLP7d97xR/Zdq1/Lt9kKTHAHszfpzt66uFGkLS4/ouH0M5QrOa7Q9VihQNyQp8xOj7V8oK/BX0daVu1Aq2LxvosXdfrTADPl07QFS1FuUMbM+9wJNs3yWppe+rppvtSXoL8FbKiuHVfR9aCbi4TqqJjcKuoM63uree8yvlmMp9tm9rt4/qg1J9hatrrLctZZTYo4CXSWruiAdt/24fdAzwM+DlwALKbqvmJjrY/uPArc9KughIAR8p4CNmgL+1/YraIZZSs8WH7QtqZ4iqjgMukXRqd70DcLykFYGf1os1ztsozfaeIelGumZ7dSONcRzwHeCTwPv67t/R31BT0qq2b5nucEOMypGEPwLfbnVOfZ8fS/oHYBlJTwX2pXSij4fmdMqxiWsozWlb1ezv9iGeYntnSa+0fXT3/X5W7VCDVEbY9syhHPEataM/8QjJFvqIESfpi8Chtq+pnWUqo9DpW9I1jF95uQ24HPjYkKfiMUNI2pzS/FHARbYvrxxpnFFptjeVYVvsK+UYlSMJxwLzKF3oj2p1Bnw3ZeCfKWMORSmMPjqKTUAlXWL7OZUzjMTxgwl+t+9h+4aauYaRdJntZ0m6kLJb6HfAZa2dLZd0Xt/lfZR/04Nbbw4Z0yMFfMSI6is0HwU8FfgVZQu9ALf4S1/ScpQGPE8GVgNup2Rt5ry+pIOA+ykriVDmRYtSJG1pe4da2SK61aL+ZnvbUZrtPYMyuquVZnuTauh88XeAt1P+7TbrVg73tr1N5WjjSJoL7EYZz2fgKOB423dUDTaiVPZ77w6sY3uBpLWANW1fVjnaYpIOBM61fXbtLEuj27E0p+WvSUlvoDwI2xD4CvBY4IO2j6iZa5CkdWz/auDe2ravr5Up2pECPmJESZp0jnJLq9o9ks5kyXn9+3v3bR9cK9MgSRfbft6we5Kusb1hrWwRo9JsbyoNrcA3vyuon6S/AfYA3kk5t/sU4HO2D62c6ygmPjNu23tPZ56lIelwyrb0rWyvJ2lV4GzbW1SOtpikHYFjKbtt7mXJA/q5VYMN6B7O78SSs/pAs810xxXBLRbGEzQCXWR781qZoh05Ax8xolp9gTmFUTiv/1hJz7Z9KYCkZ1Ge0EO7TXli9hiVZnsjoVvhemnrK4eS5lNW3telNOF6lu2bui3r1wJVC3jgjCH31qI8aFhmeqMstWd3uy6uBLB9i6RH1w414GDK0Ylr3PaK26l0R3lov5nuSYwfD3gi0ERhLOkZlA75K0t6dd+H5tLXNT9mtxTwETGdfiBpw8bP678B+HK3sglwB7B39wL/k/ViRQCj02xvKlXblHfzyq/uexD6f4GdJP0aeEdrq3GUo0efsX1h/03bd0qqPv7K9km997tdDf8EvAA4ADiyVq4p3Nt1ee81Xlud9hrFXQf8uPHiHUbg4fwIFcZPB7YHVqH8fO+5A3hjjUDRnmyhj4hH3Iie11+Z8jPy1oH7e9o+uk6qiJFptre37SMH7h1g+33d+6v1d6Wfbt2Iu+d0BfD2lHGcuwGbAjvbfnmtbMNIOtD2e6e6V5Ok9SgN7DYFPgUca7vZXUuSdgd2pazGHk15SPIB2ydUDdZH0leAdSiTHRavbLc2Rm4UmulKeiXwKmA+pYdIzx3A1203NS1B0jzbC2vniDalgI+IR9wontefSCtndyNa1jWHO9b217rrw4DlWjkLLelHtjfu3v8y8HPbB3bXzX2PT3AetpkO5ZJOoDRX/DTwTfp6nADUfFgzmW5V9iWUh2HnttbdX9K/DLtv+yPTnWUykn5K6cdwPe0/nB+JwljSY4C9KbsGFu8QsF19x03Uly30EfGIG6UCfSlU3fobMSJeDZwm6QFgG+BPtt9aOVM/dcdk7qQUcIf1fayZ7bSS3kIZdbVut2ugZyXg4jqphtqCssvq3ZTjCLDkZ6Upq8hNkfQc4Ce2P99dr9Tf/6S2bnv/U23vUTvLUph0aoOkVW3fMl1hprCjpJ8AdwFnAhsD77R9bN1Y4xwD/Ax4ObCAMjGhqQdMUU9W4CMiHoQWV+ciWiFptb7LlYBTKIXmh6Cdldju3Pg/UUZZ3tQ7vytpU+DTtl9SM19Pd5RnVUr/jff1feiO/n/LxgqkCUla3/ZPaueAMsoQ2Kx3vlzSHODyln6+d1MndrD9lyk/uWEt/d6UdJXtTboO/68C3gWc19uR04reqM3eThtJywJn2d6qdraoLyvwEREPTlbgIya2iLHjxHqz6rejoZVY21/uiqPHAz/q+9DvKN3egfoFp+3bKN29d5viU89lfGftFh1DOznV3xzO9gOSWntdfANwsaTTgD/3brZ2Bn4ptPR7c9nuz22B423/SWop3mL3dn/eKmkDys+mJ9eLEy1p7QdVRETrWtq2GtEU22t3K5nzbDf9vWL7RuDGgXu/Hfi0lgrOyTRZgQzRUs5fSdoXOLy7fiulwWpL/rt7m0PZ0TKqWtrue7qkn1G20L+1mz5wd+VMw3xR0qrAByhN9x4LfLBupGhFttBHRPSRtN+Q27cBi2xfNc1xIkaSpIW259XO8dfqbWOtnWMqLW1RnkxLOSU9HvgcsBWlwDyXchb6pqrBhpC0EqUp3P/UzvJQtPT/DuXICXC77fu7EZwr2f5d97GtbZ9TNyFIWntwnOWwezE7zakdICKiMc8E9gGe2L29CXgR8CVJ76mYK2KUnC1pJzW6N/VByCrHDGX7Jtuvtf1422vY/ofWindJG3Rn9X8M/ETSIknr1871EDT1c8D2Lbbv797/c6947xxYKdagk4bcO3HaU0STsoU+ImKsx1EaG/0PLB7jcyLwAsr53oMqZosYFfsBKwL3S7qLJWOl5taNNWM1VSBNoplmbN3W6TdSzhUvfj3c2JiuLwL72T4PQNKLgC8Bz62YaRxJ6wK/sX1Pl3Ej4Ku2b+0+pYmmkEup6vdSN9pwfWBlSa/u+9BcGpqQEXVlBT4iYqy1GPsi817gSbbvosy3jYgp2F7J9hzby9qe212PYvHeRMEpae8h9w7ou2yiQFKxh6QPdddrSXpW7+O2n1Mv3TinAisD3wX+o++tJSv2incA2+dTHoy15iTKw7qnAEcCawPH9T7YyvSJpVR7183Tge2BVYAd+t42ozxwisgKfETEgOOASySd2l3vABzfnZP7ab1YEaNF0nzKzhWA822fUTPPMN0W/92BdWwvkLQWsKbty6CpgvM1ku62/TUASYcBy/U+2FCBdBjwAOVc+QLgDkpxt0XNUBNYwfZ7a4eYwq8kfZDSTBFgD6DFM9AP2L6vG832WduHdlv/40GyfSpwqqR5thfWzhNtygp8REQf2x+lPOW+ldK8bh/bC7pzcrtXDRcxIrrV4XdQHnr9FHjHwIpxKw4D5rFkTNsdwOfrxZnQq4HXSdpN0leBv9getyrfgGfbfhtdV+9uNv2j60aa0BmStq0dYgqvB1YHTu7e/oa+MYcNuVfSbsCeQO9B3bKTfH7LbqgdoLOjpLmSlpV0rqSbJe1RO1S0IV3oIyL6SDoE+IbtH9TOEjGqJF0NbGL7ge56GeBK2xvVTTZWrzt2f7d5ST+yvXHtbACSVuu7XAk4hTLK8kPQ1Mo7AJIupZzP/mH377o6cHaLnfwl3UHZjv6X7q2ZPg2SjrH9j5LeYfuQ2nmmIunvKc1fF9o+XtLawK62m3toN3CuvOc24JqWmhhKusr2Jt2uhlcB7wLOa+VnU9SVLfQREWNdAXxA0tOAb1GK+csrZ4oYRasAvQJz5Yo5JnNv93DBsLix2QN1I42xiLFncgVs170ZWKdGqEl8jvJzcw1JHwdeQ5lj3RzbLc9V31zSk4DXdzsuxjRWa+3Bje2fAvv2XV8PNFe8d/am7Lrp9RZ4EXAJ8DRJC2wfM9FfnGa9HQzbAsfb/tPoD/WIh0sK+IiIPraPBo7uVr52Ag6UtJbtp1aOFjFKPgFcIel8SvHxAuD9VRMN13TBaXttSXOAebYvrp1nKra/JmkRS5rqvcr2tTUzTaSv/8Hatj8q6e+AJ/T6H1T2BeBMygOaRYwt4Jt7cCPpeoY0f7PdVM7OA8B6tn8PIGkN4HDg2cCFLOk3UNvpkn4G3AW8tXu4eHflTNGIbKGPiBii65y8K2Xr2k9t71A3UcTokHQMcB1wC/CfwKUDs5ab0Y1t6hWc32ux4JS00Pa82jmWhqTNgC0pBd3Ftq+oHGkoSYfTNdyzvZ6kVSnb/ZtpuCfpcNtvmeTjq3Z9BqqS9Li+y8cAOwOr2f5QpUgTknSN7Q37rkXZPr9B/1GaFnRfk7fbvr9rpLtS7+eopK1tn1M3YdSSJnYREX0kHSjpOkoH5R8Dm6d4j3jQjur+nA/8K/B5Se+omGcyKwDLUF4TLV85y0TOlrSTGt9D242POxpYjdJw7ShJzexoGNB8w73JivfOudMSZAq2/9j3dqPtz1ImEbTo+5LOkLSnpD0p4wQv7ArkW+tGG8v2Lbbv797/88BD0AMrxYoGZAU+IqKPpH0oY4/WYeyYpgurhYoYQd3Z8i2AF1MaXN1l+xl1U43VFZw7U77nRdlxc4Ltj9XMNaiv4dr9lC21zTRc6yfpWmBT23d318sDV9her26y8Uap4d5EWlkx7nZd9MwBngm8pcWGa91DsJ2A51G+jy4CTvKIFUSt/N9HHTkDHxEx1v3A94C/Ba4CngMspN3VhIjmSDqXUnAuBL4PbNFSh+c+uzG24DyA0siyqQK+8YZr/W6gbKHundVdDvhltTST6/U/eHyL/Q+WUitF58F9799HmVW/S6Usk+oK9RO7t1HWyv99VJACPiJirH0pq4aX2H5xdz72I5UzRYyaq4HNgQ0oI5pu7c5x31U31jg3MCIFp6T5lGaAAOfbPmOyz6/kHuAnks6hFBhbAxdJ+hyA7X0n+8vTaaDhnmi44d4I2Nv2r/pvdKPkmtONkTsQeDzl/73J3SwRk0kBHxEx1t2275aEpOVs/0zS02uHihgltt8FIOmxwF6UM/Fr0ncspREjUXB2OwO2AL7W3XqHpC1tv69irGG+1b31nF8px4S6CSM9NwHH93+stRFtU2ilJ8KJwGZD7m1eIctUDgJ2mAEPa26oHSDqSQEfETHWbyStApwCnCPpFuC/qyaKGDGS3g48n/IC/tfAlylb6VvTfMHZ2RbYxPYDAJKOBq4EWivg/wh8u5ezUYsoD2sErEWZlCBgFcrEhGZWjiWtC/zG9j2SXgRsBHzV9q3dp7xkgr86LbodausDK3cr2z1zKTtbWvT7USjeB/49e26jdMy/yfawj8cskQI+IqKP7R27dz8s6TxgZco83ohYestTus8vsn1f7TCTGIWCs2cVoLc6vHLFHJN5LXCIpJOAo1oslGyvDSDpC8Bptr/dXW8DvLRmtiFOAp4p6SnAkcBpwHGUBzo0sFvg6cD2lK/N/mktdwBvrBFoKVwu6RuUh/T39G7aPrlaouH2BuYB53XXLwIuAZ4maYHtVubVRwXpQh8RERGzkqRjKS+Smy04ASS9FjiAskNAlLPw77f99Zq5hpE0l9IccC/KSvdRwPG276gabICkRbY3H7h3ue1n1so0SNIVXYf8/SnHuw5tsfu4pHm2F9bOsTQkHTXktm2/ftrDTELS6cAbbP++u14DOBx4A3Ch7Q1q5ou6sgIfERERs5LtPfoKzqMktVpwbkc5hnALZZv3ewdmQjfD9u3dCvzywDuBHYH9JX3O9qFVw411czej/ljKg4Y9KDsyWnKvpN2APVmywr1sxTwTuVLS2yjb6RdvnW+tKAawvVftDEvpyb3ivXMT8DTbf5J0b61Q0YY5tQNERERE1GL7dsoK/NeBJ1AKzisk/Z+qwcbqrRrOpxxN+Lykd1TMM5Sk+ZK+RRnFuSzwLNvbABsD764abrzdgNUpPRBOoXQl361moCH2ouwQ+bjt67vO7sdWzjTMMZQmlS8HLqCMYW3pARiS3tP9eaikzw2+1c43xPclnSFpT0l7AqcCF0paEbi1brSoLVvoIyIiYlbqRrPtBaxLKUKOtn2TpBWAa20/qWrAPpKWoXSifzGwD3CX7WfUTTWWpK8C/277wiEfe4ntcyvEikdYb1u/pKttbyRpWeAs21vVztYjaQfbp3fF8Di2j57uTJORJGAn4HmUYzMXASc5hVuQLfQRERExe70G+MxgwWn7TknNbP+VdC6wIrCQ0s1/C9s31U011G8H/y0lHWj7va0U75I+a/ud3RnjccWQ7fkVYg0l6XqGZ1ynQpzJ9LZ03yppA+B3wJPrxRnP9unduxfYvqH/Y5K2mP5Ek+sK9RO7t4gxUsBHRETEbNV8wdm5mjKSbwPKKKlbJS20fVfdWONsDbx34N42Q+7V1Ove/emqKZZOf0O9xwA7A6tN8Lk1fVHSqsAHKJ3yHwt8sG6kCZ0kab7tGwEkvRD4N2DDurHG6sbIHUg52qHuzbbnVg0WTcgW+oiIiJiVel2+B+5dbXujWpkmI+mxlC3/7wbWtL1c5UgASHoL8FbKUYRf9H1oJeBi23tUCbaUuuLz72xfXTvLVCRdZHvL2jn6SVrb9vVT3WtBt9p+GKUp4GbAJ4AdbP9X1WADJP2CkqvJyRhRV1bgIyIiYlbpLzgl9RdtKwEX10k1MUlvB55PWYX/NaUj/ferhhrrOOA7wCeB9/Xdv6N/VrmkVW3fMt3hhpF0PqUp4KOAq4A/SLrA9n41c/WT1P9waQ5lRX6lSnEmcxKlGO53IuXrtSm2fyhpX+Bs4G5ga9t/qBxrmN+neI+JpICPiIiI2WbUCs7lKd3nF9m+r3aYQbZvo2ztn6qL+7mML/RqWbkbefcG4Cjb/zLwMKcFB/e9fx9wPbBLpSzjSHoGZXTcyt2W75659I2Ta8GQngcrUL5mj5TUVO+DzuWSvkGZkHBP76btk6slimakgI+IiIhZZdQKTtufqp3hYaLaAfo8StITKAXxP9cOM4G9bf+q/0Y3Sq4VTwe2B1ZhyZx6KCPk3lgj0CRGoedBv7nAncDL+u4ZSAEfKeAjIiIiJtBSwTkTtNR4aQFwFuWM/g8lrQNcVznToBNpeGu67VOBUyXNs72wdp7J2L6g976kNSgjGQEua3Gig+29ameIdqWAj4iIiBiupYIzHka2TwBO6Lv+FWXudnWjtDW9s6OknwB3AWcCGwPvtH1s3VjjSdoF+BRwPuUB3aGS9rfdxLg2Se+xfZCkQxk+QnDfCrGiMSngIyIiImI6NLOjQdLTgMOBNWxvIGkjYL7tj1WOBqO1NR3gZbbfI2lH4DeUcXfnAc0V8JTjElv0Vt0lrQ58l3bmrfca111eNUU0LQV8RERExHDNFJyjQNLeto8cuHeA7V6jwJdUiDWRLwH7A0cA2L5a0nFA9QJ+lLamd5bt/twWON72n6Rmv3XmDGyZ/yOlw38TbJ/evXuB7Rv6P9aNwIto5ws2IiIiYjpJ2nvIvQP6LlsqOEfBayTt3ruQdBiweu+6v8N/A1awfdnAvdY6/F8p6W2SDpP05d5b7VBDnC7pZ5Qxd+d2q9p3V840kTMlnSXpdZJeB/wHZSJFa06S9MTehaQXUsZHRqSAj4iIiFlrlArOUfBq4HWSdpP0VeAvtsc9JGnEzZLWpTtnLOk1wG/rRhrnGGBN4OXABcDfUrbRN6XbYTEPeKbteynd01/Z+7ikrWtlG2R7f+CLwEaUs/pftP2euqmG2gc4RdKakrYFDqHscIhAdvqzRERExOwjaXngNMrK1jbAn2y/s2qoESRptb7LlSizqy8GPgRtPgjpus5/EXgucAtlxvrutn9dNVgfSVfa3lTS1bY3krQscJbtrWpnezAkXWG7+jjGfpLm0neUuNGv0XmUIx53A9vZ/kPlSNGInIGPiIiIWWWg4HwDSwrOBZJWa/HFfOMWMbZjtoDtujcD69QINZmu6/xLJa1I2ZF6F7Ar0EwBD9zb/XmrpA2A3wFPrhfnIWvmQLykN1NGCN4FPEDJ1szXqKTTGfu9tAJwG3CkJGzPr5MsWpIV+IiIiJhVJF3P+IKzx7abeDE/SiTNAebZvrh2lsl0K69vA54InErpQP424N3Aj2y/cpK/Pq0kvQE4CdgQ+ArwWOCDto+omevBamkFXtJ1lK/Tm2tnGaY76z6h/nn2MXulgI+IiIhZZ1QKzlEiaaHtebVzTEbSqZQt8wspTQpXBR4NvMP2VRWjjSNpbdvXT3WvdY0V8GcCr7Z9Z+0sU5G0BtDrPH/ZQPf8mMVSwEdERMSsNAoF5yiR9BHgauBkN/oCU9I1tjfs3l8GuBlYy3ZzzeGGFb6SFtnevFamh0LSybZfXTsHgKRNgaOAS4F7evdt71st1BCSdgE+BZxP2SH0fGB/263Mq4+KcgY+IiIiZquzJe1EwwXniNkPWBG4X9JddOeLbc+tG2uM3rlybN8v6frWindJzwDWB1aW1F/4zgUeUyfVxAYy9twGXGP7plaK984RwPeAayhn4Fv1z8AWvVX3bjTfd4EU8JECPiIiImatUSg4R4btlWpnWAobS7q9e1/A8t11S//3Twe2B1YBdui7fwfwxhqBprA3ZYzced31i4BLgKdJWmD7mFrBhrjP9n61QyyFOQNb5v9Ixn9HJwV8REREzEojUnCOFEnzgRd0l+fbPqNmnkG2l6mdYSq2TwVOlTTP9sLaeZbCA8B6tn8Pi89uHw48G7iQMs++FedJehNwOmO30Lc2eeJMSWcBx3fXuwLfqZgnGpIz8BERETFrtV5wjhJJB1Cabn2tu7UbsMj2++qlGl2SDgI+Rhl5diawMfBO28dWDTagv69Ady3K9vkNerPsK8Ybo5tA0bO4CGpx8kR3vOd5lN0hF9r+VuVI0YgU8BERETErpeB8eEm6GtjE9gPd9TLAlbY3qptsNEm6yvYmknYEXgW8CzjP9sZ1k40l6TBgLeCE7tZOwG+A/YEzbL+4VrZBXXO4M23fLumDwGbAR21fUTnaUN3Yw8U7phvcKRAVZAt9REREzFbbMrbgPBq4EkgB/9CtAvSKjJUr5pgJlu3+3BY43vafyuJ2c95GKdp7q8VfBU7qGkM2U7x3PmD7m5K2BLYGDmbJdv9mSHozsICy++IBuh4NQHM7BWL6pYCPiIiI2WwVUnA+XD4BXCHpfErB8QLg/VUTjbbTJf2MUsS9tetEfnflTON0hfqJjEaH9Pu7P7cDvmD7VEkfrphnIu8G1rd9c+0g0Z5soY+IiIhZSdJrgQNYMmv5BcD7bX+9Zq5RJekY4DrgFuA/gUtt/65uqtEmaVXg9m7k3YrASr1/U0lb2z6nbsLFY+QOBB5P+T5qqaP/GJLOAG4EXgpsTnk4clmDxxLOBF5t+87aWaI9KeAjIiJiVkrB+fCStBWwJfB8ylbfqyjNtw6pmWumknSF7c0ayPELYAfb19bOMhVJKwCvoDTZu07SE4ANbZ9dOdoYkjYFjgIuZWy3/H2rhYpmpICPiIiIWSkF58Ova1y3BeXs8z7AXbafUTfVzNRKh3dJF9t+Xu0cM4mky4CLgGsoZ+ABsH10tVDRjBTwERERMWul4Hz4SDoXWBFYCHwfuMj2TXVTzVwNrcAfAqwJnMLY1eKTa2UadZJ+YPu5tXNEm9LELiIiImalIQXnFik4/ypXU84VbwDcBtwqaaHtu+rGikfYXOBO4GV99wykgH/ozpP0JuB0xj4UyRi5yAp8REREzE6SPkMpOO8BLgYuBFJw/pUkPRbYi9JJe03by1WONCNJOtn2q2vniIefpOv7LhcXa7YzRi5SwEdERMTsloLz4SHp7ZR+ApsDv6Y8EPm+7e9VDTaiuu7ug26jNGCrvlNE0ntsHyTpUPqKzJ40XHvoJO0CnGn7dkkfBDYDPmr7isrRogHZQh8RERGz0pCC88uUrfTx0CwP/CuwyPZ9tcPMAHsD84DzuusXAZcAT5O0wPYxtYJ1el3nL6+aYmb6gO1vStoS2Bo4GDgceHbdWNGCFPARERExW6XgfBjZ/lTtDDPMA8B6tn8PIGkNlhRxFwJVC3jbp3fvXmD7hv6PSdpi+hPNKPd3f24HfMH2qZI+XDFPNGRO7QARERERNdj+lO1LU7xHo57cK947NwFP6xqZ3Vsp0zAnSXpi70LSCym7WeKhu1HSEcAuwLclLUfqtuhkBT4iIiIioj3fl3QGcEJ3vRNwoaQVgVurpRpvH+AUSTtQzmp/Ati2bqSRtwvwCuDTtm+V9ARg/8qZohFpYhcRERER0RhJohTtzwMEXASc5AZfvEuaBxwB3A1sZ/sPlSNFzFgp4CMiIiIi4kGRdDpju8//PfBb4BYA2/Nr5IqY6bKFPiIiIiKiMd0YuQOBx1NW4AXY9tyqwZb4dO0AEbNRVuAjIiIiIhoj6RfADravnfKTK+s65Pc6z1/Wwpz6iJkq3QwjIiIiItrz+xEp3ncBLgN2pjRfu1TSa+qmipi5sgIfEREREdEYSYcAawKnAPf07ts+uVamYST9CNi6t+ouaXXgu7Y3rpssYmbKGfiIiIiIiPbMBe4EXtZ3z0BTBTwwZ2DL/B/JLt+IR0wK+IiIiIiIxtjeq3aGpXSmpLOA47vrXYHvVMwTMaNlC31ERERERCMkvcf2QZIOZeyYNgBs71sh1qQk9c+rv9D2typHipixsgIfEREREdGOXuO6y6umeBBsnyTpHLraQtJqtv9UOVbEjJQV+IiIiIiIxkh6su0bBu5tYfuHlSINJenNwALgLuABlsyrX6dqsIgZKgV8RERERERjJC0C5tu+sbt+IfBvtjesm2wsSdcB82zfXDtLxGyQDpEREREREe3ZBzhF0pqStgUOAbatnGmYX1K65UfENMgKfEREREREgyTNA44A7ga2s/2HypHGkbQpcBRwKWPn1TfXbC9iJkgTu4iIiIiIRkg6nbHd51cAbgOOlITt+XWSTegI4HvANZQz8BHxCEoBHxERERHRjk/XDvAg3Wd7v9ohImaLbKGPiIiIiGiQpDWALbrLy2zfVDPPMJI+DvwaOJ2xW+gzRi7iEZACPiIiIiKiMZJ2AT4FnE8ZzfZ8YH/bJ9bMNUjS9UNuZ4xcxCMkBXxERERERGMk/QjYurfqLml14Lu2N66bLCJqyhi5iIiIiIj2zBnYMv9HGnztLmlnSSt1739A0sldZ/qIeAQ090MgIiIiIiI4U9JZkl4n6XXAfwDfqZxpmA/avkPSlsDLgaOBL1TOFDFjZQt9RERERESDJO0EPI9yBv5C29+qHGkcSVfa3lTSJ4FrbB/Xu1c7W8RMlAI+IiIiIqJRkubSN/q5te7uks4AbgReCmwO3EXpmJ+z+hGPgBTwERERERGNkfRmYAGlIH6AsgrfXHd3SSsAr6Csvl8n6QnAhrbPrhwtYkZKAR8RERER0RhJ1wHzbN9cO0tEtCNN7CIiIiIi2vNL4M7aISKiLVmBj4iIiIhoTDeK7SjgUuCe3n3b+1YLFRHVPWrqT4mIiIiIiGl2BPA94BrKGfiIiBTwERERERENus/2frVDRERbcgY+IiIiIqI950l6k6QnSFqt91Y7VETUlTPwERERERGNkXT9kNvNjZGLiOmVAj4iIiIiIiJiBGQLfUREREREYyTtLGml7v0PSDq560wfEbNYCviIiIiIiPZ80PYdkrYEXg4cDXyhcqaIqCwFfEREREREe+7v/twOONz2qcCjK+aJiAakgI+IiIiIaM+Nko4AdgG+LWk58to9YtZLE7uIiIiIiMZIWgF4BXCN7eskPQHY0PbZlaNFREUp4CMiIiIiIiJGQLbhRERERERERIyAFPARERERERERIyAFfERERERERMQISAEfERERERERMQJSwEdERERERESMgP8PkSYARFXaSjgAAAAASUVORK5CYII=\\n","text/plain":["<Figure size 1152x1008 with 2 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# remove features\\n","relevant_features = cor[\\"stroke\\"].copy()\\n","del relevant_features[\\"ever_married_No\\"]\\n","del relevant_features[\\"ever_married_Yes\\"]\\n","del relevant_features[\\"work_type_children\\"]\\n","del relevant_features[\\"Residence_type_Rural\\"]\\n","del relevant_features[\\"gender_Female\\"]\\n","\\n","# build corr matrix\\n","plt.figure(figsize=(16, 14))\\n","relevant_cor = hot_data[relevant_features.index].corr()\\n","relevant_cor = abs(relevant_cor)\\n","sns.heatmap(relevant_cor, annot=True, cmap=plt.cm.Blues)\\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"bbdBK1MBPtuK"},"source":["Now we update our hot_data to only use these selected features."]},{"cell_type":"code","execution_count":32,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":461},"id":"8hUS-EqZPs0R","outputId":"b3f7a614-4796-4245-978a-d4377faa9e4a"},"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>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Other</th>\\n","      <th>work_type_Private</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th>0</th>\\n","      <td>1.051331</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>2.706111</td>\\n","      <td>0.990957</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1</th>\\n","      <td>0.785993</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>2.121351</td>\\n","      <td>0.288746</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2</th>\\n","      <td>1.626231</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>-0.005028</td>\\n","      <td>0.460623</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3</th>\\n","      <td>0.255317</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.437217</td>\\n","      <td>0.706388</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4</th>\\n","      <td>1.582008</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.501038</td>\\n","      <td>-0.638849</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>...</th>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","      <td>...