app v0

app.py

@@ -0,0 +1,149 @@
+import csv
+import numpy as np
+import pandas as pd
+import gradio as gr
+import joblib as jb
+import os.path as path
+
+import warnings
+warnings.filterwarnings("ignore")
+
+# Loading features names
+file_csv = path.join("model" ,"model_features.csv")
+with open(file_csv) as f:
+    reader = csv.reader(f)
+    data = list(reader)
+
+features = data[0]
+
+# Creating a list with accident types
+accident_type_list = [None,
+                      "type_ATROPELAMENTO",
+                      "type_CHOQUE",
+                      "type_COLISÃO",
+                      "type_OUTROS"]
+
+# Loading the scaler
+file_scaler_feridos = path.join("model" ,"scaler_feridos.pkl")
+scaler_feridos = jb.load(file_scaler_feridos)
+
+# Loading the model
+file_model_feridos = path.join("model" ,"model_feridos.pkl")
+model_feridos = jb.load(file_model_feridos)
+
+def fit_inputs_injured(latitude, 
+        longitude,
+        caminhao,
+        moto,
+        cars,
+        transport,
+        others,
+        holiday,
+        week_day,
+        hour_day,
+        accident_type) -> np.array:
+    """This function will process data input
+    from use to use in the model"""
+    input_dict = {col: False for col in features}
+
+    input_dict["latitude"] = latitude
+    input_dict["longitude"] = longitude
+    input_dict["caminhao"] = caminhao
+    input_dict["moto"] = moto
+    input_dict["cars"] = cars
+    input_dict["transport"] = transport
+    input_dict["others"] = others
+    input_dict["holiday"] = holiday
+
+    if week_day != 0:
+        input_dict["day_" + str(week_day)] = True
+
+    if hour_day != 0:
+        input_dict["hour_" + str(hour_day)] = True
+
+    if accident_type != 0:
+        input_dict[accident_type_list[accident_type]] = True
+
+    input_series = pd.Series(input_dict)
+
+    input_array = input_series.to_numpy().reshape(1,-1)
+
+    input_scaled = scaler_feridos.transform(input_array)
+
+    return input_scaled
+
+def predict(
+        latitude, 
+        longitude,
+        caminhao,
+        moto,
+        cars,
+        transport,
+        others,
+        holiday,
+        week_day,
+        hour_day,
+        accident_type) -> dict:
+    """This function will be call by gradio
+    when on submit action."""
+
+    input_to_predict = fit_inputs_injured(latitude, 
+        longitude,
+        caminhao,
+        moto,
+        cars,
+        transport,
+        others,
+        holiday,
+        week_day,
+        hour_day,
+        accident_type)
+
+    predic_injured = model_feridos.predict_proba(input_to_predict)
+
+    return {"No": predic_injured[0][0], "Yes": predic_injured[0][1]}
+
+demo = gr.Interface(
+    fn=predict,
+    inputs=[gr.Slider(
+                minimum=-31.054,
+                maximum=-29.054,
+                step=0.001,
+                value=-30.054,
+                label="Latitude"),
+            gr.Slider(
+                minimum=-52.196,
+                maximum=-50.196,
+                step=0.001,
+                value=-51.196,
+                label="Longitude"),
+            gr.Checkbox(label="Trucks involved?"),
+            gr.Checkbox(label="Motorcycle involved?"),    
+            gr.Checkbox(label="Cars involved?"),
+            gr.Checkbox(label="Bus involved?"),
+            gr.Checkbox(label="Other vehicle (i.e. scooter) involved?"),
+            gr.Checkbox(label="Is holiday?"),
+            gr.Radio(
+                choices=["Sun", "Mon",
+                         "Tue", "Wed",
+                         "Thu", "Fri",
+                         "Sat"],
+                type="index",
+                label="Day of Week"),
+            gr.Slider(
+                minimum=0,
+                maximum=23,
+                step=1,
+                label="Hour"),
+            gr.Dropdown(
+                choices=["Violent Collision",
+                         "Running over",
+                         "Shock",
+                         "Collision",
+                         "Other"],
+                type="index",
+                label="Accident type")],
+    outputs=gr.Label(
+        label="Are there people injured?"))
+
+demo.launch()

model/model_features.csv

@@ -0,0 +1 @@
+latitude,longitude,caminhao,moto,cars,transport,others,holiday,day_1,day_2,day_3,day_4,day_5,day_6,hour_1,hour_2,hour_3,hour_4,hour_5,hour_6,hour_7,hour_8,hour_9,hour_10,hour_11,hour_12,hour_13,hour_14,hour_15,hour_16,hour_17,hour_18,hour_19,hour_20,hour_21,hour_22,hour_23,type_ATROPELAMENTO,type_CHOQUE,type_COLIS�O,type_OUTROS

model/model_feridos.ipynb

@@ -0,0 +1,1659 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 1. Introduction\n",
+    "\n",
+    "This notebook was written to train Porto Alegre Traffic Accidents Data after the first cleaning, processing, and transforming step. This was made in a notebook in the `data` folder. In truth, we will have 3 models.\n",
+    "\n",
+    "1. Predict the probability of injured people.\n",
+    "\n",
+    "2. Predict the probability of seriously injured people.\n",
+    "\n",
+    "3. Predict the probability of dead people in the event or after it.\n",
+    "\n",
+    "The path to training the models will be the same, just make some filtering on data and analyze the results properly."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 2. Data Loading"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>0</th>\n",
+       "      <th>1</th>\n",
+       "      <th>2</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>latitude</th>\n",
+       "      <td>-30.009614</td>\n",
+       "      <td>-30.0403</td>\n",
+       "      <td>-30.069</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>longitude</th>\n",
+       "      <td>-51.185581</td>\n",
+       "      <td>-51.1958</td>\n",
+       "      <td>-51.1437</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>feridos</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>feridos_gr</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>fatais</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>caminhao</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>moto</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>cars</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>transport</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>others</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>holiday</th>\n",
+       "      <td>False</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_2</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_3</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_4</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_5</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_6</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_2</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_3</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_4</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_5</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_6</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_7</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_8</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_9</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_10</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_11</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_12</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_13</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_14</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_15</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_16</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_17</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_18</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_19</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_20</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_21</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_22</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_23</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_ATROPELAMENTO</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_CHOQUE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_COLISÃO</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_OUTROS</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                            0        1        2\n",
+       "latitude           -30.009614 -30.0403  -30.069\n",
+       "longitude          -51.185581 -51.1958 -51.1437\n",
+       "feridos                  True     True     True\n",
+       "feridos_gr              False    False    False\n",
+       "fatais                  False    False    False\n",
+       "caminhao                False    False    False\n",
+       "moto                     True     True    False\n",
+       "cars                     True     True     True\n",
+       "transport               False    False    False\n",
+       "others                  False    False    False\n",
+       "holiday                 False     True     True\n",
+       "day_1                       0        0        0\n",
+       "day_2                       0        0        0\n",
+       "day_3                       0        0        0\n",
+       "day_4                       0        0        0\n",
+       "day_5                       1        0        0\n",
+       "day_6                       0        1        1\n",
+       "hour_1                      0        0        0\n",
+       "hour_2                      0        0        0\n",
+       "hour_3                      0        0        0\n",
+       "hour_4                      0        0        0\n",
+       "hour_5                      0        0        0\n",
+       "hour_6                      0        0        0\n",
+       "hour_7                      0        0        0\n",
+       "hour_8                      0        0        0\n",
+       "hour_9                      0        0        0\n",
+       "hour_10                     0        1        0\n",
+       "hour_11                     0        0        0\n",
+       "hour_12                     0        0        0\n",
+       "hour_13                     0        0        0\n",
+       "hour_14                     0        0        0\n",
+       "hour_15                     0        0        0\n",
+       "hour_16                     0        0        0\n",
+       "hour_17                     0        0        0\n",
+       "hour_18                     0        0        0\n",
+       "hour_19                     1        0        1\n",
+       "hour_20                     0        0        0\n",
+       "hour_21                     0        0        0\n",
+       "hour_22                     0        0        0\n",
+       "hour_23                     0        0        0\n",
+       "type_ATROPELAMENTO          0        0        1\n",
+       "type_CHOQUE                 0        0        0\n",
+       "type_COLISÃO                0        0        0\n",
+       "type_OUTROS                 0        0        0"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import os.path as path\n",
+    "from pandas import read_csv\n",
+    "\n",
+    "file_csv =  path.abspath(\"../\")\n",
+    "\n",
+    "file_csv = path.join(file_csv, \"data\" ,\"accidents_trans.csv\")\n",
+    "\n",
+    "accidents_trans = read_csv(file_csv)\n",
+    "\n",
+    "accidents_trans.head(3).T"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 3. Data Preparation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import joblib as jb # Use to save the model to deploy\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "from sklearn.model_selection import train_test_split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Our model to predict the probability of feridos will be create with 68218 rows and 41 features.\n"
+     ]
+    }
+   ],
+   "source": [
+    "outputs = [\"feridos\", \"feridos_gr\", \"fatais\"]\n",
+    "inputs = [col for col in accidents_trans.columns if col not in outputs]\n",
+    "\n",
+    "X = accidents_trans[inputs].copy()\n",
+    "Y = accidents_trans[outputs].copy()\n",
+    "\n",
+    "# Filtering data considering the output\n",
+    "output = \"feridos\"\n",
+    "\n",
+    "if output == \"feridos_gr\":\n",
+    "    X = X[Y[\"feridos\"]]\n",
+    "    Y = Y.loc[Y[\"feridos\"], \"feridos_gr\"]\n",
+    "elif output == \"fatais\":\n",
+    "    X = X[Y[\"feridos_gr\"]]\n",
+    "    Y = Y.loc[Y[\"feridos_gr\"], \"fatais\"]\n",
+    "else:\n",
+    "    Y = Y[\"feridos\"]\n",
+    "\n",
+    "print(f\"Our model to predict the probability of \" \\\n",
+    "      f\"{output} will be create with {X.shape[0]} \" \\\n",
+    "      f\"rows and {X.shape[1]} features.\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import csv\n",
+    "\n",
+    "with open(\"model_features.csv\", 'w') as f:\n",
+    "    writer = csv.writer(f)\n",
+    "    writer.writerow(X.columns)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Considering that we will use models scaling sensitive, we will need to scale our data first. Beside this, we will need to save our scaler for future use."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['c:\\\\Users\\\\grego\\\\OneDrive\\\\Documentos\\\\Documentos Pessoais\\\\00_DataCamp\\\\09_VSC\\\\poa_car_accidents\\\\poa_car_accidents\\\\model\\\\scaler_feridos.pkl']"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Setting the random state using my luck number :-)\n",
+    "lucky_num = 7\n",
+    "\n",
+    "# X_train and y_train to train our model\n",
+    "X_train, X_test, y_train, y_test = train_test_split(\n",
+    "    X,\n",
+    "    Y,\n",
+    "    test_size=0.30,\n",
+    "    random_state=lucky_num,\n",
+    "    shuffle=True,  # Used because our data is sort by date\n",
+    "    stratify=Y)  # Used because our data is unbalanced\n",
+    "\n",
+    "# Scaling\n",
+    "scaler = StandardScaler()\n",
+    "X_train = scaler.fit_transform(X_train)\n",
+    "X_test = scaler.transform(X_test)\n",
+    "\n",
+    "# Saving scaler\n",
+    "file_name = \"scaler_\" + output + '.pkl'\n",
+    "jb.dump(scaler, path.join(path.abspath(\"./\"), file_name))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 4. Data Modeling\n",
+    "\n",
+    "We will create and use cross-validation to evaluate the following models:\n",
+    "\n",
+    "- Logistic Regression;\n",
+    "\n",
+    "- Gaussian Naive Bayes;\n",
+    "\n",
+    "- K Neighbors;\n",
+    "\n",
+    "- Random Forest;\n",
+    "\n",
+    "- Gradient Boosting; and,\n",
+    "\n",
+    "- XGBoost.\n",
+    "\n",
+    "We will use two scores to select and evaluate our models:\n",
+    "\n",
+    "- F1 score: composition between the precision (how much our model correct classify every true label) and recall (how moch our model correct indicate true labels); and,\n",
+    "\n",
+    "- Brier score: average between the correct and the predict probability.\n",
+    "\n",
+    "However, we will see other metrics to support our decision:\n",
+    "\n",
+    "- Accurancy;\n",
+    "\n",
+    "- ROC_AOC; and,\n",
+    "\n",
+    "- Log loss (an other way to quantify the quality of probability predictions).\n",
+    "\n",
+    "And, before you go, we will find for each model if there is a hyperparameter to deal with the unbalanced output."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import xgboost as xgb\n",
+    "from sklearn.naive_bayes import GaussianNB\n",
+    "from sklearn.neighbors import KNeighborsClassifier\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.model_selection import cross_validate \n",
+    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier, VotingClassifier\n",
+    "from sklearn.metrics import accuracy_score, recall_score, precision_score, roc_auc_score, f1_score, brier_score_loss, log_loss\n",
+    "\n",
+    "scores = [\"accuracy\", \"f1\", \"precision\", \"recall\", \"roc_auc\", \"neg_brier_score\",\"neg_log_loss\"]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def eval_model(cls) -> tuple:\n",
+    "    \"\"\"This function will calculate the metrics\n",
+    "    to evaluate a classification model.\n",
+    "    \"\"\"\n",
+    "    # Predicting labels and probabilities\n",
+    "    y_pred = cls.predict(X_test)\n",
+    "    y_prob = cls.predict_proba(X_test)[:,1]\n",
+    "\n",
+    "    # Calculating scores\n",
+    "    accurancy = accuracy_score(y_test, y_pred)\n",
