{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[]},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","source":["### Table of content\n","1. Problem statement\n","2. Importing data and necessary libraries\n","3. Statistical information\n","4. Exploratory Data Analysis"],"metadata":{"id":"P4iuYfow2xs_"}},{"cell_type":"markdown","source":["1. **Porblem statement**:\n","The problem at hand is to develop a machine learning-based fraud detection system for credit card transactions to mitigate the rising incidence of fraud cases associated with credit cards. As the most popular mode of payment for electronic commerce, credit card usage has surged, leading to an increase in fraudulent activities. The challenge is to accurately identify fraudulent transactions from an imbalanced dataset, using various algorithms such as logistic regression, Naive Bayes, random forest, and ensemble classifiers with boosting techniques. The project also involves conducting a comprehensive review of existing and proposed models for credit card fraud detection and performing a comparative analysis of these techniques.\n"],"metadata":{"id":"MVV1Pyge4Q4V"}},{"cell_type":"markdown","source":["2. **Importing data and necessary libraries**\n","\n","The link to the datasets is provided in the link https://www.kaggle.com/datasets/ealaxi/paysim1"],"metadata":{"id":"FYt0T--r4gxc"}},{"cell_type":"code","execution_count":30,"metadata":{"id":"0xMy0bdJxC3l","executionInfo":{"status":"ok","timestamp":1684413963822,"user_tz":-60,"elapsed":479,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"outputs":[],"source":["#Import libraries\n","import pandas as pd\n","import numpy as np\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","from sklearn.linear_model import LogisticRegression\n","from sklearn.naive_bayes import MultinomialNB\n","from sklearn.neighbors import KNeighborsClassifier\n","import xgboost as xgb\n","from sklearn.model_selection import train_test_split, GridSearchCV, StratifiedKFold\n","from sklearn.preprocessing import MinMaxScaler\n","from imblearn.over_sampling import SMOTE\n","from imblearn.pipeline import Pipeline as imbpipeline\n","from imblearn.pipeline import Pipeline\n","from sklearn.model_selection import StratifiedShuffleSplit\n","from imblearn.over_sampling import RandomOverSampler,SMOTE\n","splitter = StratifiedShuffleSplit(n_splits=10,random_state = 42,test_size=0.2)\n","from imblearn.under_sampling import RandomUnderSampler\n","from sklearn.model_selection import cross_validate\n","from sklearn.model_selection import RepeatedStratifiedKFold\n","from sklearn.model_selection import cross_val_score\n","from sklearn.compose import ColumnTransformer\n","from sklearn.pipeline import Pipeline\n","from sklearn.impute import SimpleImputer\n","from sklearn.pipeline import Pipeline\n","from sklearn.preprocessing import OneHotEncoder, StandardScaler\n","from sklearn.ensemble import RandomForestClassifier\n","import joblib\n","import os\n","%matplotlib inline"]},{"cell_type":"code","source":[],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"POqpmyP72996","executionInfo":{"status":"ok","timestamp":1684413420085,"user_tz":-60,"elapsed":9,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"274eb4f6-ed5e-435a-d0b6-877210dbc179"},"execution_count":24,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['\"]"]},"metadata":{},"execution_count":24}]},{"cell_type":"code","source":["models"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":36},"id":"QaehNvdM3HAN","executionInfo":{"status":"ok","timestamp":1684413383591,"user_tz":-60,"elapsed":18,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"78f4d5e0-5ec7-4de6-cf6f-356f7796f5f2"},"execution_count":23,"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":23}]},{"cell_type":"code","source":["# Set random seed for Python's random module\n","random.seed(42)\n","# Set random seed for NumPy\n","np.random.seed(42)"],"metadata":{"id":"wBFIo8jj26Sj"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["#Extract dataset to colab root folder\n","import zipfile\n","path_to_zip_file = \"/content/drive/MyDrive/Predictiong_cvss_client/Fraud_Credit_Card/Data/Fraud.csv.zip\"\n","directory_to_extract_to = \"/content\"\n","with zipfile.ZipFile(path_to_zip_file, 'r') as zip_ref:\n"," zip_ref.extractall(directory_to_extract_to)"],"metadata":{"id":"U_trndDN1HZz","executionInfo":{"status":"ok","timestamp":1684412543685,"user_tz":-60,"elapsed":10219,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":2,"outputs":[]},{"cell_type":"code","source":["from google.colab import drive\n","drive.mount('/content/drive')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"OnUDQFBsY1SN","executionInfo":{"status":"ok","timestamp":1684412533481,"user_tz":-60,"elapsed":35089,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"46a3a7e4-dc7f-43fd-8ebb-9929e4354866"},"execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}]},{"cell_type":"code","source":["#load Datasets\n","df = pd.read_csv(\"/content/Fraud.csv\")\n","df.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"0Zylx71d7uMp","executionInfo":{"status":"ok","timestamp":1684412655950,"user_tz":-60,"elapsed":17316,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"1c0073f9-7c73-4d6d-f8c5-59e6b08ba084"},"execution_count":4,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount nameOrig oldbalanceOrg newbalanceOrig \\\n","0 1 PAYMENT 9839.64 C1231006815 170136.0 160296.36 \n","1 1 PAYMENT 1864.28 C1666544295 21249.0 19384.72 \n","2 1 TRANSFER 181.00 C1305486145 181.0 0.00 \n","3 1 CASH_OUT 181.00 C840083671 181.0 0.00 \n","4 1 PAYMENT 11668.14 C2048537720 41554.0 29885.86 \n","\n"," nameDest oldbalanceDest newbalanceDest isFraud isFlaggedFraud \n","0 M1979787155 0.0 0.0 0 0 \n","1 M2044282225 0.0 0.0 0 0 \n","2 C553264065 0.0 0.0 1 0 \n","3 C38997010 21182.0 0.0 1 0 \n","4 M1230701703 0.0 0.0 0 0 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountnameOrigoldbalanceOrgnewbalanceOrignameDestoldbalanceDestnewbalanceDestisFraudisFlaggedFraud
01PAYMENT9839.64C1231006815170136.0160296.36M19797871550.00.000
11PAYMENT1864.28C166654429521249.019384.72M20442822250.00.000
21TRANSFER181.00C1305486145181.00.00C5532640650.00.010
31CASH_OUT181.00C840083671181.00.00C3899701021182.00.010
41PAYMENT11668.14C204853772041554.029885.86M12307017030.00.000
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":4}]},{"cell_type":"markdown","source":["**Datasets description**\n","Data for the case is available in CSV format having 6362620 rows and 10 columns.\n","\n","Content\n","Data for the case is available in CSV format having 6362620 rows and 10 columns.\n","\n","Data Dictionary:\n","\n","step - maps a unit of time in the real world. In this case 1 step is 1 hour of time. Total steps 744 (30 days simulation).\n","\n","type - CASH-IN, CASH-OUT, DEBIT, PAYMENT and TRANSFER.\n","\n","amount - amount of the transaction in local currency.\n","\n","nameOrig - customer who started the transaction\n","\n","oldbalanceOrg - initial balance before the transaction\n","\n","newbalanceOrig - new balance after the transaction\n","\n","nameDest - customer who is the recipient of the transaction\n","\n","oldbalanceDest - initial balance recipient before the transaction. Note that there is not information for customers that start with M (Merchants).\n","\n","newbalanceDest - new balance recipient after the transaction. Note that there is not information for customers that start with M (Merchants).\n","\n","isFraud - This is the transactions made by the fraudulent agents inside the simulation. In this specific dataset the fraudulent behavior of the agents aims to profit by taking control or customers accounts and try to empty the funds by transferring to another account and then cashing out of the system.\n","\n","isFlaggedFraud - The business model aims to control massive transfers from one account to another and flags illegal attempts. An illegal attempt in this dataset is an attempt to transfer more than 200.000 in a single transaction."],"metadata":{"id":"4FYORKbr-4M9"}},{"cell_type":"code","source":["df.shape"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"9LcJMdfh-cUt","executionInfo":{"status":"ok","timestamp":1671738788390,"user_tz":-60,"elapsed":431,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"f12355d4-223c-4247-fe17-8602f05733de"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(6362620, 11)"]},"metadata":{},"execution_count":5}]},{"cell_type":"code","source":["df.columns"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"9nubctwK-cHc","executionInfo":{"status":"ok","timestamp":1671738791527,"user_tz":-60,"elapsed":584,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"5915bb45-2949-43d8-8d66-816dbe02f548"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Index(['step', 'type', 'amount', 'nameOrig', 'oldbalanceOrg', 'newbalanceOrig',\n"," 'nameDest', 'oldbalanceDest', 'newbalanceDest', 'isFraud',\n"," 'isFlaggedFraud'],\n"," dtype='object')"]},"metadata":{},"execution_count":6}]},{"cell_type":"code","source":["df.info()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"FlEdIvIk-ky8","executionInfo":{"status":"ok","timestamp":1671738794623,"user_tz":-60,"elapsed":35,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"0901383a-5462-46d9-ba13-520a0f52113c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["\n","RangeIndex: 6362620 entries, 0 to 6362619\n","Data columns (total 11 columns):\n"," # Column Dtype \n","--- ------ ----- \n"," 0 step int64 \n"," 1 type object \n"," 2 amount float64\n"," 3 nameOrig object \n"," 4 oldbalanceOrg float64\n"," 5 newbalanceOrig float64\n"," 6 nameDest object \n"," 7 oldbalanceDest float64\n"," 8 newbalanceDest float64\n"," 9 isFraud int64 \n"," 10 isFlaggedFraud int64 \n","dtypes: float64(5), int64(3), object(3)\n","memory usage: 534.0+ MB\n"]}]},{"cell_type":"code","source":["df[\"isFlaggedFraud\"].unique()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8nMJ0LKt-kn1","executionInfo":{"status":"ok","timestamp":1671664687917,"user_tz":-60,"elapsed":446,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"0dc877b8-0d3d-4864-b233-8a122198da4f"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([0, 1])"]},"metadata":{},"execution_count":14}]},{"cell_type":"markdown","source":["### 3. Statistical Information"],"metadata":{"id":"FjHtp1X971hX"}},{"cell_type":"code","source":["df.describe()"],"metadata":{"id":"NPeQRqKe7uFw","colab":{"base_uri":"https://localhost:8080/","height":300},"executionInfo":{"status":"ok","timestamp":1671738820298,"user_tz":-60,"elapsed":2543,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"d1da548b-da44-43f8-ba5e-afa55e9019ad"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step amount oldbalanceOrg newbalanceOrig \\\n","count 6.362620e+06 6.362620e+06 6.362620e+06 6.362620e+06 \n","mean 2.433972e+02 1.798619e+05 8.338831e+05 8.551137e+05 \n","std 1.423320e+02 6.038582e+05 2.888243e+06 2.924049e+06 \n","min 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 \n","25% 1.560000e+02 1.338957e+04 0.000000e+00 0.000000e+00 \n","50% 2.390000e+02 7.487194e+04 1.420800e+04 0.000000e+00 \n","75% 3.350000e+02 2.087215e+05 1.073152e+05 1.442584e+05 \n","max 7.430000e+02 9.244552e+07 5.958504e+07 4.958504e+07 \n","\n"," oldbalanceDest newbalanceDest isFraud isFlaggedFraud \n","count 6.362620e+06 6.362620e+06 6.362620e+06 6.362620e+06 \n","mean 1.100702e+06 1.224996e+06 1.290820e-03 2.514687e-06 \n","std 3.399180e+06 3.674129e+06 3.590480e-02 1.585775e-03 \n","min 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 \n","25% 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 \n","50% 1.327057e+05 2.146614e+05 0.000000e+00 0.000000e+00 \n","75% 9.430367e+05 1.111909e+06 0.000000e+00 0.000000e+00 \n","max 3.560159e+08 3.561793e+08 1.000000e+00 1.000000e+00 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
stepamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudisFlaggedFraud
count6.362620e+066.362620e+066.362620e+066.362620e+066.362620e+066.362620e+066.362620e+066.362620e+06
mean2.433972e+021.798619e+058.338831e+058.551137e+051.100702e+061.224996e+061.290820e-032.514687e-06
std1.423320e+026.038582e+052.888243e+062.924049e+063.399180e+063.674129e+063.590480e-021.585775e-03
min1.000000e+000.000000e+000.000000e+000.000000e+000.000000e+000.000000e+000.000000e+000.000000e+00
25%1.560000e+021.338957e+040.000000e+000.000000e+000.000000e+000.000000e+000.000000e+000.000000e+00
50%2.390000e+027.487194e+041.420800e+040.000000e+001.327057e+052.146614e+050.000000e+000.000000e+00
75%3.350000e+022.087215e+051.073152e+051.442584e+059.430367e+051.111909e+060.000000e+000.000000e+00
max7.430000e+029.244552e+075.958504e+074.958504e+073.560159e+083.561793e+081.000000e+001.000000e+00
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":8}]},{"cell_type":"code","source":[],"metadata":{"id":"KzCEQyYz7t6k"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["#### 4. Exploratory Data Analysis\n","** Missing values\n","** Categorical features\n","** Numerical features\n","** Target variable analysis\n","** Boxplots and histogram"],"metadata":{"id":"JN17jHy_795g"}},{"cell_type":"markdown","source":["**Missing Value**"],"metadata":{"id":"TbQbVQKrZnw4"}},{"cell_type":"code","source":["df.isnull().sum()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1h5UThBZ9bSR","executionInfo":{"status":"ok","timestamp":1684180169669,"user_tz":-60,"elapsed":3521,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"1c1248d8-d7be-4587-fde0-afefab47523f"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["step 0\n","type 0\n","amount 0\n","nameOrig 0\n","oldbalanceOrg 0\n","newbalanceOrig 0\n","nameDest 0\n","oldbalanceDest 0\n","newbalanceDest 0\n","isFraud 0\n","isFlaggedFraud 0\n","dtype: int64"]},"metadata":{},"execution_count":5}]},{"cell_type":"markdown","source":["Comment: No missing values"],"metadata":{"id":"MbMn2Hkp9nzC"}},{"cell_type":"markdown","source":["**Numerical features**"],"metadata":{"id":"tniCjUPvZyCu"}},{"cell_type":"code","source":["#Print all numerical features\n","numFeatures = [f for f in df.columns if df[f].dtype != \"O\"]\n","numFeatures"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"nr30QdSlZ16Z","executionInfo":{"status":"ok","timestamp":1684189735918,"user_tz":-60,"elapsed":3,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"77b27618-6fb0-4177-d7c5-62558ba2a719"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['step',\n"," 'amount',\n"," 'oldbalanceOrg',\n"," 'newbalanceOrig',\n"," 'oldbalanceDest',\n"," 'newbalanceDest',\n"," 'isFraud',\n"," 'isFlaggedFraud']"]},"metadata":{},"execution_count":5}]},{"cell_type":"code","source":["for f in numFeatures: # Print data types of numerical features\n"," print(f,\":\",df[f].dtype)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2u3hyibJaIPa","executionInfo":{"status":"ok","timestamp":1684189737204,"user_tz":-60,"elapsed":5,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"d79a98a3-6ada-4b60-ce87-db7a48043e7c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["step : int64\n","amount : float64\n","oldbalanceOrg : float64\n","newbalanceOrig : float64\n","oldbalanceDest : float64\n","newbalanceDest : float64\n","isFraud : int64\n","isFlaggedFraud : int64\n"]}]},{"cell_type":"code","source":["#Continuos numerical features \n","contFeat = [f for f in numFeatures if df[f].dtype == \"float64\"]\n","contFeat"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"6VoRei6OaH9c","executionInfo":{"status":"ok","timestamp":1684189740045,"user_tz":-60,"elapsed":7,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"8069c8c0-529d-435b-85c5-d179f9836f33"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['amount',\n"," 'oldbalanceOrg',\n"," 'newbalanceOrig',\n"," 'oldbalanceDest',\n"," 'newbalanceDest']"]},"metadata":{},"execution_count":7}]},{"cell_type":"code","source":["plt.figure(figsize=(10,6))\n","plt.hist(df[\"newbalanceOrig\"], 30, range=[0, 15000000],align='mid')#Plot distirbution newbalanceorig\n","plt.title(\" NewbalanceOrig Distribution\")\n","import os \n","os.chdir(\"/content/drive/MyDrive/Predictiong_cvss_client/FraudlentTransact/images\")\n","plt.savefig(\"fig1.png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":563},"id":"rgL3wzDfopk1","executionInfo":{"status":"ok","timestamp":1684180254233,"user_tz":-60,"elapsed":3995,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"2c5816d5-93b0-4c59-8723-bc1b1bea3dd6"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["The fig above shows that most customers have a balance ranging from 0 to half a miliion value after transactions.Reducing the xlimits can give us a better information about what is happening."],"metadata":{"id":"X7iUfgwnCEd6"}},{"cell_type":"code","source":["plt.figure(figsize=(10,6))\n","plt.hist(df[\"amount\"], 30, range=[0, 15000000],align='mid')\n","plt.title(\"Amount Distribution\")\n","plt.savefig(\"fig2.png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":403},"id":"0ZhCx3ofpdyB","executionInfo":{"status":"ok","timestamp":1671900846869,"user_tz":-60,"elapsed":837,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"77ed2124-9fab-4334-9305-ba464a5c060c"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["The fig above shows that most csutomers have a balance ranging from 0 to over 2 million in thier local currency.Just few customers have above this value"],"metadata":{"id":"I_M5v7zPC62b"}},{"cell_type":"code","source":["plt.figure(figsize=(10,6))\n","plt.hist(df[\"oldbalanceOrg\"], 30, range=[0, 15000000],align='mid')\n","plt.title(\" oldbalanceOrg Distribution\")\n","plt.savefig(\"fig3.png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":403},"id":"NpcfYeBnpewK","executionInfo":{"status":"ok","timestamp":1671900951468,"user_tz":-60,"elapsed":1537,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"694fdf0f-a97c-4f20-9852-7e6a528b193e"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Fig 3 shows the same similar distribution occurs as that of new balance after transactions"],"metadata":{"id":"Cjl73lBZEK3v"}},{"cell_type":"code","source":["plt.figure(figsize=(10,6))\n","plt.hist(df[\"oldbalanceDest\"], 30, range=[0, 15000000],align='mid')\n","plt.title(\" oldbalanceDest Distribution\")\n","plt.savefig(\"fig4.png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":403},"id":"ZdLtWTvYpf1z","executionInfo":{"status":"ok","timestamp":1671901160539,"user_tz":-60,"elapsed":959,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"5bef30ae-bbda-45ae-8f13-23743fcda536"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["fig 4 shows that the distribution of the recipient balance after transaction is rights skewed whereby just most values ranges from o to over a value of 4 million "],"metadata":{"id":"FPKn8wlfEzom"}},{"cell_type":"code","source":["plt.figure(figsize=(10,6))\n","plt.hist(df[\"newbalanceDest\"], 30, range=[0, 15000000],align='mid')\n","plt.title(\"newbalanceDest Distribution\")\n","plt.savefig(\"fig5 .png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":403},"id":"5Eq5vpHekRWS","executionInfo":{"status":"ok","timestamp":1671901644282,"user_tz":-60,"elapsed":1172,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"92e8e070-e3cf-44df-ebf1-b23c6c6712fa"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["fig 5 which shows the distribution as of the recipient balance after transaction shows a similiar distributions as compared to that of the balance of recipient after transacions."],"metadata":{"id":"X0809ePTGpkr"}},{"cell_type":"markdown","source":["**Outliers**"],"metadata":{"id":"3W5eiFwTHp1f"}},{"cell_type":"code","source":["\n","plt.boxplot(df[\"newbalanceDest\"])\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":276},"id":"X6aVp1ycHtYy","executionInfo":{"status":"ok","timestamp":1671902106713,"user_tz":-60,"elapsed":1882,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"20fa2191-bdea-4728-f901-ab22d33d8b91"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":["# Set the figure size\n","plt.rcParams[\"figure.figsize\"] = [7.50, 3.50]\n","plt.rcParams[\"figure.autolayout\"] = True\n","\n","# Pandas dataframe\n","\n","# Plot the dataframe\n","ax = df[contFeat].plot(kind='box', title='boxplot')\n","plt.ylabel(\"Value\")\n","\n","# Display the plot\n","plt.savefig(\"boxplots\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":261},"id":"zBfXQkUuIvyQ","executionInfo":{"status":"ok","timestamp":1671902372230,"user_tz":-60,"elapsed":13642,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"637066f4-1b61-4983-d039-3769019d4377"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAAhQAAAD0CAYAAADDob9OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5wcZZ3v8c93Jk0GMhgIiRFDQlyFNRBuGi7msBziagJZNtFdWMmymkAw4tHILrpZNYsSjiwCEo8YXeQSQVdHvKBEZAWUcbkpGjAXAuiyQkiIwIQJmoszmcz8zh9VM8xM5pb01FS6+/t+vfrV1U8/XfXrp6q7f/3UU1WKCMzMzMyKUZV3AGZmZlb6nFCYmZlZ0ZxQmJmZWdGcUJiZmVnRnFCYmZlZ0ZxQmJmZWdGcUJhZryQ9K+kdQ7i8WyR9ZqiWZ2aDxwmFmZUkSSHpTXnHYWYJJxRmZmZWNCcUZtafEyU9IWmLpK9KqgGQ9H5JT0tqlLRC0uvT8n+X9L32F0u6StJPlThd0kZJn5S0Od2lcl5vC+5jGfenVVZL2ibpPRm+fzMbACcUZtaf84AZwBuBI4F/lfR24Erg74BDgfXAt9L6HwWOkTRP0l8A84G58ep5/l8HjAbGAXOBGyT9efeF9rWMiDgtrXZcRNRGxG2D+5bNbE+VZEIhabmklyQ9PoC6EyTVS/q1pDWSZg5FjGZlZFlEbIiIRuAKYA5JkrE8Ih6LiGbgE8DbJE2MiB3Ae4GlwH8ACyNiY7d5XhoRzRHxX8CPSJKG7npdRgbv0cyKVJIJBXALcMYA6/4r8O2IOAE4F/hyVkGZlakNnabXA69Pb+vbCyNiG/AySa8DEfEI8DtAwLe7zW9LRGzvYZ7d9bkMM9u3lGRCERH3A42dyyS9UdKPJT0q6QFJb26vDrwmnR4JbBrCUM3KwfhO0xNIPkObgMPbCyWNAA4Bnk8ffwgYntZb1G1+B6f1u8+zuz6XYWb7lpJMKHpxA0nX6luBj/FqT8RlwD9I2gjcBSzMJzyzkvUhSYdJGgUsBm4D6oDzJR0vaTjwb8AjEfGspCOBzwD/QLLrY5Gk47vNc4mk/dIxFmcB3+lhub0uI33+ReDPBvetmtneGpZ3AINBUi0wFfiOpPbi4en9HOCWiLhW0tuAr0uaHBFtOYRqVoq+CdxDsgviDuAzEbFD0qXA94CDgYeBcyUNIxk3cVVErAaQ9EmSz92UdH4vAFtIeiB2ABdFxFPdFxoRP+lpGZ2qXAbcKml/YEFEdN+1YmZDSK8OvC4t6cCsOyNisqTXAL+JiEN7qLcOOCMiNqSPfwecEhEvDWW8ZgaSTgf+IyIOyzsWMxtcZbHLIyL+CDwj6RyA9Hj349KnnwP+Mi2fBNQADbkEamZmVqZKMqGQVAf8HPjz9CQ580kOMZsvaTWwDpidVv8o8P60vA6YF6XaLWNmZraPKtldHmZmZrbvKMkeCjMzM9u3lNxRHqNHj46JEyfmHYaZmVlFevTRRzdHxJju5SWXUEycOJGVK1fmHYaZmVlFkrS+p3Lv8jAzM7OiOaEwMzOzojmhMDMzs6I5oTAzM7OiOaEwM6sAdXV1TJ48merqaiZPnkxdXV3eIVWUhQsXUlNTgyRqampYuLD8rlNZckd5mJnZnqmrq2Pu3Lm0tLQAsG7dOubOnQvAnDlz8gytIixcuJBly5Z1PG5ubu54/MUvfjGvsAadeyjMzMrc+eefT0tLC7NmzaKhoYFZs2bR0tLC+eefn3doFaE9eZg6dSqbNm1i6tSpXcrLRWYJhaQaSb+UtFrSOklLeqgzT1KDpFXp7cKs4jEzq1TNzc2cddZZ3HHHHYwePZo77riDs846i+bm5rxDqxgnnXQSDz30EIceeigPPfQQJ510Ut4hDboseyiagbdHxHHA8cAZkk7pod5tEXF8erspw3jMzCrW/Pnz+3xs2TrzzDP7fFwOMksoIrEtfVhIb74SmZlZDs477zzq6+tpaWmhvr6e8847L++QKsqSJUtYunQpO3bsYOnSpSxZslunfcnL9GqjkqqBR4E3AV+KiH/p9vw84EqgAfgt8E8RsaGH+SwAFgBMmDDhrevX93jWTzMz68Gxxx7L2rVrOfDAA9m+fTsjRoxg69atHHPMMaxZsybv8MrehAkT2LBhA5KIiI778ePH89xzz+Ud3h6T9GhETOlenumgzIhojYjjgcOAkyRN7lblh8DEiDgWuBe4tZf53BARUyJiypgxu12PxMzM+rBmzRqOOeYYtm7dSltbm5OJIfbcc88xfvx42v/Al3Iy0ZchOWw0Il6RVA+cATzeqfzlTtVuAq4einjMzCqNk4d8lVvy0JMsj/IYI+mgdHp/4J3AU93qHNrp4SzgyaziMTMzs+xk2UNxKHBrOo6iCvh2RNwp6XJgZUSsAD4iaRawC2gE5mUYj5mZmWUk00GZWZgyZUqsXLky7zDMzMwqUi6DMs3MzKwyOKEwMzOzojmhMDOrAL7aaL58tVEzMyt5vtpovirlaqMelGlmVuZqampobm6mtraWbdu2ddwPHz6cpqamvMMre1VVVUQEs2bN4uabb2b+/PmsWLECSbS1teUd3h7zoEwzswrV3NxMTU0NK1asYOfOnaxYsaIjybDsRQQzZ87scrXXmTNnUmp/6PvjhMLMrALMnDmzYz/+woULmTlzZt4hVZSampouY1hqamryDmnQOaEwM6sAt99+OxdccAFbt27lggsu4Pbbb887pIpy++23c9ppp9HY2Mhpp51Wlu3vMRRmZmVOEgC1tbUdVxvdtm0bQNl1u++L2ncvVVVV0dbW1nFfqmNYPIbCzKxCSaJQKLBt2zYigm3btlEoFDoSDctWS0sLkydP7hiA2dbWxuTJkzuOuikXTijMzMrcUUcdxaJFizj66KOpqqri6KOPZtGiRRx11FF5h1YRJk2axHXXXUdEdNyuu+46Jk2alHdog8rnoTAzK3OLFy/mAx/4AE1NTbS1tfHb3/6W6667jq985St5h1YRFi9ezIwZM7r0SBQKBW699dYcoxp87qEwMytzDz/8MNu3b2fUqFEAjBo1iu3bt/Pwww/nHFlluPLKK2lpaaG2thZJ1NbW0tLSwpVXXpl3aIPKCYWZWZm78cYbueaaa3jhhReICF544QWuueYabrzxxrxDqwhr165l1qxZbN26lba2NrZu3cqsWbNYu3Zt3qENKh/lYWZW5iSxfft2DjjggI6yHTt2MGLECB/lMQQk0dDQwOjRozvKNm/ezJgxY0qy/X2Uh5lZhRo+fDjTp0/vcnGq6dOnM3z48LxDqxgnnnhil/Y/8cQT8w5p0GWWUEiqkfRLSaslrZO0pIc6wyXdJulpSY9ImphVPGZmlerII4/koYce6jjVdnNzMw899BBHHnlkzpFVhlGjRvHss892af9nn322Y0xLuciyh6IZeHtEHAccD5wh6ZRudeYDWyLiTcDngasyjMfMrCI9/vjjQHKRqs737eWWrcbGxj0qL1WZJRSR2JY+LKS37juLZgPtx818F/hL+UwrZmaDKiK48MILaW1tJSJobW3lwgsvLMn996Xq5JNP7nIeipNPPjnvkAZdpmMoJFVLWgW8BNwbEY90qzIO2AAQEbuAPwCHZBmTmVklmjhxYp+PLVszZszo83E5yPTEVhHRChwv6SDg+5ImR8Qe97FJWgAsAJgwYcIgR2lmVv4uvfRS9t9/fy666CKuv/56Lr300rxDqiiXX345I0eO7Gj/yy+/PO+QBt2QHTYq6VPAjoj4XKeyu4HLIuLnkoYBLwBjoo+gfNiomdmemTFjBvfcc89u5dOnT+fuu+/OIaLKMmHCBDZs2LBb+fjx43nuuedyiKg4Q37YqKQxac8EkvYH3gk81a3aCmBuOn02cF9fyYSZme25efPmUSgUupQVCgXmzZuXT0AVZvbs2XtUXqoy66GQdCzJgMtqksTl2xFxuaTLgZURsUJSDfB14ASgETg3In7X13zdQ2FmtmfGjx/Ptm3bOOigg1i/fj2HH344r7zyCrW1tT3+c7bBVSgUqKmpYfTo0R3tv3nzZpqamkryiqND3kMREWsi4oSIODYiJkfE5Wn5pyJiRTrdFBHnRMSbIuKk/pIJMzPbcxs3bqSmpobly5fT3NzM8uXLqampYePGjXmHVhF27drFiBEjurT/iBEj2LVrV96hDSqfKdPMrAJccsklTJs2jUKhwLRp07jkkkvyDqmizJ49u0v7l9vuDnBCYWZWEa699lrq6+tpaWmhvr6ea6+9Nu+QKspNN93E0qVL2bFjB0uXLuWmm27KO6RB54uDmZmVOY+hyNf48eNpbGykpaWFlpYWCoUChUKBUaNGlWT7++JgZmYV6uqrr6a1tZXnn3+eiOD555+ntbWVq6++Ou/QKsLVV1/NiBEjGDduHFVVVYwbN44RI0aUXfs7oTAzM8vQnDlzOOGEE1i/fj1tbW2sX7+eE044gTlz5uQd2qByQmFmVuYWLVpEa2trl7LW1lYWLVqUU0SVZeHChdx7771dLs527733snDhwpwjG1weQ2FmVuYkIYnXvva1vPjii4wdO5aXXnqp40JVlq1hw4bR2trK2LFjO9r/xRdfpLq6uiQPHfUYCjOzCjZixAjq6urYuXMndXV1jBgxIu+QKkZraysjR47s0v4jR47crdeo1DmhMDOrAMOGDevzsWXrnHPO6XIeinPOOSfvkAadtygzswrQ0tLCBRdc0HHYaCme8rmULV++nEmTJnVcbXT58uV5hzToPIbCzKzMHXLIIWzZsoWqqipaW1uprq6mra2Ngw8+mJdffjnv8MreIYccQmNjI8OGDWPXrl0d96NGjSrJ9vcYCjOzCvX3f//3QDI4s/N9e7lla9myZRx44IFd2v/AAw9k2bJlOUc2uJxQmJmVufr6embPnk11dTUA1dXVzJ49m/r6+pwjqwxz5sxh7ty5XQ4bnTt3btmdh8K7PMzMylxVVRXDhg3rMm6iUCiwa9cu2tracoysMtTV1fHe9763y1Ed1dXVfP3rXy/JpKK3XR5OKMzMylx7V3ttbS3btm3ruAd8Hooh0J681dTU0NTU1HHfPckrFR5DYWZW4ZYsWcL27dtZsmRJ3qFUlF27dlEoFLjrrrvYuXMnd911V0eSUU7cQ2FmVuYkUSgUdtvl0dLS4h6KIdB+ptLObd3+uBTbf8h7KCSNl1Qv6QlJ6yRd3EOd0yX9QdKq9PaprOIxM6tk3bvWS7GrvZR1TxxKMZHoT5YnttoFfDQiHpN0IPCopHsj4olu9R6IiLMyjMPMzMwyllkPRUT8PiIeS6e3Ak8C47JanpmZmeVnSAZlSpoInAA80sPTb5O0WtJ/Sjq6l9cvkLRS0sqGhoYMIzUzK09jx47l6KOPpqqqiqOPPpqxY8fmHVJFqamp6dL+NTU1eYc06DJPKCTVAt8D/jEi/tjt6ceAwyPiOOCLwA96mkdE3BARUyJiypgxY7IN2MysDL344osAPPPMM10e29Boampi06ZNrFq1ik2bNtHU1JR3SIMu06M8JBWAO4G7I2LpAOo/C0yJiM291fFRHmZme6a2tpbt27fvVj5ixIiO81FYdno7RNTnoRj4AgXcDDzZWzIh6XVpPSSdlMZTeldKMTPbh51//vl7VG6D66KLLtqj8lKVWQ+FpFOBB4C1QPu5XT8JTACIiOslfRj4IMkRIX8CLomIh/uar3sozMz2TE1NDc3NzbuVDx8+vCy73vc15db+vfVQZHbYaEQ8CKifOsuA8rrcmpnZPqb9x6yqqop77rmH6dOn09bW1uOPnA2+zu3c+QRX5db+PvW2mVmFKBQKvOMd76BQKOQdSsUqxxNatXNCYWZWIWbMmEFDQwMzZszIO5SKNHXqVDZt2sTUqVPzDiUTvpaHmVmZa7/aaE9K7TegFJVb+/tqo2Zmxg9+0OPpfmyI3HLLLXmHkBn3UJiZlbly+4dcasqt/d1DYWZWwaqqqjpOtz127Fiqqvz1b4Mry6uNmpnZPqKtra3jdNs+7bZlwSmqmVmZ6603wr0UNpi8NZmZlbm2trY9KjfbG04ozMwqRPsJrXxiq6E3ZswYhg8fDiSn3C7HK2c7oTAzqxC1tbVd7m3oNDQ0dEnoGhoaco5o8DmhMDOrEK+88kqXexta7ZeKL9dLxjuhMDOrEO3nPCjFcx/Yvs8JhZmZmRXNCYWZmZkVzQmFmVmFaD/vhM8/kY9yb//M3pWk8ZLqJT0haZ2ki3uoI0nXSXpa0hpJb8kqHjOzSlZdXd1x3om2tjaqq6tzjqjytB8qWo6HjEK2PRS7gI9GxFHAKcCHJB3Vrc6ZwBHpbQHw7xnGY2ZWsVpbWzn44IOpqqri4IMPprW1Ne+QKk5jY2OX+3KTWUIREb+PiMfS6a3Ak8C4btVmA1+LxC+AgyQdmlVMZmaVbMuWLbS1tbFly5a8Q6lILS0tXe7LzYATCkkH7O1CJE0ETgAe6fbUOGBDp8cb2T3pQNICSSslrSzHk4GYmZmVun4TCklTJT0BPJU+Pk7Slwe6AEm1wPeAf4yIP+5NkBFxQ0RMiYgp5brvyczMrJQNpIfi88AM4GWAiFgNnDaQmUsqkCQT34iI23uo8jwwvtPjw9IyMzMzKyED2uURERu6FfU7mkeSgJuBJyNiaS/VVgDvS4/2OAX4Q0T8fiAxmZmZlZJyvzjbsAHU2SBpKhBpj8PFJAMs+/O/gPcCayWtSss+CUwAiIjrgbuAmcDTwA7g/D0L38zMrDSU+6DMgSQUFwFfIBks+TxwD/Ch/l4UEQ8C6qdODGReZmZmtm/rN6GIiM3AeUMQi5mZmZWofhMKSV8Fdrs0XURckElEZmZmVnIGssvjzk7TNcC7gU3ZhGNmZmalaCC7PL7X+bGkOuDBzCIyMzOzkrM3p94+AnjtYAdiZmZmpWsgYyi2koyhUHr/AvAvGcdlZmZmJWQguzwOHIpAzMzMrHT1mlBIektfL2y/kqiZmZlZXz0U1/bxXABvH+RYzMzMrET1mlBExLShDMTMzMxK10DOQ4GkycBRJOehACAivpZVUGZmZlZaBnKUx6eB00kSiruAM0nOQ+GEwszMzICBnYfibOAvgRci4nzgOGBkplGZmZlZSRlIQtEUEW3ALkmvAV4CxmcblpmZmZWSvg4b/RJQB/xS0kHAjcCjwDbg50MTnpmZmZWCvsZQ/Ba4Bng9sJ0kuXgn8JqIWDMEsZmZmVmJ6HWXR0R8ISLeBpwGvAwsB34MvFvSEf3NWNJySS9JeryX50+X9AdJq9Lbp/byPZiZmVnO+h1DERHrI+KqiDgBmAO8C3hqAPO+BTijnzoPRMTx6e3yAczTzMzM9kH9JhSShkn6a0nfAP4T+A3wN/29LiLuBxqLD9HMzMz2dX0NynwnSY/ETOCXwLeABRGxfRCX/zZJq4FNwMciYl0vsSwAFgBMmDBhEBdvZmZmg6GvQZmfAL4JfDQitmSw7MeAwyNim6SZwA+AHsdmRMQNwA0AU6ZMiQxiMTMzsyL0dS2PTC/+FRF/7DR9l6QvSxodEZuzXK6ZmZkNvoGc2CoTkl4nSen0SWksL+cVj5mZme29AV0cbG9IqiO5BshoSRuBTwMFgIi4nuSU3h+UtAv4E3BuRHh3hpmZWQnKLKGIiDn9PL8MWJbV8s3MzGzo5LbLw8zMzMqHEwozMzMrmhMKMzMzK5oTCjMzMyuaEwozMzMrmhMKMzMzK5oTCjMzMyuaEwozMzMrmhMKMzMzK5oTCjMzMyuaEwozMzMrmhMKMzMzK5oTCjMzMyuaEwozMzMrmhMKMzMzK5oTCjMzMytaZgmFpOWSXpL0eC/PS9J1kp6WtEbSW7KKxczMzLKVZQ/FLcAZfTx/JnBEelsA/HuGsZiZmVmGMksoIuJ+oLGPKrOBr0XiF8BBkg7NKh4zMzPLTp5jKMYBGzo93piW7UbSAkkrJa1saGgYkuDMzMxs4EpiUGZE3BARUyJiypgxY/IOZ6/U1dUxefJkqqurmTx5MnV1dXmHZGZmNmiG5bjs54HxnR4flpaVnbq6OhYvXszNN9/MqaeeyoMPPsj8+fMBmDNnTs7RmZmZFS/PHooVwPvSoz1OAf4QEb/PMZ7MXHHFFdx8881MmzaNQqHAtGnTuPnmm7niiivyDs3MzGxQKCKymbFUB5wOjAZeBD4NFAAi4npJApaRHAmyAzg/Ilb2N98pU6bEypX9VtunVFdX09TURKFQ6ChraWmhpqaG1tbWHCMzs0qQfN32LKvfAHtVubW/pEcjYkr38sx2eUREn335kbTih7Ja/r5k0qRJPPjgg0ybNq2j7MEHH2TSpEk5RmVmZjZ4SmJQZqlbvHgx8+fPp76+npaWFurr65k/fz6LFy/OOzQzM7NBkeegzIrRPvBy4cKFPPnkk0yaNIkrrrjCAzLNzKxsZDaGIiulOIbCzCxP5bYPv9SUW/v3NobCuzzMzMysaE4ozMzMrGhOKIaIz5RpZmblzAnFEKirq+Piiy9m+/btAGzfvp2LL77YSYWZmZUND8ocAuPHj6exsZGWlhZaWlooFAoUCgVGjRrFhg0b+p+BmVkRym1QYKkpt/Yf8hNb2as2btwIJGfMBGhra2PHjh3s2LEjz7DMzMwGjXd5DKHRo0d3uTczMysXTiiG0Isvvtjl3szMrFw4oTAzM7OiOaGwiuFDd83MsuOEwiqCD901M8uWEwqrCIsWLWLYsGEsX76cpqYmli9fzrBhw1i0aFHeoZmZlQUnFFYRNm7cyK233sq0adMoFApMmzaNW2+9teOQXjMzK06mCYWkMyT9RtLTkj7ew/PzJDVIWpXeLswyHqts9913X5cxFPfdd1/eIZmZlY3MEgpJ1cCXgDOBo4A5ko7qoeptEXF8erspq3isso0aNYqrrrqKzZs3ExFs3ryZq666ilGjRuUdWsXwoFiz8pZlD8VJwNMR8buI2Al8C5id4fLM+tSeSHS+t6HhQbFm5S/LhGIc0PlCFRvTsu7+VtIaSd+VND7DeKyCNTY2st9++9Ha2gpAa2sr++23H42NjTlHVhk8KNas/OU9KPOHwMSIOBa4F7i1p0qSFkhaKWllQ0PDkAZo5WPnzp19PrbsbNy4kRNPPJEzzzyT/fbbjzPPPJMTTzzRg2LNykhmVxuV9DbgsoiYkT7+BEBEXNlL/WqgMSJG9jXfUrzaaLldaa4UeR3ky+2fL7d/vsqt/Xu72miWPRS/Ao6Q9AZJ+wHnAiu6BXVop4ezgCczjMfMzMwyktnlyyNil6QPA3cD1cDyiFgn6XJgZUSsAD4iaRawC2gE5mUVj5mZmWUns10eWfEuD9sbXgf5cvvny+2fr3Jr/zx2eZiZmVmFcEJhZmZmRXNCYWZmZkVzQmFmZmZFc0JhZmZmRXNCYWZmZkVzQmFmZmZFc0JhZmZmRXNCYWZmZkVzQmFmZmZFc0JhZmZmRXNCYWZmZkVzQmFmZmZFc0JhZmZmRXNCYWZmZkVzQmFmZmZFc0JhZmZmRcs0oZB0hqTfSHpa0sd7eH64pNvS5x+RNDHLeMzMzCwbw7KasaRq4EvAO4GNwK8krYiIJzpVmw9siYg3SToXuAp4T1YxmZmZDaaJH/9R5q9/9rN/VdQyhkpmCQVwEvB0RPwOQNK3gNlA54RiNnBZOv1dYJkkRURkGJeZ2T7ruCX38Ic/tQzZ8or9Qexu5P4FVn96+qDOcygNdfsPxJ6uo7zWgbL67ZZ0NnBGRFyYPn4vcHJEfLhTncfTOhvTx/+T1tncbV4LgAUAEyZMeOv69esziRngmFuPyWzeQ2Xt3LV5h1AUr4N8uf3z5fbPVzm0P2S7DiQ9GhFTdisvhYSisylTpsTKlSsziTkrknp9zp0xQ8PrIF9u/3y5/fPX0zoo1bbvLaHIclDm88D4To8PS8t6rCNpGDASeDnDmMzMKk5vP1yl+oNWiiJit1u5yTKh+BVwhKQ3SNoPOBdY0a3OCmBuOn02cF85jp/whzl/Xgf5cvvnrxJ+0CxfmQ3KjIhdkj4M3A1UA8sjYp2ky4GVEbECuBn4uqSngUaSpKMs+cObP6+DfLn9zcpblkd5EBF3AXd1K/tUp+km4JwsYzAzM7Ps+UyZZmZmVjQnFGZmZlY0JxRmZmZWtMzOQ5EVSQ1Adme2yt5ooNfzbNiQ8DrIl9s/X27/fJVD+x8eEWO6F5ZcQlHqJK3s6YQgNnS8DvLl9s+X2z9f5dz+3uVhZmZmRXNCYWZmZkVzQjH0bsg7APM6yJnbP19u/3yVbft7DIWZmZkVzT0UZmZmVjQnFGZmZlY0JxQlRtK7JB2VdxyDQdK2XspvkXR2Ov2spNF7MM95kpYNVowDWN67JK2R9KSktZLeNVTLHiqSLpP0sT18TY/rNguSDpN0h6T/lvQ/kr6QXuG4p7qvl/TdoYqtlxhKertPl9Ug6ddpm98taepezut4STMHO8bBVALbf6ukVZLWSVot6aOS9uq3XdIni4nFCUXpeRdQFglFqZN0HPA5YHZETAJmAZ+TdGwPdTO9EF+lkiTgduAHEXEEcCRQC1zRQ91hEbEpIs4e4jDL0W0RcULa5p8Fbpc0aS/mczywTycUJeBPEXF8RBwNvBM4E/j0Xs7LCcVQkfQDSY+mmeCCtGybpGvSsp9IOknSzyT9TtKstE6NpK+m/2B/LWlaWt7lX4WkOyWd3mm+V6QZ5y8kjU3/BcwCrkkz0jcOeSPsJUmXSHo8vf1jt+ckaZmk30j6CfDabi9flLbdLyW9KX3NX0t6JG3Pn0ga28Mye6yT/uNY3mk9faTTa96X9jislvT1tGyMpO9J+lV6+19p9Y8B/xYRzwCk91cC/5y+7meS/p+klcDFkk5M570q3WYeH4Sm3Y2kiWmPyY3pdnmPpP0lvVHSj9Nt+AFJb5ZULemZdB0clP7bOS2dz/2Sjkhne5ykn6f/SN+fPl8r6aeSHkvXz+weYumxTm8xps+9KV1fq9PXvTEt/+e0/ddIWpIu4u1AU0R8NV0HrcA/ARdIOiD9jK2QdB/w03S5j6fzO0DStyU9Ien76bYyqCccKtPtvouIqCc5cqH9O3G37SwtPydth9Xptl3mKY4AAAbrSURBVLUfcDnwnvQz8Z69a+Xd3n8lbf/d18VL6Xr4cPqeqpV817S/7gPpvA5N39+qdJ38haTPAvunZd/Yq8aPCN8GeANGpff7A48DhwABnJmWfx+4BygAxwGr0vKPAsvT6TcDzwE1wDxgWaf53wmcnk4H8Nfp9NXAv6bTtwBn590We9hubwXWAiNI/j2uA04AtqXP/w1wL1ANvB54pf09As8Ci9Pp9wF3ptMH8+pRShcC16bTHW3aR53LgIeB4SSnwX05XWdHA78FRndb398ETk2nJwBPptOPAcd1e6/HAY+l0z8DvtzpuceBt6XTnwUez6i9JwK7gOPTx98G/gH4KXBEWnYycF86/eP0vZ8F/ApYnLbNM53aazXJdj8a2JCup2HAa9I6o4GnO7V3+7rtsU5vMabTjwDvTqdrgAOA6SQ/WiL5I3QncBrwEeDzPbTBr4Fj0+1hY6d1ObG93UkSwq+k05PTeKZ4u+93u+9YVqf3+i7gP9Pp3raztcC4dPqg3ubl7X/g23/nZXVrg1eAsSTJRftvx3BgJfAGkt+k9u2rGjiwt3ntyc3dsHvmI5LenU6PB44AdpJskJB8YJojokXSWpKNBuBU4IsAEfGUpPUkXbN92Umy0QA8StKVVapOBb4fEdsBJN0O/EWn508D6iL5d7lJyb/Jzuo63X8+nT4MuE3SocB+wDM9LLevOj+KiGagWdJLJB++twPfiYjNABHRmNZ9B3CUpPbXvkZS7QDf+23pez6I5EP787T8myRfYFl5JiJWpdOPkmyLU4HvdHofw9P7B0jWwRtIeljeD/wXyZdruzsi4k/AnyTVAycBPwL+Lf1H1waMI2nHFzq9Tr3U6TFGSQeS/Oh8HyAimgAkTSf5Uv11Wr+W5PM3EPd2WpednQp8IV3O45LWDHB+A1VJ273S91hL79vZQ8Atkr5NspsqS5Wy/d/fTztMB45VOjYHGJm+7lfAckkFkt2Fq3qbwZ5wQjFASnZFvIPkH+YOST8jyR5bIk3tSDaYZoCIaFP/+8130XW3U02n6c7zbaWy11X0MP1FYGlErEjXzWU9vK6vOs2dpvtr3yrglPYPdztJT5D8C13dqfitJP9E223vY75Z6v7+xgKvRMTxPdS9H/ggyb+uT5Hssjmd5Iu2XfcT1gRwHjAGeGuaRD9L122Yfup0j3H/Pt6PgCsj4itdCqV3AGd3K3sNyT/qp4G3kN86KNa+ut33VPcE4Mn0NT1uZxFxkaSTgb8CHpX01j6WXayK2P57rCj9WTq/l9LXLYyIu3uodxrJurhF0tKI+Fp/8+6Px1AM3EhgS5pMvBk4ZQ9e+wDJhoWkI0m+7H5D0q15vKQqSeNJst7+bAUO3JPA9wEPAO9Sss96BPBuun5Y7yfZj1qd/qua1u317+l03/4PfyTwfDo9t5flDqROZ/cB50g6BEDSqLT8HmBheyVJ7V9KnwM+IWliWj6RZFDTtd1nHBGvAFvTL1SAcwcQz2D6I/CMpHOgY//9celzvyT599aW/nisAj5A138/s5WMBTqE5Mv2VyTt+1L6RTkNOLyH5Q6kToeI2ApsVHq0jKThkg4A7iYZF1Gblo+T9FqSbuwDJL0vLa8maf9bImJHP23yEPB36euOAo7pp/6eKtftvgtJ/5uka/3GiOh1O5P0xoh4JCI+BTSQ9PIO1fdZuW7/XUgaA1xPshsp0td9MO2JQNKRkkZIOhx4MSJuBG4iSbwBWtrr7g0nFAP3Y2CYpCdJ9n//Yg9e+2WgKt0NchswL+12fIikO/IJ4DqSffL9+Rbwz0oGXJXEoMyIeIxk7McvSfYP3hQRv+5U5fvAf5O0w9d49cuz3cFpd/TFJAPuIPnX9R1Jj9L7pYAHUqdznOtIjg74L0mrgaXpUx8BpigZ1PQEcFFafxXwL8APJT0F/BBY1Ef34XzgRkmrSPar/6G/mAbZecD89L2tA2YDpNviBl7dph8g+ZJf2+m1a4D6tM7/jYhNwDdI2mUtyX7+p3pY5kDqdPdekt2La0j2+b8uIu4h2U3083Re3yXZhRQkP9TnSPpvkrEATQxstPqXgTHpOv0MSZsM2jop1+0+1T6Q8rckbf23EfFk+lyP2xnJYPK1SgbFPkzSs1dPsltl0AZl9qHstv+0fvtAynXAT0gSwfZBmzeRbF+Ppe3+FZJeqdOB1ZJ+TZKwfiGtfwOwRns5KNOn3jYbIpJqI2JbOv1x4NCIuDjnsCpW2ptRiIimNDn/CfDnEbEz59DMSlIl75c3G2p/JekTJJ+79SQj3C0/BwD1aRevgP/jZMJs77mHwszMzIrmMRRmZmZWNCcUZmZmVjQnFGZmZlY0JxRmZmZWNCcUZmZmVrT/D88Hb5aznAUJAAAAAElFTkSuQmCC\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["The figure above shows five different bxoplots which indicate a lot of outliers present in the features named"],"metadata":{"id":"-VnHP2GzJYrU"}},{"cell_type":"code","source":["data = df.copy()\n","for f in contFeat: # log features\n"," data[f] = np.log(data[f] + 1)\n","\n","# Set the figure size\n","plt.rcParams[\"figure.figsize\"] = [7.50, 3.50]\n","plt.rcParams[\"figure.autolayout\"] = True\n","\n","# Pandas dataframe\n","\n","# Plot the dataframe)\n","ax = data[contFeat].plot(kind='box', title='boxplot')\n","plt.ylabel(\"Value\")\n","\n","# Display the plot\n","plt.savefig(\"boxplots2afterlog\")\n","plt.title(\"Boxplots after logging\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":261},"id":"5pwaXiw_J6TQ","executionInfo":{"status":"ok","timestamp":1671902687684,"user_tz":-60,"elapsed":4222,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"c24fabf8-14bf-4d9b-e4d4-c21b173f8946"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["After logging down the features,the above figures eliminates the outiers to some but only the amount still have some outlliers"],"metadata":{"id":"di3N1e_6KoC1"}},{"cell_type":"markdown","source":["**Pairplots bewteen of continuous features**"],"metadata":{"id":"gw3N4Hn7K27Z"}},{"cell_type":"markdown","source":["**Making correlation plots**"],"metadata":{"id":"3f6hYPpzM4Wr"}},{"cell_type":"code","source":["dataplot = sns.heatmap(data[contFeat].corr(), cmap=\"YlGnBu\", annot=True)\n","plt.title(\"correlation plot\")\n"," \n","# displaying heatmap\n","plt.savefig(\"corrplot.png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":261},"id":"njPa0hFYMRa0","executionInfo":{"status":"ok","timestamp":1671904214922,"user_tz":-60,"elapsed":1594,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"85a5960f-cd3f-444b-e941-b25f5d20e9e1"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAAdMAAAD0CAYAAADAIMyXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3gU1frA8e+bQkmFhDQIJTSpEgRBVBCkFymWHzYQLNiwXiteERWwoIJXQEGvjctVgQuI9CIQRUCqNEGQGiAJBJIQasr7+2OGsJtO2iZwPs+zDzsz75l5Z8nu2XNm9hxRVQzDMAzDKDg3VydgGIZhGGWdqUwNwzAMo5BMZWoYhmEYhWQqU8MwDMMoJFOZGoZhGEYhmcrUMAzDMArJVKaGUUaJSHsRiS5E+c9E5PWizCmfx90vIp1K+riGUZxMZWoYVwERGSQivzquU9XHVPVtV+WUl8J+WTCMkmQqU8MoBUTEIz/rDMMonUxlahiFJCLVRWSmiBwTkXgRGW+vdxORf4rIARGJE5FvRcTf3lZLRFREHhKRg8DPdutxlYiMFZF4YISIlBeRD0TkoIjE2l2zFXPI4xUR+VtETonIDhHpZ69vCHwGtBGRZBFJsNd/LSIjHco/IiJ7ROSEiMwRkaoO21REHhOR3SKSICITRERyyGOEiMwQkR/sXDaKSLMcYsuLyDgROWI/xtnrvIEFQFU752THfAyjtDGVqWEUgoi4A3OBA0AtoBrwvb15kP3oANQGfIDxmXZxC9AQ6Govtwb2AiHAKOBdoD4QCdS19z88h3T+BtoC/sCbwH9EJExV/wQeA1arqo+qVsrmPG4F3gH+Dwizz+f7TGG9gOuBa+24ruSsDzAdCAD+C8wWEc9s4l4DbrDPrxnQCvinqp4GugNH7Jx9VPVILsczDJcylalhFE4roCrwoqqeVtVzqnrx2uR9wEequldVk4FXgbszdd+OsMudtZePqOonqpoKnAOGAM+p6glVPQWMBu7OLhFVna6qR1Q1XVV/AHbb+eXHfcCXqrpRVc/bubYRkVoOMe+qaoKqHgSWY1WAOdmgqjNUNQX4CKiAVWlmd9y3VDVOVY9hfQkYkM+cDaPUMNdkDKNwqgMH7Movs6pYLbyLDmC950Ic1h3KVMZxOQjwAjY49KgK4J5dIiIyEHgeq4UMVku4Sp5ncCnXjRcXVDXZ7mquBuy3V8c4xJ+x95+TjPNQ1XT7RqLsummze41Md65R5piWqWEUziGgRg43Cx0Bajos1wBSgViHdZmnbXJcPg6cBRqraiX74a+qWSoxEakJfA4MBQLtrtxtWJVvdsfJNVf7mmUgcDiPcjmp7rAvNyDcPkaux8V6jS7GmSmtjDLDVKaGUTi/A0eBd0XEW0QqiMhN9rbvgOdEJEJEfLC6aH/IoRWbhaqmY1WQY0UkGEBEqolIdtcqvbEqn2N23GCgicP2WCBcRMrlcLjvgMEiEiki5e1c16rq/vzkmo0WInK7/SXjWeA8sCaH4/5TRIJEpArW9eD/OOQcePGmLcMozUxlahiFoKppwG1YNwcdBKKB/vbmL4EpQBSwD+sa6FOXeYiXgT3AGhFJApYC12STxw7gQ2A1ViXUFFjlEPIzsB2IEZHj2ZRfCrwO/A/ry0Edcrg2m08/Yr0OJ7Gugd5uXz/NbCSwHtgCbMXqah5p57QTq7Lda99BbLp/jVJLzOTghmEUJREZAdRV1ftdnYthlBTTMjUMwzCMQjKVqWEYhnHVEJEv7UFUtuWwXUTkX/YAJltE5Lp87dd08xqGYRhXCxFpByQD36pqk2y298C6t6EH1iAqH6tq67z2a1qmhmEYxlVDVaOAE7mE9MGqaFVV1wCVRCQsr/2aQRtKgYo17rkquwcO/nWPq1NwmaAKDVydgsvUfOegq1NwGffDp1ydgsvsHd8v27GcCyqnz81zh75/FGvksIsmq+rky9h1NZwHT4m21x3NrZCpTA3DMIwyxxoLJCu74rycyrNImMrUMAzDKHPcim+GwsM4jOCFNXpXniOBmWumhmEYRpnj5uae7aMIzAEG2nf13gAkqmquXbxgWqaGYRhGGZT9cNj5KSffAe2BKvYEDG8AngCq+hkwH+tO3j1YEzoMzs9+TWVqGIZhlDkF7eZV1VzvfFTr96JPXu5+TWVqGIZhlDlubqWr+ipd2RiGYRhGPrhJkVwfLTKmMjUMwzDKHNMyvUqISF/gL3tqrFLpszGP0r1jc47FJ9Gy80uuTqfQ1qzaycfvzSE9PZ1e/Vox4KFbnbZfuJDKyNe+Z9ef0fj5e/HW+/cTVi0gY3vM0ZMM6PcBgx/vzL0PtAdg9PBp/Ba1g8oBPkyZ+UJJnk6BqSqjRk1m5coNVKhQnnfffYbGjetmiRs79ltmz15OUlIymzZNz1i/bt02Ro/+nF279vPRRy/RrdtNWcqWRrfUDuSNTvVxdxO+33yYT9ccyBLTs0Ewz7WtjSr8GZfM03Os4Vm/6R9J86r+rI9O4MHpf5R06oXWrmEww++8Fjc3YdpvB/hsyV9ZYno0r8YzPRqgwM7DiTz79XoAXu7TmA5NQnET4dedcbw1Y0sJZ18wBb0BqbiYn8YUn75AI1cnkZsp01fSZ+C7rk6jSKSlpfPR6Fl8MPEh/jPrBZYu3My+v2OdYubO+h1fv4r8MPcV+t/fjk/HzXfaPv6Dn2h9s/PIRD36tOTDTx8u9vyLUlTUBvbvP8LixZN4++0nGTHi02zjOnRoxfTpH2ZZHxYWxDvvPEuvXrcUd6pFxk3g7S7X8MC0zXSavJrejUKpF+jtFFOrckWebBPB7VPW0/mLNby5dFfGtslrDvDcT9tLOu0i4Sbw5v81Y/DE3+g6cim3tQinbqivU0ytIG8e71Kfuz6KotuoZbxtV5jXRQTQonYgPUYvo9uopVxbszKt61VxxWlcNjc3j2wfLsvHZUcuRiIyW0Q2iMh2ERlir0sWkTH2uqUi0kpEVojIXhHpbcdUEJGvRGSriGwSkQ72+kEiMt5h/3NFpL3DfkeJyB8iskZEQkTkRqA3MEZENotInRJ/EfJh1e87OZGQ7Oo0isSf2w4SXr0K1cID8fT0oFO3SH5d4fzh+Ovy7XTv3QKA9p2bsuH33Vyc6CHq522EVQsgok6IU5nIFrXx8/MqmZMoIsuWraFv31sRESIjG5CUdJq4uKxDkUZGNiA4OCDL+vDwEBo0iMDNrUhHfytWkVX92X/yLIcSzpKSrvz0Zyyd6wc5xdwTWY1vNx4i6VwqAPFnLs1VvurASU5fSCvRnItKs1oBHDh+mkPxZ0hJU+ZujKbztc5Dyfa/sRZTovaSdNY65/jkCwAoUN7TDU8PN8p5uOPpLhxPOl/Sp1AgbuKR7cNl+bjsyMXrQVVtAbQEnhaRQMAb+FlVGwOngJFAZ6Af8JZd7kmsO6ObAvcA34hIhTyO5Q2sUdVmQBTwiKr+hvXD3xdVNVJV/y7i8zMyORaXRHBopYzloGB/jsUmZopJzIjx8HDH26cCiQlnOHPmPFO/Ws7gxzqXaM7FJTY2ntDQS62L0NBAYmPjXZhR8Qv1Kc/RpHMZy0dPnSPUt7xTTESAFxEBXvxvQEtmDbyeW2oHlnSaxSLUvwJHT57NWD568iwh/s4fWxHBPkQE+zDtuXb87x+30K5hMACb9p1gze7jrB3VnbWjuxP1Zxx/x5aN8YOLcdCGguXjsiMXr6dF5A9gDdawUPWAC8BCe/tWYKWqptjPa9nrbwb+A6CqO4EDQP08jnUBmGs/3+Cwr1yJyBARWS8i61OT9+SniFFMvvx0Mf93fzu8vMrnHWyUWR5uQq3KXvSfuoGnf9zKu90b4le+dF13Ky4e7m7UCvbh3o9/4Zmv1zH63ub4VvSkZhVv6ob4cuM/F9LmtQW0qR/E9XXKxpeM0tYyveL+kuzu105AG1U9IyIrgApAil6avDUdOA+gqumS95XsVJy/eDh+7XPcbxr5fE0dB2O+WmeNKUpBwX7ExSRkLB+LSyQoxD9TjD9xMQkEh1QiNTWN08nn8K/kxY6th1ixdCufjptH8qmziAjly3lyxz1l48YbgKlT5zFt2iIAmjatR0zM8YxtMTHxhISUjQ/IgopJPk+Y36W3ZZhvBWJOOXdXHk06z+YjiaSmK4cSz7HvxBlqBXix5WhSSadbpGISzxFWuWLGcljlisQmnnOOSTjL5v0nSE1XouPPsD8umYggb1rXC2LT/hOcsbu4V26PoXlEAOv+Lv09Ga6sOLNzJbZM/YGTdkXaALjhMsr+AtwHICL1gRrALmA/ECkibiJSHWiVj32dAnzzjDKKRIPG1Tl08DhHok+QkpLK0oWbuekW5/u/bmrfiAVzNgCwYslWrmtVFxFh4tdPMGPBMGYsGMZd97VlwMO3lqmKFOC++3ry44//4scf/0WnTjcwe/bPqCqbN+/E19cr22ujV5I/jiQRUbki1f0r4Okm3NYwhCW7jznFLN59jBtqVgagckVPIgK8OJhwNrvdlSlbDpykVpAP4YFeeLoLva4LZ+kW56FkF/9xhNb1rGvIlb3LUSvYh4PxZzhy8gyt61bB3U3wcBNa16vCnpgy0s3r7pntw1VKV9VeNBYCj4nIn1gV4ZrLKDsR+FREtmK1Rgep6nkRWQXsA3YAfwIb87Gv74HPReRp4M7SeN30m0+eom2bhlSp7MueteN5+6MZfPPDClenVSAeHu48/2pfnn/8c9LT0+nZtxW164byxYRFNGgczs3tG9OrXyvefu17+vd6Fz8/L0a8f1+e+33j5alsXv83CQmn6dd5JA893oVet+fnu5Tr3HJLS1auXE/nzkOoWLE8o0c/k7GtT5+n+fHHfwHw/vtfMXfuSs6ePU+7doO4664uPPXUvWzZ8hdDh44mKSmZ5cvX8cknU5k3b6KrTidf0lQZvmQX397dHHcRpm05wu7jp3m+bW22HE1i6Z7jrNwbT7uIAJY+cgNp6TD6590k2DfkTL+/BXUCvfH2dGfNkzfz0vwdRO3Lbf7o0iMtXRkx7Q++efIm3ASmrznA7phTPNuzIVsPnmTZ1hii/oyjbcMQFr3WkXRV3p29jYTTF1iw6TBt6gexYFhHVJWoP+P4eVuMq08pX0pby1Qu9VAarnK1dvOaycGvTmZy8KtTUU8OXv+Gidl+bv615gmX3IZeuqp2wzAMw8gPt9J1ldJUpoZhGEaZo+6l63fQpatqNwzDMIz8cJfsH/kgIt1EZJeI7BGRV7LZXkNEltuD92wRkR557dNUpoZhGEbZ4ybZP/IgIu7ABKA71pCv94hI5qFf/wlMU9XmwN1YN6fmns5ln4BhGIZhuJq7W/aPvLUC9qjqXlW9gPXLiz6ZYhTws5/7A0fy2qmpTA3DMIyyJ4duXsfR5ezHkEwlqwGHHJaj7XWORgD3i0g0MB94Kq90zA1IhmEYRpmjOXTpOo4uVwj3AF+r6oci0gaYIiJNVDU9pwKmMi0FrtbfW9ao/52rU3CZaq17ujoFl9kyNSjvoCtUmpqP3CJT8Lt5D2ON2X5RuL3O0UNANwBVXW1PeFIFiMtpp6ab1zAMwyh7Cn7NdB1QT0QiRKQc1g1GczLFHAQ6AohIQ6zx2I+RC/M1yTAMwyh7CjjfrqqmishQYBHgDnypqttF5C1gvarOAf6BNRzsc1g3Iw3SPIYLNJWpYRiGUfYUYtAGVZ2PdWOR47rhDs93AJc124WpTA3DMIyyp4At0+JiKlPDMAyjzNH8XR8tMaYyNQzDMMqeUjY2r6lMDcMwjLLHw7RMDcMwDKNQStusMcVemYpIsqr6ZLP+a2Cuqs4Qkf1AS1U9ns99DrLjhxZlrrkcry/wFuAJpAKvq+rskjj25VqzaicfvzeH9PR0evVrxYCHbnXafuFCKiNf+55df0bj5+/FW+/fT1i1gIztMUdPMqDfBwx+vDP3PtAegNHDp/Fb1A4qB/gwZeYLJXk6xeKzMY/SvWNzjsUn0bLzS65Op8i1iwzjn4Ovx91NmLZsD5Nmb3fa/toDLWjdJASAiuU8CPSvwHWDpgGw64d72XUwAYCjx8/w6HsrSjT3/Fj96w4+fG8m6Wnp9Lm9DQ883Nlp+4ULKYwY9h927jiEfyVvRo0ZRNVqgQB8/cVi5sxcg5u7G/945Q7a3NSQA/tiGfbi1xnlj0QfZ8iTPbhnQAeGvfAVB/Zbv9NPPnUWH9+KTJ3xcomda25W/7qTce/9SFp6Or1vb83AbN7rb732HTt3ROPv78XIMQMIqxbA76v/YuK4eaSkpOHp6c7Q53vRsnU9Tp8+x+ODLo3nHhebQNeeLXju5czD1pYS5ppp2SIizYAPgM6quk9EIoAlIrJXVbdkivVQ1VSXJAqkpaXz0ehZjJ00hOAQfx6+91/c3L4xEXVCMmLmzvodX7+K/DD3FZYu2Myn4+bz1pj7M7aP/+AnWt/cwGm/Pfq05I57bmTka9+X2LkUpynTV/LZN4v4YuwTrk6lyLm5CSMeasUDby8j5sQZZr7TnWXro9kTnZgRM+qbDRnPB3S7hkYRlTOWz11Io/eLTr8YKFXS0tJ5f9R0xk9+kuDQSjxw9we07dCE2nXCMmLmzFyDr58XM+cPZ/GCDYwfO4fRHwxm799HWbxgI9/PfpVjcUkMfWQ8M+a+Ts2IkIwKMi0tnZ4dX6d9x2YAjP5gcMZ+x42ZhY9PhZI94RykpaXz4ehZfDzZeq8/eM/HtG3fiIg6oRkxP81ci69fRWbMe5UlCzYxYdw8Ro4ZgH8lb8Z88iBBwf78vfsozz7+OT8tHY63dwW+nf58RvlB/cfSvmMTV5xe/niUrpZpkVbtIvK8iGyzH89m2iYiMt6eQ24pEJyp+EsislVEfheRunaZ20RkrT2n3FIRCclUJscYERkhIl+KyAoR2SsiTzuUGWjPUfeHiEyx1wWJyP9EZJ39uPgboxeA0aq6D8D+9x3gRbvcChEZJyLrgWdE5Hp735tFZIyIbCuClzZf/tx2kPDqVagWHoinpwedukXy6wrnVsmvy7fTvXcLANp3bsqG33dz8bfIUT9vI6xagFPlCxDZojZ+fl4lcxIlYNXvOzmRkOzqNIpFs7qBHIg5xaG4ZFJS05m3aj+dWobnGH/bzbWYu2p/ySVYSNu3HiC8RhDVqlfB09ODLt2vI2r5VqeYlcu30rN3KwBu7RzJurV/oapELd9Kl+7XUa6cJ9XCAwmvEcT2rQecyq5bu4vw6lUIqxrgtF5VWbpoE116tCjeE8ynHdsOEl4j0Om9HrXc+b3+y4rt9OjdEoAOna9l/VrrvX5Nw2oEBfsDULtuKOfPpXDhgnMb4OD+Y5w8kUxki9olc0IFUfARkIpFkR1ZRFoAg4HWwA3AIyLS3CGkH3AN1vxxA4EbM+0iUVWbAuOBcfa6X4Eb7Dnlvgey65PLLaYB0BVryp03RMRTRBpjzVV3q6o2A56xYz8Gxqrq9cAdwBf2+sbABpytt9dfVE5VW6rqh8BXwKOqGgmkZZNvsTkWl0RwaKWM5aBgf47FJmaKScyI8fBwx9unAokJZzhz5jxTv1rO4Mecu8yMsiUkwIuj8WcylmNOnCEkMPsvQlWreBMe7MPqbbEZ68p7ujPr3e7MGNWVTtfnXAm7yrG4BEIc/saDQypl+zce4vA37uNTgcSE0xyLTSQkpLJz2bgEp7JLFmykS/esFeamDX8TEOhLjZqZ2wCucSw2keCQTK9DXGKWmJAQx9ehIokJZ5xili/ZwjUNwylXzrmTcsnCTXTsGolI6Wr9OVJ3yfbhKkXZzXszMEtVTwOIyEygrcP2dsB3qpoGHBGRnzOV/87h37H283DgBxEJA8oB+7I5bm4x81T1PHBeROKAEOBWYPrF67OqesKO7QQ0cvjj8RORLNd6c/CDfc6VAF9VXW2v/y/QK7sC9rRAQwA+GP8EAx/qms9DFY8vP13M/93fDi+v8i7Nwyg5vW6qycI1B0hPvzRK2i1PzCL2xFmqB/sw5Y1O/HUwgYOxV2YrPrOUlFSiVmzjiWduy7Jt8YINdC0lrdKisndPDBPHzWfcpEeybFu6cDNvjL7XBVldBnPNNEeazfNPgI9UdY6ItMeaYy6z3GLOOzxPI/fzdcNq4Z5zXCkiO4AWwB8Oq1sAjn0qp3PZb7Ycpwk6dm5OrmM+5ldQsB9xMZe+aR+LSyQoxD9TjD9xMQkEh1QiNTWN08nn8K/kxY6th1ixdCufjptH8qmziAjly3lyxz2XNaKW4WKxJ84Q5tASDQ3wIjb+TLaxvW6qxRtf/J6p/FkADsUls3ZHLI0iAkpVZRoUXIlYh7/xuNiEbP/GY2MSCAmtTGpqGsnJ5/Cv5E1QiD+xsSedywZfat399ssOGjQMJ7CKn9P+UlPTWLF0C9/8UHpuvgsK8ScuNtPrEOyfJSY2NoHg0Er263AW/0rW30ZcTAKvPPc1r4+6m/DqVZzK7d51hLS0dBo0Kn09E05K2d28RVm1/wL0FREvEfHG6tb9xWF7FNBfRNztVmSHTOX7O/x7sWXnz6WpcR7I4bj5iXH0M3CXiAQCiMjFiyOLcZgAVkQi7acfAK+KSC17fS1gGPBh5h2ragJwSkRa26vuzkc+RaZB4+ocOnicI9EnSElJZenCzdx0SyOnmJvaN2LBHKvXesWSrVzXqi4iwsSvn2DGgmHMWDCMu+5ry4CHbzUVaRm0ZU88NcN8CQ/2xtPDjZ431WLZ+ugscbWr+uHnXY5Nf126gd7Puxzl7N/uVfYtT4trgpxuXCoNGjWpwaEDxzgcHU9KSiqLF2ykbfumTjHt2jdh3hzrS8LPSzbTslU9RIS27ZuyeMFGLlxI4XB0PIcOHKNx05oZ5Rbn0MW7bs0uakYEExJaOcs2V2nYuDqHDhzniP06LF24mbbtGzvF3Ny+MfPnrAes7twW9nv9VNJZ/jH03zzxTE+aNY/Isu8lCzbRuVvzLOtLG/Vwy/bhKkXWMlXVjfbPXS5+1f1CVTc5dJvOwupi3YE1vc3qTLuoLCJbsFqTFyf4HAFMF5GTWJVg1v/5/MU45rldREYBK0UkDdgEDAKeBibYOXhgVf6PqepmEXkZ+ElEPIEU4CVV3ZzDIR7Cmm0gHVgJlNinkYeHO8+/2pfnH/+c9PR0evZtRe26oXwxYRENGodzc/vG9OrXirdf+57+vd7Fz8+LEe/fl+d+33h5KpvX/01Cwmn6dR7JQ493odftrUrgjIrHN588Rds2DalS2Zc9a8fz9kcz+OaHFa5Oq0ikpStv/nsdX73WEXc3Yfryv9kdncgz/a9l298nMirWXjfVYt5v+53K1qnmx8hHW5OeDm5uMGn29lJXmXp4uPPisDt5+rGJpKelc1u/G6hTN4xJ4+fRsHEN2nVoSu/b2/DGq1O4vcdb+Pl7Mer9QQDUqRtGp67N6d9nNO4e7rz02l24212FZ8+cZ+3qnbw6vH+WYy5esLHU3Hh0kYeHO/8Y1o9nH/+c9DSlV9/rqV03lMkTFtKwUXXadmjMbf1a8eaw77iz5zv4+Xvx9vvWXfszvl9F9MHjfDlpCV9OWgLAuM8eISDQF4Bli/7gw4kPu+zc8q2UtUwlj1lljMskIj6qmmw/fwUIU9VncitTVN28ZY2ZHPzqtOGqnhz8gqtTcJmA8rcVae1X681F2X5u7n+ja57HEZFuWDedumM1/N7NJub/sBprCvyhqrleRC5N10yvFD1F5FWs1/YAVqvXMAzDKEIFHeheRNyBCUBnIBpYJyJz7GnXLsbUA14FblLVkyKS523cpjItYqr6A/bdvYZhGEYxKXg3bytgj6ruBRCR74E+WJcgL3oEmKCqJwFUNS6vnZaue4sNwzAMIz883LJ9iMgQEVnv8BiSqWQ14JDDcrS9zlF9oL6IrBKRNXa3cO7pFO5sDMMwDKPk5TRAg+PPDgvBA6gHtMcayyBKRJrav9jIlmmZGoZhGGVPwYcTPAxUd1gO59LPKy+KBuaoaoo9hOxfWJVrjkxlahiGYZQ5bu7ZP/JhHVBPRCJEpBzWeABzMsXMxmqVIiJVsLp99+a2U9PNaxiGYZQ57gW8AUlVU0VkKLAI66cxX9rjD7wFrFfVOfa2LvYIeGnAi6oan9t+ze9MS4W/rsr/hLp3/Z530BXq8Np5rk7BZY7sznugkCtVpfK59hRe0YRrivR3pvUmRWX7ubn70XYuGc3BtEwNwzCMMiefXbolxlSmhmEYRplTyiaNMZWpYRiGUfa4uZWusXlNZWoYhmGUOaab1zAMwzAKyc108xqGYRhG4ZhrpoZhGIZRSG6lbD5Tl1WmIjICSFbVDy6jTLKq+hRfVk7HCseapqcR1khRc7F+uJtlQkIRqQr8S1XvLInc8ktVGTVqMitXbqBChfK8++4zNG5cN0vc2LHfMnv2cpKSktm0aXrG+nXrtjF69Ofs2rWfjz56iW7dbirJ9AulXWQY/xx8Pe5uwrRle5g0e7vT9tceaEHrJiEAVCznQaB/Ba4bNA2AXT/cy66D1hCcR4+f4dH3VpRo7sXpszGP0r1jc47FJ9Gy80uuTqfQVv+6k7HvzSY9PZ3et7dm4EMdnbZfuJDKm6/9l107ovHz92bkmAFUrRbA2tW7mDhuPqkpqXh4evDU871o2dr5N6AvPPVvjkSf4L+zXizJUyo0633/OVEr11OhQnneefdZGjeukyVu7Ngp/Gi/7zdumuaCTAuntHXzlrJ0SgcREWAmMFtV62ENJeUDjMom1kNVj5S2ihQgKmoD+/cfYfHiSbz99pOMGPFptnEdOrRi+vQPs6wPCwvinXeepVevW4o71SLl5iaMeKgVD436mW7P/USvm2pRN9zfKWbUNxvo/eJ8er84n28X7GLR2oMZ285dSMvYdiVVpABTpq+kz8As8yCXSWlp6XwweiZjP32E72a/xOIFm9j3d4xTzJyZa/Hz82LGvGHcM6AdE8bNBaBSJW8++ORBps58keEj7+bN1/7rVG750i14eZUvsXMpSlFRGziw/wiLFk/irbef5M0c3/fXM216vtsypU7Bh+YtHvk+tIjUEpE/ReRzEdkuIotFpKKI1BGRhSKyQUR+EZEGIuIuIvvEUklE0kSknb2fKHviVYBmIlDMkAYAACAASURBVLJaRHaLyCP2dh8RWSYiG0Vkq4j0ySaXbGNyytHeVldElorIH3a5Ovb6F0VknYhsEZE37UPcCpxT1a8AVDUNeA54UES8RGSQiMwRkZ+BZfZxt9n78xKRaSKyQ0RmichaEWl5+f81hbds2Rr69r0VESEysgFJSaeJizuRJS4ysgHBwQFZ1oeHh9CgQUSpuwU9L83qBnIg5hSH4pJJSU1n3qr9dGoZnmP8bTfXYu6q/SWXoAut+n0nJxKSXZ1Gkdix7SDhNQKpFh6Ip6cHnbs1J2q5cw/ELyu20aO39fbr0Pla1q/djapyTcNwgoKtL1i164Zy/lwKFy6kAnDmzHm+m7KSwUM6lewJFZFly9bSp2+HAr/vy4pCjM1bPPlcZnw9rAlTGwMJwB1YU908paotgBeAiXblswuri/RmYCPQVkTKA9VVdbe9v2uxKq42wHC7u/Qc0E9VrwM6AB/aLUVHucVklyPAVHt9M+BG4KiIdLHjWwGRQAu70m8MbHA8oKomAQeBi/2k1wF3qmrmZtsTwElVbQS8DrTI81UtJrGx8YSGVslYDg0NJDY21+ElrwghAV4cjT+TsRxz4gwhgV7Zxlat4k14sA+rt8VmrCvv6c6sd7szY1RXOl2fcyVsuNax2ESCQyplLAeH+HMsLjFTTBIhdoyHhzs+PhVJTDjtFLN8yRbqNwynXDnrqtfk8Qu5d2B7ylcoV8xnUDxiY+MJCw3KWL5S3/dubpLtw1Uu95rpPlXdbD/fANTCqpimO9R3F/tGfgHaARHAO1gzl6/EGrH/oh9V9SxwVkSWY1Vq84DRdqWWjjVpawjg2H8jOcRkm6OI+ALVVHUWgKqeA7Ar0y7AJjvehzym2XGwRFWzft2zvjx8bB9nm4hsyef+DBfodVNNFq45QHr6pWE+b3liFrEnzlI92Icpb3Tir4MJHIy9MlpzhrO9e2KYMG4eH0+y5o/+a+dhog8d59mX+nDkcHZvb6O0KOvXTM87PE8DAoAEVY10eDS0t0cBbbEqyPlAJawpbX5x2EfmgYoVuA8IAlqoaiQQC1TIFJdbTOYcc/vCIMA7DrnXVdV/AzvI1KIUET+gBrDHXuX89fYyOc4GP3nyD4XZlZOpU+fRp8/T9OnzNEFBAcTEHM/YFhMTT0hIYJEdq7SKPXGGMIeWaGiAF7EOLVVHvW6qxU+/7s9U/iwAh+KSWbsjlkYRZbcr7EoWFOJPXOyluZrjYhMzum4vxfgRa8ekpqaRnHwW/0reVnxMAi8/9xXDR91DeHWrB2frHwfYuSOavt1G8ugD4zl44BiPPzixhM6o4KZOnUffPs/Qt88zBAcFcDTmWMa2K/V9X2avmeYgCdgnIneBdeOOiDSzt/2O1WpNt1uCm4FHsSrZi/qISAURCcSqaNcB/kCcqqaISAegZjbHzU9MBlU9BUSLSF87z/Ii4oU1zc6DIuJjr68mIsHAMsBLRAba692BD4GvVTX7T+VLVgH/Z5drBDTNIafJqtpSVVsOGdI/j13m33339eTHH//Fjz/+i06dbmD27J9RVTZv3omvr1eZvkaSX1v2xFMzzJfwYG88PdzoeVMtlq2PzhJXu6offt7l2PTXpS8cft7lKOdhvS0q+5anxTVB7IlOzFLWcL2Gjatz6MBxjkTHk5KSypKFm2jbvrFTTNv2jZk/Zz1gdee2bFUPEeFU0lmeH/oFTzzTk2bNIzLi7+h/I3OXvcHshf9k0jdDqVEziE+/fKJEz6sg7ruvJ7N//JjZP35Mx06t+XH28iv+fe/hptk+XKUo6vH7gIdE5A9gO9AHQFXPA4eANXbcL4AvsNWh7BZguR3ztqoewbq22VJEtgIDgZ3ZHDM/MZkNAJ62u11/A0JVdTHwX2C1va8ZgK9a89L1A+4Skd1Ys6yfA4bl4zgTgSB7HryRWK+JSz6Nb7mlJdWrh9K58xBef308b7zxeMa2Pn2eznj+/vtf0a7dIM6ePU+7doP45BPrzsYtW/6iXbtBLFy4ijfemEDPnqX/QwUgLV1589/r+Oq1jiwaexvzVx9gd3Qiz/S/lo4ONyL1uqkW837b71S2TjU/Zr3XnZ/G9OQ/Izoxafb2K6oy/eaTp1gx+y3q1w5jz9rxPNC/vatTKjAPD3deGHY7zzw+mbv7vE/HLpHUrhvK5AkLiVq+DYDb+rUmMeE0d/YczXdTVvLEsz0BmP79r0QfjOfLSUsYcNeHDLjrQ07En3Ll6RSZi+/7Lp0f5fXXJzD8jccytvXt80zG8zHvf8Ut7QZz9ux5bmk3OON9X1Z4SPaP/BCRbiKyS0T2iMgrucTdISKan5tIzXymRcxuxXqq6jn7juGlwDXZ/T71EjOf6dXGzGd6dTLzmRadfkt/yfZzc1antrkex/6M/gvoDERj9Yjeo6o7MsX5Yt3DUw4Yqqrrc9uvGQGp6HkBy0XEE+ua7BO5V6SGYRjG5XIveJduK2CPqu4FEJHvsXpUd2SKext4D8jXqB2mMi1i9vVZl/yu1DAM42qRU5euiAwBhjismqyqkx2Wq2FdgrwoGmidaR/XYf2Mc56ImMrUMAzDuDLldLORXXFOznZjPoiIG/ARMOiy8inoAQ3DMAzDVfJ7s1E2DgPVHZbD7XUX+QJNgBX2+AmhwBwR6Z3bdVNTmRqGYRhlTiF+BrMOqCciEViV6N3AvRc3qmoikDF0nIisAF4wNyAZhmEYVxwPKVhlqqqpIjIUa5wBd+BLVd0uIm8B61V1ToHyKVA2hmEYhuFCHoUYJUFV52ONzOe4bngOse3zlU/B0zEMwzAM1yhoy7S4mMq0FKj5zsG8g65AW6YG5R10hUrXq3fggqr1pro6BZcJq3K9q1Nwmb0brynS/RXiBqRiYSpTwzAMo8wpxKANxcJUpoZhGEaZY1qmhmEYhlFIrpwhJjumMjUMwzDKHE/TMjUMwzCMwjEtU8MwDMMoJM+imI27CJnK1DAMwyhzruiWqYgkq6pPNuu/Buaq6gwR2Q+0VNXj+dznIDt+aFHmmsuxxmBNyeMD7AXeVNXfCrCvSKCqPdKGS9xSO5A3OtXH3U34fvNhPl1zIEtMzwbBPNe2NqrwZ1wyT8/ZBsA3/SNpXtWf9dEJPDj9j5JOPd9W/7qDD9+bSXpaOn1ub8MDD3d22n7hQgojhv2HnTsO4V/Jm1FjBlG1WiAAX3+xmDkz1+Dm7sY/XrmDNjc15MC+WIa9+HVG+SPRxxnyZA/uGdCBYS98xYH9cQAknzqLj29Fps54ucTONTerf93J2Pdmk56eTu/bWzPwoY5O2y9cSOXN1/7Lrh3R+Pl7M3LMAKpWC2Dt6l1MHDef1JRUPDw9eOr5XrRs7TyB9QtP/Zsj0Sf476x8zURVan025lG6d2zOsfgkWnZ+ydXpFKl2N9Zi+AvtcXN3Y9qsrXz29Tqn7VVDfRnzZjf8fMvj7i68/69fWbFqHx4ebrzzemeaNAjB3UOYNXcHn361LoejlC7mmmnp98PFiltEOgAzRaSDqv55mfuJxJrX1CWVqZvA212u4b7vNxGTdI45g1qxdPdxdsefzoipVbkiT7aJ4PYp60k6l0qgl2fGtslrDlDB0537mldzRfr5kpaWzvujpjN+8pMEh1bigbs/oG2HJtSuE5YRM2fmGnz9vJg5fziLF2xg/Ng5jP5gMHv/PsriBRv5fvarHItLYugj45kx93VqRoRkVJBpaen07Pg67Ts2A2D0B4Mz9jtuzCx8fCqU7AnnIC0tnQ9Gz+Rfkx8lOMSfwfeMo237xkTUCc2ImTNzLX5+XsyYN4wlCzYxYdxcRo0ZSKVK3nzwyYMEBfvz9+6jPPv4ZH5a+kZGueVLt+DlVd4Vp1XkpkxfyWffLOKLsU+4OpUi5eYmvPnyrQx84n/ExJ5i9n/uY+nKv9mz70RGzJMPt2b+kl1MnbGFuhEBfPlJP9r1+jc9OtWnXDl3uvf/lgoVPFg84wHmLNzF4aNJLjyj/CnvXrpapgXudRaR50Vkm/14NtM2EZHxIrJLRJYCwZmKvyQiW0XkdxGpa5e5TUTWisgmEVkqIiHZHDPbGBEZISJfisgKEdkrIk87lBkoIltE5A8RmWKvCxKR/4nIOvtxU3bnqKrLsebFG2KXqyMiC0Vkg4j8IiIN7PV32a/DHyISJSLlgLeA/iKyWUT6F+xVLrjIqv7sP3mWQwlnSUlXfvozls71nUccuieyGt9uPETSuVQA4s+kZGxbdeAkpy+klWjOl2v71gOE1wiiWvUqeHp60KX7dUQt3+oUs3L5Vnr2bgXArZ0jWbf2L1SVqOVb6dL9OsqV86RaeCDhNYLYvtW55b5u7S7Cq1chrGqA03pVZemiTXTp0aJ4TzCfdmw7SHiNQKqFB+Lp6UHnbs2JWr7dKeaXFdvo0duas75D52tZv3Y3qso1DcMJCvYHoHbdUM6fS+HCBevv4cyZ83w3ZSWDh3Qq2RMqJqt+38mJhGRXp1HkmjUJ5UB0AocOJ5KSms7cRTvp3L6OU4wq+HhbX4p8fcsTe+y0vV7xquiJu7tQobwHKSnpJJ++UOLnUBAebtk/XJZPQQqJSAtgMNbs5AKsFZGVDiH9gGuARkAIsAP40mF7oqo2FZGBwDigF/ArcIOqqog8DLwE/CPToXOLaQB0wJqLbpeIfArUB/4J3Kiqx0Xk4qfix8BYVf1VRGpgzR7QMIfT3Qg8aj+fDDymqrtFpDUwEbgVGA50VdXDIlJJVS+IyHBKqHs6O6E+5TmadC5j+eipczSv6u8UExHgBcD/BrTETYRxv+5l5d74Es2zMI7FJRASWiljOTikEtu3HMgUk5gR4+Hhjo9PBRITTnMsNpEm19ZyKnssLsGp7JIFG+nSPWuFuWnD3wQE+lKjZubviK5xLDaR4BDH18Gf7VsPZopJIiTE8XWoSGLCaSpVvnRVZvmSLdRvGE65ctbHwuTxC7l3YHvKVyhXAmdhFFRokA9HY05lLB+NSyaySZhTzMeTVvPthDsYeHckXhU9GfD4DAAWLNtNp/Z1WLP4USpW8GTkhytIdPjcKM08S9k104LW4zcDs1T1tKomAzOBtg7b2wHfqWqaqh4Bfs5U/juHf9vYz8OBRSKyFXgRaJzNcXOLmaeq5+1rsXFYlfitwPSL12dV9WK/RydgvIhsBuYAfiKS5VqvTQDs7TcC0+1yk4CLf7GrgK9F5BGsKX3yJCJDRGS9iKxP/n1efooUOQ83oVZlL/pP3cDTP27l3e4N8Stvev4BUlJSiVqxjY5dIrNsW7xgA11LSau0qOzdE8OEcfN4ZfidAPy18zDRh47TvmNTF2dmFIXeXa9hxk/buan75zz49Cw+fLs7ItCscSjpaUqbrpO5pdcXPHx/C6pX8897h6WAh2T/yA8R6Wb3nO4RkVey2f68iOywezWXiUjNvPbpqkaxZvP8E2C8qjbFaglmd0Eqt5jzDs/TyL3V7YbVwo20H9XsLwXZaQ78aZdJcCgTqaoNAVT1MawWcHVgg4gE5nJs7DKTVbWlqrb0adUzr/DLFpN8njC/Sy9PmG8FYk6dd4o5mnSepbuPkZquHEo8x74TZ6hlt1bLgqDgSsTGXGpNxsUmEBTinynGPyMmNTWN5ORz+FfyJijEn9jYk85lgy+17n77ZQcNGoYTWMXPaX+pqWmsWLqFTl2bF8cpFUhQiD9xsY6vQ2JG1+2lGD9iYx1fh7P4V/K24mMSePm5rxg+6h7Cq1tzIm/94wA7d0TTt9tIHn1gPAcPHOPxByeW0BkZlyPmWDJhob4Zy2HBPsTGnXKKuatvE+Yv2QXApi1HKV/OnYBKFendvQErV+8nNTWd+JNn2fDHEZo2ynKFrVTydNNsH3kREXdgAtAdq/f0HhFplClsE1bP4rXADOD9vPZb0Mr0F6CviHiJiDdWt+4vDtujsK4XuotIGFb3q6P+Dv+utp/7Y816DvBADsfNT4yjn4G7LlZuDt28i4GnLgbZd95mISK3YF0v/VxVk4B9InKXvU1EpJn9vI6qrrXnwzuGVamewupydok/jiQRUbki1f0r4Okm3NYwhCW7jznFLN59jBtqVgagckVPIgK8OJhw1hXpFkijJjU4dOAYh6PjSUlJZfGCjbRt79ySate+CfPm/A7Az0s207JVPUSEtu2bsnjBRi5cSOFwdDyHDhyjcdNLXz4X59DFu27NLmpGBBMSWrl4T+4yNGxcnUMHjnPEfh2WLNxE2/bOHTtt2zdm/pz1gNWde/F1OJV0lueHfsETz/SkWfOIjPg7+t/I3GVvMHvhP5n0zVBq1Azi0y+vrBt3rhRbtsdQq3olwqv64enhRq+uDVi6cq9TzJGYU9zYqgYAdSICKF/eg/iTZzly9BQ3Xl8dgIoVPIhsGsbe/SeyHKM0KsQ101bAHlXdq6oXgO+BPo4BqrpcVc/Yi2uwekVzz+fy0s840Eb75y6/26u+UNVNIhlt7FlYXaw7gINcqjAvqiwiW7Bak/fY60ZgdaGexKoEI8gqPzGOeW4XkVHAShFJw/q2MQh4Gphg5+CBVfk/ZhfrLyI3A17APuAOhzt57wM+FZF/Ap5Y/wl/AGNEpB5Wl/Aye91B4BW7S/gdVf0ht1yLWpoqw5fs4tu7m+MuwrQtR9h9/DTPt63NlqNJLN1znJV742kXEcDSR24gLR1G/7ybhLPWTUjT729BnUBvvD3dWfPkzbw0fwdR+0rXm8zDw50Xh93J049NJD0tndv63UCdumFMGj+Pho1r0K5DU3rf3oY3Xp3C7T3ews/fi1HvDwKgTt0wOnVtTv8+o3H3cOel1+7C3d16J549c561q3fy6vCs940tXrCx1Nx4dJGHhzsvDLudZx6fTHqa0qtvK2rXDWXyhIU0aBROuw5NuK1fa94c9l/u7DkaP38v3n5/AADTv/+V6IPxfDlpCV9OWgLAx58NISDQZd8Di803nzxF2zYNqVLZlz1rx/P2RzP45ocVrk6r0NLSlBHvLeebCXfg5iZMn7ON3XvjefaxG9m6I4ZlUXsZ/dFKRr/emQfva4Gq8uIbiwCYMm0z74/oysLpAxERZszZzs7d+frVosvlNGiDiAzBvmnUNllVJzssVwMOOSxHY93/k5OHgAV55SOqpesi7tWo5jtLr8r/hC3/KN13CxendE3JO+gKZeYzvTrt3fh8kf4ydOnh+dl+bnaq1iPX44jInUA3VX3YXh4AtM7uZlERuR8YCtyiquczb3dk7jYxDMMwypxCDCd4GOtS3EXhXLp8mEFEOgGvkY+KFExlahiGYZRBhRhOcB1QT0QisCrRu4F7HQNEpDnWLza6qWpcvvIpaDaGYRiG4SoFnRxcVVNFZCjW+ALuwJf2/TVvAetVdQ7WsLI+WPfoABxU1d655lOwdAzDMAzDddwKcQXWHjN9fqZ1wx2eX/awX6YyNQzDMMocdzPQvWEYhmEUjpuUrh9BmMrUMAzDKHMK081bHExlWgq4Hz6Vd9AVqPnzsP7Dq/NPsHL5+q5OwWWu5t9aHj1eNuYKLQtMN69h2K7WitQwjMIzLVPDMAzDKKSC/jSmuJjK1DAMwyhzzA1IhmEYhlFI5pqpYRiGYRSSuWZqGIZhGIVkKlPDMAzDKCTTzWsYhmEYhXTV34AkIiOAZFX94DLKJKuqT/Fl5XSsNGAr4AmkAt8CY1U1vQD7Gqaqo4s4xXxr1zCY4Xdei5ubMO23A3y25K8sMT2aV+OZHg1QYOfhRJ79ej0AL/dpTIcmobiJ8OvOON6asaWEs798q3/dybj3fiQtPZ3et7dm4EO3Om2/cCGVt177jp07ovH392LkmAGEVQvg99V/MXHcPFJS0vD0dGfo871o2boep0+f4/FBEzPKx8Um0LVnC557uU9Jn1qBqSqjRn1O1Mr1VKhQnnfefZbGjetkiRs7dgo/zl5OUlIyGzdNc0GmRaPdjbUY/kJ73NzdmDZrK5997TxIQtVQX8a82Q0/3/K4uwvv/+tXVqzah4eHG++83pkmDUJw9xBmzd3Bp19dOQMsfDbmUbp3bM6x+CRadn7J1ekUidLWzVvw6VWvXGdVNVJVGwOdge7AGwXc17CiS+vyuAm8+X/NGDzxN7qOXMptLcKpG+rrFFMryJvHu9Tnro+i6DZqGW/bFeZ1EQG0qB1Ij9HL6DZqKdfWrEzrelVccRr5lpaWzoejZ/HRpw/z3ewXWbJgE/v+jnGK+WnmWnz9KjJj3qvcPaAdE8bNA8C/kjdjPnmQqTNf4PWRd/Pma98B4O1dgW+nP5/xCA2rTPuOTUr83AojKmoDB/YfYdHiSbz19pO8OeLTbOM6dLieadPz/f22VHJzE958+VYGPzWLrnd8zW3dGlA3IsAp5smHWzN/yS5uu/c/PP3KPN561frC1aNTfcqVc6d7/2/pfd9U7rnjWqqF+bniNIrFlOkr6TPwXVenUaTcJftHfohINxHZJSJ7ROSVbLaXF5Ef7O1rRaRWXvvMszIVkVoi8qeIfC4i20VksYhUFJE6IrJQRDaIyC8i0kBE3EVkn1gqiUiaiLSz9xMlIvXs3TYTkdUisltEHrG3+4jIMhHZKCJbRSTL1/+cYnLK0d5WV0SWisgfdrk69voXRWSdiGwRkTezO3d7UtghwFD7nNxFZIxDuUftfYXZ57dZRLaJSFsReReoaK+bmtfrXNSa1QrgwPHTHIo/Q0qaMndjNJ2vDXOK6X9jLaZE7SXpbAoA8ckXAFCgvKcbnh5ulPNwx9NdOJ6U50TzLrVj20HCawRSLTwQT08POnWLJGr5dqeYX1Zsp0fvlgB06Hwt69fuRlW5pmE1goL9AahdN5Tz51K4cCHVqezB/cc4eSKZyBa1S+aEisiyZWvp07cDIkJkZAOSkk4TF3ciS1xkZAOCgwOy2UPZ0axJKAeiEzh0OJGU1HTmLtpJ5/bOrXBV8PEuD4Cvb3lij5221yteFT1xdxcqlPcgJSWd5NMXSvwcisuq33dyIiHZ1WkUKcEt20ee5UTcgQlYDaVGwD0i0ihT2EPASVWtC4wF3strv/nt5q0H3KOqj4jINOAOYDDwmKruFpHWwERVvVVEdtkJRgAbgbYishaobscCXAvcAHgDm0RkHhAH9FPVJBGpAqwRkTmq6tgxfi67mFxy/A8wFXhXVWeJSAXATUS62PGtAAHmiEg7VY3KfOKqutd+8YOBPkCiql4vIuWBVSKyGLgdWKSqo+xYL1X9RUSGqmpkPl/jIhXqX4GjJ89mLB89eZbIWpWdYiKCrZ7zac+1w91N+Hj+n0T9GcemfSdYs/s4a0d1R0T4Nmovf8eW7vGDj8UmEhxSKWM5OKQS27ceyBITYsd4eLjj41ORxIQzVKrsnRGzfMkWrmkYTrlyzm+NJQs30bFrJPbfb5kRGxtPWGhQxnJoaCCxsfFlvuLMTmiQD0djLv2dHo1LJrKJ8xfIjyet5tsJdzDw7ki8Knoy4PEZACxYtptO7euwZvGjVKzgycgPV5CYdK5E8zcuj/VRWyCtgD2qutfaj3yP9dm+wyGmDzDCfj4DGC8ikqk+cpLfbt59qrrZfr4BqAXciDUL+WZgEnDxr/YXoJ39eAe4GbgecLwA8aOqnlXV48ByLlVqo0VkC7AUqAaEZMojt5gsOYqIL1BNVWcBqOo5VT0DdLEfm7Aq/AZYlWteugAD7XNeCwTa5dYBg+3rwU1VNc+aR0SGiMh6EVmftH1xPg5d9Dzc3agV7MO9H//CM1+vY/S9zfGt6EnNKt7UDfHlxn8upM1rC2hTP4jr6wS6JMeStHdPDBPHzefl4Xdk2bZ04Wa69GjugqyMotS76zXM+Gk7N3X/nAefnsWHb3dHBJo1DiU9TWnTdTK39PqCh+9vQfVq/q5O18hFQVumWPXGIYflaHtdtjGqmgokYn3e5yi/LVPHPr40rAosIYdWVxTwOFAVGA68CLTHqmQvyly7K3AfEAS0UNUUEdkPVMgUl1tM5hwr5nI+AryjqpNyibECRWrb+4uzyz2lqouyiWsH9AS+FpGPVPXb3ParqpOByQC1h84q8tvSYhLPEVb50ksQVrkisYnO37RjEs6yef8JUtOV6Pgz7I9LJiLIm9b1gti0/wRnLqQBsHJ7DM0jAlj3d3xRp1lkgkL8iYtNyFiOi03I6Lp1jImNTSA4tBKpqWkkJ5/Fv5KXFR+TwCvPfc3ro+4mvLrz9eHdu46QlpZOg0bhxX8iRWDq1HlMn2Z9QWvatB5HY45lbIuJiSck5Mr8YhRzLJkwh/sCwoJ9iI1z/l57V98mDB46E4BNW45Svpw7AZUq0rt7A1au3k9qajrxJ8+y4Y8jNG0UwqHDiSV5CsZlyKllKiJDsC7PXTTZ/rwtVgW9ASkJ2CcidwHY1xOb2dt+x2q1pqvqOWAz8ChWJXtRHxGpICKBWBXtOsAfiLMryQ5AzWyOm5+YDHYLMVpE+tp5lhcRL2AR8KCI+Njrq4lIcObyIhIEfAaMt5v3i4DHRcTT3l5f/r+9+46XqyrXOP57DqEFCB0shBqaoUtAEAhFQKQrKhoUaTakiYiggpSLV68oiMqly0Us4OUSxBAIxSRShFCSUCI9lBCBSAk9hOf+sfeQyeFQ9JxZK9nr/fI5n8zeZ5I8w+dk1qx3v2ttaSFJKwD/sH0WcDawQf1HzGg9N7UJk59hxaUXZrkl+zPvPGKnDZbj6glPzPacq8ZPYeNVqxLg4gvNx4rLLMwj015iyjMvsfGgpZinS/TrEhuvuhT3T52zy7xrDh7Io5OfZspj05gx43WuHnkHm285eLbnbLblYEZcVnUrXzdqAh/eaBCSmP78yxz+jXP4+iE7su76K73lzx51xe1s+/G5Z1Y6bNiOXDr8VC4dfirbfGxjhl96Hba5445JLLJI/0aWeAEm3DWVFQcuxnIfGMC8/brYafs1uHr0g7M9Z8rU6Wy60fIArLLSEsw/fz+mPfMyS6QGRQAAFHVJREFUU56YzqZDBgKw4AL9WG/t9/Pgw2+9thzmHF1v85/tM21v2PbVfSB9HBjYdrxcfa7H50jqRzX2vONsojdLY4YBp0v6HtUykt8D422/KulR4Kb6eWOBz1EtN2mZQFXeXQo4wfaUuknnT5ImAuOAST38ne/lOd19AThD0vHADODTtq+StCZwY30N7AVgL6rZ54J1Gbe1NOYC4Kf1n3U2VYn7NlW/8SlgN6oPBEdImlH/WV+sn38mMEHSbbaHvYesfWbmG+YHF43n/AM/Spfg4psmc9/U6Ry645pMfOQZrpk4lTH3PMnmay7Lld/dhjds/vPSO3n2xde44vbH2WS1pbni6G2wzZh7nuTaO6e++1+aUb9+83D40btz6NfO4o2ZZqfdhrDyoPdx5i9HsuaHBrL5VoPZefeNOO7o37HHjj9kwKL9OeHHewHwx99fz2OPPM25Z4zi3DNGAXDKfx/AEktWs5xrrhzPyb/aP9tr642hQzdkzOhb2W7br7DAgvNz0kkHv/m93XY9hEuHnwrAf/34PC6/fAwvv/wqQ7fYhz0+vS0HHfT5XLH/LTNnmh/86DrO/+Wn6OoSF192J/c9OI1Dv7opE++eyjVjHuSkn47mpO9vy77DPoxtjji2KjJdcNEd/PgH2zPy4i8iiT9edheT7ns68yvqO+efdhCbb7ImSy2+CPf/7Rec8NM/cv4f/pI7Vq/04prpLcCqklaiGjT3BLr/sF8G7A3cCOwBXPtO10sB9C7fDwl0osw7Nyj5fqYl3xx8lQ3+nDtCNiXfHPzlR37Xp917L70+tsf3zf79Nn/Xv0fSJ4BTgHmAc+vm0eOBcbYvq5tVLwDWB/4J7NlqWHo75b6bhRBCmGuJf3tmiu0RwIhu545pe/wK8Ol/5c+MwTSEEMJcp0tz1vA1Z6UJIYQQ3gNpztrALwbTEEIIc53elHk7IQbTEEIIc533uEFDMjGYhhBCmOv0YmlMR8RgGkIIYa4TM9MQQgihl+a0mWls2hCQ9OUUe1fOaUp93RCvPV576Gtz1jw55PLld39KI5X6uiFee6lKfu0dFYNpCCGE0EsxmIYQQgi9FINpgPq+qgUq9XVDvPZSlfzaOyoakEIIIYReiplpCCGE0EsxmIYQQgi9FINpCCGE0EsxmBZI0krv5VwITSDpLTd57ulcCL0RDUgFknSb7Q26nbvV9odzZUpF0kSg+w/9c8A44ETb09KnSkPSBj2cfg6YbPv11HlSeZuf97ecaypJF9j+wrudC70Te/MWRNIawGBgUUmfbPvWAGCBPKmSuwKYCfy2Pt4T6A9MBX4N7JwnVhK/AjYAJgAC1gLuovp5+Jrtq3KG62uSdgA+AXxQ0s/bvjUAaOyHhx4Mbj9Qtalt4z84pxaDaVlWB3YCFmP2QWM6cECWROl9rNuMZGJrliJpr2yp0pgC7Gf7LgBJHwKOB74NXAI0ajCler3jgF2AW9vOTwcOy5IoIUlHAUcDC0p6vnUaeI1Yb9rnosxbIEmb2L4xd44cJI0HDrB9c308BDjb9rqSbre9ft6EnSPpTttr9XRO0h2218uVrZMkzWt7Rv14cWCg7QmZYyUj6Ye2j8qdo+liZlqm+yUdDaxI28+A7X2zJUpnP+A8SQvXx9OB/SQtBPwwX6wk7pJ0OvD7+vizwN2S5gdm5IvVcaMk7UL1s34r8KSkG2w3fnZau1zSQrZfrKsvGwCn2p6cO1iTxMy0QJJuAMZSvbHMbJ23/b/ZQiVQXys62PbPJC0KYPu5zLGSkbQg8HVgs/rU9VTXUV8B+tt+IVe2TmpVHCTtTzUrPVbSBNvr5M6WgqQJwLrAOlR9AWcDn7E9NGeuponBtEBNLum9G0k3294od46QTt3BvR1wPvBd27cUNpi2egKOAR63fU5J3cypRJm3TJdL+oTtEbmDZHC9pF8AfwBebJ20fVu+SJ0l6SLbn3mbZUEUMKgcD1wJXF8PpCsD92XOlNL0uhnpC8DmkrqAeTNnapyYmRZI0nRgIeBVqmtlAmx7QNZgCUi6rofTtr118jCJSHq/7SckrdDT9+PaWbNJeh/weeAW22MlLQ9saft/MkdrlBhMQyhAfb34attb5c6SmqTVgNOBZevO5XWAXWyfmDlaMvUHqVVtXy2pPzCP7em5czVJbCdYIElb9PSVO1enSVpL0vmSxtVf50taO3euFGzPBN5oNV4V5izgKOqO5XpZzJ5ZEyUk6QDgj8AZ9akPApfmS9RMcc20TEe0PV4A2Iiqs7fJpc5dgZ9QLX85uT69IXCJpG/ZHp4tXDovUG1SMYrZrxcfnC9SEv1t3yyp/VxJOyAdSPVv/G8Atu+TtEzeSM0Tg2mBbM+2ZZ6kgcApmeKkcjywre2H285NkHQtMLz+arpL6q/SPC1pFermK0l7AE/kjZTUq7Zfa32YkNSPHhrRQu/EYBoAHgPWzB2iw/p1G0gBsP2wpFI6G/8ADKof32/7lZxhEjqQavu8NSQ9DjwEDMsbKanR9SYtC0ralmqt8Z8yZ2qcaEAqkKTTmPXJtAtYD3jYdmP3pq23EdzZ9iPdzq8A/KnJy0PqmchJwL7AZKru7YHAeVTrLpu8+9Gb6l2uukprvKmXwuxHtdZWVMuEzna8+fepGEwLJGnvtsPXqQbS63PlSUHSbsCPqQaV1qbnGwLfAY603diGDEk/AxYBDmsNJJIGUF1Dftn2ITnzdZKk1YEvA2vUp+4BzrR9b75U6UlaGsD2U7mzNFUMpoWSNB+wWn349xJmJ5LWBQ5n1i2p7gJOtj0+X6rOk3QfsFr3mUi9XGaS7VXzJOssSZtQXSM+A7idala2PtUdkj5p+6aM8TpO1UXSY4FvMGvlxkzgNNvHZwvWUDGYFkjSllRbqz3MrJLf3rbHZIyVVGvj79w5UpB0r+3V/tXvze0kXQH8yPZfup0fCnzH9g5ZgiUi6ZvADsCXbT9Un1uZas3tSNs/y5mvaWKdaZlOBrazPdT2FsD2QBH/sCRtIuluqnIfktaV9KvMsTrtbklf7H6yvoPIpAx5Ulml+0AKYHs0sHL6OMl9AfhcayAFsP0gsBfwlp+H0DvRzVumeW3/vXVg+96COlpPofrwcBmA7fEFbFhxINV62n2Z/XrxgsDu2VJ13js1GpVQlZjX9tPdT9p+qqB/78nEYFqmcZLOBn5THw8DxmXMk5TtR7st4J/5ds9tAtuPAxtL2ppZ14tH2L4mY6wUBkr6eQ/nRbULUNO99m9+L/wbYjAt09eoZiutnW/GUt3XsgSPStoUcP3p/BDqkm/T2b5W0gxgkO1rJC0FLNJeBmyYI97heyV8eFxX0vP149anR9ePF8gTqbmiASkUpR5ATgU+RvWmchVwiO1pWYMlIOlYqvLu6rZXk/QB4GLbH80cLQlJ/W2/lDtHaKZoQCqQpJ0k3S7pn5KelzS97RNso9l+2vYw28vaXsb2XiUMpLXdgV2orxfankK1/rTR2prOJtXHJTSdzUbSZpL2qR8vJWml3JmaJgbTMp0C7A0saXuA7UVKuJcpQH2nmMXajheXdG7OTAm9Vq81be1Ru1DmPKm0ms6mQdV0BjS96exNdUXiSKo75wDMx6x+idBHYjAt06PAnYVuJ7aO7WdbB7afoVrIX4KLJJ0BLFbflutqqtuTNZ7tR7udanTTWTdFViRSiwakMn0bGCFpNPBq66Ttn+aLlEyXpMXrQRRJS1DIvwPbP6k3On8eWB04xvaozLFSKLbprPaabUsqrSKRVBFvIuEt/oPq3pYLUJV8SnIycKOki6kakPag+v/RePV1srGtAVTSgpJW7OluOg3zVaqmsw8Cj1M1nR2YNVFa3SsS+1JIRSKl6OYtkKQ7ba+VO0cukgYDW9WH19q+O2eeVCSNAza1/Vp9PB9wve0heZOFTqsrEm/eNaaQikRScc20TCMkbZc7REaTqDZAvwx4QdLymfOk0q81kALUjxtfmSi86ay9InGE7W8Bf5W0Yt5UzRODaZm+BoyU9HJpS2MkHQT8AxgFXA78uf61BE9J2qV1IGlX4C3bzTVQyU1nABcDb7Qdz6zPhT4U10wLZHuRuvFmVcrbCeUQqk0LSllb2u6rwIWSfkFV7nuUMjY8L7bprPaWikRd4g99qKQfqFCTtD/VoLIccAfwEeAGYJucuRJ5FHgud4gcbD8AfETSwvXxC5kjpVJs01ntKUm72L4MiqpIJBUNSAWSNBEYAtxkez1JawAn2f5k5mgdJ+kcqmUhf6awZUGS5gc+BaxI2wfpEm4UXWrTGYCkVYALgQ/QVpGwfX/WYA0TM9MyvWL7FUlImt/2JEmr5w6VyCP113wU0HzTzXCqWfmttH2QKMQk4Bnq9zxJy9t+JG+kNAquSCQVg2mZHqu7Gy8FRkl6BpicOVMSto/LnSGj5Wx/PHeI1Oqms2OpGs9mUs3ODKyTM1cq3SsSrdsPllCRSCnKvIWTNBRYFBjZ3qTQVJKWptoBajBtzVe2t84WKhFJZwKn2Z6YO0tKku4HNi606QxJI5lVkXhzG0XbJ2cL1UAxMy2c7dG5MyR2IfAHYCeq7ta9gaeyJkpnM+BLkh6iKvMKsO2mz9CKbTqrFVmRSC0G01CaJW2fI+mQ+oPEaEm35A6VyA65A2TyIPAXScU1ndVukLR2aRWJ1GIwDaWZUf/6hKQdgSnAEhnzJGN7MoCkZShrfXHJTWdQbkUiqbhmGooiaSdgLDAQOA0YABzXWoPXZPXuRydTLZF4ElgBuMf24KzBQkdJWqGn860PV6FvxGAaQiEkjQe2Bq62vb6krYC9bO+XOVpHldx01q57RaKUpUGpRJk3FEHSaVTLIXpk++CEcXKZYXuapC5JXbavk3RK7lAJlNx09rYVCaoPF6GPxGAaSjEud4A5wLP1wv0xVHv0Pgm8mDlTCiU3nQGcQLVl6GwVicyZGicG01AE2+fnzjAH2BV4BTgMGEa1vriEhfvFNp3VSq1IJBWDaShKff3sSOBDFHb9zHb7LLSkDxcnSloUOJxZTWeH5Y2UVKkViaSiASkURdJVVNfPvkXb9TPbR2YN1kGSptPz9eLWEokBiSOFhCQtRFWRELMqEheWuiNUp8RgGooi6VbbH5Y0obXOTtIttofkzhb6VjSdhZSizBtKU/r1s5KWSBTddBYVibRiZhqKEps2xKYNIXRCDKYhFKLwTRuKbDprV1BFIouu3AFCSEnSypL+JOlpSU9KGi5p5dy5EplRN528uUQC2DB3qAQupNqkYCXgOOBhoJh1ppJ2kXQf8BAwmur1X5E1VAPFYBpK81vgIuB9VOXOi4HfZU2UTmuJxFiqJRKnUsYSiSVtn0P1YWK07X2pZuilaG3acK/tlYBtgJvyRmqeGExDafrbvsD26/XXbyjnDiq7Ai8DhwIjgQeAnbMmSmO2pjNJ61NW01mpFYmkops3FEFS683zCknfAX5P1en4WWBEtmAJ2X5R0vuAjYB/AlcWstYwNm2YvSIRmzZ0QDQghSLU93I01bKA7my78ddNJe0PHANcS/X/YShwvO1zswYLHRWbNqQRg2kIhZD0d2DT1puopCWBG2yvnjdZZ9UNZqcCmwBvADcCh9l+MGuwhNoqEgZusT01c6TGiTJvKIKkT77T921fkipLRtOA6W3H0+tzTfdb4JfA7vXxnlRNZxtnS5RQDxWJ0yRFRaKPxcw0FEHSefXDZYBNqd5YALaimp3tlCVYApK+WT9cD1gbGE41Q9kVmGD7S5miJdG+dWTbufG2182VKaVSKxKpxcw0FMH2PvDmRvcfsv1Effx+4NcZo6WwSP3rA/VXy/AMWZKJprM3lVqRSCpmpqEoku6xvWbbcRdwV/u5ppK0iu0H3v2ZzVB601npFYnUYmYaSnONpCuZtVHDZ4GrM+ZJ6VxJy1Ht/jMWGGN7YuZMHVNvUFCyIisSucTMNBSnbkbavD4cY/v/cuZJSdJ8wBBgS+ArwMK2G7mBQTSdVUqrSOQSg2kIhZC0GdWHiM2BxYA7gLG2G7mdYslNZ+0kjQaKqUjkEoNpKELbvR3F7Pd4LObejpJeB24FfgiMsP1a5khJ1E1ne3dvOrO9fd5k6ZRUkcglrpmGIthuXT9C0nrMXuYdnydVcksBHwW2AA6W9AZwo+3v543VcQNbA2ntH8DyucKk1kNF4nKqGWroQzGYhqJIOhg4ALiEalZ6gaSzbJ+WN1nn2X5W0oNUN0Zfjqr0OW/eVEmU3HQG8BcKrEikFmXeUBRJE4BNbL9YHy9ENTtb551/59yvHkgnAX8FxgA3l/LGWnjT2WLMqkgMod5SsYCKRFIxMw2lETCz7XgmPa9DbKJBtt/IHSKHunO3iO7d7gquSCQVg2kozXnA3yS1Zia7AedkzJPSIEmnA8vaXkvSOsAutk/MHawToums0q0icTqwTykViZSizBuKI2kDYLP6cKzt23PmSaVeInEEcIbt9etzd9peK2+yziu46Yz6huBFViRS6sodIITUbN9m++f1VxEDaa2/7Zu7nXs9S5KE6qazC6i6mZemajo7KG+qpAZJukbSnQCS1pH0vdyhmiYG0xDK8bSkVahLnpL2AJ5459/SCPsDH7F9rO1jqO5rekDmTCmdBRwFzACwPYHqNnShD8U10xDKcSBwJrCGpMeBh4BheSMlUXLTGdQVCWm2l9z4ikRqMZiGUI7HqRqwrgOWAJ4H9gaOzxkqgZKbzqDcikRS0YAUQiEkjQSeBW6jbaZm++RsoRIptekMQNLKVBWJTYFnqCsStidnDdYwMZiGUIhSOnfD7CTND+wBrMisioRtN70ikVQ0IIVQjhskrZ07REhuOLAzVQPSFOAF4MWsiRooZqYhFELS3cAgqjLfq8zavKDxWymWLCoSaUQDUgjl2CF3gJDFDZLWjnuYdlbMTEMIocGiIpFGDKYhhNBgklbo6Xx08/atGExDCCGEXopu3hBCCKGXYjANIYQQeikG0xBCCKGXYjANIYQQeun/AeiI4/uRvc/mAAAAAElFTkSuQmCC\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["The above correlation shows that there is a positive relationship between the oldbalance of the customer before transaction and after transaction as expected and this is also the same for recipient.\n","The amonunt of owned by the customer in local currency is also positively related to the old balance of the recipeint before transaction and the new balance after transaction.\n","\n","\n"],"metadata":{"id":"BevNUOSFNC3o"}},{"cell_type":"code","source":["contFeat"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"L8jYJNzrRMMI","executionInfo":{"status":"ok","timestamp":1671904421696,"user_tz":-60,"elapsed":10,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"9fe1b2b5-b7f8-4fce-fdec-b0486fc3e7bd"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['amount',\n"," 'oldbalanceOrg',\n"," 'newbalanceOrig',\n"," 'oldbalanceDest',\n"," 'newbalanceDest']"]},"metadata":{},"execution_count":39}]},{"cell_type":"markdown","source":["### Predicting Feature isFraud(This is the transactions made by the fraudulent agents inside the simulation. )"],"metadata":{"id":"vEEQsaUySnbN"}},{"cell_type":"markdown","source":["### Feature Engineering"],"metadata":{"id":"baRL7b9CWpqJ"}},{"cell_type":"markdown","source":["**Feature Creation**"],"metadata":{"id":"Q_FbtY_BU6RV"}},{"cell_type":"code","source":["#Estimating amount transfered by customer\n","df[\"transferAmt\"] = df[\"oldbalanceOrg\"] - df[\"newbalanceOrig\"]\n","df.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"67MwiA7wNyeW","executionInfo":{"status":"ok","timestamp":1684412730947,"user_tz":-60,"elapsed":575,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"12ad95f9-5ccc-4e1a-d3ca-adf2ca91c8f1"},"execution_count":9,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount nameOrig oldbalanceOrg newbalanceOrig \\\n","0 1 PAYMENT 9839.64 C1231006815 170136.0 160296.36 \n","1 1 PAYMENT 1864.28 C1666544295 21249.0 19384.72 \n","2 1 TRANSFER 181.00 C1305486145 181.0 0.00 \n","3 1 CASH_OUT 181.00 C840083671 181.0 0.00 \n","4 1 PAYMENT 11668.14 C2048537720 41554.0 29885.86 \n","\n"," nameDest oldbalanceDest newbalanceDest isFraud isFlaggedFraud \\\n","0 M1979787155 0.0 0.0 0 0 \n","1 M2044282225 0.0 0.0 0 0 \n","2 C553264065 0.0 0.0 1 0 \n","3 C38997010 21182.0 0.0 1 0 \n","4 M1230701703 0.0 0.0 0 0 \n","\n"," transferAmt \n","0 9839.64 \n","1 1864.28 \n","2 181.00 \n","3 181.00 \n","4 11668.14 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountnameOrigoldbalanceOrgnewbalanceOrignameDestoldbalanceDestnewbalanceDestisFraudisFlaggedFraudtransferAmt
01PAYMENT9839.64C1231006815170136.0160296.36M19797871550.00.0009839.64
11PAYMENT1864.28C166654429521249.019384.72M20442822250.00.0001864.28
21TRANSFER181.00C1305486145181.00.00C5532640650.00.010181.00
31CASH_OUT181.00C840083671181.00.00C3899701021182.00.010181.00
41PAYMENT11668.14C204853772041554.029885.86M12307017030.00.00011668.14
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":9}]},{"cell_type":"code","source":["#Dropping unnecessary columns like the name of the customer and recipeint\n","df = df.drop([\"nameOrig\",\"nameDest\"],axis=1)\n","df.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"BoudYxzLVyTa","executionInfo":{"status":"ok","timestamp":1684412731486,"user_tz":-60,"elapsed":545,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"abf7d924-3123-4d87-de93-07224e2eb14b"},"execution_count":10,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount oldbalanceOrg newbalanceOrig oldbalanceDest \\\n","0 1 PAYMENT 9839.64 170136.0 160296.36 0.0 \n","1 1 PAYMENT 1864.28 21249.0 19384.72 0.0 \n","2 1 TRANSFER 181.00 181.0 0.00 0.0 \n","3 1 CASH_OUT 181.00 181.0 0.00 21182.0 \n","4 1 PAYMENT 11668.14 41554.0 29885.86 0.0 \n","\n"," newbalanceDest isFraud isFlaggedFraud transferAmt \n","0 0.0 0 0 9839.64 \n","1 0.0 0 0 1864.28 \n","2 0.0 1 0 181.00 \n","3 0.0 1 0 181.00 \n","4 0.0 0 0 11668.14 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudisFlaggedFraudtransferAmt
01PAYMENT9839.64170136.0160296.360.00.0009839.64
11PAYMENT1864.2821249.019384.720.00.0001864.28
21TRANSFER181.00181.00.000.00.010181.00
31CASH_OUT181.00181.00.0021182.00.010181.00
41PAYMENT11668.1441554.029885.860.00.00011668.14
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":10}]},{"cell_type":"markdown","source":["\n","\n","**Handling class imbalance**"],"metadata":{"id":"ORDU3J_ZStwu"}},{"cell_type":"code","source":["data = df.copy()"],"metadata":{"id":"A8CLR6jpOONp","executionInfo":{"status":"ok","timestamp":1684412735742,"user_tz":-60,"elapsed":1166,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":11,"outputs":[]},{"cell_type":"code","source":["data.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"uxT0gjwZVWdt","executionInfo":{"status":"ok","timestamp":1684412735744,"user_tz":-60,"elapsed":17,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"7ceb657f-763b-4dad-cbdf-2b78b61a1eca"},"execution_count":12,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount oldbalanceOrg newbalanceOrig oldbalanceDest \\\n","0 1 PAYMENT 9839.64 170136.0 160296.36 0.0 \n","1 1 PAYMENT 1864.28 21249.0 19384.72 0.0 \n","2 1 TRANSFER 181.00 181.0 0.00 0.0 \n","3 1 CASH_OUT 181.00 181.0 0.00 21182.0 \n","4 1 PAYMENT 11668.14 41554.0 29885.86 0.0 \n","\n"," newbalanceDest isFraud isFlaggedFraud transferAmt \n","0 0.0 0 0 9839.64 \n","1 0.0 0 0 1864.28 \n","2 0.0 1 0 181.00 \n","3 0.0 1 0 181.00 \n","4 0.0 0 0 11668.14 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudisFlaggedFraudtransferAmt
01PAYMENT9839.64170136.0160296.360.00.0009839.64
11PAYMENT1864.2821249.019384.720.00.0001864.28
21TRANSFER181.00181.00.000.00.010181.00
31CASH_OUT181.00181.00.0021182.00.010181.00
41PAYMENT11668.1441554.029885.860.00.00011668.14
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":12}]},{"cell_type":"code","source":["data.columns"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"50QCGlfTVnDu","executionInfo":{"status":"ok","timestamp":1684412736442,"user_tz":-60,"elapsed":713,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"721ca9ae-239d-4774-d47c-ff155b918e9c"},"execution_count":13,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Index(['step', 'type', 'amount', 'oldbalanceOrg', 'newbalanceOrig',\n"," 'oldbalanceDest', 'newbalanceDest', 'isFraud', 'isFlaggedFraud',\n"," 'transferAmt'],\n"," dtype='object')"]},"metadata":{},"execution_count":13}]},{"cell_type":"code","source":["#Selecting nearly equal samples of categories to handle class imbalance\n","dfIsFraud = data.loc[df[\"isFraud\"] == 1]\n","dfnotfraud = data.loc[df[\"isFraud\"] == 0]\n","dfnew = dfnotfraud.sample(8500)\n","newdata = pd.concat([dfIsFraud,dfnew],axis=0)\n","newdata = newdata.reset_index(drop=True)\n","newdata.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"Fbxzcbv2aNmB","executionInfo":{"status":"ok","timestamp":1684412737145,"user_tz":-60,"elapsed":714,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"d486a35c-c4bd-4530-9b62-aa73b70785dd"},"execution_count":14,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount oldbalanceOrg newbalanceOrig oldbalanceDest \\\n","0 1 TRANSFER 181.0 181.0 0.0 0.0 \n","1 1 CASH_OUT 181.0 181.0 0.0 21182.0 \n","2 1 TRANSFER 2806.0 2806.0 0.0 0.0 \n","3 1 CASH_OUT 2806.0 2806.0 0.0 26202.0 \n","4 1 TRANSFER 20128.0 20128.0 0.0 0.0 \n","\n"," newbalanceDest isFraud isFlaggedFraud transferAmt \n","0 0.0 1 0 181.0 \n","1 0.0 1 0 181.0 \n","2 0.0 1 0 2806.0 \n","3 0.0 1 0 2806.0 \n","4 0.0 1 0 20128.0 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudisFlaggedFraudtransferAmt
01TRANSFER181.0181.00.00.00.010181.0
11CASH_OUT181.0181.00.021182.00.010181.0
21TRANSFER2806.02806.00.00.00.0102806.0
31CASH_OUT2806.02806.00.026202.00.0102806.0
41TRANSFER20128.020128.00.00.00.01020128.0
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":14}]},{"cell_type":"markdown","source":["Data preprocessing\n","- Splitting data into train and test\n","- Performing log transformation on numerical features\n","- Standard scaling of features\n","- One hot encoding on categorical features"],"metadata":{"id":"FhO_dFDyZxzi"}},{"cell_type":"code","source":["import os\n","def transform_features(df):\n","\n"," numerical_features = ['amount', 'oldbalanceOrg', 'newbalanceOrig','oldbalanceDest', 'newbalanceDest','transferAmt']\n"," categorical_features = [\"type\"]\n","\n"," X = df[['amount', 'oldbalanceOrg', 'newbalanceOrig','oldbalanceDest', 'newbalanceDest','transferAmt','type']]\n"," Y = df[['isFraud']]\n"," for feature in numerical_features:\n"," df[feature] = np.log(df[feature] + 1)\n"," splitter = StratifiedShuffleSplit(n_splits=5,random_state = 42,test_size=0.2)\n"," for train,test in splitter.split(X,Y):\n"," x_train = X.loc[train]\n"," y_train = np.ravel(Y.loc[train].values)\n"," x_test = X.loc[test]\n"," y_test = np.ravel(Y.loc[test].values)\n","\n"," \n","\n"," numerical_pipeline = Pipeline(\n"," steps=[(\"imputer\",SimpleImputer(strategy='median')),\n"," (\"scaler\",MinMaxScaler())]\n"," )\n"," cat_pipeline = Pipeline(steps=[(\"imputer\",SimpleImputer(strategy=\"most_frequent\")),\n"," (\"one_hot_encoder\",OneHotEncoder())])\n"," \n"," preprocessing_obj = ColumnTransformer([(\"num_pippeline\",numerical_pipeline,numerical_features),\n"," (\"cat_pipeline\",cat_pipeline,categorical_features)])\n"," \n"," input_features_train_arr = preprocessing_obj.fit_transform(x_train)\n"," input_features_test_arr = preprocessing_obj.fit_transform(x_test)\n","\n"," #save preprocessing object\n","\n"," path = \"/content/drive/MyDrive/Predictiong_cvss_client/Fraud_Credit_Card/Artifacts\"\n"," os.chdir(path)\n"," \n"," joblib.dump(preprocessing_obj, 'preprocessor.pkl')\n","\n"," return (input_features_train_arr,\n"," y_train,\n"," input_features_test_arr,\n"," y_test,\n"," preprocessing_obj)"],"metadata":{"id":"rmxJMwGuEaKx","executionInfo":{"status":"ok","timestamp":1684414456207,"user_tz":-60,"elapsed":11,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":52,"outputs":[]},{"cell_type":"code","source":[" x_train, y_train, x_test,y_test, preprocessing_obj = transform_features(df=newdata)"],"metadata":{"id":"3hMT0GhOhmbC","executionInfo":{"status":"ok","timestamp":1684414457505,"user_tz":-60,"elapsed":1306,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":53,"outputs":[]},{"cell_type":"code","source":["y_train"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"_7RgHb1u61PD","executionInfo":{"status":"ok","timestamp":1684414405403,"user_tz":-60,"elapsed":3,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"0073a3f7-8971-404b-d56e-dac03fbdb6cd"},"execution_count":48,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([1, 0, 1, ..., 1, 1, 0])"]},"metadata":{},"execution_count":48}]},{"cell_type":"code","source":["#Create function to transform single data point\n","\n","def transform_single_data_point(single_data,preprocessing_obj):\n"," # Create a DataFrame from the single data point\n"," single_data_df = pd.DataFrame([single_data_point], index=[0])\n","\n"," # Transform the single data point using the ColumnTransformer\n"," transformed_data_point_arr = preprocessing_obj.transform(single_data_df)\n","\n"," return transformed_data_point_arr\n"],"metadata":{"id":"664QCn3aUKzf","executionInfo":{"status":"ok","timestamp":1684412746770,"user_tz":-60,"elapsed":15,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":18,"outputs":[]},{"cell_type":"code","source":["single_data_point = {\"amount\":3425,\"oldbalanceOrg\":2637,\"newbalanceOrig\":13242,\"oldbalanceDest\":12132,\"newbalanceDest\":1245,\"transferAmt\":123,\"type\":\"PAYMENT\"}\n","tranformed_single_data = transform_single_data_point(single_data = single_data_point,preprocessing_obj = preprocessing_obj)\n","print(tranformed_single_data)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8TPYovt7UKYb","executionInfo":{"status":"ok","timestamp":1684412746771,"user_tz":-60,"elapsed":15,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"28745e44-addf-49ff-b657-ff2e5a2442fe"},"execution_count":19,"outputs":[{"output_type":"stream","name":"stdout","text":["[[-0.44271576 -0.37096874 -0.1959874 -0.34073718 -0.44051823 -0.38377669\n"," 0. 