</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5105</th>\\n","      <td>1.626231</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>-0.494609</td>\\n","      <td>0.430013</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5106</th>\\n","      <td>1.670454</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.420734</td>\\n","      <td>1.430746</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5107</th>\\n","      <td>-0.363806</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>-0.511393</td>\\n","      <td>0.214859</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5108</th>\\n","      <td>0.343763</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.328127</td>\\n","      <td>-0.431890</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>5109</th>\\n","      <td>0.034201</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>-0.460822</td>\\n","      <td>-0.354280</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","<p>5002 rows \xd7 17 columns</p>\\n","</div>"],"text/plain":["           age  hypertension  heart_disease  avg_glucose_level       bmi  \\\\\\n","0     1.051331           0.0            1.0           2.706111  0.990957   \\n","1     0.785993           0.0            0.0           2.121351  0.288746   \\n","2     1.626231           0.0            1.0          -0.005028  0.460623   \\n","3     0.255317           0.0            0.0           1.437217  0.706388   \\n","4     1.582008           1.0            0.0           1.501038 -0.638849   \\n","...        ...           ...            ...                ...       ...   \\n","5105  1.626231           1.0            0.0          -0.494609  0.430013   \\n","5106  1.670454           0.0            0.0           0.420734  1.430746   \\n","5107 -0.363806           0.0            0.0          -0.511393  0.214859   \\n","5108  0.343763           0.0            0.0           1.328127 -0.431890   \\n","5109  0.034201           0.0            0.0          -0.460822 -0.354280   \\n","\\n","      stroke  gender_Male  gender_Other  work_type_Private  \\\\\\n","0        1.0          1.0           0.0                1.0   \\n","1        1.0          0.0           0.0                0.0   \\n","2        1.0          1.0           0.0                1.0   \\n","3        1.0          0.0           0.0                1.0   \\n","4        1.0          0.0           0.0                0.0   \\n","...      ...          ...           ...                ...   \\n","5105     0.0          0.0           0.0                1.0   \\n","5106     0.0          0.0           0.0                0.0   \\n","5107     0.0          0.0           0.0                0.0   \\n","5108     0.0          1.0           0.0                1.0   \\n","5109     0.0          0.0           0.0                0.0   \\n","\\n","      work_type_Self-employed  work_type_Govt_job  work_type_Never_worked  \\\\\\n","0                         0.0                 0.0                     0.0   \\n","1                         1.0                 0.0                     0.0   \\n","2                         0.0                 0.0                     0.0   \\n","3                         0.0                 0.0                     0.0   \\n","4                         1.0                 0.0                     0.0   \\n","...                       ...                 ...                     ...   \\n","5105                      0.0                 0.0                     0.0   \\n","5106                      1.0                 0.0                     0.0   \\n","5107                      1.0                 0.0                     0.0   \\n","5108                      0.0                 0.0                     0.0   \\n","5109                      0.0                 1.0                     0.0   \\n","\\n","      Residence_type_Urban  smoking_status_formerly smoked  \\\\\\n","0                      1.0                             1.0   \\n","1                      0.0                             0.0   \\n","2                      0.0                             0.0   \\n","3                      1.0                             0.0   \\n","4                      0.0                             0.0   \\n","...                    ...                             ...   \\n","5105                   1.0                             0.0   \\n","5106                   1.0                             0.0   \\n","5107                   0.0                             0.0   \\n","5108                   0.0                             1.0   \\n","5109                   1.0                             0.0   \\n","\\n","      smoking_status_never smoked  smoking_status_smokes  \\\\\\n","0                             0.0                    0.0   \\n","1                             1.0                    0.0   \\n","2                             1.0                    0.0   \\n","3                             0.0                    1.0   \\n","4                             1.0                    0.0   \\n","...                           ...                    ...   \\n","5105                          1.0                    0.0   \\n","5106                          1.0                    0.0   \\n","5107                          1.0                    0.0   \\n","5108                          0.0                    0.0   \\n","5109                          0.0                    0.0   \\n","\\n","      smoking_status_Unknown  \\n","0                        0.0  \\n","1                        0.0  \\n","2                        0.0  \\n","3                        0.0  \\n","4                        0.0  \\n","...                      ...  \\n","5105                     0.0  \\n","5106                     0.0  \\n","5107                     0.0  \\n","5108                     0.0  \\n","5109                     1.0  \\n","\\n","[5002 rows x 17 columns]"]},"execution_count":32,"metadata":{},"output_type":"execute_result"}],"source":["# update hot_data\\n","original_hot_data = hot_data.copy()\\n","hot_data = hot_data[relevant_features.index]\\n","hot_data"]},{"cell_type":"markdown","metadata":{"id":"ZeCbWZQNKN8g"},"source":["## Data Mining (Testing Different Models)"]},{"cell_type":"code","execution_count":33,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"lz7wGiUMKZAs","outputId":"17224bc2-8c7c-451a-e8ad-52b376f33217"},"outputs":[{"name":"stdout","output_type":"stream","text":["(5002, 16)\\n","(5002,)\\n"]}],"source":["# Separating labels from the rest of the data\\n","data_Y = data[\'stroke\']\\n","data_X = data.drop([\'stroke\'], axis=1)\\n","\\n","hot_data_Y = hot_data[\'stroke\']\\n","hot_data_X = hot_data.drop([\'stroke\'], axis=1)\\n","print(hot_data_X.shape)\\n","print(hot_data_Y.shape)"]},{"cell_type":"markdown","metadata":{"id":"cqrBxQsNM__D"},"source":["**Decision Trees**"]},{"cell_type":"code","execution_count":34,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"LiMHusykNEf-","outputId":"52a818f1-5f7c-46ba-fd21-9b17a873d39c"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Using criterion: Entropy\\n","Accuracy: 91.22343313373256\\n","\\n","Using criterion: Gini\\n","Accuracy: 90.94395209580838\\n","\\n","Using criterion: max_depth of 10\\n","Accuracy: 93.3424750499002\\n","\\n","Using criterion: max_depth of 1\\n","Accuracy: 95.22191616766466\\n","\\n","Using criterion: min_samples_split = 20\\n","Accuracy: 93.02287425149703\\n","\\n","Using criterion: min_samples_leaf = 10\\n","Accuracy: 92.2431137724551\\n","\\n","Using criterion: min_impurity_decrease = .9\\n","Accuracy: 95.22191616766466\\n"]}],"source":["from sklearn.model_selection import train_test_split\\n","from sklearn.tree import DecisionTreeClassifier\\n","from sklearn.metrics import accuracy_score \\n","from sklearn.model_selection import cross_val_score\\n","\\n","print(\\"\\\\nUsing criterion: Entropy\\")\\n","clf = DecisionTreeClassifier(criterion=\'entropy\')\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")\\n","\\n","print(\\"\\\\nUsing criterion: Gini\\")\\n","clf = DecisionTreeClassifier(criterion=\'gini\')\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")\\n","\\n","print(\\"\\\\nUsing criterion: max_depth of 10\\")\\n","clf = DecisionTreeClassifier(criterion=\'entropy\', max_depth=10)\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")\\n","\\n","print(\\"\\\\nUsing criterion: max_depth of 1\\")\\n","clf = DecisionTreeClassifier(criterion=\'entropy\', max_depth=1)\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")\\n","\\n","print(\\"\\\\nUsing criterion: min_samples_split = 20\\")\\n","clf = DecisionTreeClassifier(criterion=\'entropy\', min_samples_split=20)\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")\\n","\\n","print(\\"\\\\nUsing criterion: min_samples_leaf = 10\\")\\n","clf = DecisionTreeClassifier(criterion=\'entropy\', min_samples_split=10)\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")\\n","\\n","print(\\"\\\\nUsing criterion: min_impurity_decrease = .9\\")\\n","clf = DecisionTreeClassifier(criterion=\'entropy\', min_impurity_decrease=.9)\\n","scores = cross_val_score(clf, hot_data_X, hot_data_Y, cv=10)\\n","print(f\\"Accuracy: {scores.mean() * 100}\\")"]},{"cell_type":"markdown","metadata":{"id":"AN9Cs59VTXGV"},"source":["So far, the best accuracy was produced by decreasing the max depth or using a min_impurity_decrease. We can further try more combinations using a GridSearchCV."]},{"cell_type":"code","execution_count":35,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1ESEFA2gQmmy","outputId":"d78db3e4-7610-46b1-b260-d2bef1951c1c"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'max_depth\': 1, \'max_features\': 1, \'min_impurity_decrease\': 0.1, \'min_samples_leaf\': 1, \'min_samples_split\': 5}\\n","Accuracy: 95.22191808191808\\n"]}],"source":["from sklearn.model_selection import GridSearchCV\\n","from sklearn.metrics import classification_report\\n","\\n","params = {\\n","    \\"max_depth\\": [1, 3, 5, 10, 20], \\n","    \\"min_samples_split\\": [5, 10, 20], \\n","    \\"min_samples_leaf\\": [1, 3, 5, 10, 20], \\n","    \\"max_features\\": [1, 3, 5, 10, 15],\\n","    \\"min_impurity_decrease\\": [.1, .3, .5, .9]}\\n","\\n","grid_search = GridSearchCV(clf, params, cv=5, scoring=\\"accuracy\\")\\n","grid_search.fit(hot_data_X, hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":36,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"UVn6fXzTcr6w","outputId":"8070c71c-a97d-4e54-8773-73fc82f4899e"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.95      1.00      0.98      4763\\n","         1.0       0.00      0.00      0.00       239\\n","\\n","    accuracy                           0.95      5002\\n","   macro avg       0.48      0.50      0.49      5002\\n","weighted avg       0.91      0.95      0.93      5002\\n","\\n"]},{"name":"stderr","output_type":"stream","text":["/home/vineeth/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\\n","  _warn_prf(average, modifier, msg_start, len(result))\\n"]}],"source":["preds = grid_search.predict(hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"teFnv_yNeNUt"},"source":["The \\"best\\" decision tree simply chooses the \\"no stroke\\" label every almost every time, which produces a poor f-score. We can try to fix this by undersampling the majority class."]},{"cell_type":"code","execution_count":37,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":310},"id":"Srrqh-Dsgd--","outputId":"e33441a9-3a71-49ca-a7c9-cdab499a4bb3"},"outputs":[{"name":"stdout","output_type":"stream","text":["number of records with stroke: 239\\n","(478, 17)\\n"]},{"data":{"text/html":["<div>\\n","<style scoped>\\n","    .dataframe tbody tr th:only-of-type {\\n","        vertical-align: middle;\\n","    }\\n","\\n","    .dataframe tbody tr th {\\n","        vertical-align: top;\\n","    }\\n","\\n","    .dataframe thead th {\\n","        text-align: right;\\n","    }\\n","</style>\\n","<table border=\\"1\\" class=\\"dataframe\\">\\n","  <thead>\\n","    <tr style=\\"text-align: right;\\">\\n","      <th></th>\\n","      <th></th>\\n","      <th>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Other</th>\\n","      <th>work_type_Private</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","    <tr>\\n","      <th>stroke</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 rowspan=\\"5\\" valign=\\"top\\">0.0</th>\\n","      <th>932</th>\\n","      <td>-0.850259</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>-0.114339</td>\\n","      <td>-0.134385</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2784</th>\\n","      <td>0.211094</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.463354</td>\\n","      <td>0.783998</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1380</th>\\n","      <td>-1.469382</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.136745</td>\\n","      <td>-1.402013</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2867</th>\\n","      <td>1.095554</td>\\n","      <td>1.0</td>\\n","      <td>1.0</td>\\n","      <td>2.807914</td>\\n","      <td>0.680276</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>2806</th>\\n","      <td>0.564878</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>-0.530605</td>\\n","      <td>0.460623</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","</div>"],"text/plain":["                  age  hypertension  heart_disease  avg_glucose_level  \\\\\\n","stroke                                                                  \\n","0.0    932  -0.850259           0.0            0.0          -0.114339   \\n","       2784  0.211094           0.0            0.0           0.463354   \\n","       1380 -1.469382           0.0            0.0           0.136745   \\n","       2867  1.095554           1.0            1.0           2.807914   \\n","       2806  0.564878           0.0            0.0          -0.530605   \\n","\\n","                  bmi  stroke  gender_Male  gender_Other  work_type_Private  \\\\\\n","stroke                                                                        \\n","0.0    932  -0.134385     0.0          0.0           0.0                0.0   \\n","       2784  0.783998     0.0          1.0           0.0                1.0   \\n","       1380 -1.402013     0.0          0.0           0.0                0.0   \\n","       2867  0.680276     0.0          0.0           0.0                1.0   \\n","       2806  0.460623     0.0          0.0           0.0                1.0   \\n","\\n","             work_type_Self-employed  work_type_Govt_job  \\\\\\n","stroke                                                     \\n","0.0    932                       0.0                 1.0   \\n","       2784                      0.0                 0.0   \\n","       1380                      0.0                 0.0   \\n","       2867                      0.0                 0.0   \\n","       2806                      0.0                 0.0   \\n","\\n","             work_type_Never_worked  Residence_type_Urban  \\\\\\n","stroke                                                      \\n","0.0    932                      0.0                   0.0   \\n","       2784                     0.0                   0.0   \\n","       1380                     0.0                   1.0   \\n","       2867                     0.0                   0.0   \\n","       2806                     0.0                   1.0   \\n","\\n","             smoking_status_formerly smoked  smoking_status_never smoked  \\\\\\n","stroke                                                                     \\n","0.0    932                              0.0                          1.0   \\n","       2784                             0.0                          0.0   \\n","       1380                             0.0                          0.0   \\n","       2867                             0.0                          0.0   \\n","       2806                             0.0                          0.0   \\n","\\n","             smoking_status_smokes  smoking_status_Unknown  \\n","stroke                                                      \\n","0.0    932                     0.0                     0.0  \\n","       2784                    0.0                     1.0  \\n","       1380                    0.0                     1.0  \\n","       2867                    0.0                     1.0  \\n","       2806                    1.0                     0.0  "]},"execution_count":37,"metadata":{},"output_type":"execute_result"}],"source":["num_stroke = int(hot_data[\\"stroke\\"].sum())\\n","print(f\\"number of records with stroke: {num_stroke}\\")\\n","\\n","# sample equally\\n","balanced_hot_data = hot_data.groupby(\'stroke\').apply(lambda x: x.sample(num_stroke))\\n","balanced_hot_data_Y = balanced_hot_data[\'stroke\']\\n","balanced_hot_data_X = balanced_hot_data.drop([\'stroke\'], axis=1)\\n","print(balanced_hot_data.shape)\\n","balanced_hot_data.head()"]},{"cell_type":"code","execution_count":38,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8IGDfbZ7i1kG","outputId":"955384ff-2d77-4a33-a9c3-4be8ad5d8ed9"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'max_depth\': 1, \'max_features\': 10, \'min_impurity_decrease\': 0.1, \'min_samples_leaf\': 1, \'min_samples_split\': 10}\\n","Accuracy: 72.3859649122807\\n"]}],"source":["# Generating another Decision Tree using balanced_hot_data\\n","params = {\\n","    \\"max_depth\\": [1, 3, 5, 10, 20], \\n","    \\"min_samples_split\\": [5, 10, 20], \\n","    \\"min_samples_leaf\\": [1, 3, 5, 10, 20], \\n","    \\"max_features\\": [1, 3, 5, 10, 15],\\n","    \\"min_impurity_decrease\\": [.1, .3, .5, .9]}\\n","\\n","grid_search = GridSearchCV(clf, params, cv=5, scoring=\\"accuracy\\")\\n","grid_search.fit(balanced_hot_data_X, balanced_hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":39,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mFZlnAUSjI2O","outputId":"100608a6-4101-48af-9328-2370aa45c5de"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.50      1.00      0.67       239\\n","         1.0       0.00      0.00      0.00       239\\n","\\n","    accuracy                           0.50       478\\n","   macro avg       0.25      0.50      0.33       478\\n","weighted avg       0.25      0.50      0.33       478\\n","\\n"]},{"name":"stderr","output_type":"stream","text":["/home/vineeth/anaconda3/lib/python3.8/site-packages/sklearn/metrics/_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\\n","  _warn_prf(average, modifier, msg_start, len(result))\\n"]}],"source":["preds = grid_search.predict(balanced_hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(balanced_hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"G9joFb8Lj9HE"},"source":["The average f-score is greatly improved now! However, we only have a 75% accuracy."]},{"cell_type":"markdown","metadata":{"id":"TCMTHdNIkGEO"},"source":["**Naive Bayes**"]},{"cell_type":"code","execution_count":40,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2DXyRByekVi7","outputId":"2fc84131-5244-49f6-946a-eb6a068c3874"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.96      0.19      0.32       239\\n","         1.0       0.55      0.99      0.71       239\\n","\\n","    accuracy                           0.59       478\\n","   macro avg       0.75      0.59      0.51       478\\n","weighted avg       0.75      0.59      0.51       478\\n","\\n"]}],"source":["from sklearn.naive_bayes import GaussianNB\\n","# from sklearn.metrics import confusion_matrix\\n","from sklearn.metrics import classification_report\\n","from sklearn.model_selection import cross_val_predict\\n","\\n","# Use a gaussian Naive Bayes with 10-fold cross validation\\n","gnb = GaussianNB()\\n","preds = cross_val_predict(gnb, balanced_hot_data_X, balanced_hot_data_Y, cv=10)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(balanced_hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"Yk3eT67GlVrL"},"source":["Standard gaussian Naive Bayes produces a lower accuracy and f-score compared to the decision tree."]},{"cell_type":"markdown","metadata":{"id":"CkbD0oahlpqD"},"source":["**K-Nearest Neighbors**\\n","\\n","\\n","Since KNN suffers from the curse of dimensionality, we will have to perform dimensionality reduction. We use a GridSearchCV to perform hyperparamter tuning."]},{"cell_type":"code","execution_count":41,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"qxOhLpoynG9m","outputId":"b4c83445-4649-44de-8f26-b995de99a421"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'knn__n_neighbors\': 10, \'pca__n_components\': 6}\\n","Accuracy: 95.22191808191808\\n"]}],"source":["from sklearn.decomposition import PCA\\n","from sklearn.neighbors import KNeighborsClassifier\\n","from sklearn.pipeline import Pipeline\\n","\\n","pca = PCA()\\n","knn = KNeighborsClassifier(n_neighbors=7)\\n","pline = Pipeline(steps=[(\\"pca\\", pca), (\\"knn\\", knn)])\\n","\\n","param_grid = {\\n","    \'pca__n_components\': list(range(3, 15, 3)),\\n","    \'knn__n_neighbors\': list(range(1, 25))\\n","}\\n","\\n","grid_search = GridSearchCV(pline, param_grid, cv=5, scoring=\\"accuracy\\")\\n","grid_search.fit(hot_data_X, hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":42,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1q5mfnvgo3AG","outputId":"db4e4941-3966-4f27-eafb-1ae333b10634"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.95      1.00      0.98      4763\\n","         1.0       0.00      0.00      0.00       239\\n","\\n","    accuracy                           0.95      5002\\n","   macro avg       0.48      0.50      0.49      5002\\n","weighted avg       0.91      0.95      0.93      5002\\n","\\n"]}],"source":["preds = grid_search.predict(hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"4COvsis-o-QO"},"source":["Again, this produces a poor f-score so we will try again on the balanced one-hot-encoded dataset"]},{"cell_type":"code","execution_count":43,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"PwtMABWkpEMa","outputId":"f999f71d-6cc3-451d-b2cf-5ac3266f30b5"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'knn__n_neighbors\': 19, \'pca__n_components\': 9}\\n","Accuracy: 74.90789473684211\\n"]}],"source":["from sklearn.decomposition import PCA\\n","from sklearn.neighbors import KNeighborsClassifier\\n","from sklearn.pipeline import Pipeline\\n","\\n","grid_search.fit(balanced_hot_data_X, balanced_hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":44,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"YFCYe4wTpFis","outputId":"e6b27676-80b8-455b-f06f-f4dd99955635"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.81      0.66      0.73       239\\n","         1.0       0.71      0.84      0.77       239\\n","\\n","    accuracy                           0.75       478\\n","   macro avg       0.76      0.75      0.75       478\\n","weighted avg       0.76      0.75      0.75       478\\n","\\n"]}],"source":["preds = grid_search.predict(balanced_hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(balanced_hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"2lUu_uzNpcj8"},"source":["This model appears to be more successful than the one built using a decision tree."]},{"cell_type":"markdown","metadata":{"id":"dUS1oU2sqMkK"},"source":["**Support Vector Machines**\\n","\\n","We again use a GridSearchCV to perform hyperparamter tuning. We will also use the balanced one-hot-encoded dataset."]},{"cell_type":"code","execution_count":45,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"LIxtKf-VqQXa","outputId":"81c902cc-a3b4-4f5c-ad08-8f49773ff05d"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'pca__n_components\': 12, \'svc__kernel\': \'linear\'}\\n","Accuracy: 75.31140350877192\\n"]}],"source":["from sklearn.svm import SVC\\n","\\n","pca = PCA()\\n","svc = SVC()\\n","pline = Pipeline(steps=[(\\"pca\\", pca), (\\"svc\\", svc)])\\n","\\n","param_grid = {\\n","    \'pca__n_components\': list(range(3, 15, 3)),\\n","    \'svc__kernel\': [\'linear\', \'rbf\', \'poly\']\\n","}\\n","\\n","grid_search = GridSearchCV(pline, param_grid, cv=5, scoring=\\"accuracy\\")\\n","grid_search.fit(balanced_hot_data_X, balanced_hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":46,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GpnQV6oHrALX","outputId":"a8eb2c33-e967-444f-8f8a-70ef8534cb1a"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.82      0.70      0.76       239\\n","         1.0       0.74      0.84      0.79       239\\n","\\n","    accuracy                           0.77       478\\n","   macro avg       0.78      0.77      0.77       478\\n","weighted avg       0.78      0.77      0.77       478\\n","\\n"]}],"source":["preds = grid_search.predict(balanced_hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(balanced_hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"i80fUhAhVt8h"},"source":["This SVM model appears to perform even better than KNN, running faster and producing a higher f-score."]},{"cell_type":"markdown","metadata":{"id":"cgyh0ov4sHc-"},"source":["**Random Forests**"]},{"cell_type":"code","execution_count":47,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"fQC5ekYxsLQQ","outputId":"998acc85-cd44-452c-adb6-c6ef89f1a5ab"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'max_depth\': 60, \'max_features\': \'log2\', \'min_samples_leaf\': 31}\\n","Accuracy: 76.15350877192981\\n"]}],"source":["from sklearn.ensemble import RandomForestClassifier\\n","\\n","rfc = RandomForestClassifier()\\n","param_grid = {\\n","    \'max_depth\': list(range(5, 61, 5)),\\n","    \'min_samples_leaf\': [2, 31, 4],\\n","    \'max_features\': [\\"sqrt\\", \\"log2\\"]\\n","}\\n","\\n","grid_search = GridSearchCV(rfc, param_grid, cv=5, scoring=\\"accuracy\\")\\n","grid_search.fit(balanced_hot_data_X, balanced_hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":48,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1nFqYrPAshG0","outputId":"da7e013c-77bf-4fcd-9959-d1853fdae8e5"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.86      0.66      0.75       239\\n","         1.0       0.72      0.89      0.80       239\\n","\\n","    accuracy                           0.78       478\\n","   macro avg       0.79      0.78      0.77       478\\n","weighted avg       0.79      0.78      0.77       478\\n","\\n"]}],"source":["preds = grid_search.predict(balanced_hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(balanced_hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"MLfyXjxik9Li"},"source":["Since Random Forests don\'t suffer from the curse of dimensionality, let\'s try it with all features. Again, first we undersample the majority class."]},{"cell_type":"code","execution_count":49,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":310},"id":"F3PTIMUflK2c","outputId":"64cbabce-20f8-4a27-bc70-73bc16c15fb8"},"outputs":[{"name":"stdout","output_type":"stream","text":["number of records with stroke: 239\\n","(478, 22)\\n"]},{"data":{"text/html":["<div>\\n","<style scoped>\\n","    .dataframe tbody tr th:only-of-type {\\n","        vertical-align: middle;\\n","    }\\n","\\n","    .dataframe tbody tr th {\\n","        vertical-align: top;\\n","    }\\n","\\n","    .dataframe thead th {\\n","        text-align: right;\\n","    }\\n","</style>\\n","<table border=\\"1\\" class=\\"dataframe\\">\\n","  <thead>\\n","    <tr style=\\"text-align: right;\\">\\n","      <th></th>\\n","      <th></th>\\n","      <th>age</th>\\n","      <th>hypertension</th>\\n","      <th>heart_disease</th>\\n","      <th>avg_glucose_level</th>\\n","      <th>bmi</th>\\n","      <th>stroke</th>\\n","      <th>gender_Male</th>\\n","      <th>gender_Female</th>\\n","      <th>gender_Other</th>\\n","      <th>ever_married_Yes</th>\\n","      <th>...</th>\\n","      <th>work_type_Self-employed</th>\\n","      <th>work_type_Govt_job</th>\\n","      <th>work_type_children</th>\\n","      <th>work_type_Never_worked</th>\\n","      <th>Residence_type_Urban</th>\\n","      <th>Residence_type_Rural</th>\\n","      <th>smoking_status_formerly smoked</th>\\n","      <th>smoking_status_never smoked</th>\\n","      <th>smoking_status_smokes</th>\\n","      <th>smoking_status_Unknown</th>\\n","    </tr>\\n","    <tr>\\n","      <th>stroke</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","      <th></th>\\n","      <th></th>\\n","    </tr>\\n","  </thead>\\n","  <tbody>\\n","    <tr>\\n","      <th rowspan=\\"5\\" valign=\\"top\\">0.0</th>\\n","      <th>1152</th>\\n","      <td>0.122648</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.704943</td>\\n","      <td>-0.600045</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4411</th>\\n","      <td>0.609101</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.137408</td>\\n","      <td>-0.069711</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>4076</th>\\n","      <td>0.255317</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>-0.750111</td>\\n","      <td>0.240729</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>1382</th>\\n","      <td>1.095554</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>2.894258</td>\\n","      <td>-0.302540</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","    <tr>\\n","      <th>3452</th>\\n","      <td>0.476432</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>-1.090852</td>\\n","      <td>-0.263735</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>...