Final_Project_Update_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 831,
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:17.062338200Z",
     "start_time": "2023-08-08T18:17:17.003341200Z"
    }
   },
   "outputs": [],
   "source": [
    "#import necessary libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import plotly.express as px\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from imblearn.under_sampling import NearMiss\n",
    "from sklearn.metrics import accuracy_score, precision_score, classification_report\n",
    "from sklearn.metrics import recall_score, f1_score, average_precision_score\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.metrics import PrecisionRecallDisplay, precision_recall_curve\n",
    "\n",
    "from matplotlib import gridspec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 832,
   "outputs": [],
   "source": [
    "#load dataset\n",
    "data = pd.read_csv(\"./creditcard.csv\")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.745838400Z",
     "start_time": "2023-08-08T18:17:17.011338600Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 833,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(284807, 31)\n"
     ]
    }
   ],
   "source": [
    "# get the shape of the data frame\n",
    "print(data.shape)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.761875800Z",
     "start_time": "2023-08-08T18:17:18.747340400Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 833,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.806149300Z",
     "start_time": "2023-08-08T18:17:18.762376600Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 834,
   "outputs": [
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "data": [
        {
         "domain": {
          "x": [
           0.0,
           1.0
          ],
          "y": [
           0.0,
           1.0
          ]
         },
         "hovertemplate": "label=%{label}<br>value=%{value}<extra></extra>",
         "labels": [
          "Valid",
          "Fraud"
         ],
         "legendgroup": "",
         "name": "",
         "showlegend": true,
         "values": [
          284315,
          492
         ],
         "type": "pie"
        }
       ],
       "layout": {
        "template": {
         "data": {
          "histogram2dcontour": [
           {
            "type": "histogram2dcontour",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0.0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1.0,
              "#f0f921"
             ]
            ]
           }
          ],
          "choropleth": [
           {
            "type": "choropleth",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            }
           }
          ],
          "histogram2d": [
           {
            "type": "histogram2d",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0.0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1.0,
              "#f0f921"
             ]
            ]
           }
          ],
          "heatmap": [
           {
            "type": "heatmap",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0.0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1.0,
              "#f0f921"
             ]
            ]
           }
          ],
          "heatmapgl": [
           {
            "type": "heatmapgl",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0.0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1.0,
              "#f0f921"
             ]
            ]
           }
          ],
          "contourcarpet": [
           {
            "type": "contourcarpet",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            }
           }
          ],
          "contour": [
           {
            "type": "contour",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0.0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1.0,
              "#f0f921"
             ]
            ]
           }
          ],
          "surface": [
           {
            "type": "surface",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            },
            "colorscale": [
             [
              0.0,
              "#0d0887"
             ],
             [
              0.1111111111111111,
              "#46039f"
             ],
             [
              0.2222222222222222,
              "#7201a8"
             ],
             [
              0.3333333333333333,
              "#9c179e"
             ],
             [
              0.4444444444444444,
              "#bd3786"
             ],
             [
              0.5555555555555556,
              "#d8576b"
             ],
             [
              0.6666666666666666,
              "#ed7953"
             ],
             [
              0.7777777777777778,
              "#fb9f3a"
             ],
             [
              0.8888888888888888,
              "#fdca26"
             ],
             [
              1.0,
              "#f0f921"
             ]
            ]
           }
          ],
          "mesh3d": [
           {
            "type": "mesh3d",
            "colorbar": {
             "outlinewidth": 0,
             "ticks": ""
            }
           }
          ],
          "scatter": [
           {
            "marker": {
             "line": {
              "color": "#283442"
             }
            },
            "type": "scatter"
           }
          ],
          "parcoords": [
           {
            "type": "parcoords",
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "scatterpolargl": [
           {
            "type": "scatterpolargl",
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "bar": [
           {
            "error_x": {
             "color": "#f2f5fa"
            },
            "error_y": {
             "color": "#f2f5fa"
            },
            "marker": {
             "line": {
              "color": "rgb(17,17,17)",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "bar"
           }
          ],
          "scattergeo": [
           {
            "type": "scattergeo",
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "scatterpolar": [
           {
            "type": "scatterpolar",
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "histogram": [
           {
            "marker": {
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "histogram"
           }
          ],
          "scattergl": [
           {
            "marker": {
             "line": {
              "color": "#283442"
             }
            },
            "type": "scattergl"
           }
          ],
          "scatter3d": [
           {
            "type": "scatter3d",
            "line": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            },
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "scattermapbox": [
           {
            "type": "scattermapbox",
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "scatterternary": [
           {
            "type": "scatterternary",
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "scattercarpet": [
           {
            "type": "scattercarpet",
            "marker": {
             "colorbar": {
              "outlinewidth": 0,
              "ticks": ""
             }
            }
           }
          ],
          "carpet": [
           {
            "aaxis": {
             "endlinecolor": "#A2B1C6",
             "gridcolor": "#506784",
             "linecolor": "#506784",
             "minorgridcolor": "#506784",
             "startlinecolor": "#A2B1C6"
            },
            "baxis": {
             "endlinecolor": "#A2B1C6",
             "gridcolor": "#506784",
             "linecolor": "#506784",
             "minorgridcolor": "#506784",
             "startlinecolor": "#A2B1C6"
            },
            "type": "carpet"
           }
          ],
          "table": [
           {
            "cells": {
             "fill": {
              "color": "#506784"
             },
             "line": {
              "color": "rgb(17,17,17)"
             }
            },
            "header": {
             "fill": {
              "color": "#2a3f5f"
             },
             "line": {
              "color": "rgb(17,17,17)"
             }
            },
            "type": "table"
           }
          ],
          "barpolar": [
           {
            "marker": {
             "line": {
              "color": "rgb(17,17,17)",
              "width": 0.5
             },
             "pattern": {
              "fillmode": "overlay",
              "size": 10,
              "solidity": 0.2
             }
            },
            "type": "barpolar"
           }
          ],
          "pie": [
           {
            "automargin": true,
            "type": "pie"
           }
          ]
         },
