5param_estimation_with_emukit.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import emcee\n",
    "from scipy.optimize import minimize\n",
    "import corner\n",
    "import ptemcee\n",
    "import subprocess\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import run_simple_energy_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "simulation=run_simple_energy_loss.run_simulation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "pT_range = (1, 10, 10) # Compute more observable than needed to avoid boundry unstabilities.\n",
    "pT_obs = [3, 5, 7] # p_T Observables we will use for the analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from emukit.core import ParameterSpace, ContinuousParameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "prior_ranges= {'alpha_s':(0.05,0.5), 'exponent_inel':(-3,3), 'exponent_el':(-3,3), 'scale_inel':(0.1, 0.5), 'scale_el':(0.1, 0.5)}\n",
    "true_values= {'alpha_s': 0.2, 'exponent_inel': -1, 'exponent_el': 1, 'scale_inel': 0.3, 'scale_el': 0.3}\n",
    "name_to_position =  {'alpha_s':0, 'exponent_inel':1, 'exponent_el':2, 'scale_inel':3, 'scale_el':4}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# We will generate observable for these model parametr values and consider them as experimental data\n",
    "# We will generate observables for these model parameter values and consider them as experimental data\n",
    "truth = np.array([true_values['alpha_s'],  true_values['exponent_inel'],\n",
    "                 true_values['exponent_el'],true_values['scale_inel'],true_values['scale_el']]).reshape(1,-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 1/1 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning: dopri5: step size becomes too small\n",
      "  self.messages.get(istate, unexpected_istate_msg)))\n",
      "/Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning: dopri5: larger nsteps is needed\n",
      "  self.messages.get(istate, unexpected_istate_msg)))\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning: The integral is probably divergent, or slowly convergent.\n",
      "  res_quad2b=scipy.integrate.quad(lambda u, p=p: p*integrand_middle(p,u), 0, delta, limit=npts, epsabs=epsabs, epsrel=epsrel)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning: The integral is probably divergent, or slowly convergent.\n",
      "  res_quad2a=scipy.integrate.quad(vec_integrand, pmin, p, limit=npts, epsabs=epsabs, epsrel=epsrel)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Experimental observations')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set seed for reproducibility.\n",
    "seed = 42\n",
    "np.random.seed(seed)\n",
    "\n",
    "# Generate some synthetic data from the model.\n",
    "x = np.array(pT_obs)\n",
    "N = len(x)\n",
    "obserr = 0.05 # Amplitude of noise term\n",
    "obs = simulation(truth, *pT_range)[:,pT_obs] # True model\n",
    "obs += obserr * np.random.randn(N) # Add some random noise\n",
    "\n",
    "# Now plot it to see what the data looks like\n",
    "fig, ax = plt.subplots(figsize=(8,8))\n",
    "\n",
    "ax.errorbar(x.flatten(), obs.flatten(), yerr=obserr, fmt=\"xk\", capsize=0)\n",
    "#ax.plot(x.flatten(), obs.flatten(), \"k\", alpha=0.3, lw=3)\n",
    "#ax.set_xlim(0, 10)\n",
    "ax.set_xlabel(r\"$p_T$\")\n",
    "ax.set_ylabel(r\"$R_{AA}$\")\n",
    "ax.set_title('Experimental observations')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameter_space = ParameterSpace([ContinuousParameter('alpha_s', 0.05,0.5 ),\n",
    "                                  ContinuousParameter('exponent_inel', -3, 3),\n",
    "                                  ContinuousParameter('exponent_el', -3, 3),\n",
    "                                  ContinuousParameter('scale_inel', 0.1, 0.5),\n",
    "                                  ContinuousParameter('scale_el', 0.1, 0.5)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X (5, 5)\n"
     ]
    }
   ],
   "source": [
    "from emukit.core.initial_designs import RandomDesign\n",
    "\n",
    "design = RandomDesign(parameter_space)\n",
    "num_data_points = 5\n",
    "X = design.get_samples(num_data_points)\n",
    "print(f'Shape of X {X.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(design, pT=pT_obs[0]):\n",
    "    \"\"\"returns simulation output for a design matrix\"\"\"\n",
    "    print(design)\n",
    "    observables=simulation(design,*pT_range)\n",
    "    #Let's just return only one obsrvable for now. The middle p_T bin\n",
    "    return observables[:,pT]\n",
    "\n",
    "def f1(design, pT=pT_obs[0]):\n",
    "    \"\"\"returns simulation output for a design matrix\"\"\"\n",
    "    print(design)\n",
    "    observables=simulation(design,*pT_range)\n",
    "    #Let's just return only one obsrvable for now. The middle p_T bin\n",
    "    return observables[:,pT].reshape(-1,1)\n",
    "\n",
    "def f2(design, pT=pT_obs[1]):\n",
    "    \"\"\"returns simulation output for a design matrix\"\"\"\n",
    "    print(design)\n",
    "    observables=simulation(design,*pT_range)\n",
    "    #Let's just return only one obsrvable for now. The middle p_T bin\n",
    "    return observables[:,pT].reshape(-1,1)\n",
    "\n",
    "def f3(design, pT=pT_obs[2]):\n",
    "    \"\"\"returns simulation output for a design matrix\"\"\"\n",
    "    print(design)\n",
    "    observables=simulation(design,*pT_range)\n",
    "    #Let's just return only one obsrvable for now. The middle p_T bin\n",
    "    return observables[:,pT].reshape(-1,1)\n",
    "\n",
    "fun=np.array([f1, f2, f3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.12020839  1.24843547 -1.9090502   0.21649166  0.28242799]\n",
      " [ 0.12019753 -2.87649303 -1.89957294  0.34474116  0.41407038]\n",
      " [ 0.07613763  2.81945911 -1.17454654  0.15579754  0.17986951]\n",
      " [ 0.43977927  1.99465584  0.14853859  0.21685786  0.30569378]\n",
      " [ 0.32050176 -1.72596534 -0.40832989  0.24654474  0.33696583]]\n",
      "Working on 1/5 design\n",
      "Working on 2/5 design\n",
      "Working on 3/5 design\n",
      "Working on 4/5 design\n",
      "Working on 5/5 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:42: RuntimeWarning: invalid value encountered in log\n",
      "  log_P_g_prev=np.log(P_g)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:74: IntegrationWarning: The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      "  res_quad2c=scipy.integrate.quad(vec_integrand, p, pmax, limit=npts, epsabs=epsabs, epsrel=epsrel)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:79: IntegrationWarning: The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      "  res_quad2c=scipy.integrate.quad(vec_integrand, p*(1+delta), pmax, limit=npts, epsabs=epsabs, epsrel=epsrel)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning: The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      "  res_quad2b=scipy.integrate.quad(lambda u, p=p: p*integrand_middle(p,u), 0, delta, limit=npts, epsabs=epsabs, epsrel=epsrel)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:77: IntegrationWarning: The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      "  res_quad2a=scipy.integrate.quad(vec_integrand, pmin, p*(1-delta), limit=npts, epsabs=epsabs, epsrel=epsrel)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning: The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      "  res_quad2a=scipy.integrate.quad(vec_integrand, pmin, p, limit=npts, epsabs=epsabs, epsrel=epsrel)\n",
      "/Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:74: IntegrationWarning: Extremely bad integrand behavior occurs at some points of the\n",
      "  integration interval.\n",
      "  res_quad2c=scipy.integrate.quad(vec_integrand, p, pmax, limit=npts, epsabs=epsabs, epsrel=epsrel)\n"
     ]
    }
   ],
   "source": [
    "#Generate 5 randome observables\n",
    "Y = f(X,pT_obs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from GPy.kern import RBF, White\n",
    "from GPy.models import GPRegression\n",
    "from emukit.model_wrappers import GPyModelWrapper\n",
    "from multiprocessing import Pool"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let's build a emulator for each observable (3 in total) and query 5 points for each emulator at a time. Then train all 3 emulators on these new 5 points that we get in each iteration. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 1/30, f = -0.517570421816393\n",
      "Optimization restart 2/30, f = -0.5175747604149712\n",
      "Optimization restart 3/30, f = -0.5175746063830977\n",
      "Optimization restart 4/30, f = 0.2116361865511216\n",
      "Optimization restart 5/30, f = 3.7854579870774465\n",
      "Optimization restart 6/30, f = -0.5175580628978849\n",
      "Optimization restart 7/30, f = -0.5175747161419957\n",
      "Optimization restart 8/30, f = -0.5175747258556558\n",
      "Optimization restart 9/30, f = -0.5175646946937569\n",
      "Optimization restart 10/30, f = -0.517572795787645\n",
      "Optimization restart 11/30, f = -0.5175746838216861\n",
      "Optimization restart 12/30, f = 3.785458013298841\n",
      "Optimization restart 13/30, f = 3.785457200859773\n",
      "Optimization restart 14/30, f = -0.5175747586068393\n",
      "Optimization restart 15/30, f = 0.2439208648515816\n",
      "Optimization restart 16/30, f = -0.5175747629383332\n",
      "Optimization restart 17/30, f = -0.5175747704436229\n",
      "Optimization restart 18/30, f = 0.09529531154006454\n",
      "Optimization restart 19/30, f = 0.24392091118951997\n",
      "Optimization restart 20/30, f = -0.5175558135555653\n",
      "Optimization restart 21/30, f = 0.21163619391384403\n",
      "Optimization restart 22/30, f = 3.7854530199974246\n",
      "Optimization restart 23/30, f = -0.5175746506982999\n",
      "Optimization restart 24/30, f = -0.5175484254369622\n",
      "Optimization restart 25/30, f = -0.51757474996535\n",
      "Optimization restart 26/30, f = -0.5175632653859905\n",
      "Optimization restart 27/30, f = -0.5175747611434245\n",
      "Optimization restart 28/30, f = -0.5175747526758729\n",
      "Optimization restart 29/30, f = -0.5175744330523662\n",
      "Optimization restart 30/30, f = 3.7854568761436673\n",
      "Optimization restart 1/30, f = -0.46611975331687905\n",
      "Optimization restart 2/30, f = -0.14061605337975447\n",
      "Optimization restart 3/30, f = 3.4976398501580586\n",
      "Optimization restart 4/30, f = -0.46608417942037095\n",
      "Optimization restart 5/30, f = -0.46611891674571915\n",
      "Optimization restart 6/30, f = -0.14061573179169518\n",
      "Optimization restart 7/30, f = 3.497768410158757\n",
      "Optimization restart 8/30, f = -0.4661201308824965\n",
      "Optimization restart 9/30, f = -0.4661201331675269\n",
      "Optimization restart 10/30, f = -0.14061609997638147\n",
      "Optimization restart 11/30, f = -0.4661201325264237\n",
      "Optimization restart 12/30, f = -0.14061610563061233\n",
      "Optimization restart 13/30, f = -0.4661125452917947\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/GPy/kern/src/stationary.py:166: RuntimeWarning:overflow encountered in true_divide\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/GPy/kern/src/stationary.py:137: RuntimeWarning:overflow encountered in square\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/GPy/kern/src/stationary.py:138: RuntimeWarning:invalid value encountered in add\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/GPy/kern/src/rbf.py:76: RuntimeWarning:invalid value encountered in multiply\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 14/30, f = -0.4661199749129157\n",
      "Optimization restart 15/30, f = -0.46606990856893526\n",
      "Optimization restart 16/30, f = -0.46612012872900976\n",
      "Optimization restart 17/30, f = -0.46612011138415\n",
      "Optimization restart 18/30, f = 3.497640463387766\n",
      "Optimization restart 19/30, f = -0.46611456220678793\n",
      "Optimization restart 20/30, f = -0.4660823637671605\n",
      "Optimization restart 21/30, f = -0.46611420459025865\n",
      "Optimization restart 22/30, f = -0.1406160134376755\n",
      "Optimization restart 23/30, f = -0.466120060058703\n",
      "Optimization restart 24/30, f = -0.14061600124747287\n",
      "Optimization restart 25/30, f = 3.4977667205524856\n",
      "Optimization restart 26/30, f = -0.46612012289817395\n",
      "Optimization restart 27/30, f = -0.4661198683205421\n",
      "Optimization restart 28/30, f = -0.14061612819193137\n",
      "Optimization restart 29/30, f = -0.46611819765024487\n",
      "Optimization restart 30/30, f = 3.4977680675267644\n",
      "Optimization restart 1/30, f = -0.44678043874928175\n",
      "Optimization restart 2/30, f = -0.44677899706758817\n",
      "Optimization restart 3/30, f = 0.2504204587498142\n",
      "Optimization restart 4/30, f = -0.2496368428390392\n",
      "Optimization restart 5/30, f = 3.405143870562505\n",
      "Optimization restart 6/30, f = -0.24963680630233354\n",
      "Optimization restart 7/30, f = 3.4051438914358894\n",
      "Optimization restart 8/30, f = -0.44677056596518083\n",
      "Optimization restart 9/30, f = 3.405142825010367\n",
      "Optimization restart 10/30, f = -0.44678031591704714\n",
      "Optimization restart 11/30, f = -0.24926233765417294\n",
      "Optimization restart 12/30, f = -0.44678048143785576\n",
      "Optimization restart 13/30, f = 0.2504205315879484\n",
      "Optimization restart 14/30, f = -0.44678047343131766\n",
      "Optimization restart 15/30, f = -0.24963659034959962\n",
      "Optimization restart 16/30, f = -0.44678043663983\n",
      "Optimization restart 17/30, f = -0.44678046700156226\n",
      "Optimization restart 18/30, f = -0.44676393980902773\n",
      "Optimization restart 19/30, f = -0.24963654863300588\n",
      "Optimization restart 20/30, f = 3.2074099719713263\n",
      "Optimization restart 21/30, f = 3.4051438914359404\n",
      "Optimization restart 22/30, f = -0.44676414216714777\n",
      "Optimization restart 23/30, f = -0.4467707770063982\n",
      "Optimization restart 24/30, f = -0.4467317976934635\n",
      "Optimization restart 25/30, f = -0.4467751881688642\n",
      "Optimization restart 26/30, f = -0.24963686731731194\n",
      "Optimization restart 27/30, f = 3.4051438914359498\n",
      "Optimization restart 28/30, f = -0.24963684662359276\n",
      "Optimization restart 29/30, f = -0.2496084082966803\n",
      "Optimization restart 30/30, f = -0.44678049294341093\n"
     ]
    }
   ],
   "source": [
    "# size of possible variation of parameters\n",
    "ptp = np.array([prior_ranges['alpha_s'][1],prior_ranges['exponent_inel'][1],prior_ranges['exponent_el'][1],prior_ranges['scale_inel'][1],prior_ranges['scale_el'][1]])\\\n",
    "-np.array([prior_ranges['alpha_s'][0],prior_ranges['exponent_inel'][0],prior_ranges['exponent_el'][0],prior_ranges['scale_inel'][0],prior_ranges['scale_el'][0]])\n",
    "\n",
    "gp_array = []\n",
    "for i,pt in enumerate(pT_obs):\n",
    "    rbf_kern =      RBF(  input_dim=5,\n",
    "                          variance= 1,\n",
    "                          lengthscale=ptp,\n",
    "                          ARD=True)\n",
    "    my_kernel = rbf_kern\n",
    "    model_gpy = GPRegression(X=X,Y=Y[:, i].reshape(-1,1),kernel=my_kernel, noise_var=0.01**2)\n",
    "    model_gpy.optimize_restarts(num_restarts = 30)\n",
    "    model_emukit = GPyModelWrapper(model_gpy, n_restarts=30)\n",
    "    gp_array.append(model_emukit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from emukit.experimental_design.acquisitions import IntegratedVarianceReduction, ModelVariance\n",
    "from emukit.experimental_design import ExperimentalDesignLoop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Name : GP regression\n",
      "Objective : -0.5175747704436229\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.26437635834107004  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "\n",
      "Name : GP regression\n",
      "Objective : -0.4661201331675269\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |       0.250101844615355  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "\n",
      "Name : GP regression\n",
      "Objective : -0.44678049294341093\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.24262005133796255  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n"
     ]
    }
   ],
   "source": [
    "integrated_var = []\n",
    "for g in gp_array:\n",
    "    print(g.model)\n",
    "    integrated_var.append(IntegratedVarianceReduction(model = g, space=parameter_space))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from emukit.core.optimization import GradientAcquisitionOptimizer\n",
    "#optimizer = GradientAcquisitionOptimizer(parameter_space)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5, 5)\n",
      "(5, 3)\n"
     ]
    }
   ],
   "source": [
    "print(X.shape)\n",
    "print(Y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trainng 0 gp for R_AA at 3 GeV\n",
      "Optimization restart 1/30, f = -0.5175747706933396\n",
      "Optimization restart 2/30, f = 0.24392107144897235\n",
      "Optimization restart 3/30, f = -0.5175682419340992\n",
      "Optimization restart 4/30, f = 3.785458013294597\n",
      "Optimization restart 5/30, f = 0.0952953264090195\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/paramz/transformations.py:111: RuntimeWarning:overflow encountered in expm1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 6/30, f = -0.5175743177870475\n",
      "Optimization restart 7/30, f = 0.21163618214910374\n",
      "Optimization restart 8/30, f = -0.5175747429227351\n",
      "Optimization restart 9/30, f = 3.7854580062975614\n",
      "Optimization restart 10/30, f = -0.5175710863371705\n",
      "Optimization restart 11/30, f = -0.5175651364251581\n",
      "Optimization restart 12/30, f = 3.785450406720013\n",
      "Optimization restart 13/30, f = 0.24392113971057405\n",
      "Optimization restart 14/30, f = -0.5175745699103866\n",
      "Optimization restart 15/30, f = -0.517540499508073\n",
      "Optimization restart 16/30, f = -0.517574772760288\n",
      "Optimization restart 17/30, f = -0.5175747258780032\n",
      "Optimization restart 18/30, f = 0.21163620860724874\n",
      "Optimization restart 19/30, f = -0.5175697139906208\n",
      "Optimization restart 20/30, f = -0.517574591133461\n",
      "Optimization restart 21/30, f = -0.5175675811964258\n",
      "Optimization restart 22/30, f = -0.5175747262586357\n",
      "Optimization restart 23/30, f = 0.2116361888127547\n",
      "Optimization restart 24/30, f = 3.785458011620706\n",
      "Optimization restart 25/30, f = -0.517574122850383\n",
      "Optimization restart 26/30, f = -0.5175652915369282\n",
      "Optimization restart 27/30, f = -0.5175255242742773\n",
      "Optimization restart 28/30, f = 3.7854563116366355\n",
      "Optimization restart 29/30, f = -0.5175747354325186\n",
      "Optimization restart 30/30, f = -0.5175747133377522\n",
      "[[ 0.46567774 -0.82704713  1.23280509  0.39693004  0.47902359]\n",
      " [ 0.2356259   2.29061779 -0.60663823  0.19767491  0.44612357]\n",
      " [ 0.33114635 -1.05463496  2.8477772   0.34984988  0.28832218]\n",
      " [ 0.1590015   1.29271346 -0.9928159   0.22145948  0.14327099]\n",
      " [ 0.43461398 -2.41599554 -1.21529473  0.37836793  0.26422344]\n",
      " [ 0.09931404  0.11982645  2.3634211   0.156057    0.14000565]\n",
      " [ 0.2173792  -2.30454487  2.72526094  0.41412492  0.10706217]\n",
      " [ 0.21283051  1.65140541 -0.83545359  0.19152585  0.12506686]\n",
      " [ 0.40554268  0.32966993  1.90936359  0.35705111  0.34703106]\n",
      " [ 0.39941283  2.26435576  2.37321357  0.49476208  0.38426327]]\n",
      "Working on 1/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: step size becomes too small\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: larger nsteps is needed\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:42: RuntimeWarning:invalid value encountered in log\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:74: IntegrationWarning:The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning:The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:79: IntegrationWarning:The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:77: IntegrationWarning:The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:The occurrence of roundoff error is detected, which prevents \n",