</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","      <td>1.0</td>\\n","      <td>0.0</td>\\n","      <td>0.0</td>\\n","    </tr>\\n","  </tbody>\\n","</table>\\n","<p>5 rows \xd7 22 columns</p>\\n","</div>"],"text/plain":["                  age  hypertension  heart_disease  avg_glucose_level  \\\\\\n","stroke                                                                  \\n","0.0    1152  0.122648           0.0            0.0           0.704943   \\n","       4411  0.609101           0.0            1.0           0.137408   \\n","       4076  0.255317           0.0            0.0          -0.750111   \\n","       1382  1.095554           0.0            0.0           2.894258   \\n","       3452  0.476432           0.0            0.0          -1.090852   \\n","\\n","                  bmi  stroke  gender_Male  gender_Female  gender_Other  \\\\\\n","stroke                                                                    \\n","0.0    1152 -0.600045     0.0          1.0            0.0           0.0   \\n","       4411 -0.069711     0.0          1.0            0.0           0.0   \\n","       4076  0.240729     0.0          0.0            1.0           0.0   \\n","       1382 -0.302540     0.0          0.0            1.0           0.0   \\n","       3452 -0.263735     0.0          0.0            1.0           0.0   \\n","\\n","             ever_married_Yes  ...  work_type_Self-employed  \\\\\\n","stroke                         ...                            \\n","0.0    1152               1.0  ...                      0.0   \\n","       4411               1.0  ...                      1.0   \\n","       4076               1.0  ...                      0.0   \\n","       1382               1.0  ...                      0.0   \\n","       3452               1.0  ...                      1.0   \\n","\\n","             work_type_Govt_job  work_type_children  work_type_Never_worked  \\\\\\n","stroke                                                                        \\n","0.0    1152                 0.0                 0.0                     0.0   \\n","       4411                 0.0                 0.0                     0.0   \\n","       4076                 0.0                 0.0                     0.0   \\n","       1382                 0.0                 0.0                     0.0   \\n","       3452                 0.0                 0.0                     0.0   \\n","\\n","             Residence_type_Urban  Residence_type_Rural  \\\\\\n","stroke                                                    \\n","0.0    1152                   0.0                   1.0   \\n","       4411                   1.0                   0.0   \\n","       4076                   1.0                   0.0   \\n","       1382                   1.0                   0.0   \\n","       3452                   1.0                   0.0   \\n","\\n","             smoking_status_formerly smoked  smoking_status_never smoked  \\\\\\n","stroke                                                                     \\n","0.0    1152                             0.0                          1.0   \\n","       4411                             0.0                          1.0   \\n","       4076                             0.0                          0.0   \\n","       1382                             0.0                          0.0   \\n","       3452                             0.0                          1.0   \\n","\\n","             smoking_status_smokes  smoking_status_Unknown  \\n","stroke                                                      \\n","0.0    1152                    0.0                     0.0  \\n","       4411                    0.0                     0.0  \\n","       4076                    0.0                     1.0  \\n","       1382                    1.0                     0.0  \\n","       3452                    0.0                     0.0  \\n","\\n","[5 rows x 22 columns]"]},"execution_count":49,"metadata":{},"output_type":"execute_result"}],"source":["num_stroke = int(original_hot_data[\\"stroke\\"].sum())\\n","print(f\\"number of records with stroke: {num_stroke}\\")\\n","\\n","# sample equally\\n","balanced_original_hot_data = original_hot_data.groupby(\'stroke\').apply(lambda x: x.sample(num_stroke))\\n","balanced_original_hot_data_Y = balanced_original_hot_data[\'stroke\']\\n","balanced_original_hot_data_X = balanced_original_hot_data.drop([\'stroke\'], axis=1)\\n","print(balanced_original_hot_data.shape)\\n","balanced_original_hot_data.head()"]},{"cell_type":"code","execution_count":55,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"hAsW47NRlVh2","outputId":"6863474c-f539-4b17-ce31-d4648eb80404"},"outputs":[{"name":"stdout","output_type":"stream","text":["{\'max_depth\': 20, \'max_features\': \'sqrt\', \'min_samples_leaf\': 2}\\n","Accuracy: 77.00877192982456\\n"]}],"source":["rfc = RandomForestClassifier()\\n","param_grid = {\\n","    \'max_depth\': list(range(5, 61, 5)),\\n","    \'min_samples_leaf\': [2, 31, 4],\\n","    \'max_features\': [\\"sqrt\\", \\"log2\\"]\\n","}\\n","\\n","grid_search = GridSearchCV(rfc, param_grid, cv=5, scoring=\\"accuracy\\")\\n","grid_search.fit(balanced_original_hot_data_X, balanced_original_hot_data_Y)\\n","print(grid_search.best_params_)\\n","print(f\\"Accuracy: {grid_search.best_score_*100}\\")"]},{"cell_type":"code","execution_count":56,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"35X5kMgTlcFm","outputId":"ba967a3e-226d-4a8e-c741-cd16a4fa608a"},"outputs":[{"name":"stdout","output_type":"stream","text":["\\n","Classification Report:\\n","              precision    recall  f1-score   support\\n","\\n","         0.0       0.94      0.91      0.92       239\\n","         1.0       0.91      0.94      0.93       239\\n","\\n","    accuracy                           0.92       478\\n","   macro avg       0.92      0.92      0.92       478\\n","weighted avg       0.92      0.92      0.92       478\\n","\\n"]}],"source":["preds = grid_search.predict(balanced_original_hot_data_X)\\n","print(\\"\\\\nClassification Report:\\")\\n","print(classification_report(balanced_original_hot_data_Y, preds))"]},{"cell_type":"markdown","metadata":{"id":"ZWKc8nGOt8bz"},"source":["Random Forests produce the best f-score and accuracy by far, with an f-score of 0.92!"]}],"metadata":{"colab":{"collapsed_sections":[],"name":"project-notebook.ipynb","provenance":[]},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.8.5"}},"nbformat":4,"nbformat_minor":1}')},147:function(t,e,n){},787:function(t,e,n){"use strict";n.r(e);var a=n(0),d=n.n(a),s=n(136),r=n.n(s),o=(n(147),n(1));function l(){return Object(o.jsx)("section",{id:"about",children:Object(o.jsxs)("div",{className:"container mx-auto flex px-10 py-20 md:flex-row flex-col items-center",children:[Object(o.jsxs)("div",{className:"lg:flex-grow md:w-1/2 lg:pr-24 md:pr-16 flex flex-col md:items-start md:text-left mb-16 md:mb-0 items-center text-center",children:[Object(o.jsxs)("h1",{className:"title-font sm:text-4xl text-3xl mb-4 font-medium text-white",children:["Hello, I'm Nalin.",Object(o.jsx)("br",{className:"lg:inline-block"}),"An aspiring Software Engineer"]}),Object(o.jsx)("p",{className:"mb-8 leading-relaxed",children:"As a CS student at UT who is interested in solving the problems of today and tomorrow in the digital sphere, I am willing to quickly learn on the fly and also be proactive by taking full advantage of my free time to gain new skills."}),Object(o.jsxs)("div",{className:"flex justify-center",children:[Object(o.jsx)("a",{href:"#contact",className:"inline-flex text-white bg-green-500 border-0 py-2 px-6 focus:outline-none hover:bg-green-600 rounded text-lg",children:"Contact"}),Object(o.jsx)("a",{href:"#projects",className:"ml-4 inline-flex text-white bg-cyan-500 border-0 py-2 px-6 focus:outline-none hover:bg-gray-700 hover:text-white rounded text-lg",children:"Projects"}),Object(o.jsx)("a",{href:"./resume.pdf",className:"ml-4 inline-flex text-white bg-blue-500 border-0 py-2 px-6 focus:outline-none hover:bg-blue-600 rounded text-lg",children:"Resume"})]})]}),Object(o.jsx)("div",{className:"lg:max-w-lg lg:w-full md:w-1/2 w-5/6",children:Object(o.jsx)("img",{className:"object-cover object-center rounded",alt:"hero",src:"./coding.svg"})})]})})}var i=n(13),c=n(142),h=n(5),f=(n(149),n(83));function u(){var t=d.a.useState(""),e=Object(i.default)(t,2),n=(e[0],e[1]),a=d.a.useState(""),s=Object(i.default)(a,2),r=(s[0],s[1]),l=d.a.useState(""),u=Object(i.default)(l,2),p=(u[0],u[1]);return Object(o.jsx)("section",{id:"contact",className:"text-gray-400 bg-gray-900 body-font",children:Object(o.jsxs)("div",{className:"container px-5 py-10 mx-auto lg:px-40 items-center",children:[Object(o.jsx)(h.e,{className:"mx-auto flex w-10 mb-4"}),Object(o.jsxs)("form",{name:"contact",onSubmit:function(t){t.preventDefault(),c.a.sendForm("service_dcipdbe","template_5jcmb3h",t.target,"user_sI6OHl2gPmtOoCyZiA0Tw").then((function(t){console.log(t.text),f.a.fire({icon:"success",title:"Message Sent Successfully"})}),(function(t){console.log(t.text),f.a.fire({icon:"error",title:"Something Went Wrong",text:t.text})})),t.target.reset()},className:"",children:[Object(o.jsx)("h2",{className:"text-white text-center sm:text-4xl text-3xl mb-1 font-medium title-font",children:"Contact Me"}),Object(o.jsx)("p",{className:"text-center leading-relaxed mb-5",children:"If you want to contact me just send an email I would love to talk. Hopefully, it won't get sent to junk JK :p."}),Object(o.jsxs)("div",{className:"relative mb-4",children:[Object(o.jsx)("label",{htmlFor:"name",className:"leading-7 text-sm text-gray-400",children:"Name"}),Object(o.jsx)("input",{type:"text",id:"from_name",name:"from_name",className:"w-full bg-gray-800 rounded border border-gray-700 focus:border-indigo-500 focus:ring-2 focus:ring-indigo-900 text-base outline-none text-gray-100 py-1 px-3 leading-8 transition-colors duration-200 ease-in-out",onChange:function(t){return n(t.target.value)}})]}),Object(o.jsxs)("div",{className:"relative mb-4",children:[Object(o.jsx)("label",{htmlFor:"email",className:"leading-7 text-sm text-gray-400",children:"Email"}),Object(o.jsx)("input",{type:"email",id:"reply_to",name:"reply_to",className:"w-full bg-gray-800 rounded border border-gray-700 focus:border-indigo-500 focus:ring-2 focus:ring-indigo-900 text-base outline-none text-gray-100 py-1 px-3 leading-8 transition-colors duration-200 ease-in-out",onChange:function(t){return r(t.target.value)}})]}),Object(o.jsxs)("div",{className:"relative mb-4",children:[Object(o.jsx)("label",{htmlFor:"message",className:"leading-7 text-sm text-gray-400",children:"Message"}),Object(o.jsx)("textarea",{id:"message",name:"message",className:"w-full bg-gray-800 rounded border border-gray-700 focus:border-indigo-500 focus:ring-2 focus:ring-indigo-900 h-32 text-base outline-none text-gray-100 py-1 px-3 resize-none leading-6 transition-colors duration-200 ease-in-out",onChange:function(t){return p(t.target.value)}})]}),Object(o.jsx)("button",{type:"submit",className:"text-white bg-indigo-500 border-0 py-2 px-6 focus:outline-none hover:bg-indigo-600 rounded text-lg",children:"Submit"})]})]})})}var p=[{title:"TuxKart Ice Hockey Image Agent",subtitle:"Showcase Image detection",description:"An agent that was trained using an FCN to mask out the puck and detect the center. This was combined with a hand made controller. As a result the agent successfully beat the competing agents. Scoring goals in almost every match allowing us to place top 10.",image:"./tuxkart.gif",link:"/projects/0"},{title:"Smoothed Particle Hydrodynamics Simulation",subtitle:"Showcase Fluid Simulation",description:"A class project to implement Smoothed Particle Hydrodynamics. This is a way of simulating fluids by representing them as discrete particles. Then calculating the force of pressure and gravity using the nearby kernel of particles. This demo is of a simplistic version of that.",image:"./sph.gif",link:"/projects/9"},{title:"CNN Fruit Classifier",subtitle:"Keras Showcase",description:"Exploration into CNN classification using the Keras API and fruit-360 dataset from Kaggle. In this project using the Tensorflow Keras API I create a series of CNN models and evaluated the best one on the Fruit-360 dataset from Kaggle ",image:"./fruitClassifier.png",link:"/projects/1"},{title:"Point Cloud Stream Compression",subtitle:"Showcase FRI Project",description:"An application that runs using the ROS framework that compresses a depth camera output. The color data is compressed using standard compression techniques. The depth data is compressed using a custom lz4. This is reconstructed into a point cloud.",image:"./compression_node.png",link:"/projects/2"},{title:"VR Hallway Experiment",subtitle:"Showcase BWI Project",description:"Using previous data extrapolate gaze correlation with movement and if possible try to train a virtual agent such that it can read the gaze cues of a person in the VR environment and clone the behavior of a normal human.",image:"./vr.gif",link:"/projects/5"},{title:"Pacman Q-Learning",subtitle:"Showcase Reinforcement Learning",description:"A reinforcement agent that learned to estimate the value of certain action state pairs by manipulating weights of hand coded features. This was then used to populate and expectimax tree. This resulted in a fifth place sumbission.",image:"./pacman.gif",link:"/projects/7"},{title:"Old Website",subtitle:"Showcase of Express and PugJS",description:"I have created this website from scratch. The website uses nodejs as a web server and serves content using an express middleware. In addition the contact system below uses the nodemailer library to send e-mails and Send-Grid is the email service provider.",image:"./website.jpg",link:"/projects/4"},{title:"Stroke Classification",subtitle:"Showcase Data Mining",description:"This is a python notebook showcasing a full pipeline of data cleaning, feature creation and classification. Primarily, utilizes scipy and covers a wide variety of models for classification. In this Notebook we perform classification using KNN, SVM and Trees.",image:"./data-mining.png",link:"/projects/3"},{title:"NBAToday Website",subtitle:"Showcase React, Flask and PostgreSQL",description:"A website designed to show interesting stats for the NBA. This contains a react frontend. The data is stored in a PostegreSQL database. This is distributed using a fully fledged and documented API served by a Flask Backend.",image:"./nba.png",link:"/projects/6"},{title:"Asynchronous Copy",subtitle:"Showcase Asynchronous I/O",description:"A final project to create a faster version of cp -r by using asynchronous IO. For this project I leveraged the original linux aio library. I then compared it to an version with the new kernel level io_uring library including several optimizations.",image:"./aio.png",link:"/projects/8"}],A=["JavaScript","React","Flask","Java","Python","C/C++","Pytorch","Tensorflow","SpringBoot","Tomcat/Apache Servers","Gradle","AWS","GCP","Nginx"];function m(){return Object(o.jsx)("section",{id:"projects",className:"text-gray-400 bg-gray-900 body-font",children:Object(o.jsxs)("div",{className:"container px-5 py-10 mx-auto text-center lg:px-40",children:[Object(o.jsxs)("div",{className:"flex flex-col w-full mb-20",children:[Object(o.jsx)(h.g,{className:"mx-auto inline-block w-10 mb-4"}),Object(o.jsx)("h1",{className:"sm:text-4xl text-3xl font-medium title-font mb-4 text-white",children:"Projects I've Created"}),Object(o.jsx)("p",{className:"lg:w-2/3 mx-auto leading-relaxed text-base",children:"A showcase of what I have accomplished either in my free time or in class. This is meant to demonstrate my skill with different types of technologies."