2_Modelling_and_Evaluating_AdaBoost.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4e3e9a4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from joblib import Parallel,delayed,dump\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.impute import SimpleImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "af7cd3c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "from cust_funcs import prepare_dataset,ams_curve_tuning,linear_impute_transform,linear_imputer,ams_curve,prepare_training_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "adb38c43",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv('Data/training.csv')\n",
    "X,weights,y=prepare_dataset(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "789e8cfa",
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={\n",
    "    'learning_rate':[0.01,0.05,0.1,1,1.2],\n",
    "    'n_estimators':[50,100,150,200]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "746b455f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean ams = 2.0391210125094763, lr = 0.01, n = 50\n",
      "Mean ams = 2.037570166907507, lr = 0.01, n = 100\n",
      "Mean ams = 2.6348026068674586, lr = 0.01, n = 150\n",
      "Mean ams = 2.670455126448005, lr = 0.01, n = 200\n",
      "Mean ams = 2.7007001969083553, lr = 0.05, n = 50\n",
      "Mean ams = 2.822087110308913, lr = 0.05, n = 100\n",
      "Mean ams = 2.8714812991701795, lr = 0.05, n = 150\n",
      "Mean ams = 2.8843340284920487, lr = 0.05, n = 200\n",
      "Mean ams = 2.8189427835569907, lr = 0.1, n = 50\n",
      "Mean ams = 2.888825043471292, lr = 0.1, n = 100\n",
      "Mean ams = 2.8835949926171827, lr = 0.1, n = 150\n",
      "Mean ams = 2.9347700096259812, lr = 0.1, n = 200\n",
      "Mean ams = 2.4854615112535674, lr = 1, n = 50\n",
      "Mean ams = 2.5059857503925276, lr = 1, n = 100\n",
      "Mean ams = 2.5272338280015623, lr = 1, n = 150\n",
      "Mean ams = 2.5432565970706893, lr = 1, n = 200\n",
      "Mean ams = 2.446782436974116, lr = 1.2, n = 50\n",
      "Mean ams = 2.4979825256615786, lr = 1.2, n = 100\n",
      "Mean ams = 2.524806181815711, lr = 1.2, n = 150\n",
      "Mean ams = 2.5410602454730977, lr = 1.2, n = 200\n",
      "BEST SCORES: ams = 2.9347700096259812, lr = 0.1, n = 200\n"
     ]
    }
   ],
   "source": [
    "random_states=np.random.randint(10,100,8)\n",
    "AMS_vals=[];N=[];LR=[]\n",
    "for lr in parameters['learning_rate']:\n",
    "    for n in parameters['n_estimators']:\n",
    "        ADB=AdaBoostClassifier(n_estimators=n,learning_rate=lr)\n",
    "        trial_vals=Parallel(n_jobs=4)(delayed(ams_curve_tuning)(X,y,weights,lr,n,random_states[i],ADB) for i in range(8))\n",
    "        AMS_vals.append(np.mean(trial_vals));N.append(n);LR.append(lr) \n",
    "        \n",
    "        print('Mean ams = {}, lr = {}, n = {}'.format(np.mean(trial_vals),lr,n))\n",
    "        \n",
    "LR=np.array(LR);AMS_vals=np.array(AMS_vals);N=np.array(N)        \n",
    "print('BEST SCORES: ams = {}, lr = {}, n = {}'.format(max(AMS_vals),\n",
    "                                                      LR[AMS_vals==max(AMS_vals)][0],N[AMS_vals==max(AMS_vals)][0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6feeb8b",
   "metadata": {},
   "source": [
    "### Fitting and Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9d7fffc0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "X_test,weights,y_test=prepare_dataset(pd.read_csv('Data/testing.csv',index_col=0))\n",
    "\n",
    "X_train,_,y_train=prepare_dataset(pd.read_csv('Data/training.csv'))\n",
    "\n",
    "X_test=prepare_training_features(X_test,X_dep=X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3ffb7bc3",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>AdaBoostClassifier(learning_rate=0.1, n_estimators=200)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">AdaBoostClassifier</label><div class=\"sk-toggleable__content\"><pre>AdaBoostClassifier(learning_rate=0.1, n_estimators=200)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "AdaBoostClassifier(learning_rate=0.1, n_estimators=200)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ADA=AdaBoostClassifier(n_estimators=200,learning_rate=0.1)\n",
    "ADA.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bd427f4f",
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction_proba=ADA.predict_proba(X_test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "32c61fbb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ams_dat,max_ams=ams_curve(prediction_proba,y_test,weights,test_ratio=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "923d16a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.savetxt('Data/ADA_curve.csv',ams_dat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "70b4d118",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final AMS = 2.9662737575747773\n"
     ]
    }
   ],
   "source": [
    "print('Final AMS = {}'.format(max_ams))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}