+    "    f1 = f1_score(y_test, y_pred)\n",
+    "    recall = recall_score(y_test, y_pred)\n",
+    "    precision = precision_score(y_test, y_pred)\n",
+    "    roc_auc = roc_auc_score(y_test, y_prob)  # https://datascience.stackexchange.com/questions/114394/does-roc-auc-different-between-crossval-and-test-set-indicate-overfitting-or-oth\n",
+    "    brier_score = brier_score_loss(y_test, y_prob)\n",
+    "    log_loss_value = log_loss(y_test, y_prob)\n",
+    "\n",
+    "    return accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value\n",
+    "\n",
+    "def create_model(name: str, cls) -> list:\n",
+    "    \"\"\"This function will create some models\n",
+    "    and return scores to evaluate it.\"\"\"\n",
+    "    # Ftting model\n",
+    "    cls.fit(X_train, y_train)\n",
+    "\n",
+    "    # Using cross-validation to evaluate the model fitted\n",
+    "    cls_cross = cross_validate(\n",
+    "        estimator=cls,\n",
+    "        X=X_train,\n",
+    "        y=y_train,\n",
+    "        cv=5,\n",
+    "        scoring=scores)\n",
+    "\n",
+    "    df_cv = pd.DataFrame.from_dict(cls_cross, orient='index', columns=[\"CV\"+str(i) for i in range(1,6)])\n",
+    "\n",
+    "    # Calculating score to test set\n",
+    "    accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value = eval_model(cls)\n",
+    "\n",
+    "    # Filling a dataframe to better presentation\n",
+    "    df_cv.at[\"test_accuracy\", \"TestSet\"] = accurancy\n",
+    "    df_cv.at[\"test_f1\", \"TestSet\"] = f1\n",
+    "    df_cv.at[\"test_recall\", \"TestSet\"] = recall\n",
+    "    df_cv.at[\"test_precision\", \"TestSet\"] = precision\n",
+    "    df_cv.at[\"test_roc_auc\", \"TestSet\"] = roc_auc\n",
+    "    df_cv.at[\"test_neg_brier_score\", \"TestSet\"] = -brier_score\n",
+    "    df_cv.at[\"test_neg_log_loss\", \"TestSet\"] = -log_loss_value\n",
+    "\n",
+    "    caption = f\"{name} Validation Scores\"\n",
+    "\n",
+    "    display(df_cv.style.set_caption(caption))\n",
+    "\n",
+    "    return [accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_31097\">\n",
+       "  <caption>LR Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_31097_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_31097_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_31097_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_31097_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_31097_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_31097_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_31097_row0_col0\" class=\"data row0 col0\" >0.082354</td>\n",
+       "      <td id=\"T_31097_row0_col1\" class=\"data row0 col1\" >0.080257</td>\n",
+       "      <td id=\"T_31097_row0_col2\" class=\"data row0 col2\" >0.089329</td>\n",
+       "      <td id=\"T_31097_row0_col3\" class=\"data row0 col3\" >0.094720</td>\n",
+       "      <td id=\"T_31097_row0_col4\" class=\"data row0 col4\" >0.087742</td>\n",
+       "      <td id=\"T_31097_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_31097_row1_col0\" class=\"data row1 col0\" >0.016066</td>\n",
+       "      <td id=\"T_31097_row1_col1\" class=\"data row1 col1\" >0.017635</td>\n",
+       "      <td id=\"T_31097_row1_col2\" class=\"data row1 col2\" >0.020100</td>\n",
+       "      <td id=\"T_31097_row1_col3\" class=\"data row1 col3\" >0.018260</td>\n",
+       "      <td id=\"T_31097_row1_col4\" class=\"data row1 col4\" >0.018356</td>\n",
+       "      <td id=\"T_31097_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_31097_row2_col0\" class=\"data row2 col0\" >0.869228</td>\n",
+       "      <td id=\"T_31097_row2_col1\" class=\"data row2 col1\" >0.868391</td>\n",
+       "      <td id=\"T_31097_row2_col2\" class=\"data row2 col2\" >0.872356</td>\n",
+       "      <td id=\"T_31097_row2_col3\" class=\"data row2 col3\" >0.869005</td>\n",
+       "      <td id=\"T_31097_row2_col4\" class=\"data row2 col4\" >0.867539</td>\n",
+       "      <td id=\"T_31097_row2_col5\" class=\"data row2 col5\" >0.865924</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_31097_row3_col0\" class=\"data row3 col0\" >0.817584</td>\n",
+       "      <td id=\"T_31097_row3_col1\" class=\"data row3 col1\" >0.818116</td>\n",
+       "      <td id=\"T_31097_row3_col2\" class=\"data row3 col2\" >0.823920</td>\n",
+       "      <td id=\"T_31097_row3_col3\" class=\"data row3 col3\" >0.819611</td>\n",
+       "      <td id=\"T_31097_row3_col4\" class=\"data row3 col4\" >0.817011</td>\n",
+       "      <td id=\"T_31097_row3_col5\" class=\"data row3 col5\" >0.814469</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_31097_row4_col0\" class=\"data row4 col0\" >0.854135</td>\n",
+       "      <td id=\"T_31097_row4_col1\" class=\"data row4 col1\" >0.846154</td>\n",
+       "      <td id=\"T_31097_row4_col2\" class=\"data row4 col2\" >0.850582</td>\n",
+       "      <td id=\"T_31097_row4_col3\" class=\"data row4 col3\" >0.844326</td>\n",
+       "      <td id=\"T_31097_row4_col4\" class=\"data row4 col4\" >0.844498</td>\n",
+       "      <td id=\"T_31097_row4_col5\" class=\"data row4 col5\" >0.843439</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_31097_row5_col0\" class=\"data row5 col0\" >0.784034</td>\n",
+       "      <td id=\"T_31097_row5_col1\" class=\"data row5 col1\" >0.791877</td>\n",
+       "      <td id=\"T_31097_row5_col2\" class=\"data row5 col2\" >0.798880</td>\n",
+       "      <td id=\"T_31097_row5_col3\" class=\"data row5 col3\" >0.796301</td>\n",
+       "      <td id=\"T_31097_row5_col4\" class=\"data row5 col4\" >0.791258</td>\n",
+       "      <td id=\"T_31097_row5_col5\" class=\"data row5 col5\" >0.787423</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_31097_row6_col0\" class=\"data row6 col0\" >0.903418</td>\n",
+       "      <td id=\"T_31097_row6_col1\" class=\"data row6 col1\" >0.904970</td>\n",
+       "      <td id=\"T_31097_row6_col2\" class=\"data row6 col2\" >0.906377</td>\n",
+       "      <td id=\"T_31097_row6_col3\" class=\"data row6 col3\" >0.902405</td>\n",
+       "      <td id=\"T_31097_row6_col4\" class=\"data row6 col4\" >0.906939</td>\n",
+       "      <td id=\"T_31097_row6_col5\" class=\"data row6 col5\" >0.904458</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_31097_row7_col0\" class=\"data row7 col0\" >-0.109808</td>\n",
+       "      <td id=\"T_31097_row7_col1\" class=\"data row7 col1\" >-0.109221</td>\n",
+       "      <td id=\"T_31097_row7_col2\" class=\"data row7 col2\" >-0.106382</td>\n",
+       "      <td id=\"T_31097_row7_col3\" class=\"data row7 col3\" >-0.110939</td>\n",
+       "      <td id=\"T_31097_row7_col4\" class=\"data row7 col4\" >-0.109709</td>\n",
+       "      <td id=\"T_31097_row7_col5\" class=\"data row7 col5\" >-0.110435</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_31097_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_31097_row8_col0\" class=\"data row8 col0\" >-0.370200</td>\n",
+       "      <td id=\"T_31097_row8_col1\" class=\"data row8 col1\" >-0.366684</td>\n",
+       "      <td id=\"T_31097_row8_col2\" class=\"data row8 col2\" >-0.360534</td>\n",
+       "      <td id=\"T_31097_row8_col3\" class=\"data row8 col3\" >-0.372374</td>\n",
+       "      <td id=\"T_31097_row8_col4\" class=\"data row8 col4\" >-0.367532</td>\n",
+       "      <td id=\"T_31097_row8_col5\" class=\"data row8 col5\" >-0.370350</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a6427220>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_750d2\">\n",
+       "  <caption>NB Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_750d2_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_750d2_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_750d2_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_750d2_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_750d2_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_750d2_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_750d2_row0_col0\" class=\"data row0 col0\" >0.035410</td>\n",
+       "      <td id=\"T_750d2_row0_col1\" class=\"data row0 col1\" >0.030015</td>\n",
+       "      <td id=\"T_750d2_row0_col2\" class=\"data row0 col2\" >0.032639</td>\n",
+       "      <td id=\"T_750d2_row0_col3\" class=\"data row0 col3\" >0.029752</td>\n",
+       "      <td id=\"T_750d2_row0_col4\" class=\"data row0 col4\" >0.030653</td>\n",
+       "      <td id=\"T_750d2_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_750d2_row1_col0\" class=\"data row1 col0\" >0.037826</td>\n",
+       "      <td id=\"T_750d2_row1_col1\" class=\"data row1 col1\" >0.040993</td>\n",
+       "      <td id=\"T_750d2_row1_col2\" class=\"data row1 col2\" >0.032376</td>\n",
+       "      <td id=\"T_750d2_row1_col3\" class=\"data row1 col3\" >0.030767</td>\n",
+       "      <td id=\"T_750d2_row1_col4\" class=\"data row1 col4\" >0.028092</td>\n",
+       "      <td id=\"T_750d2_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_750d2_row2_col0\" class=\"data row2 col0\" >0.768401</td>\n",
+       "      <td id=\"T_750d2_row2_col1\" class=\"data row2 col1\" >0.763376</td>\n",
+       "      <td id=\"T_750d2_row2_col2\" class=\"data row2 col2\" >0.765131</td>\n",
+       "      <td id=\"T_750d2_row2_col3\" class=\"data row2 col3\" >0.771518</td>\n",
+       "      <td id=\"T_750d2_row2_col4\" class=\"data row2 col4\" >0.772251</td>\n",
+       "      <td id=\"T_750d2_row2_col5\" class=\"data row2 col5\" >0.766637</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_750d2_row3_col0\" class=\"data row3 col0\" >0.667068</td>\n",
+       "      <td id=\"T_750d2_row3_col1\" class=\"data row3 col1\" >0.654223</td>\n",
+       "      <td id=\"T_750d2_row3_col2\" class=\"data row3 col2\" >0.660922</td>\n",
+       "      <td id=\"T_750d2_row3_col3\" class=\"data row3 col3\" >0.675876</td>\n",
+       "      <td id=\"T_750d2_row3_col4\" class=\"data row3 col4\" >0.668899</td>\n",
+       "      <td id=\"T_750d2_row3_col5\" class=\"data row3 col5\" >0.664795</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_750d2_row4_col0\" class=\"data row4 col0\" >0.720885</td>\n",
+       "      <td id=\"T_750d2_row4_col1\" class=\"data row4 col1\" >0.720836</td>\n",
+       "      <td id=\"T_750d2_row4_col2\" class=\"data row4 col2\" >0.717898</td>\n",
+       "      <td id=\"T_750d2_row4_col3\" class=\"data row4 col3\" >0.719254</td>\n",
+       "      <td id=\"T_750d2_row4_col4\" class=\"data row4 col4\" >0.732333</td>\n",
+       "      <td id=\"T_750d2_row4_col5\" class=\"data row4 col5\" >0.717684</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_750d2_row5_col0\" class=\"data row5 col0\" >0.620728</td>\n",
+       "      <td id=\"T_750d2_row5_col1\" class=\"data row5 col1\" >0.598880</td>\n",
+       "      <td id=\"T_750d2_row5_col2\" class=\"data row5 col2\" >0.612325</td>\n",
+       "      <td id=\"T_750d2_row5_col3\" class=\"data row5 col3\" >0.637433</td>\n",
+       "      <td id=\"T_750d2_row5_col4\" class=\"data row5 col4\" >0.615579</td>\n",
+       "      <td id=\"T_750d2_row5_col5\" class=\"data row5 col5\" >0.619166</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_750d2_row6_col0\" class=\"data row6 col0\" >0.852290</td>\n",
+       "      <td id=\"T_750d2_row6_col1\" class=\"data row6 col1\" >0.847184</td>\n",
+       "      <td id=\"T_750d2_row6_col2\" class=\"data row6 col2\" >0.843733</td>\n",
+       "      <td id=\"T_750d2_row6_col3\" class=\"data row6 col3\" >0.851873</td>\n",
+       "      <td id=\"T_750d2_row6_col4\" class=\"data row6 col4\" >0.856047</td>\n",
+       "      <td id=\"T_750d2_row6_col5\" class=\"data row6 col5\" >0.848834</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_750d2_row7_col0\" class=\"data row7 col0\" >-0.206596</td>\n",
+       "      <td id=\"T_750d2_row7_col1\" class=\"data row7 col1\" >-0.210362</td>\n",
+       "      <td id=\"T_750d2_row7_col2\" class=\"data row7 col2\" >-0.211968</td>\n",
+       "      <td id=\"T_750d2_row7_col3\" class=\"data row7 col3\" >-0.204214</td>\n",
+       "      <td id=\"T_750d2_row7_col4\" class=\"data row7 col4\" >-0.202682</td>\n",
+       "      <td id=\"T_750d2_row7_col5\" class=\"data row7 col5\" >-0.208278</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_750d2_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_750d2_row8_col0\" class=\"data row8 col0\" >-1.668014</td>\n",
+       "      <td id=\"T_750d2_row8_col1\" class=\"data row8 col1\" >-1.788896</td>\n",
+       "      <td id=\"T_750d2_row8_col2\" class=\"data row8 col2\" >-1.917438</td>\n",
+       "      <td id=\"T_750d2_row8_col3\" class=\"data row8 col3\" >-1.662381</td>\n",
+       "      <td id=\"T_750d2_row8_col4\" class=\"data row8 col4\" >-1.670358</td>\n",
+       "      <td id=\"T_750d2_row8_col5\" class=\"data row8 col5\" >-1.761326</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a6bb4820>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_2d5a3\">\n",
+       "  <caption>KNN Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_2d5a3_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_2d5a3_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_2d5a3_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_2d5a3_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_2d5a3_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_2d5a3_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_2d5a3_row0_col0\" class=\"data row0 col0\" >0.010002</td>\n",
+       "      <td id=\"T_2d5a3_row0_col1\" class=\"data row0 col1\" >0.011312</td>\n",
+       "      <td id=\"T_2d5a3_row0_col2\" class=\"data row0 col2\" >0.011621</td>\n",
+       "      <td id=\"T_2d5a3_row0_col3\" class=\"data row0 col3\" >0.013843</td>\n",
+       "      <td id=\"T_2d5a3_row0_col4\" class=\"data row0 col4\" >0.011473</td>\n",
+       "      <td id=\"T_2d5a3_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_2d5a3_row1_col0\" class=\"data row1 col0\" >1.660269</td>\n",
+       "      <td id=\"T_2d5a3_row1_col1\" class=\"data row1 col1\" >1.360570</td>\n",
+       "      <td id=\"T_2d5a3_row1_col2\" class=\"data row1 col2\" >1.651296</td>\n",
+       "      <td id=\"T_2d5a3_row1_col3\" class=\"data row1 col3\" >1.734129</td>\n",
+       "      <td id=\"T_2d5a3_row1_col4\" class=\"data row1 col4\" >1.823339</td>\n",
+       "      <td id=\"T_2d5a3_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_2d5a3_row2_col0\" class=\"data row2 col0\" >0.842320</td>\n",
+       "      <td id=\"T_2d5a3_row2_col1\" class=\"data row2 col1\" >0.848707</td>\n",
+       "      <td id=\"T_2d5a3_row2_col2\" class=\"data row2 col2\" >0.847330</td>\n",
+       "      <td id=\"T_2d5a3_row2_col3\" class=\"data row2 col3\" >0.842723</td>\n",
+       "      <td id=\"T_2d5a3_row2_col4\" class=\"data row2 col4\" >0.847749</td>\n",
+       "      <td id=\"T_2d5a3_row2_col5\" class=\"data row2 col5\" >0.843692</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_2d5a3_row3_col0\" class=\"data row3 col0\" >0.776492</td>\n",
+       "      <td id=\"T_2d5a3_row3_col1\" class=\"data row3 col1\" >0.787218</td>\n",
+       "      <td id=\"T_2d5a3_row3_col2\" class=\"data row3 col2\" >0.783551</td>\n",
+       "      <td id=\"T_2d5a3_row3_col3\" class=\"data row3 col3\" >0.779053</td>\n",
+       "      <td id=\"T_2d5a3_row3_col4\" class=\"data row3 col4\" >0.786365</td>\n",
+       "      <td id=\"T_2d5a3_row3_col5\" class=\"data row3 col5\" >0.779698</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_2d5a3_row4_col0\" class=\"data row4 col0\" >0.825758</td>\n",
+       "      <td id=\"T_2d5a3_row4_col1\" class=\"data row4 col1\" >0.829867</td>\n",
+       "      <td id=\"T_2d5a3_row4_col2\" class=\"data row4 col2\" >0.833544</td>\n",
+       "      <td id=\"T_2d5a3_row4_col3\" class=\"data row4 col3\" >0.820068</td>\n",
+       "      <td id=\"T_2d5a3_row4_col4\" class=\"data row4 col4\" >0.826691</td>\n",
+       "      <td id=\"T_2d5a3_row4_col5\" class=\"data row4 col5\" >0.823778</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_2d5a3_row5_col0\" class=\"data row5 col0\" >0.732773</td>\n",
+       "      <td id=\"T_2d5a3_row5_col1\" class=\"data row5 col1\" >0.748739</td>\n",
+       "      <td id=\"T_2d5a3_row5_col2\" class=\"data row5 col2\" >0.739216</td>\n",
+       "      <td id=\"T_2d5a3_row5_col3\" class=\"data row5 col3\" >0.741945</td>\n",
+       "      <td id=\"T_2d5a3_row5_col4\" class=\"data row5 col4\" >0.749790</td>\n",