      "  the requested tolerance from being achieved.  The error may be \n",
      "  underestimated.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:74: IntegrationWarning:Extremely bad integrand behavior occurs at some points of the\n",
      "  integration interval.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:Extremely bad integrand behavior occurs at some points of the\n",
      "  integration interval.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For model parameters [ 0.46567774 -0.82704713  1.23280509  0.39693004  0.47902359] takes 93.980797290802 S\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "For model parameters [ 0.43461398 -2.41599554 -1.21529473  0.37836793  0.26422344] takes 214.48797297477722 S\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "For model parameters [ 0.2173792  -2.30454487  2.72526094  0.41412492  0.10706217] takes 61.47925090789795 S\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -6.138089338758288\n",
      "Optimization restart 2/30, f = -13.818389675295252\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/GPy/kern/src/stationary.py:243: RuntimeWarning:invalid value encountered in true_divide\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 3/30, f = -14.753815953923398\n",
      "Optimization restart 4/30, f = -11.64250901871094\n",
      "Optimization restart 5/30, f = -14.753816884130236\n",
      "Optimization restart 6/30, f = -14.971824933402477\n",
      "Optimization restart 7/30, f = -14.753814368097586\n",
      "Optimization restart 8/30, f = -14.753781752562482\n",
      "Optimization restart 9/30, f = -14.753816912846577\n",
      "Optimization restart 10/30, f = -14.753816479416264\n",
      "Optimization restart 11/30, f = -14.753816826121861\n",
      "Optimization restart 12/30, f = -14.753816938846906\n",
      "Optimization restart 13/30, f = -14.75381666584746\n",
      "Optimization restart 14/30, f = -14.753816383107157\n",
      "Optimization restart 15/30, f = -14.753816982785915\n",
      "Optimization restart 16/30, f = -14.753816551214975\n",
      "Optimization restart 17/30, f = -14.753816564398809\n",
      "Optimization restart 18/30, f = -14.75381667580757\n",
      "Optimization restart 19/30, f = -14.753816868646824\n",
      "Optimization restart 20/30, f = -14.753816434347549\n",
      "Optimization restart 21/30, f = -14.753816667792115\n",
      "Optimization restart 22/30, f = -14.753816401110578\n",
      "Optimization restart 23/30, f = -14.753816216350565\n",
      "Optimization restart 24/30, f = -14.753816958910075\n",
      "Optimization restart 25/30, f = -14.753797941943263\n",
      "Optimization restart 26/30, f = -14.97182493870314\n",
      "Optimization restart 27/30, f = -14.753816968062903\n",
      "Optimization restart 28/30, f = -14.75381640329006\n",
      "Optimization restart 29/30, f = -14.753816604959312\n",
      "Optimization restart 30/30, f = -14.971792101647889\n",
      "[[ 0.46567774 -0.82704713  1.23280509  0.39693004  0.47902359]\n",
      " [ 0.2356259   2.29061779 -0.60663823  0.19767491  0.44612357]\n",
      " [ 0.33114635 -1.05463496  2.8477772   0.34984988  0.28832218]\n",
      " [ 0.1590015   1.29271346 -0.9928159   0.22145948  0.14327099]\n",
      " [ 0.43461398 -2.41599554 -1.21529473  0.37836793  0.26422344]\n",
      " [ 0.09931404  0.11982645  2.3634211   0.156057    0.14000565]\n",
      " [ 0.2173792  -2.30454487  2.72526094  0.41412492  0.10706217]\n",
      " [ 0.21283051  1.65140541 -0.83545359  0.19152585  0.12506686]\n",
      " [ 0.40554268  0.32966993  1.90936359  0.35705111  0.34703106]\n",
      " [ 0.39941283  2.26435576  2.37321357  0.49476208  0.38426327]]\n",
      "Working on 1/10 design\n",
      "For model parameters [ 0.46567774 -0.82704713  1.23280509  0.39693004  0.47902359] takes 90.15564107894897 S\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "For model parameters [ 0.43461398 -2.41599554 -1.21529473  0.37836793  0.26422344] takes 213.90244007110596 S\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(15, 5)\n",
      "(15, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : 17818.875321191506\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |       0.250101844615355  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "Optimization restart 1/30, f = -6.516916440973636\n",
      "Optimization restart 2/30, f = -16.489487490320265\n",
      "Optimization restart 3/30, f = -17.16348277255409\n",
      "Optimization restart 4/30, f = -17.16348315686232\n",
      "Optimization restart 5/30, f = -17.163481982158203\n",
      "Optimization restart 6/30, f = -17.16348293380204\n",
      "Optimization restart 7/30, f = -17.163483097622333\n",
      "Optimization restart 8/30, f = -17.163476636405985\n",
      "Optimization restart 9/30, f = -17.163482467348953\n",
      "Optimization restart 10/30, f = -17.163482057670947\n",
      "Optimization restart 11/30, f = -17.16347677693444\n",
      "Optimization restart 12/30, f = -17.163482598254802\n",
      "Optimization restart 13/30, f = -17.163482798432618\n",
      "Optimization restart 14/30, f = -17.163483177397527\n",
      "Optimization restart 15/30, f = -17.163481950253516\n",
      "Optimization restart 16/30, f = -17.163481807815117\n",
      "Optimization restart 17/30, f = -17.163482740521456\n",
      "Optimization restart 18/30, f = -17.163481679299203\n",
      "Optimization restart 19/30, f = -17.16348277207814\n",
      "Optimization restart 20/30, f = -17.163483131348883\n",
      "Optimization restart 21/30, f = -17.16347618389938\n",
      "Optimization restart 22/30, f = -17.163481428335025\n",
      "Optimization restart 23/30, f = -17.16348336109311\n",
      "Optimization restart 24/30, f = -16.016425167089874\n",
      "Optimization restart 25/30, f = -17.16348264509208\n",
      "Optimization restart 26/30, f = -17.163482746652253\n",
      "Optimization restart 27/30, f = -17.163480885002897\n",
      "Optimization restart 28/30, f = -16.072125202576935\n",
      "Optimization restart 29/30, f = -17.16348301194808\n",
      "Optimization restart 30/30, f = -17.163480150461535\n",
      "Time it takes to do optimization 3.130350112915039\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -17.16348336109311\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.13970609334103276  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "(15, 5)\n",
      "(15, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : 12900.581988498365\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.24262005133796255  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "Optimization restart 1/30, f = -6.6483125451119935\n",
      "Optimization restart 2/30, f = -17.750938845281635\n",
      "Optimization restart 3/30, f = -17.75093823331103\n",
      "Optimization restart 4/30, f = -17.750939003773333\n",
      "Optimization restart 5/30, f = -17.750934568983393\n",
      "Optimization restart 6/30, f = -17.750939468204574\n",
      "Optimization restart 7/30, f = -17.75093893486575\n",
      "Optimization restart 8/30, f = -17.750939389146097\n",
      "Optimization restart 9/30, f = -17.750935304541343\n",
      "Optimization restart 10/30, f = -17.750935450426667\n",
      "Optimization restart 11/30, f = 0.46181791779209114\n",
      "Optimization restart 12/30, f = -17.750938752467135\n",
      "Optimization restart 13/30, f = -17.75093830815515\n",
      "Optimization restart 14/30, f = -17.750938449568032\n",
      "Optimization restart 15/30, f = -17.750936808411453\n",
      "Optimization restart 16/30, f = -17.750938250014826\n",
      "Optimization restart 17/30, f = -12.540303556374173\n",
      "Optimization restart 18/30, f = -17.75093874205817\n",
      "Optimization restart 19/30, f = -17.750937486077582\n",
      "Optimization restart 20/30, f = -17.750939497839212\n",
      "Optimization restart 21/30, f = 0.46181791023206475\n",
      "Optimization restart 22/30, f = -17.334078396626523\n",
      "Optimization restart 23/30, f = -17.750939376048933\n",
      "Optimization restart 24/30, f = -17.75093887489643\n",
      "Optimization restart 25/30, f = -17.750939369610165\n",
      "Optimization restart 26/30, f = -17.75093890346584\n",
      "Optimization restart 27/30, f = -17.750938825610355\n",
      "Optimization restart 28/30, f = -16.54265312350941\n",
      "Optimization restart 29/30, f = -17.75093198727429\n",
      "Optimization restart 30/30, f = -17.750937909766257\n",
      "Time it takes to do optimization 2.949838876724243\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -17.750939497839212\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.13343837831511846  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "Trainng 1 gp for R_AA at 5 GeV\n",
      "Optimization restart 1/30, f = -17.163483365493263\n",
      "Optimization restart 2/30, f = -17.163482898082925\n",
      "Optimization restart 3/30, f = -17.163482666033786\n",
      "Optimization restart 4/30, f = -17.163482840779793\n",
      "Optimization restart 5/30, f = 0.5774354100171202\n",
      "Optimization restart 6/30, f = -17.163483228709836\n",
      "Optimization restart 7/30, f = -17.16348348404167\n",
      "Optimization restart 8/30, f = -17.163479043995814\n",
      "Optimization restart 9/30, f = -17.16348059717422\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 10/30, f = -17.163482791375834\n",
      "Optimization restart 11/30, f = -17.163482731294174\n",
      "Optimization restart 12/30, f = -17.1634831024002\n",
      "Optimization restart 13/30, f = -17.16347658105058\n",
      "Optimization restart 14/30, f = 0.5789232994617821\n",
      "Optimization restart 15/30, f = -17.16348321404698\n",
      "Optimization restart 16/30, f = -17.163482844543395\n",
      "Optimization restart 17/30, f = 4.09402762691813\n",
      "Optimization restart 18/30, f = -17.16348211212651\n",
      "Optimization restart 19/30, f = -17.16348235289358\n",
      "Optimization restart 20/30, f = -17.163477903685724\n",
      "Optimization restart 21/30, f = -17.16348267388715\n",
      "Optimization restart 22/30, f = -17.163481899018088\n",
      "Optimization restart 23/30, f = -17.16346604979422\n",
      "Optimization restart 24/30, f = -17.163479900179436\n",
      "Optimization restart 25/30, f = -17.16348102051186\n",
      "Optimization restart 26/30, f = -17.16347142974507\n",
      "Optimization restart 27/30, f = -17.16348217775628\n",
      "Optimization restart 28/30, f = -17.163482354742982\n",
      "Optimization restart 29/30, f = -17.16348220443828\n",
      "Optimization restart 30/30, f = -17.16348216364697\n",
      "[[ 0.09365796 -0.2437481   0.49219249  0.46074186  0.4564673 ]\n",
      " [ 0.41012037 -2.01173275 -1.24872147  0.12244752  0.46599591]\n",
      " [ 0.34165651 -0.12688508 -0.67818537  0.12081528  0.19372681]\n",
      " [ 0.30586156  2.84042101 -1.41478702  0.49834718  0.22853155]\n",
      " [ 0.43266459  1.98071415  0.80354337  0.21282431  0.47628578]\n",
      " [ 0.14432483 -0.64207329 -1.33342289  0.47852881  0.17098934]\n",
      " [ 0.15262508  0.77539923  0.03113096  0.2161989   0.38626734]\n",
      " [ 0.4204725   0.35548163  1.92068533  0.3253488   0.10245958]\n",
      " [ 0.45717028 -0.64104918  1.99276274  0.12229443  0.27619653]\n",
      " [ 0.07498604 -0.8327942   0.16031992  0.29096345  0.41003471]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -17.518590043706\n",
      "Optimization restart 2/30, f = -30.197866464271055\n",
      "Optimization restart 3/30, f = -30.197866367324135\n",
      "Optimization restart 4/30, f = -30.197865954887867\n",
      "Optimization restart 5/30, f = -30.19786616564179\n",
      "Optimization restart 6/30, f = -30.19786644102926\n",
      "Optimization restart 7/30, f = -30.197866510497924\n",
      "Optimization restart 8/30, f = -30.19786651668642\n",
      "Optimization restart 9/30, f = -30.197866527817187\n",
      "Optimization restart 10/30, f = -30.197866542280007\n",
      "Optimization restart 11/30, f = -30.19786654794031\n",
      "Optimization restart 12/30, f = -30.197866553151968\n",
      "Optimization restart 13/30, f = -30.197866532833157\n",
      "Optimization restart 14/30, f = -30.197866531667838\n",
      "Optimization restart 15/30, f = -30.19786611664284\n",
      "Optimization restart 16/30, f = -30.197865991714835\n",
      "Optimization restart 17/30, f = -30.19786654910244\n",
      "Optimization restart 18/30, f = -31.295008275830604\n",
      "Optimization restart 19/30, f = -30.19786656576288\n",
      "Optimization restart 20/30, f = -30.19786655726216\n",
      "Optimization restart 21/30, f = -30.197853311482813\n",
      "Optimization restart 22/30, f = -30.197866451600063\n",
      "Optimization restart 23/30, f = -30.197866335420958\n",
      "Optimization restart 24/30, f = -30.197865502505998\n",
      "Optimization restart 25/30, f = -30.197866513707936\n",
      "Optimization restart 26/30, f = -30.19786576424385\n",
      "Optimization restart 27/30, f = -30.19786654613113\n",
      "Optimization restart 28/30, f = -30.197866513216976\n",
      "Optimization restart 29/30, f = -30.197866458470646\n",
      "Optimization restart 30/30, f = -30.197866518746334\n",
      "[[ 0.09365796 -0.2437481   0.49219249  0.46074186  0.4564673 ]\n",
      " [ 0.41012037 -2.01173275 -1.24872147  0.12244752  0.46599591]\n",
      " [ 0.34165651 -0.12688508 -0.67818537  0.12081528  0.19372681]\n",
      " [ 0.30586156  2.84042101 -1.41478702  0.49834718  0.22853155]\n",
      " [ 0.43266459  1.98071415  0.80354337  0.21282431  0.47628578]\n",
      " [ 0.14432483 -0.64207329 -1.33342289  0.47852881  0.17098934]\n",
      " [ 0.15262508  0.77539923  0.03113096  0.2161989   0.38626734]\n",
      " [ 0.4204725   0.35548163  1.92068533  0.3253488   0.10245958]\n",
      " [ 0.45717028 -0.64104918  1.99276274  0.12229443  0.27619653]\n",
      " [ 0.07498604 -0.8327942   0.16031992  0.29096345  0.41003471]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(25, 5)\n",
      "(25, 1)\n",
      "Optimizing the GP 0\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : 30.137909315010546\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.11004647001881544  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  7.686916771520381e-05  |      +ve      |        \n",
      "Optimization restart 1/30, f = -28.22538960234486\n",
      "Optimization restart 2/30, f = -27.53713594149969\n",
      "Optimization restart 3/30, f = -27.537135937010355\n",
      "Optimization restart 4/30, f = -27.53713594154226\n",
      "Optimization restart 5/30, f = -27.53713595266524\n",
      "Optimization restart 6/30, f = -27.53713595541432\n",
      "Optimization restart 7/30, f = -27.537135750228007\n",
      "Optimization restart 8/30, f = -27.537135955403404\n",
      "Optimization restart 9/30, f = -27.5371359553735\n",
      "Optimization restart 10/30, f = -27.53713595097037\n",
      "Optimization restart 11/30, f = -27.53713595548274\n",
      "Optimization restart 12/30, f = -27.53713595096479\n",
      "Optimization restart 13/30, f = -27.537135954298833\n",
      "Optimization restart 14/30, f = -27.53713595474457\n",
      "Optimization restart 15/30, f = -27.537135860912567\n",
      "Optimization restart 16/30, f = -27.53713593363487\n",
      "Optimization restart 17/30, f = -27.537135955441986\n",
      "Optimization restart 18/30, f = -27.537135925105588\n",
      "Optimization restart 19/30, f = -27.537135746832167\n",
      "Optimization restart 20/30, f = -27.537135916200345\n",
      "Optimization restart 21/30, f = -27.53713591244052\n",
      "Optimization restart 22/30, f = -27.53713595227644\n",
      "Optimization restart 23/30, f = -27.537135933831593\n",
      "Optimization restart 24/30, f = -27.537135950023917\n",
      "Optimization restart 25/30, f = -27.537135951714685\n",
      "Optimization restart 26/30, f = -27.537135941947994\n",
      "Optimization restart 27/30, f = -27.537135947566025\n",
      "Optimization restart 28/30, f = -27.537135951688896\n",
      "Optimization restart 29/30, f = -27.53713595376563\n",
      "Optimization restart 30/30, f = -27.53713594545848\n",
      "Time it takes to do optimization 1.8593721389770508\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -28.22538960234486\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                 value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |   0.13208691892415952  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                  (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  3.41164020107084e-27  |      +ve      |        \n",
      "(25, 5)\n",
      "(25, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : 349.1094074565523\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.13343837831511846  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.562684646268137e-309  |      +ve      |        \n",
      "Optimization restart 1/30, f = -18.075247914794858\n",
      "Optimization restart 2/30, f = -30.91238495471289\n",
      "Optimization restart 3/30, f = -30.912273028717717\n",
      "Optimization restart 4/30, f = -30.912234845799137\n",
      "Optimization restart 5/30, f = -31.75126709458825\n",
      "Optimization restart 6/30, f = -30.912323122412598\n",
      "Optimization restart 7/30, f = -30.912385299494005\n",
      "Optimization restart 8/30, f = -30.91238485163775\n",
      "Optimization restart 9/30, f = -30.9123853627227\n",
      "Optimization restart 10/30, f = -30.912384524733888\n",
      "Optimization restart 11/30, f = -30.91238452360007\n",
      "Optimization restart 12/30, f = -30.912373342679068\n",
      "Optimization restart 13/30, f = -30.91233111895399\n",
      "Optimization restart 14/30, f = -30.912346476083\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 15/30, f = -30.912242093172488\n",
      "Optimization restart 16/30, f = -30.912350467370533\n",
      "Optimization restart 17/30, f = -30.912365483685907\n",
      "Optimization restart 18/30, f = -30.91238474478003\n",
      "Optimization restart 19/30, f = -30.91237965120997\n",
      "Optimization restart 20/30, f = -30.91238525017612\n",
      "Optimization restart 21/30, f = -30.91238423352622\n",
      "Optimization restart 22/30, f = -30.912384473915452\n",
      "Optimization restart 23/30, f = -30.9122293485244\n",
      "Optimization restart 24/30, f = -30.91235922676645\n",
      "Optimization restart 25/30, f = -30.912359136491695\n",
      "Optimization restart 26/30, f = -30.912242889943546\n",
      "Optimization restart 27/30, f = -30.91233259971894\n",
      "Optimization restart 28/30, f = -30.912376499379977\n",
      "Optimization restart 29/30, f = -30.912245334404723\n",
      "Optimization restart 30/30, f = -30.912367293675565\n",
      "Time it takes to do optimization 2.015885829925537\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -31.75126709458825\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.28837215917173725  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.0005463960975873863  |      +ve      |        \n",
      "Trainng 2 gp for R_AA at 7 GeV\n",
      "Optimization restart 1/30, f = -31.751267105647287\n",
      "Optimization restart 2/30, f = -30.912348363511118\n",
      "Optimization restart 3/30, f = -30.912270969299172\n",
      "Optimization restart 4/30, f = -30.912273751938162\n",
      "Optimization restart 5/30, f = -30.912298388892037\n",
      "Optimization restart 6/30, f = -30.912385100823197\n",
      "Optimization restart 7/30, f = -30.912384938337464\n",
      "Optimization restart 8/30, f = -30.912384230177906\n",
      "Optimization restart 9/30, f = -30.91237137498871\n",
      "Optimization restart 10/30, f = -30.912369056153494\n",
      "Optimization restart 11/30, f = -30.91238389847652\n",
      "Optimization restart 12/30, f = -30.912317518718\n",
      "Optimization restart 13/30, f = -30.911906532425817\n",
      "Optimization restart 14/30, f = -31.751266091595603\n",
      "Optimization restart 15/30, f = -30.91238491862332\n",
      "Optimization restart 16/30, f = -30.912385132304166\n",
      "Optimization restart 17/30, f = -30.912361460650477\n",
      "Optimization restart 18/30, f = -30.912290224320063\n",
      "Optimization restart 19/30, f = -30.91217897737816\n",
      "Optimization restart 20/30, f = -30.9123231221253\n",
      "Optimization restart 21/30, f = -30.912186593234885\n",