         "layout": {
          "autotypenumbers": "strict",
          "colorway": [
           "#636efa",
           "#EF553B",
           "#00cc96",
           "#ab63fa",
           "#FFA15A",
           "#19d3f3",
           "#FF6692",
           "#B6E880",
           "#FF97FF",
           "#FECB52"
          ],
          "font": {
           "color": "#f2f5fa"
          },
          "hovermode": "closest",
          "hoverlabel": {
           "align": "left"
          },
          "paper_bgcolor": "rgb(17,17,17)",
          "plot_bgcolor": "rgb(17,17,17)",
          "polar": {
           "bgcolor": "rgb(17,17,17)",
           "angularaxis": {
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "ticks": ""
           },
           "radialaxis": {
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "ticks": ""
           }
          },
          "ternary": {
           "bgcolor": "rgb(17,17,17)",
           "aaxis": {
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "ticks": ""
           },
           "baxis": {
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "ticks": ""
           },
           "caxis": {
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "ticks": ""
           }
          },
          "coloraxis": {
           "colorbar": {
            "outlinewidth": 0,
            "ticks": ""
           }
          },
          "colorscale": {
           "sequential": [
            [
             0.0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1.0,
             "#f0f921"
            ]
           ],
           "sequentialminus": [
            [
             0.0,
             "#0d0887"
            ],
            [
             0.1111111111111111,
             "#46039f"
            ],
            [
             0.2222222222222222,
             "#7201a8"
            ],
            [
             0.3333333333333333,
             "#9c179e"
            ],
            [
             0.4444444444444444,
             "#bd3786"
            ],
            [
             0.5555555555555556,
             "#d8576b"
            ],
            [
             0.6666666666666666,
             "#ed7953"
            ],
            [
             0.7777777777777778,
             "#fb9f3a"
            ],
            [
             0.8888888888888888,
             "#fdca26"
            ],
            [
             1.0,
             "#f0f921"
            ]
           ],
           "diverging": [
            [
             0,
             "#8e0152"
            ],
            [
             0.1,
             "#c51b7d"
            ],
            [
             0.2,
             "#de77ae"
            ],
            [
             0.3,
             "#f1b6da"
            ],
            [
             0.4,
             "#fde0ef"
            ],
            [
             0.5,
             "#f7f7f7"
            ],
            [
             0.6,
             "#e6f5d0"
            ],
            [
             0.7,
             "#b8e186"
            ],
            [
             0.8,
             "#7fbc41"
            ],
            [
             0.9,
             "#4d9221"
            ],
            [
             1,
             "#276419"
            ]
           ]
          },
          "xaxis": {
           "gridcolor": "#283442",
           "linecolor": "#506784",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "#283442",
           "automargin": true,
           "zerolinewidth": 2
          },
          "yaxis": {
           "gridcolor": "#283442",
           "linecolor": "#506784",
           "ticks": "",
           "title": {
            "standoff": 15
           },
           "zerolinecolor": "#283442",
           "automargin": true,
           "zerolinewidth": 2
          },
          "scene": {
           "xaxis": {
            "backgroundcolor": "rgb(17,17,17)",
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "#C8D4E3",
            "gridwidth": 2
           },
           "yaxis": {
            "backgroundcolor": "rgb(17,17,17)",
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "#C8D4E3",
            "gridwidth": 2
           },
           "zaxis": {
            "backgroundcolor": "rgb(17,17,17)",
            "gridcolor": "#506784",
            "linecolor": "#506784",
            "showbackground": true,
            "ticks": "",
            "zerolinecolor": "#C8D4E3",
            "gridwidth": 2
           }
          },
          "shapedefaults": {
           "line": {
            "color": "#f2f5fa"
           }
          },
          "annotationdefaults": {
           "arrowcolor": "#f2f5fa",
           "arrowhead": 0,
           "arrowwidth": 1
          },
          "geo": {
           "bgcolor": "rgb(17,17,17)",
           "landcolor": "rgb(17,17,17)",
           "subunitcolor": "#506784",
           "showland": true,
           "showlakes": true,
           "lakecolor": "rgb(17,17,17)"
          },
          "title": {
           "x": 0.05
          },
          "updatemenudefaults": {
           "bgcolor": "#506784",
           "borderwidth": 0
          },
          "sliderdefaults": {
           "bgcolor": "#C8D4E3",
           "borderwidth": 1,
           "bordercolor": "rgb(17,17,17)",
           "tickwidth": 0
          },
          "mapbox": {
           "style": "dark"
          }
         }
        },
        "legend": {
         "tracegroupgap": 0
        },
        "title": {
         "text": "Fraud vs Valid transactions"
        },
        "piecolorway": [
         "orange",
         "blue"
        ],
        "height": 400,
        "width": 600
       },
       "config": {
        "plotlyServerURL": "https://plot.ly"
       }
      },
      "text/html": "<div>                            <div id=\"87752ab8-a058-4564-81ec-d4cf91591609\" class=\"plotly-graph-div\" style=\"height:400px; width:600px;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"87752ab8-a058-4564-81ec-d4cf91591609\")) {                    Plotly.newPlot(                        \"87752ab8-a058-4564-81ec-d4cf91591609\",                        [{\"domain\":{\"x\":[0.0,1.0],\"y\":[0.0,1.0]},\"hovertemplate\":\"label=%{label}\\u003cbr\\u003evalue=%{value}\\u003cextra\\u003e\\u003c\\u002fextra\\u003e\",\"labels\":[\"Valid\",\"Fraud\"],\"legendgroup\":\"\",\"name\":\"\",\"showlegend\":true,\"values\":[284315,492],\"type\":\"pie\"}],                        {\"template\":{\"data\":{\"histogram2dcontour\":[{\"type\":\"histogram2dcontour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"choropleth\":[{\"type\":\"choropleth\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"histogram2d\":[{\"type\":\"histogram2d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmap\":[{\"type\":\"heatmap\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"heatmapgl\":[{\"type\":\"heatmapgl\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"contourcarpet\":[{\"type\":\"contourcarpet\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"contour\":[{\"type\":\"contour\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"surface\":[{\"type\":\"surface\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]}],\"mesh3d\":[{\"type\":\"mesh3d\",\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}],\"scatter\":[{\"marker\":{\"line\":{\"color\":\"#283442\"}},\"type\":\"scatter\"}],\"parcoords\":[{\"type\":\"parcoords\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolargl\":[{\"type\":\"scatterpolargl\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"bar\":[{\"error_x\":{\"color\":\"#f2f5fa\"},\"error_y\":{\"color\":\"#f2f5fa\"},\"marker\":{\"line\":{\"color\":\"rgb(17,17,17)\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"scattergeo\":[{\"type\":\"scattergeo\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterpolar\":[{\"type\":\"scatterpolar\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"scattergl\":[{\"marker\":{\"line\":{\"color\":\"#283442\"}},\"type\":\"scattergl\"}],\"scatter3d\":[{\"type\":\"scatter3d\",\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattermapbox\":[{\"type\":\"scattermapbox\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scatterternary\":[{\"type\":\"scatterternary\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"scattercarpet\":[{\"type\":\"scattercarpet\",\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}}}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#A2B1C6\",\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"minorgridcolor\":\"#506784\",\"startlinecolor\":\"#A2B1C6\"},\"baxis\":{\"endlinecolor\":\"#A2B1C6\",\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"minorgridcolor\":\"#506784\",\"startlinecolor\":\"#A2B1C6\"},\"type\":\"carpet\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#506784\"},\"line\":{\"color\":\"rgb(17,17,17)\"}},\"header