})]}),Object(o.jsx)("div",{className:"flex flex-wrap -m-4",children:p.map((function(t){return Object(o.jsx)("a",{href:t.link,className:"sm:w-1/2 w-100 p-4",children:Object(o.jsxs)("div",{className:"flex relative",children:[Object(o.jsx)("img",{alt:"gallery",className:"absolute inset-0 w-full h-full object-strech object-center",src:t.image}),Object(o.jsxs)("div",{className:"px-8 py-10 relative z-10 w-full border-4 border-gray-800 bg-gray-900 opacity-0 hover:opacity-100",children:[Object(o.jsx)("h2",{className:"tracking-widest text-sm title-font font-medium text-green-400 mb-1",children:t.subtitle}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-3",children:t.title}),Object(o.jsx)("p",{className:"leading-relaxed",children:t.description})]})]})},t.image)}))})]})})}function b(){return Object(o.jsx)("section",{id:"skills",children:Object(o.jsxs)("div",{className:"container px-5 py-10 mx-auto bg-gray-900",children:[Object(o.jsxs)("div",{className:"text-center mb-20",children:[Object(o.jsx)(h.f,{className:"w-10 inline-block mb-4"}),Object(o.jsx)("h1",{className:"sm:text-4xl text-3xl font-medium title-font text-white mb-4",children:"Skills & Technologies"}),Object(o.jsx)("p",{className:"text-base leading-relaxed xl:w-2/4 lg:w-3/4 mx-auto",children:"I am constantly picking up new skills and working with new technologies. Here is a sample of some of the technologies I have learned to work with."})]}),Object(o.jsx)("div",{className:"flex flex-wrap lg:w-4/5 sm:mx-auto sm:mb-2 -mx-2",children:A.map((function(t){return Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.c,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsx)("span",{className:"title-font font-medium text-white",children:t})]})},t)}))})]})})}function v(){return Object(o.jsx)("section",{id:"experiences",children:Object(o.jsxs)("div",{className:"container px-5 py-10 mx-auto",children:[Object(o.jsxs)("div",{className:"text-center mb-10",children:[Object(o.jsx)(h.d,{className:"w-10 inline-block mb-4"}),Object(o.jsx)("h1",{className:"sm:text-4xl text-3xl font-medium title-font text-white mb-4",children:"Experience"})]}),Object(o.jsxs)("div",{className:"flex flex-wrap lg:w-4/5 sm:mx-auto sm:mb-2 -mx-2",children:[Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.a,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsxs)("span",{children:[Object(o.jsx)("div",{className:"title-font font-medium text-white",children:"Masters of Computer Science UT 2023"}),Object(o.jsx)("div",{children:"Currently in the process of obtaining my Masters degree by Spring 2023 in the integrated 5-year program at the University of Texas at Austin."})]})]})}),Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.a,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsxs)("span",{children:[Object(o.jsx)("div",{className:"title-font font-medium text-white",children:"Bachelors of Math and Computer Science UT 2022"}),Object(o.jsx)("div",{children:"Graduating as Double Major Math and Computer Science Spring 2022 at the University of Texas at Austin."})]})]})}),Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.h,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsxs)("span",{children:[Object(o.jsx)("div",{className:"title-font font-medium text-white",children:"SDE Intern Amazon 2022"}),Object(o.jsx)("div",{children:"Implemented and tested production ready and highly requested features for Application Load Balancer as an intern for the Elastic Load Balancing team. Utilized low level networking constructs with emphasis on TCP/TLS and the HTTP/2 protocols."})]})]})}),Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.h,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsxs)("span",{children:[Object(o.jsx)("div",{className:"title-font font-medium text-white",children:"Software Engineer Intern Trend Micro 2021"}),Object(o.jsx)("div",{children:"Spearheaded a PoC that shifted cloud services to a local on premise environment; utilizing Java EE, SpringBoot, Gradle, Tomcat Servers, and AWS."})]})]})}),Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.h,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsxs)("span",{children:[Object(o.jsx)("div",{className:"title-font font-medium text-white",children:"FRI Peer Mentor 2021"}),Object(o.jsx)("div",{children:"Encouraged and supported Freshmen College students to enter research as a part of the FRI program by providing guidance as a peer mentor."})]})]})}),Object(o.jsx)("div",{className:"p-2 sm:w-1/2 w-full",children:Object(o.jsxs)("div",{className:"bg-gray-800 rounded flex p-4 h-full items-center",children:[Object(o.jsx)(h.h,{className:"text-cyan-500 w-6 h-6 flex-shrink-0 mr-4"}),Object(o.jsxs)("span",{children:[Object(o.jsx)("div",{className:"title-font font-medium text-white",children:"Research Volunteer at UT BWI Lab 2019-2021"}),Object(o.jsx)("div",{children:"Engaged in ground breaking research as an undergraduate student conducted in the BWI Lab with a focus on the topic of HRI (Human-Robot Interaction)"})]})]})})]})]})})}function g(){return Object(o.jsx)("header",{className:"bg-gray-800 sticky top-0 z-10 w-full",children:Object(o.jsxs)("div",{className:"container mx-auto flex flex-wrap p-2 md:p-5 flex-col md:flex-row items-center",children:[Object(o.jsx)("div",{className:"title-font font-medium text-white mb-1 md:mb-0",children:Object(o.jsx)("a",{href:"/#about",className:"ml-3 text-xl",children:"Nalin Mahajan"})}),Object(o.jsxs)("nav",{className:"md:mr-auto md:ml-4 md:py-1 md:pl-4 md:border-l md:border-gray-700\tflex flex-wrap items-center text-base justify-center",children:[Object(o.jsx)("a",{href:"/",className:"mr-5 hover:text-white",children:"Home"}),Object(o.jsx)("a",{href:"/#experiences",className:"mr-5 hover:text-white",children:"Experience"}),Object(o.jsx)("a",{href:"/#projects",className:"mr-5 hover:text-white",children:"Projects"}),Object(o.jsx)("a",{href:"/#skills",className:"mr-5 hover:text-white",children:"Skills"}),Object(o.jsx)("a",{href:"https://github.com/nalin29",className:"mr-5 hover:text-white",children:"Github"}),Object(o.jsx)("a",{href:"https://www.linkedin.com/in/nalin-mahajan-b7b449183",className:"mr-5 hover:text-white",children:"Linkedin"})]}),Object(o.jsx)("div",{className:"hidden md:block",children:Object(o.jsxs)("a",{href:"/#contact",className:"inline-flex items-center bg-gray-800 border-0 py-1 px-3 focus:outline-none hover:bg-gray-700 rounded text-base mt-1 md:mt-0",children:["Contact Me",Object(o.jsx)(h.b,{className:"w-4 h-4 ml-1"})]})})]})})}function P(){return Object(o.jsxs)("div",{children:[Object(o.jsx)(g,{}),Object(o.jsx)(l,{}),Object(o.jsx)(v,{}),Object(o.jsx)(m,{}),Object(o.jsx)(b,{}),Object(o.jsx)(u,{})]})}function w(){return Object(o.jsx)("section",{id:"NoPage",children:Object(o.jsxs)("div",{className:"flex items-center justify-center w-screen h-screen bg-gray-800 flex-col",children:[Object(o.jsxs)("div",{children:[Object(o.jsxs)("p",{class:"text-5xl text-white md:text-7xl lg:text-9xl text-center",children:["404 ",Object(o.jsx)("br",{className:"lg:inline-block"})]})," ",Object(o.jsx)("p",{className:"mb-8 leading-relaxed text-center",children:"Sorry this Page Does Not Exist"})]}),Object(o.jsx)("div",{children:Object(o.jsx)("a",{className:"ml-4 inline-flex text-white bg-blue-500 border-0 py-2 px-6 focus:outline-none hover:bg-blue-600 rounded text-lg",href:"/",children:"Back To Home Page"})})]})})}var y=n(2),k=n(137),j=n.n(k),S=n(141),z=[{id:0,title:"TuxKart Ice Hockey Image Agent",body:Object(o.jsxs)("div",{className:"bg-gray-900",children:[Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Introduction"}),Object(o.jsxs)("p",{children:["For our final project in CS342, we were tasked with programming a machine learning model that could play ice hockey in the SuperTuxKart racing game. Our model needed to be able to win against various pre-written agents in 2v2 tournaments where points are earned by scoring the hockey puck into the other team\u2019s goal. After the game\u2019s time limit, the winner is the team with the most goals scored. We elected to work on this project by building out an image-based agent because we were more comfortable tuning the vision component of our agent. See the Project here:\xa0",Object(o.jsx)("a",{className:"mr-5 decoration-blue underline hover:text-white",href:"https://github.com/nalin29/nn_final",children:"https://github.com/nalin29/nn_final"})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Model"}),Object(o.jsx)("p",{children:"Our model is a fully convolutional network with 32 channels, 64 channels, and 128 channels along with up-convolutional channels in reverse channel size. We gathered our starting data for training the model by running the given agents against each other. This method gave us good results in the initial training, but when we wrote our controller, it struggled, and there were a lot of edge cases that we were having trouble with solving. To combat this, we did some dagger-style runs where we collected data from one run and trained a new model using it while getting rid of the older data. We continued this process 12 times before we thought our model was good enough. We further tried to improve our model by testing out a few different loss functions, and we saw the best results when using a binary cross-entropy loss function."}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Controller"}),Object(o.jsx)("p",{children:"For our controller, we decided to go with a hand-tuned one rather than one of the provided ones. Our controller detects the puck and estimates the location of its center point. To do this, we built off of previous code from the third homework to try and segment out the puck from the current image. After segmenting the puck, we utilized a similar peak detection process to that from the fifth homework. If the puck is on the screen, then we treat it similarly to the aim point from the fifth homework assignment. Once the puck has been located, we determine the angle to the other team\u2019s goal by taking the dot product and the arccosine. This math is easy to do because we have the location of both our players, and the enemy goal doesn\u2019t move. In the case that the angle is very large or very small, our controller aims at the center of the puck. Otherwise, we try to offset the aim by hitting slightly to the left or right of the puck depending on the angle. The amount that we hit the puck by is determined by an inverse function of the distance to the goal because as the players get closer to the goal, the angle becomes more extreme. After a bit of testing, we realized that we were able to get similar results to this by hard-coding a constant amount left or right when we are within a small enough threshold distance from the enemy goal. When the puck is not on the screen, the controller looks to see if it has seen the puck recently. In this case that it has, the controller assumes that the puck is in the last location that it was seen at and plays accordingly. When the controller hasn\u2019t seen the puck recently, it enters a \u201clost mode\u201d where the player reverses backward towards their own goal to try and widen the amount of the map that they can see. To do this, we calculate the angle that we have to travel in to get to our goal and go backward in that direction. One problem that we noticed with this is that once we reached our goal, the puck could be too far away to see. While in this \u201clost mode,\u201d our player kind of became a goalie in that it stayed around the goal. To fix this, we added a timer for the time we spent near the goal without seeing the puck. After the timer ran out, our player would travel towards the opponent's goal looking for the puck."