Example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Surrogate Modeling Regression Example\n",
    "This example illustrates the difference between Kriging and the (Batch) XGBoost algorithm. \n",
    "The aim is to show the difference between Kriging and XGBoost on the exact same examples in\n",
    "the standard setting of regression. We do not expect a great difference in performance.\n",
    "Note that there is no iterative training in this example. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup the Environment\n",
    "Note that this code has only been tested on Python 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:100% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%config InlineBackend.figure_format = 'retina'\n",
    "from IPython.core.display import display, HTML\n",
    "display(HTML(\"<style>.container { width:100% !important; }</style>\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from functions import *\n",
    "%matplotlib inline\n",
    "import seaborn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initialize the parameters and constants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Set the ABM Evaluation Budget\n",
    "budget = 500\n",
    "\n",
    "# Set out-of-sample test and montecarlo sizes\n",
    "test_size = 100\n",
    "montecarlos = 100\n",
    "\n",
    "# Get an on out-of-sample test set that does not have combinations from the\n",
    "# batch or iterative experiments\n",
    "final_test_size = (test_size * montecarlos)\n",
    "\n",
    "# Set the ABM parameters and support\n",
    "islands_exploration_range = np.array([\n",
    "    (0.0, 10),  # rho\n",
    "    (0.8, 2.0),  # alpha\n",
    "    (0.0, 1.0),  # phi\n",
    "    (0.0, 1.0),  # pi\n",
    "    (0.0, 1.0)])  # eps\n",
    "\n",
    "param_dims = islands_exploration_range.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the entire Budget in Batch and Generate unique out-of-sample test set\n",
    "Draw these samples according to a pseudo-random, well spaced generator. Here we use Sobol sequences because of their known coverage and compute performance. Note that the out of sample set is taken uniformly at random from the support. If you take points that are structured by the sobol sequence generator, then the performance evaluation will not be blind and will actually bias the Sobol structure. The aim of the out of sample set is to demonstrate out of sample points that can come from any point in the support of the parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "load_data = True\n",
    "\n",
    "if load_data: # This is only for the budget = 500 setting\n",
    "    evaluated_set_X_batch = pd.read_csv('X.csv',index_col=0).values\n",
    "    evaluated_set_y_batch = pd.read_csv('y.csv',index_col=0).values\n",
    "    oos_set = pd.read_csv('X_oos.csv',index_col=0).values\n",
    "    y_test = pd.read_csv('y_oos.csv',index_col=0).values\n",
    "else:\n",
    "    # Generate Sobol samples for training set\n",
    "    n_dimensions = islands_exploration_range.shape[0]\n",
    "    evaluated_set_X_batch = get_sobol_samples(n_dimensions, budget, islands_exploration_range)\n",
    "    evaluated_set_y_batch = evaluate_islands_on_set(evaluated_set_X_batch)\n",
    "    \n",
    "    pd.DataFrame(evaluated_set_X_batch).to_csv(\"X.csv\")\n",
    "    pd.DataFrame(evaluated_set_y_batch).to_csv(\"y.csv\")\n",
    "    \n",
    "    # Build Out-of-sample set\n",
    "    oos_set = get_unirand_samples(n_dimensions, final_test_size*budget, islands_exploration_range)\n",
    "    selections = []\n",
    "    for i, v in enumerate(oos_set):\n",
    "        if (v not in evaluated_set_X_batch):\n",
    "            selections.append(i)\n",
    "    oos_set = unique_rows(oos_set[selections])[:final_test_size]\n",
    "\n",
    "    # Evaluate the test set for the ABM response\n",
    "    y_test = evaluate_islands_on_set(oos_set)\n",
    "    \n",
    "    pd.DataFrame(oos_set).to_csv(\"X_oos.csv\")\n",
    "    pd.DataFrame(y_test).to_csv(\"y_oos.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compute the Kriging Surrogate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianProcessRegressor(alpha=1e-10, copy_X_train=True, kernel=None,\n",
       "             n_restarts_optimizer=0, normalize_y=False,\n",
       "             optimizer='fmin_l_bfgs_b', random_state=0)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We use the default Gaussian Kernel and L-BFGS optimizer.\n",
    "surrogate_models_kriging = GaussianProcessRegressor(random_state=0)\n",
    "surrogate_models_kriging.fit(evaluated_set_X_batch, evaluated_set_y_batch)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compute the XGBoost Surrogate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration No: 1 started. Evaluating function at random point.\n",