\":{\"fill\":{\"color\":\"#2a3f5f\"},\"line\":{\"color\":\"rgb(17,17,17)\"}},\"type\":\"table\"}],\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"rgb(17,17,17)\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}]},\"layout\":{\"autotypenumbers\":\"strict\",\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#f2f5fa\"},\"hovermode\":\"closest\",\"hoverlabel\":{\"align\":\"left\"},\"paper_bgcolor\":\"rgb(17,17,17)\",\"plot_bgcolor\":\"rgb(17,17,17)\",\"polar\":{\"bgcolor\":\"rgb(17,17,17)\",\"angularaxis\":{\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"ticks\":\"\"},\"radialaxis\":{\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"ticks\":\"\"}},\"ternary\":{\"bgcolor\":\"rgb(17,17,17)\",\"aaxis\":{\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"ticks\":\"\"},\"caxis\":{\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"ticks\":\"\"}},\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]]},\"xaxis\":{\"gridcolor\":\"#283442\",\"linecolor\":\"#506784\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"#283442\",\"automargin\":true,\"zerolinewidth\":2},\"yaxis\":{\"gridcolor\":\"#283442\",\"linecolor\":\"#506784\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"#283442\",\"automargin\":true,\"zerolinewidth\":2},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"rgb(17,17,17)\",\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"#C8D4E3\",\"gridwidth\":2},\"yaxis\":{\"backgroundcolor\":\"rgb(17,17,17)\",\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"#C8D4E3\",\"gridwidth\":2},\"zaxis\":{\"backgroundcolor\":\"rgb(17,17,17)\",\"gridcolor\":\"#506784\",\"linecolor\":\"#506784\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"#C8D4E3\",\"gridwidth\":2}},\"shapedefaults\":{\"line\":{\"color\":\"#f2f5fa\"}},\"annotationdefaults\":{\"arrowcolor\":\"#f2f5fa\",\"arrowhead\":0,\"arrowwidth\":1},\"geo\":{\"bgcolor\":\"rgb(17,17,17)\",\"landcolor\":\"rgb(17,17,17)\",\"subunitcolor\":\"#506784\",\"showland\":true,\"showlakes\":true,\"lakecolor\":\"rgb(17,17,17)\"},\"title\":{\"x\":0.05},\"updatemenudefaults\":{\"bgcolor\":\"#506784\",\"borderwidth\":0},\"sliderdefaults\":{\"bgcolor\":\"#C8D4E3\",\"borderwidth\":1,\"bordercolor\":\"rgb(17,17,17)\",\"tickwidth\":0},\"mapbox\":{\"style\":\"dark\"}}},\"legend\":{\"tracegroupgap\":0},\"title\":{\"text\":\"Fraud vs Valid transactions\"},\"piecolorway\":[\"orange\",\"blue\"],\"height\":400,\"width\":600},                        {\"responsive\": true}                    ).then(function(){\n                            \nvar gd = document.getElementById('87752ab8-a058-4564-81ec-d4cf91591609');\nvar x = new MutationObserver(function (mutations, observer) {{\n        var display = window.getComputedStyle(gd).display;\n        if (!display || display === 'none') {{\n            console.log([gd, 'removed!']);\n            Plotly.purge(gd);\n            observer.disconnect();\n        }}\n}});\n\n// Listen for the removal of the full notebook cells\nvar notebookContainer = gd.closest('#notebook-container');\nif (notebookContainer) {{\n    x.observe(notebookContainer, {childList: true});\n}}\n\n// Listen for the clearing of the current output cell\nvar outputEl = gd.closest('.output');\nif (outputEl) {{\n    x.observe(outputEl, {childList: true});\n}}\n\n                        })                };                });            </script>        </div>"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create labels\n",
    "labels=[\"Valid\",\"Fraud\"]\n",
    "\n",
    "# get count list\n",
    "li = data[\"Class\"].value_counts().tolist()\n",
    "values = [li[0], li[1]]\n",
    "\n",
    "# plot data\n",
    "fig = px.pie(values=data['Class'].value_counts(),names=labels ,width=600,\n",
    "             height=400,color_discrete_sequence=[\"orange\",\"blue\"],\n",
    "             title=\"Fraud vs Valid transactions\")\n",
    "fig.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.823275500Z",
     "start_time": "2023-08-08T18:17:18.779651300Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 835,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fraud Cases: 492\n",
      "Valid Transactions: 284315\n"
     ]
    }
   ],
   "source": [
    "# data has a class imbalance since there are\n",
    "# many more valid than fraudulent transactions\n",
    "fraud = data[data['Class'] == 1]\n",
    "valid = data[data['Class'] == 0]\n",
    "\n",
    "print(f\"Fraud Cases: {len(data[data['Class'] == 1])}\")\n",
    "print(f\"Valid Transactions: {len(data[data['Class'] == 0])}\")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.888775600Z",
     "start_time": "2023-08-08T18:17:18.824276700Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 836,
   "outputs": [
    {
     "data": {
      "text/plain": "count     492.000000\nmean      122.211321\nstd       256.683288\nmin         0.000000\n25%         1.000000\n50%         9.250000\n75%       105.890000\nmax      2125.870000\nName: Amount, dtype: float64"
     },
     "execution_count": 836,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#the amount details for fraudulent transaction\n",
    "fraud.Amount.describe()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.900276500Z",
     "start_time": "2023-08-08T18:17:18.886276Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 837,
   "outputs": [
    {
     "data": {
      "text/plain": "count    284315.000000\nmean         88.291022\nstd         250.105092\nmin           0.000000\n25%           5.650000\n50%          22.000000\n75%          77.050000\nmax       25691.160000\nName: Amount, dtype: float64"
     },
     "execution_count": 837,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#the amount details for normal transaction\n",
    "valid.Amount.describe()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:18.945276100Z",
     "start_time": "2023-08-08T18:17:18.901276400Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 838,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 864x648 with 2 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the correlation matrix\n",
    "corr = data.corr()\n",
    "# could use abs value for correlation as we can treat negatively and positively correlated as both being important\n",
    "#corr = data.corr().abs()\n",
    "fig = plt.figure(figsize = (12, 9))\n",
    "sns.color_palette(\"mako\", as_cmap=True)\n",
    "sns.heatmap(corr, cmap=\"mako\", square = True)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:19.814776Z",
     "start_time": "2023-08-08T18:17:18.917275800Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 839,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "V4     0.133447\n",
      "V11    0.154876\n",
      "dtype: float64\n",
      "[0.9999999999999998, 0.15487564474394697, 0.13344748623900424, -0.10134729859508473, -0.11148525388904085, -0.18725659151429688, -0.19296082706741338, -0.19653894030401806, -0.21688294364102748, -0.2605929248772168, -0.3025436958044035, -0.32648106724371373] [0.9999999999999998, 0.15487564474394697, 0.13344748623900424, -0.10134729859508473, -0.11148525388904085, -0.18725659151429688, -0.19296082706741338, -0.19653894030401806, -0.21688294364102748, -0.2605929248772168, -0.3025436958044035, -0.32648106724371373]\n"
     ]
    }
   ],
   "source": [
    "colors = ['#C4C7DF','#4C4C4C', \"blue\"]\n",
    "mat = corr.unstack()\n",
    "mat = mat[mat.between(0.1, 1, inclusive='neither')]\n",
    "\n",
    "print(mat['Class'])\n",
    "corr = data.corrwith(data['Class']).sort_values(ascending = False)\n",
    "\n",
    "corr.columns = ['Correlation']\n",
    "res = []\n",
    "for i in range(len(corr)):\n",
    "    if corr.iloc[i] > 0.1 or corr.iloc[i] < -0.1:\n",
    "        res.append(corr.iloc[i])\n",
    "\n",
    "\n",
    "\n",
    "corr = [x for x in corr if x > 0.1 or x < -0.1]\n",
    "print(corr, res)\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:19.954167100Z",
     "start_time": "2023-08-08T18:17:19.819275400Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 840,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(284807, 30)\n",
      "(284807,)\n"
     ]
    }
   ],
   "source": [
    "#separating the X and the Y values\n",
    "x = data.drop(['Class'], axis = 1)\n",
    "\n",
    "# only use the features with an abs value correlation above a certain threshold, we have to decide what matters and doesn't even a weak correlation could be useful\n",
    "#X = data[['V3','V4','V7','V10','V11','V12','V14','V16','V17']].copy(deep = True)\n",
    "y = data[\"Class\"]\n",
    "print(x.shape)\n",
    "print(y.shape)\n",
    "# getting just the values for the sake of processing\n",
    "# (its a numpy array with no columns)\n",
    "x_data = x.values\n",