}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Experimentation"}),Object(o.jsx)("p",{children:"While coming up with a solution, we tried experimenting with a few different strategies. One example of this is that we messed around with choosing different karts for our members. While testing them out in different trial runs, we found that choosing Wilber increased the probability of hitting the puck with our strategy because of his bigger hitbox. After trying different combinations, we concluded that having both our players use the Wilber kart led to the best results. Another thing we tried was having the two players on our team perform different functions. We realized that our \u201clost mode\u201d made the cars resemble goalies, and we thought that it would be interesting to see how our team would perform if we made one of the players act as a goalie for the entire game. To do this, we messed with the lost mode to make the goalie car immediately go back to its goal. This goalie car would then stay within a small range of the goal for the rest of the game trying to block the enemy cars from shooting the puck into its goal. The strategy seemed to work, but after some analysis, we found out that having both of our players on the offensive for the entire game led to the number of goals our team scored being much higher. We attempted to optimize the goalie strategy a little bit more before completely disregarding it by trying different karts for both positions. Surprisingly, having the larger Kart, Wilber, still seemed to outperform strategies with a mix of kart types, but the combination of Wilber as goalie and Tux as the offensive player was close. Because of these findings, we ditched the one-goalie strategy for the double offense one. Some additional things that we experimented with were using parameter search for the various player hyperparameters like target speed and drift threshold. This ended up giving us pretty suboptimal results, so we switched to manually tuning the values until we were satisfied with the performance. Something else that we tried was calculating the distance between a player and the puck by analyzing the size of the puck on the screen; the closer the puck, the larger the number of pixels it was. We did this by using the rectangular nature of the puck to make a bounding box with the maximum and minimum dimensions of the puck. This turned out to not be super useful, so we scrapped it in our final implementation."})]})},{id:1,title:"CNN Fruit Classifier",body:Object(o.jsxs)("div",{children:[Object(o.jsxs)("p",{className:" mb-8 leading-relaxed text-justify",children:["Exploration into CNN classification using the Keras API and fruit-360 dataset from Kaggle. In this project using the Tensorflow Keras API I create a series of CNN models and evaluated the best one on the Fruit-360 dataset from Kaggle which contains 131 fruits and vegetables to classify. In addition I implement the 3 best models into the pytorch framework. Ultimately the best model was a transfer learning model based on pretrained weights from ResNet50. The python notebook is located in my git repo:\xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://github.com/nalin29/Fruit-Classification-Project",children:"https://github.com/nalin29/Fruit-Classification-Project"})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Custom CNN Model"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"A custom model that was settled is a 4 layer convolutional network with 2 fully conntected layers. The leaky RELU Activation function was used and trained using dropout to regularize."}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"https://raw.githubusercontent.com/nalin29/Fruit-Classification-Project/master/Images/confusion_matrix1.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Confusion Matrix"})})]}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"https://github.com/nalin29/Fruit-Classification-Project/blob/master/Images/plot_data1.png?raw=true"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Loss/Accuracy per Epoch"})})]}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"https://github.com/nalin29/Fruit-Classification-Project/blob/master/Images/test_images1.png?raw=true"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Test Image Batch"})})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center pt-6",children:"Transfer Model"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"Transfer Learning is done on ResNet50 pretrained on Imagenet. The features computed by ResNet is average pooled and passed into two fully connected layers where softmax layer outputs predictions."}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"https://raw.githubusercontent.com/nalin29/Fruit-Classification-Project/master/Images/confusion_matrix2.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Confusion Matrix"})})]}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"https://github.com/nalin29/Fruit-Classification-Project/blob/master/Images/data_plot2.png?raw=true"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Loss/Accuracy per Epoch"})})]}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"https://raw.githubusercontent.com/nalin29/Fruit-Classification-Project/master/Images/test_images2.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Test Image Batch"})})]})]})},{id:2,title:"Point Cloud Stream Compression",body:Object(o.jsxs)("div",{className:"item-center",children:[Object(o.jsxs)("p",{className:" mb-8 leading-relaxed text-justify",children:["This was a FRI I research project done with a partner. The goal was to create a compression algorithim by which point cloud data could be compressed in real time, frame by frame, through an application run in ROS. The end result was a combination of lz4 and garden variety video compression techniques. First the point cloud data broadcast from a kinect camera is split into two channels. A 32 bit grayscale depth map and a 16 bit color image. The depthmap due to limitations could not be compressed through standard image compression. Thus, lz4 was adopted to compress frame by frame costructing an lz4 archive. For the color images opencv was simply applied with the ability to tweak codec. The end result was a successful compression of the point cloud where it could then be reconstructed on playack. To see the relevant code visit the git repo:\xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://github.com/nalin29/compress_point_cloud",children:"https://github.com/nalin29/compress_point_cloud"}),"In addition an informal white paper may be found: ",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"/proj2/FRI_Final_Paper.pdf",children:" here "})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Technology"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"The point cloud data worked on by the labis often generated by a Kinect style device which generates anRGB image and a depth map in addition to a point cloud. We wrote a node that subscribes to the rectified RGB image anddepth map sensor data. The ROS Node then utilizes OpenCV\u2019sVideoWriter  class  to  compress  the  RGB  image  using  a  userspecified  encoding  such  as  FFmpeg. Concurrently,  the  depthmap  is  compressed  using  LZ4  frame  compression  and  theoutput of both of these is stored locally in two separate files.Figure  1  shows  an  abstracted  visualization  of  this  process. We then wrote another node that accepts the two compressed files  and  generates  the  original  RGB  and  depth  frame  which is published through ROS. Another node then reconstructs thepoint  cloud  from  the  given  RGB  image  and  depth map. The node then broadcasts the reconstructed point cloud to a topic.This  simulates  rosbag  usage  where  the  rosbag  is  a recording of  selected  topics. Then,  when  ran,  it  publishes  the  recordeddata  to  those  same  topics  allowing  users  to  run simulations that  utilize  point  cloud  data.  Figure  2  shows  an  abstracted visualization of this process."}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"/proj2/compression_node.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Figure 1"})}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain",src:"/proj2/broadcast_node.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Figure 2"})})]})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Results"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"The  results  of  different  encoders  on  total  RGB image compression can be seen in Table I. Since we used a variety of  lossy  and  lossless  encoders,  some  of  the  results  heavilyfavor  the  lossy  encoders. All  of  the  encoders  were  testedusing  the  same  image  stream.  File  size  represents  the  sizeof  the  compressed  video  file. While  we  got  the  best  resultsfrom  H.264,  its  important  to  note  that  this  is  a  lossy  codec. It   is   still   uncertain   whether   a   lossy   compression   has   aninadvertent  effect  on  the  usage  of  the  reconstructed  pointcloud. If  the  integrity  of  the  color  portion  of  the  point  cloudneeds to be maintained, a lossless encoder should be used. The data  suggests  that,  for  the  purposes  of  lossless  compression, FFmpeg should be utilized. The lossy compression algorithms feature the highest com-pression  ratio\u2019s  by  a  substantial  margin, but  one  thing  to consider  is  whether  the  point  cloud  is  still  usable  for  the purpose that the original would be. At the time of writing, our tests on this were inconclusive. It also worth considering that the RGB  image is a relatively small portion  of  the  total  size of the original rosbag which means that savings here are not as significant, making the choice of lossy or lossless encoding only provide marginal savings. Table 2 shows the results of our depth map compression using LZ4. The other substantial part of our savings comes from compressing the depth map. In our tests, we achieved an average compression ratio of 4.4565, which is comparable to the FFmpeg encoder's compression ratio. While this isn't as drastic as the savings of lossy image compression, one thing to consider is that LZ4 is a lossless compression algorithm. Although, other lossless compression algorithms can achieve better compression ratios LZ4 is unique in that it has a low compression and decompression time which is important for being able to hit refresh rate targets. Further works can investigate other compatible compression algorithms for more saving or to support higher frame rates. Tables 3 and 4 show our aggregate results when considering FFmpeg and H.264 for RGB image compression and LZ4 for depth map compression, respectively. The original file size is the size of just the original point cloud data from the rosbag and the compressed size is the aggregate of data needed to reconstruct the point cloud. Compression ratio is calculated as before. This data shows that using a lossy RGB encoder, specifically H.264, we were able to achieve a 25-33 compression ratio which is a substantial decrease in required file size as a 6GB file now only required 192.8 MB. We hypothesize that this could scale up with larger data files providing even greater space efficiency. A lossless encoder only achieves a 14-19 compression ratio which is much lower than lossy compression. However, compared to the original file size, there is still substantial space savings as a 6GB file can be compressed into a 333 MB file."}),Object(o.jsxs)("div",{className:"place-items-center",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-cover object-fill mx-auto",src:"/proj2/table_1.jpg"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Table 1"})}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-cover mx-auto",src:"/proj2/table_2.jpg"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Table 2"})})]}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-cover mx-auto",src:"/proj2/table_3and4.jpg"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Table 3 and 4"})})]})]})]})},{id:3,title:"Stroke Database Classification",body:Object(o.jsx)("div",{className:"w-full flex-auto text-sm align-left whitespace-pre bg-white text-black pr-6",children:Object(o.jsx)(j.a,{rawIpynb:S})})},{id:4,title:"Previous Website",body:Object(o.jsxs)("div",{children:[Object(o.jsxs)("p",{className:" mb-8 leading-relaxed text-justify",children:["In order to learn more about web development and boost my online presence I have created this website from scratch. The website itself uses an html template that was provided copywrite free from HTML5UP which has then been converted to pugjs by me and then the CSS design has been tweaked slightly. The website uses nodejs as a web server and serves content using an express middleware. The website is dynamically rendered using pugjs. In addition the contact system below uses the nodemailer library to send e-mails and Send-Grid is the email service provider. Interested take a look at my repository:\xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://github.com/nalin29/website",children:"https://github.com/nalin29/website"})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Hosting"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"The website is currently hosting through the Google Cloud Platform. This website is run as a compute engine instance as part of an instance group. This instance group can be accessed through an https load balancer that will intercept ssl traffic and create an internal route to the instances while balancing the load on each instance. An http load balancer re-routes http requests to https. On the instance itself an nginx reverse proxy is utilized to direct traffic to the appropiate port."}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"https://cloud.google.com/load-balancing/images/ilb-l7-tiers.svg"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Example of GCP Setup"})})]})]})},{id:5,title:"VR Hallway Experiment",body:Object(o.jsxs)("div",{children:[Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"This an ongoing research project conducted by the BWI lab at UT. The goal of this experiment is to follow up on the results of a previous project that identified gaze as one of the earliest and most accurate indicators of where a person might move. The goal of this experiment is to identify and log the gaze cues when two people pass each other. Then using this data extrapolate gaze correlation with movement and if possible try to train a virtual agent such that it can read the gaze cues of a person in the VR environment and clone the behavior of a normal human."