      "Optimization restart 22/30, f = -30.912385337505707\n",
      "Optimization restart 23/30, f = -30.91238532321659\n",
      "Optimization restart 24/30, f = -30.912206229338167\n",
      "Optimization restart 25/30, f = -30.91231679571591\n",
      "Optimization restart 26/30, f = -34.617204964119665\n",
      "Optimization restart 27/30, f = -30.912385262410105\n",
      "Optimization restart 28/30, f = -30.91236580603114\n",
      "Optimization restart 29/30, f = -30.912333110718276\n",
      "Optimization restart 30/30, f = -30.91238393458789\n",
      "[[ 0.27588522 -0.41965221 -1.73687677  0.49398551  0.38401834]\n",
      " [ 0.0770518  -1.38177661  0.99381752  0.44147186  0.39714139]\n",
      " [ 0.49170431  1.53028969  2.70206263  0.29636362  0.26017405]\n",
      " [ 0.29008003 -1.87056374 -2.94526144  0.2809383   0.30972636]\n",
      " [ 0.48897648 -1.98964086  1.84803414  0.11506875  0.35813632]\n",
      " [ 0.1509756   0.92494399 -2.51896657  0.4397286   0.26058794]\n",
      " [ 0.24083186 -1.75820082  1.94641819  0.24072393  0.23408034]\n",
      " [ 0.48551457  0.81148581  0.97876708  0.11733068  0.2537689 ]\n",
      " [ 0.06685335 -0.38730924 -2.8571111   0.19014289  0.38687257]\n",
      " [ 0.17704349 -0.99848254 -2.5418031   0.41386771  0.15609164]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "For model parameters [0.48551457 0.81148581 0.97876708 0.11733068 0.2537689 ] takes 80.98744893074036 S\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -30.950299605482293\n",
      "Optimization restart 2/30, f = -52.50671916853217\n",
      "Optimization restart 3/30, f = -52.50671761448635\n",
      "Optimization restart 4/30, f = -52.50671922198383\n",
      "Optimization restart 5/30, f = -52.5067094532715\n",
      "Optimization restart 6/30, f = -46.3401911995816\n",
      "Optimization restart 7/30, f = -52.506718590165775\n",
      "Optimization restart 8/30, f = -52.50671953439916\n",
      "Optimization restart 9/30, f = -52.506719362830225\n",
      "Optimization restart 10/30, f = -52.50671847199816\n",
      "Optimization restart 11/30, f = -52.50671961742614\n",
      "Optimization restart 12/30, f = -52.50671918594765\n",
      "Optimization restart 13/30, f = -52.506719276014735\n",
      "Optimization restart 14/30, f = -52.50671938254483\n",
      "Optimization restart 15/30, f = -52.50671904528354\n",
      "Optimization restart 16/30, f = -52.50671938399884\n",
      "Optimization restart 17/30, f = -52.50671957181602\n",
      "Optimization restart 18/30, f = -52.50671927757052\n",
      "Optimization restart 19/30, f = -52.50671936708674\n",
      "Optimization restart 20/30, f = -52.506718968238076\n",
      "Optimization restart 21/30, f = -52.50671866142041\n",
      "Optimization restart 22/30, f = -52.5067187373145\n",
      "Optimization restart 23/30, f = -52.50671875207137\n",
      "Optimization restart 24/30, f = -52.50671961066294\n",
      "Optimization restart 25/30, f = -48.30975876372779\n",
      "Optimization restart 26/30, f = -52.50671946709353\n",
      "Optimization restart 27/30, f = -46.34823518183596\n",
      "Optimization restart 28/30, f = -52.50671926799329\n",
      "Optimization restart 29/30, f = -52.50671843899338\n",
      "Optimization restart 30/30, f = -48.30975872904595\n",
      "[[ 0.27588522 -0.41965221 -1.73687677  0.49398551  0.38401834]\n",
      " [ 0.0770518  -1.38177661  0.99381752  0.44147186  0.39714139]\n",
      " [ 0.49170431  1.53028969  2.70206263  0.29636362  0.26017405]\n",
      " [ 0.29008003 -1.87056374 -2.94526144  0.2809383   0.30972636]\n",
      " [ 0.48897648 -1.98964086  1.84803414  0.11506875  0.35813632]\n",
      " [ 0.1509756   0.92494399 -2.51896657  0.4397286   0.26058794]\n",
      " [ 0.24083186 -1.75820082  1.94641819  0.24072393  0.23408034]\n",
      " [ 0.48551457  0.81148581  0.97876708  0.11733068  0.2537689 ]\n",
      " [ 0.06685335 -0.38730924 -2.8571111   0.19014289  0.38687257]\n",
      " [ 0.17704349 -0.99848254 -2.5418031   0.41386771  0.15609164]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "For model parameters [0.48551457 0.81148581 0.97876708 0.11733068 0.2537689 ] takes 81.13158917427063 S\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(35, 5)\n",
      "(35, 1)\n",
      "Optimizing the GP 0\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : 200.3959395275453\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                 value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |   0.13208691892415952  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                  (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  3.41164020107084e-27  |      +ve      |        \n",
      "Optimization restart 1/30, f = -30.82432797102932\n",
      "Optimization restart 2/30, f = -46.078822564765794\n",
      "Optimization restart 3/30, f = -42.96847569863864\n",
      "Optimization restart 4/30, f = -42.96847566234588\n",
      "Optimization restart 5/30, f = -42.968475695208355\n",
      "Optimization restart 6/30, f = -42.968475700499255\n",
      "Optimization restart 7/30, f = -42.96847568671498\n",
      "Optimization restart 8/30, f = -42.968475700174466\n",
      "Optimization restart 9/30, f = -42.9684756996798\n",
      "Optimization restart 10/30, f = -42.96847570026748\n",
      "Optimization restart 11/30, f = -42.96847570052545\n",
      "Optimization restart 12/30, f = -42.96847568271477\n",
      "Optimization restart 13/30, f = -45.359503286829295\n",
      "Optimization restart 14/30, f = -42.96847547907838\n",
      "Optimization restart 15/30, f = -42.96847570010997\n",
      "Optimization restart 16/30, f = -42.96847567126848\n",
      "Optimization restart 17/30, f = -41.13857863431869\n",
      "Optimization restart 18/30, f = -45.96772157404274\n",
      "Optimization restart 19/30, f = -42.96847568154497\n",
      "Optimization restart 20/30, f = -45.97416590157519\n",
      "Optimization restart 21/30, f = -42.96847568051034\n",
      "Optimization restart 22/30, f = -41.13820597393918\n",
      "Optimization restart 23/30, f = -42.96847568811398\n",
      "Optimization restart 24/30, f = -42.96847569232728\n",
      "Optimization restart 25/30, f = -42.96847569861478\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 26/30, f = -42.968475690415566\n",
      "Optimization restart 27/30, f = -42.96847570053728\n",
      "Optimization restart 28/30, f = -42.96847570049235\n",
      "Optimization restart 29/30, f = -42.96847568297032\n",
      "Optimization restart 30/30, f = -42.96847570053117\n",
      "Time it takes to do optimization 2.0739948749542236\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -46.078822564765794\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.16967529375898344  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  8.928981379271784e-05  |      +ve      |        \n",
      "(35, 5)\n",
      "(35, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -42.33904271240725\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.28122420378172175  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.0006091397528945777  |      +ve      |        \n",
      "Optimization restart 1/30, f = -45.071244655998846\n",
      "Optimization restart 2/30, f = -47.80145920741411\n",
      "Optimization restart 3/30, f = -51.042919432758566\n",
      "Optimization restart 4/30, f = -51.04291930552024\n",
      "Optimization restart 5/30, f = -51.04291938844719\n",
      "Optimization restart 6/30, f = -51.04291935386081\n",
      "Optimization restart 7/30, f = -51.04291929765099\n",
      "Optimization restart 8/30, f = -51.042919101376214\n",
      "Optimization restart 9/30, f = -51.04291917296249\n",
      "Optimization restart 10/30, f = -51.04291949969511\n",
      "Optimization restart 11/30, f = -51.04291950728831\n",
      "Optimization restart 12/30, f = -51.04291950825332\n",
      "Optimization restart 13/30, f = -51.0429193022411\n",
      "Optimization restart 14/30, f = -51.04291945396839\n",
      "Optimization restart 15/30, f = -51.042919316919765\n",
      "Optimization restart 16/30, f = -51.04291925441408\n",
      "Optimization restart 17/30, f = -51.042918142321355\n",
      "Optimization restart 18/30, f = -51.04291940257209\n",
      "Optimization restart 19/30, f = -51.04291902259871\n",
      "Optimization restart 20/30, f = -51.0429186340495\n",
      "Optimization restart 21/30, f = -51.04291955520348\n",
      "Optimization restart 22/30, f = -51.042917990233235\n",
      "Optimization restart 23/30, f = -51.04291895625441\n",
      "Optimization restart 24/30, f = -51.042919366611855\n",
      "Optimization restart 25/30, f = -51.04291928682389\n",
      "Optimization restart 26/30, f = -51.04291769415736\n",
      "Optimization restart 27/30, f = -51.042919060823294\n",
      "Optimization restart 28/30, f = -51.04291832389186\n",
      "Optimization restart 29/30, f = -51.04291938648537\n",
      "Optimization restart 30/30, f = -51.04291923056098\n",
      "Time it takes to do optimization 4.0196919441223145\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -51.04291955520348\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.2772399987221634  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00017075736420523913  |      +ve      |        \n",
      "Trainng 0 gp for R_AA at 3 GeV\n",
      "Optimization restart 1/30, f = -46.07882256553306\n",
      "Optimization restart 2/30, f = -45.359503403017605\n",
      "Optimization restart 3/30, f = -45.360078666278184\n",
      "Optimization restart 4/30, f = -42.96847569558747\n",
      "Optimization restart 5/30, f = -42.96847561606357\n",
      "Optimization restart 6/30, f = -42.96847570013781\n",
      "Optimization restart 7/30, f = -42.96847569708057\n",
      "Optimization restart 8/30, f = -42.968475489682895\n",
      "Optimization restart 9/30, f = -42.96847570034879\n",
      "Optimization restart 10/30, f = -42.9684755691035\n",
      "Optimization restart 11/30, f = -45.359503283043594\n",
      "Optimization restart 12/30, f = -42.96847569891138\n",
      "Optimization restart 13/30, f = -42.96847569637874\n",
      "Optimization restart 14/30, f = -42.968475690898515\n",
      "Optimization restart 15/30, f = -42.96847569428198\n",
      "Optimization restart 16/30, f = -42.96847569504055\n",
      "Optimization restart 17/30, f = -41.13822301501831\n",
      "Optimization restart 18/30, f = -42.968475635746465\n",
      "Optimization restart 19/30, f = -42.96847569955614\n",
      "Optimization restart 20/30, f = -42.96847570038247\n",
      "Optimization restart 21/30, f = -42.9684757004715\n",
      "Optimization restart 22/30, f = -41.13820516055971\n",
      "Optimization restart 23/30, f = -42.96847569287574\n",
      "Optimization restart 24/30, f = -42.968475695084656\n",
      "Optimization restart 25/30, f = -42.96847566741295\n",
      "Optimization restart 26/30, f = -42.96847569116915\n",
      "Optimization restart 27/30, f = -45.3595043362897\n",
      "Optimization restart 28/30, f = -42.96847565420066\n",
      "Optimization restart 29/30, f = -42.968475654958596\n",
      "Optimization restart 30/30, f = -42.968475698786456\n",
      "[[ 0.10854599  2.66074369 -0.46724267  0.42215333  0.28555688]\n",
      " [ 0.11649683 -2.46882925 -0.02861155  0.12590137  0.28112056]\n",
      " [ 0.28341355  1.1020585   2.14257018  0.32720823  0.29056061]\n",
      " [ 0.32334057 -2.84570604  2.14499821  0.15515128  0.36656636]\n",
      " [ 0.05        1.46515018 -2.46655194  0.39343884  0.26264012]\n",
      " [ 0.45310564  2.58630495 -2.84724112  0.47541358  0.44320221]\n",
      " [ 0.36330843 -2.42308869 -1.13692455  0.46260405  0.46852945]\n",
      " [ 0.33789293  2.81509449  2.31183344  0.15650205  0.27589539]\n",
      " [ 0.48036087  2.94029209  0.85671791  0.2136848   0.30638634]\n",
      " [ 0.05375381 -2.68928936  0.84847326  0.22040335  0.38622796]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "For model parameters [ 0.45310564  2.58630495 -2.84724112  0.47541358  0.44320221] takes 98.000155210495 S\n",
      "Working on 7/10 design\n",
      "For model parameters [ 0.36330843 -2.42308869 -1.13692455  0.46260405  0.46852945] takes 199.98584699630737 S\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -49.12543418907852\n",
      "Optimization restart 2/30, f = -49.00703298972695\n",
      "Optimization restart 3/30, f = -49.0071307058083\n",
      "Optimization restart 4/30, f = -49.00703566630642\n",
      "Optimization restart 5/30, f = -49.007033023011786\n",
      "Optimization restart 6/30, f = -49.00703302148969\n",
      "Optimization restart 7/30, f = -49.00703510559907\n",
      "Optimization restart 8/30, f = -49.00703299628404\n",
      "Optimization restart 9/30, f = -49.00703303921616\n",
      "Optimization restart 10/30, f = -49.00703299788613\n",
      "Optimization restart 11/30, f = -49.00703294224232\n",
      "Optimization restart 12/30, f = -49.00703299751805\n",
      "Optimization restart 13/30, f = -49.00703299160499\n",
      "Optimization restart 14/30, f = -49.00703299734255\n",
      "Optimization restart 15/30, f = -49.007032981406596\n",
      "Optimization restart 16/30, f = -49.00703299802059\n",
      "Optimization restart 17/30, f = -49.00703298948775\n",
      "Optimization restart 18/30, f = -49.00703299480699\n",
      "Optimization restart 19/30, f = -37.391008902984915\n",
      "Optimization restart 20/30, f = -49.007033053168456\n",
      "Optimization restart 21/30, f = -49.00703298762744\n",
      "Optimization restart 22/30, f = -49.007033007052364\n",
      "Optimization restart 23/30, f = -49.00703299726171\n",
      "Optimization restart 24/30, f = -49.0070329947938\n",
      "Optimization restart 25/30, f = -49.0070330013287\n",
      "Optimization restart 26/30, f = -49.00703299227632\n",
      "Optimization restart 27/30, f = -37.38913242142087\n",
      "Optimization restart 28/30, f = -49.007091594113135\n",
      "Optimization restart 29/30, f = -49.00703299862107\n",
      "Optimization restart 30/30, f = -49.00703278313548\n",
      "[[ 0.10854599  2.66074369 -0.46724267  0.42215333  0.28555688]\n",
      " [ 0.11649683 -2.46882925 -0.02861155  0.12590137  0.28112056]\n",
      " [ 0.28341355  1.1020585   2.14257018  0.32720823  0.29056061]\n",
      " [ 0.32334057 -2.84570604  2.14499821  0.15515128  0.36656636]\n",
      " [ 0.05        1.46515018 -2.46655194  0.39343884  0.26264012]\n",
      " [ 0.45310564  2.58630495 -2.84724112  0.47541358  0.44320221]\n",
      " [ 0.36330843 -2.42308869 -1.13692455  0.46260405  0.46852945]\n",
      " [ 0.33789293  2.81509449  2.31183344  0.15650205  0.27589539]\n",
      " [ 0.48036087  2.94029209  0.85671791  0.2136848   0.30638634]\n",
      " [ 0.05375381 -2.68928936  0.84847326  0.22040335  0.38622796]]\n",
      "Working on 1/10 design\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "For model parameters [ 0.45310564  2.58630495 -2.84724112  0.47541358  0.44320221] takes 97.33517479896545 S\n",
      "Working on 7/10 design\n",
      "For model parameters [ 0.36330843 -2.42308869 -1.13692455  0.46260405  0.46852945] takes 199.8520519733429 S\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(45, 5)\n",
      "(45, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -19.580735128189076\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.2772399987221634  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00017075736420523913  |      +ve      |        \n",
      "Optimization restart 1/30, f = -52.16851110106548\n",
      "Optimization restart 2/30, f = -52.47475867057047\n",
      "Optimization restart 3/30, f = -52.47475898598972\n",
      "Optimization restart 4/30, f = -52.4747592078581\n",
      "Optimization restart 5/30, f = -52.47475867889049\n",
      "Optimization restart 6/30, f = -52.474759373535456\n",
      "Optimization restart 7/30, f = -52.47475867874265\n",
      "Optimization restart 8/30, f = -52.474966921371305\n",
      "Optimization restart 9/30, f = -52.47475865127751\n",
      "Optimization restart 10/30, f = -52.47475868528913\n",
      "Optimization restart 11/30, f = -52.474758691899275\n",
      "Optimization restart 12/30, f = -52.47475868535879\n",
      "Optimization restart 13/30, f = -52.47476352562616\n",
      "Optimization restart 14/30, f = -52.474758678399\n",
      "Optimization restart 15/30, f = -52.47475890514123\n",
      "Optimization restart 16/30, f = -38.28442611931207\n",
      "Optimization restart 17/30, f = -52.47475867873216\n",
      "Optimization restart 18/30, f = -52.47475867860001\n",
      "Optimization restart 19/30, f = -52.47475867661033\n",
      "Optimization restart 20/30, f = -52.47475867645524\n",
      "Optimization restart 21/30, f = -52.47475866446693\n",
      "Optimization restart 22/30, f = -52.47475867894158\n",
      "Optimization restart 23/30, f = -52.474758705110204\n",
      "Optimization restart 24/30, f = -52.474758678948305\n",
      "Optimization restart 25/30, f = -52.47475867738816\n",
      "Optimization restart 26/30, f = -52.474760051733526\n",
      "Optimization restart 27/30, f = -52.47475884929462\n",
      "Optimization restart 28/30, f = -52.47476138692499\n",
      "Optimization restart 29/30, f = -52.474758665032454\n",
      "Optimization restart 30/30, f = -52.47475871030732\n",
      "Time it takes to do optimization 2.509119987487793\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -52.474966921371305\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.22303843778410085  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  7.615639415974227e-09  |      +ve      |        \n",
      "(45, 5)\n",
      "(45, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -11.251816303893094\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.3268417835417073  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00015562692094288184  |      +ve      |        \n",
      "Optimization restart 1/30, f = -52.679826872084305\n",
      "Optimization restart 2/30, f = -52.946546535020715\n",
      "Optimization restart 3/30, f = -53.00452924077385\n",
      "Optimization restart 4/30, f = -52.94654656375565\n",
      "Optimization restart 5/30, f = -52.946547548099865\n",
      "Optimization restart 6/30, f = -52.94654654966767\n",
      "Optimization restart 7/30, f = -52.94654652734465\n",
      "Optimization restart 8/30, f = -52.9465465229281\n",
      "Optimization restart 9/30, f = -52.94654748458518\n",
      "Optimization restart 10/30, f = -52.94654654369399\n",
      "Optimization restart 11/30, f = -52.946546544927656\n",
      "Optimization restart 12/30, f = -52.9465465332444\n",
      "Optimization restart 13/30, f = -52.946546543171905\n",
      "Optimization restart 14/30, f = -52.94654662166883\n",
      "Optimization restart 15/30, f = -52.94654683531481\n",
      "Optimization restart 16/30, f = -52.94654651291394\n",
      "Optimization restart 17/30, f = -52.9465465630431\n",
      "Optimization restart 18/30, f = -25.102425452414543\n",
      "Optimization restart 19/30, f = -52.94654654374551\n",
      "Optimization restart 20/30, f = -52.946546592138304\n",
      "Optimization restart 21/30, f = -53.00452923865742\n",
      "Optimization restart 22/30, f = -52.94655184650848\n",
      "Optimization restart 23/30, f = -52.94654822467265\n",
      "Optimization restart 24/30, f = -52.94654654879438\n",
      "Optimization restart 25/30, f = -52.94667133428985\n",
      "Optimization restart 26/30, f = -52.946546541524235\n",
      "Optimization restart 27/30, f = -53.0045292395064\n",
      "Optimization restart 28/30, f = -52.94654784418095\n",
      "Optimization restart 29/30, f = -52.946546538776374\n",
      "Optimization restart 30/30, f = -53.00452921194476\n",
      "Time it takes to do optimization 2.542513132095337\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -53.00452924077385\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.21622341208160972  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.990271154883409e-06  |      +ve      |        \n",
      "Trainng 1 gp for R_AA at 5 GeV\n",
      "Optimization restart 1/30, f = -52.47496693352507\n",
      "Optimization restart 2/30, f = -52.474758678941186\n",
      "Optimization restart 3/30, f = -52.474768442388\n",
      "Optimization restart 4/30, f = -52.47475867881125\n",