    "y_data = y.values"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:20.006168200Z",
     "start_time": "2023-08-08T18:17:19.955168400Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 841,
   "outputs": [],
   "source": [
    "#training and testing data bifurcation\n",
    "\n",
    "\n",
    "#split the data into training and testing sets\n",
    "# for later, don't run SMOTE on test data, only training.\n",
    "x_train, x_test, y_train, y_test = train_test_split(x_data, y_data, test_size = 0.2, random_state = 42)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:20.109168400Z",
     "start_time": "2023-08-08T18:17:19.986168900Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 842,
   "outputs": [],
   "source": [
    "# store overall results here\n",
    "final_results = {}"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:20.126420300Z",
     "start_time": "2023-08-08T18:17:20.110668300Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 843,
   "outputs": [],
   "source": [
    "# baseline models\n",
    "# FYI, this cell might take a bit to finish loading,\n",
    "# import classifier\n",
    "\n",
    "\n",
    "# create the classifier object, the more estimators\n",
    "# the longer it will take to run, diminishing returns at a point\n",
    "# later we will more finely tune the model's parameters along with\n",
    "rfc = RandomForestClassifier(n_estimators=10, max_depth=13, min_samples_split=9, random_state=42)\n",
    "\n",
    "# train the data\n",
    "rfc.fit(x_train, y_train)\n",
    "\n",
    "# get predictions\n",
    "y_pred = rfc.predict(x_test)\n",
    "\n",
    "\n",
    "\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:47.899560700Z",
     "start_time": "2023-08-08T18:17:20.126920700Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 844,
   "outputs": [],
   "source": [
    "# base line xgb\n",
    "# train xgb model\n",
    "xgb = XGBClassifier(n_estimators=10, max_delta_step=1, eval_metric='aucpr', random_state=42)\n",
    "\n",
    "# train the data\n",
    "xgb.fit(x_train, y_train)\n",
    "\n",
    "# get predictions\n",
    "y_pred_xgb = xgb.predict(x_test)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:49.729029600Z",
     "start_time": "2023-08-08T18:17:47.901534500Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 845,
   "outputs": [],
   "source": [
    "# base line results\n",
    "results = {\n",
    "    \"Accuracy\": accuracy_score(y_test, y_pred),\n",
    "    \"Precision\": precision_score(y_test, y_pred),\n",
    "    \"Recall\": recall_score(y_test, y_pred),\n",
    "    \"F1 Score\": f1_score(y_test, y_pred),\n",
    "    \"AUPRC\": average_precision_score(y_test, y_pred),\n",
    "}\n",
    "\n",
    "final_results[\"RFC_BASELINE\"] = results.copy()\n",
    "\n",
    "\n",
    "# solve for accuracy, precision, recall, f1 score and auprc\n",
    "results_xgb = {\n",
    "    \"Accuracy\": accuracy_score(y_test, y_pred_xgb),\n",
    "    \"Precision\": precision_score(y_test, y_pred_xgb),\n",
    "    \"Recall\": recall_score(y_test, y_pred_xgb),\n",
    "    \"F1 Score\": f1_score(y_test, y_pred_xgb),\n",
    "    \"AUPRC\": average_precision_score(y_test, y_pred_xgb),\n",
    "}\n",
    "\n",
    "final_results[\"XGB_BASELINE\"] = results_xgb.copy()\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:49.821030500Z",
     "start_time": "2023-08-08T18:17:49.732030600Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 846,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "________________________________\n",
      "Random Forest Classifier Results\n",
      "________________________________\n",
      "Accuracy: 0.9995435553526912\n",
      "Precision: 0.95\n",
      "Recall: 0.7755102040816326\n",
      "F1 Score: 0.8539325842696629\n",
      "AUPRC: 0.7371209162714276\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 360x216 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# show results\n",
    "print(\"________________________________\")\n",
    "print(\"Random Forest Classifier Results\")\n",
    "print(\"________________________________\")\n",
    "for [k, v] in results.items():\n",
    "    print(f\"{k}: {v}\")\n",
    "\n",
    "vals = results.values()\n",
    "names = results.keys()\n",
    "\n",
    "fig = plt.figure(figsize = (5, 3))\n",
    "plt.bar(names, vals)\n",
    "plt.show()\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:49.899531Z",
     "start_time": "2023-08-08T18:17:49.823531Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 847,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "precision, recall, _ = precision_recall_curve(y_test, y_pred)\n",
    "disp = PrecisionRecallDisplay(precision=precision, recall=recall)\n",
    "disp.plot()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:50.007531900Z",
     "start_time": "2023-08-08T18:17:49.901034500Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 848,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "________________________________\n",
      "XGB Classifier Results\n",
      "________________________________\n",
      "Accuracy: 0.9996137776061234\n",
      "Precision: 0.9634146341463414\n",
      "Recall: 0.8061224489795918\n",
      "F1 Score: 0.8777777777777778\n",
      "AUPRC: 0.7769637199646289\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 360x216 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"________________________________\")\n",
    "print(\"XGB Classifier Results\")\n",
    "print(\"________________________________\")\n",
    "for [k, v] in results_xgb.items():\n",
    "    print(f\"{k}: {v}\")\n",
    "\n",
    "vals = results_xgb.values()\n",
    "names = results_xgb.keys()\n",
    "\n",
    "fig = plt.figure(figsize = (5, 3))\n",
    "plt.bar(names, vals)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:50.100249400Z",
     "start_time": "2023-08-08T18:17:50.009531Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 849,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "precision, recall, _ = precision_recall_curve(y_test, y_pred_xgb)\n",
    "disp = PrecisionRecallDisplay(precision=precision, recall=recall)\n",
    "disp.plot()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:17:50.208343900Z",
     "start_time": "2023-08-08T18:17:50.101750200Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 850,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--RFC + SMOTE--\n",
      "Fold: 1 | Valid: 284315 | Fraud: 284315\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     56863\n",
      "           1       0.89      0.77      0.83        99\n",
      "\n",
      "    accuracy                           1.00     56962\n",
      "   macro avg       0.95      0.88      0.91     56962\n",
      "weighted avg       1.00      1.00      1.00     56962\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9994382219725431\n",
      "Precision: 0.8941176470588236\n",
      "Recall: 0.7676767676767676\n",
      "F1 Score: 0.826086956521739\n",
      "AUPRC: 0.6867971231741092\n",
      "________________________________\n",
      "Fold: 2 | Valid: 284315 | Fraud: 284315\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     56863\n",
      "           1       0.96      0.75      0.84        99\n",
      "\n",
      "    accuracy                           1.00     56962\n",
      "   macro avg       0.98      0.87      0.92     56962\n",
      "weighted avg       1.00      1.00      1.00     56962\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9995084442259752\n",
      "Precision: 0.961038961038961\n",
      "Recall: 0.7474747474747475\n",
      "F1 Score: 0.8409090909090909\n",
      "AUPRC: 0.7187912437999419\n",
      "________________________________\n",
      "Fold: 3 | Valid: 284315 | Fraud: 284315\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     56863\n",
      "           1       0.96      0.78      0.86        98\n",
      "\n",
      "    accuracy                           1.00     56961\n",
      "   macro avg       0.98      0.89      0.93     56961\n",
      "weighted avg       1.00      1.00      1.00     56961\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9995611032109689\n",
      "Precision: 0.9620253164556962\n",
      "Recall: 0.7755102040816326\n",
      "F1 Score: 0.8587570621468926\n",
      "AUPRC: 0.7464466786706015\n",
      "________________________________\n",
      "Fold: 4 | Valid: 284315 | Fraud: 284315\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     56863\n",
      "           1       0.95      0.77      0.85        98\n",
      "\n",
      "    accuracy                           1.00     56961\n",
      "   macro avg       0.97      0.88      0.92     56961\n",
      "weighted avg       1.00      1.00      1.00     56961\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9995259914678464\n",
