HIV Incidence Predictor

HIV Incidence Predictor.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>AMAT_fac</th>\n",
       "      <th>HIVincidence</th>\n",
       "      <th>MH_fac</th>\n",
       "      <th>Med_AMAT_fac</th>\n",
       "      <th>Med_MH_fac</th>\n",
       "      <th>Med_SA_fac</th>\n",
       "      <th>Med_SMAT_fac</th>\n",
       "      <th>Med_TMAT_fac</th>\n",
       "      <th>Population</th>\n",
       "      <th>SA_fac</th>\n",
       "      <th>...</th>\n",
       "      <th>SD</th>\n",
       "      <th>TN</th>\n",
       "      <th>TX</th>\n",
       "      <th>UT</th>\n",
       "      <th>VA</th>\n",
       "      <th>VT</th>\n",
       "      <th>WA</th>\n",
       "      <th>WI</th>\n",
       "      <th>WV</th>\n",
       "      <th>WY</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>10.9</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>55035.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>8.7</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>203690.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>26270.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22561.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>10.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>57676.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 82 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   AMAT_fac  HIVincidence  MH_fac  Med_AMAT_fac  Med_MH_fac  Med_SA_fac  \\\n",
       "0       0.0          10.9     1.0           0.0         1.0         2.0   \n",
       "1       0.0           8.7     4.0           0.0         3.0         1.0   \n",
       "2       0.0           0.0     1.0           0.0         1.0         1.0   \n",
       "3       0.0           0.0     0.0           0.0         0.0         1.0   \n",
       "4       0.0          10.4     1.0           0.0         1.0         1.0   \n",
       "\n",
       "   Med_SMAT_fac  Med_TMAT_fac  Population  SA_fac  ...  SD  TN  TX  UT  VA  \\\n",
       "0           0.0           0.0     55035.0     2.0  ...   0   0   0   0   0   \n",
       "1           0.0           0.0    203690.0     2.0  ...   0   0   0   0   0   \n",
       "2           0.0           0.0     26270.0     1.0  ...   0   0   0   0   0   \n",
       "3           1.0           0.0     22561.0     1.0  ...   0   0   0   0   0   \n",
       "4           0.0           0.0     57676.0     1.0  ...   0   0   0   0   0   \n",
       "\n",
       "   VT  WA  WI  WV  WY  \n",
       "0   0   0   0   0   0  \n",
       "1   0   0   0   0   0  \n",
       "2   0   0   0   0   0  \n",
       "3   0   0   0   0   0  \n",
       "4   0   0   0   0   0  \n",
       "\n",
       "[5 rows x 82 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('HIV_df.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Separate into X and y\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "    y = df.pop('HIVincidence')\n",
    "except:\n",
    "    pass\n",
    "X = df.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How many observations have Zero incidence?\n",
    "\n",
    "## ~75%"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.254341164453524"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(y>0).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scale all Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler as SS\n",
    "X = SS().fit_transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train/Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split as TTS\n",
    "X_train, X_test, y_train, y_test = TTS(X,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make a very simple model to predict `Zero` or `Non-Zero` HIV Incidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression as LR\n",
    "model = LR(max_iter = 1000).fit(X_train,y_train>0)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Score the Model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9168937329700273, 0.9210884353741496)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.score(X_train, y_train>0), model.score(X_test, y_test>0) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 90% Accuracy is a reasonable starting point for such a simple model.\n",
    " \n",
    "## Now we'll look at only the portion of the data with non-zero HIV Incidence and make a regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "HIV_X = X[model.predict(X)>0]\n",
    "HIV_y = y[model.predict(X)>0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(656, 81)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "HIV_X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "HIV_X_train, HIV_X_test, HIV_y_train, HIV_y_test = TTS(HIV_X,HIV_y, random_state=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fit 100 different LASSO Models with alpha in the range of $10^{-2}$ to $10$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.metrics import mean_squared_error as mse\n",
    "\n",
    "train_score, test_score = [],[]\n",
    "coefs = []\n",
    "\n",
    "alphas = np.logspace(-2, 1, 100)\n",
    "\n",
    "for alpha in alphas:\n",
    "    regression = Lasso(alpha = alpha, max_iter = 10000)\n",
    "    regression.fit(HIV_X_train, HIV_y_train)\n",
    "    \n",
    "    train_score.append(mse(HIV_y_train, regression.predict(HIV_X_train), squared = False))    \n",
    "    test_score.append(mse(HIV_y_test, regression.predict(HIV_X_test), squared = False))\n",
    "\n",
    "    coefs.append(regression.coef_)\n",
    "    \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot the Train and Test RMSE Error.\n",
    "## Identify the alpha value that yields the lowest Test RMSE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(alphas, train_score, label = 'train')   \n",
    "plt.plot(alphas, test_score, label = 'test')\n",
    "plt.legend()\n",
    "best_alpha = alphas[np.argmin(test_score)]\n",
    "plt.axvline(best_alpha)\n",
    "plt.xlabel('LASSO Alpha')\n",
    "plt.ylabel('RMSE')\n",
    "plt.xscale('log');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1873817422860384"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "best_alpha"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot the coefficients and a vertical line to indicate the best alpha value.\n",
    "\n",
    "## Notice that at this alpha value, some of our coefficients are zero"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(alphas, coefs)\n",
    "plt.axvline(best_alpha)\n",
    "plt.title('Coefficients')\n",
    "plt.xscale('log')\n",
    "plt.xlabel('LASSO Alpha');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build our final model based on the alpha value that gives lowest test RMSE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Lasso(alpha=0.1873817422860384, max_iter=10000)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model = Lasso(alpha = best_alpha, max_iter = 10000)\n",
    "best_model.fit(HIV_X_train, HIV_y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use this best model to make predictions and compare to actuals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(best_model.predict(HIV_X_train), HIV_y_train, label = 'Train')\n",
    "plt.scatter(best_model.predict(HIV_X_test), HIV_y_test, label = 'Test')\n",
    "plt.legend()\n",
    "plt.plot([0,120],[0,120])\n",
    "plt.xlabel('Predictions')\n",
    "plt.ylabel('Actuals');\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

# Repo