      "Optimization restart 5/30, f = -52.47475867879763\n",
      "Optimization restart 6/30, f = -52.47475868280218\n",
      "Optimization restart 7/30, f = -52.474790997725776\n",
      "Optimization restart 8/30, f = -52.47475862493653\n",
      "Optimization restart 9/30, f = -52.587107845803814\n",
      "Optimization restart 10/30, f = -52.474758676335846\n",
      "Optimization restart 11/30, f = -52.587107842808905\n",
      "Optimization restart 12/30, f = -38.28442651570114\n",
      "Optimization restart 13/30, f = -52.47475870683388\n",
      "Optimization restart 14/30, f = -52.474758676785335\n",
      "Optimization restart 15/30, f = -52.474768465074675\n",
      "Optimization restart 16/30, f = -52.58710785608752\n",
      "Optimization restart 17/30, f = -52.4747586981847\n",
      "Optimization restart 18/30, f = -52.58710773580392\n",
      "Optimization restart 19/30, f = -52.474758678915805\n",
      "Optimization restart 20/30, f = -52.474760273447345\n",
      "Optimization restart 21/30, f = -52.587107850390986\n",
      "Optimization restart 22/30, f = -52.474759761928794\n",
      "Optimization restart 23/30, f = -52.47475864403901\n",
      "Optimization restart 24/30, f = -52.47475867837136\n",
      "Optimization restart 25/30, f = -52.474758683058724\n",
      "Optimization restart 26/30, f = -52.474758678899505\n",
      "Optimization restart 27/30, f = -52.47475868832241\n",
      "Optimization restart 28/30, f = -52.47475868005191\n",
      "Optimization restart 29/30, f = -52.58710765407581\n",
      "Optimization restart 30/30, f = -52.47475868694376\n",
      "[[ 0.37315583  2.53298227  0.00746171  0.32584685  0.425702  ]\n",
      " [ 0.45213063 -0.0759928   1.20009699  0.49521434  0.45754614]\n",
      " [ 0.19557226  2.05986054  2.19398766  0.33388413  0.44228233]\n",
      " [ 0.46611709 -2.21868726  0.05692199  0.43126418  0.30159158]\n",
      " [ 0.16498168  2.06796776 -0.00423536  0.12981474  0.24285195]\n",
      " [ 0.45094917  0.2194594  -2.48750102  0.2373112   0.33698198]\n",
      " [ 0.23263287 -1.4224408   0.72436626  0.13599267  0.25251965]\n",
      " [ 0.42071911 -1.52408417 -2.00584061  0.41297371  0.188994  ]\n",
      " [ 0.36082141  1.80253081 -2.2342661   0.42242934  0.29618033]\n",
      " [ 0.07733291  0.90317145  2.85581942  0.30269901  0.43866022]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "For model parameters [ 0.46611709 -2.21868726  0.05692199  0.43126418  0.30159158] takes 278.751412153244 S\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "For model parameters [ 0.45094917  0.2194594  -2.48750102  0.2373112   0.33698198] takes 98.51221108436584 S\n",
      "Working on 7/10 design\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 8/10 design\n",
      "For model parameters [ 0.42071911 -1.52408417 -2.00584061  0.41297371  0.188994  ] takes 99.82377409934998 S\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -68.28910695717559\n",
      "Optimization restart 2/30, f = -68.28910703507012\n",
      "Optimization restart 3/30, f = -68.28910700720891\n",
      "Optimization restart 4/30, f = -68.05551226262598\n",
      "Optimization restart 5/30, f = -68.28910702753734\n",
      "Optimization restart 6/30, f = -68.28910703754116\n",
      "Optimization restart 7/30, f = -68.28910703863379\n",
      "Optimization restart 8/30, f = -68.28910703777362\n",
      "Optimization restart 9/30, f = -68.28910702746546\n",
      "Optimization restart 10/30, f = -48.47631667498602\n",
      "Optimization restart 11/30, f = -68.0554655759691\n",
      "Optimization restart 12/30, f = -68.289106976818\n",
      "Optimization restart 13/30, f = -68.05551283835455\n",
      "Optimization restart 14/30, f = -68.28910700105368\n",
      "Optimization restart 15/30, f = -48.481402247307756\n",
      "Optimization restart 16/30, f = -48.486884262862475\n",
      "Optimization restart 17/30, f = -68.28910692427289\n",
      "Optimization restart 18/30, f = -48.48688290409914\n",
      "Optimization restart 19/30, f = -68.2891067562839\n",
      "Optimization restart 20/30, f = -68.28910703778239\n",
      "Optimization restart 21/30, f = -68.28910703882752\n",
      "Optimization restart 22/30, f = -68.28910703854481\n",
      "Optimization restart 23/30, f = -68.28910701250597\n",
      "Optimization restart 24/30, f = -68.28910687897152\n",
      "Optimization restart 25/30, f = -68.28910701677589\n",
      "Optimization restart 26/30, f = -48.479807737421076\n",
      "Optimization restart 27/30, f = -68.28910699362115\n",
      "Optimization restart 28/30, f = -68.05550653356536\n",
      "Optimization restart 29/30, f = -68.05561182634273\n",
      "Optimization restart 30/30, f = -68.28910703879578\n",
      "[[ 0.37315583  2.53298227  0.00746171  0.32584685  0.425702  ]\n",
      " [ 0.45213063 -0.0759928   1.20009699  0.49521434  0.45754614]\n",
      " [ 0.19557226  2.05986054  2.19398766  0.33388413  0.44228233]\n",
      " [ 0.46611709 -2.21868726  0.05692199  0.43126418  0.30159158]\n",
      " [ 0.16498168  2.06796776 -0.00423536  0.12981474  0.24285195]\n",
      " [ 0.45094917  0.2194594  -2.48750102  0.2373112   0.33698198]\n",
      " [ 0.23263287 -1.4224408   0.72436626  0.13599267  0.25251965]\n",
      " [ 0.42071911 -1.52408417 -2.00584061  0.41297371  0.188994  ]\n",
      " [ 0.36082141  1.80253081 -2.2342661   0.42242934  0.29618033]\n",
      " [ 0.07733291  0.90317145  2.85581942  0.30269901  0.43866022]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.45213063 -0.0759928   1.20009699  0.49521434  0.45754614] takes 60.306254863739014 S\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "For model parameters [ 0.46611709 -2.21868726  0.05692199  0.43126418  0.30159158] takes 278.9079370498657 S\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "For model parameters [ 0.45094917  0.2194594  -2.48750102  0.2373112   0.33698198] takes 98.66041588783264 S\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "For model parameters [ 0.42071911 -1.52408417 -2.00584061  0.41297371  0.188994  ] takes 100.27241396903992 S\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(55, 5)\n",
      "(55, 1)\n",
      "Optimizing the GP 0\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -64.49138068595026\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.2307865167846895  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  2.383141731083315e-05  |      +ve      |        \n",
      "Optimization restart 1/30, f = -66.69167029324493\n",
      "Optimization restart 2/30, f = -66.00108350257989\n",
      "Optimization restart 3/30, f = -66.69167016829859\n",
      "Optimization restart 4/30, f = -65.99922607687085\n",
      "Optimization restart 5/30, f = -66.69167030415068\n",
      "Optimization restart 6/30, f = -65.9991669776097\n",
      "Optimization restart 7/30, f = -66.6916703034941\n",
      "Optimization restart 8/30, f = -65.9995714967402\n",
      "Optimization restart 9/30, f = -65.99976511787412\n",
      "Optimization restart 10/30, f = -65.99914226527667\n",
      "Optimization restart 11/30, f = -66.69167030095002\n",
      "Optimization restart 12/30, f = -66.6916702710713\n",
      "Optimization restart 13/30, f = -47.87632768164567\n",
      "Optimization restart 14/30, f = -66.6916702866792\n",
      "Optimization restart 15/30, f = -66.69167030044011\n",
      "Optimization restart 16/30, f = -65.99960622180785\n",
      "Optimization restart 17/30, f = -66.69167030180834\n",
      "Optimization restart 18/30, f = -66.69167029196487\n",
      "Optimization restart 19/30, f = -65.99914477683213\n",
      "Optimization restart 20/30, f = -66.69167030529165\n",
      "Optimization restart 21/30, f = -66.69167027437828\n",
      "Optimization restart 22/30, f = -66.69166792900955\n",
      "Optimization restart 23/30, f = -66.69167030497691\n",
      "Optimization restart 24/30, f = -66.69167030146141\n",
      "Optimization restart 25/30, f = -66.6916702929128\n",
      "Optimization restart 26/30, f = -65.99928795331121\n",
      "Optimization restart 27/30, f = -65.99973670651542\n",
      "Optimization restart 28/30, f = -65.9995522886161\n",
      "Optimization restart 29/30, f = -66.00026437682925\n",
      "Optimization restart 30/30, f = -66.69167029959554\n",
      "Time it takes to do optimization 3.1098902225494385\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -66.69167030529165\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.22442518905554046  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.054449009560498e-05  |      +ve      |        \n",
      "(55, 5)\n",
      "(55, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -64.62286583261594\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.21622341208160972  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.990271154883409e-06  |      +ve      |        \n",
      "Optimization restart 1/30, f = -68.22252635645239\n",
      "Optimization restart 2/30, f = -68.22252674260388\n",
      "Optimization restart 3/30, f = -68.22252727704355\n",
      "Optimization restart 4/30, f = -68.22216486545048\n",
      "Optimization restart 5/30, f = -68.22252679709341\n",
      "Optimization restart 6/30, f = -68.22252691238388\n",
      "Optimization restart 7/30, f = -68.22252733939273\n",
      "Optimization restart 8/30, f = -68.22252692644753\n",
      "Optimization restart 9/30, f = -68.22241665596407\n",
      "Optimization restart 10/30, f = -68.22252689046799\n",
      "Optimization restart 11/30, f = -68.2225273067197\n",
      "Optimization restart 12/30, f = -68.22252702009035\n",
      "Optimization restart 13/30, f = -68.2225273357506\n",
      "Optimization restart 14/30, f = -68.22252715786456\n",
      "Optimization restart 15/30, f = -68.22252713874033\n",
      "Optimization restart 16/30, f = -48.4258538982493\n",
      "Optimization restart 17/30, f = -68.22252694506747\n",
      "Optimization restart 18/30, f = -68.22252621156628\n",
      "Optimization restart 19/30, f = -68.22252360265551\n",
      "Optimization restart 20/30, f = -68.22252700684554\n",
      "Optimization restart 21/30, f = -68.22252713473672\n",
      "Optimization restart 22/30, f = -68.22252060530717\n",
      "Optimization restart 23/30, f = -68.22252311195501\n",
      "Optimization restart 24/30, f = -68.22167654933992\n",
      "Optimization restart 25/30, f = -68.22252709918274\n",
      "Optimization restart 26/30, f = -68.22252723142094\n",
      "Optimization restart 27/30, f = -68.22252725058013\n",
      "Optimization restart 28/30, f = -68.22252714730595\n",
      "Optimization restart 29/30, f = -68.22252727387469\n",
      "Optimization restart 30/30, f = -68.22252106324268\n",
      "Time it takes to do optimization 3.6022489070892334\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -68.22252733939273\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.17960129005716674  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.821814971912773e-12  |      +ve      |        \n",
      "Trainng 2 gp for R_AA at 7 GeV\n",
      "Optimization restart 1/30, f = -68.22252734019226\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 2/30, f = -48.41809685438861\n",
      "Optimization restart 3/30, f = -68.22227435317416\n",
      "Optimization restart 4/30, f = -68.2225273059012\n",
      "Optimization restart 5/30, f = -68.22252615223198\n",
      "Optimization restart 6/30, f = -68.22252726317026\n",
      "Optimization restart 7/30, f = -68.22252114733024\n",
      "Optimization restart 8/30, f = -68.22206424149968\n",
      "Optimization restart 9/30, f = -68.22252543507858\n",
      "Optimization restart 10/30, f = -68.22252711015575\n",
      "Optimization restart 11/30, f = -68.22181406837026\n",
      "Optimization restart 12/30, f = -68.22252706491673\n",
      "Optimization restart 13/30, f = -68.22252592172435\n",
      "Optimization restart 14/30, f = -68.22181550084507\n",
      "Optimization restart 15/30, f = -68.2225274077919\n",
      "Optimization restart 16/30, f = -68.22252709668267\n",
      "Optimization restart 17/30, f = -68.22252715449483\n",
      "Optimization restart 18/30, f = -68.22252735717048\n",
      "Optimization restart 19/30, f = -68.22252740978354\n",
      "Optimization restart 20/30, f = -48.41557055652723\n",
      "Optimization restart 21/30, f = -68.22252581116274\n",
      "Optimization restart 22/30, f = -68.22252726077974\n",
      "Optimization restart 23/30, f = -68.22252570427165\n",
      "Optimization restart 24/30, f = -48.42585588158665\n",
      "Optimization restart 25/30, f = -68.22252726035302\n",
      "Optimization restart 26/30, f = -68.22244427438181\n",
      "Optimization restart 27/30, f = -68.22252668132086\n",
      "Optimization restart 28/30, f = -48.41591187515691\n",
      "Optimization restart 29/30, f = -68.2225271132704\n",
      "Optimization restart 30/30, f = -68.22252727747797\n",
      "[[ 0.20721191  1.18359245  1.61022134  0.39968364  0.28954666]\n",
      " [ 0.27329797 -2.41436368 -2.62008905  0.44442013  0.42018927]\n",
      " [ 0.32451707  0.1787443  -2.87130241  0.27296643  0.15544984]\n",
      " [ 0.15273197 -1.4840438   2.30233477  0.39870732  0.13599448]\n",
      " [ 0.15669291 -0.70108695 -0.96412075  0.16513625  0.19384647]\n",
      " [ 0.42103427 -1.59071016  0.52843562  0.40123282  0.30391914]\n",
      " [ 0.12356627  2.84460461 -1.50766717  0.35879786  0.34135593]\n",
      " [ 0.48122397  1.05757782 -1.11082298  0.39800606  0.2301795 ]\n",
      " [ 0.41409942  2.20808756 -1.14311015  0.16352415  0.33860383]\n",
      " [ 0.23839289 -0.67619294  2.6366139   0.15061298  0.28348797]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.27329797 -2.41436368 -2.62008905  0.44442013  0.42018927] takes 108.17821073532104 S\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "For model parameters [ 0.42103427 -1.59071016  0.52843562  0.40123282  0.30391914] takes 111.52449989318848 S\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -85.10526825582102\n",
      "Optimization restart 2/30, f = -85.105268242555\n",
      "Optimization restart 3/30, f = -85.10526823857019\n",
      "Optimization restart 4/30, f = -85.10526824628695\n",
      "Optimization restart 5/30, f = -85.10526825772833\n",
      "Optimization restart 6/30, f = -85.10526825833317\n",
      "Optimization restart 7/30, f = -85.10526888979078\n",
      "Optimization restart 8/30, f = -85.10527886375739\n",
      "Optimization restart 9/30, f = -85.10527335918894\n",
      "Optimization restart 10/30, f = -85.10526824125031\n",
      "Optimization restart 11/30, f = -85.1052682565884\n",
      "Optimization restart 12/30, f = -85.10526825659554\n",
      "Optimization restart 13/30, f = -85.1052682453321\n",
      "Optimization restart 14/30, f = -59.82727887272636\n",
      "Optimization restart 15/30, f = -59.82725058825012\n",
      "Optimization restart 16/30, f = -85.10526825599742\n",
      "Optimization restart 17/30, f = -85.10526825635768\n",
      "Optimization restart 18/30, f = -85.10526825962873\n",
      "Optimization restart 19/30, f = -85.10526825573749\n",
      "Optimization restart 20/30, f = -85.10526825658802\n",
      "Optimization restart 21/30, f = -85.10526825655857\n",
      "Optimization restart 22/30, f = -85.10526825185511\n",
      "Optimization restart 23/30, f = -85.10526825562577\n",
      "Optimization restart 24/30, f = -85.10526822187038\n",
      "Optimization restart 25/30, f = -85.10526956456658\n",
      "Optimization restart 26/30, f = -85.10526826631099\n",
      "Optimization restart 27/30, f = -85.10526827645504\n",
      "Optimization restart 28/30, f = -85.10526825577253\n",
      "Optimization restart 29/30, f = -85.10526822170826\n",
      "Optimization restart 30/30, f = -85.10526825436416\n",
      "[[ 0.20721191  1.18359245  1.61022134  0.39968364  0.28954666]\n",
      " [ 0.27329797 -2.41436368 -2.62008905  0.44442013  0.42018927]\n",
      " [ 0.32451707  0.1787443  -2.87130241  0.27296643  0.15544984]\n",
      " [ 0.15273197 -1.4840438   2.30233477  0.39870732  0.13599448]\n",
      " [ 0.15669291 -0.70108695 -0.96412075  0.16513625  0.19384647]\n",
      " [ 0.42103427 -1.59071016  0.52843562  0.40123282  0.30391914]\n",
      " [ 0.12356627  2.84460461 -1.50766717  0.35879786  0.34135593]\n",
      " [ 0.48122397  1.05757782 -1.11082298  0.39800606  0.2301795 ]\n",
      " [ 0.41409942  2.20808756 -1.14311015  0.16352415  0.33860383]\n",
      " [ 0.23839289 -0.67619294  2.6366139   0.15061298  0.28348797]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.27329797 -2.41436368 -2.62008905  0.44442013  0.42018927] takes 107.8488130569458 S\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "For model parameters [ 0.42103427 -1.59071016  0.52843562  0.40123282  0.30391914] takes 111.7251980304718 S\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(65, 5)\n",
      "(65, 1)\n",
      "Optimizing the GP 0\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -76.76501211787159\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.22442518905554046  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  5.054449009560498e-05  |      +ve      |        \n",
      "Optimization restart 1/30, f = -86.11225431031713\n",
      "Optimization restart 2/30, f = -86.11230711091264\n",
      "Optimization restart 3/30, f = -86.11230741373875\n",
      "Optimization restart 4/30, f = -86.11230740564316\n",
      "Optimization restart 5/30, f = -86.11230671668864\n",
      "Optimization restart 6/30, f = -86.11230773401124\n",
      "Optimization restart 7/30, f = -86.11230603570633\n",
      "Optimization restart 8/30, f = -86.11230367075763\n",
      "Optimization restart 9/30, f = -86.11230788713681\n",
      "Optimization restart 10/30, f = -86.1123077030784\n",
      "Optimization restart 11/30, f = -86.11230779276923\n",
      "Optimization restart 12/30, f = -86.11230770881537\n",
      "Optimization restart 13/30, f = -86.1123077192312\n",
      "Optimization restart 14/30, f = -86.11215290542492\n",
      "Optimization restart 15/30, f = -86.11230773082788\n",
      "Optimization restart 16/30, f = -86.11230699076913\n",
      "Optimization restart 17/30, f = -86.1123077754058\n",
      "Optimization restart 18/30, f = -86.11230759584063\n",
      "Optimization restart 19/30, f = -86.11230101308489\n",
      "Optimization restart 20/30, f = -86.11230745304981\n",
      "Optimization restart 21/30, f = -86.11230760244976\n",
      "Optimization restart 22/30, f = -86.11230464772018\n",
      "Optimization restart 23/30, f = -86.11230756410413\n",
      "Optimization restart 24/30, f = -86.11230742583663\n",
      "Optimization restart 25/30, f = -86.11230772564502\n",
      "Optimization restart 26/30, f = -86.11230769344792\n",
      "Optimization restart 27/30, f = -86.11230734580494\n",
      "Optimization restart 28/30, f = -86.11230789879403\n",
      "Optimization restart 29/30, f = -86.11230733955864\n",
      "Optimization restart 30/30, f = -86.11230729855824\n",
      "Time it takes to do optimization 4.526043176651001\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -86.11230789879403\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.16004329987952423  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  1.5124642062364335e-15  |      +ve      |        \n",
      "(65, 5)\n",
      "(65, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -76.7267836222891\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                 value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |   0.20648700388241348  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                  (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  3.61628247808548e-05  |      +ve      |        \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 1/30, f = -84.82768544159464\n",
      "Optimization restart 2/30, f = -84.8076978242158\n",
      "Optimization restart 3/30, f = -84.82768523917147\n",
      "Optimization restart 4/30, f = -84.82765229089252\n",
      "Optimization restart 5/30, f = -84.8276650793381\n",
      "Optimization restart 6/30, f = -84.82684276905931\n",
      "Optimization restart 7/30, f = -84.82766909584231\n",
      "Optimization restart 8/30, f = -84.82764397051668\n",
      "Optimization restart 9/30, f = -84.82768565389634\n",
      "Optimization restart 10/30, f = -84.80769752428256\n",
      "Optimization restart 11/30, f = -84.80769767067974\n",