      "Precision: 0.9493670886075949\n",
      "Recall: 0.7653061224489796\n",
      "F1 Score: 0.8474576271186441\n",
      "AUPRC: 0.7269602304088639\n",
      "________________________________\n",
      "Fold: 5 | Valid: 284315 | Fraud: 284315\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     56863\n",
      "           1       0.94      0.78      0.85        98\n",
      "\n",
      "    accuracy                           1.00     56961\n",
      "   macro avg       0.97      0.89      0.92     56961\n",
      "weighted avg       1.00      1.00      1.00     56961\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9995259914678464\n",
      "Precision: 0.9382716049382716\n",
      "Recall: 0.7755102040816326\n",
      "F1 Score: 0.8491620111731844\n",
      "AUPRC: 0.7280254330040273\n",
      "________________________________\n",
      "\n",
      "--Final Results--\n",
      "Avg Accuracy: 0.999511950469036\n",
      "Avg Precision: 0.9409641236198694\n",
      "Avg Recall: 0.766295609152752\n",
      "Avg F1 Score: 0.8444745495739102\n",
      "Avg AUPRC: 0.7214041418115088\n"
     ]
    }
   ],
   "source": [
    "# smote only random forest classifier\n",
    "print(\"--RFC + SMOTE--\")\n",
    "# separate data\n",
    "x = data.drop('Class', axis=1)\n",
    "y = data['Class']\n",
    "\n",
    "# using stratified kfold for imbalanced dataset to validate\n",
    "sfk = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
    "\n",
    "fold = 1\n",
    "\n",
    "# used to track results after all k folds validations\n",
    "res_acc = []\n",
    "res_prec = []\n",
    "res_rec = []\n",
    "res_f1 = []\n",
    "res_auprc = []\n",
    "\n",
    "\n",
    "# loop over every fold\n",
    "for train_idx, test_idx in sfk.split(x, y):\n",
    "\n",
    "    x = data.drop('Class', axis=1)\n",
    "    y = data['Class']\n",
    "\n",
    "    # use smote to oversample data\n",
    "    smt = SMOTE(random_state=13)\n",
    "    x, y = smt.fit_resample(x, y)\n",
    "    \n",
    "    print(f\"Fold: {fold} | Valid: {np.bincount(y)[0]} | Fraud: {np.bincount(y)[1]}\")\n",
    "    \n",
    "    # get the training data\n",
    "    x_train, x_test = x.loc[train_idx], x.loc[test_idx]\n",
    "    y_train, y_test = y[train_idx], y[test_idx] \n",
    "    \n",
    "    # create the forest classifier model\n",
    "    rfc = RandomForestClassifier(n_estimators=10, max_depth=10, min_samples_split=6, random_state=13)\n",
    "    \n",
    "    # train the model on the data\n",
    "    rfc.fit(x_train, y_train)\n",
    "    \n",
    "    # get predictions\n",
    "    y_pred = rfc.predict(x_test)\n",
    "    \n",
    "    print(classification_report(y_test, y_pred))\n",
    "    \n",
    "    # solve for accuracy, precision, recall, f1 score and auprc and store for avg\n",
    "    res_acc += [accuracy_score(y_test, y_pred)]\n",
    "    res_prec += [precision_score(y_test, y_pred)]\n",
    "    res_rec += [recall_score(y_test, y_pred)]\n",
    "    res_f1 += [f1_score(y_test, y_pred)]\n",
    "    res_auprc += [average_precision_score(y_test, y_pred)]\n",
    "    \n",
    "    # print current results\n",
    "    results = {\n",
    "        \"Accuracy\": res_acc[-1],\n",
    "        \"Precision\": res_prec[-1],\n",
    "        \"Recall\": res_rec[-1],\n",
    "        \"F1 Score\": res_f1[-1],\n",
    "        \"AUPRC\": res_auprc[-1],\n",
    "    }\n",
    "    \n",
    "    print(\"________________________________\")\n",
    "    for [k, v] in results.items():\n",
    "        print(f\"{k}: {v}\")\n",
    "    print(\"________________________________\")\n",
    "    #precision, recall, _ = precision_recall_curve(y_test, y_pred)\n",
    "    #disp = PrecisionRecallDisplay(precision=precision, recall=recall)\n",
    "    #disp.plot()\n",
    "    #plt.show()\n",
    "    \n",
    "    \n",
    "    fold += 1\n",
    "\n",
    "# show avg results after k folds validation\n",
    "print(\"\\n--Final Results--\")\n",
    "print(f\"Avg Accuracy: {np.mean(res_acc)}\")\n",
    "print(f\"Avg Precision: {np.mean(res_prec)}\")\n",
    "print(f\"Avg Recall: {np.mean(res_rec)}\")\n",
    "print(f\"Avg F1 Score: {np.mean(res_f1)}\")\n",
    "print(f\"Avg AUPRC: {np.mean(res_auprc)}\")   \n",
    "\n",
    "final_results[\"RFC_SMOTE\"] = {\n",
    "    \"Accuracy\": np.mean(res_acc),\n",
    "    \"Precision\": np.mean(res_prec),\n",
    "    \"Recall\": np.mean(res_rec),\n",
    "    \"F1 Score\": np.mean(res_f1),\n",
    "    \"AUPRC\": np.mean(res_auprc),\n",
    "}\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:19:45.232884700Z",
     "start_time": "2023-08-08T18:17:50.214312200Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 851,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--RFC + NearMiss--\n",
      "Fold: 1 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.99      0.97       197\n",
      "           1       0.98      0.90      0.94        99\n",
      "\n",
      "    accuracy                           0.96       296\n",
      "   macro avg       0.96      0.94      0.95       296\n",
      "weighted avg       0.96      0.96      0.96       296\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9594594594594594\n",
      "Precision: 0.978021978021978\n",
      "Recall: 0.898989898989899\n",
      "F1 Score: 0.9368421052631579\n",
      "AUPRC: 0.913015663015663\n",
      "________________________________\n",
      "Fold: 2 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.99      0.97       197\n",
      "           1       0.99      0.90      0.94        98\n",
      "\n",
      "    accuracy                           0.96       295\n",
      "   macro avg       0.97      0.95      0.96       295\n",
      "weighted avg       0.96      0.96      0.96       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9627118644067797\n",
      "Precision: 0.9887640449438202\n",
      "Recall: 0.8979591836734694\n",
      "F1 Score: 0.9411764705882354\n",
      "AUPRC: 0.9217680597281762\n",
      "________________________________\n",
      "Fold: 3 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.99      0.98       197\n",
      "           1       0.98      0.95      0.96        98\n",
      "\n",
      "    accuracy                           0.98       295\n",
      "   macro avg       0.98      0.97      0.97       295\n",
      "weighted avg       0.98      0.98      0.98       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.976271186440678\n",
      "Precision: 0.9789473684210527\n",
      "Recall: 0.9489795918367347\n",
      "F1 Score: 0.9637305699481866\n",
      "AUPRC: 0.945950226656229\n",
      "________________________________\n",
      "Fold: 4 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      1.00      0.98       197\n",
      "           1       1.00      0.90      0.95        98\n",
      "\n",
      "    accuracy                           0.97       295\n",
      "   macro avg       0.98      0.95      0.96       295\n",
      "weighted avg       0.97      0.97      0.97       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9661016949152542\n",
      "Precision: 1.0\n",
      "Recall: 0.8979591836734694\n",
      "F1 Score: 0.9462365591397849\n",
      "AUPRC: 0.9318574887582152\n",
      "________________________________\n",
      "Fold: 5 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.99      0.98       196\n",
      "           1       0.99      0.93      0.96        99\n",
      "\n",
      "    accuracy                           0.97       295\n",
      "   macro avg       0.98      0.96      0.97       295\n",
      "weighted avg       0.97      0.97      0.97       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9728813559322034\n",
      "Precision: 0.989247311827957\n",
      "Recall: 0.9292929292929293\n",
      "F1 Score: 0.9583333333333333\n",
      "AUPRC: 0.94302934576308\n",
      "________________________________\n",
      "\n",
      "--Final Results--\n",
      "Avg Accuracy: 0.967485112230875\n",
      "Avg Precision: 0.9869961406429615\n",
      "Avg Recall: 0.9146361574933003\n",
      "Avg F1 Score: 0.9492638076545395\n",
      "Avg AUPRC: 0.9311241567842726\n"
     ]
    }
   ],
   "source": [
    "print(\"--RFC + NearMiss--\")\n",
    "\n",
    "# separate data\n",
    "x = data.drop('Class', axis=1)\n",
    "y = data['Class']\n",
    "\n",
    "# under sample with near miss  \n",
    "nm = NearMiss(sampling_strategy=0.5)\n",
    "x, y = nm.fit_resample(x, y)\n",
    "\n",
    "# using stratified kfold for imbalanced dataset to validate\n",
    "sfk = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
    "\n",
    "fold = 1\n",
    "\n",
    "# used to track results after all k folds validations\n",