      "Iteration No: 1 ended. Evaluation done at random point.\n",
      "Time taken: 1.0216\n",
      "Function value obtained: 0.0152\n",
      "Current minimum: 0.0152\n",
      "Iteration No: 2 started. Evaluating function at random point.\n",
      "Iteration No: 2 ended. Evaluation done at random point.\n",
      "Time taken: 0.9079\n",
      "Function value obtained: 0.0181\n",
      "Current minimum: 0.0152\n",
      "Iteration No: 3 started. Evaluating function at random point.\n",
      "Iteration No: 3 ended. Evaluation done at random point.\n",
      "Time taken: 0.7801\n",
      "Function value obtained: 0.0306\n",
      "Current minimum: 0.0152\n",
      "Iteration No: 4 started. Evaluating function at random point.\n",
      "Iteration No: 4 ended. Evaluation done at random point.\n",
      "Time taken: 0.9032\n",
      "Function value obtained: 0.0181\n",
      "Current minimum: 0.0152\n",
      "Iteration No: 5 started. Evaluating function at random point.\n",
      "Iteration No: 5 ended. Evaluation done at random point.\n",
      "Time taken: 0.2662\n",
      "Function value obtained: 0.0142\n",
      "Current minimum: 0.0142\n",
      "Iteration No: 6 started. Evaluating function at random point.\n",
      "Iteration No: 6 ended. Evaluation done at random point.\n",
      "Time taken: 0.4642\n",
      "Function value obtained: 0.0161\n",
      "Current minimum: 0.0142\n",
      "Iteration No: 7 started. Evaluating function at random point.\n",
      "Iteration No: 7 ended. Evaluation done at random point.\n",
      "Time taken: 0.9978\n",
      "Function value obtained: 0.0150\n",
      "Current minimum: 0.0142\n",
      "Iteration No: 8 started. Evaluating function at random point.\n",
      "Iteration No: 8 ended. Evaluation done at random point.\n",
      "Time taken: 1.3553\n",
      "Function value obtained: 0.0090\n",
      "Current minimum: 0.0090\n",
      "Iteration No: 9 started. Evaluating function at random point.\n",
      "Iteration No: 9 ended. Evaluation done at random point.\n",
      "Time taken: 0.7153\n",
      "Function value obtained: 0.0110\n",
      "Current minimum: 0.0090\n",
      "Iteration No: 10 started. Evaluating function at random point.\n",
      "Iteration No: 10 ended. Evaluation done at random point.\n",
      "Time taken: 0.7433\n",
      "Function value obtained: 0.0108\n",
      "Current minimum: 0.0090\n"
     ]
    }
   ],
   "source": [
    "# This surrogate will not have multiple iterations. It will run on the entire budget of evaluations.\n",
    "surrogate_model_XGBoost = fit_surrogate_model(evaluated_set_X_batch, evaluated_set_y_batch)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluate both surrogates on the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test = [None] * 2\n",
    "y_hat_test[0] = surrogate_models_kriging.predict(oos_set)\n",
    "y_hat_test[1] = surrogate_model_XGBoost.predict(oos_set)\n",
    "\n",
    "# MSE performance\n",
    "mse_perf = np.zeros((2, montecarlos))\n",
    "for sur_idx in range(len(y_hat_test)):\n",
    "    for i in range(montecarlos):\n",
    "        mse_perf[sur_idx, i] = mean_squared_error(y_test[i * test_size:(i + 1) * test_size],\n",
    "                                                  y_hat_test[int(sur_idx)][i * test_size:(i + 1) * test_size])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot the out-of-sample Monte Carlo performance densities for each of the methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABWQAAAKXCAYAAAAFLEOIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f35178a14a8>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 331,
       "width": 690
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "experiment_labels = [\"Kriging\", \"XGBoost (Batch)\"]\n",
    "\n",
    "mse_perf = pd.DataFrame(mse_perf, index=experiment_labels)\n",
    "\n",
    "k_label = \"Kriging: Mean \" + '{:2.5f}'.format(mse_perf.iloc[0, :].mean()) + \", Variance \" + '{:2.5f}'.format(mse_perf.iloc[0, :].var())\n",
    "xgb_label = \"XGBoost: Mean \" + '{:2.5f}'.format(mse_perf.iloc[1, :].mean()) + \", Variance \" + '{:2.5f}'.format(mse_perf.iloc[1, :].var())\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12, 5))\n",
    "sns.distplot(mse_perf.iloc[0, :], label=k_label, ax=ax)\n",
    "sns.distplot(mse_perf.iloc[1, :], label=xgb_label, ax=ax)\n",
    "\n",
    "plt.title(\"Out-Of-Sample Prediction Performance\")\n",
    "plt.xlabel('Mean-Squared Error')\n",
    "plt.yticks([])\n",
    "\n",
    "plt.legend()\n",
    "\n",
    "fig.savefig(\"xgboost_kriging_ba_comparison_\" + str(budget) + \".png\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.48006'"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'{:2.5f}'.format(mse_perf.iloc[0, :].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}