0. 0. 1. 0. ]]\n"]}]},{"cell_type":"markdown","source":[],"metadata":{"id":"5-aONn7402VX"}},{"cell_type":"markdown","source":["### Model building\n","- Cross validation\n","- Training and evalutaion with best model\n","- Saving trained models"],"metadata":{"id":"lUlhuai0fdbK"}},{"cell_type":"markdown","source":["Cross validation"],"metadata":{"id":"sMz61DpN1WQb"}},{"cell_type":"code","source":["#Define function to cross validate single model\n","def run_cross_validation(model, X, y, cv=5):\n"," accuracy = cross_val_score(model, X, y, scoring='accuracy', cv=cv)\n"," f1 = cross_val_score(model, X, y, scoring='f1_macro', cv=cv)\n"," auc_roc = cross_val_score(model, X, y, scoring='roc_auc', cv=cv)\n"," \n"," return accuracy, f1, auc_roc"],"metadata":{"id":"8Pw1T4G01nxy","executionInfo":{"status":"ok","timestamp":1684414462342,"user_tz":-60,"elapsed":3,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":54,"outputs":[]},{"cell_type":"code","source":["def run_cross_validation_models(model_list, X, y, cv=5):\n"," results = []\n"," \n"," for model_name,model in model_list.items():\n"," accuracy, f1, auc_roc = run_cross_validation(model, X, y, cv)\n"," \n"," result = {\n"," 'Model': model_name,\n"," 'Accuracy': accuracy.mean(),\n"," 'F1 Score': f1.mean(),\n"," 'AUC-ROC Score': auc_roc.mean()\n"," }\n"," \n"," results.append(result)\n"," \n"," return pd.DataFrame(results)"],"metadata":{"id":"_auttbR92DRD","executionInfo":{"status":"ok","timestamp":1684414463501,"user_tz":-60,"elapsed":11,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":55,"outputs":[]},{"cell_type":"code","source":["models = {\"logistic_reg\":LogisticRegression(random_state=42),\"Naive_Bayes\":MultinomialNB(),\n"," \"K-Nearest-Neighbour\":KNeighborsClassifier(),\"xgboost\":xgb.XGBClassifier(random_state=42),\n"," \"Random_Forest\":RandomForestClassifier(\n"," random_state=42\n"," )\n","\n"," }\n","results_df = run_cross_validation_models(model_list=models,\n"," X=x_train,\n"," y=y_train,\n"," cv=5)\n"],"metadata":{"id":"pUwQoNcY4UOT","executionInfo":{"status":"ok","timestamp":1684414512794,"user_tz":-60,"elapsed":49303,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":56,"outputs":[]},{"cell_type":"code","source":["path = \"/content/drive/MyDrive/Predictiong_cvss_client/Fraud_Credit_Card/Results\"\n","os.chdir(path)\n","results_df.to_csv(\"Crossval_score_df.csv\",index=False)"],"metadata":{"id":"P_gfqM0FCKxp","executionInfo":{"status":"ok","timestamp":1684416359213,"user_tz":-60,"elapsed":1009,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":71,"outputs":[]},{"cell_type":"code","source":["results_df"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"-TKGgTq_7kM0","executionInfo":{"status":"ok","timestamp":1684414554619,"user_tz":-60,"elapsed":26,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"bee0b31e-d1e0-4cec-cfe1-31324ac5dbfe"},"execution_count":57,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" Model Accuracy F1 Score AUC-ROC Score\n","0 logistic_reg 0.920793 0.920559 0.982434\n","1 Naive_Bayes 0.867464 0.866323 0.934478\n","2 K-Nearest-Neighbour 0.986836 0.986835 0.994222\n","3 xgboost 0.993269 0.993267 0.998833\n","4 Random_Forest 0.992595 0.992594 0.998731"],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
ModelAccuracyF1 ScoreAUC-ROC Score
0logistic_reg0.9207930.9205590.982434
1Naive_Bayes0.8674640.8663230.934478
2K-Nearest-Neighbour0.9868360.9868350.994222
3xgboost0.9932690.9932670.998833
4Random_Forest0.9925950.9925940.998731
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":57}]},{"cell_type":"markdown","source":["Comment:The best model so far is xgboost"],"metadata":{"id":"j2CXJ9ymAHKm"}},{"cell_type":"markdown","source":["### Training and evaluation with different models"],"metadata":{"id":"wL6rNAKKAbWR"}},{"cell_type":"markdown","source":["Training and evaluation with Logistic Regression"],"metadata":{"id":"iUOgW1T6vxLn"}},{"cell_type":"code","source":["def train_model(model):\n"," model.fit(x_train,y_train)\n"," preds = model.predict(x_test)\n"," return model,preds,y_test"],"metadata":{"id":"nHiHrgW0o_Ef","executionInfo":{"status":"ok","timestamp":1684416017770,"user_tz":-60,"elapsed":340,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":58,"outputs":[]},{"cell_type":"code","source":["lreg = LogisticRegression()\n","model,predictions,true_values = train_model(model=lreg)"],"metadata":{"id":"RrojUJEVo--z","executionInfo":{"status":"ok","timestamp":1684416556731,"user_tz":-60,"elapsed":777,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}}},"execution_count":78,"outputs":[]},{"cell_type":"code","source":["from sklearn import metrics\n","from sklearn.metrics import classification_report\n","cp = classification_report(true_values,predictions)\n","print(\"Classification report of logreg\")\n","print(cp) "],"metadata":{"id":"H-WTjs9IyvHX","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1684416557266,"user_tz":-60,"elapsed":10,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"ed311e96-f5b0-418e-a8ff-5a1c1314b711"},"execution_count":79,"outputs":[{"output_type":"stream","name":"stdout","text":["Classification report of logreg\n"," precision recall f1-score support\n","\n"," 0 0.99 0.85 0.91 1700\n"," 1 0.86 0.99 0.92 1643\n","\n"," accuracy 0.92 3343\n"," macro avg 0.93 0.92 0.92 3343\n","weighted avg 0.93 0.92 0.92 3343\n","\n"]}]},{"cell_type":"code","source":["confusion_matrix = metrics.confusion_matrix(true_values, predictions)\n","cm_display = metrics.ConfusionMatrixDisplay(confusion_matrix = confusion_matrix, display_labels = [False, True])\n","print(\"Confusion matrix of logisttic regression\")\n","cm_display.plot()\n","plt.savefig(\"confusion_matrix_log_reg.png\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":468},"id":"wnswYJl5dt_w","executionInfo":{"status":"ok","timestamp":1684416561045,"user_tz":-60,"elapsed":1338,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"1027b6cc-1bcc-4949-f176-ce9c2b5e3c76"},"execution_count":80,"outputs":[{"output_type":"stream","name":"stdout","text":["Confusion matrix of logisttic regression\n"]},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["**Training and evaluation with XGBOOST**"],"metadata":{"id":"OF2wCSCizSNb"}},{"cell_type":"code","source":["from xgboost import XGBClassifier\n","xgb = XGBClassifier()\n","model,predictions,true_values = train_model(model=xgb)\n","from sklearn.metrics import classification_report\n","cp = classification_report(true_values,predictions)\n","print(\"Classification report of XGBoost\")\n","print(cp) "],"metadata":{"id":"-v5nmKe6yu46","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1684416585252,"user_tz":-60,"elapsed":3229,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"4e64a286-1c93-4217-f56b-7fda02225b77"},"execution_count":83,"outputs":[{"output_type":"stream","name":"stdout","text":["Classification report of XGBoost\n"," precision recall f1-score support\n","\n"," 0 0.81 1.00 0.90 1700\n"," 1 1.00 0.76 0.86 1643\n","\n"," accuracy 0.88 3343\n"," macro avg 0.90 0.88 0.88 3343\n","weighted avg 0.90 0.88 0.88 3343\n","\n"]}]},{"cell_type":"code","source":["confusion_matrix = metrics.confusion_matrix(true_values, predictions)\n","cm_display = metrics.ConfusionMatrixDisplay(confusion_matrix = confusion_matrix, display_labels = [False, True])\n","print(\"Confusion matrix of XGBOOST\")\n","cm_display.plot()\n","plt.savefig(\"confusion_matrix_xgb.png\")\n","plt.show()"],"metadata":{"id":"PWyeFqiLehaL","colab":{"base_uri":"https://localhost:8080/","height":468},"executionInfo":{"status":"ok","timestamp":1684416593425,"user_tz":-60,"elapsed":1542,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"65e5c4f7-90dc-4971-b749-96d1568d8bb9"},"execution_count":84,"outputs":[{"output_type":"stream","name":"stdout","text":["Confusion matrix of XGBOOST\n"]},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["**Training and evaluation with Guassian NaiveBayes**"],"metadata":{"id":"LVn5QA0F0NZA"}},{"cell_type":"code","source":["from sklearn.naive_bayes import GaussianNB\n","model = GaussianNB()\n","model,predictions,true_values = train_model(model)\n","from sklearn.metrics import classification_report\n","cp = classification_report(true_values,predictions)\n","print(\"Classification report of Guassain NaiveBayes\")\n","print(cp) "],"metadata":{"id":"z6UCwkKnyuzM","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1684416637733,"user_tz":-60,"elapsed":653,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"92a33873-4326-427f-ecc9-07eb1fad40c4"},"execution_count":85,"outputs":[{"output_type":"stream","name":"stdout","text":["Classification report of Guassain NaiveBayes\n"," precision recall f1-score support\n","\n"," 0 1.00 0.76 0.86 1700\n"," 1 0.80 1.00 0.89 1643\n","\n"," accuracy 0.87 3343\n"," macro avg 0.90 0.88 0.87 3343\n","weighted avg 0.90 0.87 0.87 3343\n","\n"]}]},{"cell_type":"code","source":["confusion_matrix = metrics.confusion_matrix(true_values, predictions)\n","\n","cm_display = metrics.ConfusionMatrixDisplay(confusion_matrix = confusion_matrix, display_labels = [False, True])\n","print(\"Confusion matrix of Guassain NaiveBayes\")\n","cm_display.plot()\n","plt.savefig(\"confusion_matrix_naive_bayes.png\")\n","plt.show()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":468},"id":"MbURB8xpeqhU","executionInfo":{"status":"ok","timestamp":1684416639139,"user_tz":-60,"elapsed":1415,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"79827463-be15-41a4-97e7-7124a08511bd"},"execution_count":86,"outputs":[{"output_type":"stream","name":"stdout","text":["Confusion matrix of Guassain NaiveBayes\n"]},{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAAh4AAAGwCAYAAADmPWxJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABItUlEQVR4nO3deVxWZf7/8dcNyCJyg6hAJCpmKpZaaaNUbkWiWbk1jUUuDelkkomTS79ytyjbXKak1aUvljVNlpom5eSSZEpDpiKpaVgKNqEiGut9fn8w3HWn3oH34Wbp/exxHg/OOdd17s/hEfLh+lzXORbDMAxERERE3MCjpgMQERGRPw4lHiIiIuI2SjxERETEbZR4iIiIiNso8RARERG3UeIhIiIibqPEQ0RERNzGq6YDqAtsNhtHjx4lICAAi8VS0+GIiEgVGYbB6dOnCQ8Px8Oj+v7mLiwspLi42OXreHt74+vra0JEtY8Sj0o4evQoERERNR2GiIi46MiRIzRv3rxarl1YWEhky0bkHC9z+VphYWEcOnSoXiYfSjwqISAgAID4dbfi7d+ghqMRqR7pT19T0yGIVJuykkLS1z9u//e8OhQXF5NzvIzv0lthDbj4UZX80zZadjlMcXGxEo8/qoryird/A3waKfGQ+smrQf37B07kt9xRLm8UYKFRwMV/jo36XdJX4iEiImKiMsNGmQtvQSszbOYFUwsp8RARETGRDQMbF595uNK3LtByWhERkTps8+bN3HbbbYSHh2OxWFi1atU5bTIzM7n99tsJDAzE39+fa6+9luzsbPv5wsJCxo0bR5MmTWjUqBFDhw4lNzfX4RrZ2dkMGDCAhg0bEhISwqRJkygtLa1yvEo8RERETGQz4b+qOHPmDJ07d+aFF1447/mDBw9yww030L59ez799FN27drFtGnTHCauJiYmsnr1at555x02bdrE0aNHGTJkiP18WVkZAwYMoLi4mG3btrFs2TKWLl3K9OnTq/z9UalFRETERGWGQZlx8eWSqvbt378//fv3v+D5Rx99lFtuuYV58+bZj1122WX2r0+dOsVrr73GihUruPHGGwFYsmQJUVFRfP7553Tv3p0NGzawd+9ePv74Y0JDQ7nqqquYM2cOU6ZMYebMmXh7e1c6Xo14iIiI1EL5+fkOW1FRUZWvYbPZWLt2LW3btiU2NpaQkBC6devmUI5JT0+npKSEmJgY+7H27dvTokUL0tLSAEhLS6Njx46Ehoba28TGxpKfn8+ePXuqFJMSDxERERNVTC51ZQOIiIggMDDQviUlJVU5luPHj1NQUMCTTz5Jv3792LBhA4MHD2bIkCFs2rQJgJycHLy9vQkKCnLoGxoaSk5Ojr3Nr5OOivMV56pCpRYRERET2TAoM2FVy5EjR7BarfbjPj4+Vb+WrXy+yMCBA0lMTATgqquuYtu2bSQnJ9OrV6+LjvNiacRDRESkFrJarQ7bxSQeTZs2xcvLiw4dOjgcj4qKsq9qCQsLo7i4mJMnTzq0yc3NJSwszN7mt6tcKvYr2lSWEg8RERETmVVqMYO3tzfXXnstWVlZDse/+eYbWrZsCUCXLl1o0KABn3zyif18VlYW2dnZREdHAxAdHc3XX3/N8ePH7W1SU1OxWq3nJDW/R6UWERERE7l7VUtBQQEHDhyw7x86dIiMjAyCg4Np0aIFkyZN4i9/+Qs9e/akT58+rF+/ntWrV/Ppp58CEBgYSHx8PBMnTiQ4OBir1cqDDz5IdHQ03bt3B6Bv37506NCB4cOHM2/ePHJycnjssccYN25clUdilHiIiIjUYTt37qRPnz72/YkTJwIwcuRIli5dyuDBg0lOTiYpKYnx48fTrl073n33XW644QZ7n+effx4PDw+GDh1KUVERsbGxvPjii/bznp6erFmzhrFjxxIdHY2/vz8jR45k9uzZVY7XYhgupGV/EPn5+QQGBjJ282C9JE7qrS/mXlvTIYhUm9KSQr5YPY1Tp045TNg0U8Xvin2ZoQS48Hba06dttI/KrdZYa5JGPERERExU5uKqFlf61gVKPERERExUZuDi22nNi6U20qoWERERcRuNeIiIiJjI9r/Nlf71mRIPERERE9mwUIbFpf71mUotIiIi4jYa8RARETGRzSjfXOlfnynxEBERMVGZi6UWV/rWBSq1iIiIiNtoxENERMREGvFwTomHiIiIiWyGBZvhwqoWF/rWBSq1iIiIiNtoxENERMREKrU4p8RDRETERGV4UOZCQaHMxFhqIyUeIiIiJjJcnONhaI6HiIiIiDk04iEiImIizfFwTomHiIiIicoMD8oMF+Z41PNHpqvUIiIiIm6jEQ8RERET2bBgc+Hvehv1e8hDiYeIiIiJNMfDOZVaRERExG004iEiImIi1yeXqtQiIiIilVQ+x8OFl8Sp1CIiIiJiDo14iIiImMjm4rtatKpFREREKk1zPJxT4iEiImIiGx56jocTmuMhIiIibqMRDxEREROVGRbKXHi1vSt96wIlHiIiIiYqc3FyaZlKLSIiIiLm0IiHiIiIiWyGBzYXVrXYtKpFREREKkulFudUahEREanDNm/ezG233UZ4eDgWi4VVq1ZdsO3999+PxWJh/vz5Dsfz8vKIi4vDarUSFBREfHw8BQUFDm127dpFjx498PX1JSIignnz5l1UvEo8RERETGTjl5UtF7PZqvh5Z86coXPnzrzwwgtO27333nt8/vnnhIeHn3MuLi6OPXv2kJqaypo1a9i8eTNjxoyxn8/Pz6dv3760bNmS9PR0nn76aWbOnMnLL79cxWhVahERETGV6w8Qq1rf/v37079/f6dtfvjhBx588EE++ugjBgwY4HAuMzOT9evXs2PHDrp27QrAokWLuOWWW3jmmWcIDw8nJSWF4uJiXn/9dby9vbniiivIyMjgueeec0hQKkMjHiIiIrVQfn6+w1ZUVHRR17HZbAwfPpxJkyZxxRVXnHM+LS2NoKAge9IBEBMTg4eHB9u3b7e36dmzJ97e3vY2sbGxZGVlceLEiSrFo8RDRETERBXvanFlA4iIiCAwMNC+JSUlXVQ8Tz31FF5eXowfP/6853NycggJCXE45uXlRXBwMDk5OfY2oaGhDm0q9ivaVJZKLSIiIiayYcHGxT99tKLvkSNHsFqt9uM+Pj5VvlZ6ejoLFizgyy+/xGKpHU9E1YiHiIiIicwa8bBarQ7bxSQeW7Zs4fjx47Ro0QIvLy+8vLz47rvv+Pvf/06rVq0ACAsL4/jx4w79SktLycvLIywszN4mNzfXoU3FfkWbylLiISIiUk8NHz6cXbt2kZGRYd/Cw8OZNGkSH330EQDR0dGcPHmS9PR0e7+NGzdis9no1q2bvc3mzZspKSmxt0lNTaVdu3Y0bty4SjGp1CIiImIi1x8gVrW+BQUFHDhwwL5/6NAhMjIyCA4OpkWLFjRp0sShfYMGDQgLC6Ndu3YAREVF0a9fP0aPHk1ycjIlJSUkJCQwbNgw+9Lbu+++m1mzZhEfH8+UKVPYvXs3CxYs4Pnnn6/y/SnxEBERMZHNsGBz4Q2zVe27c+dO+vTpY9+fOHEiACNHjmTp0qWVukZKSgoJCQncdNNNeHh4MHToUBYuXGg/