      "Optimization restart 12/30, f = -84.80769731704598\n",
      "Optimization restart 13/30, f = -84.80769744226708\n",
      "Optimization restart 14/30, f = -84.82750374984843\n",
      "Optimization restart 15/30, f = -84.82767494916035\n",
      "Optimization restart 16/30, f = -84.80769786683636\n",
      "Optimization restart 17/30, f = -59.85904580534579\n",
      "Optimization restart 18/30, f = -84.8076978285539\n",
      "Optimization restart 19/30, f = -84.80769793970111\n",
      "Optimization restart 20/30, f = -84.807695789271\n",
      "Optimization restart 21/30, f = -84.80769699267071\n",
      "Optimization restart 22/30, f = -84.82753398262946\n",
      "Optimization restart 23/30, f = -84.8076944956668\n",
      "Optimization restart 24/30, f = -84.82761281874782\n",
      "Optimization restart 25/30, f = -84.82765767814408\n",
      "Optimization restart 26/30, f = -84.82764475002615\n",
      "Optimization restart 27/30, f = -84.80769381070618\n",
      "Optimization restart 28/30, f = -84.80769783277778\n",
      "Optimization restart 29/30, f = -84.82768459019917\n",
      "Optimization restart 30/30, f = -84.80769795763709\n",
      "Time it takes to do optimization 3.996548891067505\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -84.82768565389634\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.1756016943970174  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  4.164966594315066e-12  |      +ve      |        \n",
      "Trainng 0 gp for R_AA at 3 GeV\n",
      "Optimization restart 1/30, f = -86.11230790349097\n",
      "Optimization restart 2/30, f = -86.11230626530009\n",
      "Optimization restart 3/30, f = -86.1123075445182\n",
      "Optimization restart 4/30, f = -86.11230751661702\n",
      "Optimization restart 5/30, f = -86.11230586190644\n",
      "Optimization restart 6/30, f = -86.11230720572273\n",
      "Optimization restart 7/30, f = -86.11230779185027\n",
      "Optimization restart 8/30, f = -86.11230500171871\n",
      "Optimization restart 9/30, f = -86.11230100744007\n",
      "Optimization restart 10/30, f = -86.11230687618612\n",
      "Optimization restart 11/30, f = -86.112307384356\n",
      "Optimization restart 12/30, f = -86.1114789061595\n",
      "Optimization restart 13/30, f = -86.11230658625084\n",
      "Optimization restart 14/30, f = -86.11230746146629\n",
      "Optimization restart 15/30, f = -86.11230723750455\n",
      "Optimization restart 16/30, f = -86.11230732742497\n",
      "Optimization restart 17/30, f = -86.11230735239633\n",
      "Optimization restart 18/30, f = -86.11230765703563\n",
      "Optimization restart 19/30, f = -86.11230745938732\n",
      "Optimization restart 20/30, f = -86.11230750251583\n",
      "Optimization restart 21/30, f = -86.11230772500562\n",
      "Optimization restart 22/30, f = -86.1123054423175\n",
      "Optimization restart 23/30, f = -86.11230757009642\n",
      "Optimization restart 24/30, f = -86.11223633661609\n",
      "Optimization restart 25/30, f = -86.1123076558333\n",
      "Optimization restart 26/30, f = -86.11230366194873\n",
      "Optimization restart 27/30, f = -86.11230606913048\n",
      "Optimization restart 28/30, f = -86.11230726346945\n",
      "Optimization restart 29/30, f = -86.11230663766861\n",
      "Optimization restart 30/30, f = -86.11230733418915\n",
      "[[ 0.15492536  1.80254214  2.08865081  0.17480233  0.20570096]\n",
      " [ 0.46941975 -2.59083867 -0.69658744  0.20840293  0.34813485]\n",
      " [ 0.28600507 -1.69397426  2.77783603  0.49622435  0.37073032]\n",
      " [ 0.23875894 -0.3816663  -0.34854411  0.38400351  0.45945268]\n",
      " [ 0.44937493 -1.33486246  1.3507906   0.32059235  0.33343752]\n",
      " [ 0.08628658 -0.34665506 -1.32062796  0.10661126  0.39130178]\n",
      " [ 0.27962622 -2.4010189  -1.36730209  0.13859322  0.41933343]\n",
      " [ 0.20553215  2.28822326  2.66315297  0.49593009  0.39814146]\n",
      " [ 0.26157877 -2.25262808 -1.75731965  0.26583247  0.46149186]\n",
      " [ 0.14508772 -2.40358254  1.37310407  0.46594691  0.37279815]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.46941975 -2.59083867 -0.69658744  0.20840293  0.34813485] takes 80.91592192649841 S\n",
      "Working on 3/10 design\n",
      "For model parameters [ 0.28600507 -1.69397426  2.77783603  0.49622435  0.37073032] takes 87.72575092315674 S\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "For model parameters [ 0.44937493 -1.33486246  1.3507906   0.32059235  0.33343752] takes 86.53295731544495 S\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -95.8667010691162\n",
      "Optimization restart 2/30, f = -99.3317056535452\n",
      "Optimization restart 3/30, f = -99.33170566378243\n",
      "Optimization restart 4/30, f = -99.33170564135796\n",
      "Optimization restart 5/30, f = -99.33170564246677\n",
      "Optimization restart 6/30, f = -99.33170558819447\n",
      "Optimization restart 7/30, f = -95.8668721134637\n",
      "Optimization restart 8/30, f = -99.33170566154169\n",
      "Optimization restart 9/30, f = -99.3317055938445\n",
      "Optimization restart 10/30, f = -99.33170505138766\n",
      "Optimization restart 11/30, f = -99.33170561624485\n",
      "Optimization restart 12/30, f = -99.33170566116894\n",
      "Optimization restart 13/30, f = -99.33170562587958\n",
      "Optimization restart 14/30, f = -99.33170557274173\n",
      "Optimization restart 15/30, f = -95.86697315674184\n",
      "Optimization restart 16/30, f = -99.33170565099537\n",
      "Optimization restart 17/30, f = -99.33170565906937\n",
      "Optimization restart 18/30, f = -99.3316569340359\n",
      "Optimization restart 19/30, f = -99.33170566612174\n",
      "Optimization restart 20/30, f = -99.33170562951106\n",
      "Optimization restart 21/30, f = -99.33170562207732\n",
      "Optimization restart 22/30, f = -99.3317056505133\n",
      "Optimization restart 23/30, f = -99.33170550195044\n",
      "Optimization restart 24/30, f = -99.33170565007512\n",
      "Optimization restart 25/30, f = -99.33170516797152\n",
      "Optimization restart 26/30, f = -99.33170564878534\n",
      "Optimization restart 27/30, f = -99.33170561492791\n",
      "Optimization restart 28/30, f = -99.33170563550395\n",
      "Optimization restart 29/30, f = -99.33170555007976\n",
      "Optimization restart 30/30, f = -99.33170554933906\n",
      "[[ 0.15492536  1.80254214  2.08865081  0.17480233  0.20570096]\n",
      " [ 0.46941975 -2.59083867 -0.69658744  0.20840293  0.34813485]\n",
      " [ 0.28600507 -1.69397426  2.77783603  0.49622435  0.37073032]\n",
      " [ 0.23875894 -0.3816663  -0.34854411  0.38400351  0.45945268]\n",
      " [ 0.44937493 -1.33486246  1.3507906   0.32059235  0.33343752]\n",
      " [ 0.08628658 -0.34665506 -1.32062796  0.10661126  0.39130178]\n",
      " [ 0.27962622 -2.4010189  -1.36730209  0.13859322  0.41933343]\n",
      " [ 0.20553215  2.28822326  2.66315297  0.49593009  0.39814146]\n",
      " [ 0.26157877 -2.25262808 -1.75731965  0.26583247  0.46149186]\n",
      " [ 0.14508772 -2.40358254  1.37310407  0.46594691  0.37279815]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.46941975 -2.59083867 -0.69658744  0.20840293  0.34813485] takes 80.14905095100403 S\n",
      "Working on 3/10 design\n",
      "For model parameters [ 0.28600507 -1.69397426  2.77783603  0.49622435  0.37073032] takes 86.76875686645508 S\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "For model parameters [ 0.44937493 -1.33486246  1.3507906   0.32059235  0.33343752] takes 86.17456793785095 S\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(75, 5)\n",
      "(75, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -75.6473998416287\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.1756016943970174  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  4.164966594315066e-12  |      +ve      |        \n",
      "Optimization restart 1/30, f = -98.31299474649819\n",
      "Optimization restart 2/30, f = -101.80814587319577\n",
      "Optimization restart 3/30, f = -101.80814541851232\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 4/30, f = -101.80814577110684\n",
      "Optimization restart 5/30, f = -101.80814564979457\n",
      "Optimization restart 6/30, f = -101.80814589150728\n",
      "Optimization restart 7/30, f = -101.80788607669358\n",
      "Optimization restart 8/30, f = -101.80814589290574\n",
      "Optimization restart 9/30, f = -101.80814589456543\n",
      "Optimization restart 10/30, f = -101.80814589561597\n",
      "Optimization restart 11/30, f = -101.80814589526969\n",
      "Optimization restart 12/30, f = -101.8081457582794\n",
      "Optimization restart 13/30, f = -101.80814589592416\n",
      "Optimization restart 14/30, f = -30.191545034086616\n",
      "Optimization restart 15/30, f = -101.80814579827573\n",
      "Optimization restart 16/30, f = -101.80814588932171\n",
      "Optimization restart 17/30, f = -101.80814579855672\n",
      "Optimization restart 18/30, f = -101.80812063528253\n",
      "Optimization restart 19/30, f = -101.72192941904571\n",
      "Optimization restart 20/30, f = -101.80814562664156\n",
      "Optimization restart 21/30, f = -101.80814589045937\n",
      "Optimization restart 22/30, f = -101.80814589013849\n",
      "Optimization restart 23/30, f = -101.80814486637352\n",
      "Optimization restart 24/30, f = -101.80814587987138\n",
      "Optimization restart 25/30, f = -101.80814589559367\n",
      "Optimization restart 26/30, f = -101.80814588818802\n",
      "Optimization restart 27/30, f = -101.80814589251277\n",
      "Optimization restart 28/30, f = -101.80814589609648\n",
      "Optimization restart 29/30, f = -101.80814589035441\n",
      "Optimization restart 30/30, f = -101.80814578267206\n",
      "Time it takes to do optimization 3.846181869506836\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -101.80814589609648\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.19699647243911725  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.0002510454541148768  |      +ve      |        \n",
      "(75, 5)\n",
      "(75, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -91.77756800300351\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.12848512370628792  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  7.469588449269823e-10  |      +ve      |        \n",
      "Optimization restart 1/30, f = -99.50319978676744\n",
      "Optimization restart 2/30, f = -103.7155884144658\n",
      "Optimization restart 3/30, f = -103.65485227799525\n",
      "Optimization restart 4/30, f = -103.71558850999935\n",
      "Optimization restart 5/30, f = -103.71558851103978\n",
      "Optimization restart 6/30, f = -103.71558847572621\n",
      "Optimization restart 7/30, f = -103.7155885145585\n",
      "Optimization restart 8/30, f = -103.71558802830896\n",
      "Optimization restart 9/30, f = -103.71558839084778\n",
      "Optimization restart 10/30, f = -103.6350796741034\n",
      "Optimization restart 11/30, f = -103.62839782795663\n",
      "Optimization restart 12/30, f = -103.71558850233775\n",
      "Optimization restart 13/30, f = -103.71558840390503\n",
      "Optimization restart 14/30, f = -103.71558812323997\n",
      "Optimization restart 15/30, f = -103.7155884468777\n",
      "Optimization restart 16/30, f = -103.71558848858183\n",
      "Optimization restart 17/30, f = -103.71558851410401\n",
      "Optimization restart 18/30, f = -103.71558832846728\n",
      "Optimization restart 19/30, f = -103.71558834639953\n",
      "Optimization restart 20/30, f = -103.7155885029915\n",
      "Optimization restart 21/30, f = -103.71558851441648\n",
      "Optimization restart 22/30, f = -103.63383534222478\n",
      "Optimization restart 23/30, f = -19.432416737375704\n",
      "Optimization restart 24/30, f = -103.7155884566428\n",
      "Optimization restart 25/30, f = -103.71558844825273\n",
      "Optimization restart 26/30, f = -103.71558850486267\n",
      "Optimization restart 27/30, f = -103.7155884742339\n",
      "Optimization restart 28/30, f = -103.71558849647077\n",
      "Optimization restart 29/30, f = -103.71558851534783\n",
      "Optimization restart 30/30, f = -64.4167393240779\n",
      "Time it takes to do optimization 4.375380039215088\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -103.71558851534783\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.1938660845188217  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00025661562468379836  |      +ve      |        \n",
      "Trainng 1 gp for R_AA at 5 GeV\n",
      "Optimization restart 1/30, f = -101.80814589609648\n",
      "Optimization restart 2/30, f = -101.80814575276833\n",
      "Optimization restart 3/30, f = -101.80814589555398\n",
      "Optimization restart 4/30, f = -98.19675076660135\n",
      "Optimization restart 5/30, f = -101.80814574235717\n",
      "Optimization restart 6/30, f = -101.8081458655183\n",
      "Optimization restart 7/30, f = -101.80814588986718\n",
      "Optimization restart 8/30, f = -98.19913147600928\n",
      "Optimization restart 9/30, f = -101.80814589383417\n",
      "Optimization restart 10/30, f = -101.80814589300813\n",
      "Optimization restart 11/30, f = -101.80814589505945\n",
      "Optimization restart 12/30, f = -101.80814589489898\n",
      "Optimization restart 13/30, f = -101.8081458694417\n",
      "Optimization restart 14/30, f = -101.80814589606683\n",
      "Optimization restart 15/30, f = -101.80814585696422\n",
      "Optimization restart 16/30, f = -101.80814589339602\n",
      "Optimization restart 17/30, f = -101.80814589585867\n",
      "Optimization restart 18/30, f = -101.80814589436403\n",
      "Optimization restart 19/30, f = -101.8081457711917\n",
      "Optimization restart 20/30, f = -101.80814589523948\n",
      "Optimization restart 21/30, f = -101.80814587620644\n",
      "Optimization restart 22/30, f = -101.80814589457574\n",
      "Optimization restart 23/30, f = -101.80814585120382\n",
      "Optimization restart 24/30, f = -101.80814586046756\n",
      "Optimization restart 25/30, f = -101.8081458957747\n",
      "Optimization restart 26/30, f = -101.80814581750838\n",
      "Optimization restart 27/30, f = -101.80786288775806\n",
      "Optimization restart 28/30, f = -101.8081457881908\n",
      "Optimization restart 29/30, f = -101.80814588843023\n",
      "Optimization restart 30/30, f = -101.80814589608035\n",
      "[[ 0.27728115  0.87055939 -0.54899899  0.1256951   0.3632088 ]\n",
      " [ 0.0513998  -2.94904513 -1.99709923  0.45639034  0.40527121]\n",
      " [ 0.197498   -2.86556605  1.33600849  0.18453419  0.44568977]\n",
      " [ 0.5        -1.3919409  -2.7756736   0.30342381  0.19884851]\n",
      " [ 0.2601877   2.98705961 -2.32159143  0.36309977  0.11299373]\n",
      " [ 0.05213941  1.25269375  0.26356234  0.139182    0.30903412]\n",
      " [ 0.34463494 -2.88781489  2.72957249  0.3081426   0.17090272]\n",
      " [ 0.0703001   2.66945266  2.34415785  0.47234024  0.44203004]\n",
      " [ 0.49417052  2.29825003  2.40891531  0.41457114  0.14171497]\n",
      " [ 0.37460495  1.52217145  2.62959511  0.20235626  0.43154197]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "For model parameters [ 0.5        -1.3919409  -2.7756736   0.30342381  0.19884851] takes 99.910395860672 S\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "For model parameters [ 0.34463494 -2.88781489  2.72957249  0.3081426   0.17090272] takes 100.48165082931519 S\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -117.44897688161959\n",
      "Optimization restart 2/30, f = -117.45282046757609\n",
      "Optimization restart 3/30, f = -117.45282091570908\n",
      "Optimization restart 4/30, f = -117.45282111106329\n",
      "Optimization restart 5/30, f = -117.45281755746643\n",
      "Optimization restart 6/30, f = -117.4528210342292\n",
      "Optimization restart 7/30, f = -117.45282110278978\n",
      "Optimization restart 8/30, f = -117.45282108717184\n",
      "Optimization restart 9/30, f = -117.45282104073988\n",
      "Optimization restart 10/30, f = -117.4528210832132\n",
      "Optimization restart 11/30, f = -117.4528208216966\n",
      "Optimization restart 12/30, f = -117.452821113568\n",
      "Optimization restart 13/30, f = -117.45282109104744\n",
      "Optimization restart 14/30, f = -117.45282090334875\n",
      "Optimization restart 15/30, f = -117.45282110917168\n",
      "Optimization restart 16/30, f = -117.45256780052483\n",
      "Optimization restart 17/30, f = -117.45282106737706\n",
      "Optimization restart 18/30, f = -117.45282102200204\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 19/30, f = -115.54965424538554\n",
      "Optimization restart 20/30, f = -117.452821107397\n",
      "Optimization restart 21/30, f = -117.4528210354799\n",
      "Optimization restart 22/30, f = -117.45282108398582\n",
      "Optimization restart 23/30, f = -117.45282111283761\n",
      "Optimization restart 24/30, f = -117.4528211097388\n",
      "Optimization restart 25/30, f = -117.45282111313068\n",
      "Optimization restart 26/30, f = -117.45282111001114\n",
      "Optimization restart 27/30, f = -117.45282100115392\n",
      "Optimization restart 28/30, f = -117.37507632685127\n",
      "Optimization restart 29/30, f = -117.45282110840668\n",
      "Optimization restart 30/30, f = -117.45282088581888\n",
      "[[ 0.27728115  0.87055939 -0.54899899  0.1256951   0.3632088 ]\n",
      " [ 0.0513998  -2.94904513 -1.99709923  0.45639034  0.40527121]\n",
      " [ 0.197498   -2.86556605  1.33600849  0.18453419  0.44568977]\n",
      " [ 0.5        -1.3919409  -2.7756736   0.30342381  0.19884851]\n",
      " [ 0.2601877   2.98705961 -2.32159143  0.36309977  0.11299373]\n",
      " [ 0.05213941  1.25269375  0.26356234  0.139182    0.30903412]\n",
      " [ 0.34463494 -2.88781489  2.72957249  0.3081426   0.17090272]\n",
      " [ 0.0703001   2.66945266  2.34415785  0.47234024  0.44203004]\n",
      " [ 0.49417052  2.29825003  2.40891531  0.41457114  0.14171497]\n",
      " [ 0.37460495  1.52217145  2.62959511  0.20235626  0.43154197]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "For model parameters [ 0.5        -1.3919409  -2.7756736   0.30342381  0.19884851] takes 99.90730690956116 S\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "For model parameters [ 0.34463494 -2.88781489  2.72957249  0.3081426   0.17090272] takes 101.69977617263794 S\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(85, 5)\n",
      "(85, 1)\n",
      "Optimizing the GP 0\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -108.59374626156011\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.20709449401353447  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00022784792875472858  |      +ve      |        \n",
      "Optimization restart 1/30, f = -113.87069413516457\n",
      "Optimization restart 2/30, f = -113.8706941178559\n",
      "Optimization restart 3/30, f = -113.87069417269213\n",
      "Optimization restart 4/30, f = -113.87069415695063\n",
      "Optimization restart 5/30, f = -113.87069417551524\n",
      "Optimization restart 6/30, f = -113.87069363904396\n",
      "Optimization restart 7/30, f = -113.87069406421404\n",
      "Optimization restart 8/30, f = -113.87069394181766\n",
      "Optimization restart 9/30, f = -113.87069411722388\n",
      "Optimization restart 10/30, f = -112.08879903245288\n",
      "Optimization restart 11/30, f = -113.87069414486072\n",
      "Optimization restart 12/30, f = -113.87069416427288\n",
      "Optimization restart 13/30, f = -113.87069385187179\n",
      "Optimization restart 14/30, f = -113.87069396250875\n",
      "Optimization restart 15/30, f = -113.87069404702956\n",
      "Optimization restart 16/30, f = -113.87069417524899\n",
      "Optimization restart 17/30, f = -113.87069417465078\n",
      "Optimization restart 18/30, f = -113.87069417214703\n",
      "Optimization restart 19/30, f = -113.87069417577595\n",
      "Optimization restart 20/30, f = -113.87069416793376\n",
      "Optimization restart 21/30, f = -113.87069414735868\n",
      "Optimization restart 22/30, f = -113.87069416985156\n",
      "Optimization restart 23/30, f = -113.87069372284232\n",
      "Optimization restart 24/30, f = -113.8706941749571\n",
      "Optimization restart 25/30, f = -113.87069417562972\n",
      "Optimization restart 26/30, f = -113.87069412367788\n",