+       "      <td id=\"T_2d5a3_row5_col5\" class=\"data row5 col5\" >0.740097</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_2d5a3_row6_col0\" class=\"data row6 col0\" >0.867330</td>\n",
+       "      <td id=\"T_2d5a3_row6_col1\" class=\"data row6 col1\" >0.869924</td>\n",
+       "      <td id=\"T_2d5a3_row6_col2\" class=\"data row6 col2\" >0.872951</td>\n",
+       "      <td id=\"T_2d5a3_row6_col3\" class=\"data row6 col3\" >0.866868</td>\n",
+       "      <td id=\"T_2d5a3_row6_col4\" class=\"data row6 col4\" >0.872277</td>\n",
+       "      <td id=\"T_2d5a3_row6_col5\" class=\"data row6 col5\" >0.872155</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_2d5a3_row7_col0\" class=\"data row7 col0\" >-0.130989</td>\n",
+       "      <td id=\"T_2d5a3_row7_col1\" class=\"data row7 col1\" >-0.127425</td>\n",
+       "      <td id=\"T_2d5a3_row7_col2\" class=\"data row7 col2\" >-0.126777</td>\n",
+       "      <td id=\"T_2d5a3_row7_col3\" class=\"data row7 col3\" >-0.130655</td>\n",
+       "      <td id=\"T_2d5a3_row7_col4\" class=\"data row7 col4\" >-0.127401</td>\n",
+       "      <td id=\"T_2d5a3_row7_col5\" class=\"data row7 col5\" >-0.128215</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_2d5a3_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_2d5a3_row8_col0\" class=\"data row8 col0\" >-2.083997</td>\n",
+       "      <td id=\"T_2d5a3_row8_col1\" class=\"data row8 col1\" >-1.959589</td>\n",
+       "      <td id=\"T_2d5a3_row8_col2\" class=\"data row8 col2\" >-1.815403</td>\n",
+       "      <td id=\"T_2d5a3_row8_col3\" class=\"data row8 col3\" >-2.007178</td>\n",
+       "      <td id=\"T_2d5a3_row8_col4\" class=\"data row8 col4\" >-1.929602</td>\n",
+       "      <td id=\"T_2d5a3_row8_col5\" class=\"data row8 col5\" >-1.877810</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a6bb4820>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_08606\">\n",
+       "  <caption>RF Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_08606_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_08606_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_08606_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_08606_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_08606_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_08606_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_08606_row0_col0\" class=\"data row0 col0\" >4.099665</td>\n",
+       "      <td id=\"T_08606_row0_col1\" class=\"data row0 col1\" >4.061200</td>\n",
+       "      <td id=\"T_08606_row0_col2\" class=\"data row0 col2\" >4.090116</td>\n",
+       "      <td id=\"T_08606_row0_col3\" class=\"data row0 col3\" >4.055705</td>\n",
+       "      <td id=\"T_08606_row0_col4\" class=\"data row0 col4\" >4.050387</td>\n",
+       "      <td id=\"T_08606_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_08606_row1_col0\" class=\"data row1 col0\" >0.390365</td>\n",
+       "      <td id=\"T_08606_row1_col1\" class=\"data row1 col1\" >0.389244</td>\n",
+       "      <td id=\"T_08606_row1_col2\" class=\"data row1 col2\" >0.392108</td>\n",
+       "      <td id=\"T_08606_row1_col3\" class=\"data row1 col3\" >0.387358</td>\n",
+       "      <td id=\"T_08606_row1_col4\" class=\"data row1 col4\" >0.400155</td>\n",
+       "      <td id=\"T_08606_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_08606_row2_col0\" class=\"data row2 col0\" >0.856141</td>\n",
+       "      <td id=\"T_08606_row2_col1\" class=\"data row2 col1\" >0.859282</td>\n",
+       "      <td id=\"T_08606_row2_col2\" class=\"data row2 col2\" >0.861571</td>\n",
+       "      <td id=\"T_08606_row2_col3\" class=\"data row2 col3\" >0.853508</td>\n",
+       "      <td id=\"T_08606_row2_col4\" class=\"data row2 col4\" >0.855393</td>\n",
+       "      <td id=\"T_08606_row2_col5\" class=\"data row2 col5\" >0.856152</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_08606_row3_col0\" class=\"data row3 col0\" >0.800349</td>\n",
+       "      <td id=\"T_08606_row3_col1\" class=\"data row3 col1\" >0.805217</td>\n",
+       "      <td id=\"T_08606_row3_col2\" class=\"data row3 col2\" >0.807681</td>\n",
+       "      <td id=\"T_08606_row3_col3\" class=\"data row3 col3\" >0.798676</td>\n",
+       "      <td id=\"T_08606_row3_col4\" class=\"data row3 col4\" >0.799477</td>\n",
+       "      <td id=\"T_08606_row3_col5\" class=\"data row3 col5\" >0.800623</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_08606_row4_col0\" class=\"data row4 col0\" >0.831522</td>\n",
+       "      <td id=\"T_08606_row4_col1\" class=\"data row4 col1\" >0.834234</td>\n",
+       "      <td id=\"T_08606_row4_col2\" class=\"data row4 col2\" >0.840194</td>\n",
+       "      <td id=\"T_08606_row4_col3\" class=\"data row4 col3\" >0.821006</td>\n",
+       "      <td id=\"T_08606_row4_col4\" class=\"data row4 col4\" >0.829717</td>\n",
+       "      <td id=\"T_08606_row4_col5\" class=\"data row4 col5\" >0.830547</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_08606_row5_col0\" class=\"data row5 col0\" >0.771429</td>\n",
+       "      <td id=\"T_08606_row5_col1\" class=\"data row5 col1\" >0.778151</td>\n",
+       "      <td id=\"T_08606_row5_col2\" class=\"data row5 col2\" >0.777591</td>\n",
+       "      <td id=\"T_08606_row5_col3\" class=\"data row5 col3\" >0.777529</td>\n",
+       "      <td id=\"T_08606_row5_col4\" class=\"data row5 col4\" >0.771365</td>\n",
+       "      <td id=\"T_08606_row5_col5\" class=\"data row5 col5\" >0.772781</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_08606_row6_col0\" class=\"data row6 col0\" >0.890122</td>\n",
+       "      <td id=\"T_08606_row6_col1\" class=\"data row6 col1\" >0.890561</td>\n",
+       "      <td id=\"T_08606_row6_col2\" class=\"data row6 col2\" >0.897321</td>\n",
+       "      <td id=\"T_08606_row6_col3\" class=\"data row6 col3\" >0.887396</td>\n",
+       "      <td id=\"T_08606_row6_col4\" class=\"data row6 col4\" >0.891078</td>\n",
+       "      <td id=\"T_08606_row6_col5\" class=\"data row6 col5\" >0.893466</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_08606_row7_col0\" class=\"data row7 col0\" >-0.116884</td>\n",
+       "      <td id=\"T_08606_row7_col1\" class=\"data row7 col1\" >-0.114867</td>\n",
+       "      <td id=\"T_08606_row7_col2\" class=\"data row7 col2\" >-0.111343</td>\n",
+       "      <td id=\"T_08606_row7_col3\" class=\"data row7 col3\" >-0.117719</td>\n",
+       "      <td id=\"T_08606_row7_col4\" class=\"data row7 col4\" >-0.116295</td>\n",
+       "      <td id=\"T_08606_row7_col5\" class=\"data row7 col5\" >-0.115285</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_08606_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_08606_row8_col0\" class=\"data row8 col0\" >-0.607395</td>\n",
+       "      <td id=\"T_08606_row8_col1\" class=\"data row8 col1\" >-0.579640</td>\n",
+       "      <td id=\"T_08606_row8_col2\" class=\"data row8 col2\" >-0.536542</td>\n",
+       "      <td id=\"T_08606_row8_col3\" class=\"data row8 col3\" >-0.614554</td>\n",
+       "      <td id=\"T_08606_row8_col4\" class=\"data row8 col4\" >-0.631888</td>\n",
+       "      <td id=\"T_08606_row8_col5\" class=\"data row8 col5\" >-0.562042</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a6bb4820>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_14134\">\n",
+       "  <caption>GBC Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_14134_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_14134_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_14134_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_14134_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_14134_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_14134_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_14134_row0_col0\" class=\"data row0 col0\" >4.591437</td>\n",
+       "      <td id=\"T_14134_row0_col1\" class=\"data row0 col1\" >4.437213</td>\n",
+       "      <td id=\"T_14134_row0_col2\" class=\"data row0 col2\" >4.121067</td>\n",
+       "      <td id=\"T_14134_row0_col3\" class=\"data row0 col3\" >4.142180</td>\n",
+       "      <td id=\"T_14134_row0_col4\" class=\"data row0 col4\" >4.113901</td>\n",
+       "      <td id=\"T_14134_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_14134_row1_col0\" class=\"data row1 col0\" >0.055993</td>\n",
+       "      <td id=\"T_14134_row1_col1\" class=\"data row1 col1\" >0.048113</td>\n",
+       "      <td id=\"T_14134_row1_col2\" class=\"data row1 col2\" >0.049492</td>\n",
+       "      <td id=\"T_14134_row1_col3\" class=\"data row1 col3\" >0.050163</td>\n",
+       "      <td id=\"T_14134_row1_col4\" class=\"data row1 col4\" >0.055706</td>\n",
+       "      <td id=\"T_14134_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_14134_row2_col0\" class=\"data row2 col0\" >0.871113</td>\n",
+       "      <td id=\"T_14134_row2_col1\" class=\"data row2 col1\" >0.873207</td>\n",
+       "      <td id=\"T_14134_row2_col2\" class=\"data row2 col2\" >0.878639</td>\n",
+       "      <td id=\"T_14134_row2_col3\" class=\"data row2 col3\" >0.870052</td>\n",
+       "      <td id=\"T_14134_row2_col4\" class=\"data row2 col4\" >0.870157</td>\n",
+       "      <td id=\"T_14134_row2_col5\" class=\"data row2 col5\" >0.871054</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_14134_row3_col0\" class=\"data row3 col0\" >0.817169</td>\n",
+       "      <td id=\"T_14134_row3_col1\" class=\"data row3 col1\" >0.820831</td>\n",
+       "      <td id=\"T_14134_row3_col2\" class=\"data row3 col2\" >0.827709</td>\n",
+       "      <td id=\"T_14134_row3_col3\" class=\"data row3 col3\" >0.817043</td>\n",
+       "      <td id=\"T_14134_row3_col4\" class=\"data row3 col4\" >0.817109</td>\n",
+       "      <td id=\"T_14134_row3_col5\" class=\"data row3 col5\" >0.817560</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_14134_row4_col0\" class=\"data row4 col0\" >0.869744</td>\n",
+       "      <td id=\"T_14134_row4_col1\" class=\"data row4 col1\" >0.869865</td>\n",
+       "      <td id=\"T_14134_row4_col2\" class=\"data row4 col2\" >0.881850</td>\n",
+       "      <td id=\"T_14134_row4_col3\" class=\"data row4 col3\" >0.862166</td>\n",
+       "      <td id=\"T_14134_row4_col4\" class=\"data row4 col4\" >0.862660</td>\n",
+       "      <td id=\"T_14134_row4_col5\" class=\"data row4 col5\" >0.867518</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_14134_row5_col0\" class=\"data row5 col0\" >0.770588</td>\n",
+       "      <td id=\"T_14134_row5_col1\" class=\"data row5 col1\" >0.777031</td>\n",
+       "      <td id=\"T_14134_row5_col2\" class=\"data row5 col2\" >0.779832</td>\n",
+       "      <td id=\"T_14134_row5_col3\" class=\"data row5 col3\" >0.776408</td>\n",
+       "      <td id=\"T_14134_row5_col4\" class=\"data row5 col4\" >0.776128</td>\n",
+       "      <td id=\"T_14134_row5_col5\" class=\"data row5 col5\" >0.773042</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_14134_row6_col0\" class=\"data row6 col0\" >0.907041</td>\n",
+       "      <td id=\"T_14134_row6_col1\" class=\"data row6 col1\" >0.908041</td>\n",
+       "      <td id=\"T_14134_row6_col2\" class=\"data row6 col2\" >0.911930</td>\n",
+       "      <td id=\"T_14134_row6_col3\" class=\"data row6 col3\" >0.906283</td>\n",
+       "      <td id=\"T_14134_row6_col4\" class=\"data row6 col4\" >0.909348</td>\n",
+       "      <td id=\"T_14134_row6_col5\" class=\"data row6 col5\" >0.908648</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_14134_row7_col0\" class=\"data row7 col0\" >-0.105054</td>\n",
+       "      <td id=\"T_14134_row7_col1\" class=\"data row7 col1\" >-0.103463</td>\n",
+       "      <td id=\"T_14134_row7_col2\" class=\"data row7 col2\" >-0.099338</td>\n",
+       "      <td id=\"T_14134_row7_col3\" class=\"data row7 col3\" >-0.104658</td>\n",
+       "      <td id=\"T_14134_row7_col4\" class=\"data row7 col4\" >-0.104459</td>\n",
+       "      <td id=\"T_14134_row7_col5\" class=\"data row7 col5\" >-0.104280</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_14134_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_14134_row8_col0\" class=\"data row8 col0\" >-0.352792</td>\n",
+       "      <td id=\"T_14134_row8_col1\" class=\"data row8 col1\" >-0.348499</td>\n",
+       "      <td id=\"T_14134_row8_col2\" class=\"data row8 col2\" >-0.338605</td>\n",
+       "      <td id=\"T_14134_row8_col3\" class=\"data row8 col3\" >-0.351285</td>\n",
+       "      <td id=\"T_14134_row8_col4\" class=\"data row8 col4\" >-0.350193</td>\n",
+       "      <td id=\"T_14134_row8_col5\" class=\"data row8 col5\" >-0.350152</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a86dd240>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_25121\">\n",
+       "  <caption>XGB Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_25121_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_25121_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_25121_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_25121_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_25121_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_25121_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_25121_row0_col0\" class=\"data row0 col0\" >3.802029</td>\n",
+       "      <td id=\"T_25121_row0_col1\" class=\"data row0 col1\" >3.036764</td>\n",
+       "      <td id=\"T_25121_row0_col2\" class=\"data row0 col2\" >2.979647</td>\n",
+       "      <td id=\"T_25121_row0_col3\" class=\"data row0 col3\" >2.177232</td>\n",
+       "      <td id=\"T_25121_row0_col4\" class=\"data row0 col4\" >2.287098</td>\n",
+       "      <td id=\"T_25121_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_25121_row1_col0\" class=\"data row1 col0\" >0.069013</td>\n",
+       "      <td id=\"T_25121_row1_col1\" class=\"data row1 col1\" >0.071819</td>\n",
+       "      <td id=\"T_25121_row1_col2\" class=\"data row1 col2\" >0.049402</td>\n",
+       "      <td id=\"T_25121_row1_col3\" class=\"data row1 col3\" >0.057279</td>\n",
+       "      <td id=\"T_25121_row1_col4\" class=\"data row1 col4\" >0.050020</td>\n",
+       "      <td id=\"T_25121_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_25121_row2_col0\" class=\"data row2 col0\" >0.860224</td>\n",
+       "      <td id=\"T_25121_row2_col1\" class=\"data row2 col1\" >0.851848</td>\n",
+       "      <td id=\"T_25121_row2_col2\" class=\"data row2 col2\" >0.854136</td>\n",
+       "      <td id=\"T_25121_row2_col3\" class=\"data row2 col3\" >0.853298</td>\n",
+       "      <td id=\"T_25121_row2_col4\" class=\"data row2 col4\" >0.856021</td>\n",
+       "      <td id=\"T_25121_row2_col5\" class=\"data row2 col5\" >0.854344</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_25121_row3_col0\" class=\"data row3 col0\" >0.814145</td>\n",
+       "      <td id=\"T_25121_row3_col1\" class=\"data row3 col1\" >0.804747</td>\n",
+       "      <td id=\"T_25121_row3_col2\" class=\"data row3 col2\" >0.808259</td>\n",
+       "      <td id=\"T_25121_row3_col3\" class=\"data row3 col3\" >0.807370</td>\n",
+       "      <td id=\"T_25121_row3_col4\" class=\"data row3 col4\" >0.810371</td>\n",
+       "      <td id=\"T_25121_row3_col5\" class=\"data row3 col5\" >0.808283</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_25121_row4_col0\" class=\"data row4 col0\" >0.809300</td>\n",
+       "      <td id=\"T_25121_row4_col1\" class=\"data row4 col1\" >0.793038</td>\n",
+       "      <td id=\"T_25121_row4_col2\" class=\"data row4 col2\" >0.794587</td>\n",
+       "      <td id=\"T_25121_row4_col3\" class=\"data row4 col3\" >0.792657</td>\n",
+       "      <td id=\"T_25121_row4_col4\" class=\"data row4 col4\" >0.797936</td>\n",
+       "      <td id=\"T_25121_row4_col5\" class=\"data row4 col5\" >0.795443</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_25121_row5_col0\" class=\"data row5 col0\" >0.819048</td>\n",
+       "      <td id=\"T_25121_row5_col1\" class=\"data row5 col1\" >0.816807</td>\n",
+       "      <td id=\"T_25121_row5_col2\" class=\"data row5 col2\" >0.822409</td>\n",
+       "      <td id=\"T_25121_row5_col3\" class=\"data row5 col3\" >0.822639</td>\n",
+       "      <td id=\"T_25121_row5_col4\" class=\"data row5 col4\" >0.823200</td>\n",
+       "      <td id=\"T_25121_row5_col5\" class=\"data row5 col5\" >0.821545</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_25121_row6_col0\" class=\"data row6 col0\" >0.908407</td>\n",