}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Technology"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:"The VR environment is built in UNITY. An HTC-VIVE is used as the primary vr set due to its integrated gaze tracking. The environment consists of a rigged model that uses the Sranipal Framework to move the eye models to the appropiate location given the gaze of a human. Then to provide full body Motion a full body solver provided throught the Final IK package is used since it provides an accurate represnetation of real life movements."}),Object(o.jsxs)("div",{className:"",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"/vr.gif"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Full Body Tracking Demo"})})]})]})},{id:6,title:"NBAToday Website PoC",body:Object(o.jsxs)("div",{children:[Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Introduction"}),Object(o.jsxs)("p",{className:" mb-8 leading-relaxed text-justify",children:["NBA fans want current information about games, teams, and players in a cohesive, concise, and friendly manner. But acquiring such information is often easier said than done. Hence, we are creating NBAToday, a user-friendly web application with a responsive design that presents a rich assortment of information over games, teams, and players in the NBA. In addition, we provide live news to let our users know of what's happening in the world of NBA right as they access the site. Through NBAToday, we aim to provide some of the most desired information to fans so they can better engage with the sport. See the repository at:\xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://gitlab.com/mmishra14/cs373-idb",children:"https://gitlab.com/mmishra14/cs373-idb"})]}),Object(o.jsxs)("div",{className:"p-4",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"/nba.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Home Page"})})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"API Design"}),Object(o.jsx)("p",{className:" mb-8 leading-relaxed text-justify",children:'Our RESTful API is designed to aggregate and forward data from each of our three source APIs (see Tools > Backend below) to the frontend in a way that best reflects the structure of the site. For the pages that require varying levels of detail about each pillar instance (of games, teams, or players), we specify endpoints that return all attributes for one or more instances. For pagination, we specify endpoints that return a requested "page" of instances, and additionally we allow the request to specify the size of each page for added flexibility. To the same end, we allow requests to instance endpoints to specify an attribute by which to sort the data, and a direction (ascending or descending) for this sorting; this is necessary because pagination limits the frontend to a single page of instance data at a time, and so the frontend would otherwise only be able to sort some limited number of the instances. The frontend will have little to no data processing or reformatting to do. The endpoints with the path variable :id get a single instance with that ID. The IDs are internally defined nonnegative integers, and they are not guaranteed to be contiguous or in any particular order. These endpoints also have no URL query parameters for sorting or pagination. The endpoints with no :id path variable will request multiple instances according to the sorting and pagination URL query parameters specified. Successful requests are met with status code 200. Bad requests are met with status code 400. There are no silent failures. Bad requests are any and all that include invalid paths, parameter keys, or parameter values. Detailed descriptions are given below:'}),Object(o.jsx)("h2",{className:"title-font text-lg font-small text-white mb-7 text-center",children:"GET /api/players"}),Object(o.jsx)("p",{children:"This api call returns a multitude of player objects. The response will be made up of a JSON body that contains a list of player objects. Note that this api call supports query parameters. The specifiable query parameters control the page number and page size as well as support filtering and sorting by specific player attributes."}),Object(o.jsx)("br",{}),Object(o.jsx)("h2",{className:"title-font text-lg font-small text-white mb-7 text-center",children:"GET /api/teams"}),Object(o.jsx)("p",{children:"This api call returns a multitude of team objects. The response will be made up of a JSON body that contains a list of team objects. Note that this api call supports query parameters. The specifiable query parameters control the page number and page size as well as support filtering and sorting by specific team attributes."}),Object(o.jsx)("br",{}),Object(o.jsx)("h2",{className:"title-font text-lg font-small text-white mb-7 text-center",children:"GET /api/games"}),Object(o.jsx)("p",{children:"This api call returns a multitude of game objects. The response will be made up of a JSON body that contains a list of game objects. Note that this api call supports query parameters. The specifiable query parameters control the page number and page size as well as support filtering and sorting by specific game attributes."}),Object(o.jsx)("br",{}),Object(o.jsx)("h2",{className:"title-font text-lg font-small text-white mb-7 text-center",children:"Searching"}),Object(o.jsx)("p",{children:"Searching can also be utilized on each pillar type (Teams, Players, Games), through a query parameter. Searching looks for matching strings within each model, as well as some of the relationships to see if a match is found. For instance, Players have team names shown but not stored in the database. Instead we access the relationship and search if the name of the team contains a query word. The only non-searchable relationship is the games fields for players and teams. This was done to primarily target search results, as searching for a team name in the player pillar, for instance, would result in all players showing up as almost all players have played against all other teams. Searching is done by executing two search types when interacting with a multi-word query: AND as well as OR search. The order of words and capitalization is not taken into account. In addition searching includes non-string fields."}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Frontend"}),Object(o.jsx)("p",{children:'From the perspective of the UI, we took a simple page-based approach in the frontend to display information, where each page would display a different set of information. As a result, the user can access the following pages: Home, Players, Games, Team, APIs and Sources, and About Us. From the perspective of coding, we took a component-based approach. Firstly, this approach was taken largely because of the fact that ReactJs itself is a component-based language. Furthermore, taking the component based approach enabled us to develop the frontend in a modularized fashion, which sped up development and also allowed common components to be reused, thus reducing code complexity and redundancy (a common issue found in classic HTML-based approaches). Coming to the stylistic choices taken for our UI, in our homepage, we decided to separate features of our website into two separate Navigation Toolbars (Navbar). The Navbar on the left side of the home page contains links to the Home, Players, Teams, and Games. For the top Navbar, we have links to the APIs and Sources and the About page. The reason we decided to have more than one Navigation Toolbar is that we wanted to keep the functionality of the website distinct from the more general information about the project and its group members. That way, users can completely focus on exploring NBAToday. In addition, the homepage also features a section that covers the latest news regarding the NBA, a list containing upcoming matches, and a compilation of featured NBA players. The homepage serves as a simple hub that contains all of the most important information that users can quickly look at and more detailed data and statistics can be pulled up through the links if the user decides to do so. Both the latest news and upcoming games sections of the homepage are created using text formatting and hyperlinking. The Carousel component was the most appropriate way for displaying multiple players because it allowed us to show images of different players and would auto-scroll to other players after a short period of time. In addition, the Carousel allows for users to manually look through the featured players by clicking on the arrows located to the left and right of the pictures. In the Players and Teams pages, we decided to use a grid (refer to Tools > Frontend below) of Cards (a react-bootstrap component). We decided to use the Card component to display information regarding the players and teams because it is the modern approach for displaying information regarding people and organizations. Furthermore, the the "More Info" button on the Card gave us a simple and elegant way to indicate to the user about how to access more information regarding a player or a team. In the Games page, we displayed the information about games using a table instead of cards. We took the classic table approach because it enabled us to condense the rich information about a Game into an elegant style, and overall present the rich information regarding games in a concise, organized, and eye-appealing manner to the end user. To hint to the user that more information regarding a game can be accessed about a game, we ensured that each row (containing information about a specific game) was highlighted while being hovered over. For the instance pages of Players, Games, and Teams, we adopted a media-first approach. By this, we mean that we ensured that we would display some sort of media along with the required attributes right as any instance page loaded. The overall organization of each instance page was arbitrary, with the goal of being eye-appealing and user friendly in mind; and responsiveness of these pages was once again supported through a grid layout (refer to Tools > Frontend below). APIs and Sources and About Us pages also used a Card-based approach to organize and display their information for the same reasons as outlined above, and their responsiveness was once again enabled through the grid layout. For the Provider Visualizations page, we simply rendered a bar graph and doughnut graph beneath each other. The bar graph provides details about the number of books by page count. The doughnut graph provides details about the number of books by alphabet. We chose these two types of graphs because they are visually appealing while being informative; and we rendered the graphs on top of one another to provide a more visually-appealing UI to the user. For the global search page, we logically split the page into 3 horizontal sections, where each section would display the search results for Players, Teams, and Games, in that order. We implemented this specific order because both Players and Teams use Cards to present their information, while Games uses a table. To avoid conflicts in presentation style, we chose to display Games at the bottom. At the bottom of the page, we have a pagination bar that enables the user to go through the different search results for global search. We chose to not have a pagination bar per pillar to reduce the clutter on the screen. As a result of only one pagination bar, we also include basic filtering options, where the user is able to filter by Players, Teams, and Games to further narrow their search.'}),Object(o.jsx)("br",{}),Object(o.jsx)("p",{children:'From the perspective of the UI, we took a simple page-based approach in the frontend to display information, where each page would display a different set of information. As a result, the user can access the following pages: Home, Players, Games, Team, APIs and Sources, and About Us. From the perspective of coding, we took a component-based approach. Firstly, this approach was taken largely because of the fact that ReactJs itself is a component-based language. Furthermore, taking the component based approach enabled us to develop the frontend in a modularized fashion, which sped up development and also allowed common components to be reused, thus reducing code complexity and redundancy (a common issue found in classic HTML-based approaches). Coming to the stylistic choices taken for our UI, in our homepage, we decided to separate features of our website into two separate Navigation Toolbars (Navbar). The Navbar on the left side of the home page contains links to the Home, Players, Teams, and Games. For the top Navbar, we have links to the APIs and Sources and the About page. The reason we decided to have more than one Navigation Toolbar is that we wanted to keep the functionality of the website distinct from the more general information about the project and its group members. That way, users can completely focus on exploring NBAToday. In addition, the homepage also features a section that covers the latest news regarding the NBA, a list containing upcoming matches, and a compilation of featured NBA players. The homepage serves as a simple hub that contains all of the most important information that users can quickly look at and more detailed data and statistics can be pulled up through the links if the user decides to do so. Both the latest news and upcoming games sections of the homepage are created using text formatting and hyperlinking. The Carousel component was the most appropriate way for displaying multiple players because it allowed us to show images of different players and would auto-scroll to other players after a short period of time. In addition, the Carousel allows for users to manually look through the featured players by clicking on the arrows located to the left and right of the pictures. In the Players and Teams pages, we decided to use a grid (refer to Tools > Frontend below) of Cards (a react-bootstrap component). We decided to use the Card component to display information regarding the players and teams because it is the modern approach for displaying information regarding people and organizations. Furthermore, the the "More Info" button on the