      "Optimization restart 27/30, f = -113.87069413369609\n",
      "Optimization restart 28/30, f = -113.87069409127388\n",
      "Optimization restart 29/30, f = -113.87069417258442\n",
      "Optimization restart 30/30, f = -113.87069389653928\n",
      "Time it takes to do optimization 3.066087007522583\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -113.87069417577595\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.16590771112641556  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00014099088708005217  |      +ve      |        \n",
      "(85, 5)\n",
      "(85, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -119.87096922809081\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.1938660845188217  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00025661562468379836  |      +ve      |        \n",
      "Optimization restart 1/30, f = -120.4323073900984\n",
      "Optimization restart 2/30, f = -120.42397138814276\n",
      "Optimization restart 3/30, f = -120.43263038294461\n",
      "Optimization restart 4/30, f = -120.42926323720431\n",
      "Optimization restart 5/30, f = -120.42018665788083\n",
      "Optimization restart 6/30, f = -120.43263035498858\n",
      "Optimization restart 7/30, f = -116.79259581577631\n",
      "Optimization restart 8/30, f = -120.426323627422\n",
      "Optimization restart 9/30, f = -120.43262998027653\n",
      "Optimization restart 10/30, f = -120.43262990642745\n",
      "Optimization restart 11/30, f = -120.4326303625153\n",
      "Optimization restart 12/30, f = -120.41166136221115\n",
      "Optimization restart 13/30, f = -116.79259550725101\n",
      "Optimization restart 14/30, f = -120.4324804771095\n",
      "Optimization restart 15/30, f = -120.43016481955907\n",
      "Optimization restart 16/30, f = -116.79225099737863\n",
      "Optimization restart 17/30, f = -120.42488983456334\n",
      "Optimization restart 18/30, f = -72.92881321105097\n",
      "Optimization restart 19/30, f = -116.79259536384939\n",
      "Optimization restart 20/30, f = -120.43166693238882\n",
      "Optimization restart 21/30, f = -120.43248970875942\n",
      "Optimization restart 22/30, f = -116.79259525666652\n",
      "Optimization restart 23/30, f = -120.42481471093797\n",
      "Optimization restart 24/30, f = -120.43263038441613\n",
      "Optimization restart 25/30, f = -120.42913700948887\n",
      "Optimization restart 26/30, f = -120.40560564875673\n",
      "Optimization restart 27/30, f = -116.79259506812522\n",
      "Optimization restart 28/30, f = -120.41694986576105\n",
      "Optimization restart 29/30, f = -72.92882329619918\n",
      "Optimization restart 30/30, f = -116.79259274186624\n",
      "Time it takes to do optimization 4.064213037490845\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -120.43263038441613\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.2054742371086251  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00030307598403886806  |      +ve      |        \n",
      "Trainng 2 gp for R_AA at 7 GeV\n",
      "Optimization restart 1/30, f = -120.43263038479361\n",
      "Optimization restart 2/30, f = -120.40887611348043\n",
      "Optimization restart 3/30, f = -116.79259504603118\n",
      "Optimization restart 4/30, f = -116.79259532754538\n",
      "Optimization restart 5/30, f = -120.4192746461263\n",
      "Optimization restart 6/30, f = -116.79248227092913\n",
      "Optimization restart 7/30, f = -116.79259552857545\n",
      "Optimization restart 8/30, f = -120.42728487652138\n",
      "Optimization restart 9/30, f = -116.79248973134317\n",
      "Optimization restart 10/30, f = -116.79259556624973\n",
      "Optimization restart 11/30, f = -120.43262718801081\n",
      "Optimization restart 12/30, f = -120.42599928334869\n",
      "Optimization restart 13/30, f = -116.7925949713896\n",
      "Optimization restart 14/30, f = -120.41535653615011\n",
      "Optimization restart 15/30, f = -120.43262748981981\n",
      "Optimization restart 16/30, f = -72.92882301242349\n",
      "Optimization restart 17/30, f = -116.79259533824887\n",
      "Optimization restart 18/30, f = -120.424097602376\n",
      "Optimization restart 19/30, f = -116.79259518200965\n",
      "Optimization restart 20/30, f = -116.79259367604735\n",
      "Optimization restart 21/30, f = -116.79259526986306\n",
      "Optimization restart 22/30, f = -116.7925344119462\n",
      "Optimization restart 23/30, f = -120.43112270563815\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 24/30, f = -120.42649695928236\n",
      "Optimization restart 25/30, f = -116.7925912629752\n",
      "Optimization restart 26/30, f = -116.79259494444624\n",
      "Optimization restart 27/30, f = -116.7925937278209\n",
      "Optimization restart 28/30, f = -116.79259503816633\n",
      "Optimization restart 29/30, f = -120.43263035494549\n",
      "Optimization restart 30/30, f = -116.79259503283279\n",
      "[[ 0.33614374  0.6903296   0.66141064  0.49225251  0.47905288]\n",
      " [ 0.05992731  2.86702524 -0.08501915  0.29495999  0.23185522]\n",
      " [ 0.24884015  2.91699549 -0.55373538  0.1456469   0.47149217]\n",
      " [ 0.44456955  1.9888233   0.79646431  0.10305734  0.36239242]\n",
      " [ 0.49376508 -2.64950195  2.14931667  0.44996862  0.32884975]\n",
      " [ 0.407681    2.94087412  2.02007716  0.38046025  0.3337377 ]\n",
      " [ 0.05426177 -1.87175966 -1.49325813  0.1006395   0.16508655]\n",
      " [ 0.3840981  -1.1383857   2.7000507   0.20290489  0.33462974]\n",
      " [ 0.13366212 -1.45043938  0.72715925  0.25147682  0.27743218]\n",
      " [ 0.16704194  2.94028955  0.8913548   0.19837293  0.3133901 ]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "For model parameters [0.44456955 1.9888233  0.79646431 0.10305734 0.36239242] takes 128.20711088180542 S\n",
      "Working on 5/10 design\n",
      "For model parameters [ 0.49376508 -2.64950195  2.14931667  0.44996862  0.32884975] takes more than 10 mins\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -139.43082718291086\n",
      "Optimization restart 2/30, f = -138.85352485272105\n",
      "Optimization restart 3/30, f = -138.8535248146774\n",
      "Optimization restart 4/30, f = -138.0174432172728\n",
      "Optimization restart 5/30, f = -138.85349680232514\n",
      "Optimization restart 6/30, f = -138.85352426141392\n",
      "Optimization restart 7/30, f = -138.85352469414212\n",
      "Optimization restart 8/30, f = -138.0174432186501\n",
      "Optimization restart 9/30, f = -138.85352467024086\n",
      "Optimization restart 10/30, f = -138.85352035703318\n",
      "Optimization restart 11/30, f = -138.85352441765284\n",
      "Optimization restart 12/30, f = -138.85352108268552\n",
      "Optimization restart 13/30, f = -138.85352453274504\n",
      "Optimization restart 14/30, f = -138.85352459911314\n",
      "Optimization restart 15/30, f = -138.85352447614594\n",
      "Optimization restart 16/30, f = -138.85352320542577\n",
      "Optimization restart 17/30, f = -138.8535229433783\n",
      "Optimization restart 18/30, f = -138.8535245972835\n",
      "Optimization restart 19/30, f = -138.85352499445378\n",
      "Optimization restart 20/30, f = -138.8535242726885\n",
      "Optimization restart 21/30, f = -138.85352472550028\n",
      "Optimization restart 22/30, f = -138.8535226681534\n",
      "Optimization restart 23/30, f = -138.85352449602527\n",
      "Optimization restart 24/30, f = -138.8535245754772\n",
      "Optimization restart 25/30, f = -138.85351606399905\n",
      "Optimization restart 26/30, f = -138.85352430941677\n",
      "Optimization restart 27/30, f = -138.8535234260504\n",
      "Optimization restart 28/30, f = -138.85351890199243\n",
      "Optimization restart 29/30, f = -138.85352470746435\n",
      "Optimization restart 30/30, f = -138.85352432174125\n",
      "[[ 0.33614374  0.6903296   0.66141064  0.49225251  0.47905288]\n",
      " [ 0.05992731  2.86702524 -0.08501915  0.29495999  0.23185522]\n",
      " [ 0.24884015  2.91699549 -0.55373538  0.1456469   0.47149217]\n",
      " [ 0.44456955  1.9888233   0.79646431  0.10305734  0.36239242]\n",
      " [ 0.49376508 -2.64950195  2.14931667  0.44996862  0.32884975]\n",
      " [ 0.407681    2.94087412  2.02007716  0.38046025  0.3337377 ]\n",
      " [ 0.05426177 -1.87175966 -1.49325813  0.1006395   0.16508655]\n",
      " [ 0.3840981  -1.1383857   2.7000507   0.20290489  0.33462974]\n",
      " [ 0.13366212 -1.45043938  0.72715925  0.25147682  0.27743218]\n",
      " [ 0.16704194  2.94028955  0.8913548   0.19837293  0.3133901 ]]\n",
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "For model parameters [0.44456955 1.9888233  0.79646431 0.10305734 0.36239242] takes 128.50901007652283 S\n",
      "Working on 5/10 design\n",
      "For model parameters [ 0.49376508 -2.64950195  2.14931667  0.44996862  0.32884975] takes more than 10 mins\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(95, 5)\n",
      "(95, 1)\n",
      "Optimizing the GP 0\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -133.9141229779221\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.16590771112641556  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00014099088708005217  |      +ve      |        \n",
      "Optimization restart 1/30, f = -135.08477072363553\n",
      "Optimization restart 2/30, f = -135.08477098800103\n",
      "Optimization restart 3/30, f = -135.08477086248706\n",
      "Optimization restart 4/30, f = -135.08477100790756\n",
      "Optimization restart 5/30, f = -135.0847709850869\n",
      "Optimization restart 6/30, f = -135.08477086154807\n",
      "Optimization restart 7/30, f = -135.08477091723995\n",
      "Optimization restart 8/30, f = -135.08477094490496\n",
      "Optimization restart 9/30, f = -135.0847693970104\n",
      "Optimization restart 10/30, f = -135.08477027874608\n",
      "Optimization restart 11/30, f = -135.08477091569674\n",
      "Optimization restart 12/30, f = -135.08477100774547\n",
      "Optimization restart 13/30, f = -135.084771001828\n",
      "Optimization restart 14/30, f = -135.08477099431963\n",
      "Optimization restart 15/30, f = -135.08477018528316\n",
      "Optimization restart 16/30, f = -135.08477098380365\n",
      "Optimization restart 17/30, f = -135.08477056680132\n",
      "Optimization restart 18/30, f = -135.08477097069857\n",
      "Optimization restart 19/30, f = -135.084770810013\n",
      "Optimization restart 20/30, f = -135.08477096140115\n",
      "Optimization restart 21/30, f = -135.0847708310376\n",
      "Optimization restart 22/30, f = -135.08477089361452\n",
      "Optimization restart 23/30, f = -135.08477096980448\n",
      "Optimization restart 24/30, f = -135.0847709905683\n",
      "Optimization restart 25/30, f = -135.08477081778773\n",
      "Optimization restart 26/30, f = -135.0847710048572\n",
      "Optimization restart 27/30, f = -135.08477088351566\n",
      "Optimization restart 28/30, f = -135.08477091730447\n",
      "Optimization restart 29/30, f = -135.08477057404613\n",
      "Optimization restart 30/30, f = -135.08477100738526\n",
      "Time it takes to do optimization 3.2325398921966553\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -135.08477100790756\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.14897058818407075  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00011444718242652622  |      +ve      |        \n",
      "(95, 5)\n",
      "(95, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -135.22716583203746\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.20365779038392676  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00032704825378330137  |      +ve      |        \n",
      "Optimization restart 1/30, f = -136.52615155208906\n",
      "Optimization restart 2/30, f = -137.30965857520994\n",
      "Optimization restart 3/30, f = -137.30965872102405\n",
      "Optimization restart 4/30, f = -137.30965851026588\n",
      "Optimization restart 5/30, f = -137.30965872924008\n",
      "Optimization restart 6/30, f = -136.52650353648124\n",
      "Optimization restart 7/30, f = -137.30965873277745\n",
      "Optimization restart 8/30, f = -137.30965873094746\n",
      "Optimization restart 9/30, f = -137.30965871294126\n",
      "Optimization restart 10/30, f = -137.30965867751664\n",
      "Optimization restart 11/30, f = -137.30965868363097\n",
      "Optimization restart 12/30, f = -137.30965870831886\n",
      "Optimization restart 13/30, f = -137.3096583221107\n",
      "Optimization restart 14/30, f = -137.30965867061332\n",
      "Optimization restart 15/30, f = -137.30965838076798\n",
      "Optimization restart 16/30, f = -137.02566965736972\n",
      "Optimization restart 17/30, f = -137.30965872648943\n",
      "Optimization restart 18/30, f = -137.30965810955917\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 19/30, f = -137.30965873432447\n",
      "Optimization restart 20/30, f = -137.30965873249343\n",
      "Optimization restart 21/30, f = -137.30965870913232\n",
      "Optimization restart 22/30, f = -137.30965870937763\n",
      "Optimization restart 23/30, f = -137.3096587334818\n",
      "Optimization restart 24/30, f = -137.30965833160602\n",
      "Optimization restart 25/30, f = -137.30965805966954\n",
      "Optimization restart 26/30, f = -137.3096584381655\n",
      "Optimization restart 27/30, f = -137.30965841377343\n",
      "Optimization restart 28/30, f = -17.53824327577874\n",
      "Optimization restart 29/30, f = -137.3096586518829\n",
      "Optimization restart 30/30, f = -137.30965873409468\n",
      "Time it takes to do optimization 3.4261062145233154\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -137.30965873432447\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.125679203228355  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  2.734166660939177e-05  |      +ve      |        \n",
      "Trainng 0 gp for R_AA at 3 GeV\n",
      "Optimization restart 1/30, f = -135.08477100794786\n",
      "Optimization restart 2/30, f = -135.08477100743707\n",
      "Optimization restart 3/30, f = -135.08477098603444\n",
      "Optimization restart 4/30, f = -132.85393108692702\n",
      "Optimization restart 5/30, f = -135.0847710032964\n",
      "Optimization restart 6/30, f = -135.0847709667101\n",
      "Optimization restart 7/30, f = -113.43471560349775\n",
      "Optimization restart 8/30, f = -135.08477096895416\n",
      "Optimization restart 9/30, f = -135.0847709837432\n",
      "Optimization restart 10/30, f = -135.08477099778761\n",
      "Optimization restart 11/30, f = -135.08477097304134\n",
      "Optimization restart 12/30, f = -135.08477094486915\n",
      "Optimization restart 13/30, f = -135.08477100330666\n",
      "Optimization restart 14/30, f = -135.08477096428507\n",
      "Optimization restart 15/30, f = -135.08477086553825\n",
      "Optimization restart 16/30, f = -135.08477100309037\n",
      "Optimization restart 17/30, f = -135.08477024558306\n",
      "Optimization restart 18/30, f = -135.08477097135096\n",
      "Optimization restart 19/30, f = -135.0847700799303\n",
      "Optimization restart 20/30, f = -135.08477096852815\n",
      "Optimization restart 21/30, f = -135.08477065508754\n",
      "Optimization restart 22/30, f = -135.0847708983136\n",
      "Optimization restart 23/30, f = -135.0847708893093\n",
      "Optimization restart 24/30, f = -135.08477099997555\n",
      "Optimization restart 25/30, f = -135.08477100063766\n",
      "Optimization restart 26/30, f = -135.08477099101023\n",
      "Optimization restart 27/30, f = -135.08477099637844\n",
      "Optimization restart 28/30, f = -135.08477086517166\n",
      "Optimization restart 29/30, f = -135.08476495901806\n",
      "Optimization restart 30/30, f = -135.08477082528435\n",
      "[[ 0.48065949  2.39777458 -2.26806392  0.27147681  0.41054358]\n",
      " [ 0.49845704  1.16864462 -2.40716525  0.11408599  0.34816877]\n",
      " [ 0.23170891  0.32304633 -2.4986695   0.23017368  0.44541719]\n",
      " [ 0.05568936 -1.40035924  1.60934406  0.47364983  0.11021991]\n",
      " [ 0.36045656 -0.01657237  0.52673261  0.44655463  0.14032583]\n",
      " [ 0.0883465   2.91376223 -2.56831336  0.45731893  0.44309717]\n",
      " [ 0.26287634  1.72901448 -2.94785581  0.12131059  0.19049594]\n",
      " [ 0.4755844  -1.03144363 -2.6357408   0.47645597  0.24217227]\n",
      " [ 0.10526143 -2.86103619 -2.36185797  0.18543616  0.1080111 ]\n",
      " [ 0.35646783 -2.98868554 -2.9243399   0.20739873  0.12546668]]\n",
      "Working on 1/10 design\n",
      "For model parameters [ 0.48065949  2.39777458 -2.26806392  0.27147681  0.41054358] takes 80.84480690956116 S\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.49845704  1.16864462 -2.40716525  0.11408599  0.34816877] takes 147.51336812973022 S\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "For model parameters [ 0.4755844  -1.03144363 -2.6357408   0.47645597  0.24217227] takes 120.80578517913818 S\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "Optimization restart 1/30, f = -136.39692807812102\n",
      "Optimization restart 2/30, f = -136.3969280783959\n",
      "Optimization restart 3/30, f = -136.3969280764047\n",
      "Optimization restart 4/30, f = -136.39692805548685\n",
      "Optimization restart 5/30, f = -136.39692807764624\n",
      "Optimization restart 6/30, f = -136.39692799841998\n",
      "Optimization restart 7/30, f = -136.39692808513\n",
      "Optimization restart 8/30, f = -136.3969280547817\n",
      "Optimization restart 9/30, f = -136.3969280833153\n",
      "Optimization restart 10/30, f = -136.39692808596385\n",
      "Optimization restart 11/30, f = -136.39692808668548\n",
      "Optimization restart 12/30, f = -136.39692802269835\n",
      "Optimization restart 13/30, f = -136.396928083401\n",
      "Optimization restart 14/30, f = -136.3969280858568\n",
      "Optimization restart 15/30, f = -136.39692808865811\n",
      "Optimization restart 16/30, f = -136.39692808551268\n",
      "Optimization restart 17/30, f = -136.39692804222582\n",
      "Optimization restart 18/30, f = -136.39692788312908\n",
      "Optimization restart 19/30, f = -136.39692807055235\n",
      "Optimization restart 20/30, f = -136.39692808863617\n",
      "Optimization restart 21/30, f = -136.39692808650253\n",
      "Optimization restart 22/30, f = -136.39692806646465\n",
      "Optimization restart 23/30, f = -136.39692808483662\n",
      "Optimization restart 24/30, f = -136.39692806539762\n",
      "Optimization restart 25/30, f = -136.3969280886609\n",
      "Optimization restart 26/30, f = -136.39692801942277\n",
      "Optimization restart 27/30, f = -136.3969280826922\n",
      "Optimization restart 28/30, f = -136.39692808455047\n",
      "Optimization restart 29/30, f = -129.75410051580982\n",
      "Optimization restart 30/30, f = -136.39692808166808\n",
      "[[ 0.48065949  2.39777458 -2.26806392  0.27147681  0.41054358]\n",
      " [ 0.49845704  1.16864462 -2.40716525  0.11408599  0.34816877]\n",
      " [ 0.23170891  0.32304633 -2.4986695   0.23017368  0.44541719]\n",
      " [ 0.05568936 -1.40035924  1.60934406  0.47364983  0.11021991]\n",
      " [ 0.36045656 -0.01657237  0.52673261  0.44655463  0.14032583]\n",
      " [ 0.0883465   2.91376223 -2.56831336  0.45731893  0.44309717]\n",
      " [ 0.26287634  1.72901448 -2.94785581  0.12131059  0.19049594]\n",
      " [ 0.4755844  -1.03144363 -2.6357408   0.47645597  0.24217227]\n",
      " [ 0.10526143 -2.86103619 -2.36185797  0.18543616  0.1080111 ]\n",
      " [ 0.35646783 -2.98868554 -2.9243399   0.20739873  0.12546668]]\n",
      "Working on 1/10 design\n",
      "For model parameters [ 0.48065949  2.39777458 -2.26806392  0.27147681  0.41054358] takes 80.18999123573303 S\n",
      "Working on 2/10 design\n",
      "For model parameters [ 0.49845704  1.16864462 -2.40716525  0.11408599  0.34816877] takes 146.7813949584961 S\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "For model parameters [ 0.4755844  -1.03144363 -2.6357408   0.47645597  0.24217227] takes 119.43931317329407 S\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n",
      "(105, 5)\n",
      "(105, 1)\n",
      "Optimizing the GP 1\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -120.71517460284024\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.125679203228355  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  2.734166660939177e-05  |      +ve      |        \n",