+       "      <td id=\"T_25121_row6_col1\" class=\"data row6 col1\" >0.906379</td>\n",
+       "      <td id=\"T_25121_row6_col2\" class=\"data row6 col2\" >0.910833</td>\n",
+       "      <td id=\"T_25121_row6_col3\" class=\"data row6 col3\" >0.907507</td>\n",
+       "      <td id=\"T_25121_row6_col4\" class=\"data row6 col4\" >0.908959</td>\n",
+       "      <td id=\"T_25121_row6_col5\" class=\"data row6 col5\" >0.908681</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_25121_row7_col0\" class=\"data row7 col0\" >-0.116893</td>\n",
+       "      <td id=\"T_25121_row7_col1\" class=\"data row7 col1\" >-0.119319</td>\n",
+       "      <td id=\"T_25121_row7_col2\" class=\"data row7 col2\" >-0.116034</td>\n",
+       "      <td id=\"T_25121_row7_col3\" class=\"data row7 col3\" >-0.119313</td>\n",
+       "      <td id=\"T_25121_row7_col4\" class=\"data row7 col4\" >-0.118294</td>\n",
+       "      <td id=\"T_25121_row7_col5\" class=\"data row7 col5\" >-0.118266</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_25121_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_25121_row8_col0\" class=\"data row8 col0\" >-0.393473</td>\n",
+       "      <td id=\"T_25121_row8_col1\" class=\"data row8 col1\" >-0.395306</td>\n",
+       "      <td id=\"T_25121_row8_col2\" class=\"data row8 col2\" >-0.384403</td>\n",
+       "      <td id=\"T_25121_row8_col3\" class=\"data row8 col3\" >-0.397352</td>\n",
+       "      <td id=\"T_25121_row8_col4\" class=\"data row8 col4\" >-0.394224</td>\n",
+       "      <td id=\"T_25121_row8_col5\" class=\"data row8 col5\" >-0.392001</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a8656f80>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# XGB hyperparameter that deals with unbalanced\n",
+    "scale_pos_weight = Y.mean()**-1\n",
+    "\n",
+    "# Creating the model objects\n",
+    "cls_lr = LogisticRegression(\n",
+    "            class_weight=\"balanced\",  # Hyperparameter to deal with unbalanced output\n",
+    "            random_state=lucky_num)\n",
+    "# cls_svm = SVC(random_state=lucky_num)  # Remove due its resource consumption and worst results\n",
+    "cls_NB = GaussianNB()\n",
+    "cls_knn = KNeighborsClassifier()\n",
+    "cls_rf = RandomForestClassifier(\n",
+    "            random_state=lucky_num,\n",
+    "            class_weight=\"balanced_subsample\")  # Hyperparameter to deal with unbalanced output\n",
+    "cls_gbc = GradientBoostingClassifier(random_state=lucky_num)\n",
+    "cls_xgb = xgb.XGBClassifier(\n",
+    "            objective=\"binary:logistic\",\n",
+    "            verbose=None,\n",
+    "            random_state=lucky_num,\n",
+    "            scale_pos_weight = scale_pos_weight)\n",
+    "\n",
+    "# Lists to iterate on our modeling function\n",
+    "cls_name = [\"LR\", \"NB\", \"KNN\", \"RF\", \"GBC\", \"XGB\"]\n",
+    "cls_list = [cls_lr, cls_NB, cls_knn, cls_rf, cls_gbc, cls_xgb]\n",
+    "\n",
+    "mdl_summaries = []\n",
+    "for name, inst in zip(cls_name, cls_list):\n",
+    "    mdl_list = create_model(name, inst)\n",
+    "    mdl_list = [name] + mdl_list\n",
+    "    mdl_summaries.append(mdl_list)\n",
+    "\n",
+    "df_mdl = pd.DataFrame(\n",
+    "            mdl_summaries,\n",
+    "            columns=[\n",
+    "                \"model\",\n",
+    "                \"test_accuracy\",\n",
+    "                \"test_f1\",\n",
+    "                \"test_precision\",\n",
+    "                \"test_recall\",\n",
+    "                \"test_roc_auc\",\n",
+    "                \"test_brier\",\n",
+    "                \"test_log_loss\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_3ba63\">\n",
+       "  <caption>Test set validation scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_3ba63_level0_col0\" class=\"col_heading level0 col0\" >model</th>\n",
+       "      <th id=\"T_3ba63_level0_col1\" class=\"col_heading level0 col1\" >test_accuracy</th>\n",
+       "      <th id=\"T_3ba63_level0_col2\" class=\"col_heading level0 col2\" >test_f1</th>\n",
+       "      <th id=\"T_3ba63_level0_col3\" class=\"col_heading level0 col3\" >test_precision</th>\n",
+       "      <th id=\"T_3ba63_level0_col4\" class=\"col_heading level0 col4\" >test_recall</th>\n",
+       "      <th id=\"T_3ba63_level0_col5\" class=\"col_heading level0 col5\" >test_roc_auc</th>\n",
+       "      <th id=\"T_3ba63_level0_col6\" class=\"col_heading level0 col6\" >test_brier</th>\n",
+       "      <th id=\"T_3ba63_level0_col7\" class=\"col_heading level0 col7\" >test_log_loss</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_3ba63_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
+       "      <td id=\"T_3ba63_row0_col0\" class=\"data row0 col0\" >GBC</td>\n",
+       "      <td id=\"T_3ba63_row0_col1\" class=\"data row0 col1\" >0.871054</td>\n",
+       "      <td id=\"T_3ba63_row0_col2\" class=\"data row0 col2\" >0.817560</td>\n",
+       "      <td id=\"T_3ba63_row0_col3\" class=\"data row0 col3\" >0.867518</td>\n",
+       "      <td id=\"T_3ba63_row0_col4\" class=\"data row0 col4\" >0.773042</td>\n",
+       "      <td id=\"T_3ba63_row0_col5\" class=\"data row0 col5\" >0.908648</td>\n",
+       "      <td id=\"T_3ba63_row0_col6\" class=\"data row0 col6\" >0.104280</td>\n",
+       "      <td id=\"T_3ba63_row0_col7\" class=\"data row0 col7\" >0.350152</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_3ba63_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
+       "      <td id=\"T_3ba63_row1_col0\" class=\"data row1 col0\" >LR</td>\n",
+       "      <td id=\"T_3ba63_row1_col1\" class=\"data row1 col1\" >0.865924</td>\n",
+       "      <td id=\"T_3ba63_row1_col2\" class=\"data row1 col2\" >0.814469</td>\n",
+       "      <td id=\"T_3ba63_row1_col3\" class=\"data row1 col3\" >0.843439</td>\n",
+       "      <td id=\"T_3ba63_row1_col4\" class=\"data row1 col4\" >0.787423</td>\n",
+       "      <td id=\"T_3ba63_row1_col5\" class=\"data row1 col5\" >0.904458</td>\n",
+       "      <td id=\"T_3ba63_row1_col6\" class=\"data row1 col6\" >0.110435</td>\n",
+       "      <td id=\"T_3ba63_row1_col7\" class=\"data row1 col7\" >0.370350</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_3ba63_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
+       "      <td id=\"T_3ba63_row2_col0\" class=\"data row2 col0\" >XGB</td>\n",
+       "      <td id=\"T_3ba63_row2_col1\" class=\"data row2 col1\" >0.854344</td>\n",
+       "      <td id=\"T_3ba63_row2_col2\" class=\"data row2 col2\" >0.808283</td>\n",
+       "      <td id=\"T_3ba63_row2_col3\" class=\"data row2 col3\" >0.795443</td>\n",
+       "      <td id=\"T_3ba63_row2_col4\" class=\"data row2 col4\" >0.821545</td>\n",
+       "      <td id=\"T_3ba63_row2_col5\" class=\"data row2 col5\" >0.908681</td>\n",
+       "      <td id=\"T_3ba63_row2_col6\" class=\"data row2 col6\" >0.118266</td>\n",
+       "      <td id=\"T_3ba63_row2_col7\" class=\"data row2 col7\" >0.392001</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_3ba63_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
+       "      <td id=\"T_3ba63_row3_col0\" class=\"data row3 col0\" >RF</td>\n",
+       "      <td id=\"T_3ba63_row3_col1\" class=\"data row3 col1\" >0.856152</td>\n",
+       "      <td id=\"T_3ba63_row3_col2\" class=\"data row3 col2\" >0.800623</td>\n",
+       "      <td id=\"T_3ba63_row3_col3\" class=\"data row3 col3\" >0.830547</td>\n",
+       "      <td id=\"T_3ba63_row3_col4\" class=\"data row3 col4\" >0.772781</td>\n",
+       "      <td id=\"T_3ba63_row3_col5\" class=\"data row3 col5\" >0.893466</td>\n",
+       "      <td id=\"T_3ba63_row3_col6\" class=\"data row3 col6\" >0.115285</td>\n",
+       "      <td id=\"T_3ba63_row3_col7\" class=\"data row3 col7\" >0.562042</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_3ba63_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
+       "      <td id=\"T_3ba63_row4_col0\" class=\"data row4 col0\" >KNN</td>\n",
+       "      <td id=\"T_3ba63_row4_col1\" class=\"data row4 col1\" >0.843692</td>\n",
+       "      <td id=\"T_3ba63_row4_col2\" class=\"data row4 col2\" >0.779698</td>\n",
+       "      <td id=\"T_3ba63_row4_col3\" class=\"data row4 col3\" >0.823778</td>\n",
+       "      <td id=\"T_3ba63_row4_col4\" class=\"data row4 col4\" >0.740097</td>\n",
+       "      <td id=\"T_3ba63_row4_col5\" class=\"data row4 col5\" >0.872155</td>\n",
+       "      <td id=\"T_3ba63_row4_col6\" class=\"data row4 col6\" >0.128215</td>\n",
+       "      <td id=\"T_3ba63_row4_col7\" class=\"data row4 col7\" >1.877810</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_3ba63_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
+       "      <td id=\"T_3ba63_row5_col0\" class=\"data row5 col0\" >NB</td>\n",
+       "      <td id=\"T_3ba63_row5_col1\" class=\"data row5 col1\" >0.766637</td>\n",
+       "      <td id=\"T_3ba63_row5_col2\" class=\"data row5 col2\" >0.664795</td>\n",
+       "      <td id=\"T_3ba63_row5_col3\" class=\"data row5 col3\" >0.717684</td>\n",
+       "      <td id=\"T_3ba63_row5_col4\" class=\"data row5 col4\" >0.619166</td>\n",
+       "      <td id=\"T_3ba63_row5_col5\" class=\"data row5 col5\" >0.848834</td>\n",
+       "      <td id=\"T_3ba63_row5_col6\" class=\"data row5 col6\" >0.208278</td>\n",
+       "      <td id=\"T_3ba63_row5_col7\" class=\"data row5 col7\" >1.761326</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1a8656bc0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "df_mdl.sort_values(\n",
+    "        \"test_f1\",\n",
+    "        ascending=False,\n",
+    "        inplace=True,\n",
+    "        ignore_index=True)\n",
+    "\n",
+    "display(df_mdl.style.set_caption(\"Test set validation scores\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "GBC, LR, XGB and RF preset great results! We have two ways here: hyperparameters tunning or creating a composite model. Let's begin with the composite model.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_04769\">\n",
+       "  <caption>Test set validation scores for Composite Model</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_04769_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_04769_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_04769_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_04769_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_04769_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_04769_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_04769_row0_col0\" class=\"data row0 col0\" >10.109613</td>\n",
+       "      <td id=\"T_04769_row0_col1\" class=\"data row0 col1\" >11.766011</td>\n",
+       "      <td id=\"T_04769_row0_col2\" class=\"data row0 col2\" >11.450818</td>\n",
+       "      <td id=\"T_04769_row0_col3\" class=\"data row0 col3\" >11.737634</td>\n",
+       "      <td id=\"T_04769_row0_col4\" class=\"data row0 col4\" >12.702598</td>\n",
+       "      <td id=\"T_04769_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_04769_row1_col0\" class=\"data row1 col0\" >0.490518</td>\n",
+       "      <td id=\"T_04769_row1_col1\" class=\"data row1 col1\" >0.532695</td>\n",
+       "      <td id=\"T_04769_row1_col2\" class=\"data row1 col2\" >0.529459</td>\n",
+       "      <td id=\"T_04769_row1_col3\" class=\"data row1 col3\" >0.549051</td>\n",
+       "      <td id=\"T_04769_row1_col4\" class=\"data row1 col4\" >0.586749</td>\n",
+       "      <td id=\"T_04769_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_04769_row2_col0\" class=\"data row2 col0\" >0.870799</td>\n",
+       "      <td id=\"T_04769_row2_col1\" class=\"data row2 col1\" >0.871532</td>\n",
+       "      <td id=\"T_04769_row2_col2\" class=\"data row2 col2\" >0.875497</td>\n",
+       "      <td id=\"T_04769_row2_col3\" class=\"data row2 col3\" >0.869948</td>\n",
+       "      <td id=\"T_04769_row2_col4\" class=\"data row2 col4\" >0.869215</td>\n",
+       "      <td id=\"T_04769_row2_col5\" class=\"data row2 col5\" >0.869002</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_04769_row3_col0\" class=\"data row3 col0\" >0.818689</td>\n",
+       "      <td id=\"T_04769_row3_col1\" class=\"data row3 col1\" >0.820797</td>\n",
+       "      <td id=\"T_04769_row3_col2\" class=\"data row3 col2\" >0.826297</td>\n",
+       "      <td id=\"T_04769_row3_col3\" class=\"data row3 col3\" >0.819319</td>\n",
+       "      <td id=\"T_04769_row3_col4\" class=\"data row3 col4\" >0.817531</td>\n",
+       "      <td id=\"T_04769_row3_col5\" class=\"data row3 col5\" >0.817283</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_04769_row4_col0\" class=\"data row4 col0\" >0.860939</td>\n",
+       "      <td id=\"T_04769_row4_col1\" class=\"data row4 col1\" >0.857492</td>\n",
+       "      <td id=\"T_04769_row4_col2\" class=\"data row4 col2\" >0.863511</td>\n",
+       "      <td id=\"T_04769_row4_col3\" class=\"data row4 col3\" >0.852042</td>\n",
+       "      <td id=\"T_04769_row4_col4\" class=\"data row4 col4\" >0.854090</td>\n",
+       "      <td id=\"T_04769_row4_col5\" class=\"data row4 col5\" >0.853645</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_04769_row5_col0\" class=\"data row5 col0\" >0.780392</td>\n",
+       "      <td id=\"T_04769_row5_col1\" class=\"data row5 col1\" >0.787115</td>\n",
+       "      <td id=\"T_04769_row5_col2\" class=\"data row5 col2\" >0.792157</td>\n",
+       "      <td id=\"T_04769_row5_col3\" class=\"data row5 col3\" >0.789017</td>\n",
+       "      <td id=\"T_04769_row5_col4\" class=\"data row5 col4\" >0.783973</td>\n",
+       "      <td id=\"T_04769_row5_col5\" class=\"data row5 col5\" >0.783893</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_04769_row6_col0\" class=\"data row6 col0\" >0.909022</td>\n",
+       "      <td id=\"T_04769_row6_col1\" class=\"data row6 col1\" >0.908890</td>\n",
+       "      <td id=\"T_04769_row6_col2\" class=\"data row6 col2\" >0.912418</td>\n",
+       "      <td id=\"T_04769_row6_col3\" class=\"data row6 col3\" >0.907315</td>\n",
+       "      <td id=\"T_04769_row6_col4\" class=\"data row6 col4\" >0.910340</td>\n",
+       "      <td id=\"T_04769_row6_col5\" class=\"data row6 col5\" >0.910500</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_04769_row7_col0\" class=\"data row7 col0\" >-0.105818</td>\n",
+       "      <td id=\"T_04769_row7_col1\" class=\"data row7 col1\" >-0.105354</td>\n",
+       "      <td id=\"T_04769_row7_col2\" class=\"data row7 col2\" >-0.101743</td>\n",
+       "      <td id=\"T_04769_row7_col3\" class=\"data row7 col3\" >-0.106567</td>\n",
+       "      <td id=\"T_04769_row7_col4\" class=\"data row7 col4\" >-0.105957</td>\n",
+       "      <td id=\"T_04769_row7_col5\" class=\"data row7 col5\" >-0.105743</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_04769_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_04769_row8_col0\" class=\"data row8 col0\" >-0.356051</td>\n",
+       "      <td id=\"T_04769_row8_col1\" class=\"data row8 col1\" >-0.353269</td>\n",
+       "      <td id=\"T_04769_row8_col2\" class=\"data row8 col2\" >-0.344184</td>\n",
+       "      <td id=\"T_04769_row8_col3\" class=\"data row8 col3\" >-0.357062</td>\n",
+       "      <td id=\"T_04769_row8_col4\" class=\"data row8 col4\" >-0.355010</td>\n",
+       "      <td id=\"T_04769_row8_col5\" class=\"data row8 col5\" >-0.353621</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x1a1859d3520>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Selecting the models\n",
+    "cls_name = [\"GBC\", \"XGB\", \"LR\", \"RF\",]\n",
+    "cls_list = [cls_gbc, cls_xgb, cls_lr, cls_rf]\n",
+    "\n",
+    "# Training the voting classifier\n",
+    "cls_vot = VotingClassifier([*zip(cls_name, cls_list)], voting=\"soft\")\n",
+    "cls_vot.fit(X_train, y_train)\n",
+    "\n",
+    "# Using cross-validation to evaluate the model fitted\n",
+    "cls_cross = cross_validate(\n",
+    "    estimator=cls_vot,\n",
+    "    X=X_train,\n",
+    "    y=y_train,\n",
+    "    cv=5,\n",
+    "    scoring=scores)\n",
+    "\n",
+    "df_vot = pd.DataFrame.from_dict(cls_cross, orient='index', columns=[\"CV\"+str(i) for i in range(1,6)])\n",
+    "\n",
+    "# Calculating score to test set\n",
+    "accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value = eval_model(cls_vot)\n",
+    "\n",
+    "# Filling a dataframe to better presentation\n",