    "res_acc = []\n",
    "res_prec = []\n",
    "res_rec = []\n",
    "res_f1 = []\n",
    "res_auprc = []\n",
    "\n",
    "\n",
    "# loop over every fold\n",
    "for train_idx, test_idx in sfk.split(x, y):\n",
    "    x = data.drop('Class', axis=1)\n",
    "    y = data['Class']\n",
    "\n",
    "    # under sample with near miss  \n",
    "    nm = NearMiss(sampling_strategy=0.5)\n",
    "    x, y = nm.fit_resample(x, y)\n",
    "    \n",
    "    print(f\"Fold: {fold} | Valid: {np.bincount(y)[0]} | Fraud: {np.bincount(y)[1]}\")\n",
    "    \n",
    "    # get the training data\n",
    "    x_train, x_test = x.loc[train_idx], x.loc[test_idx]\n",
    "    y_train, y_test = y[train_idx], y[test_idx] \n",
    "\n",
    "    # create the forest classifier model\n",
    "    rfc = RandomForestClassifier(n_estimators=10, max_depth=10, min_samples_split=9, random_state=13)\n",
    "    \n",
    "    # train the model on the data\n",
    "    rfc.fit(x_train, y_train)\n",
    "    \n",
    "    # get predictions\n",
    "    y_pred = rfc.predict(x_test)\n",
    "    print(classification_report(y_test, y_pred))\n",
    "\n",
    "    # solve for accuracy, precision, recall, f1 score and auprc and store for avg\n",
    "    res_acc += [accuracy_score(y_test, y_pred)]\n",
    "    res_prec += [precision_score(y_test, y_pred)]\n",
    "    res_rec += [recall_score(y_test, y_pred)]\n",
    "    res_f1 += [f1_score(y_test, y_pred)]\n",
    "    res_auprc += [average_precision_score(y_test, y_pred)]\n",
    "    \n",
    "    # print current results\n",
    "    results = {\n",
    "        \"Accuracy\": res_acc[-1],\n",
    "        \"Precision\": res_prec[-1],\n",
    "        \"Recall\": res_rec[-1],\n",
    "        \"F1 Score\": res_f1[-1],\n",
    "        \"AUPRC\": res_auprc[-1],\n",
    "    }\n",
    "    \n",
    "    print(\"________________________________\")\n",
    "    for [k, v] in results.items():\n",
    "        print(f\"{k}: {v}\")\n",
    "    print(\"________________________________\")\n",
    "    fold += 1\n",
    "\n",
    "# show avg results after k folds validation\n",
    "print(\"\\n--Final Results--\")\n",
    "print(f\"Avg Accuracy: {np.mean(res_acc)}\")\n",
    "print(f\"Avg Precision: {np.mean(res_prec)}\")\n",
    "print(f\"Avg Recall: {np.mean(res_rec)}\")\n",
    "print(f\"Avg F1 Score: {np.mean(res_f1)}\")\n",
    "print(f\"Avg AUPRC: {np.mean(res_auprc)}\")\n",
    "\n",
    "\n",
    "final_results[\"RFC_NearMiss\"] = {\n",
    "    \"Accuracy\": np.mean(res_acc),\n",
    "    \"Precision\": np.mean(res_prec),\n",
    "    \"Recall\": np.mean(res_rec),\n",
    "    \"F1 Score\": np.mean(res_f1),\n",
    "    \"AUPRC\": np.mean(res_auprc),\n",
    "}\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:19:48.398383600Z",
     "start_time": "2023-08-08T18:19:45.239385Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 852,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--RFC + SMOTE + NearMiss--\n",
      "Fold: 1 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99     45491\n",
      "           1       1.00      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68236\n",
      "   macro avg       0.99      0.99      0.99     68236\n",
      "weighted avg       0.99      0.99      0.99     68236\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9899759657658714\n",
      "Precision: 0.9981483990425868\n",
      "Recall: 0.9717300505605627\n",
      "F1 Score: 0.9847620744965248\n",
      "AUPRC: 0.979353972649511\n",
      "________________________________\n",
      "Fold: 2 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.98      1.00      0.99     45491\n",
      "           1       1.00      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68236\n",
      "   macro avg       0.99      0.98      0.99     68236\n",
      "weighted avg       0.99      0.99      0.99     68236\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9891992496629345\n",
      "Precision: 0.9985050285403643\n",
      "Recall: 0.9690481424488898\n",
      "F1 Score: 0.9835560811263081\n",
      "AUPRC: 0.9779165777832469\n",
      "________________________________\n",
      "Fold: 3 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99     45490\n",
      "           1       1.00      0.97      0.99     22746\n",
      "\n",
      "    accuracy                           0.99     68236\n",
      "   macro avg       0.99      0.99      0.99     68236\n",
      "weighted avg       0.99      0.99      0.99     68236\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.990269066182074\n",
      "Precision: 0.9988704138803542\n",
      "Recall: 0.9719071485096281\n",
      "F1 Score: 0.9852043317438388\n",
      "AUPRC: 0.9801738539827599\n",
      "________________________________\n",
      "Fold: 4 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.98      1.00      0.99     45490\n",
      "           1       1.00      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68235\n",
      "   macro avg       0.99      0.98      0.99     68235\n",
      "weighted avg       0.99      0.99      0.99     68235\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.989169780904228\n",
      "Precision: 0.9982791413821211\n",
      "Recall: 0.9691800395691361\n",
      "F1 Score: 0.9835143998036897\n",
      "AUPRC: 0.9777855378893887\n",
      "________________________________\n",
      "Fold: 5 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.98      1.00      0.99     45490\n",
      "           1       1.00      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68235\n",
      "   macro avg       0.99      0.98      0.99     68235\n",
      "weighted avg       0.99      0.99      0.99     68235\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9886568476588261\n",
      "Precision: 0.9987741203178206\n",
      "Recall: 0.9671576170586942\n",
      "F1 Score: 0.9827116372570918\n",
      "AUPRC: 0.9769194591669123\n",
      "________________________________\n",
      "\n",
      "--Final Results--\n",
      "Avg Accuracy: 0.9894541820347869\n",
      "Avg Precision: 0.9985154206326495\n",
      "Avg Recall: 0.9698045996293821\n",
      "Avg F1 Score: 0.9839497048854907\n",
      "Avg AUPRC: 0.9784298802943636\n"
     ]
    }
   ],
   "source": [
    "print(\"--RFC + SMOTE + NearMiss--\")\n",
    "\n",
    "# separate data\n",
    "x = data.drop('Class', axis=1)\n",
    "y = data['Class']\n",
    "\n",
    "smt = SMOTE(random_state=13, sampling_strategy=0.4)\n",
    "x, y = smt.fit_resample(x, y)\n",
    "\n",
    "# under sample with near miss  \n",
    "nm = NearMiss(sampling_strategy=0.5)\n",
    "x, y = nm.fit_resample(x, y)\n",
    "\n",
    "# using stratified kfold for imbalanced dataset to validate\n",
    "sfk = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
    "\n",
    "fold = 1\n",
    "\n",
    "# used to track results after all k folds validations\n",
    "res_acc = []\n",
    "res_prec = []\n",
    "res_rec = []\n",
    "res_f1 = []\n",
    "res_auprc = []\n",
    "\n",
    "\n",
    "# loop over every fold\n",
    "for train_idx, test_idx in sfk.split(x, y):\n",
    "    x = data.drop('Class', axis=1)\n",
    "    y = data['Class']\n",
    "\n",
    "    smt = SMOTE(random_state=13, sampling_strategy=0.4)\n",
    "    x, y = smt.fit_resample(x, y)\n",
    "    \n",
    "    # under sample with near miss  \n",
    "    nm = NearMiss(sampling_strategy=0.5)\n",
    "    x, y = nm.fit_resample(x, y)\n",
    "    \n",
    "    print(f\"Fold: {fold} | Valid: {np.bincount(y)[0]} | Fraud: {np.bincount(y)[1]}\")\n",
    "    \n",
    "    # get the training data\n",
    "    x_train, x_test = x.loc[train_idx], x.loc[test_idx]\n",
    "    y_train, y_test = y[train_idx], y[test_idx] \n",
    "\n",
    "    # create the forest classifier model\n",
    "    rfc = RandomForestClassifier(n_estimators=10, max_depth=10, min_samples_split=9, random_state=13)\n",
    "    \n",
    "    # train the model on the data\n",
    "    rfc.fit(x_train, y_train)\n",
    "    \n",
    "    # get predictions\n",
    "    y_pred = rfc.predict(x_test)\n",
    "    print(classification_report(y_test, y_pred))\n",
    "\n",
    "    # solve for accuracy, precision, recall, f1 score and auprc and store for avg\n",
    "    res_acc += [accuracy_score(y_test, y_pred)]\n",
    "    res_prec += [precision_score(y_test, y_pred)]\n",
    "    res_rec += [recall_score(y_test, y_pred)]\n",
    "    res_f1 += [f1_score(y_test, y_pred)]\n",
    "    res_auprc += [average_precision_score(y_test, y_pred)]\n",
    "    \n",
    "    # print current results\n",
    "    results = {\n",
    "        \"Accuracy\": res_acc[-1],\n",
    "        \"Precision\": res_prec[-1],\n",
    "        \"Recall\": res_rec[-1],\n",