      "Optimization restart 1/30, f = -140.94321779910595\n",
      "Optimization restart 2/30, f = -140.9432182149425\n",
      "Optimization restart 3/30, f = -140.94321826120583\n",
      "Optimization restart 4/30, f = -140.94321824761386\n",
      "Optimization restart 5/30, f = -140.94321807850253\n",
      "Optimization restart 6/30, f = -136.98740326536358\n",
      "Optimization restart 7/30, f = -140.94321569290443\n",
      "Optimization restart 8/30, f = -140.9432180915736\n",
      "Optimization restart 9/30, f = -140.94321673364468\n",
      "Optimization restart 10/30, f = -140.94321826211916\n",
      "Optimization restart 11/30, f = -140.94321825848817\n",
      "Optimization restart 12/30, f = -140.94321823319893\n",
      "Optimization restart 13/30, f = -140.94321825853302\n",
      "Optimization restart 14/30, f = -140.94321826021215\n",
      "Optimization restart 15/30, f = -140.94321823175017\n",
      "Optimization restart 16/30, f = -140.94321644959246\n",
      "Optimization restart 17/30, f = -140.9432182484077\n",
      "Optimization restart 18/30, f = -140.94321805400733\n",
      "Optimization restart 19/30, f = -140.9432182033578\n",
      "Optimization restart 20/30, f = -140.94321824694066\n",
      "Optimization restart 21/30, f = -140.94321826038967\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization restart 22/30, f = -140.9432181785667\n",
      "Optimization restart 23/30, f = -140.9432182593971\n",
      "Optimization restart 24/30, f = -140.94321825951627\n",
      "Optimization restart 25/30, f = -140.94321795193073\n",
      "Optimization restart 26/30, f = -140.94321825907707\n",
      "Optimization restart 27/30, f = -140.9432182001454\n",
      "Optimization restart 28/30, f = -140.94321826160194\n",
      "Optimization restart 29/30, f = -140.94321819731172\n",
      "Optimization restart 30/30, f = -140.94321823171995\n",
      "Time it takes to do optimization 3.08777117729187\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -140.94321826211916\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |    0.11897004344595252  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.0006400439478961759  |      +ve      |        \n",
      "(105, 5)\n",
      "(105, 1)\n",
      "Optimizing the GP 2\n",
      "\n",
      " before optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -134.640009134306\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                   value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |      0.2073991369886202  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                    (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.00043729452925814693  |      +ve      |        \n",
      "Optimization restart 1/30, f = -143.48420020971304\n",
      "Optimization restart 2/30, f = -144.9412657897082\n",
      "Optimization restart 3/30, f = -144.94126578874398\n",
      "Optimization restart 4/30, f = -144.94126561437668\n",
      "Optimization restart 5/30, f = -144.94126576044843\n",
      "Optimization restart 6/30, f = -144.94126578987243\n",
      "Optimization restart 7/30, f = -144.9412657817773\n",
      "Optimization restart 8/30, f = -144.9412657892463\n",
      "Optimization restart 9/30, f = -144.94126569389425\n",
      "Optimization restart 10/30, f = -144.9412657701516\n",
      "Optimization restart 11/30, f = -144.9412657648514\n",
      "Optimization restart 12/30, f = -144.94126569905507\n",
      "Optimization restart 13/30, f = -144.94126577733664\n",
      "Optimization restart 14/30, f = -144.941265771897\n",
      "Optimization restart 15/30, f = -144.94126579063595\n",
      "Optimization restart 16/30, f = -144.94126579159703\n",
      "Optimization restart 17/30, f = -144.94126575612955\n",
      "Optimization restart 18/30, f = -144.9412657895862\n",
      "Optimization restart 19/30, f = -140.51644479904758\n",
      "Optimization restart 20/30, f = -144.9412657861634\n",
      "Optimization restart 21/30, f = -144.94126573096676\n",
      "Optimization restart 22/30, f = -144.94126579150753\n",
      "Optimization restart 23/30, f = -144.94126572750648\n",
      "Optimization restart 24/30, f = -144.94126578390512\n",
      "Optimization restart 25/30, f = -144.94126578617445\n",
      "Optimization restart 26/30, f = -144.94126577772857\n",
      "Optimization restart 27/30, f = -144.94126579158802\n",
      "Optimization restart 28/30, f = -144.94126579162503\n",
      "Optimization restart 29/30, f = -144.94126577836755\n",
      "Optimization restart 30/30, f = -144.94126574600332\n",
      "Time it takes to do optimization 4.692286252975464\n",
      "\n",
      " after optimization\n",
      "\n",
      "Name : GP regression\n",
      "Objective : -144.94126579162503\n",
      "Number of Parameters : 7\n",
      "Number of Optimization Parameters : 7\n",
      "Updates : True\n",
      "Parameters:\n",
      "  \u001b[1mGP_regression.         \u001b[0;0m  |                  value  |  constraints  |  priors\n",
      "  \u001b[1mrbf.variance           \u001b[0;0m  |     0.1666934296079104  |      +ve      |        \n",
      "  \u001b[1mrbf.lengthscale        \u001b[0;0m  |                   (5,)  |      +ve      |        \n",
      "  \u001b[1mGaussian_noise.variance\u001b[0;0m  |  0.0008120758036561749  |      +ve      |        \n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "for i in range(0,10):\n",
    "#model_variance = ModelVariance(model = model_emukit)\n",
    "    gp = i%3\n",
    "    print (f'Trainng {gp} gp for R_AA at {pT_obs[gp]} GeV')\n",
    "    integrated_var = IntegratedVarianceReduction(model = gp_array[gp], space=parameter_space)\n",
    "    expdesign_loop = ExperimentalDesignLoop(model = gp_array[gp],\n",
    "                                         space = parameter_space,\n",
    "                                         acquisition = integrated_var,\n",
    "                                         batch_size = 10)\n",
    "    ed=expdesign_loop.run_loop(fun[gp],1)\n",
    "    X_latest=gp_array[gp].X[-10:,:]\n",
    "    Y_latest= f(X_latest, pT=pT_obs)         \n",
    "    for k in range(0,3):\n",
    "        if k!=gp:\n",
    "            X = np.append(gp_array[k].X, X_latest, axis=0)\n",
    "            \n",
    "            Y = np.append(gp_array[k].Y, Y_latest[:,k].reshape(-1,1), axis=0)\n",
    "            print(X.shape)\n",
    "            print(Y.shape)\n",
    "            gp_array[k].set_data(X, Y)\n",
    "            print(f'Optimizing the GP {k}')\n",
    "            print('\\n before optimization')\n",
    "            print(gp_array[k].model)\n",
    "            st = time.time()\n",
    "            gp_array[k].optimize()\n",
    "            et = time.time()\n",
    "            print(f'Time it takes to do optimization {et-st}')\n",
    "            print('\\n after optimization')\n",
    "            print(gp_array[k].model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def emu_predict0(X):\n",
    "    y, variance_dy = gp_array[0].predict(X)\n",
    "    dy = np.sqrt(variance_dy)\n",
    "    #y = scaler_0.inverse_transform(scaled_y)\n",
    "    #dy = scaled_dy* scaler_0.scale_\n",
    "    return y, dy.flatten()\n",
    "\n",
    "def emu_predict1(X):\n",
    "    y, variance_dy = gp_array[1].predict(X)\n",
    "    dy = np.sqrt(variance_dy)\n",
    "    #y = scaler_0.inverse_transform(scaled_y)\n",
    "    #dy = scaled_dy* scaler_0.scale_\n",
    "    return y, dy.flatten()\n",
    "\n",
    "def emu_predict2(X):\n",
    "    y, variance_dy = gp_array[2].predict(X)\n",
    "    dy = np.sqrt(variance_dy)\n",
    "    #y = scaler_0.inverse_transform(scaled_y)\n",
    "    #dy = scaled_dy* scaler_0.scale_\n",
    "    return y, dy.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 1/10 design\n",
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "For model parameters [ 0.41363636 -1.          1.          0.3         0.3       ] takes 65.27234387397766 S\n",
      "Working on 10/10 design\n",
      "For model parameters [ 0.45909091 -1.          1.          0.3         0.3       ] takes 82.3524899482727 S\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/ipykernel_launcher.py:32: MatplotlibDeprecationWarning:Passing the pad parameter of tight_layout() positionally is deprecated since Matplotlib 3.3; the parameter will become keyword-only two minor releases later.\n"
     ]
    },
    {
     "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": [
    "#make a regular grid to plot our Emulator predictions\n",
    "n_plot_pts = 100\n",
    "sim_pnt = 10\n",
    "test_points = np.array([[x,true_values['exponent_inel'],\n",
    "                 true_values['exponent_el'],true_values['scale_inel'],true_values['scale_el']] for x in np.linspace(prior_ranges['alpha_s'][0],prior_ranges['alpha_s'][1],n_plot_pts)])\n",
    "simulation_out = simulation(test_points[np.arange(0,n_plot_pts,sim_pnt),:],*pT_range)[:,pT_obs]\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),0], simulation_out[:,0],'ro', label='sim pT_obs[0]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),0], simulation_out[:,1],'go', label='sim pT_obs[1]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),0], simulation_out[:,2], 'bo', label='sim pT_obs[2]')\n",
    "\n",
    "\n",
    "#get the GP Emulator's predictions of both the mean and std. deviation\n",
    "gp_y0, gp_dy0 = emu_predict0(test_points)\n",
    "gp_y1, gp_dy1 = emu_predict1(test_points)\n",
    "gp_y2, gp_dy2 = emu_predict2(test_points)\n",
    "\n",
    "\n",
    "plt.plot(test_points[:,0], gp_y0[:,0], color='red', label='pT_obs[0]')\n",
    "plt.plot(test_points[:,0], gp_y1[:,0], color='green', label='pT_obs[1]')\n",
    "plt.plot(test_points[:,0], gp_y2[:,0], color='blue', label='pT_obs[2]')\n",
    "plt.fill_between(test_points[:,0], y1 = gp_y0[:,0] - 2.*gp_dy0, y2 = gp_y0[:,0] + 2.*gp_dy0, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='red')\n",
    "plt.fill_between(test_points[:,0], y1 = gp_y1[:,0] - 2.*gp_dy1, y2 = gp_y1[:,0] + 2.*gp_dy1, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='green')\n",
    "plt.fill_between(test_points[:,0], y1 = gp_y2[:,0] - 2.*gp_dy2, y2 = gp_y2[:,0] + 2.*gp_dy2, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='blue')\n",
    "\n",
    "plt.xlabel(r'$\\alpha_s$')\n",
    "plt.ylabel(r'$R_{AA}$')\n",
    "plt.title(r'GP Emulator and Model Training Points')\n",
    "plt.legend()\n",
    "plt.tight_layout(True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 1/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: step size becomes too small\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: larger nsteps is needed\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/ipykernel_launcher.py:32: MatplotlibDeprecationWarning:Passing the pad parameter of tight_layout() positionally is deprecated since Matplotlib 3.3; the parameter will become keyword-only two minor releases later.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaoAAAEaCAYAAABARRODAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAACPBUlEQVR4nOy9d5gc1Zm+fZ/OaXpyHuUAykISEgIBElFkkTFgY3/2ssZhbZzwGtuL8TqwtvfntcGwXhtjMMlEAyKDMEFCOec4OYfumc5ddb4/TvdoNJrQk0dS3dfVl2a6q6tO17Tqqfec931eIaXEwMDAwMBgtGIa6QEYGBgYGBj0hCFUBgYGBgajGkOoDAwMDAxGNYZQGRgYGBiMagyhMjAwMDAY1RhCZWBgYGAwqjGEymDUIYT4QAjxpZEex2AghHhMCPGfo/WYQogjQoiLhng8PxBC/Gmwtx0uhBBtQoiJIz2OUxlDqE5ihBC3CCHWCiECQoi6xM9fEUKIxOuPCSGiif+ITUKId4QQp3ezr47bJh9bh/cTdTmuIb/QDhVCiM8LIaQQ4r87Pb8i8fxjIzQ0hBBvdPg7xzr97R/py76klD+XUqZ049GXbftK4pwGEp+hUgjx30IIcwpj8kgpD/XhGJMHPlqDjhhCdZIihPg28D/Ar4ACIB/4MnAOYOuw6X9JKT1ACVAHPNbDbv8r8Z82+ZgzJIMfJoQQlpEeA3AQuLnTWD4H7Buh8QAgpbws+XcGnuTYv/2Xk9uNknPYF+YkPtOFwK3Av4zweAxSwBCqkxAhRDpwP/AVKeXzUspWqdgspbxNShnp/B4pZRB4CpjZj+ONT9xJfkEIUS6EaBZCfFkIcaYQYpsQokUI8WCH7e8TQvyti/cfd9ETQkwSQrwvhGgUQjQIIZ4UQmQkXnsCGAu8mrhL/l7i+auFEDsTx/1ACDGtw/6OCCHuEUJsAwLdHPN/Ep/DL4TYKIQ4t9PY/y6EeFwI0Zo4zoIOr58hhNiUeO1ZwNHL6asBtgOXJt6fBZwNvNJpTD19ph6PKYS4UgixJfHe1UKI2b2MqUcSf6uvCiH2A/sTz/V2zv6W+Dn5t75DCFGW+Jve289tnUKIvya+b7uFEN8TQlSk8hmklHuAj0h834UQ/yKEOCDUzMIrQoiiTp93cuLnx4QQDwkhVibO91ohxKTEax8m3rI18X28WQiRI4R4LXHum4QQHwkhjOtuHzFO2MnJYsAO/CPVNwghPMBtwOYBHHcRMAW4GfgtcC9wETADuEkIcX4/9imAXwBFwDRgDHAfgJTys0AZcFXiTv+/hBBTgaeBbwK5wOsoIesYRX4GuALIkFLGuzjmemAukIUS7+eEEB0v/lcDzwAZKEF5ECBxjJeBJxLvfQ64PoXP+DgqigK4BfV3a7+Z6Okz9XZMIcQ84FHgX4Fs4H+BV4QQ9hTG1RMrUH/v6YnfeztnnVkCnIaKbH7cUXj7sO1/AOOBicDFwO2pDl4IMR04F9gshLgA9R27CSgESlF/3+74DPATIBM4APwMQEp5XuL1OYnv47PAt4EK1N8tH/gBYPjW9RFDqE5OcoCGjhfhxJ10ixAiJIQ4r8O23xFCtKD+w3mAz/ew3+8k9pF8/LXT6z+VUoallG8DAeBpKWWdlLISdfd6Rl8/iJTygJTyHSllREpZD/w30JPg3QysTLwnBvwacKKilCS/k1KWSylD3Rzzb1LKRillXEr5G5Ton9Zhk4+llK9LKTWUQCSnQM8CrMBvpZQxKeXzqAt4b7wELBUqEv4cSrhS/Uy9HfNfgP+VUq6VUmpSyr+iRPCsFMbVE7+QUjYlz2EK56wzP5FShqSUW4GtHD2Hfdn2JuDnUspmKWUF8LsUxr1JCNEMvAr8CfgL6gbtUSnlpsRsw78Di4UQ47vZx4tSynWJ/19PogS6O2Io8RuX+Pt8JA2D1T5jCNXJSSOQ03FaS0p5tpQyI/Fax7/7r6WUGVLKAinl1VLKgz3sN7lt8nFHp9drO/wc6uJ3T18/iBAiTwjxjFCL337gbygh7o4i1B0xAFJKHSgHijtsU97LMb+dmEryJUQ8vdMxazr8HAQciXNdBFR2uhCV0guJi/1K4IdAjpTykz58pt6OOQ74dscbDFRUWsTAOOYcpnDOOtP5HPb03ehu26JO4+jx75pgnpQyU0o5SUr5w8S57Hx+21D/T4q72Udfxv4r1E3g20KIQ0KI76cwRoNOGEJ1crIGddd8zUgPpBsCgKvD7wU9bPsL1FTJbCmlFzW9Izq83vnutAp1cQZACCFQF+bKHt7TTmJt5R7U3XpmQtx9nY7ZHdVAceKYScam8D5QUdS3URFaZ3r6TL0dsxz4WacbDJeU8ukUx9Ud7edwgOdsIFSjkoCSjOnnfjqfXzdqmrSy23ekSGJ9+NtSyonAVcC3hBAXDnS/pxqGUJ2ESClbUHPofxBC3CCE8AghTEKIuYB7RAen2AKcJ4QYm5ju+vcetk0D2oAWIUQx8N1Or9ei1iiS/B24QghxoRDCirr4R4DVKY4tDYgD9YBFCPFjwJvie9ck3vtvQgiLEOI6YGGK7/0nap3l91281tNn6u2Y/wd8WQixSCjcQogrhBBpKY4rFQZyzgbC34F/F0JkJr4bX+vnfp4CviCEmJtYu/s5sFZKeaQf+zrm+yhUIsvkxI2EH9ASD4M+YAjVSYqU8r+AbwHfQ6Wd16IW0u8h9Yt2Z74njq2jaujn2N4BngW2ARuB13rY/CfAPNQd+krgxU6v/wL4YWJa6ztSyr2oqOv3QAPqLvYqKWU0xeG9BbyBSg8vBcKkNqVE4hjXodb5mlFrS53H2917pZTyPSllUxevdfuZejumlHIDap3qwcTrB+h5HbI/9PucDZD7UYkKh4F3gefpkISSKlLK94AfAS+gorRJqKSW/nAf8NfE9/EmVHLRu6ibrTXAH6SUH/Rz36cswljXMzAwOBkQQtwF3CKl7E92qcEoxoioDAwMTkiEEIVCiHMS09qnoaZEXxrpcRkMPidaVbmBgYFBEhtqOnsC0IKqffrDSA7IYGgwpv4MDAwMDEY1xtSfgYGBgcGoxhAqAwMDA4NRzQm/RpWTkyPHjx8/0sMwMDAwMBgAGzdubJBS5nb12gkvVOPHj2fDhg0jPQwDAwMDgwEghOjWbsyY+jMwMDAwGNUYQmVgYGBgMKoxhMrAwMDAYFRjCJWBgYGBwajGECoDAwMDg1GNIVQGBgYGBqMaQ6gMDAwMDEY1hlAZGBgYGPQfKSEYBF0fskOc8AW/BgYGBgbDSCgELS3Q2AiVlVBdDeEwrFgBBQVDckhDqAwMDAwMuiYaVaLU3KxEqbIS2tpACPVwucDrhUhERVZDhCFUBgYGBgagaeDzKWGqroaKCvVzUoCcTnC7ISNj2IdmCJWBgYHBqUggoCKlujooL4faWrXOJCU4HEqUCgtV5DTCGEJlYGBgcLITjytRampSkVJFhVprAjCbweOBvDz18yjEECoDAwODk41ktFRbC6WlKmpK4nQqYcrKGrnx9RFDqAwMDAxOZHRdrS0lo6XycpXwAEejpcJCMJ241UiGUBkYGBicSCSn8erroaxMZeLFYmptKRktpaeP9CgHFUOoDAwMDEYzsZiqWaqtVcJUVXU0E8/tVlN4lpP7Un5yfzoDAwODE41oVAlTTQ0cOaLWl6RUU3cejyqqPYGn8fqDIVQGBgYGI0kyYqquVsJUW6ueN5kgLe2UFKbOGEJlYGBgMJxomkp8qKmBQ4fUv7quEh/S0qCoaFTULo0mDKEyMDAwGEqkVA4PdXVw+LDKyovHj2bkGRFTrxhCZWBgYDDYBIMqK+/IESVO4bB63u2G3NxRW1g7WhlWoRJCLAf+BzADf5JS/rKLbZYCvwWsQIOU8vxhHKKBgYFB34nHj7qJHzigfhZCWRF5vZCdPdIjPKEZNqESQpiBh4CLgQpgvRDiFSnlrg7bZAB/AJZLKcuEEHnDNT4DAwODPtHaqhIfDh5U7g+6rqbw0tOhpGSkR3dSMZwR1ULggJTyEIAQ4hngGmBXh21uBV6UUpYBSCnrjtuLgYGBwUigadDQoNwf9u1TbhCgWl2MYp+8k4HhFKpioLzD7xXAok7bTAWsQogPgDTgf6SUj3fekRDiTuBOgLFjxw7JYA0MDAwIhY5GTYcOKbEym1XUVFw80qMbUeJSozneRm28mcrwfhZFzieDwiE51nAKVVf5lp07bVmA+cCFgBNYI4T4VEq575g3SflH4I8ACxYsGLpuXQYGBqcWyQy9qirYu1eJlBDKmign56R3gOiJuNRoirdSF2/hcKSOqlgTEokAotF6ZmthMobo2MN51iuAMR1+LwGqutimQUoZAAJCiA+BOcA+DAwMDIaC5JReWZkSp6Shq9eroqZTtKZJkxpN8Tbq4z6OtFbQWn0ET4MPb0MbE5tCnNEUwNXUirPRj72llbpLvjlkYxlOoVoPTBFCTAAqgVtQa1Id+QfwoBDCAthQU4P/bxjHaGBgcCoQi6lo6fBhtd4Ui6loKT39pDN0TRW9rRV/+QEClUcIVB6G6mrS6nyMq/cz3Rc8dluziVC2l1B2Og3TxlHv1CnS9SEb27AJlZQyLoT4GvAWKj39USnlTiHElxOvPyKl3C2EeBPYBuioFPYdwzVGAwODk5hIRLlA7N+v1pt0HWw2yMwEq3WkRzc86LoqPK6oQFZUEC07jFZRjrmqGrs/QAa0T98Fs7wE8zNpnDOF8rxMgjkZBHPTCeZkEEn3gOlopFlVf5CrvWlDNuxhnXCVUr4OvN7puUc6/f4r4FfDOS4DA4OTlFBIrTft26em9nT91MjSk1LZNJWWHn2UlSEryhHhCKCSBvQ0B76CDNrmjCdSlEeoIIe2giyCuRnottEj3qfuyqCBgcHJSSikCm/37FGp5KAcIU5WqyJNU7ZMhw6pqczko7W1fZNoRhqNxZnUL5lKU1EWweJc9JJiZNrQRUGDiSFUBgYGJz5Jcdq9+6g4paWd8J1tjyMeV5HhgQPqkSw2jkYBkDYb8THF+BfMoLzYQ1mhi8aSbHSPG6/ZhcNkG+EP0D8MoTIwMDgxCYePRk7l5Wq6K+k+fjKIk5Sq9ce+fWpdLbm2lhAl3G7khAlEl19M87g8DhQ72ZMliZvBhCDN5MJtspNzEmQtGkJlYGBw4hCNqov3nj3K8FXXlQP5yRA5hUJKjPbsUY+9e49O39ntMGkSXHYZ8ckTaRifx5FMwYFYLW16GCklbpONHLMLsxj+86DrQyuGhlAZGBiMbjRNZeslIwtNOznWnBob1VTl7t2wa5daV0qmeJeUwMKFcNppyKlTaSnOpkb3cyBSTVWsEUkZloiZDLObdKt72IceiVpoa3PS4ndT15BBeU0JS86zUDhpaI5nCJWBgcHoQ0rVJuPAARVdRCLKifxEzdaTUtVt7dwJO3aof2tq1Gt2O0ydCjfeCKefDqedRtRlpz7upzRax4HIEQL+PQgEaSYnBZYsTJ2m84pW72Dac6twNvoIZaez+8ZlVJ09c9CGHgrbaW1z0tScRl1jBoGgA4FEmCRORxQtbkHTBuVwXWIIlYGBweihpUVFFjt3qmkvi0W1yDgR65xqa2H7dvXYsUMJL6h1tBkz4PLL1b8TJoDFgl8LUhNtZn90NxWNDUjAipl0s5tMq6fbwxSt3sGcR1diicYAcDX6mPPoSoB+iZWmCwIBB61tThqa0qlvzCAWM4MAi0XD6YiQk+U/9k1iaJ3sDKEyMDAYWUIhlcm2Y4e6uJvNkJV14jlE+HywbRts3ar+TUZMXi/MnAnXXaf+HTMGTCY0qdEQ91MROcxefxV+LQgSPGZHl1FTd0x7blW7SCWxRGNMe25VSkIVj5toDTjx+d3UN2TQ1JKGLgUCsNliuF1hzOahc51IBUOoDAwMhp94XF3Id+1SmWxSKmEaM6b3944WolEV+W3Zoh6HD6vn3W4lSFdfDbNmwdix7X6BET1GbbyRQ5EaDoariaFhFibSTW6Krf1rruhs9PXp+WjUQmvASYtPrS+1+D0gQZgkDnuMdG8Ak2l0eX0bQmVgYDA8JN0SDhxQF/hIRLmSnygZe1KqGq1Nm9Rj504lVhYLTJsGt98Oc+eq7LwO62htWojqWDP7IpVURBuR6DiEjSxLGhYx8PW2UHY6ri5EKZStItJI1EJrm4vmFg819Zm0trkQQiKExOWIkJXROup9dw2hMjAwGFpCIVWUunWrynSzWk+cdadQSE3jbdigxCm5zlRSApdeqkTqo4/UtGVNDeTlIadMoSXeRmWsgd3hChrifpCQZnZSYMlMeUovVXbfuOyYNSqAmNXK++ddx0efzqA14EIgMZl1XI4o2Zn+US9MnTGEysDAYPDR9aP1Tvv3q2gkI+PEmNqrrlbCtH69EqB4XEV+c+aozLx581T24QcfwEMPqcgQoL4e7cHfs6ZtN9sXjUMgyDC7KbHmDOlwDy+YS2ubgzNefRuvr5kmdw7/mHc7mzPPxSm6SHw4ATGEysDAYPDw+5Wtz7ZtEAyeGFN7mqaKa9euVQJVnmhEXlICV14J8+fD9OnHRYDyiScQSZFKYI7GOOOFT2heMn/IhhuLmfG3uWhqSaOmTk3lYZ7LC9ffhNMZwW6LA5BBYMjGMNwYQmVgYDAw4nFlZbRtm1rDMZnU1F5W1kiPrHsiEdi8WYnT+vVKYC0WlS6+fDksWKAEthOa1KiL+zgcqWVxchqwE67GwY1gNE1l5TW3eKipy6TFlwZCYhISlzMyIlN5EklQtNJgqqLRVMWhgkp0uXDIjmcIlYGBQf9oblaJEdu3qwu/xzO6O+L6/UqU1qxRWXrRqMrQW7AAFi2CM85Qv3ciKU6HIjXsCVUQQ8OKmTnZXtxdiFIyiaG/SAmBoINmn4fa+kwam7zoukAIidMZJWuYhSlEgDpzOfXmcupNldSbK6g3VRIytbVvY3bYaIg0DNkYDKEyMDBInVhMTY1t3676PI32gtzGRvj0UyVOO3aotbOcHLj4YjjrLBVBWY6/DGpSoz7u52Ckmr3hSiJ6DLvJSqYlDWsiU2/PjRccl8QQt1nZfeOyPg8zErHia3VR15BBTX0msagFIcBujw5burhEp8lUS425lFpTGbXmUurM5bSamtu3sUsXuVoxp8cXkKMVkaMXk60V0VDqIm953pCNzRAqAwOD3mluVkkR27erSMTrHb3RU309rF6tHnv2qBCluBiuv16J0+TJXY5blzr1iWm93eEKwjKKDYsSJ8vxaeTJYtr+WBdpusDf6qKpOY3quixaW92AxGrVcDnDWDyhAZ+GntDRaTRVUWU+TLX5sBIncykxodbcTNJMjl7EuPh08rQS8vWx5GolpGkZCCRIQJeABClplEPon4QhVAYGBt2RXHvaskX9a7GoaGSko6cPPoAnnoCGBjWez35WRUarV8PHH6vECIDx4+Ezn4Gzz1ZFt10gpaRRa+VwpJZd4TKCegQLZrItaeQIb69DqTp7Zso2RaGwDZ/fTXVtJnUNmei6wGRS60xDOZ0nkfhEA1WWQ1SZD1JpPkSN+TAxodqFWHU7+fES5oYXkx8dQ0GshJxYARY6/Z2lBFNQ1YiZzGA2qfVIYQKPExzOofkAGEJlYGDQmdZWFT1t3arqiLxelQE3GqKnLlLC+X//T11EQfnm3X47LFmi+lJ1g08LcCRSx45wKT4tgEVYyDJ5yLIOXsfbjlFTVU02rQEXAI4hns6LEaHKdIgK0z4qLQeotBwiYFZraWZpoTA6hrmhsynUJlDEFLIs4zDZbOC2Q4YVLGYlQJ0FyWRCNbDvgvKQWqMcIgyhMjAwOFr3tG2bKs4VQkUrOUNbA9Rn/vrXoyKVREpwueA3v1FTfN0Q1COURerYGS6nNtaMSZjIMqcxxpo7aMOLRC20JJIgauoy0TSTippc4SGoZ5Kg6bRpjVSY9lJmO0CF/SA11nJ0obz5svQCJjKPEtMMiqwzyLNOwWx1qOh4BPpW9RdDqAwMTmVCIeW1t3mzyopzu0df3VM4rNLIP/xQJUd0RSjUpUhF9RhVsSZ2hcspjdQBkGF2M8Y2OOKUzNBrak6jqjabZl8aSInNFifNE8RsHqSoSdNUIks8hs/URJl9P6W2A5S5D9BkqQXAgo0i0+mcZVvCGPtciq2zcZkyBuf4I8ywCpUQYjnwP4AZ+JOU8pedXl8K/ANIuDvyopTy/uEco4HBKUFDg0o02LlTXW2zskaXa4SmqbWxDz5QIhUOq+jO6VSi1JkOkZ8uderiPvaFK9kbriQuNdwmB0XW7EGxL9J1ga/VRUNTOpXV2YRCDkjUNGVlDHStSUJMg3hMrRECflMzpe5DHEk7SKllNy1CCZNdeBhrnctc642Mtc2l0HI6ZjFKsy8HyLAJlRDCDDwEXAxUAOuFEK9IKXd12vQjKeWVwzUuA4NThmRyxKZNaprPbof8/NHTiFBKFd2tWqWip5YWFeGdf756TJ+unu+4RgXqc3z2s7TE2zgUrWV76AhBPYpdWMixeAfF+DUeN9Hi91Bbn0FVTTbxuBmzScflDpM9kCk9TVNZlAlRAgg54xzJOsIR+14Os5UmWQGAQ3gZZz2DhbZbGWudR55lEqZB+GwnAsMZUS0EDkgpDwEIIZ4BrgE6C5WBgcFgEgio5IgtW5St0Whrp9HQAP/8J7z/vqrRsljgzDNh6VJVjNsxy3DpUvVvIutPz8mm7parWD3HRm3Th5gT606DkRQRjVpo9nmoqsmiriETTTdhtcbxuPvZn0mXavouFlHp3VKC3YaW7qbCU8Eh8zYO65uoiu8BJFacjLOewTzbdYy3LiDPMrlbYdoRepNVgYfx6bWkm/JZ5r6Lmc7lA/r8o4nhFKpioLzD7xXAoi62WyyE2ApUAd+RUu7svIEQ4k7gToCx3aSdGhic8tTXq6m9PXtUckR2tnqMBiIRVYT7/vsqu1BK1Yb9K19RGXs9ZJDp559H7Tmz2R0qZ3+kGh1Juh6jxDbwxI9IxEpjs4fK6hwamtORUvQ/S0/XVbQUSxQEC8CTBrm5NDtbOGTaysH4Oo7ENhLVggjNTLF1Bue6vsgE25kUW2diFr1foneE3mRl6y+IEQbAp9ewsvUXAEMqVuGQmaZaB031Dg7sN6PfMWSHGlah6mrmtvNffhMwTkrZJoS4HHgZmHLcm6T8I/BHgAULFoyuDl8GBiOJpim/vQ0bVLdchwMKCkbH9J6USjTfe0+1xgiFlAv5zTerSKmHdHIAvxbkUKSGraHDBPQITmEj35KBeYDZa+GIlcamNCqqc2lqTkMicPanT5OuQaTDNJ7ZrFL709OJO8yUmfdwMPo6B6KraQyXApBuKmSm/VIm2s9ivHUBDlPfU7xXBR5uF6kkMcKsCjw8YKGSElpbrNRXu2iodtBQ66SxxkFjnYOA39ZxSxrqGuC0AR2uW4ZTqCqAjvMNJaioqR0ppb/Dz68LIf4ghMiRUg6diZSBwclAKKSm9zZvVlN9QzG911WhbXIqricaG9W603vvqTUyh0MV4V54oSrU7SHDMCbjVEWb2BY+QkW0AYEgx