+    "df_vot.at[\"test_accuracy\", \"TestSet\"] = accurancy\n",
+    "df_vot.at[\"test_f1\", \"TestSet\"] = f1\n",
+    "df_vot.at[\"test_recall\", \"TestSet\"] = recall\n",
+    "df_vot.at[\"test_precision\", \"TestSet\"] = precision\n",
+    "df_vot.at[\"test_roc_auc\", \"TestSet\"] = roc_auc\n",
+    "df_vot.at[\"test_neg_brier_score\", \"TestSet\"] = -brier_score\n",
+    "df_vot.at[\"test_neg_log_loss\", \"TestSet\"] = -log_loss_value\n",
+    "\n",
+    "display(df_vot.style.set_caption(\"Test set validation scores for Composite Model\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The composite model does not present any evidence of overfitting. For now, we will use it on our app."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['c:\\\\Users\\\\grego\\\\OneDrive\\\\Documentos\\\\Documentos Pessoais\\\\00_DataCamp\\\\09_VSC\\\\poa_car_accidents\\\\poa_car_accidents\\\\model\\\\model_feridos.pkl']"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Saving\n",
+    "file_name = \"model_\" + output + '.pkl'\n",
+    "jb.dump(cls_vot, path.join(path.abspath(\"./\"), file_name))"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.10.6 64-bit",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.6"
+  },
+  "orig_nbformat": 4,
+  "vscode": {
+   "interpreter": {
+    "hash": "1372d04dbd71fdc5436c5d6e671c1b9287e750e86143c81b5a7ba0acaf653c5e"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

model/model_feridos.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee61d7ec89de8ff0cf49904823a9a892556807449852976ede436fa059ed765a
+size 248858704

model/model_feridos_gr.ipynb

@@ -0,0 +1,1659 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 1. Introduction\n",
+    "\n",
+    "This notebook was written to train Porto Alegre Traffic Accidents Data after the first cleaning, processing, and transforming step. This was made in a notebook in the `data` folder. In truth, we will have 3 models.\n",
+    "\n",
+    "1. Predict the probability of injured people.\n",
+    "\n",
+    "2. Predict the probability of seriously injured people.\n",
+    "\n",
+    "3. Predict the probability of dead people in the event or after it.\n",
+    "\n",
+    "The path to training the models will be the same, just make some filtering on data and analyze the results properly."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 2. Data Loading"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>0</th>\n",
+       "      <th>1</th>\n",
+       "      <th>2</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>latitude</th>\n",
+       "      <td>-30.009614</td>\n",
+       "      <td>-30.0403</td>\n",
+       "      <td>-30.069</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>longitude</th>\n",
+       "      <td>-51.185581</td>\n",
+       "      <td>-51.1958</td>\n",
+       "      <td>-51.1437</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>feridos</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>feridos_gr</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>fatais</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>caminhao</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>moto</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>cars</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>transport</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>others</th>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>holiday</th>\n",
+       "      <td>False</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_2</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_3</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_4</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_5</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>day_6</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_2</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_3</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_4</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_5</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_6</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_7</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_8</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_9</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_10</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_11</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_12</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_13</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_14</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_15</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_16</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_17</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_18</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_19</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_20</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_21</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_22</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>hour_23</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_ATROPELAMENTO</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_CHOQUE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_COLISÃO</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>type_OUTROS</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                            0        1        2\n",
+       "latitude           -30.009614 -30.0403  -30.069\n",
+       "longitude          -51.185581 -51.1958 -51.1437\n",
+       "feridos                  True     True     True\n",
+       "feridos_gr              False    False    False\n",
+       "fatais                  False    False    False\n",
+       "caminhao                False    False    False\n",
+       "moto                     True     True    False\n",
+       "cars                     True     True     True\n",
+       "transport               False    False    False\n",
+       "others                  False    False    False\n",
+       "holiday                 False     True     True\n",
+       "day_1                       0        0        0\n",
+       "day_2                       0        0        0\n",
+       "day_3                       0        0        0\n",
+       "day_4                       0        0        0\n",
+       "day_5                       1        0        0\n",
+       "day_6                       0        1        1\n",
+       "hour_1                      0        0        0\n",
+       "hour_2                      0        0        0\n",
+       "hour_3                      0        0        0\n",
+       "hour_4                      0        0        0\n",
+       "hour_5                      0        0        0\n",
+       "hour_6                      0        0        0\n",
+       "hour_7                      0        0        0\n",
+       "hour_8                      0        0        0\n",
+       "hour_9                      0        0        0\n",
+       "hour_10                     0        1        0\n",
+       "hour_11                     0        0        0\n",
+       "hour_12                     0        0        0\n",
+       "hour_13                     0        0        0\n",
+       "hour_14                     0        0        0\n",
+       "hour_15                     0        0        0\n",
+       "hour_16                     0        0        0\n",
+       "hour_17                     0        0        0\n",
+       "hour_18                     0        0        0\n",
+       "hour_19                     1        0        1\n",
+       "hour_20                     0        0        0\n",
+       "hour_21                     0        0        0\n",
+       "hour_22                     0        0        0\n",
+       "hour_23                     0        0        0\n",
+       "type_ATROPELAMENTO          0        0        1\n",
+       "type_CHOQUE                 0        0        0\n",
+       "type_COLISÃO                0        0        0\n",
+       "type_OUTROS                 0        0        0"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import os.path as path\n",
+    "from pandas import read_csv\n",
+    "\n",
+    "file_csv =  path.abspath(\"../\")\n",
+    "\n",
+    "file_csv = path.join(file_csv, \"data\" ,\"accidents_trans.csv\")\n",
+    "\n",
+    "accidents_trans = read_csv(file_csv)\n",
+    "\n",
+    "accidents_trans.head(3).T"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 3. Data Preparation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import joblib as jb # Use to save the model to deploy\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "from sklearn.model_selection import train_test_split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Our model to predict the probability of feridos_gr will be create with 25497 rows and 41 features.\n"
+     ]
+    }
+   ],
+   "source": [
+    "outputs = [\"feridos\", \"feridos_gr\", \"fatais\"]\n",
+    "inputs = [col for col in accidents_trans.columns if col not in outputs]\n",
+    "\n",
+    "X = accidents_trans[inputs].copy()\n",
+    "Y = accidents_trans[outputs].copy()\n",
+    "\n",
+    "# Filtering data considering the output\n",
+    "output = \"feridos_gr\"\n",
+    "\n",
+    "if output == \"feridos_gr\":\n",
+    "    X = X[Y[\"feridos\"]]\n",
+    "    Y = Y.loc[Y[\"feridos\"], \"feridos_gr\"]\n",
+    "elif output == \"fatais\":\n",
+    "    X = X[Y[\"feridos_gr\"]]\n",
+    "    Y = Y.loc[Y[\"feridos_gr\"], \"fatais\"]\n",
+    "else:\n",
+    "    Y = Y[\"feridos\"]\n",
+    "\n",
+    "print(f\"Our model to predict the probability of \" \\\n",
+    "      f\"{output} will be create with {X.shape[0]} \" \\\n",
+    "      f\"rows and {X.shape[1]} features.\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import csv\n",
+    "\n",
+    "with open(\"model_features.csv\", 'w') as f:\n",
+    "    writer = csv.writer(f)\n",
+    "    writer.writerow(X.columns)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Considering that we will use models scaling sensitive, we will need to scale our data first. Beside this, we will need to save our scaler for future use."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['c:\\\\Users\\\\grego\\\\OneDrive\\\\Documentos\\\\Documentos Pessoais\\\\00_DataCamp\\\\09_VSC\\\\poa_car_accidents\\\\poa_car_accidents\\\\model\\\\scaler_feridos_gr.pkl']"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Setting the random state using my luck number :-)\n",
+    "lucky_num = 7\n",
+    "\n",
+    "# X_train and y_train to train our model\n",
+    "X_train, X_test, y_train, y_test = train_test_split(\n",
+    "    X,\n",
+    "    Y,\n",
+    "    test_size=0.30,\n",
+    "    random_state=lucky_num,\n",
+    "    shuffle=True,  # Used because our data is sort by date\n",
+    "    stratify=Y)  # Used because our data is unbalanced\n",
+    "\n",
+    "# Scaling\n",
+    "scaler = StandardScaler()\n",
+    "X_train = scaler.fit_transform(X_train)\n",
+    "X_test = scaler.transform(X_test)\n",
+    "\n",
+    "# Saving scaler\n",
+    "file_name = \"scaler_\" + output + '.pkl'\n",
+    "jb.dump(scaler, path.join(path.abspath(\"./\"), file_name))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 4. Data Modeling\n",
+    "\n",
+    "We will create and use cross-validation to evaluate the following models:\n",
+    "\n",
+    "- Logistic Regression;\n",
+    "\n",
+    "- Gaussian Naive Bayes;\n",
+    "\n",
+    "- K Neighbors;\n",
+    "\n",
+    "- Random Forest;\n",
+    "\n",
+    "- Gradient Boosting; and,\n",
+    "\n",
+    "- XGBoost.\n",
+    "\n",
+    "We will use two scores to select and evaluate our models:\n",
+    "\n",
+    "- F1 score: composition between the precision (how much our model correct classify every true label) and recall (how moch our model correct indicate true labels); and,\n",
+    "\n",
+    "- Brier score: average between the correct and the predict probability.\n",
+    "\n",
+    "However, we will see other metrics to support our decision:\n",
+    "\n",
+    "- Accurancy;\n",
+    "\n",
+    "- ROC_AOC; and,\n",
+    "\n",
+    "- Log loss (an other way to quantify the quality of probability predictions).\n",
+    "\n",
+    "And, before you go, we will find for each model if there is a hyperparameter to deal with the unbalanced output."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import xgboost as xgb\n",
+    "from sklearn.naive_bayes import GaussianNB\n",
+    "from sklearn.neighbors import KNeighborsClassifier\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.model_selection import cross_validate \n",
+    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier, VotingClassifier\n",
+    "from sklearn.metrics import accuracy_score, recall_score, precision_score, roc_auc_score, f1_score, brier_score_loss, log_loss\n",
+    "\n",
+    "scores = [\"accuracy\", \"f1\", \"precision\", \"recall\", \"roc_auc\", \"neg_brier_score\",\"neg_log_loss\"]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def eval_model(cls) -> tuple:\n",
+    "    \"\"\"This function will calculate the metrics\n",
+    "    to evaluate a classification model.\n",
+    "    \"\"\"\n",
+    "    # Predicting labels and probabilities\n",
+    "    y_pred = cls.predict(X_test)\n",
+    "    y_prob = cls.predict_proba(X_test)[:,1]\n",
+    "\n",
+    "    # Calculating scores\n",
+    "    accurancy = accuracy_score(y_test, y_pred)\n",
+    "    f1 = f1_score(y_test, y_pred)\n",
+    "    recall = recall_score(y_test, y_pred)\n",
+    "    precision = precision_score(y_test, y_pred)\n",
+    "    roc_auc = roc_auc_score(y_test, y_prob)  # https://datascience.stackexchange.com/questions/114394/does-roc-auc-different-between-crossval-and-test-set-indicate-overfitting-or-oth\n",
+    "    brier_score = brier_score_loss(y_test, y_prob)\n",
+    "    log_loss_value = log_loss(y_test, y_prob)\n",
+    "\n",
+    "    return accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value\n",
+    "\n",
+    "def create_model(name: str, cls) -> list:\n",
+    "    \"\"\"This function will create some models\n",
+    "    and return scores to evaluate it.\"\"\"\n",
+    "    # Ftting model\n",
+    "    cls.fit(X_train, y_train)\n",
+    "\n",
+    "    # Using cross-validation to evaluate the model fitted\n",
+    "    cls_cross = cross_validate(\n",
+    "        estimator=cls,\n",
+    "        X=X_train,\n",
+    "        y=y_train,\n",
+    "        cv=5,\n",
+    "        scoring=scores)\n",
+    "\n",
+    "    df_cv = pd.DataFrame.from_dict(cls_cross, orient='index', columns=[\"CV\"+str(i) for i in range(1,6)])\n",
+    "\n",
+    "    # Calculating score to test set\n",
+    "    accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value = eval_model(cls)\n",
+    "\n",
+    "    # Filling a dataframe to better presentation\n",
+    "    df_cv.at[\"test_accuracy\", \"TestSet\"] = accurancy\n",
+    "    df_cv.at[\"test_f1\", \"TestSet\"] = f1\n",
+    "    df_cv.at[\"test_recall\", \"TestSet\"] = recall\n",
+    "    df_cv.at[\"test_precision\", \"TestSet\"] = precision\n",
+    "    df_cv.at[\"test_roc_auc\", \"TestSet\"] = roc_auc\n",
+    "    df_cv.at[\"test_neg_brier_score\", \"TestSet\"] = -brier_score\n",
+    "    df_cv.at[\"test_neg_log_loss\", \"TestSet\"] = -log_loss_value\n",
+    "\n",
+    "    caption = f\"{name} Validation Scores\"\n",
+    "\n",
+    "    display(df_cv.style.set_caption(caption))\n",
+    "\n",
+    "    return [accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_440e0\">\n",
+       "  <caption>LR Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_440e0_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_440e0_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_440e0_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_440e0_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_440e0_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_440e0_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_440e0_row0_col0\" class=\"data row0 col0\" >0.037210</td>\n",
+       "      <td id=\"T_440e0_row0_col1\" class=\"data row0 col1\" >0.031787</td>\n",
+       "      <td id=\"T_440e0_row0_col2\" class=\"data row0 col2\" >0.031054</td>\n",
+       "      <td id=\"T_440e0_row0_col3\" class=\"data row0 col3\" >0.030418</td>\n",
+       "      <td id=\"T_440e0_row0_col4\" class=\"data row0 col4\" >0.031932</td>\n",