Card gave us a simple and elegant way to indicate to the user about how to access more information regarding a player or a team. In the Games page, we displayed the information about games using a table instead of cards. We took the classic table approach because it enabled us to condense the rich information about a Game into an elegant style, and overall present the rich information regarding games in a concise, organized, and eye-appealing manner to the end user. To hint to the user that more information regarding a game can be accessed about a game, we ensured that each row (containing information about a specific game) was highlighted while being hovered over. For the instance pages of Players, Games, and Teams, we adopted a media-first approach. By this, we mean that we ensured that we would display some sort of media along with the required attributes right as any instance page loaded. The overall organization of each instance page was arbitrary, with the goal of being eye-appealing and user friendly in mind; and responsiveness of these pages was once again supported through a grid layout (refer to Tools > Frontend below). APIs and Sources and About Us pages also used a Card-based approach to organize and display their information for the same reasons as outlined above, and their responsiveness was once again enabled through the grid layout. For the Provider Visualizations page, we simply rendered a bar graph and doughnut graph beneath each other. The bar graph provides details about the number of books by page count. The doughnut graph provides details about the number of books by alphabet. We chose these two types of graphs because they are visually appealing while being informative; and we rendered the graphs on top of one another to provide a more visually-appealing UI to the user. For the global search page, we logically split the page into 3 horizontal sections, where each section would display the search results for Players, Teams, and Games, in that order. We implemented this specific order because both Players and Teams use Cards to present their information, while Games uses a table. To avoid conflicts in presentation style, we chose to display Games at the bottom. At the bottom of the page, we have a pagination bar that enables the user to go through the different search results for global search. We chose to not have a pagination bar per pillar to reduce the clutter on the screen. As a result of only one pagination bar, we also include basic filtering options, where the user is able to filter by Players, Teams, and Games to further narrow their search.'}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Backend"}),Object(o.jsx)("p",{children:"The Backend is built off of the three key tools Flask, SQL-alchemy and PostgreSQL. Flask provides the actual routing and server for executing and managing api calls. The flask application is defined as a module (backend>app). The api calls are defined in main.py. Flask is useful as being in Python means it is simpler to code and beginner friendly in addition powerful python libraries can be leveraged such as its integration with SQL-alchemy. SQL-Alchemy is the second essential tool for the backend. SQL-Alchemy provides a fast, intuitive a powerful way to manage and interact with our PostgreSQL database. Through the ability to define models and retrieve as python objects as well as execute queries with filters applied as well as built in pagination functions. PostgreSQL is the database that stores and manages the data used for the website. This database type integrates well with SQL-alchemy provides a good platform for a relational database and is well documented."}),Object(o.jsxs)("div",{className:"p-4",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"https://gitlab.com/mmishra14/cs373-idb/-/wikis/uploads/4d2e5941ae3ccd11af467ab191e3bdfa/uml.JPG"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"UML Diagram of Database Relationships"})})]}),Object(o.jsx)("br",{}),Object(o.jsxs)("p",{children:["The current design uses three APIS: \xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://stats.nba.com/stats/",children:"https://stats.nba.com/stats/,"}),Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://www.balldontlie.io",children:"https://www.balldontlie.io"}),"and \xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://newsdata.io/",children:"https://newsdata.io/"})]}),Object(o.jsx)("br",{}),Object(o.jsx)("p",{children:"Both NBA statistics APIs operate mostly with their internal IDs for players, teams, games, and more, which sometimes makes it difficult to enumerate data and to map between the APIs. The basic approach was to start with a known list of current team IDs for stats.nba.com, find their current rosters (list of player IDs), search for these player names on balldontlie.io to obtain the player IDs for balldontlie.io, and then search for all games played in the most recent season on balldontlie.io. Certain statistics, namely sums and averages of various scores for players and teams, are not made directly available, but require us to perform the aggregations on the data we enumerate with the methods described in the previous paragraph. Data was scraped through a programmatic but lengthy sequence. First since NBA-Stats is not well documented we had to experiment and find our own endpoints. These are defined in endpoints.py. Then by running scrape.py we are able to firstly, scrape team data and note down player data. During this time a sleep period is entered to prevent us from reaching rate limits. Then in a similar fashion the players and games data is scraped and saved to players.json, teams.json, games.json. Then certain stats such as total wins are generated and then saved again to the JSON. In addition since we have two disparate sources: NBA-stats and ball don't lie we preprocess by storing and matching the different ids for player, teams and games in apidefs.py."}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Hosting"}),Object(o.jsx)("p",{children:"Deployment is done through Google App Engine. This is a service offered by GCP that provides a flexible and hassle-free way to deploy complex web applications. Google App Engine is leveraged through the use of its services model. The services model allows specific routes for different services through a 'dispatch.yaml' file. In the case of our deployment we have two services default and backend. Default simply refers to the frontend which is set up to be routed to in all cases except for backend routes. The backend service is specified as serving routes: */api/*. For instance: https://nbatoday.xyz/api/players. The frontend service is specified as serving the default route (i.e all other routes). For instance https://nbatoday.xyz. The backend service makes use of a PostgreSQL database. This database is hosted using Google Cloud SQL allowing Google App Engine with the correct connection URL to interact with the database. To deploy make use of: make deploy_stack"}),Object(o.jsx)("br",{}),Object(o.jsx)("p",{children:"The database is deployed using Google Cloud SQL. This is setup as a PostgreSQL using the default setup. There are two databases nbatoday and test. The first database contains all the data for all the various api calls. The second database is a database that is used for the unit testing."}),Object(o.jsxs)("div",{className:"p-4",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"https://gitlab.com/mmishra14/cs373-idb/-/wikis/uploads/1fb114077fbc225e3e2a3d4ce54f94f9/nbatoday.png"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Hosting Diagram"})})]}),Object(o.jsx)("br",{}),Object(o.jsx)("h1",{className:"title-font text-lg font-medium text-white mb-7 text-center",children:"Domain Name"}),Object(o.jsx)("p",{children:"The current website domain: nbatoday.xyz is supplied by NameCheap Domain Name Service Provider. To link the domain to our webapplication running on Google App Engine a series of records are applied to the domain. Namely, the google txt verification record that verifies that the domain provided is controlled by the user. In addition to four A and four AAAA records that direct the domain to the required Google IP addresses as well as domain servers. Then a CNAME record is applied allowing users to access the same domain using www.nbatoday.xyz. Once these changes propagate the Google App Engine auto-generates and auto-renews SSL Certificates for the application. This allows for secure browsing over HTTPS with Google as the Certificate Authority. The rest of the interaction between users and the website is handled automatically by GCP including: load balancing, redirecting and protection."})]})},{id:7,title:"Pacman Q-Learning Agent",body:Object(o.jsxs)("div",{children:["This project was the optional extra credit for Artificial Intelligence survey course. The game is set up such that there are two sides of the map. The goal is to eat all pellets on the opponenets side of the map. Agents becomes pacman when crossing to the opponents side and Ghosts on their side. To acheive this I implemented from scratch a Q-learning agent. This agent used a variety of features that could be tweaked to estimate the value of next states. The agent was given rewards based on the number of pellets remaining, winning, eating ghosts and the distance from the closest pellet. Additionally, a negative reward was applied when losing, being eaten or being close to a ghost. Ghosts or oppponents can be directly detected when within 5 blocks. To improve this we implemented a particle filter then we could estimate the locations of the opponents using peaks. Then to choose action the pacmans would try to estimate the value of future states and using an expectimax tree with fixed finite horizon choose the best action. See the repository here: \xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://github.com/nalin29/AI_Contest",children:"https://github.com/nalin29/AI_Contest"}),Object(o.jsx)("br",{}),Object(o.jsxs)("div",{className:"p-4",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"/pacman.gif"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Gameplay"})})]}),"Q-Learning was run for several hundred rounds against the provided AI and across several maps. Then hand made tweaks were made to some of the features to improve performance. The result was a fifth place finish and a decent Q-learning reinforcement agent."]})},{id:8,title:"AIO File Copy",body:Object(o.jsxs)("div",{className:"w-full h-full",children:["The project was a final project for the Graduate course in Advanced Operating Systems. The project implements recursive file copy by utilizing asynchronous I/O. The goal is to be able to utilize the ability to execute many writes/reads in parallel and for multiple files in parallel thus reducing time. This was accomplished first by using the slower and older aio library in Glib which executes in user-space. Then we implemented the same idea using the new io_uring library which allows for zero-copy asynchrounous I/O using syscall. See the repository here: \xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://github.com/nalin29/copy_async",children:"https://github.com/nalin29/copy_async"}),Object(o.jsx)("br",{}),Object(o.jsx)("br",{}),Object(o.jsx)("div",{className:"h-screen",children:Object(o.jsx)("iframe",{src:"https://drive.google.com/viewerng/viewer?embedded=true&url=nalinmahajan.com/aio.pdf#toolbar=0&scrollbar=0",frameBorder:"0",scrolling:"auto",height:"85%",width:"100%"})})]})},{id:9,title:"Smoothed Particle Hydrodynamics",body:Object(o.jsxs)("div",{className:"w-full h-full",children:["This was a final project for the Graduate Physical Simulation class. In this demo I have implemented a basic verson of SPH. Fluid is represented by these particles and we calculate the necessary forces on the fluid such as pressure and gravity by taking a sample at each particle. We then weight the sample utilzing a kernel and then we can use the specified equation to calculate the forces of pressure, viscosity and gravity. Then we can apply time integraton to get the next time step See the repository here: \xa0",Object(o.jsx)("a",{className:"mr-5 underline hover:text-white",href:"https://github.com/nalin29/sph_final",children:"https://github.com/nalin29/sph_final"}),Object(o.jsx)("br",{}),Object(o.jsx)("br",{}),Object(o.jsxs)("div",{className:"p-4",children:[Object(o.jsx)("img",{alt:"gallery",className:"object-contain mx-auto",src:"/sph.gif"}),Object(o.jsx)("div",{className:"p-2",children:Object(o.jsx)("h1",{className:"title-font text-lg font-small text-white mb-1 text-center",children:"Demo"})})]}),Object(o.jsx)("div",{className:"h-screen",children:Object(o.jsx)("iframe",{src:"https://drive.google.com/viewerng/viewer?embedded=true&url=nalinmahajan.com/sph.pdf#toolbar=0&scrollbar=0",frameBorder:"0",scrolling:"auto",height:"85%",width:"100%"})})]})}];function X(){var t=Object(y.h)().id,e=z[t],n=e.title,a=e.body;return Object(o.jsx)("section",{id:"projects",className:"text-gray-200 min-h-screen w-full bg-gray-900",children:Object(o.jsxs)("div",{className:"container px-5 py-10 mx-auto lg:px-40",children:[Object(o.jsx)("h1",{className:"title-font text-center sm:text-4xl text-3xl mb-4 font-medium text-white",children:n}),Object(o.jsx)("div",{className:"sm:mx-auto p-6 bg-gray-900",children:a})]})})}function H(){var t=Object(y.h)().id;return"undefined"==typeof z[t]?Object(o.jsx)(w,{}):Object(o.jsxs)("div",{className:"bg-gray-900 w-full h-full",children:[Object(o.jsx)(g,{}),Object(o.jsx)(X,{})]})}var O=n(30);function x(){return Object(o.jsx)(O.a,{children:Object(o.jsx)("main",{className:"text-gray-400 bg-gray-900 body-font w-full h-full",children:Object(o.jsxs)(y.d,{children:[Object(o.jsx)(y.b,{path:"/",element:Object(o.jsx)(P,{})}),Object(o.jsx)(y.b,{path:"/home",element:Object(o.jsx)(y.a,{to:"/"})}),Object(o.jsx)(y.b,{path:"/index",element:Object(o.jsx)(y.a,{to:"/"})}),Object(o.jsx)(y.b,{path:"/projects/:id",element:Object(o.jsx)(H,{})}),Object(o.jsx)(y.b,{path:"*",element:Object(o.jsx)(w,{})})]})})})}r.a.render(Object(o.jsx)(d.a.StrictMode,{children:Object(o.jsx)(x,{})}),document.getElementById("root"))}},[[787,1,2]]]);
//# sourceMappingURL=main.3ee525fa.chunk.js.map