    "        \"F1 Score\": res_f1[-1],\n",
    "        \"AUPRC\": res_auprc[-1],\n",
    "    }\n",
    "    \n",
    "    print(\"________________________________\")\n",
    "    for [k, v] in results.items():\n",
    "        print(f\"{k}: {v}\")\n",
    "    print(\"________________________________\")\n",
    "    fold += 1\n",
    "\n",
    "    \n",
    "# show avg results after k folds validation\n",
    "print(\"\\n--Final Results--\")\n",
    "print(f\"Avg Accuracy: {np.mean(res_acc)}\")\n",
    "print(f\"Avg Precision: {np.mean(res_prec)}\")\n",
    "print(f\"Avg Recall: {np.mean(res_rec)}\")\n",
    "print(f\"Avg F1 Score: {np.mean(res_f1)}\")\n",
    "print(f\"Avg AUPRC: {np.mean(res_auprc)}\")\n",
    "\n",
    "final_results[\"RFC_SMOTE_NearMiss\"] = {\n",
    "    \"Accuracy\": np.mean(res_acc),\n",
    "    \"Precision\": np.mean(res_prec),\n",
    "    \"Recall\": np.mean(res_rec),\n",
    "    \"F1 Score\": np.mean(res_f1),\n",
    "    \"AUPRC\": np.mean(res_auprc),\n",
    "}"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:24:33.701884300Z",
     "start_time": "2023-08-08T18:19:48.401385600Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 853,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--XGB + NearMiss--\n",
      "Fold: 1 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.99      0.97       197\n",
      "           1       0.99      0.90      0.94        99\n",
      "\n",
      "    accuracy                           0.96       296\n",
      "   macro avg       0.97      0.95      0.96       296\n",
      "weighted avg       0.96      0.96      0.96       296\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9628378378378378\n",
      "Precision: 0.9888888888888889\n",
      "Recall: 0.898989898989899\n",
      "F1 Score: 0.9417989417989419\n",
      "AUPRC: 0.9227849061182395\n",
      "________________________________\n",
      "Fold: 2 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.99      0.98       197\n",
      "           1       0.99      0.91      0.95        98\n",
      "\n",
      "    accuracy                           0.97       295\n",
      "   macro avg       0.97      0.95      0.96       295\n",
      "weighted avg       0.97      0.97      0.97       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9661016949152542\n",
      "Precision: 0.9888888888888889\n",
      "Recall: 0.9081632653061225\n",
      "F1 Score: 0.946808510638298\n",
      "AUPRC: 0.9285810369345479\n",
      "________________________________\n",
      "Fold: 3 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.99      0.98       197\n",
      "           1       0.99      0.94      0.96        98\n",
      "\n",
      "    accuracy                           0.98       295\n",
      "   macro avg       0.98      0.97      0.97       295\n",
      "weighted avg       0.98      0.98      0.98       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.976271186440678\n",
      "Precision: 0.989247311827957\n",
      "Recall: 0.9387755102040817\n",
      "F1 Score: 0.9633507853403142\n",
      "AUPRC: 0.9490201329301541\n",
      "________________________________\n",
      "Fold: 4 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.98      0.97       197\n",
      "           1       0.97      0.91      0.94        98\n",
      "\n",
      "    accuracy                           0.96       295\n",
      "   macro avg       0.96      0.95      0.95       295\n",
      "weighted avg       0.96      0.96      0.96       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9593220338983051\n",
      "Precision: 0.967391304347826\n",
      "Recall: 0.9081632653061225\n",
      "F1 Score: 0.9368421052631578\n",
      "AUPRC: 0.9090577203615418\n",
      "________________________________\n",
      "Fold: 5 | Valid: 984 | Fraud: 492\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.99      0.98       196\n",
      "           1       0.98      0.94      0.96        99\n",
      "\n",
      "    accuracy                           0.97       295\n",
      "   macro avg       0.97      0.96      0.97       295\n",
      "weighted avg       0.97      0.97      0.97       295\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9728813559322034\n",
      "Precision: 0.9789473684210527\n",
      "Recall: 0.9393939393939394\n",
      "F1 Score: 0.9587628865979383\n",
      "AUPRC: 0.9399562079312304\n",
      "________________________________\n",
      "\n",
      "--Final Results--\n",
      "Avg Accuracy: 0.9674828218048557\n",
      "Avg Precision: 0.9826727524749227\n",
      "Avg Recall: 0.9186971758400329\n",
      "Avg F1 Score: 0.94951264592773\n",
      "Avg AUPRC: 0.9298800008551427\n"
     ]
    }
   ],
   "source": [
    "print(\"--XGB + NearMiss--\")\n",
    "\n",
    "# near miss only\n",
    "# separate data\n",
    "x = data.drop('Class', axis=1)\n",
    "y = data['Class']\n",
    "\n",
    "\n",
    "# under sample with near miss  \n",
    "nm = NearMiss(sampling_strategy=0.5)\n",
    "x, y = nm.fit_resample(x, y)\n",
    "\n",
    "# using stratified kfold for imbalanced dataset to validate\n",
    "sfk = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
    "\n",
    "fold = 1\n",
    "\n",
    "# used to track results after all k folds validations\n",
    "res_acc = []\n",
    "res_prec = []\n",
    "res_rec = []\n",
    "res_f1 = []\n",
    "res_auprc = []\n",
    "\n",
    "# loop over every fold\n",
    "for train_idx, test_idx in sfk.split(x, y):\n",
    "\n",
    "    print(f\"Fold: {fold} | Valid: {np.bincount(y)[0]} | Fraud: {np.bincount(y)[1]}\")\n",
    "    \n",
    "    # get the training data\n",
    "    x_train, x_test = x.loc[train_idx], x.loc[test_idx]\n",
    "    y_train, y_test = y[train_idx], y[test_idx] \n",
    "    \n",
    "    # create the forest classifier model\n",
    "    xgb = XGBClassifier(n_estimators=10, max_delta_step=1, eval_metric='aucpr', random_state=42)\n",
    "\n",
    "    # train the data\n",
    "    xgb.fit(x_train, y_train)\n",
    "\n",
    "    # get predictions\n",
    "    y_pred_xgb = xgb.predict(x_test)\n",
    "    print(classification_report(y_test, y_pred_xgb))\n",
    "\n",
    "    # solve for accuracy, precision, recall, f1 score and auprc and store for avg\n",
    "    res_acc += [accuracy_score(y_test, y_pred_xgb)]\n",
    "    res_prec += [precision_score(y_test, y_pred_xgb)]\n",
    "    res_rec += [recall_score(y_test, y_pred_xgb)]\n",
    "    res_f1 += [f1_score(y_test, y_pred_xgb)]\n",
    "    res_auprc += [average_precision_score(y_test, y_pred_xgb)]\n",
    "    \n",
    "    # print current results\n",
    "    results = {\n",
    "        \"Accuracy\": res_acc[-1],\n",
    "        \"Precision\": res_prec[-1],\n",
    "        \"Recall\": res_rec[-1],\n",
    "        \"F1 Score\": res_f1[-1],\n",
    "        \"AUPRC\": res_auprc[-1],\n",
    "    }\n",
    "    \n",
    "    print(\"________________________________\")\n",
    "    for [k, v] in results.items():\n",
    "        print(f\"{k}: {v}\")\n",
    "    print(\"________________________________\")\n",
    "    \n",
    "    #precision, recall, _ = precision_recall_curve(y_test, y_pred_xgb)\n",
    "    #disp = PrecisionRecallDisplay(precision=precision, recall=recall)\n",
    "    #disp.plot()\n",
    "    #plt.show()\n",
    "    \n",
    "    fold += 1\n",
    "    \n",
    "# show avg results after k folds validation\n",
    "print(\"\\n--Final Results--\")\n",
    "print(f\"Avg Accuracy: {np.mean(res_acc)}\")\n",
    "print(f\"Avg Precision: {np.mean(res_prec)}\")\n",
    "print(f\"Avg Recall: {np.mean(res_rec)}\")\n",
    "print(f\"Avg F1 Score: {np.mean(res_f1)}\")\n",
    "print(f\"Avg AUPRC: {np.mean(res_auprc)}\")\n",
    "\n",
    "final_results[\"XGB_NearMiss\"] = {\n",
    "    \"Accuracy\": np.mean(res_acc),\n",
    "    \"Precision\": np.mean(res_prec),\n",
    "    \"Recall\": np.mean(res_rec),\n",
    "    \"F1 Score\": np.mean(res_f1),\n",
    "    \"AUPRC\": np.mean(res_auprc),\n",
    "}"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:24:34.353884800Z",
     "start_time": "2023-08-08T18:24:33.707384800Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 854,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--XGB + SMOTE + NearMiss--\n",
      "Fold: 1 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99     45491\n",
      "           1       0.99      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68236\n",
      "   macro avg       0.99      0.98      0.99     68236\n",
      "weighted avg       0.99      0.99      0.99     68236\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.988349258455947\n",
      "Precision: 0.9938354931605472\n",
      "Recall: 0.9710705649593318\n",