+       "      <td id=\"T_440e0_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_440e0_row1_col0\" class=\"data row1 col0\" >0.000000</td>\n",
+       "      <td id=\"T_440e0_row1_col1\" class=\"data row1 col1\" >0.008581</td>\n",
+       "      <td id=\"T_440e0_row1_col2\" class=\"data row1 col2\" >0.008867</td>\n",
+       "      <td id=\"T_440e0_row1_col3\" class=\"data row1 col3\" >0.009067</td>\n",
+       "      <td id=\"T_440e0_row1_col4\" class=\"data row1 col4\" >0.008065</td>\n",
+       "      <td id=\"T_440e0_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_440e0_row2_col0\" class=\"data row2 col0\" >0.676751</td>\n",
+       "      <td id=\"T_440e0_row2_col1\" class=\"data row2 col1\" >0.650420</td>\n",
+       "      <td id=\"T_440e0_row2_col2\" class=\"data row2 col2\" >0.651723</td>\n",
+       "      <td id=\"T_440e0_row2_col3\" class=\"data row2 col3\" >0.662370</td>\n",
+       "      <td id=\"T_440e0_row2_col4\" class=\"data row2 col4\" >0.641636</td>\n",
+       "      <td id=\"T_440e0_row2_col5\" class=\"data row2 col5\" >0.660523</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_440e0_row3_col0\" class=\"data row3 col0\" >0.437622</td>\n",
+       "      <td id=\"T_440e0_row3_col1\" class=\"data row3 col1\" >0.422222</td>\n",
+       "      <td id=\"T_440e0_row3_col2\" class=\"data row3 col2\" >0.413403</td>\n",
+       "      <td id=\"T_440e0_row3_col3\" class=\"data row3 col3\" >0.424271</td>\n",
+       "      <td id=\"T_440e0_row3_col4\" class=\"data row3 col4\" >0.401497</td>\n",
+       "      <td id=\"T_440e0_row3_col5\" class=\"data row3 col5\" >0.414431</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_440e0_row4_col0\" class=\"data row4 col0\" >0.352433</td>\n",
+       "      <td id=\"T_440e0_row4_col1\" class=\"data row4 col1\" >0.329957</td>\n",
+       "      <td id=\"T_440e0_row4_col2\" class=\"data row4 col2\" >0.326379</td>\n",
+       "      <td id=\"T_440e0_row4_col3\" class=\"data row4 col3\" >0.337386</td>\n",
+       "      <td id=\"T_440e0_row4_col4\" class=\"data row4 col4\" >0.315673</td>\n",
+       "      <td id=\"T_440e0_row4_col5\" class=\"data row4 col5\" >0.331889</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_440e0_row5_col0\" class=\"data row5 col0\" >0.577121</td>\n",
+       "      <td id=\"T_440e0_row5_col1\" class=\"data row5 col1\" >0.586118</td>\n",
+       "      <td id=\"T_440e0_row5_col2\" class=\"data row5 col2\" >0.563707</td>\n",
+       "      <td id=\"T_440e0_row5_col3\" class=\"data row5 col3\" >0.571429</td>\n",
+       "      <td id=\"T_440e0_row5_col4\" class=\"data row5 col4\" >0.551414</td>\n",
+       "      <td id=\"T_440e0_row5_col5\" class=\"data row5 col5\" >0.551621</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_440e0_row6_col0\" class=\"data row6 col0\" >0.688461</td>\n",
+       "      <td id=\"T_440e0_row6_col1\" class=\"data row6 col1\" >0.667854</td>\n",
+       "      <td id=\"T_440e0_row6_col2\" class=\"data row6 col2\" >0.663118</td>\n",
+       "      <td id=\"T_440e0_row6_col3\" class=\"data row6 col3\" >0.671007</td>\n",
+       "      <td id=\"T_440e0_row6_col4\" class=\"data row6 col4\" >0.644010</td>\n",
+       "      <td id=\"T_440e0_row6_col5\" class=\"data row6 col5\" >0.667540</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_440e0_row7_col0\" class=\"data row7 col0\" >-0.221720</td>\n",
+       "      <td id=\"T_440e0_row7_col1\" class=\"data row7 col1\" >-0.228903</td>\n",
+       "      <td id=\"T_440e0_row7_col2\" class=\"data row7 col2\" >-0.228032</td>\n",
+       "      <td id=\"T_440e0_row7_col3\" class=\"data row7 col3\" >-0.226401</td>\n",
+       "      <td id=\"T_440e0_row7_col4\" class=\"data row7 col4\" >-0.231747</td>\n",
+       "      <td id=\"T_440e0_row7_col5\" class=\"data row7 col5\" >-0.225709</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_440e0_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_440e0_row8_col0\" class=\"data row8 col0\" >-0.635799</td>\n",
+       "      <td id=\"T_440e0_row8_col1\" class=\"data row8 col1\" >-0.651183</td>\n",
+       "      <td id=\"T_440e0_row8_col2\" class=\"data row8 col2\" >-0.649342</td>\n",
+       "      <td id=\"T_440e0_row8_col3\" class=\"data row8 col3\" >-0.646086</td>\n",
+       "      <td id=\"T_440e0_row8_col4\" class=\"data row8 col4\" >-0.657217</td>\n",
+       "      <td id=\"T_440e0_row8_col5\" class=\"data row8 col5\" >-0.644816</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e08e87ca0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_0ee79\">\n",
+       "  <caption>NB Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_0ee79_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_0ee79_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_0ee79_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_0ee79_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_0ee79_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_0ee79_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_0ee79_row0_col0\" class=\"data row0 col0\" >0.014398</td>\n",
+       "      <td id=\"T_0ee79_row0_col1\" class=\"data row0 col1\" >0.009557</td>\n",
+       "      <td id=\"T_0ee79_row0_col2\" class=\"data row0 col2\" >0.006821</td>\n",
+       "      <td id=\"T_0ee79_row0_col3\" class=\"data row0 col3\" >0.007424</td>\n",
+       "      <td id=\"T_0ee79_row0_col4\" class=\"data row0 col4\" >0.012280</td>\n",
+       "      <td id=\"T_0ee79_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_0ee79_row1_col0\" class=\"data row1 col0\" >0.011809</td>\n",
+       "      <td id=\"T_0ee79_row1_col1\" class=\"data row1 col1\" >0.011134</td>\n",
+       "      <td id=\"T_0ee79_row1_col2\" class=\"data row1 col2\" >0.017092</td>\n",
+       "      <td id=\"T_0ee79_row1_col3\" class=\"data row1 col3\" >0.011901</td>\n",
+       "      <td id=\"T_0ee79_row1_col4\" class=\"data row1 col4\" >0.011843</td>\n",
+       "      <td id=\"T_0ee79_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_0ee79_row2_col0\" class=\"data row2 col0\" >0.713725</td>\n",
+       "      <td id=\"T_0ee79_row2_col1\" class=\"data row2 col1\" >0.696639</td>\n",
+       "      <td id=\"T_0ee79_row2_col2\" class=\"data row2 col2\" >0.695433</td>\n",
+       "      <td id=\"T_0ee79_row2_col3\" class=\"data row2 col3\" >0.695153</td>\n",
+       "      <td id=\"T_0ee79_row2_col4\" class=\"data row2 col4\" >0.678902</td>\n",
+       "      <td id=\"T_0ee79_row2_col5\" class=\"data row2 col5\" >0.699608</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_0ee79_row3_col0\" class=\"data row3 col0\" >0.398115</td>\n",
+       "      <td id=\"T_0ee79_row3_col1\" class=\"data row3 col1\" >0.385706</td>\n",
+       "      <td id=\"T_0ee79_row3_col2\" class=\"data row3 col2\" >0.386222</td>\n",
+       "      <td id=\"T_0ee79_row3_col3\" class=\"data row3 col3\" >0.373993</td>\n",
+       "      <td id=\"T_0ee79_row3_col4\" class=\"data row3 col4\" >0.358343</td>\n",
+       "      <td id=\"T_0ee79_row3_col5\" class=\"data row3 col5\" >0.381260</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_0ee79_row4_col0\" class=\"data row4 col0\" >0.367391</td>\n",
+       "      <td id=\"T_0ee79_row4_col1\" class=\"data row4 col1\" >0.345178</td>\n",
+       "      <td id=\"T_0ee79_row4_col2\" class=\"data row4 col2\" >0.344064</td>\n",
+       "      <td id=\"T_0ee79_row4_col3\" class=\"data row4 col3\" >0.338189</td>\n",
+       "      <td id=\"T_0ee79_row4_col4\" class=\"data row4 col4\" >0.317460</td>\n",
+       "      <td id=\"T_0ee79_row4_col5\" class=\"data row4 col5\" >0.345703</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_0ee79_row5_col0\" class=\"data row5 col0\" >0.434447</td>\n",
+       "      <td id=\"T_0ee79_row5_col1\" class=\"data row5 col1\" >0.437018</td>\n",
+       "      <td id=\"T_0ee79_row5_col2\" class=\"data row5 col2\" >0.440154</td>\n",
+       "      <td id=\"T_0ee79_row5_col3\" class=\"data row5 col3\" >0.418275</td>\n",
+       "      <td id=\"T_0ee79_row5_col4\" class=\"data row5 col4\" >0.411311</td>\n",
+       "      <td id=\"T_0ee79_row5_col5\" class=\"data row5 col5\" >0.424970</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_0ee79_row6_col0\" class=\"data row6 col0\" >0.658495</td>\n",
+       "      <td id=\"T_0ee79_row6_col1\" class=\"data row6 col1\" >0.637091</td>\n",
+       "      <td id=\"T_0ee79_row6_col2\" class=\"data row6 col2\" >0.633805</td>\n",
+       "      <td id=\"T_0ee79_row6_col3\" class=\"data row6 col3\" >0.633238</td>\n",
+       "      <td id=\"T_0ee79_row6_col4\" class=\"data row6 col4\" >0.609990</td>\n",
+       "      <td id=\"T_0ee79_row6_col5\" class=\"data row6 col5\" >0.634031</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_0ee79_row7_col0\" class=\"data row7 col0\" >-0.251232</td>\n",
+       "      <td id=\"T_0ee79_row7_col1\" class=\"data row7 col1\" >-0.260890</td>\n",
+       "      <td id=\"T_0ee79_row7_col2\" class=\"data row7 col2\" >-0.271004</td>\n",
+       "      <td id=\"T_0ee79_row7_col3\" class=\"data row7 col3\" >-0.273609</td>\n",
+       "      <td id=\"T_0ee79_row7_col4\" class=\"data row7 col4\" >-0.285054</td>\n",
+       "      <td id=\"T_0ee79_row7_col5\" class=\"data row7 col5\" >-0.268468</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0ee79_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_0ee79_row8_col0\" class=\"data row8 col0\" >-1.412893</td>\n",
+       "      <td id=\"T_0ee79_row8_col1\" class=\"data row8 col1\" >-1.627295</td>\n",
+       "      <td id=\"T_0ee79_row8_col2\" class=\"data row8 col2\" >-1.745289</td>\n",
+       "      <td id=\"T_0ee79_row8_col3\" class=\"data row8 col3\" >-1.752608</td>\n",
+       "      <td id=\"T_0ee79_row8_col4\" class=\"data row8 col4\" >-1.950351</td>\n",
+       "      <td id=\"T_0ee79_row8_col5\" class=\"data row8 col5\" >-1.659029</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e09deee00>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_4933a\">\n",
+       "  <caption>KNN Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_4933a_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_4933a_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_4933a_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_4933a_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_4933a_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_4933a_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_4933a_row0_col0\" class=\"data row0 col0\" >0.000000</td>\n",
+       "      <td id=\"T_4933a_row0_col1\" class=\"data row0 col1\" >0.002192</td>\n",
+       "      <td id=\"T_4933a_row0_col2\" class=\"data row0 col2\" >0.004006</td>\n",
+       "      <td id=\"T_4933a_row0_col3\" class=\"data row0 col3\" >0.004514</td>\n",
+       "      <td id=\"T_4933a_row0_col4\" class=\"data row0 col4\" >0.006648</td>\n",
+       "      <td id=\"T_4933a_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_4933a_row1_col0\" class=\"data row1 col0\" >0.241018</td>\n",
+       "      <td id=\"T_4933a_row1_col1\" class=\"data row1 col1\" >0.238426</td>\n",
+       "      <td id=\"T_4933a_row1_col2\" class=\"data row1 col2\" >0.234827</td>\n",
+       "      <td id=\"T_4933a_row1_col3\" class=\"data row1 col3\" >0.250730</td>\n",
+       "      <td id=\"T_4933a_row1_col4\" class=\"data row1 col4\" >0.503909</td>\n",
+       "      <td id=\"T_4933a_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_4933a_row2_col0\" class=\"data row2 col0\" >0.749860</td>\n",
+       "      <td id=\"T_4933a_row2_col1\" class=\"data row2 col1\" >0.757703</td>\n",
+       "      <td id=\"T_4933a_row2_col2\" class=\"data row2 col2\" >0.750911</td>\n",
+       "      <td id=\"T_4933a_row2_col3\" class=\"data row2 col3\" >0.753713</td>\n",
+       "      <td id=\"T_4933a_row2_col4\" class=\"data row2 col4\" >0.746708</td>\n",
+       "      <td id=\"T_4933a_row2_col5\" class=\"data row2 col5\" >0.754379</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_4933a_row3_col0\" class=\"data row3 col0\" >0.203390</td>\n",
+       "      <td id=\"T_4933a_row3_col1\" class=\"data row3 col1\" >0.214351</td>\n",
+       "      <td id=\"T_4933a_row3_col2\" class=\"data row3 col2\" >0.201258</td>\n",
+       "      <td id=\"T_4933a_row3_col3\" class=\"data row3 col3\" >0.208821</td>\n",
+       "      <td id=\"T_4933a_row3_col4\" class=\"data row3 col4\" >0.208406</td>\n",
+       "      <td id=\"T_4933a_row3_col5\" class=\"data row3 col5\" >0.214136</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_4933a_row4_col0\" class=\"data row4 col0\" >0.332362</td>\n",
+       "      <td id=\"T_4933a_row4_col1\" class=\"data row4 col1\" >0.365325</td>\n",
+       "      <td id=\"T_4933a_row4_col2\" class=\"data row4 col2\" >0.333333</td>\n",
+       "      <td id=\"T_4933a_row4_col3\" class=\"data row4 col3\" >0.347305</td>\n",
+       "      <td id=\"T_4933a_row4_col4\" class=\"data row4 col4\" >0.326923</td>\n",
+       "      <td id=\"T_4933a_row4_col5\" class=\"data row4 col5\" >0.353103</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_4933a_row5_col0\" class=\"data row5 col0\" >0.146530</td>\n",
+       "      <td id=\"T_4933a_row5_col1\" class=\"data row5 col1\" >0.151671</td>\n",
+       "      <td id=\"T_4933a_row5_col2\" class=\"data row5 col2\" >0.144144</td>\n",
+       "      <td id=\"T_4933a_row5_col3\" class=\"data row5 col3\" >0.149292</td>\n",
+       "      <td id=\"T_4933a_row5_col4\" class=\"data row5 col4\" >0.152956</td>\n",
+       "      <td id=\"T_4933a_row5_col5\" class=\"data row5 col5\" >0.153661</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_4933a_row6_col0\" class=\"data row6 col0\" >0.583018</td>\n",
+       "      <td id=\"T_4933a_row6_col1\" class=\"data row6 col1\" >0.588834</td>\n",
+       "      <td id=\"T_4933a_row6_col2\" class=\"data row6 col2\" >0.575426</td>\n",
+       "      <td id=\"T_4933a_row6_col3\" class=\"data row6 col3\" >0.575821</td>\n",
+       "      <td id=\"T_4933a_row6_col4\" class=\"data row6 col4\" >0.573316</td>\n",
+       "      <td id=\"T_4933a_row6_col5\" class=\"data row6 col5\" >0.582555</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_4933a_row7_col0\" class=\"data row7 col0\" >-0.190980</td>\n",
+       "      <td id=\"T_4933a_row7_col1\" class=\"data row7 col1\" >-0.186599</td>\n",
+       "      <td id=\"T_4933a_row7_col2\" class=\"data row7 col2\" >-0.191325</td>\n",
+       "      <td id=\"T_4933a_row7_col3\" class=\"data row7 col3\" >-0.190608</td>\n",
+       "      <td id=\"T_4933a_row7_col4\" class=\"data row7 col4\" >-0.193769</td>\n",
+       "      <td id=\"T_4933a_row7_col5\" class=\"data row7 col5\" >-0.189077</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_4933a_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_4933a_row8_col0\" class=\"data row8 col0\" >-2.500209</td>\n",
+       "      <td id=\"T_4933a_row8_col1\" class=\"data row8 col1\" >-2.097705</td>\n",
+       "      <td id=\"T_4933a_row8_col2\" class=\"data row8 col2\" >-2.376728</td>\n",
+       "      <td id=\"T_4933a_row8_col3\" class=\"data row8 col3\" >-2.528142</td>\n",
+       "      <td id=\"T_4933a_row8_col4\" class=\"data row8 col4\" >-2.570247</td>\n",
+       "      <td id=\"T_4933a_row8_col5\" class=\"data row8 col5\" >-2.428060</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e093db520>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_a0470\">\n",
+       "  <caption>RF Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_a0470_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_a0470_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_a0470_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_a0470_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_a0470_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_a0470_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_a0470_row0_col0\" class=\"data row0 col0\" >1.522204</td>\n",
+       "      <td id=\"T_a0470_row0_col1\" class=\"data row0 col1\" >1.489951</td>\n",
+       "      <td id=\"T_a0470_row0_col2\" class=\"data row0 col2\" >1.485231</td>\n",
+       "      <td id=\"T_a0470_row0_col3\" class=\"data row0 col3\" >1.517533</td>\n",
+       "      <td id=\"T_a0470_row0_col4\" class=\"data row0 col4\" >1.491071</td>\n",
+       "      <td id=\"T_a0470_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_a0470_row1_col0\" class=\"data row1 col0\" >0.152138</td>\n",
