Major Project-Credit Card Fraud Detection

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":8297581,"sourceType":"datasetVersion","datasetId":4929079}],"dockerImageVersionId":30698,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\n\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.model_selection import train_test_split\n\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.svm import SVC\nfrom sklearn.neural_network import MLPClassifier","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-05-03T05:50:02.728562Z","iopub.execute_input":"2024-05-03T05:50:02.729038Z","iopub.status.idle":"2024-05-03T05:50:06.831414Z","shell.execute_reply.started":"2024-05-03T05:50:02.729000Z","shell.execute_reply":"2024-05-03T05:50:06.829912Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = pd.read_csv('/kaggle/input/credit-card-data-set/cards_dataset.csv', engine='python')","metadata":{"execution":{"iopub.status.busy":"2024-05-03T05:52:31.771712Z","iopub.execute_input":"2024-05-03T05:52:31.772773Z","iopub.status.idle":"2024-05-03T05:52:46.026897Z","shell.execute_reply.started":"2024-05-03T05:52:31.772738Z","shell.execute_reply":"2024-05-03T05:52:46.025872Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:27.003159Z","iopub.execute_input":"2024-05-03T06:24:27.004680Z","iopub.status.idle":"2024-05-03T06:24:27.025346Z","shell.execute_reply.started":"2024-05-03T06:24:27.004626Z","shell.execute_reply":"2024-05-03T06:24:27.023663Z"},"trusted":true},"execution_count":22,"outputs":[{"execution_count":22,"output_type":"execute_result","data":{"text/plain":"        distance_from_home  distance_from_last_transaction  \\\n0                57.877857                        0.311140   \n1                10.829943                        0.175592   \n2                 5.091079                        0.805153   \n3                 2.247564                        5.600044   \n4                44.190936                        0.566486   \n...                    ...                             ...   \n999995            2.207101                        0.112651   \n999996           19.872726                        2.683904   \n999997            2.914857                        1.472687   \n999998            4.258729                        0.242023   \n999999           58.108125                        0.318110   \n\n        ratio_to_median_purchase_price  repeat_retailer  used_chip  \\\n0                             1.945940                1          1   \n1                             1.294219                1          0   \n2                             0.427715                1          0   \n3                             0.362663                1          1   \n4                             2.222767                1          1   \n...                                ...              ...        ...   \n999995                        1.626798                1          1   \n999996                        2.778303                1          1   \n999997                        0.218075                1          1   \n999998                        0.475822                1          0   \n999999                        0.386920                1          1   \n\n        used_pin_number  online_order  fraud  \n0                     0             0      0  \n1                     0             0      0  \n2                     0             1      0  \n3                     0             1      0  \n4                     0             1      0  \n...                 ...           ...    ...  \n999995                0             0      0  \n999996                0             0      0  \n999997                0             1      0  \n999998                0             1      0  \n999999                0             1      0  \n\n[1000000 rows x 8 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>distance_from_home</th>\n      <th>distance_from_last_transaction</th>\n      <th>ratio_to_median_purchase_price</th>\n      <th>repeat_retailer</th>\n      <th>used_chip</th>\n      <th>used_pin_number</th>\n      <th>online_order</th>\n      <th>fraud</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>57.877857</td>\n      <td>0.311140</td>\n      <td>1.945940</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>10.829943</td>\n      <td>0.175592</td>\n      <td>1.294219</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>5.091079</td>\n      <td>0.805153</td>\n      <td>0.427715</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2.247564</td>\n      <td>5.600044</td>\n      <td>0.362663</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>44.190936</td>\n      <td>0.566486</td>\n      <td>2.222767</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>999995</th>\n      <td>2.207101</td>\n      <td>0.112651</td>\n      <td>1.626798</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>999996</th>\n      <td>19.872726</td>\n      <td>2.683904</td>\n      <td>2.778303</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>999997</th>\n      <td>2.914857</td>\n      <td>1.472687</td>\n      <td>0.218075</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>999998</th>\n      <td>4.258729</td>\n      <td>0.242023</td>\n      <td>0.475822</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>999999</th>\n      <td>58.108125</td>\n      <td>0.318110</td>\n      <td>0.386920</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>1000000 rows × 8 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"data.info()","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:30.665218Z","iopub.execute_input":"2024-05-03T06:24:30.665734Z","iopub.status.idle":"2024-05-03T06:24:30.694642Z","shell.execute_reply.started":"2024-05-03T06:24:30.665696Z","shell.execute_reply":"2024-05-03T06:24:30.693277Z"},"trusted":true},"execution_count":23,"outputs":[{"name":"stdout","text":"<class 'pandas.core.frame.DataFrame'>\nRangeIndex: 1000000 entries, 0 to 999999\nData columns (total 8 columns):\n #   Column                          Non-Null Count    Dtype  \n---  ------                          --------------    -----  \n 0   distance_from_home              1000000 non-null  float64\n 1   distance_from_last_transaction  1000000 non-null  float64\n 2   ratio_to_median_purchase_price  1000000 non-null  float64\n 3   repeat_retailer                 1000000 non-null  int64  \n 4   used_chip                       1000000 non-null  int64  \n 5   used_pin_number                 1000000 non-null  int64  \n 6   online_order                    1000000 non-null  int64  \n 7   fraud                           1000000 non-null  int64  \ndtypes: float64(3), int64(5)\nmemory usage: 61.0 MB\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Splitting","metadata":{}},{"cell_type":"code","source":"def preprocess_inputs(df):\n    df = df.copy()\n    \n    y = df['fraud'].copy()\n    X = df.drop('fraud',axis = 1).copy()\n    \n    return X,y","metadata":{"execution":{"iopub.status.busy":"2024-05-03T05:53:34.102249Z","iopub.execute_input":"2024-05-03T05:53:34.102786Z","iopub.status.idle":"2024-05-03T05:53:34.111007Z","shell.execute_reply.started":"2024-05-03T05:53:34.102752Z","shell.execute_reply":"2024-05-03T05:53:34.109028Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X, y = preprocess_inputs(data)","metadata":{"execution":{"iopub.status.busy":"2024-05-03T05:53:42.759813Z","iopub.execute_input":"2024-05-03T05:53:42.760251Z","iopub.status.idle":"2024-05-03T05:53:42.858923Z","shell.execute_reply.started":"2024-05-03T05:53:42.760219Z","shell.execute_reply":"2024-05-03T05:53:42.857721Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:35.901517Z","iopub.execute_input":"2024-05-03T06:24:35.901951Z","iopub.status.idle":"2024-05-03T06:24:35.927176Z","shell.execute_reply.started":"2024-05-03T06:24:35.901919Z","shell.execute_reply":"2024-05-03T06:24:35.925169Z"},"trusted":true},"execution_count":24,"outputs":[{"execution_count":24,"output_type":"execute_result","data":{"text/plain":"        distance_from_home  distance_from_last_transaction  \\\n0                 0.477882                       -0.182849   \n1                -0.241607                       -0.188094   \n2                -0.329369                       -0.163733   \n3                -0.372854                        0.021806   \n4                 0.268572                       -0.172968   \n...                    ...                             ...   \n999995           -0.373473                       -0.190529   \n999996           -0.103318                       -0.091035   \n999997           -0.362650                       -0.137903   \n999998           -0.342098                       -0.185523   \n999999            0.481403                       -0.182579   \n\n        ratio_to_median_purchase_price  repeat_retailer  used_chip  \\\n0                             0.043491         0.366584   1.361576   \n1                            -0.189300         0.366584  -0.734443   \n2                            -0.498812         0.366584  -0.734443   \n3                            -0.522048         0.366584   1.361576   \n4                             0.142373         0.366584   1.361576   \n...                                ...              ...        ...   \n999995                       -0.070505         0.366584   1.361576   \n999996                        0.340808         0.366584   1.361576   \n999997                       -0.573694         0.366584   1.361576   \n999998                       -0.481628         0.366584  -0.734443   \n999999                       -0.513384         0.366584   1.361576   \n\n        used_pin_number  online_order  \n0             -0.334458     -1.364425  \n1             -0.334458     -1.364425  \n2             -0.334458      0.732909  \n3             -0.334458      0.732909  \n4             -0.334458      0.732909  \n...                 ...           ...  \n999995        -0.334458     -1.364425  \n999996        -0.334458     -1.364425  \n999997        -0.334458      0.732909  \n999998        -0.334458      0.732909  \n999999        -0.334458      0.732909  \n\n[1000000 rows x 7 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>distance_from_home</th>\n      <th>distance_from_last_transaction</th>\n      <th>ratio_to_median_purchase_price</th>\n      <th>repeat_retailer</th>\n      <th>used_chip</th>\n      <th>used_pin_number</th>\n      <th>online_order</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.477882</td>\n      <td>-0.182849</td>\n      <td>0.043491</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>-0.241607</td>\n      <td>-0.188094</td>\n      <td>-0.189300</td>\n      <td>0.366584</td>\n      <td>-0.734443</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>-0.329369</td>\n      <td>-0.163733</td>\n      <td>-0.498812</td>\n      <td>0.366584</td>\n      <td>-0.734443</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>-0.372854</td>\n      <td>0.021806</td>\n      <td>-0.522048</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0.268572</td>\n      <td>-0.172968</td>\n      <td>0.142373</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>999995</th>\n      <td>-0.373473</td>\n      <td>-0.190529</td>\n      <td>-0.070505</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>999996</th>\n      <td>-0.103318</td>\n      <td>-0.091035</td>\n      <td>0.340808</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>999997</th>\n      <td>-0.362650</td>\n      <td>-0.137903</td>\n      <td>-0.573694</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>999998</th>\n      <td>-0.342098</td>\n      <td>-0.185523</td>\n      <td>-0.481628</td>\n      <td>0.366584</td>\n      <td>-0.734443</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>999999</th>\n      <td>0.481403</td>\n      <td>-0.182579</td>\n      <td>-0.513384</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n  </tbody>\n</table>\n<p>1000000 rows × 7 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"y","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:39.425733Z","iopub.execute_input":"2024-05-03T06:24:39.426253Z","iopub.status.idle":"2024-05-03T06:24:39.438840Z","shell.execute_reply.started":"2024-05-03T06:24:39.426216Z","shell.execute_reply":"2024-05-03T06:24:39.436910Z"},"trusted":true},"execution_count":25,"outputs":[{"execution_count":25,"output_type":"execute_result","data":{"text/plain":"0         0\n1         0\n2         0\n3         0\n4         0\n         ..\n999995    0\n999996    0\n999997    0\n999998    0\n999999    0\nName: fraud, Length: 1000000, dtype: int64"},"metadata":{}}]},{"cell_type":"markdown","source":"# Visualization","metadata":{}},{"cell_type":"code","source":"corr = data.corr()\n\nplt.figure(figsize=(18, 15))\nsns.heatmap(corr, annot=True, vmin=-1.0, cmap='mako')\nplt.title(\"Correlation Heatmap\")\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:43.254434Z","iopub.execute_input":"2024-05-03T06:24:43.254852Z","iopub.status.idle":"2024-05-03T06:24:44.212906Z","shell.execute_reply.started":"2024-05-03T06:24:43.254823Z","shell.execute_reply":"2024-05-03T06:24:44.211648Z"},"trusted":true},"execution_count":26,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1800x1500 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"# Preprocessing","metadata":{}},{"cell_type":"code","source":"def onehot_encode(df, column_dict):\n    df = df.copy()\n    for column, prefix in column_dict.items():\n        dummies = pd.get_dummies(df[column], prefix=prefix)\n        df = pd.concat([df, dummies], axis=1)\n        df = df.drop(column, axis=1)\n    return df","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:48.742142Z","iopub.execute_input":"2024-05-03T06:24:48.742547Z","iopub.status.idle":"2024-05-03T06:24:48.750743Z","shell.execute_reply.started":"2024-05-03T06:24:48.742518Z","shell.execute_reply":"2024-05-03T06:24:48.748837Z"},"trusted":true},"execution_count":27,"outputs":[]},{"cell_type":"code","source":"def preprocess_inputs(df):\n    df = df.copy()\n    \n    # Split df into X and y\n    y = df['fraud'].copy()\n    X = df.drop('fraud', axis=1).copy()\n    \n    # Scale X with a standard scaler\n    scaler = StandardScaler()\n    X = pd.DataFrame(scaler.fit_transform(X), columns=X.columns)\n    \n    return X, y","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:51.698303Z","iopub.execute_input":"2024-05-03T06:24:51.698765Z","iopub.status.idle":"2024-05-03T06:24:51.707010Z","shell.execute_reply.started":"2024-05-03T06:24:51.698733Z","shell.execute_reply":"2024-05-03T06:24:51.705499Z"},"trusted":true},"execution_count":28,"outputs":[]},{"cell_type":"code","source":"X, y = preprocess_inputs(data)","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:54.231327Z","iopub.execute_input":"2024-05-03T06:24:54.231814Z","iopub.status.idle":"2024-05-03T06:24:54.536496Z","shell.execute_reply.started":"2024-05-03T06:24:54.231782Z","shell.execute_reply":"2024-05-03T06:24:54.534891Z"},"trusted":true},"execution_count":29,"outputs":[]},{"cell_type":"code","source":"X","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:56.220092Z","iopub.execute_input":"2024-05-03T06:24:56.220505Z","iopub.status.idle":"2024-05-03T06:24:56.242925Z","shell.execute_reply.started":"2024-05-03T06:24:56.220462Z","shell.execute_reply":"2024-05-03T06:24:56.241573Z"},"trusted":true},"execution_count":30,"outputs":[{"execution_count":30,"output_type":"execute_result","data":{"text/plain":"        distance_from_home  distance_from_last_transaction  \\\n0                 0.477882                       -0.182849   \n1                -0.241607                       -0.188094   \n2                -0.329369                       -0.163733   \n3                -0.372854                        0.021806   \n4                 0.268572                       -0.172968   \n...                    ...                             ...   \n999995           -0.373473                       -0.190529   \n999996           -0.103318                       -0.091035   \n999997           -0.362650                       -0.137903   \n999998           -0.342098                       -0.185523   \n999999            0.481403                       -0.182579   \n\n        ratio_to_median_purchase_price  repeat_retailer  used_chip  \\\n0                             0.043491         0.366584   1.361576   \n1                            -0.189300         0.366584  -0.734443   \n2                            -0.498812         0.366584  -0.734443   \n3                            -0.522048         0.366584   1.361576   \n4                             0.142373         0.366584   1.361576   \n...                                ...              ...        ...   \n999995                       -0.070505         0.366584   1.361576   \n999996                        0.340808         0.366584   1.361576   \n999997                       -0.573694         0.366584   1.361576   \n999998                       -0.481628         0.366584  -0.734443   \n999999                       -0.513384         0.366584   1.361576   \n\n        used_pin_number  online_order  \n0             -0.334458     -1.364425  \n1             -0.334458     -1.364425  \n2             -0.334458      0.732909  \n3             -0.334458      0.732909  \n4             -0.334458      0.732909  \n...                 ...           ...  \n999995        -0.334458     -1.364425  \n999996        -0.334458     -1.364425  \n999997        -0.334458      0.732909  \n999998        -0.334458      0.732909  \n999999        -0.334458      0.732909  \n\n[1000000 rows x 7 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>distance_from_home</th>\n      <th>distance_from_last_transaction</th>\n      <th>ratio_to_median_purchase_price</th>\n      <th>repeat_retailer</th>\n      <th>used_chip</th>\n      <th>used_pin_number</th>\n      <th>online_order</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.477882</td>\n      <td>-0.182849</td>\n      <td>0.043491</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>-0.241607</td>\n      <td>-0.188094</td>\n      <td>-0.189300</td>\n      <td>0.366584</td>\n      <td>-0.734443</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>-0.329369</td>\n      <td>-0.163733</td>\n      <td>-0.498812</td>\n      <td>0.366584</td>\n      <td>-0.734443</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>-0.372854</td>\n      <td>0.021806</td>\n      <td>-0.522048</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0.268572</td>\n      <td>-0.172968</td>\n      <td>0.142373</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>999995</th>\n      <td>-0.373473</td>\n      <td>-0.190529</td>\n      <td>-0.070505</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>999996</th>\n      <td>-0.103318</td>\n      <td>-0.091035</td>\n      <td>0.340808</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>-1.364425</td>\n    </tr>\n    <tr>\n      <th>999997</th>\n      <td>-0.362650</td>\n      <td>-0.137903</td>\n      <td>-0.573694</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>999998</th>\n      <td>-0.342098</td>\n      <td>-0.185523</td>\n      <td>-0.481628</td>\n      <td>0.366584</td>\n      <td>-0.734443</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n    <tr>\n      <th>999999</th>\n      <td>0.481403</td>\n      <td>-0.182579</td>\n      <td>-0.513384</td>\n      <td>0.366584</td>\n      <td>1.361576</td>\n      <td>-0.334458</td>\n      <td>0.732909</td>\n    </tr>\n  </tbody>\n</table>\n<p>1000000 rows × 7 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"y","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:24:59.864126Z","iopub.execute_input":"2024-05-03T06:24:59.864587Z","iopub.status.idle":"2024-05-03T06:24:59.879459Z","shell.execute_reply.started":"2024-05-03T06:24:59.864554Z","shell.execute_reply":"2024-05-03T06:24:59.877566Z"},"trusted":true},"execution_count":31,"outputs":[{"execution_count":31,"output_type":"execute_result","data":{"text/plain":"0         0\n1         0\n2         0\n3         0\n4         0\n         ..\n999995    0\n999996    0\n999997    0\n999998    0\n999999    0\nName: fraud, Length: 1000000, dtype: int64"},"metadata":{}}]},{"cell_type":"code","source":"{column: len(X[column].unique()) for column in X.columns}","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:25:03.974690Z","iopub.execute_input":"2024-05-03T06:25:03.975193Z","iopub.status.idle":"2024-05-03T06:25:04.259620Z","shell.execute_reply.started":"2024-05-03T06:25:03.975156Z","shell.execute_reply":"2024-05-03T06:25:04.258051Z"},"trusted":true},"execution_count":32,"outputs":[{"execution_count":32,"output_type":"execute_result","data":{"text/plain":"{'distance_from_home': 999971,\n 'distance_from_last_transaction': 999836,\n 'ratio_to_median_purchase_price': 999808,\n 'repeat_retailer': 2,\n 'used_chip': 2,\n 'used_pin_number': 2,\n 'online_order': 2}"},"metadata":{}}]},{"cell_type":"markdown","source":"# Training","metadata":{}},{"cell_type":"code","source":"X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.7, random_state=123)","metadata":{"execution":{"iopub.status.busy":"2024-05-03T05:55:50.362933Z","iopub.execute_input":"2024-05-03T05:55:50.363407Z","iopub.status.idle":"2024-05-03T05:55:50.584114Z","shell.execute_reply.started":"2024-05-03T05:55:50.363375Z","shell.execute_reply":"2024-05-03T05:55:50.582624Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"models = {\n    LogisticRegression(): \"   Logistic Regression\",\n    SVC():                \"Support Vector Machine\",\n    MLPClassifier():      \"        Neural Network\"\n  \n}\n\nfor model in models.keys():\n    model.fit(X_train, y_train)","metadata":{"execution":{"iopub.status.busy":"2024-05-03T05:56:06.250330Z","iopub.execute_input":"2024-05-03T05:56:06.250911Z","iopub.status.idle":"2024-05-03T06:11:39.900157Z","shell.execute_reply.started":"2024-05-03T05:56:06.250874Z","shell.execute_reply":"2024-05-03T06:11:39.898550Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for model, name in models.items():\n    print(name + \": {:.2f}%\".format(model.score(X_test, y_test) * 100))","metadata":{"execution":{"iopub.status.busy":"2024-05-03T06:25:10.254564Z","iopub.execute_input":"2024-05-03T06:25:10.254949Z","iopub.status.idle":"2024-05-03T06:28:23.618378Z","shell.execute_reply.started":"2024-05-03T06:25:10.254921Z","shell.execute_reply":"2024-05-03T06:28:23.616885Z"},"trusted":true},"execution_count":33,"outputs":[{"name":"stdout","text":"   Logistic Regression: 95.91%\nSupport Vector Machine: 99.77%\n        Neural Network: 99.86%\n","output_type":"stream"}]}]}