HxgYyIYNGxg3bhxdunShadOmTJ8+vcpLaUGJh4iISJ3Wu3dvjCq83+Xw4cPnHAsODmbFihVO+3Xq1IktW7ZUNbxzKPEQERExkc3FUosrDx+rC5R4iIiImMj1t9PW78Sjft+diIiI1Coa8RARETFRGRbKXHiAmCt96wIlHiIiIiZSqcW5+n13IiIiUqtoxENERMREZbhWLikzL5RaSYmHiIiIiVRqcU6Jh4iIiIl+/aK3i+1fn9XvuxMREZFaRSMeIiIiJjKwYHNhjoeh5bQiIiJSWSq1OFe/705ERERqFY14iIiImMhmWKr8avvf9q/PlHiIiIiYqMzFt9O60rcuqN93JyIiIrWKRjxERERMpFKLc0o8RERETGTDA5sLBQVX+tYF9fvuREREpFbRiIeIiIiJygwLZS6US1zpWxco8RARETGR5ng4p8RDRETERIaLb6c19ORSEREREXNoxENERMREZVgoc+FFb670rQuUeIiIiJjIZrg2T8NmmBhMLaRSi4iIiLiNRjykWpz50uC/yw1+zoTS/0KLZyxY+5T/BWCUGOQuNji9FYp/AM9G0KgbhD5ooUGzX/5KKPrOIGeBwdkMMErBtw2EjLXQ6Npf2uzuYjvns5s/YSEotn4PVUrtF3dzBvcP+oK3N17JonevA+C26zO5uesB2kb8F3+/Evo/PJKCn30c+r09ewWXNClwOJa86k+kpF7lrtDFRTYXJ5e60rcuqJOJx9KlS5kwYQInT56s6VDkAmw/g29baHy7hexJjuOGtkL4eR+E3GfBty2UnYZjTxt8l2jQ5v9+SRi+m2DgHQGRL1mw+MBPKwy+m2DQ9n1o0PSXdpfOsNDoul+u7xlQ7bcn4lT7Fse5/YZMDnwf7HDc17uU7Xsj2L43gvsHfXHB/q+u7srqbe3t+2cLG1RbrGI+GxZsLszTcKVvXVCjadWoUaOwWCznbAcOHKjJsMQEAddbCH3AA+uN5/4AeQZYiHzRg8C+FnxaWWjY0cIlUywUZkLxsfIkpfSEQXE2NLvXgu/lFnxaWAh90IJRCEUHf3u98kSkYvPwqd8/tFK7+fmUMH3Uv5m3ogenzzqOZrzz746kpF7FnsMhTq9xtqgBefkN7VthsRIPqT9qfDynX79+HDt2zGGLjIys6bDEzWwFgOWX0QrPIPBuCSfXGNh+NjBKDU68C57B4Bfl2PfoUwaZN9o4OMLGifcNDKOez8ySWi3xzq2k7YkgPav5RV8jrm8Ga55axmtT3+WumK/w9Di3pCi1V8WTS13Z6rMaTzx8fHwICwtz2BYsWEDHjh3x9/cnIiKCBx54gIKCggte46uvvqJPnz4EBARgtVrp0qULO3futJ/funUrPXr0wM/Pj4iICMaPH8+ZM2fccXtSCbYig5yFBoGx4Nmo/AfOYrEQudjCz1mwt4fBnusM/pti0GqRBU/rLz+UIfdbiHjSQqsXLVhvtHD0SYO8t2rqTuSP7qYu5fM3Xnr/Txd9jXc/vZKZr9/EQwtu5YOtUQyP/Q9jB203MUqpbhVzPFzZ6rNaeXceHh4sXLiQPXv2sGzZMjZu3MjkyZMv2D4uLo7mzZuzY8cO0tPTmTp1Kg0alA9NHjx4kH79+jF06FB27drFypUr2bp1KwkJCRe8XlFREfn5+Q6bVA+jxODIVAMMCH/kl4TCMAyOPmXgFQyRr1q4bJkFa2/4LtGg5MdfRjRCRlvwv8qCX3sLzUZZaDoCfnxDIx7ifiFBBYy/I405S2+kuPTip8+t3NiJjP3hHDzahPe3duAf/+rO0N67aeBVZmK0IjWnxieXrlmzhkaNGtn3+/fvzzvvvGPfb9WqFXPnzuX+++/nxRdfPO81srOzmTRpEu3bl0/Guvzyy+3nkpKSiIuLY8KECfZzCxcupFevXixevBhfX99zrpeUlMSsWbPMuD1xwigxyJ5qUHIMWiVb7KMdAGd2wOktEPXvX477RVko2G7j5Bpodu/5r9nwSgs/vmpgKzbw8K7fw5VSu7Rr8V+CrT/z6tR/2Y95eRp0bnOMIb32cNND8Rf1l+zewyF4eRqEBZ/myPEgEyOW6mLDxXe11PPJpTWeePTp04fFixfb9/39/fn4449JSkpi37595OfnU1paSmFhIWfPnqVhw4bnXGPixIncd999vPHGG8TExPDnP/+Zyy67DCgvw+zatYuUlBR7e8MwsNlsHDp0iKioqHOu98gjjzBx4kT7fn5+PhEREWbe9h9eRdJRfKR81YpXkOMPmq3wf1/89t9pD3A2hePnb8DTipIOcbudWeGMmHuHw7FHhm8iOzeQlA1XXfTw+eXNf6LMZuHEaT8zwhQ3MFxc1WIo8ahe/v7+tGnTxr5/+PBhbr31VsaOHcvjjz9OcHAwW7duJT4+nuLi4vMmHjNnzuTuu+9m7dq1rFu3jhkzZvDWW28xePBgCgoK+Nvf/sb48ePP6deiRYvzxuTj44OPj895z0nllJ0tTyoqFB+Fn7MMPK3QoClkTzH4eR+0nG/BKIOS/5ZnE56B4NHAQsOO5RNNf5hh0Gw0ePjAifcMSn6AgBvKr5m/2aD0J2jYESw+UPA5/Pi6QdPhNXDD8of3c5E3h445Lp8tLPLiVIGv/Xiw9SzB1rM0b1Zevm0dnsfZogbk5jXi9FlfrojMpUOr43z5TThnCxtwZetcHhyaxoYv2pzzvA+pvfR2Wudq3RyP9PR0bDYbzz77LN27d6dt27YcPXr0d/u1bduWxMRENmzYwJAhQ1iyZAkA11xzDXv37qVNmzbnbN7e3tV9O39YP++Fg3cbHLy7PKHIea786+PJBiU/wulNUJoLB+8yyIr9ZTv7VXl/r8YWWv3Dgu0sHL7f4OBwgzMZ0OI5C35t/zcB1Qvy3jH49l6Dg3cZnPiXwSUTLYSMqd8/tFJ3DbxhL0se+RdT4jYD8MLE1Sx55F/c0Ok7AEpKPbmpy0EWTVjNG4+9w4jY//D2xo48/WbPmgxbarnNmzdz2223ER4ejsViYdWqVfZzJSUlTJkyxb5gIzw8nBEjRpzzezUvL4+4uDisVitBQUHEx8efs6hj165d9OjRA19fXyIiIpg3b95FxVvjIx6/1aZNG0pKSli0aBG33XYbn332GcnJyRds//PPPzNp0iTuuOMOIiMj+f7779mxYwdDhw4FYMqUKXTv3p2EhATuu+8+/P392bt3L6mpqfzjH/9w12394TTqauHK9AsnAM7OVfDrYKHVCxduF3CdhYDrlGRI7TV+wW0O+0s+7MqSD7tesP03R5py/zODqjkqqW7ufnLpmTNn6Ny5M3/9618ZMmSIw7mzZ8/y5ZdfMm3aNDp37syJEyd46KGHuP322x1Wf8bFxXHs2DFSU1MpKSnh3nvvZcyYMaxYsQIon3LQt29fYmJiSE5O5uuvv+avf/0rQUFBjBkzpkrx1rrEo3Pnzjz33HM89dRTPPLII/Ts2ZOkpCRGjBhx3vaenp789NNPjBgxgtzcXJo2bcqQIUPsk0M7derEpk2bePTRR+nRoweGYXDZZZfxl7/8xZ23JSIifxBmlVp+u6LyQtMA+vfvT//+/c97rcDAQFJTUx2O/eMf/+BPf/oT2dnZtGjRgszMTNavX8+OHTvo2rU8MV60aBG33HILzzzzDOHh4aSkpFBcXMzrr7+Ot7c3V1xxBRkZGTz33HN1K/FYunTpeY8nJiaSmJjocGz48F8K96NGjWLUqFEAeHt78+abbzr9nGuvvZYNGza4FKuIiIg7/XZRw4wZM5g5c6bL1z116hQWi4WgoCAA0tLSCAoKsicdADExMXh4eLB9+3YGDx5MWloaPXv2dJiiEBsby1NPPcWJEydo3LhxpT+/1o14iIiI1GVmvavlyJEjWK1W+3EzFj0UFhYyZcoU7rrrLvu1c3JyCAlxfIy/l5cXwcHB5OTk2Nv89qnioaGh9nNKPERERGqIWaUWq9XqkHi4qqSkhDvvvBPDMBweY+FuSjxERETquYqk47vvvmPjxo0OCU1YWBjHjx93aF9aWkpeXh5hYWH2Nrm5uQ5tKvYr2lRWrVtOKyIiUpdVjHi4spmpIunYv38/H3/8MU2aNHE4Hx0dzcmTJ0lPT7cf27hxIzabjW7dutnbbN68mZKSEnub1NRU2rVrV6UyCyjxEBERMZW7E4+CggIyMjLIyMgA4NChQ2RkZJCdnU1JSQl33HEHO3fuJCUlhbKyMnJycsjJyaG4uBiAqKgo+vXrx+jRo/niiy/47LPPSEhIYNiwYYSHhwNw99134+3tTXx8PHv27GHlypUsWLDA4SnflaVSi4iISB22c+dO+vTpY9+vSAZGjhzJzJkz+eCDDwC46qqrHPr9+9//pnfv3gCkpKSQkJDATTfdhIeHB0OHDmXhwoX2toGBgWzYsIFx48bRpUsXmjZtyvTp06u8lBaUeIiIiJjK3Y9M7927N4aTl1g5O1chODjY/rCwC+nUqRNbtmypUmzno8RDRETERAauvWH299OEuk2Jh4iIiIn0kjjnNLlURERE3EYjHiIiIibSiIdzSjxERERMpMTDOZVaRERExG004iEiImIijXg4p8RDRETERIZhwXAheXClb12gUouIiIi4jUY8RERETGTD4tIDxFzpWxco8RARETGR5ng4p1KLiIiIuI1GPEREREykyaXOKfEQERExkUotzinxEBERMZFGPJzTHA8RERFxG414iIiImMhwsdRS30c8lHiIiIiYyAAMw7X+9ZlKLSIiIuI2GvEQERExkQ0LFj259IKUeIiIiJhIq1qcU6lFRERE3EYjHiIiIiayGRYseoDYBSnxEBERMZFhuLiqpZ4va1GpRURERNxGIx4iIiIm0uRS55R4iIiImEiJh3NKPEREREykyaXOaY6HiIiIuI1GPEREREykVS3OKfEQERExUXni4cocDxODqYVUahERERG30YiHiIiIibSqxTklHiIiIiYy/re50r8+U6lFRESkDtu8eTO33XYb4eHhWCwWVq1a5XDeMAymT5/OJZdcgp+fHzExMezfv9+hTV5eHnFxcVitVoKCgoiPj6egoMChza5du+jRowe+vr5EREQwb968i4pXiYeIiIiJKkotrmxVcebMGTp37swLL7xw3vPz5s1j4cKFJCcns337dvz9/YmNjaWwsNDeJi4ujj179pCamsqaNWvYvHkzY8aMsZ/Pz8+nb9++tGzZkvT0dJ5++mlmzpzJyy+/XOXvj0otIiIiZjKp1pKfn+9w2MfHBx8fn3Oa9+/fn/79+5//UobB/Pnzeeyxxxg4cCAAy5cvJzQ0lFWrVjFs2DAyMzNZv349O3bsoGvXrgAsWrSIW265hWeeeYbw8HBSUlIoLi7m9ddfx9vbmyuuuIKMjAyee+45hwSlMjTiISIiYiZXRzv+N+IRERFBYGCgfUtKSqpyKIcOHSInJ4eYmBj7scDAQLp160ZaWhoAaWlpBAUF2ZMOgJiYGDw8PNi+fbu9Tc+ePfH29ra3iY2NJSsrixMnTlQpJo14iIiI1EJHjhzBarXa98832vF7cnJyAAgNDXU4Hhoaaj+Xk5NDSEiIw3kvLy+Cg4Md2kRGRp5zjYpzjRs3rnRMSjxERERMZNaTS61Wq0PiUV+o1CIiImIid08udSYsLAyA3Nxch+O5ubn2c2FhYRw/ftzhfGlpKXl5eQ5tzneNX39GZSnxEBERqaciIyMJCwvjk08+sR/Lz89n+/btREdHAxAdHc3JkydJT0+3t9m4cSM2m41u3brZ22zevJmSkhJ7m9TUVNq1a1elMgso8RARETFXxQRRV7YqKCgoICMjg4yMDKB8QmlGRgbZ2dlYLBYmTJjA3Llz+eCDD/j6668ZMWIE4eHhDBo0CICoqCj69evH6NGj+eKLL/jss89ISEhg2LBhhIeHA3D33Xfj7e1NfHw8e/bsYeXKlSxYsICJEydW+dujOR4iIiImcvfbaXfu3EmfPn3s+xXJwMiRI1m6dCmTJ0/mzJkzjBkzhpMnT3LDDTewfv16fH197X1SUlJISEjgpptuwsPDg6FDh7Jw4UL7+cDAQDZs2MC4cePo0qULTZs2Zfr06VVeSgtKPEREROq03r17YzjJViwWC7Nnz2b27NkXbBMcHMyKFSucfk6nTp3YsmXLRcdZQYmHiIiImfSyFqeUeIiIiJhIb6d1rlKJxwcffFDpC95+++0XHYyIiIjUb5VKPCpmvv4ei8VCWVmZK/GIiIjUffW8XOKKSiUeNputuuMQERGpF1Rqcc6l53j8+pW6IiIiwi+TS13Z6rEqJx5lZWXMmTOHSy+9lEaNGvHtt98CMG3aNF577TXTAxQREZH6o8qJx+OPP87SpUuZN2+ew+txr7zySl599VVTgxMREal7LCZs9VeVE4/ly5fz8ssvExcXh6enp/14586d2bdvn6nBiYiI1DkqtThV5cTjhx9+oE2bNucct9lsDi+PEREREfmtKiceHTp0OO8jU//5z39y9dVXmxKUiIhInaURD6eq/OTS6dOnM3LkSH744QdsNhv/+te/yMrKYvny5axZs6Y6YhQREak7LuINs+f0r8eqPOIxcOBAVq9ezccff4y/vz/Tp08nMzOT1atXc/PNN1dHjCIiIlJPXNS7Wnr06EFqaqrZsYiIiNR5hlH1V9v/tn99dtEvidu5cyeZmZlA+byPLl26mBaUiIhInaW30zpV5cTj+++/56677uKzzz4jKCgIgJMnT3Ldddfx1ltv0bx5c7NjFBERkXqiynM87rvvPkpKSsjMzCQvL4+8vDwyMzOx2Wzcd9991RGjiIhI3VExudSVrR6r8ojHpk2b2LZtG+3atbMfa9euHYsWLaJHjx6mBiciIlLXWIzyzZX+9VmVE4+IiIjzPiisrKyM8PBwU4ISERGpszTHw6kql1qefvppHnzwQXbu3Gk/tnPnTh566CGeeeYZU4MTERGR+qVSIx6NGzfGYvml5nTmzBm6deuGl1d599LSUry8vPjrX//KoEGDqiVQERGROkEPEHOqUonH/PnzqzkMERGRekKlFqcqlXiMHDmyuuMQERGRP4CLfoAYQGFhIcXFxQ7HrFarSwGJiIjUaRrxcKrKk0vPnDlDQkICISEh+Pv707hxY4dNRETkD01vp3WqyonH5MmT2bhxI4sXL8bHx4dXX32VWbNmER4ezvLly6sjRhEREaknqlxqWb16NcuXL6d3797ce++99OjRgzZt2tCyZUtSUlKIi4urjjhFRETqBq1qcarKIx55eXm0bt0aKJ/PkZeXB8ANN9zA5s2bzY1ORESkjql4cqkrW31W5cSjdevWHDp0CID27dvz9ttvA+UjIRUvjRMRERE5nyonHvfeey9fffUVAFOnTuWFF17A19eXxMREJk2aZHqAIiIidYomlzpV5TkeiYmJ9q9jYmLYt28f6enptGnThk6dOpkanIiIiNQvLj3HA6Bly5a0bNnSjFhERETqPAsuvp3WtEhqp0olHgsXLqz0BcePH3/RwYiIiEj9VqnE4/nnn6/UxSwWS71OPDJ72vCy2Go6DJFqseXoSzUdgki1yT9to/FqN32YltM6VanJpYcOHarU9u2331Z3vCIiIrWbmyeXlpWVMW3aNCIjI/Hz8+Oyyy5jzpw5GMYvFzIMg+nTp3PJJZfg5+dHTEwM+/fvd7hOXl4ecXFxWK1WgoKCiI+Pp6Cg4GK+A05VeVWLiIiI1B5PPfUUixcv5h//+AeZmZk89dRTzJs3j0WLFtnbzJs3j4ULF5KcnMz27dvx9/cnNjaWwsJCe5u4uDj27NlDamoqa9asYfPmzYwZM8b0eF2eXCoiIiK/4uaXxG3bto2BAwcyYMAAAFq1asWbb77JF198UX45w2D+/Pk89thjDBw4EIDly5cTGhrKqlWrGDZsGJmZmaxfv54dO3bQtWtXABYtWsQtt9zCM888Q3h4uAs35EgjHiIiIiYy68ml+fn5DltRUdF5P++6667jk08+4ZtvvgHgq6++YuvWrfTv3x8ony6Rk5NDTEyMvU9gYCDdunUjLS0NgLS0NIKCguxJB5Q/MsPDw4Pt27eb+v3RiIeIiEgtFBER4bA/Y8YMZs6ceU67qVOnkp+fT/v27fH09KSsrIzHH3/c/u60nJwcAEJDQx36hYaG2s/l5OQQEhLicN7Ly4vg4GB7G7Mo8RARETGTSaWWI0eOYLVa7Yd9fHzO2/ztt98mJSWFFStWcMUVV5CRkcGECRMIDw9n5MiRLgRSPS6q1LJlyxbuueceoqOj+eGHHwB444032Lp1q6nBiYiI1DkmrWqxWq0O24USj0mTJjF16lSGDRtGx44dGT58OImJiSQlJQEQFhYGQG5urkO/3Nxc+7mwsDCOHz/ucL60tJS8vDx7G7NUOfF49913iY2Nxc/Pj//85z/2mtOpU6d44oknTA1OREREnDt79iweHo6/zj09PbHZyp87FRkZSVhYGJ988on9fH5+Ptu3byc6OhqA6OhoTp48SXp6ur3Nxo0bsdlsdOvWzdR4q5x4zJ07l+TkZF555RUaNGhgP3799dfz5ZdfmhqciIhIXWPW5NLKuu2223j88cdZu3Ythw8f5r333uO5555j8ODB5fFYLEyYMIG5c+fywQcf8PXXXzNixAjCw8MZNGgQAFFRUfTr14/Ro0fzxRdf8Nlnn5GQkMCwYcNMXdECFzHHIysri549e55zPDAwkJMnT5oRk4iISN3l5ieXLlq0iGnTpvHAAw9w/PhxwsPD+dvf/sb06dPtbSZPnsyZM2cYM2YMJ0+e5IYbbmD9+vX4+vra26SkpJCQkMBNN92Eh4cHQ4cOrdIrUyqryolHWFgYBw4coFWrVg7Ht27dSuvWrc2KS0REpG5y83M8AgICmD9/PvPnz79gG4vFwuzZs5k9e/YF2wQHB7NixYqqffhFqHKpZfTo0Tz00ENs374di8XC0aNHSUlJ4eGHH2bs2LHVEaOIiIjUE1Ue8Zg6dSo2m42bbrqJs2fP0rNnT3x8fHj44Yd58MEHqyNGERGROuNi5mn8tn99VuXEw2Kx8OijjzJp0iQOHDhAQUEBHTp0oFGjRtURn4iISN3i5lJLXXPRDxDz9vamQ4cOZsYiIiIi9VyVE48+ffpgsVx4xu3GjRtdCkhERKROc7HUohGP37jqqqsc9ktKSsjIyGD37t218tGsIiIibqVSi1NVTjyef/758x6fOXMmBQUFLgckIiIi9ddFvavlfO655x5ef/11sy4nIiJSN5n0rpb6yrS306alpTk8AU1EROSPSMtpnaty4jFkyBCHfcMwOHbsGDt37mTatGmmBSYiIiL1T5UTj8DAQId9Dw8P2rVrx+zZs+nbt69pgYmIiEj9U6XEo6ysjHvvvZeOHTvSuHHj6opJRESk7tKqFqeqNLnU09OTvn376i20IiIiF1CZ197/3lafVXlVy5VXXsm3335bHbGIiIhIPVflxGPu3Lk8/PDDrFmzhmPHjpGfn++wiYiI/OFpKe0FVXqOx+zZs/n73//OLbfcAsDtt9/u8Oh0wzCwWCyUlZWZH6WIiEhdoTkeTlU68Zg1axb3338///73v6szHhEREanHKp14GEZ5CtarV69qC0ZERKSu0wPEnKvSclpnb6UVERERVGr5HVVKPNq2bfu7yUdeXp5LAYmIiEj9VaXEY9asWec8uVRERER+oVKLc1VKPIYNG0ZISEh1xSIiIlL3qdTiVKWf46H5HSIiIuKqKq9qERERESc04uFUpRMPm81WnXGIiIjUC5rj4VyV5niIiIjI79CIh1NVfleLiIiIyMXSiIeIiIiZNOLhlBIPERERE2mOh3MqtYiIiIjbaMRDRETETCq1OKXEQ0RExEQqtTinUouIiIi4jUY8REREzKRSi1Ma8RARETGTYcJWRT/88AP33HMPTZo0wc/Pj44dO7Jz585fQjIMpk+fziWXXIKfnx8xMTHs37/f4Rp5eXnExcVhtVoJCgoiPj6egoKCqgfzO5R4iIiI1GEnTpzg+uuvp0GDBqxbt469e/fy7LPP0rhxY3ubefPmsXDhQpKTk9m+fTv+/v7ExsZSWFhobxMXF8eePXtITU1lzZo1bN68mTFjxpger0otIiIiJrL8b3Olf1U89dRTREREsGTJEvuxyMhI+9eGYTB//nwee+wxBg4cCMDy5csJDQ1l1apVDBs2jMzMTNavX8+OHTvo2rUrAIsWLeKWW27hmWeeITw83IU7cqQRDxERETOZVGrJz8932IqKis77cR988AFdu3blz3/+MyEhIVx99dW88sor9vOHDh0iJyeHmJgY+7HAwEC6detGWloaAGlpaQQFBdmTDoCYmBg8PDzYvn27Cd+UXyjxEBERMVHFclpXNoCIiAgCAwPtW1JS0nk/79tvv2Xx4sVcfvnlfPTRR4wdO5bx48ezbNkyAHJycgAIDQ116BcaGmo/l5OTQ0hIiMN5Ly8vgoOD7W3MolKLiIhILXTkyBGsVqt938fH57ztbDYbXbt25YknngDg6quvZvfu3SQnJzNy5Ei3xFoVGvEQERExk0mlFqvV6rBdKPG45JJL6NChg8OxqKgosrOzAQgLCwMgNzfXoU1ubq79XFhYGMePH3c4X1paSl5enr2NWZR4iIiImM2NS2mvv/56srKyHI598803tGzZEiifaBoWFsYnn3xiP5+fn8/27duJjo4GIDo6mpMnT5Kenm5vs3HjRmw2G926dat6UE6o1CIiIlKHJSYmct111/HEE09w55138sUXX/Dyyy/z8ssvA2CxWJgwYQJz587l8ssvJzIykmnTphEeHs6gQYOA8hGSfv36MXr0aJKTkykpKSEhIYFhw4aZuqIFlHiIiIiYyt3varn22mt57733eOSRR5g9ezaRkZHMnz+fuLg4e5vJkydz5swZxowZw8mTJ7nhhhtYv349vr6+9jYpKSkkJCRw00034eHhwdChQ1m4cOHF38gFWAzDqOcPZ3Vdfn4+gYGB9GYgXpYGNR2OSLX46GhGTYcgUm3yT9to3PZbTp065TBh09TP+N/viitHP4Gnt+/vd7iAsuJCdr/y/6o11pqkOR4iIiLiNiq1iIiImMjdpZa6RomHiIiImfR2WqdUahERERG30YiHiIiIiVRqcU6Jh4iIiJlUanFKiYeIiIiZlHg4pTkeIiIi4jYa8RARETGR5ng4p8RDRETETCq1OKVSi4iIiLiNRjxERERMZDEMLC68Bs2VvnWBEg8REREzqdTilEotIiIi4jYa8RARETGRVrU4p8RDRETETCq1OKVSi4iIiLiNRjxERERMpFKLc0o8REREzKRSi1NKPEREREykEQ/nNMdDRERE3EYjHiIiImZSqcUpJR4iIiImq+/lEleo1CIiIiJuoxEPERERMxlG+eZK/3pMiYeIiIiJtKrFOZVaRERExG004iEiImImrWpxSomHiIiIiSy28s2V/vWZSi0iIiLiNhrxkFqlSVgJ8Y8e5do+p/Hxs3H0sA/PJkawf1fDmg5NxMHXn/vzzosh7P+6IXm5DZjx2iGu63/KoU32fh9emxvOrs8bUVYKLdsWMe2VQ4Q0LwFgweTm/GdLAD/lNsCvoY2ormeIf/QoLS4vAmDDymCeTWxx3s9fuWs3QU1Lq/cm5eKo1OKUEg+pNRoFlvLc+/vZta0Rj93TmpM/eXJp62IKTnnWdGgi5yg860HrK34m9q48ZsdHnnP+6GFvJg66nH7DfmL4wzk0DCjjuyxfvH1/+a1yeaefuXHICZpdWsLpE57837Nh/L+7LmPZ9r14ekKv20/QtU++w3WfmdCCkiIPJR21mFa1OFerEg+LxeL0/IwZM5g5c6Z7ghG3u3Pccf571NvhL7zcIz41GJHIhV1742muvfH0Bc8vffIS/nRjPvdNO2Y/Ft6q2KHNLff8ZP86LAJGTjnG2Jj25B7xJrxVMT5+Bj5+vyQYJ3/y5KvPGpH47BET70RMp+d4OFWrEo9jx375AV25ciXTp08nKyvLfqxRo0b2rw3DoKysDC+vWnUL4oLuffNJ/zSAR186TKfoM/w3x4s1S5uybkWTmg5NpEpsNvjiEyt/fuA4/++u1hzY7UdYi2KGJRw/pxxTofCsBxtWBhPWoohm4SXnbfPxO8H4+Bn0GHCyGqMXqV61anJpWFiYfQsMDMRisdj39+3bR0BAAOvWraNLly74+PiwdetWRo0axaBBgxyuM2HCBHr37m3ft9lsJCUlERkZiZ+fH507d+af//znBeMoKioiPz/fYZPqd0mLYm4d8RNHD/nw/+6OZM2ypoyd8wMxf86r6dBEquTkf734+YwnK/8RQtc+p0l681uu73eK2fe1Yleav0Pb1UubMLBNRwa26cSOjVaS3jpIA+/z/8X70ZtN6DP4BD5+9fsv4rquotTiylaf1arEozKmTp3Kk08+SWZmJp06dapUn6SkJJYvX05ycjJ79uwhMTGRe+65h02bNl2wfWBgoH2LiIgw8xbkAiwecGC3H0uevISDuxuyLqUJ61Y0YcDwn36/s0gtYvxvOWR0bD5DxvzIZVf+zF8ePE63mHzWLm/q0PbGISd4cUMWz/xrP81bF/H431pRXHhu2XnvzoZk7/el3136eaj1DBO2i/Tkk09isViYMGGC/VhhYSHjxo2jSZMmNGrUiKFDh5Kbm+vQLzs7mwEDBtCwYUNCQkKYNGkSpaXVM4+oziUes2fP5uabb+ayyy4jODj4d9sXFRXxxBNP8PrrrxMbG0vr1q0ZNWoU99xzDy+99NJ5+zzyyCOcOnXKvh05onqqO+Qd9+K7b3wdjh3Z70PIpcUX6CFSO1mDy/D0MmjZttDheMTlhRz/oYHDMX+rjUtbF9Ox+xkee+UwRw748Nm6wHOuuX5FEy674iyXd/q5WmOXumvHjh289NJL5/xRnpiYyOrVq3nnnXfYtGkTR48eZciQIfbzZWVlDBgwgOLiYrZt28ayZctYunQp06dPr5Y469wEia5du1ap/YEDBzh79iw333yzw/Hi4mKuvvrq8/bx8fHBx0eTGt1t7w5/Ii4rcjh2aesijv/gXUMRiVycBt4GbTuf5fuDjv+O/PCtj30p7fkYBmBYKCl2/Jvw5zMebF4dxL2PHDt/R6lVzFrV8tsyv7PfTQUFBcTFxfHKK68wd+5c+/FTp07x2muvsWLFCm688UYAlixZQlRUFJ9//jndu3dnw4YN7N27l48//pjQ0FCuuuoq5syZw5QpU5g5cybe3ub+G1znRjz8/R3rox4eHhi/mQFcUvLLD3ZBQQEAa9euJSMjw77t3bvX6TwPcb9/vdyM9tecYdiDuYS3KqLP4BPcck8eHyxp+vudRdzs5zMeHNztx8HdfgDkHPHm4G4/jn9fPqLx5weOs+mDID5MCeaHQ968/3pTPk8N5LaR/wXg2HfevLUohP27yvvs2dGQx8e0wtvPxp9ucvyFs+n9IMrKLNw09IR7b1IuTsWqFlc2ICIiwqHsn5SUdMGPHDduHAMGDCAmJsbheHp6OiUlJQ7H27dvT4sWLUhLSwMgLS2Njh07Ehoaam8TGxtLfn4+e/bsMfM7A9TBEY/fatasGbt373Y4lpGRQYMG5T/8HTp0wMfHh+zsbHr16lUTIUolffNVQ2bHR3LvI8eIS8wl54g3ydPD+fd7jWs6NJFzfPNVQybf0ca+/9LMSwG4+c48Hp6fzfX9TzH+ye956x+hLJ7WnOatyx8edmW3MwB4+9jYvb0R773SjIJTngQ1LaVj9wKef3//Oc/oWP9mE67vf5JGgWXuu0GpcUeOHMFqtdr3LzTa8dZbb/Hll1+yY8eOc87l5OTg7e1NUFCQw/HQ0FBycnLsbX6ddFScrzhntjqfeNx44408/fTTLF++nOjoaP7v//6P3bt328soAQEBPPzwwyQmJmKz2bjhhhs4deoUn332GVarlZEjR9bwHcivbf/YyvaPrb/fUKSGdb6ugI+OZjhtE3tXHrF3nX9VVpOwUub+37eV+qz5q/dXNTypQWaVWqxWq0PicT5HjhzhoYceIjU1FV9fX6dta4s6V2r5rdjYWKZNm8bkyZO59tprOX36NCNGjHBoM2fOHKZNm0ZSUhJRUVH069ePtWvXEhl57tMGRUREXOLGVS3p6ekcP36ca665Bi8vL7y8vNi0aRMLFy7Ey8uL0NBQiouLOXnypEO/3NxcwsLCgPJHWfx2lUvFfkUbM1mM306QkHPk5+cTGBhIbwbiZWnw+x1E6qDf++tdpC7LP22jcdtvOXXq1O+OIlz0Z/zvd0V0v9l4Nbj40YfSkkLS1k+vVKynT5/mu+++czh277330r59e6ZMmUJERATNmjXjzTffZOjQoQBkZWXRvn170tLS6N69O+vWrePWW2/l2LFjhISEAPDyyy8zadIkjh8/bvpiizpfahEREalN3PmuloCAAK688kqHY/7+/jRp0sR+PD4+nokTJxIcHIzVauXBBx8kOjqa7t27A9C3b186dOjA8OHDmTdvHjk5OTz22GOMGzeuWlZ4KvEQERExk80o31zpb6Lnn38eDw8Phg4dSlFREbGxsbz44ov2856enqxZs4axY8cSHR2Nv78/I0eOZPbs2abGUUGJh4iIiJlcfPqoS32BTz/91GHf19eXF154gRdeeOGCfVq2bMmHH37o2gdXUp2fXCoiIiJ1h0Y8RERETGTBxTkepkVSOynxEBERMdOvnj560f3rMZVaRERExG004iEiImIidy6nrYuUeIiIiJiphle11HYqtYiIiIjbaMRDRETERBbDwOLCBFFX+tYFSjxERETMZPvf5kr/ekylFhEREXEbjXiIiIiYSKUW55R4iIiImEmrWpxS4iEiImImPbnUKc3xEBEREbfRiIeIiIiJ9ORS55R4iIiImEmlFqdUahERERG30YiHiIiIiSy28s2V/vWZEg8REREzqdTilEotIiIi4jYa8RARETGTHiDmlBIPERERE+mR6c6p1CIiIiJuoxEPERERM2lyqVNKPERERMxkAK4sia3feYcSDxERETNpjodzmuMhIiIibqMRDxERETMZuDjHw7RIaiUlHiIiImbS5FKnVGoRERERt9GIh4iIiJlsgMXF/vWYEg8RERETaVWLcyq1iIiIiNtoxENERMRMmlzqlEY8REREzFSReLiyVUFSUhLXXnstAQEBhISEMGjQILKyshzaFBYWMm7cOJo0aUKjRo0YOnQoubm5Dm2ys7MZMGAADRs2JCQkhEmTJlFaWuryt+O3lHiIiIjUYZs2bWLcuHF8/vnnpKamUlJSQt++fTlz5oy9TWJiIqtXr+add95h06ZNHD16lCFDhtjPl5WVMWDAAIqLi9m2bRvLli1j6dKlTJ8+3fR4LYZRz8d0TJCfn09gYCC9GYiXpUFNhyNSLT46mlHTIYhUm/zTNhq3/ZZTp05htVqr5zP+97vipqi/4+Xpc9HXKS0r4pPMZy861h9//JGQkBA2bdpEz549OXXqFM2aNWPFihXccccdAOzbt4+oqCjS0tLo3r0769at49Zbb+Xo0aOEhoYCkJyczJQpU/jxxx/x9va+6Pv5LY14iIiImMlmwkZ5IvPrraioqFIff+rUKQCCg4MBSE9Pp6SkhJiYGHub9u3b06JFC9LS0gBIS0ujY8eO9qQDIDY2lvz8fPbs2XMx34ULUuIhIiJioorltK5sABEREQQGBtq3pKSk3/1sm83GhAkTuP7667nyyisByMnJwdvbm6CgIIe2oaGh5OTk2Nv8OumoOF9xzkxa1SIiIlILHTlyxKHU4uPz++WbcePGsXv3brZu3VqdoblEiYeIiIiZTFpOa7VaqzTHIyEhgTVr1rB582aaN29uPx4WFkZxcTEnT550GPXIzc0lLCzM3uaLL75wuF7FqpeKNmZRqUVERMRMNsP1rQoMwyAhIYH33nuPjRs3EhkZ6XC+S5cuNGjQgE8++cR+LCsri+zsbKKjowGIjo7m66+/5vjx4/Y2qampWK1WOnTo4MI341wa8RAREanDxo0bx4oVK3j//fcJCAiwz8kIDAzEz8+PwMBA4uPjmThxIsHBwVitVh588EGio6Pp3r07AH379qVDhw4MHz6cefPmkZOTw2OPPca4ceMqVeKpCiUeIiIiZnLzk0sXL14MQO/evR2OL1myhFGjRgHw/PPP4+HhwdChQykqKiI2NpYXX3zR3tbT05M1a9YwduxYoqOj8ff3Z+TIkcyePfvi7+MClHiIiIiYysXEg6qXWn6Pr68vL7zwAi+88MIF27Rs2ZIPP/ywSp99MTTHQ0RERNxGIx4iIiJm0kvinFLiISIiYiabQVXLJef2r79UahERERG30YiHiIiImQxb+eZK/3pMiYeIiIiZNMfDKSUeIiIiZtIcD6c0x0NERETcRiMeIiIiZlKpxSklHiIiImYycDHxMC2SWkmlFhEREXEbjXiIiIiYSaUWp5R4iIiImMlmA1x4Foetfj/HQ6UWERERcRuNeIiIiJhJpRanlHiIiIiYSYmHUyq1iIiIiNtoxENERMRMemS6U0o8RERETGQYNgwX3jDrSt+6QImHiIiImQzDtVELzfEQERERMYdGPERERMxkuDjHo56PeCjxEBERMZPNBhYX5mnU8zkeKrWIiIiI22jEQ0RExEwqtTilxENERMREhs2G4UKppb4vp1WpRURERNxGIx4iIiJmUqnFKSUeIiIiZrIZYFHicSEqtYiIiIjbaMRDRETETIYBuPIcj/o94qHEQ0RExESGzcBwodRiKPEQERGRSjNsuDbioeW0IiIiIqbQiIeIiIiJVGpxTomHiIiImVRqcUqJRyVUZJ+llLj0TBiR2iz/dP3+x07+2PILyv//dsdogqu/K0opMS+YWkiJRyWcPn0agK18WMORiFSfxm1rOgKR6nf69GkCAwOr5dre3t6EhYWxNcf13xVhYWF4e3ubEFXtYzHqezHJBDabjaNHjxIQEIDFYqnpcP4Q8vPziYiI4MiRI1it1poOR8RU+v/b/QzD4PTp04SHh+PhUX3rKgoLCykuLnb5Ot7e3vj6+poQUe2jEY9K8PDwoHnz5jUdxh+S1WrVP8xSb+n/b/eqrpGOX/P19a23CYNZtJxWRERE3EaJh4iIiLiNEg+plXx8fJgxYwY+Pj41HYqI6fT/t/yRaXKpiIiIuI1GPERERMRtlHiIiIiI2yjxEBEREbdR4iG1ytKlSwkKCqrpMEREpJoo8ZBqMWrUKCwWyznbgQMHajo0EVOd7//zX28zZ86s6RBFahU9uVSqTb9+/ViyZInDsWbNmtVQNCLV49ixY/avV65cyfTp08nKyrIfa9Sokf1rwzAoKyvDy0v/9Mofl0Y8pNr4+PgQFhbmsC1YsICOHTvi7+9PREQEDzzwAAUFBRe8xldffUWfPn0ICAjAarXSpUsXdu7caT+/detWevTogZ+fHxEREYwfP54zZ8644/ZEABz+/w4MDMRisdj39+3bR0BAAOvWraNLly74+PiwdetWRo0axaBBgxyuM2HCBHr37m3ft9lsJCUlERkZiZ+fH507d+af//yne29OpBoo8RC38vDwYOHChezZs4dly5axceNGJk+efMH2cXFxNG/enB07dpCens7UqVNp0KABAAcPHqRfv34MHTqUXbt2sXLlSrZu3UpCQoK7bkekUqZOncqTTz5JZmYmnTp1qlSfpKQkli9fTnJyMnv27CExMZF77rmHTZs2VXO0ItVL431SbdasWeMwzNy/f3/eeecd+36rVq2YO3cu999/Py+++OJ5r5Gdnc2kSZNo3749AJdffrn9XFJSEnFxcUyYMMF+buHChfTq1YvFixfrRU1Sa8yePZubb7650u2Liop44okn+Pjjj4mOjgagdevWbN26lZdeeolevXpVV6gi1U6Jh1SbPn36sHjxYvu+v78/H3/8MUlJSezbt4/8/HxKS0spLCzk7NmzNGzY8JxrTJw4kfvuu4833niDmJgY/vznP3PZZZcB5WWYXbt2kZKSYm9vGAY2m41Dhw4RFRVV/TcpUgldu3atUvsDBw5w9uzZc5KV4uJirr76ajNDE3E7JR5Sbfz9/WnTpo19//Dhw9x6662MHTuWxx9/nODgYLZu3Up8fDzFxcXnTTxmzpzJ3Xffzdq1a1m3bh0zZszgrbfeYvDgwRQUFPC3v/2N8ePHn9OvRYsW1XpvIlXh7+/vsO/h4cFv31ZRUlJi/7pi3tPatWu59NJLHdrp/S5S1ynxELdJT0/HZrPx7LPP4uFRPr3o7bff/t1+bdu2pW3btiQmJnLXXXexZMkSBg8ezDXXXMPevXsdkhuRuqBZs2bs3r3b4VhGRoZ9/lKHDh3w8fEhOztbZRWpdzS5VNymTZs2lJSUsGjRIr799lveeOMNkpOTL9j+559/JiEhgU8//ZTvvvuOzz77jB07dthLKFOmTGHbtm0kJCSQkZHB/v37ef/99zW5VGq9G2+8kZ07d7J8+XL279/PjBkzHBKRgIAAHn74YRITE1m2bBkHDx7kyy+/ZNGiRSxbtqwGIxdxnRIPcZvOnTvz3HPP8dRTT3HllVeSkpJCUlLSBdt7enry008/MWLECNq2bcudd95J//79mTVrFgCdOnVi06ZNfPPNN/To0YOrr76a6dOnEx4e7q5bErkosbGxTJs2jcmTJ3Pttddy+vRpRowY4dBmzpw5TJs2jaSkJKKioujXrx9r164lMjKyhqIWMYfF+G2hUURERKSaaMRDRERE3EaJh4iIiLiNEg8RERFxGyUeIiIi4jZKPERERMRtlHiIiIiI2yjxEBEREbdR4iEiIiJuo8RDpI4YNWoUgwYNsu/37t2bCRMmuD2OTz/9FIvFwsmTJy/YxmKxsGrVqkpfc+bMmVx11VUuxXX48GEsFgsZGRkuXUdEqpcSDxEXjBo1CovFgsViwdvbmzZt2jB79mxKS0ur/bP/9a9/MWfOnEq1rUyyICLiDno7rYiL+vXrx5IlSygqKuLDDz9k3LhxNGjQgEceeeSctsXFxXh7e5vyucHBwaZcR0TEnTTiIeIiHx8fwsLCaNmyJWPHjiUmJoYPPvgA+KU88vjjjxMeHk67du0AOHLkCHfeeSdBQUEEBwczcOBADh8+bL9mWVkZEydOJCgoiCZNmjB58mR++1ql35ZaioqKmDJlChEREfj4+NCmTRtee+01Dh8+TJ8+fQBo3LgxFouFUaNGAWCz2UhKSiIyMhI/Pz86d+7MP//5T4fP+fDDD2nbti1+fn706dPHIc7KmjJlCm3btqVhw4a0bt2aadOmUVJSck67l156iYiICBo2bMidd97JqVOnHM6/+uqrREVF4evrS/v27XnxxRerHIuI1CwlHiIm8/Pzo7i42L7/ySefkJWVRWpqKmvWrKGkpITY2FgCAgLYsmULn332GY0aNaJfv372fs8++yxLly7l9ddfZ+vWreTl5fHee+85/dwRI0bw5ptvsnDhQjIzM3nppZdo1KgRERERvPvuuwBkZWVx7NgxFixYAEBSUhLLly8nOTmZPXv2kJiYyD333MOmTZuA8gRpyJAh3HbbbWRkZHDfffcxderUKn9PAgICWLp0KXv37mXBggW88sorPP/88w5tDhw4wNtvv83q1atZv349//nPf3jggQfs51NSUpg+fTqPP/44mZmZPPHEE0ybNk2viRepawwRuWgjR440Bg4caBiGYdhsNiM1NdXw8fExHn74Yfv50NBQo6ioyN7njTfeMNq1a2fYbDb7saKiIsPPz8/46KOPDMMwjEsuucSYN2+e/XxJSYnRvHlz+2cZhmH06tXLeOihhwzDMIysrCwDMFJTU88b57///W8DME6cOGE/VlhYaDRs2NDYtm2bQ9v4+HjjrrvuMgzDMB555BGjQ4cODuenTJlyzrV+CzDee++9C55/+umnjS5dutj3Z8yYYXh6ehrff/+9/di6desMDw8P49ixY4ZhGMZll11mrFixwuE6c+bMMaKjow3DMIxDhw4ZgPGf//zngp8rIjVPczxEXLRmzRoaNWpESUkJNpuNu+++m5kzZ9rPd+zY0WFex1dffcWBAwcICAhwuE5hYSEHDx7k1KlTHDt2jG7dutnPeXl50bVr13PKLRUyMjLw9PSkV69elY77wIEDnD17lptvvtnheHFxMVdffTUAmZmZDnEAREdHV/ozKqxcuZKFCxdy8OBBCgoKKC0txWq1OrRp0aIFl156qcPn2Gw2srKyCAgI4ODBg8THxzN69Gh7m9LSUgIDA6scj4jUHCUeIi7q06cPixcvxtvbm/DwcLy8HH+s/P39HfYLCgro0qULKSkp51yrWbNmFxWDn59flfsUFBQAsHbtWodf+FA+b8UsaWlpxMXFMWvWLGJjYwkMDOStt97i2WefrXKsr7zyyjmJkKenp2mxikj1U+Ih4iJ/f3/atGlT6fbXXHMNK1euJCQk5Jy/+itccsklbN++nZ49ewLlf9mnp6dzzTXXnLd9x44dsdlsbNq0iZiYmHPOV4y4lJWV2Y916NABHx8fsrOzLzhSEhUVZZ8oW+Hzzz///Zv8lW3bttGyZUseffRR+7HvvvvunHbZ2dkcPXqU8PBw++d4eHjQrl07QkNDCQ8P59tvvyUuLq5Kny8itYsml4q4WVxcHE2bNmXgwIFs2bKFQ4cO8emnnzJ+/Hi+//57AB566CGefPJJVq1axb59+3jggQecPoOjVatWjBw5kr/+9a+sWrXKfs23334bgJYtW2KxWFizZg0//vgjBQUFBAQE8PDDD5OYmMiyZcs4ePAgX375JYsWLbJP2Lz//vvZv38/kyZNIisrixUrVrB06dIq3e/ll19OdnY2b731FgcPHmThwoXnnSjr6+vLyJEj+eqrr9iyZQvjx4/nzjvvJCwsDIBZs2aRlJTEwoUL+eabb/j6669ZsmQJzz33XJXiEZGapcRDxM0aNmzI5s2badGiBUOGDCEqKor4+HgKCwvtIyB///vfGT58OCNHjiQ6OpqAgAAGDx7s9LqLFy/mjjvu4IEHHqB9+/aMHj2aM2fOAHDppZcya9Yspk6dSmhoKAkJCQDMmTOHadOmkZSURFRUFP369WPt2rVERkYC5fMu3n33XVatWkXnzp1JTk7miSeeqNL93n777SQmJpKQkMBVV13Ftm3bmDZt2jnt2rRpw5AhQ7jlllvo27cvnTp1clgue9999/Hqq6+yZMkSOnbsSK9evVi6dKk9VhGpGyzGhWariYiIiJhMIx4iIiLiNko8RERExG2UeIiIiIjbKPEQERERt1HiISIiIm6jxENERETcRomHiIiIuI0SDxEREXEbJR4iIiLiNko8RERExG2UeIiIiIjb/H+zA8hSQVfiJgAAAABJRU5ErkJggg==\n"},"metadata":{}}]},{"cell_type":"markdown","source":["Training the data with best model (XGBOOST)"],"metadata":{"id":"iBkwlJAEEb8G"}},{"cell_type":"code","source":["from xgboost import XGBClassifier\n","model = XGBClassifier(random_state=42)\n","path = \"/content/drive/MyDrive/Predictiong_cvss_client/Fraud_Credit_Card/Artifacts\"\n","os.chdir(path)\n","model.fit(x_train,y_train)# Save the model\n","model.save_model('xgboost_model.model')\n","\"\"\"\n","# Load the model\n","loaded_model = xgb.XGBClassifier()\n","loaded_model.load_model('xgboost_model.model')\"\"\"\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":36},"id":"qU1EljiCDpXi","executionInfo":{"status":"ok","timestamp":1684417471605,"user_tz":-60,"elapsed":2154,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"3be86fdb-34d9-472f-821e-3142e963bcd8"},"execution_count":88,"outputs":[{"output_type":"execute_result","data":{"text/plain":["\"\\n# Load the model\\nloaded_model = xgb.XGBClassifier()\\nloaded_model.load_model('xgboost_model.model')\""],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":88}]},{"cell_type":"markdown","source":["### Clustering Customers local balance and the amount transferd by the customers \n","This aims to extract important information by grouping customers based on their balance and the amount transferred by the customers which will allow us to gain more insight on the transactions that are fraudlent"],"metadata":{"id":"TGgVfRzyuozz"}},{"cell_type":"code","source":["newdata.head()"],"metadata":{"id":"DL_kjG-lK_Vs","colab":{"base_uri":"https://localhost:8080/","height":206},"executionInfo":{"status":"ok","timestamp":1672083747754,"user_tz":-60,"elapsed":1813,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"b693ef28-3906-4dc3-f29a-59b25c5928c6"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount oldbalanceOrg newbalanceOrig oldbalanceDest \\\n","0 1 4 5.204007 5.204007 0.0 0.000000 \n","1 1 1 5.204007 5.204007 0.0 9.960954 \n","2 1 4 7.939872 7.939872 0.0 0.000000 \n","3 1 1 7.939872 7.939872 0.0 10.173629 \n","4 1 4 9.909917 9.909917 0.0 0.000000 \n","\n"," newbalanceDest isFraud isFlaggedFraud transferAmt \n","0 0.0 1 0 5.204007 \n","1 0.0 1 0 5.204007 \n","2 0.0 1 0 7.939872 \n","3 0.0 1 0 7.939872 \n","4 0.0 1 0 9.909917 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudisFlaggedFraudtransferAmt
0145.2040075.2040070.00.0000000.0105.204007
1115.2040075.2040070.09.9609540.0105.204007
2147.9398727.9398720.00.0000000.0107.939872
3117.9398727.9398720.010.1736290.0107.939872
4149.9099179.9099170.00.0000000.0109.909917
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":20}]},{"cell_type":"code","source":["plt.scatter(newdata[\"amount\"],newdata[\"transferAmt\"])\n","plt.xlabel(\"Amount in local bank\")\n","plt.ylabel(\"Amount trasnfered\")\n","plt.show()"],"metadata":{"id":"XsYm8MQNK_Gf","colab":{"base_uri":"https://localhost:8080/","height":279},"executionInfo":{"status":"ok","timestamp":1672084530121,"user_tz":-60,"elapsed":850,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"eedd038e-6780-44f4-f983-28951294a218"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":["#Create kmeans object\n","from sklearn.cluster import KMeans\n","km = KMeans(n_clusters=3)\n","km"],"metadata":{"id":"R2bqbDlhK-6p","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1672084734135,"user_tz":-60,"elapsed":882,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"baaac3ae-e1f2-45b8-b19b-915ea3ebbcbd"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["KMeans(n_clusters=3)"]},"metadata":{},"execution_count":32}]},{"cell_type":"code","source":["#Scale required features"],"metadata":{"id":"poPhG3WTBX68"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["newdata.isnull().sum()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"T-zLWrblBoMu","executionInfo":{"status":"ok","timestamp":1672084894107,"user_tz":-60,"elapsed":518,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"48463b49-57c5-4896-c6dc-8c5608a94f39"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["step 0\n","type 0\n","amount 0\n","oldbalanceOrg 0\n","newbalanceOrig 0\n","oldbalanceDest 0\n","newbalanceDest 0\n","isFraud 0\n","isFlaggedFraud 0\n","transferAmt 1843\n","dtype: int64"]},"metadata":{},"execution_count":36}]},{"cell_type":"code","source":["newdata[\"transferAmt\"] = newdata[\"transferAmt\"].fillna(newdata[\"transferAmt\"].mean())"],"metadata":{"id":"GsJbPFlNBtPi"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["scaler = MinMaxScaler()\n","for feature in [\"amount\",\"transferAmt\"]:\n"," newdata[feature] = scaler.fit_transform(newdata[[feature]])"],"metadata":{"id":"RqkG7S7iQB6T"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["preds = km.fit_predict(newdata[[\"amount\",\"transferAmt\"]])\n","preds"],"metadata":{"id":"prHMHJ9yK-sM","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1672089476935,"user_tz":-60,"elapsed":2133,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"3607cd4e-7b9f-4310-c1c5-ad1afca781b4"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([0, 0, 0, ..., 0, 0, 1], dtype=int32)"]},"metadata":{},"execution_count":54}]},{"cell_type":"code","source":["newdata[\"Cluster\"] = preds"],"metadata":{"id":"jqNSNBTwMW7G"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["newdata.head()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"XUyAa6CoMdk6","executionInfo":{"status":"ok","timestamp":1672089483519,"user_tz":-60,"elapsed":750,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"c8ada826-9a54-41f8-a2c4-835b68c3f7f9"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" step type amount oldbalanceOrg newbalanceOrig oldbalanceDest \\\n","0 1 4 0.304322 5.204007 0.0 0.000000 \n","1 1 1 0.304322 5.204007 0.0 9.960954 \n","2 1 4 0.464312 7.939872 0.0 0.000000 \n","3 1 1 0.464312 7.939872 0.0 10.173629 \n","4 1 4 0.579517 9.909917 0.0 0.000000 \n","\n"," newbalanceDest isFraud isFlaggedFraud transferAmt Cluster \n","0 0.0 1 0 0.322867 0 \n","1 0.0 1 0 0.322867 0 \n","2 0.0 1 0 0.492606 0 \n","3 0.0 1 0 0.492606 0 \n","4 0.0 1 0 0.614832 0 "],"text/html":["\n","
\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
steptypeamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudisFlaggedFraudtransferAmtCluster
0140.3043225.2040070.00.0000000.0100.3228670
1110.3043225.2040070.09.9609540.0100.3228670
2140.4643127.9398720.00.0000000.0100.4926060
3110.4643127.9398720.010.1736290.0100.4926060
4140.5795179.9099170.00.0000000.0100.6148320
\n","
\n"," \n"," \n"," \n","\n"," \n","
\n","
\n"," "]},"metadata":{},"execution_count":55}]},{"cell_type":"code","source":["newdata[\"Cluster\"].unique()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"w0GB161RMfnh","executionInfo":{"status":"ok","timestamp":1672087771973,"user_tz":-60,"elapsed":589,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"ae55c513-c552-40c8-e35d-2be0ae544704"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([0, 1, 2], dtype=int32)"]},"metadata":{},"execution_count":43}]},{"cell_type":"code","source":["df1 = newdata[newdata.Cluster==0]\n","df2 = newdata[newdata.Cluster==1]\n","df3 = newdata[newdata.Cluster==2]\n","plt.scatter(df1[\"amount\"],df1[\"transferAmt\"],color=\"green\")\n","plt.scatter(df2[\"amount\"],df2[\"transferAmt\"],color=\"red\")\n","plt.scatter(df3[\"amount\"],df3[\"transferAmt\"],color=\"blue\")\n","plt.scatter(km.cluster_centers_[:,0],km.cluster_centers_[:,1],color=\"purple\",marker='*',label='centroid')\n","plt.legend()\n","\n","\n","plt.xlabel(\"Amount in local bank\")\n","plt.ylabel(\"Amount trasnfered\")\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":279},"id":"KZGl9HIjMqUY","executionInfo":{"status":"ok","timestamp":1672089716799,"user_tz":-60,"elapsed":1545,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"255de077-182c-4dda-c86f-74e6a0ce0cbd"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":["#Implementing elbow method\n","kRange = range(1,7)\n","sse = []\n","\n","for k in kRange:\n"," km = KMeans(n_clusters=k)\n"," km.fit(newdata[[\"amount\",\"transferAmt\"]])\n"," sse.append(km.inertia_)"],"metadata":{"id":"weZ5JcJxBSXN"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["#plot result\n","plt.xlabel(\"K\")\n","plt.ylabel(\"Sum of squared error\")\n","plt.plot(kRange,sse)\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":279},"id":"rvTDZ-viUwY0","executionInfo":{"status":"ok","timestamp":1672090173350,"user_tz":-60,"elapsed":21,"user":{"displayName":"Owusu Samuel","userId":"05986356755319312051"}},"outputId":"fc6f6e67-5aac-4c29-a152-c76676702a03"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Comment 3 as the number of clusters is good choice"],"metadata":{"id":"tfiKXQWcVsNT"}}]}