+       "      <td id=\"T_a0470_row1_col1\" class=\"data row1 col1\" >0.145954</td>\n",
+       "      <td id=\"T_a0470_row1_col2\" class=\"data row1 col2\" >0.156016</td>\n",
+       "      <td id=\"T_a0470_row1_col3\" class=\"data row1 col3\" >0.150085</td>\n",
+       "      <td id=\"T_a0470_row1_col4\" class=\"data row1 col4\" >0.141980</td>\n",
+       "      <td id=\"T_a0470_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_a0470_row2_col0\" class=\"data row2 col0\" >0.738655</td>\n",
+       "      <td id=\"T_a0470_row2_col1\" class=\"data row2 col1\" >0.746779</td>\n",
+       "      <td id=\"T_a0470_row2_col2\" class=\"data row2 col2\" >0.740544</td>\n",
+       "      <td id=\"T_a0470_row2_col3\" class=\"data row2 col3\" >0.734099</td>\n",
+       "      <td id=\"T_a0470_row2_col4\" class=\"data row2 col4\" >0.737461</td>\n",
+       "      <td id=\"T_a0470_row2_col5\" class=\"data row2 col5\" >0.733595</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_a0470_row3_col0\" class=\"data row3 col0\" >0.240846</td>\n",
+       "      <td id=\"T_a0470_row3_col1\" class=\"data row3 col1\" >0.262643</td>\n",
+       "      <td id=\"T_a0470_row3_col2\" class=\"data row3 col2\" >0.227045</td>\n",
+       "      <td id=\"T_a0470_row3_col3\" class=\"data row3 col3\" >0.203191</td>\n",
+       "      <td id=\"T_a0470_row3_col4\" class=\"data row3 col4\" >0.244964</td>\n",
+       "      <td id=\"T_a0470_row3_col5\" class=\"data row3 col5\" >0.242379</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_a0470_row4_col0\" class=\"data row4 col0\" >0.328160</td>\n",
+       "      <td id=\"T_a0470_row4_col1\" class=\"data row4 col1\" >0.359375</td>\n",
+       "      <td id=\"T_a0470_row4_col2\" class=\"data row4 col2\" >0.323040</td>\n",
+       "      <td id=\"T_a0470_row4_col3\" class=\"data row4 col3\" >0.292271</td>\n",
+       "      <td id=\"T_a0470_row4_col4\" class=\"data row4 col4\" >0.328294</td>\n",
+       "      <td id=\"T_a0470_row4_col5\" class=\"data row4 col5\" >0.318359</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_a0470_row5_col0\" class=\"data row5 col0\" >0.190231</td>\n",
+       "      <td id=\"T_a0470_row5_col1\" class=\"data row5 col1\" >0.206941</td>\n",
+       "      <td id=\"T_a0470_row5_col2\" class=\"data row5 col2\" >0.175032</td>\n",
+       "      <td id=\"T_a0470_row5_col3\" class=\"data row5 col3\" >0.155727</td>\n",
+       "      <td id=\"T_a0470_row5_col4\" class=\"data row5 col4\" >0.195373</td>\n",
+       "      <td id=\"T_a0470_row5_col5\" class=\"data row5 col5\" >0.195678</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_a0470_row6_col0\" class=\"data row6 col0\" >0.616397</td>\n",
+       "      <td id=\"T_a0470_row6_col1\" class=\"data row6 col1\" >0.614756</td>\n",
+       "      <td id=\"T_a0470_row6_col2\" class=\"data row6 col2\" >0.597688</td>\n",
+       "      <td id=\"T_a0470_row6_col3\" class=\"data row6 col3\" >0.602290</td>\n",
+       "      <td id=\"T_a0470_row6_col4\" class=\"data row6 col4\" >0.596575</td>\n",
+       "      <td id=\"T_a0470_row6_col5\" class=\"data row6 col5\" >0.601436</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_a0470_row7_col0\" class=\"data row7 col0\" >-0.184849</td>\n",
+       "      <td id=\"T_a0470_row7_col1\" class=\"data row7 col1\" >-0.182776</td>\n",
+       "      <td id=\"T_a0470_row7_col2\" class=\"data row7 col2\" >-0.186704</td>\n",
+       "      <td id=\"T_a0470_row7_col3\" class=\"data row7 col3\" >-0.187369</td>\n",
+       "      <td id=\"T_a0470_row7_col4\" class=\"data row7 col4\" >-0.189693</td>\n",
+       "      <td id=\"T_a0470_row7_col5\" class=\"data row7 col5\" >-0.188506</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_a0470_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_a0470_row8_col0\" class=\"data row8 col0\" >-0.711816</td>\n",
+       "      <td id=\"T_a0470_row8_col1\" class=\"data row8 col1\" >-0.673727</td>\n",
+       "      <td id=\"T_a0470_row8_col2\" class=\"data row8 col2\" >-0.766073</td>\n",
+       "      <td id=\"T_a0470_row8_col3\" class=\"data row8 col3\" >-0.719537</td>\n",
+       "      <td id=\"T_a0470_row8_col4\" class=\"data row8 col4\" >-0.775028</td>\n",
+       "      <td id=\"T_a0470_row8_col5\" class=\"data row8 col5\" >-0.743245</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e6237c3a0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_87742\">\n",
+       "  <caption>GBC Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_87742_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_87742_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_87742_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_87742_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_87742_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_87742_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_87742_row0_col0\" class=\"data row0 col0\" >1.292703</td>\n",
+       "      <td id=\"T_87742_row0_col1\" class=\"data row0 col1\" >1.329037</td>\n",
+       "      <td id=\"T_87742_row0_col2\" class=\"data row0 col2\" >1.315473</td>\n",
+       "      <td id=\"T_87742_row0_col3\" class=\"data row0 col3\" >1.299452</td>\n",
+       "      <td id=\"T_87742_row0_col4\" class=\"data row0 col4\" >1.313950</td>\n",
+       "      <td id=\"T_87742_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_87742_row1_col0\" class=\"data row1 col0\" >0.014560</td>\n",
+       "      <td id=\"T_87742_row1_col1\" class=\"data row1 col1\" >0.026127</td>\n",
+       "      <td id=\"T_87742_row1_col2\" class=\"data row1 col2\" >0.018812</td>\n",
+       "      <td id=\"T_87742_row1_col3\" class=\"data row1 col3\" >0.015999</td>\n",
+       "      <td id=\"T_87742_row1_col4\" class=\"data row1 col4\" >0.020753</td>\n",
+       "      <td id=\"T_87742_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_87742_row2_col0\" class=\"data row2 col0\" >0.783193</td>\n",
+       "      <td id=\"T_87742_row2_col1\" class=\"data row2 col1\" >0.777871</td>\n",
+       "      <td id=\"T_87742_row2_col2\" class=\"data row2 col2\" >0.782292</td>\n",
+       "      <td id=\"T_87742_row2_col3\" class=\"data row2 col3\" >0.779770</td>\n",
+       "      <td id=\"T_87742_row2_col4\" class=\"data row2 col4\" >0.780050</td>\n",
+       "      <td id=\"T_87742_row2_col5\" class=\"data row2 col5\" >0.781830</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_87742_row3_col0\" class=\"data row3 col0\" >0.112385</td>\n",
+       "      <td id=\"T_87742_row3_col1\" class=\"data row3 col1\" >0.072515</td>\n",
+       "      <td id=\"T_87742_row3_col2\" class=\"data row3 col2\" >0.091228</td>\n",
+       "      <td id=\"T_87742_row3_col3\" class=\"data row3 col3\" >0.075294</td>\n",
+       "      <td id=\"T_87742_row3_col4\" class=\"data row3 col4\" >0.081871</td>\n",
+       "      <td id=\"T_87742_row3_col5\" class=\"data row3 col5\" >0.085479</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_87742_row4_col0\" class=\"data row4 col0\" >0.521277</td>\n",
+       "      <td id=\"T_87742_row4_col1\" class=\"data row4 col1\" >0.402597</td>\n",
+       "      <td id=\"T_87742_row4_col2\" class=\"data row4 col2\" >0.500000</td>\n",
+       "      <td id=\"T_87742_row4_col3\" class=\"data row4 col3\" >0.438356</td>\n",
+       "      <td id=\"T_87742_row4_col4\" class=\"data row4 col4\" >0.454545</td>\n",
+       "      <td id=\"T_87742_row4_col5\" class=\"data row4 col5\" >0.490566</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_87742_row5_col0\" class=\"data row5 col0\" >0.062982</td>\n",
+       "      <td id=\"T_87742_row5_col1\" class=\"data row5 col1\" >0.039846</td>\n",
+       "      <td id=\"T_87742_row5_col2\" class=\"data row5 col2\" >0.050193</td>\n",
+       "      <td id=\"T_87742_row5_col3\" class=\"data row5 col3\" >0.041184</td>\n",
+       "      <td id=\"T_87742_row5_col4\" class=\"data row5 col4\" >0.044987</td>\n",
+       "      <td id=\"T_87742_row5_col5\" class=\"data row5 col5\" >0.046819</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_87742_row6_col0\" class=\"data row6 col0\" >0.684278</td>\n",
+       "      <td id=\"T_87742_row6_col1\" class=\"data row6 col1\" >0.665449</td>\n",
+       "      <td id=\"T_87742_row6_col2\" class=\"data row6 col2\" >0.670804</td>\n",
+       "      <td id=\"T_87742_row6_col3\" class=\"data row6 col3\" >0.671522</td>\n",
+       "      <td id=\"T_87742_row6_col4\" class=\"data row6 col4\" >0.654303</td>\n",
+       "      <td id=\"T_87742_row6_col5\" class=\"data row6 col5\" >0.669660</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_87742_row7_col0\" class=\"data row7 col0\" >-0.157619</td>\n",
+       "      <td id=\"T_87742_row7_col1\" class=\"data row7 col1\" >-0.160471</td>\n",
+       "      <td id=\"T_87742_row7_col2\" class=\"data row7 col2\" >-0.159413</td>\n",
+       "      <td id=\"T_87742_row7_col3\" class=\"data row7 col3\" >-0.159018</td>\n",
+       "      <td id=\"T_87742_row7_col4\" class=\"data row7 col4\" >-0.161594</td>\n",
+       "      <td id=\"T_87742_row7_col5\" class=\"data row7 col5\" >-0.159691</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_87742_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_87742_row8_col0\" class=\"data row8 col0\" >-0.488690</td>\n",
+       "      <td id=\"T_87742_row8_col1\" class=\"data row8 col1\" >-0.495751</td>\n",
+       "      <td id=\"T_87742_row8_col2\" class=\"data row8 col2\" >-0.493124</td>\n",
+       "      <td id=\"T_87742_row8_col3\" class=\"data row8 col3\" >-0.492051</td>\n",
+       "      <td id=\"T_87742_row8_col4\" class=\"data row8 col4\" >-0.498777</td>\n",
+       "      <td id=\"T_87742_row8_col5\" class=\"data row8 col5\" >-0.493648</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e0ab71420>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_0e44f\">\n",
+       "  <caption>XGB Validation Scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_0e44f_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_0e44f_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_0e44f_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_0e44f_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_0e44f_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_0e44f_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_0e44f_row0_col0\" class=\"data row0 col0\" >0.643858</td>\n",
+       "      <td id=\"T_0e44f_row0_col1\" class=\"data row0 col1\" >0.640153</td>\n",
+       "      <td id=\"T_0e44f_row0_col2\" class=\"data row0 col2\" >0.677121</td>\n",
+       "      <td id=\"T_0e44f_row0_col3\" class=\"data row0 col3\" >0.634137</td>\n",
+       "      <td id=\"T_0e44f_row0_col4\" class=\"data row0 col4\" >0.669338</td>\n",
+       "      <td id=\"T_0e44f_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_0e44f_row1_col0\" class=\"data row1 col0\" >0.023605</td>\n",
+       "      <td id=\"T_0e44f_row1_col1\" class=\"data row1 col1\" >0.016412</td>\n",
+       "      <td id=\"T_0e44f_row1_col2\" class=\"data row1 col2\" >0.020805</td>\n",
+       "      <td id=\"T_0e44f_row1_col3\" class=\"data row1 col3\" >0.015040</td>\n",
+       "      <td id=\"T_0e44f_row1_col4\" class=\"data row1 col4\" >0.028892</td>\n",
+       "      <td id=\"T_0e44f_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_0e44f_row2_col0\" class=\"data row2 col0\" >0.620168</td>\n",
+       "      <td id=\"T_0e44f_row2_col1\" class=\"data row2 col1\" >0.612605</td>\n",
+       "      <td id=\"T_0e44f_row2_col2\" class=\"data row2 col2\" >0.604371</td>\n",
+       "      <td id=\"T_0e44f_row2_col3\" class=\"data row2 col3\" >0.628748</td>\n",
+       "      <td id=\"T_0e44f_row2_col4\" class=\"data row2 col4\" >0.620062</td>\n",
+       "      <td id=\"T_0e44f_row2_col5\" class=\"data row2 col5\" >0.619216</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_0e44f_row3_col0\" class=\"data row3 col0\" >0.397869</td>\n",
+       "      <td id=\"T_0e44f_row3_col1\" class=\"data row3 col1\" >0.386696</td>\n",
+       "      <td id=\"T_0e44f_row3_col2\" class=\"data row3 col2\" >0.377974</td>\n",
+       "      <td id=\"T_0e44f_row3_col3\" class=\"data row3 col3\" >0.386290</td>\n",
+       "      <td id=\"T_0e44f_row3_col4\" class=\"data row3 col4\" >0.383076</td>\n",
+       "      <td id=\"T_0e44f_row3_col5\" class=\"data row3 col5\" >0.387639</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_0e44f_row4_col0\" class=\"data row4 col0\" >0.303935</td>\n",
+       "      <td id=\"T_0e44f_row4_col1\" class=\"data row4 col1\" >0.295193</td>\n",
+       "      <td id=\"T_0e44f_row4_col2\" class=\"data row4 col2\" >0.287341</td>\n",
+       "      <td id=\"T_0e44f_row4_col3\" class=\"data row4 col3\" >0.301737</td>\n",
+       "      <td id=\"T_0e44f_row4_col4\" class=\"data row4 col4\" >0.296479</td>\n",
+       "      <td id=\"T_0e44f_row4_col5\" class=\"data row4 col5\" >0.298285</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_0e44f_row5_col0\" class=\"data row5 col0\" >0.575835</td>\n",
+       "      <td id=\"T_0e44f_row5_col1\" class=\"data row5 col1\" >0.560411</td>\n",
+       "      <td id=\"T_0e44f_row5_col2\" class=\"data row5 col2\" >0.552124</td>\n",
+       "      <td id=\"T_0e44f_row5_col3\" class=\"data row5 col3\" >0.536680</td>\n",
+       "      <td id=\"T_0e44f_row5_col4\" class=\"data row5 col4\" >0.541131</td>\n",
+       "      <td id=\"T_0e44f_row5_col5\" class=\"data row5 col5\" >0.553421</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_0e44f_row6_col0\" class=\"data row6 col0\" >0.640317</td>\n",
+       "      <td id=\"T_0e44f_row6_col1\" class=\"data row6 col1\" >0.626210</td>\n",
+       "      <td id=\"T_0e44f_row6_col2\" class=\"data row6 col2\" >0.619924</td>\n",
+       "      <td id=\"T_0e44f_row6_col3\" class=\"data row6 col3\" >0.627078</td>\n",
+       "      <td id=\"T_0e44f_row6_col4\" class=\"data row6 col4\" >0.620747</td>\n",
+       "      <td id=\"T_0e44f_row6_col5\" class=\"data row6 col5\" >0.630028</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_0e44f_row7_col0\" class=\"data row7 col0\" >-0.231951</td>\n",
+       "      <td id=\"T_0e44f_row7_col1\" class=\"data row7 col1\" >-0.238560</td>\n",
+       "      <td id=\"T_0e44f_row7_col2\" class=\"data row7 col2\" >-0.237950</td>\n",
+       "      <td id=\"T_0e44f_row7_col3\" class=\"data row7 col3\" >-0.233297</td>\n",
+       "      <td id=\"T_0e44f_row7_col4\" class=\"data row7 col4\" >-0.239657</td>\n",
+       "      <td id=\"T_0e44f_row7_col5\" class=\"data row7 col5\" >-0.236789</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0e44f_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_0e44f_row8_col0\" class=\"data row8 col0\" >-0.660186</td>\n",
+       "      <td id=\"T_0e44f_row8_col1\" class=\"data row8 col1\" >-0.678450</td>\n",
+       "      <td id=\"T_0e44f_row8_col2\" class=\"data row8 col2\" >-0.673157</td>\n",
+       "      <td id=\"T_0e44f_row8_col3\" class=\"data row8 col3\" >-0.666774</td>\n",
+       "      <td id=\"T_0e44f_row8_col4\" class=\"data row8 col4\" >-0.681109</td>\n",
+       "      <td id=\"T_0e44f_row8_col5\" class=\"data row8 col5\" >-0.671979</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e0aab11e0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# XGB hyperparameter that deals with unbalanced\n",
+    "scale_pos_weight = Y.mean()**-1\n",
+    "\n",
+    "# Creating the model objects\n",
+    "cls_lr = LogisticRegression(\n",
+    "            class_weight=\"balanced\",  # Hyperparameter to deal with unbalanced output\n",
+    "            random_state=lucky_num)\n",
+    "# cls_svm = SVC(random_state=lucky_num)  # Remove due its resource consumption and worst results\n",
+    "cls_nb = GaussianNB()\n",
+    "cls_knn = KNeighborsClassifier()\n",
+    "cls_rf = RandomForestClassifier(\n",
+    "            random_state=lucky_num,\n",
+    "            class_weight=\"balanced_subsample\")  # Hyperparameter to deal with unbalanced output\n",
+    "cls_gbc = GradientBoostingClassifier(random_state=lucky_num)\n",
+    "cls_xgb = xgb.XGBClassifier(\n",
+    "            objective=\"binary:logistic\",\n",
+    "            verbose=None,\n",
+    "            random_state=lucky_num,\n",
+    "            scale_pos_weight = scale_pos_weight)\n",
+    "\n",
+    "# Lists to iterate on our modeling function\n",
+    "cls_name = [\"LR\", \"NB\", \"KNN\", \"RF\", \"GBC\", \"XGB\"]\n",