      "F1 Score: 0.9823211545731504\n",
      "AUPRC: 0.9747273975131139\n",
      "________________________________\n",
      "Fold: 2 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.98      1.00      0.99     45491\n",
      "           1       0.99      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68236\n",
      "   macro avg       0.99      0.98      0.99     68236\n",
      "weighted avg       0.99      0.99      0.99     68236\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9869863415206049\n",
      "Precision: 0.9932746558338975\n",
      "Recall: 0.9675093427126841\n",
      "F1 Score: 0.9802227171492205\n",
      "AUPRC: 0.9718325697777074\n",
      "________________________________\n",
      "Fold: 3 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99     45490\n",
      "           1       0.99      0.97      0.98     22746\n",
      "\n",
      "    accuracy                           0.99     68236\n",
      "   macro avg       0.99      0.98      0.99     68236\n",
      "weighted avg       0.99      0.99      0.99     68236\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9878363327275925\n",
      "Precision: 0.9933369349900955\n",
      "Recall: 0.9700167062340631\n",
      "F1 Score: 0.9815383246585702\n",
      "AUPRC: 0.9735481460522404\n",
      "________________________________\n",
      "Fold: 4 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99     45490\n",
      "           1       0.99      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68235\n",
      "   macro avg       0.99      0.98      0.99     68235\n",
      "weighted avg       0.99      0.99      0.99     68235\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9881439144134242\n",
      "Precision: 0.9935649356493564\n",
      "Recall: 0.9707188393053419\n",
      "F1 Score: 0.9820090288433742\n",
      "AUPRC: 0.9742325880062493\n",
      "________________________________\n",
      "Fold: 5 | Valid: 227452 | Fraud: 113726\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99     45490\n",
      "           1       0.99      0.97      0.98     22745\n",
      "\n",
      "    accuracy                           0.99     68235\n",
      "   macro avg       0.99      0.98      0.99     68235\n",
      "weighted avg       0.99      0.99      0.99     68235\n",
      "\n",
      "________________________________\n",
      "Accuracy: 0.9883490877115849\n",
      "Precision: 0.9942357921282536\n",
      "Recall: 0.9706748735985931\n",
      "F1 Score: 0.9823140753264666\n",
      "AUPRC: 0.974854743985092\n",
      "________________________________\n",
      "\n",
      "--Final Results--\n",
      "Avg Accuracy: 0.9879329869658307\n",
      "Avg Precision: 0.99364956235243\n",
      "Avg Recall: 0.9699980653620027\n",
      "Avg F1 Score: 0.9816810601101563\n",
      "Avg AUPRC: 0.9738390890668807\n"
     ]
    }
   ],
   "source": [
    "print(\"--XGB + SMOTE + NearMiss--\")\n",
    "\n",
    "# separate data\n",
    "x = data.drop('Class', axis=1)\n",
    "y = data['Class']\n",
    "\n",
    "\n",
    "smt = SMOTE(random_state=13, sampling_strategy=0.4)\n",
    "x, y = smt.fit_resample(x, y)\n",
    "\n",
    "# under sample with near miss  \n",
    "nm = NearMiss(sampling_strategy=0.5)\n",
    "x, y = nm.fit_resample(x, y)\n",
    "\n",
    "\n",
    "# using stratified kfold for imbalanced dataset to validate\n",
    "sfk = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
    "\n",
    "fold = 1\n",
    "\n",
    "# used to track results after all k folds validations\n",
    "res_acc = []\n",
    "res_prec = []\n",
    "res_rec = []\n",
    "res_f1 = []\n",
    "res_auprc = []\n",
    "\n",
    "# loop over every fold\n",
    "for train_idx, test_idx in sfk.split(x, y):\n",
    "\n",
    "    print(f\"Fold: {fold} | Valid: {np.bincount(y)[0]} | Fraud: {np.bincount(y)[1]}\")\n",
    "    \n",
    "    # get the training data\n",
    "    x_train, x_test = x.loc[train_idx], x.loc[test_idx]\n",
    "    y_train, y_test = y[train_idx], y[test_idx] \n",
    "    \n",
    "    # create the forest classifier model\n",
    "    xgb = XGBClassifier(n_estimators=10, max_delta_step=1, eval_metric='aucpr', random_state=42)\n",
    "\n",
    "    # train the data\n",
    "    xgb.fit(x_train, y_train)\n",
    "\n",
    "    # get predictions\n",
    "    y_pred_xgb = xgb.predict(x_test)\n",
    "    print(classification_report(y_test, y_pred_xgb))\n",
    "\n",
    "    # solve for accuracy, precision, recall, f1 score and auprc and store for avg\n",
    "    res_acc += [accuracy_score(y_test, y_pred_xgb)]\n",
    "    res_prec += [precision_score(y_test, y_pred_xgb)]\n",
    "    res_rec += [recall_score(y_test, y_pred_xgb)]\n",
    "    res_f1 += [f1_score(y_test, y_pred_xgb)]\n",
    "    res_auprc += [average_precision_score(y_test, y_pred_xgb)]\n",
    "    \n",
    "    # print current results\n",
    "    results = {\n",
    "        \"Accuracy\": res_acc[-1],\n",
    "        \"Precision\": res_prec[-1],\n",
    "        \"Recall\": res_rec[-1],\n",
    "        \"F1 Score\": res_f1[-1],\n",
    "        \"AUPRC\": res_auprc[-1],\n",
    "    }\n",
    "    \n",
    "    print(\"________________________________\")\n",
    "    for [k, v] in results.items():\n",
    "        print(f\"{k}: {v}\")\n",
    "    print(\"________________________________\")\n",
    "    \n",
    "    \n",
    "    #precision, recall, _ = precision_recall_curve(y_test, y_pred_xgb)\n",
    "    #disp = PrecisionRecallDisplay(precision=precision, recall=recall)\n",
    "    #disp.plot()\n",
    "    #plt.show()\n",
    "    \n",
    "    fold += 1\n",
    "\n",
    "# show avg results after k folds validation\n",
    "print(\"\\n--Final Results--\")\n",
    "print(f\"Avg Accuracy: {np.mean(res_acc)}\")\n",
    "print(f\"Avg Precision: {np.mean(res_prec)}\")\n",
    "print(f\"Avg Recall: {np.mean(res_rec)}\")\n",
    "print(f\"Avg F1 Score: {np.mean(res_f1)}\")\n",
    "print(f\"Avg AUPRC: {np.mean(res_auprc)}\")   \n",
    "\n",
    "final_results[\"XGB_SMOTE_NearMiss\"] = {\n",
    "    \"Accuracy\": np.mean(res_acc),\n",
    "    \"Precision\": np.mean(res_prec),\n",
    "    \"Recall\": np.mean(res_rec),\n",
    "    \"F1 Score\": np.mean(res_f1),\n",
    "    \"AUPRC\": np.mean(res_auprc),\n",
    "}"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:25:16.249886100Z",
     "start_time": "2023-08-08T18:24:34.360884600Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 855,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "________________________________\n",
      "RFC_BASELINE\n",
      "\tAccuracy: 0.9995435553526912\n",
      "\tPrecision: 0.95\n",
      "\tRecall: 0.7755102040816326\n",
      "\tF1 Score: 0.8539325842696629\n",
      "\tAUPRC: 0.7371209162714276\n",
      "\n",
      "\n",
      "XGB_BASELINE\n",
      "\tAccuracy: 0.9996137776061234\n",
      "\tPrecision: 0.9634146341463414\n",
      "\tRecall: 0.8061224489795918\n",
      "\tF1 Score: 0.8777777777777778\n",
      "\tAUPRC: 0.7769637199646289\n",
      "\n",
      "\n",
      "RFC_SMOTE\n",
      "\tAccuracy: 0.999511950469036\n",
      "\tPrecision: 0.9409641236198694\n",
      "\tRecall: 0.766295609152752\n",
      "\tF1 Score: 0.8444745495739102\n",
      "\tAUPRC: 0.7214041418115088\n",
      "\n",
      "\n",
      "RFC_NearMiss\n",
      "\tAccuracy: 0.967485112230875\n",
      "\tPrecision: 0.9869961406429615\n",
      "\tRecall: 0.9146361574933003\n",
      "\tF1 Score: 0.9492638076545395\n",
      "\tAUPRC: 0.9311241567842726\n",
      "\n",
      "\n",
      "RFC_SMOTE_NearMiss\n",
      "\tAccuracy: 0.9894541820347869\n",
      "\tPrecision: 0.9985154206326495\n",
      "\tRecall: 0.9698045996293821\n",
      "\tF1 Score: 0.9839497048854907\n",
      "\tAUPRC: 0.9784298802943636\n",
      "\n",
      "\n",
      "XGB_NearMiss\n",
      "\tAccuracy: 0.9674828218048557\n",
      "\tPrecision: 0.9826727524749227\n",
      "\tRecall: 0.9186971758400329\n",
      "\tF1 Score: 0.94951264592773\n",
      "\tAUPRC: 0.9298800008551427\n",
      "\n",
      "\n",
      "XGB_SMOTE_NearMiss\n",
      "\tAccuracy: 0.9879329869658307\n",
      "\tPrecision: 0.99364956235243\n",
      "\tRecall: 0.9699980653620027\n",
      "\tF1 Score: 0.9816810601101563\n",
      "\tAUPRC: 0.9738390890668807\n"
     ]
    }
   ],
   "source": [
    "print(\"________________________________\")\n",
    "for [k, v] in final_results.items():\n",
    "    print(k)\n",
    "    for [ke, va] in v.items():\n",
    "        print(f\"\\t{ke}: {va}\")\n",
    "    print(\"\\n\")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-08T18:25:16.309885Z",
     "start_time": "2023-08-08T18:25:16.251385600Z"
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}