+    "cls_list = [cls_lr, cls_NB, cls_knn, cls_rf, cls_gbc, cls_xgb]\n",
+    "\n",
+    "mdl_summaries = []\n",
+    "for name, inst in zip(cls_name, cls_list):\n",
+    "    mdl_list = create_model(name, inst)\n",
+    "    mdl_list = [name] + mdl_list\n",
+    "    mdl_summaries.append(mdl_list)\n",
+    "\n",
+    "df_mdl = pd.DataFrame(\n",
+    "            mdl_summaries,\n",
+    "            columns=[\n",
+    "                \"model\",\n",
+    "                \"test_accuracy\",\n",
+    "                \"test_f1\",\n",
+    "                \"test_precision\",\n",
+    "                \"test_recall\",\n",
+    "                \"test_roc_auc\",\n",
+    "                \"test_brier\",\n",
+    "                \"test_log_loss\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_0de30\">\n",
+       "  <caption>Test set validation scores</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_0de30_level0_col0\" class=\"col_heading level0 col0\" >model</th>\n",
+       "      <th id=\"T_0de30_level0_col1\" class=\"col_heading level0 col1\" >test_accuracy</th>\n",
+       "      <th id=\"T_0de30_level0_col2\" class=\"col_heading level0 col2\" >test_f1</th>\n",
+       "      <th id=\"T_0de30_level0_col3\" class=\"col_heading level0 col3\" >test_precision</th>\n",
+       "      <th id=\"T_0de30_level0_col4\" class=\"col_heading level0 col4\" >test_recall</th>\n",
+       "      <th id=\"T_0de30_level0_col5\" class=\"col_heading level0 col5\" >test_roc_auc</th>\n",
+       "      <th id=\"T_0de30_level0_col6\" class=\"col_heading level0 col6\" >test_brier</th>\n",
+       "      <th id=\"T_0de30_level0_col7\" class=\"col_heading level0 col7\" >test_log_loss</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0de30_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
+       "      <td id=\"T_0de30_row0_col0\" class=\"data row0 col0\" >LR</td>\n",
+       "      <td id=\"T_0de30_row0_col1\" class=\"data row0 col1\" >0.660523</td>\n",
+       "      <td id=\"T_0de30_row0_col2\" class=\"data row0 col2\" >0.414431</td>\n",
+       "      <td id=\"T_0de30_row0_col3\" class=\"data row0 col3\" >0.331889</td>\n",
+       "      <td id=\"T_0de30_row0_col4\" class=\"data row0 col4\" >0.551621</td>\n",
+       "      <td id=\"T_0de30_row0_col5\" class=\"data row0 col5\" >0.667540</td>\n",
+       "      <td id=\"T_0de30_row0_col6\" class=\"data row0 col6\" >0.225709</td>\n",
+       "      <td id=\"T_0de30_row0_col7\" class=\"data row0 col7\" >0.644816</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0de30_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
+       "      <td id=\"T_0de30_row1_col0\" class=\"data row1 col0\" >XGB</td>\n",
+       "      <td id=\"T_0de30_row1_col1\" class=\"data row1 col1\" >0.619216</td>\n",
+       "      <td id=\"T_0de30_row1_col2\" class=\"data row1 col2\" >0.387639</td>\n",
+       "      <td id=\"T_0de30_row1_col3\" class=\"data row1 col3\" >0.298285</td>\n",
+       "      <td id=\"T_0de30_row1_col4\" class=\"data row1 col4\" >0.553421</td>\n",
+       "      <td id=\"T_0de30_row1_col5\" class=\"data row1 col5\" >0.630028</td>\n",
+       "      <td id=\"T_0de30_row1_col6\" class=\"data row1 col6\" >0.236789</td>\n",
+       "      <td id=\"T_0de30_row1_col7\" class=\"data row1 col7\" >0.671979</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0de30_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
+       "      <td id=\"T_0de30_row2_col0\" class=\"data row2 col0\" >NB</td>\n",
+       "      <td id=\"T_0de30_row2_col1\" class=\"data row2 col1\" >0.699608</td>\n",
+       "      <td id=\"T_0de30_row2_col2\" class=\"data row2 col2\" >0.381260</td>\n",
+       "      <td id=\"T_0de30_row2_col3\" class=\"data row2 col3\" >0.345703</td>\n",
+       "      <td id=\"T_0de30_row2_col4\" class=\"data row2 col4\" >0.424970</td>\n",
+       "      <td id=\"T_0de30_row2_col5\" class=\"data row2 col5\" >0.634031</td>\n",
+       "      <td id=\"T_0de30_row2_col6\" class=\"data row2 col6\" >0.268468</td>\n",
+       "      <td id=\"T_0de30_row2_col7\" class=\"data row2 col7\" >1.659029</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0de30_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
+       "      <td id=\"T_0de30_row3_col0\" class=\"data row3 col0\" >RF</td>\n",
+       "      <td id=\"T_0de30_row3_col1\" class=\"data row3 col1\" >0.733595</td>\n",
+       "      <td id=\"T_0de30_row3_col2\" class=\"data row3 col2\" >0.242379</td>\n",
+       "      <td id=\"T_0de30_row3_col3\" class=\"data row3 col3\" >0.318359</td>\n",
+       "      <td id=\"T_0de30_row3_col4\" class=\"data row3 col4\" >0.195678</td>\n",
+       "      <td id=\"T_0de30_row3_col5\" class=\"data row3 col5\" >0.601436</td>\n",
+       "      <td id=\"T_0de30_row3_col6\" class=\"data row3 col6\" >0.188506</td>\n",
+       "      <td id=\"T_0de30_row3_col7\" class=\"data row3 col7\" >0.743245</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0de30_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
+       "      <td id=\"T_0de30_row4_col0\" class=\"data row4 col0\" >KNN</td>\n",
+       "      <td id=\"T_0de30_row4_col1\" class=\"data row4 col1\" >0.754379</td>\n",
+       "      <td id=\"T_0de30_row4_col2\" class=\"data row4 col2\" >0.214136</td>\n",
+       "      <td id=\"T_0de30_row4_col3\" class=\"data row4 col3\" >0.353103</td>\n",
+       "      <td id=\"T_0de30_row4_col4\" class=\"data row4 col4\" >0.153661</td>\n",
+       "      <td id=\"T_0de30_row4_col5\" class=\"data row4 col5\" >0.582555</td>\n",
+       "      <td id=\"T_0de30_row4_col6\" class=\"data row4 col6\" >0.189077</td>\n",
+       "      <td id=\"T_0de30_row4_col7\" class=\"data row4 col7\" >2.428060</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_0de30_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
+       "      <td id=\"T_0de30_row5_col0\" class=\"data row5 col0\" >GBC</td>\n",
+       "      <td id=\"T_0de30_row5_col1\" class=\"data row5 col1\" >0.781830</td>\n",
+       "      <td id=\"T_0de30_row5_col2\" class=\"data row5 col2\" >0.085479</td>\n",
+       "      <td id=\"T_0de30_row5_col3\" class=\"data row5 col3\" >0.490566</td>\n",
+       "      <td id=\"T_0de30_row5_col4\" class=\"data row5 col4\" >0.046819</td>\n",
+       "      <td id=\"T_0de30_row5_col5\" class=\"data row5 col5\" >0.669660</td>\n",
+       "      <td id=\"T_0de30_row5_col6\" class=\"data row5 col6\" >0.159691</td>\n",
+       "      <td id=\"T_0de30_row5_col7\" class=\"data row5 col7\" >0.493648</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e044360e0>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "df_mdl.sort_values(\n",
+    "        \"test_f1\",\n",
+    "        ascending=False,\n",
+    "        inplace=True,\n",
+    "        ignore_index=True)\n",
+    "\n",
+    "display(df_mdl.style.set_caption(\"Test set validation scores\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Any of models present good results! We will try to fit a composite model with the 3 better."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "</style>\n",
+       "<table id=\"T_ca02f\">\n",
+       "  <caption>Test set validation scores for Composite Model</caption>\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_ca02f_level0_col0\" class=\"col_heading level0 col0\" >CV1</th>\n",
+       "      <th id=\"T_ca02f_level0_col1\" class=\"col_heading level0 col1\" >CV2</th>\n",
+       "      <th id=\"T_ca02f_level0_col2\" class=\"col_heading level0 col2\" >CV3</th>\n",
+       "      <th id=\"T_ca02f_level0_col3\" class=\"col_heading level0 col3\" >CV4</th>\n",
+       "      <th id=\"T_ca02f_level0_col4\" class=\"col_heading level0 col4\" >CV5</th>\n",
+       "      <th id=\"T_ca02f_level0_col5\" class=\"col_heading level0 col5\" >TestSet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row0\" class=\"row_heading level0 row0\" >fit_time</th>\n",
+       "      <td id=\"T_ca02f_row0_col0\" class=\"data row0 col0\" >0.622165</td>\n",
+       "      <td id=\"T_ca02f_row0_col1\" class=\"data row0 col1\" >0.732543</td>\n",
+       "      <td id=\"T_ca02f_row0_col2\" class=\"data row0 col2\" >0.591849</td>\n",
+       "      <td id=\"T_ca02f_row0_col3\" class=\"data row0 col3\" >0.699149</td>\n",
+       "      <td id=\"T_ca02f_row0_col4\" class=\"data row0 col4\" >0.617794</td>\n",
+       "      <td id=\"T_ca02f_row0_col5\" class=\"data row0 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row1\" class=\"row_heading level0 row1\" >score_time</th>\n",
+       "      <td id=\"T_ca02f_row1_col0\" class=\"data row1 col0\" >0.023785</td>\n",
+       "      <td id=\"T_ca02f_row1_col1\" class=\"data row1 col1\" >0.027777</td>\n",
+       "      <td id=\"T_ca02f_row1_col2\" class=\"data row1 col2\" >0.030991</td>\n",
+       "      <td id=\"T_ca02f_row1_col3\" class=\"data row1 col3\" >0.027807</td>\n",
+       "      <td id=\"T_ca02f_row1_col4\" class=\"data row1 col4\" >0.023930</td>\n",
+       "      <td id=\"T_ca02f_row1_col5\" class=\"data row1 col5\" >nan</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row2\" class=\"row_heading level0 row2\" >test_accuracy</th>\n",
+       "      <td id=\"T_ca02f_row2_col0\" class=\"data row2 col0\" >0.714846</td>\n",
+       "      <td id=\"T_ca02f_row2_col1\" class=\"data row2 col1\" >0.701120</td>\n",
+       "      <td id=\"T_ca02f_row2_col2\" class=\"data row2 col2\" >0.695713</td>\n",
+       "      <td id=\"T_ca02f_row2_col3\" class=\"data row2 col3\" >0.693752</td>\n",
+       "      <td id=\"T_ca02f_row2_col4\" class=\"data row2 col4\" >0.677221</td>\n",
+       "      <td id=\"T_ca02f_row2_col5\" class=\"data row2 col5\" >0.699346</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row3\" class=\"row_heading level0 row3\" >test_f1</th>\n",
+       "      <td id=\"T_ca02f_row3_col0\" class=\"data row3 col0\" >0.412240</td>\n",
+       "      <td id=\"T_ca02f_row3_col1\" class=\"data row3 col1\" >0.404243</td>\n",
+       "      <td id=\"T_ca02f_row3_col2\" class=\"data row3 col2\" >0.389201</td>\n",
+       "      <td id=\"T_ca02f_row3_col3\" class=\"data row3 col3\" >0.385610</td>\n",
+       "      <td id=\"T_ca02f_row3_col4\" class=\"data row3 col4\" >0.369803</td>\n",
+       "      <td id=\"T_ca02f_row3_col5\" class=\"data row3 col5\" >0.389597</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row4\" class=\"row_heading level0 row4\" >test_precision</th>\n",
+       "      <td id=\"T_ca02f_row4_col0\" class=\"data row4 col0\" >0.374214</td>\n",
+       "      <td id=\"T_ca02f_row4_col1\" class=\"data row4 col1\" >0.357354</td>\n",
+       "      <td id=\"T_ca02f_row4_col2\" class=\"data row4 col2\" >0.345654</td>\n",
+       "      <td id=\"T_ca02f_row4_col3\" class=\"data row4 col3\" >0.342315</td>\n",
+       "      <td id=\"T_ca02f_row4_col4\" class=\"data row4 col4\" >0.321905</td>\n",
+       "      <td id=\"T_ca02f_row4_col5\" class=\"data row4 col5\" >0.349191</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row5\" class=\"row_heading level0 row5\" >test_recall</th>\n",
+       "      <td id=\"T_ca02f_row5_col0\" class=\"data row5 col0\" >0.458869</td>\n",
+       "      <td id=\"T_ca02f_row5_col1\" class=\"data row5 col1\" >0.465296</td>\n",
+       "      <td id=\"T_ca02f_row5_col2\" class=\"data row5 col2\" >0.445302</td>\n",
+       "      <td id=\"T_ca02f_row5_col3\" class=\"data row5 col3\" >0.441441</td>\n",
+       "      <td id=\"T_ca02f_row5_col4\" class=\"data row5 col4\" >0.434447</td>\n",
+       "      <td id=\"T_ca02f_row5_col5\" class=\"data row5 col5\" >0.440576</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row6\" class=\"row_heading level0 row6\" >test_roc_auc</th>\n",
+       "      <td id=\"T_ca02f_row6_col0\" class=\"data row6 col0\" >0.679117</td>\n",
+       "      <td id=\"T_ca02f_row6_col1\" class=\"data row6 col1\" >0.662225</td>\n",
+       "      <td id=\"T_ca02f_row6_col2\" class=\"data row6 col2\" >0.651689</td>\n",
+       "      <td id=\"T_ca02f_row6_col3\" class=\"data row6 col3\" >0.658722</td>\n",
+       "      <td id=\"T_ca02f_row6_col4\" class=\"data row6 col4\" >0.640881</td>\n",
+       "      <td id=\"T_ca02f_row6_col5\" class=\"data row6 col5\" >0.658847</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row7\" class=\"row_heading level0 row7\" >test_neg_brier_score</th>\n",
+       "      <td id=\"T_ca02f_row7_col0\" class=\"data row7 col0\" >-0.199904</td>\n",
+       "      <td id=\"T_ca02f_row7_col1\" class=\"data row7 col1\" >-0.208466</td>\n",
+       "      <td id=\"T_ca02f_row7_col2\" class=\"data row7 col2\" >-0.211428</td>\n",
+       "      <td id=\"T_ca02f_row7_col3\" class=\"data row7 col3\" >-0.209929</td>\n",
+       "      <td id=\"T_ca02f_row7_col4\" class=\"data row7 col4\" >-0.218236</td>\n",
+       "      <td id=\"T_ca02f_row7_col5\" class=\"data row7 col5\" >-0.208991</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca02f_level0_row8\" class=\"row_heading level0 row8\" >test_neg_log_loss</th>\n",
+       "      <td id=\"T_ca02f_row8_col0\" class=\"data row8 col0\" >-0.590700</td>\n",
+       "      <td id=\"T_ca02f_row8_col1\" class=\"data row8 col1\" >-0.611156</td>\n",
+       "      <td id=\"T_ca02f_row8_col2\" class=\"data row8 col2\" >-0.616876</td>\n",
+       "      <td id=\"T_ca02f_row8_col3\" class=\"data row8 col3\" >-0.613285</td>\n",
+       "      <td id=\"T_ca02f_row8_col4\" class=\"data row8 col4\" >-0.633314</td>\n",
+       "      <td id=\"T_ca02f_row8_col5\" class=\"data row8 col5\" >-0.611563</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21e04333100>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Selecting the models\n",
+    "cls_name = [\"LR\", \"NB\", \"XGB\"]\n",
+    "cls_list = [cls_lr, cls_nb, cls_xgb]\n",
+    "\n",
+    "# Training the voting classifier\n",
+    "cls_vot = VotingClassifier([*zip(cls_name, cls_list)], voting=\"soft\")\n",
+    "cls_vot.fit(X_train, y_train)\n",
+    "\n",
+    "# Using cross-validation to evaluate the model fitted\n",
+    "cls_cross = cross_validate(\n",
+    "    estimator=cls_vot,\n",
+    "    X=X_train,\n",
+    "    y=y_train,\n",
+    "    cv=5,\n",
+    "    scoring=scores)\n",
+    "\n",
+    "df_vot = pd.DataFrame.from_dict(cls_cross, orient='index', columns=[\"CV\"+str(i) for i in range(1,6)])\n",
+    "\n",
+    "# Calculating score to test set\n",
+    "accurancy, f1, precision, recall, roc_auc, brier_score, log_loss_value = eval_model(cls_vot)\n",
+    "\n",
+    "# Filling a dataframe to better presentation\n",
+    "df_vot.at[\"test_accuracy\", \"TestSet\"] = accurancy\n",
+    "df_vot.at[\"test_f1\", \"TestSet\"] = f1\n",
+    "df_vot.at[\"test_recall\", \"TestSet\"] = recall\n",
+    "df_vot.at[\"test_precision\", \"TestSet\"] = precision\n",
+    "df_vot.at[\"test_roc_auc\", \"TestSet\"] = roc_auc\n",
+    "df_vot.at[\"test_neg_brier_score\", \"TestSet\"] = -brier_score\n",
+    "df_vot.at[\"test_neg_log_loss\", \"TestSet\"] = -log_loss_value\n",
+    "\n",
+    "display(df_vot.style.set_caption(\"Test set validation scores for Composite Model\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The composite model is not better than neat models. Well, maybe some tuning could handle this. But this will be done in future work."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['c:\\\\Users\\\\grego\\\\OneDrive\\\\Documentos\\\\Documentos Pessoais\\\\00_DataCamp\\\\09_VSC\\\\poa_car_accidents\\\\poa_car_accidents\\\\model\\\\model_feridos.pkl']"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Saving\n",
+    "# file_name = \"model_\" + output + '.pkl'\n",
+    "# jb.dump(cls_vot, path.join(path.abspath(\"./\"), file_name))"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.10.6 64-bit",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.6"
+  },
+  "orig_nbformat": 4,
+  "vscode": {
+   "interpreter": {
+    "hash": "1372d04dbd71fdc5436c5d6e671c1b9287e750e86143c81b5a7ba0acaf653c5e"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

model/scaler_feridos.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf819a52c70bb4058de5851aedbd92da179246c132e87593420f87850f7ae1b0
+size 2387

model/scaler_feridos_gr.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:063891bf93f83929da6294c645de8d503fcec2c756cde601933800c7d3dc6598
+size 2387

requirements.txt

@@ -0,0 +1,6 @@
+joblib
+numpy
+pandas
+gradio
+scikit-learn
+xgboost=1.6.2