+wd8NReKGxrF6fmFiUOTmeErMw+iJMuIRo5GjGZzeq8p6eDy0mrpZUD0TUciHzC4cB6ojKIGRvjbPOY77yOSbazyTKPQQwwwcOn1/bp+e4ItlmorXRSX+WivspJXZUSp0j4qEw43XGyC0JMmdlCVl6Y7PwwWXlhmsMt5BUMkUoxvEK1HpgihJgAVAK3ALd23EAIUQDUSimlEGIhYAK6yUc1MDCgqQl271Y+dlIqARkK1/KuCm0fekj93JVYxePKAPbdd2HjRjUNNn26sjE65xyVvdcDTfFW9oUr2RkuJ6rHSDM7KbZkD+iiHo5YaWj0HiNOLldfxElCNKbESaKmU71eyMwElwtpt1EXP8i+6EvsD3xMVVwtv3tN+cxyXMZk29mMty3AKhz9/gxdkW7Kx6fXdPl8l59Ch+YGOzUVLmrK3dRWuKircNHqOxohOd0x8opCzFrUSG5hiJzCEDkFIdxp8S736S8f2omtYRMqKWVcCPE14C1UevqjUsqdQogvJ15/BLgBuEsIEQdCwC1SSmNqz8CgI8ni3C1bVHGuzTb02XtPPHF8oW0kop7vKFQVFfDOO2rtyedTonnddXDRRb1O7UX1GGXReraGjlAbb8EmLGSZu/bZS5VI1EJjUxrlVXk0NXuRJC2LUhQnTVOfU0t42bndkJsLaR5wONkWeoP3Ar8n4G9GYEaiAYJiy3SWuu9iiv0c8syTBxw19cQy91288slB9Pd+Ar6xkF6G6cL/YNk5k5A6NNU7qC5zU13morrMTW25qz1KMpl0cgrCjD/NT15xkPySIHlFIdze2KgwIkkyrHVUUsrXgdc7PfdIh58fBB4czjEZGJwwxGJw+LCKUJqbVcLBcFkbNXQz+97QoC7kq1fDW2/Brl1qKm/hQuVQPm9ejwKa9NrbG65gV7gcTWp4TW7GDKArbixmprEljYqqXOobM0D2pT2GhEgiagKVgZiVpabz3G4wW4jJMIeia/m0+SnK41s6vFPDjIWL3d9kgfvGfo+/z2y/FfHqWIglIiLfeOTLj/LJmghvtNiIhNRl3mLVyS8OMnNhIwVjghSMCZBbGMJiHf2xgOFMYWAw2gkEYN8+tf4UiaippuFOL8/JUdN9nXE44POfV2MsLITPfU6tPWVm9ri7ZPS0JXSYungLNmElx9z/mqd43ESzT2Xr1dRlISU4HNHUCnB1XZ3X5FqTJw0K8tWNgMMBCCJ6gAPR99nTtooDkdWJ5IXjd6wRZ03ob8MiVJomqC138ebfx6PFjr2US91MY52DuYsbKBzXRtG4ADmF4cFzyhhmDKEyMBitNDWptaddu45679lsvb9vKPjsZ48vtAWVfn322XDJJTBrVq/WS41xP3vDlewMl6FJjXSTu99ee7oulBt5XRbllblougm7LUZGelvvqeSaphJQpFSed1mZkJkBbg+Y1WUxrLexP/wmu8PvczD6KRpR3KYsZjkv53T7Up5q+bcud93XJIZUiYRMVBxKo+ygh4qDaVQecROPdS/suia4/NYjQzKW4cYQKgOD0YSUR9efjhxRwjQa0ssnT1YZeps3qzGazSop4l/+RU2L9UBMximPNLA5dKh97Wkg0VNrm5Oa+kzKKnKJRGxYrBretABms8RmN3PmOWPxZjq69vnuuORtNiuRMh3dUiKJyTAxGSYuo1zIIgSLsQoHVuHAIo7eKHxOezuxJnUsAjNe88BvKKQuiMcF8bgJLSbQNBNM1wE/ZnMLZqvEYtEJBS1I/fhPK0wSb8bw3NjE51gRoXp2727qdVuHw0FJSQnWPrSLMYTKwGA0EI9DWZmqf2poGN71p57G9Omn8MYbqmGixaKKcS+9VEVPvYzNpwXYF65iW+gIUT2G1+zq99pTOGKlviGdI+X5tLa5MJl10jwh0jzH1g+deU4JEyeX4HKlqQQGKdXUHqhoz2JJCJQgOXUn0QnLNsK6n7Bsw4YVE06cJi8OkYZNOOlqmi+k5+HTq5EdykEFgnRTIU5T39vR6zpEw2YiETPRsIl4zIzNAkJIrHYdm11LPPRjTn0oYMbXZEfKo08KIUnPiuB0D21DwySRqE5BjgOXo2fxkVLS2NhIRUUFEyZMSHn/hlAZGIwk4TAcPKgEKhAYmfWnzjQ0qMSIt99WSRt5eWrq76KLel170qVOVayJbaEjlEbrMGMix+LF2kW7996Ix000taRRVplHfUMGIHG7wuRkd++t58104HJ4EB3FyWY7Gj0lkEgisrVdnCQ6Jiy4TBk4hbdbceqI05RONOgk6HODZgVzDGd6AKcntShGSohGzERCJqIRM7Go6vckhMRm13C6o9jsGlab3uM9gRKjCK0+G1pcYLZI0tKjgyxSMuH6JDv9nHw1tRsqIQTZ2dnUd7Xe2QOGUBkYjARtbar+acsWFbkMVf1TqkipelGtXAnr1qnf58+Hyy7rNXMPIKRHOBipZnPwMK1aCI/J0a+6JynB1+qmqiabiqoc4nGzSoroMWPvaKaekBLRHjmZOvVckkRliJD0EdZb0dEQmHEKLw6TF7tw0Zs4HfOZA2ZCzRmQjGQ0G6FmKzbRfSQTjZqIhsxEwmaikUQjQiGx2XXS0mPYHBpWu96HUSicbg2nu7/t648Kj0TSZUWQABMmTEIghBmTEJiECYGKTONCYDGlNpXbn1R9Q6gMDIaTpiYlCLt3q4v/SLd2DwZVzdPrr6saqLQ0uPZaNb1XUNDr2+tjPnaHy9kdrgAkWeY0MmzuPg8jFLZR25DOkbICgiEHFrPWbT+nHdZPWOV4Dp9oJF3LYpnvKmZal0HhOHUuHccW1MZlhJD0EdL9aMQQCBwiLSFOnsTFtu+0+mzHTLcBSClo9dnaRUPTBJGEMEXC5sRaksRq0/F4lTB1nsobKpIi1JUYKeExYzKZMAsTol2UBAKREJceBtkPj96+YAiVgcFQIyXU1Kjo6fBhdSEtLBzZBInychU9rVqlst+mToVvflOtQfWSWRiXGuWRejaFDlEbb8aOtV+ee8mpvdLyPBqa0hEC3O5Qj51wd4gPWen4KzFTFACfpYmV2c9A2kRmOqe1r5vpxAnpfkLSR0yqdSy7cJNmysUh0hAMvDGkFu/6wq3FBf5mK5GwhXhMHcdk1nE449idGnaHNsR9KSV6F4IkBJiFGbPJokRJJMXI1G+xHi4MoTIwGCp0XUUp69crg1i3u38JEv312OuMpqm1sNdeU47lFgucey5ccYUSql4IaGH2R6rYHDxESI+Sbnb1K7Xc3+qkqjab8spcYnEzDnusZ6cITYNQEHTJqqIX20UqSYwIqwIPc7pjGTGp06SVE5FtAFiFA68pH6fJi6mLy53zhWfx/vw+zJUVaMUl+H9wH6Hrb07pc5gtshuxEgTarNjtGi5PDLtD4/vf+wp3fvlrTD1tWkr7ToUrly8lGo3Q0tJMOBQiv7AQgD/+5UnGj5uIxWTBkhAkIUyYurmRKCnxUFHRltIxf/nL+3j88f/j9tu/yA9+cD9SSr7//W/w9tuv4/G4+OtfH2PevHmEQiEWL17Mrl27qKqqIienf0k0SQyhMjAYbJIOEuvXKxuhgRjE9tVjryva2pTn3sqVSjBzcuD229X0Xi+p5aCm93aES9kTqkAIZQqbY/H26WNEoxbqGtI5UpGPv9WNxaSRlhbqvq+TrqtEk3hcTecVFEJGBr7WrtOffXoNv224gvP1v+ORDtymbFwiHYuwdzsm5wvPkvHtr2FKdAy2VJST8e2vAfQqVrGoCatVTwhVR7GSuNNipGUca0H06//3UI/7Sw0VJelSRUovv/4uQsALzz7D9q2b+a9f/T4RJZnoy1pbX7nrrrv5+te/A8A777zBwYP7Wbt2P3v3ruWuu+5i7dq1OJ1OtmzZwvjx4wflmIZQGRgMFuGwcpDYsEEJy2C0d0/VY68rKipU9PT++2ps06fDHXfA4sW9TjtqUqM82sDG4EFq4804hI1Ca1afpvekhBafh4rqHCqrc9B1gdsd7n5qT0r12aJRla2XnQ1Z2eBxQWKqLj3QtQErwCTb2bhNmeRbJpPKhdr78/vaRSqJKRTC+/PjoyopUetMITPhkBldMwESi0Wi6cro1WyWmG3NfO1rt1FdVYmuaXzjW/dw9YobuOHa5fzoP37OnLnzmDohnzv+vzv5+MNVpKdncM8P7uNnP/0hVRUV3PfTB7hk+RXJo6JLnU8++ZD/918/JyMzi8OHDnDW4iX86lcPYbFYMAsTdosds8mCxdTzWufzzz/Nf//3z5FScsklV/CTnzzQ/toPf/htPvpoFRkZmfz5z8+Qk5PL//7v7/jLXx7BbLZw2mnTefTRZ47b5+uv/4NbbvkcQggWLTqLlpYWqqurKUxEd4OFIVQGBgMlEFDuEVu2qGmqnBzVHn0w6Mljryt0XRXlvvqqajlvsag27ldeCZMm9Xq4kB5hf7iazaGDBPRwv5wjIhErNfUZHC4rIBh0YLXFe3aLiMWUkEoJGRkwbqyyMeqURabJOKfZlrI+/Hdkh9V7MxaWe77LGa4ViYLc1KIJc2VFj8/rOglhshAJmZFSpY7bnRoOZxS78/i1ppWvvUl+fiGPP/kCAH6/77j9B4MBFp99Lvf+6Kd88fO38Ktf3s/Tf3+Vfft2c/fX/5ULLrkUSK4pWbCbbWzdsok1a3YwduwEbrhhOW++/irXXHNDSp8ToLq6ivvuu4cPPthIRkYm1113CStXvswVV6wgEAgwe/Y8/vM/f8N//df9PPDAT/jVrx7kt7/9JVu2HMZut+PztXSz30qKi4/ejJWUlFBZWWkIlYHBqKGlRVkc7dihIoChyODrzmOv85x/JKISI155RUVSmZlw662wfLm6+PdCU7yV3eFydoTKkMg+93zSdUGL3015ZR5VtdmAxOMOdV/zpGsQDKt/nQ4YMxYy0sF6fCJHQ/wIW8KvsD30OgHZjDWeSWzTZ2Dt1xBxD3OvWMcZS/oeuWrFJVgqyo97Pl44hqY6O5GwGRCYzDpOdxyHSxXc9rTEOG3aDP7zvnv52U9/xEUXL2fRWecct43NZmPZBRcDktNPn47VZkOYBVNPn0ZFeRlOqzOR9KCm8KxmG/PmLWTChMkAXH/9Z/j004/7JFSbN69nyZKl5OSom44bb7yN1as/5IorVmAymbjuOhVB3nTT7Xz2s9cBMGPGbO688zYuv3wFV1yxosv9dpXKPhRO8YZQGRj0lYYGFT3t359YPxlCi6OuPPbsdvU8qKaEr72minNbW1XUdPfdKnuvF9HUpU51rJktoUOURuqxmSx9zt4LJ6On0kJCYTs2W6wHI1gJ4cTUntkMeblqetR5fHFtTIbZFX6PLeF/UB7bigkzU2xLyCy/lQ2P3gERe3KPbHu+iDG2w8xc2Lt9T0f8P7jvmDUqAM3hovzLP0eLm/B4YzhcquA2VSZOmsLr73zE+++9zS9/dh/nLb2Au7/97+2fX5c6FosVTWoIwGw243a68dg8mE0mNC2OrQv7pc4X/77Xp6VuRpvc97PPrmT16g95441X+PWvf8qaNTuP27aoqITKynIWLFC/V1RUUNRLO5f+YAiVgUEqJFPMN21SPaAcDtVfaWjzjI+uQ3XO+hszBn7zG/j4YzVHtWgRXHONWofq5SIWk3EOR2rZGDxIc7wVj8lJiTX14tzjoicpSfOEcHe39hSPqxR4gHQvjBuXmNo7/tzVxvaxKfwyO8JvEZFtZJnHcKH7a8xyXo7HlM3vfz2HeOTYadVY1MyqV8b0Sai0uKD+4ttp/ZGVgt/9CFttGbGCMTR8+37Mt1xLrrV/xbM1NdVkZGRy/Q234Ha5+fuzf0OTGlJKNF1X51iAx+bGZDJjNdswmyyYeymW3bRpHaWlhxkzZhwvvfQsd9xxZ5/GNX/+Ir7//W/Q2NhARkYmL7zwNHfe+XUAdF3nH/94nuuvv4Xnn3+Ks85agq7rVFaWc+65yzjrrCU8//xTBALHZwZedtnV/N//PcjVV9/C2rVrSU9PH/RpPzCEysCgZ3RdtU9ft05lzI2EB9/Speqh6yqT8B//UNONTidcfjlcdVVKxbkBLcy+SCWbg4eIyBiZZg9jbKmvP0WiFmrqMtvXnmz2HqInXUI4pETKblPnLDMDrMev3UVliJ3hd9gcepmq+E7M2Jhmv4AznNcw1nrGMQLqa+q6xqu75zuixQWhoIVw0EwsqoQhfMWthG+8CYdLw2JRUYeF1KOPzuzZvYP//MkPESaB1WLlZw/8N1azFbPJjNvmwmPzqEjK1LdL75lnLuYnP/k+u3ZtZ/Hi87jyymv79P6CgkJ+/ONfcNVVy5BScvHFl3P55dcA4Ha72bNnJ0uXzsfrTefRR59F0zTuvPN2/H4fUkruuutu0tMzjtvvJZdczjvvvM7ChZPxeFw89thf+jSuVBEnegPdBQsWyA0bNoz0MAxONjRNuZevX6/cJNLTVdvxkSASgffeU+tPVVUqqrrqKtVaw927C0RTvJUdoVJ2hcsRqPRyW4oXyqSlUXllLqv8O/k042+0WepJl9ksC9/IzFinNZhYTEVPQqhpvZwc8LihiwLb2vh+NoVeYkf4TSIyQI55AvOc1zLLsbxbU9ff/3AOvqbjxS49K8LX/3Mrl8+3MXb8lPbnNU0QDpgJBS3t4mS1aThccZwuDbNl4Nc/mZjSS15LLSazEidhaV9nGggff/wBv//9r3n22dcGPNaB8stf3ofb7eFrX/tOuxG9lOp+JDPz+Frx8ePHs2HDhuPqqHbv3s20acfWlAkhNkopF3R1XCOiMjDoSCwGhw6pCKq1dWRNYpubVe3TG2+osUyZAt/9rur/1MuamJSS6lgTW0KHORKpw26yUmDJTHn9KRY3U9+QzsEjhbS2uTiU/j4f5DxMXCQcIUQjK52PAjAzcvbR6MnpgPHjlbBbjl8ji8sIu8LvsSn0IhXx7ZixMd1xIWc4VjDGOqfX6cdlV5ez8qkJ7aIDSniWXX00KULXIRSwEA5aiEaOilNaRmSQxUlT7awEWE1WLCYrFpMZ0UeHjtGElMc+kt6+SSN6h8PDY4/9kZYWPz/4wf2YTOqrmPT9TZIs+I3FYpgGYXrciKgMDEBFLfv3K4FK1kC5XCMzlvJyePlllcWnaWr9acUKmDat1ynHuNQoi9SzPrifRq0Vj3CQYXanvP7UFnBQUZ1DaUU+mmbC4wrhcMT4fdo38Zkaj9s+PZ7F12v+U0VO2Tng7trYtSlezqbQi2wNv0ZI+skyj2W+8zpmOS7H1ceWGDvWZbHqlTH4mmykZ0VZdnU5U2a1sHdrJl//XJiM9OmAwGLVcLo0HO54+7TeQJDoaIkrt0mobDxLYn1pNFkQXXTRIiKdau8eeeQJZsyYBRwVoI5ClEQIJThms6psMJmOPoQ4+u9AZ76NiMrAoC+EQsogdtMmFU3l5g5eDVRfkBJ27oSXXlLTjTYbXHyxSpBIIYsqrEc5GKlmQ/AAAS1Cptmdcu8nTRc0NXk5XFZAQ5MXs1nDmxY8xjXCJ44XKQCfpRnmzGnvitsRXWociH7ChtALHIp+igkzp9nPZ57zesZb5/c7jXnmwiZmLmxCiwsO7kpnx7psXntyIvGYia/ctguPN4bTHcdiHbg46VJHl0lxMuGwOBLiNLTuDwPh3XfXAkqEkoIkpUq2hKNtuZKPjmI01LlB/WVYhUoIsRz4H8AM/ElK+ctutjsT+BS4WUr5/DAO0eBUoa1NCcOWLer33NyRcTHXNFizRgnU/v1qHewzn1FJEinYG7VqQXaHKtgaPoym6+RYvGTZUqt/CkesVNdmcai0gEjEhtMZITur6+SIdD0bn7mLiMqUf5xIBfRmtoT+wcbQS/j1GtJMuZzn/hfOcFxDmrl/beeTSAmVhz1sX5vNrk3ZhAIWXJ4Yc8+uZ8aCRrwZUdIyYgM6RkdxMptMOC0OlZnXz47Ew4GmHRWl5DSd2ay+0lbr0ShpNItRTwybUAkhzMBDwMVABbBeCPGKlHJXF9s9ALw1XGMzOIXw+VS32u3bE7U8eeq2crhJJki8/LJKey8shLvuggsuSCmia4z72R4qZXe4AjOCHEs6VkvvF9JkckRpRS5VNbkIJGme4HGdchNbt9c9LWMFK9P/RoyjU0pWHCxz39X+e1VsF+tDz7Er/A4aMcZZ53Ox5xtMtZ+HWQzsHDc32Nm+Npvt63JorndgsWpMndPC7IUNTJjm79AOpH+t17sSJ0vCZXy00VmUQH2F7XYlSsl1oxNRkLpjOP+HLgQOSCkPAQghngGuAXZ12u7rwAvAmcM4NoOTnaYm5Ri+Z4/63zxSbTb8ftX76bXX1M9Tp8LnP6/WoVJIkKiJN7MpeJDSSD12YaHIkoUphSm0eNxEfWM6h0oLafF7sFljZKa3dm1rpGuqT5WesDQaP56ZafMgNJZVgYfx6bWkm/JZ5r6L0x0XsC30OhtCz1MV34lNuDjDeQ3zndeTa5nYv3OUIBwys2tjFtvX5lB+MA2EZPxUP0uWV3H63CbszoE1QTq65iQxm8yjUpyS03e6fqwoORyJ3pCJSGk4qyVGguEUqmKgo19JBbCo4wZCiGLgWuACDKEyGAzq65X33YED6pazL0W6g9VeA6CuTtU/vf22iqbOPBOuuy6lAt2kQeyG4AHq4j48wpFygW4obKOyJpvDpQXEYhbc7jC52cf7zwFqESMcVle+gkKVUGLv0IRw+63wyj3QZEMfc5hd1/+GdzKvJiCbyTaP41LPd5jtuAy7ydOXM3MMug6H96Sz7dMc9m7NJB4zkZ0fYtk15cw8s5H0rGjvO+mBjtl6x645jQ5x0nUVMSVz3Mzmo5FSUphOdlHqiuEUqq5Ob+fbud8C90gptZ7+Ewoh7gTuBBg7duxgjc/gZEFKVZy7caNykXA6+16kOxjtNUC1+3jxRfjoI3X8pUtVB90UvrcxGedQuIb1wf349SAZJk9KCRJSQovfTVlFPlU12QiTjtcTxGLpIgKRUiWUxOPgdMGkieBNP84Qdse6LF57ajzxvPWw7He0Tn+eVpNGQWAZ1xStYIJ14YA83hprHWz9NIfta3NobbHhcMWZe3Y9sxc1UDguMKCLc8c6JyEENrMdq8mK2WTiuT1Pcf/qe6lsLaM4bSw/Pvtn3Hj6bf0/WDf82799ia985Vucfvr0Y57vOI0HSoxcrmPXlbpj0SKV3dfU1EQoFKK4uBiAl19+uU/tNTweD21tqfWjuu+++/i///s/vvjFL3L//fezZ88evvCFL7Bp0yZ+9rOf8Z3vqPYfJ3I/qgqgY0FKCVDVaZsFwDOJL3wOcLkQIi6lfLnjRlLKPwJ/BJWePlQDNjjBkFIVxK5fr9wkBuIiMZD2GskMvhdeUGLpdKoC3WuuOd5MtgtCeoS94Uo2BQ8SkTGyzGkpOZgnp/cOlhbib3Vjs8bJyuzGOUJLTO+Bipxy87pNLddkjLf2fkz8jluheD2E02Htv8H6rxISxUz8z629jq0rImETuzdlsWVNLhUH0xBCMmmGj0tuKGXKrJYBZu3JxLqTBAE2kw2rWdU5JT/jc3ue5Bvv3Ukors5DRWsp33hPWRMNtlj97nd/Ao5fX7JY1NejYwZeqqxdq7L7HnvsMTZs2MCDDz44qGPujrvvvrtdkLKysvjd737Hyy+/fMw2J3I/qvXAFCHEBKASuAW4teMGUsoJyZ+FEI8Br3UWKQOD40h20l23Tk2xpaWlFLH0SF/bayTHsW6dEqi9e1XW3u23qww+T+/TYa1akJ2hMraGjgCk7CARjliprM7mUGlh+/Retz2fIlGIhNVVsbhIiVQXtkagsvc2hV5iY+h5Qpc2QsNUWPkgbL0Dourz+PpoNyQlVBzysGV1Lrs2ZRGLmMnOD3HBijJmLWokLX1gGXsSSVyPA2AxWXBYbN0W4d6/+t52kUoSige5f/W9/RaqQCDAF75wE1VVFWiaxne+8yOuvvpmVqxYyn/8x68588wFjBvn4ctf/iqrVr1LZmYmP//5z/ne975HWVkZv/3tb7n66quP2ecHH3zAj3/8Y7Kzs9m7dy/nnXcef/jDH/pcSPv000/z85+rflRXXHEFDzxwtB/Vt7/9bVatWkVmZibPPPMMubm5/O53v+ORRx7BYrEwffp0nnnm+H5UeXl55OXlsXLlyn6dr1QZNqGSUsaFEF9DZfOZgUellDuFEF9OvP7IcI3F4CQhaXO0bp1ycRhIJ93OpNpeA9S02YcfKoEqL1eZhF/+Mlx4YUoZfE3xVraFjrRn8OVZ0rGksKDv87soq8yjvCoXk9DxpqUwved2Q8lE8GZ0e/teFz/AuuCzbA+/iUaUibaz0J97lOCWK0Ee+55U14wCrRa2fZrDltW5NNY6sdk1ZixoZO7iBoontA1oai+ux2mLtRHXYyCLcFodWEzWbluvJ6lsLevT86nw7rtvkp9fxN/+pi7cgYCvPWLKyFBGJ4FAgAsvXMqvf/0A1157LT/84Q9555132LVrF3fcccdxQgWwbt06du3axbhx41i+fDkvvvgiN9yQepuPqqoq7rnnHjZu3EhmZiaXXHIJL7/8MitWqH5U8+bN4ze/+Q33338/P/nJT3jwwQf55S9/yeHDqh9VS0tLv8/JYDCseblSyteB1zs916VASSk/PxxjMjgBiceP2hz5/UNjc9Rbew1Qr73zjqqBqq9XruDf/rZqsZFCRmFtrJnNwUMcitRgN9koTMHiKFmce+BIEU0tHjW9l9FL9p4EsjIhN7/b6T0pdQ5G17A2+AyHY+uwYGeO4wrOdN1ErmUiO+ZksXKnJNZBlzpbFx13+ERixOZPctm3NQNdN1EysZUrP3uI6Wc0YXP0P2tPIgnEAoTjISwmK2PTx1KUVoTdUonNnFrBdnHaWCpaS7t8vi90TIA47bRZ/PjH3+GBB+7hqquuZOnSc4Gjrg6g+lEtX74cgFmzZmG327FarcyaNYsjR450eYyFCxcycaLKovzMZz7Dxx9/3CehWr9+PUuXLiU3V00h33bbbXz44YesWKH6Ud18s+pHdfvtt3Pddaof1ezZs7nttttYsWIFK1as6NM5GWwMZwqDE4doVGXvrV+vLsCD0eq9O7prr7F0qSoWXrlSddH1+1Xm3l13wfz5va6HSSmpjDWyIXCAqlgjLpOdEmtOj4kIH7Tu4PGmVTRoPtLiOSxs/Bxz5Dhyu2tKmDSGNSWz97K7jexiMsy28OusCz5Do1ZKmimXZe67OMN57THWRskWGp2ti7pqreFvsbJ1dS6bV+fib7Lj8sQ4c1ktZ5xdT05hV/VaqRPVo7RGWgHIdecyK28WmY5MLO1Fx5Up7+vHZ//smDUqAKfFxY/P/lmP75Py6FoTHE2AsNlg0aKpbN68kddff50f/vDfueSSS/jxj398zPutVmv739tkMmFP/G1MJhPxeLzLY45EP6qVK1fy4Ycf8sorr/DTn/6UnTuP70c1XBhCZTD6iUTUms+GDernnBwlUkNNsr1GkqYm+Mtf4M03lRAsWAA33KCEqhe0hAffuuB+GuN+0syulFpsvNWwhz/6XieWMINttTTwYd6DpIWCnZzLOzQltNuVMWxmBnSzxtWmNbIh9BwbQy8Skj4KLadzjfcnTLdf1G1xbtK6qCt0HQ7uTGfzJ3ns356BlIIJp/u46Loypg4wMUKXOm3RNqJ6BKfFxem5p1PgzsdpHZgXY3IdKpWsv47iJIQSpWTaeMfguaqqiqysLG6//XY8Hg+PPfbYgMaYZN26dRw+fJhx48bx7LPPcuedfetHtWjRIr7xjW/Q0NBAZmYmTz/9NF//+tF+VM8//zy33HILTz31FEuWqH5U5eXlLFu2jCVLlvDUU0+lnBk4FBhCZTB66ejDF48rgRoJH76aGpVi/t576mp1zjlKoCZM6PWtnVPMM02994CSElp8Hg6VFfBX+/8Ssx67DhQTUVY5nlNCpUsIBdW40tKUQKV56KqtBqj1p0+DT7Ez/DYacabalrDIdRtjrXP7lV7e2mJlS4foye2NsviSauaeXU9WbqT3HfRARIvQGm1FICjyFjHGO4YMR0ava0994cbTb+s2caLjlJ7JpIpsbbaes/O2b9/Od7/7XUwmE1arlYcffnhQxrl48WK+//3vs337ds477zyuvbZv/agKCwv5xS9+wbJlqh/V5ZdfzjXXHO1HtXPnTubPn096ejrPPqv6Ud1+++34fKof1d13301GRsZx+62pqWHBggX4/X5MJhO//e1v2bVrF95BboljuKcbjD4CAZXevXmz+n2kfPjKyuD551WihMmkkiOuvTYlk9iIHmNfpJL1gf3tKeYuU88iq2km6hu9HDhcjL/Vhd0e43+Kbui2AvGHVX9Q60/Z2SqBw9l1hCGl5HBsHZ8Gn+RQdC1WHMxxXslC581kWfqeHSl1OLTHy6aP8ti3PROpq+hp3pI6ps5p6WBn1Hd0qdMabSWmR3FbPUzMnEieOw+7JbUblKKi3UyaNK33DbtB09QDVKTUUZxGqtD2gw8+4Ne//jWvvTby/ajuu+8+PB5Pe3p6bxj9qAxOPvz+oz58JtPI+fDt26cE6tNP1ZXqqqtUm43s7F7fGtQj7AmVsyl0kHjCJNZu6llkI1EL1TXZHDhSRDSaSC9PrD+ly+wuncvTtSwoLILs7tPLNRljZ/gdPg09SV38AB5TNkvddzHfeW23jQl7/GxtFrauyWHTx3k01ztweWKcdWE1Z5xTT1be4ERPJkwUe4sZ4y0h3ZExoCLiVOkoTlbr0YLbkfjqjXY8Hg9//OMf8fv93H///d1uZ/Sj6oQRUZ0ENDfDtm2wa5e6OuTmDr8Pn5RKIJ97TnkCejxw5ZXqkcI0RqsWZEeojO2hI0gg1+LF2osRa1vAQXlVDqXlBUjA6wlitWrHbLPD+gkrnY+2r1EBWLFzhef7zHRd3uV+w3obm0MvsS70LK16PTnmCZzluo2ZjkuxiL6Ztibdyjd8mMfuTVlocRNjJvuZf24dp89tHtjaEzptEbX21J/oqStSjag6i1MychoJ+8fBJulY0ZEnnniCWbNmjdCIjseIqAxOHBoaVJuNffvU2tNIGMVKqZI0nntOGdZmZCiT2OXLU2qc2BxvY3voCDvD5ViEidxeaqCS60+Hy/Kpqc/CbNJI97Z1M10mmemfB6EQqzJew2duajeDnelcftzWfq2WdcFn2BR+magMMt56Jlem3ctE21l9jkyiERM71mez8cM8aivc2BwaZ5xTz/xz68gtCvVpX8ftW4/SFmlDIilKK2Js+lgyhiF66ihONpsqKeucDHEykHSsOJkwhMpg+KmtVQkShw+rW9ni4uHvSZDsA/Xcc2ocfSzSrY/52BI6xP5INXZh7bUGStMFDY3pHDhcRLPPg8MRJbs7eyNdJtzLNSWcVd+FJ/8XmmyQFYWry6FD9l1d/ABrgn9jZ/htJDDdfiFnuW6j0Hp6n09LY62DjR/msfXTHCIhC3nFQS6/9TAzFzQOUt1TGLvFzmm5p1HoLsRhdfT+5gHQOXI6WcXpZMcQKoPhQUqorj7qw+dy9d+HbyDE4/DPf6o1qMpKJZLf+Aacf36vixJSSmrjLWwMHqA0WodT2Cm2ZPfYZiMWM1Ndl8mBw8WEIjbczjB5Od24l+saBBI1Pbm5kJvLjm3FrHxhArGourL6muysfGoCEkna3HdYE/wbB6NrsAonC5w3sND1GTLMhX06JboOB3ZksP6DfA7vScdk1pl2RjMLzq+lZOLAXSNao61oMk6uK4/Z+bPJdGQOqVu5rqs/MxxdczpZpvVOVQyhMhhadF0Jwtq1yofP4xm6It2eiEbh3XdVmnldnUot/973YPHiXq9gutSpijWxLrCPmlgzbpODEkvPRbqhsI2KqhwOlRWiaSa8niAedzfFrvG4iqBMZpVRmJ0NVrWWtOqVMe0iBYDQiE1+idccD6C1bMAtMlnq/jLzndf1OUEi2GZhy+ocNn6Yj6/JTlpGlPOvquCMc+rweLsuPE2VUDxEMBbEbDIzPmM8xd5iPLb+t//ojUhElbnl56vf09IMcTqZMITKYGjQNNViY906dQUZTB++vhAOqwLdl19W4zjtNPjXf1XFur2ECrrUKY3UsT50gIaYH6/Z2WsNlL/VSWlFPhVVuQiTTnpaELO5mymzZP8nqxXGjlM2R53auvuaEskPljDM+Suc/WvIPoDWOJnL0u5hjuMKLKJvyQc15S7Wf5DPzg3ZxGMmxk3xc9H1ZZw2u7lzd48+oaMTiLQR0SN4bGnMzp9NnjsPq3loSgtiMfUnjcXUlN6CBWrWdjhqwQ2GlwEJlRDCDsyWUq4fpPEYnOjEYsqHb/36ofPhS4VAQNkc/eMf0NoKs2bB3XfD7Nm9ClRcahyO1LAuuB+fFiTD5GaMrfv2HFJCU0sah0oLqWtIx2rVyOzOfw+Ue3kkCk5Hov9TRrdrdN6COvxT/gyL/gc8dVC5AP7+HN7ay5n/0x2png00TbB3awbrVxVQfjANq01j1qIGzjy/lrzigSVHxPU4/ogfHZ1CTyHjM8YPWXKErkNLiwpArVaYNg0mT1YzpSaTqg/vC08+Cffeq0rmxo6Fn/0Mbhv8dlR86Utf4lvf+hbTU3AxSZXR0I/qySefbHdh93g8PPzww8yZM2fk+1EJIe4GZgMZwHTgb6gWHganMpEI7N+vBCocHlofvp7w++GVV5RIBQLqNvvGG9UVrReieoyDkRrWBfcR0CJkW9J6bFSYLNDdf6gYf5sbpyNCTlY3CRJICIWVkHvSYNz4Hh0k/Fod60LPELzzZTAF4MCl8Mn34PAyrDadC249nMrZINhmYfPHuWz4MI/WFjsZ2WEuur6MOYvrcbq03nfQA6F4iEAsgMVkYXLWZIq9RQO2NeqO1lbw+dQ9xsSJ6s9ZWDiwWqcnn4Q77zzalqu0VP0Ogy9Wf/rTnwZ3h4yOflQTJkzgn//8J5mZmbzxxhvceeedrF27dlT0o1oEvCqlfFII8bCU8qeDMhKDE5NgUKV1d7Q5GuDdU79awDc3q+m9N95Qorl4sRKoSZN6PVxYj7IvXMWG4H4iMk6OOY0sW1q328diZqpqszhwuIhIxNZze/eOLTYyMpRJrNvd7b4b46WsCf6NbeHXkehMd15E7uGvsvn1S3s1g+1IbYWTdasK2LE+Gy1uYsLpPi67pZTJM1sGlGApkbRFW4loanpvbv5c8tx5HUxhB4/kupOmqaTMCy5QUY/TOTj7v/feoyKVJBhUz/dXqAKBADfddBMVFaof1Y9+9CNuvvlmli5dyq9//WsWLFiAx+Phq1/9Ku++e+L3ozr77LPbfz7rrLOoqKjo4xlLjT5/u6SUtwghrhZCPAF0/z/O4OSmtVXZHG1NdHcdLJujvraAr69XfaDeeUdd0c47T/nwpdA4MahH2B0qY1PoELrUyTb37CLRniBRWoimC7xpQdI83SRI6HoixVxXYpufD47ur7BVsV2sDj7OnsgHWLAxz7mCRa7byDQXwVxYMrf3Lrq6Dvu3ZbB2VQFl+71YbRpzFtdz5tJacgfoWq5JDX/EjyY18t35TMicQKYjc9Cn9zRN3XOEQipbb8ECFUF1YTM3YMq6aTvV3fOp8Oabb1JUVNTeSNDnO/4GJhAIsHTpUh544OTqR/XnP/+Zyy67LOUx9YV+3QZJKV8RQrwFfFcI8X9Syn8Z5HEZjFaam5WDw86dKqVqsG2OUm0BX1WlBOr999V80AUXwHXXpeTD16aF2BEqbe+k25uLRGubk9KKPMoq8xBCkuENdJ8g0THFPC9fCXg3dVnKg289qwOPcyS2HrvwcI7rDha6bsZtSj0jIBwys3V1Dus/yKel0YE3K8KF15Yx95yBT+9FtQit0TZMwsS4jHGM8Y7BbRv8+9O2NrX2ZDLBlClw+ulQUDC05XVjx6rpvq6e7y+zZs3iO9/5Dvfccw9XXnkl55577nHbnIz9qFatWsWf//xnPv7445TH1Bf6fYWRUkaEEL9CTQUanOzU1SkXiYMHVd7vULlI9NYCvqxMFel+9JE6/vLlSqBye2+Z4dMCbAseYUe4rNdOukkHiYOlBdTWZ2GzxLpvUAgqFAgEOqSY53QbYUqpszfyTz4J/pXq+G48pmwudH+Nec5rsZtST+FuqrOz/oN8tq7JJRoxM2ZSKxdeVz7g7D2AYCxIMB7AYXEyI3c6BWmF2Mx9s1/qjVgMGhvVv8mpvXHjVA34cPCznx27RgUqivtZz+2oemTq1Kls3Kj6Uf37v58a/ai2bdvGl770Jd544w2yU/DD7A+DlUzx4eAOy2BUIKWKXDZsOFqkW1w8tEW63bWAz8iAX/5SuUnYbMoo9tprU8pFboz72RI8zN5IJTZh6dFFQtcFDU0qQWK9toH12b+mdUID6TKbZeEbO/WA4miTwh5SzJNoMsaO8FusDj5Bo3aETHMJl6d9n9mOy1NOMZcSSvense79AvZtz8BkksyY38TCC2ooHBvsfQc90O69p0XIcGYyI28GWc6sQS3OlVIlRbS1qUBz9mwVQY1ESnlyHWows/5OtX5UZWVlXHfddTzxxBNMnTp1UD5rVxjJFAbHk6yBWr9e3fIOZ5FuVy3gTSY15bhli0qQuPrqlIxi62ItbAoe5FCkFofJ2qOLRDxuoqYuk/2HiwmGHJRlvc1H2X9sN4P1iUZWOh8FUGKVrIGy2xJNCjPpLoyJyTCbQ//g0+CT+PVa8i1TuNb7n0yzX4CpB1/Ajmhxwc6NWax9r4DaCjdOd4xzLq1iwXl1pGXEUtpHt/uWGr6I6jtUlFbEuIxxZDgyBrTPzkQi6qukaUoQzjtP3fOMtEP5bbcNbobfqdaP6v7776exsZGvfOUrAFgsFobCJLxf7ulCiKuBGwG3lPK6QR9VHzDc0weRaPRoDVRrq7r4eobOTaBbPvgA/vxndesNai7ouuuUk3kv45FSUh1rYn3wAJXRBlwmO1nmtG6nSiJRC5XV2Rw8UkQsbiHNE8Rui/P7tG/iM3XdXuPrVT9VNVBFxaqQuZvoLKy3siH0POuCzxCULYyxzuUc1x1Msi1Oeeom2GZh00e5bPgwnzafjZzCIAuX1TJrYQNW28A6H0T1KP6IH7NQ7hFj0sfgGsT08o41Ty6Xip4mTUrpHqPfdOXKfaJj9KMykikMQF1J9u5VKebRqLLwGYo0q96QUkVNb76pRCo9XU3vpeBkrkudylgj6wL7qI378AhHjy4SwZCdsspcjpQVAOBNC2CxHE2Q6KoHFIDP1KTmqrxpdFcD1aY3si74DBtCzxOVQSbbzuZs1x2Mtc3t8TN0pLHWwdr389n2aQ7xmJmJ03xcdfthJk73DXjmVdU/tWE3O5iZO2PQ15+S0ZOuq2Bz1qyRMcY3GHxGbT8qIcRngf8GIsC9Usq/CiHOAq4ALgeukVKmlDwvhFgO/A9gBv4kpfxlp9evAX4K6EAc+KaUssc0EiOiGgA+H+zYoR6g1odsg7tgnhLJVhvPPqtafmRnqwjqkkt6dTLXpEZppJ51wX00xdtIN7vwmrsXNX+ri8PleVRV52AySbxpgS5bbHQbUZkK+HrOP7rcd4tWxZrg39gaeo04UabbL+Rs1x0UWFObu0+uP619t4D9OzIxW3RmLWxg0QW1A26tIZEEogHCWgiv3cvkzCnkunMHbf1JShU9tbWpMrE5c1T0lNZ9OdqQcDJGVH3lVO1H9WOUIB0GviaEeAeYBjyFEpJURcoMPARcDFQA64UQr0gpd3XY7D3gFSmlFELMBv4O9L1XgUHP1NWpRoX796skgPz8kbnd1XWVHPH3vx9ttfGVr6hWG73UZMVknEPhGtYH9+PXg2SaPN3aHEkJzT4PBw4X0dCYjtUaJyuztfvIREqW+a5hZfoTxExH13+sOFjmvuu4zRvih/kk+Dg7wm8hEMx2XMHZrttTbvOuaYJdifWnmnI3Lk+Mcy+vZMH5tbjTBmYOq0sdf9RPXI+R68pjTtYcshxZg1b/1Dl6uuACI3oaaU7VflRtSS8/IcRPgFpgqpSypY/HWggckFIeSuzrGeAaoF2opJQd00rcwIndfng0kXQx37BBtdtwOlUa9XD3gQK1ov7RRyrNvLy8T602onqM/ZEq1gcPENIjZJnTGGPteopP9YDycuBwMT6/W/WA6tbiiGP6QM3MvgSchayK/RmfXttlw8Lq2G4+CT7Gnsg/sWDjTOcNnOW6Da85P6XTEA6a2fRxLus/KKC1xUZOQYjLbz08KOtPyQQJXeqM8Y5hXMY4vPbBWRySUjlV+f3qa3TmmWo2dLijJ4NTh1SEqkAIcSewN/Go6IdIARQD5R1+r6CLGiwhxLXAL4A81PSiwUCIRlW0smGDmurzekfGgw9UKveqVaoXVE2NKpr5znfgnHN6vQUP6RH2hCvYFDxEVMbINaeTbe36ytgxgy8UsuN2h8nJ9ne/846NCnNzVWTncMK6+fDKj45pWCjPbKQstoVPgn/hUHQtduFhievznOm6GbcpM6XT0NJoY92qArZ8ouqfxp/m44pbDzNpuq+7vIyUiekx/BE/JmFiYsZExqSXDJr/XrLuKR5X9zjnnjs6MvcMTn5S+Yr9B6pu6jZgFpAmhHgX2AxsllI+leKxurqPPe62UUr5EvCSEOI81HrVRcftSAnnnQBjB1JGfjITCKgEic2blViNlEksqOO/845ykmhoUIsXP/gBLFzYa0TXpoXYFSpnS+gQEsgxe7GZuv7adpXB17NA6RAMqG9hXi7k5revie1Yl8XKpzo2LLTx6qd7+LDgPpocG3GJTJa5v8J85/U4UizSrSp18+l7BezelIVAMn1+E2ddVEPBmIHVPwFEtAitET82s51pudMoSisatASJtjZ1j2OxqMSI005TCaEGBsNFr0Ilpfxjx9+FECUo4ZoFXIZaq0qFCqDjlbIEqOrhuB8KISYJIXKklA2dXvsj8EdQyRQpHv/UoKFBJUfs3q1EYKQSJOBoL6iXXlJ1UKefDl/9Ksyb12vRsE8LsD1Uyo5QKSYEuT24SARDNsoq87rN4DuOjjZHBQWQmwPWY5M22hsWCh2mvQjn/hytcDPNrSVcmvMd5jqvwip6t1CQOhzYmc6n7xZSut+L3RFn0QU1nLm0lvSsaK/v742kg4TL4mZOwVzy3fmDYhCr6yp6CodVbsuFF6oAeKS+Sic61157LYcPH6atrY36+nomTJgAwB/+8IdjjF17o7t076547LHH+O53v8s111zT7t7+i1/8gj//+c+YzWZ+97vfcemllwKwbNky1q9fzwcffMCCBV3mM4wo/TGlrUCJzut9fOt6YIoQYgJQCdwC3NpxAyHEZOBgIpliHmADus4TNjhKcv1p0yb1r90+sivanXtBzZ4N3/62uh3vRaCSLhL7IpVYhYWCHlwk/K1OjpTnU1mdaFLobesyg68dTVNTfEL0anPkaxGqUeGSX0LuHmiYCi8/itx+K2f+vneD2HhMsH1dNmvfK6Shxok3M8JF15dxxtl12J09iGgKKAfzNiLxMBkO5SCR7crGNNB5Q1RyREODWoc67TSYMUPNhA6lGcmpwEsvvQQMf03UzTff3N7+Y9euXTzzzDPs3LmTqqoqLrroIvbt24fZbGbVqlUs7a1DwQgybLPLUsq4EOJrwFuo9PRHpZQ7hRBfTrz+CHA98DkhRAwIATfL/lQknypEIqpAd+NGtbKdlgYlJSN3VemqF9RNN6lIqgeklNTGW9gcPMjhSC0Ok42iblwk2j34jhRS15CRyODrIUECOgiUSS2q5OR0a3MUlxG2hF5FfPNGpLcMaubAc8/CrutBmknPinT5viShoJmNH+ax/oN8An4b+SUBVnzhINPmNfUsoimgo9MaaSWmR8lz5zMpc96gNSj0+9X0ntMJZ52lkiN66EZi0ANHjhxh+fLlLFq0iM2bNzN16lQef/xxXL3UAnbkvffe4zvf+Q7xeJwzzzyThx9+uN0T8Fe/+hWrVq0C4KmnnmLy5Mk899xz/OQnP8FsNpOens6HHx7vavePf/yDW265BbvdzoQJE5g8eTLr1q1j8eLFg/PBh5BhXQaVUr5Op0gsIVDJnx8AHuj8PoNOtLSoHlDbt6uV7ZFcfwLVNOjll9U0Xx96QelSpyrWxIbAAapijbhNDkqsOV1eeHVd0NjkZd+hYlr8Hhz2XjL4QJ2bYFAtrowpgazsbgUqogfYFH6JtcGnaNMbybKdge/Z36Ptvork8qrVprHs6vIu39/SaGPd+wVsXp1LLGJm4vQWzrroEBNO62WMKaBLHV/EhyY1ir3FTMiYMCgZfJqmpvciERWAL1lykiVHfPObqoB8MJk7F377214327t3L3/+858555xz+P/+v/+PP/zhDym7OYTDYT7/+c/z3nvvMXXqVD73uc/x8MMP881vfhMAr9fLunXrePzxx/nmN7/Ja6+9xv33389bb71FcXFxty05KisrOeuss9p/LykpobKyMqUxjTQny1fy5EfXVVr51q1w5IiassrJGdmrSl0dvPji0V5Q556rBKqXBJfORbpes7NbF4l43ERtfSb7DxcRDDpwuXpoUpiko1HsuHFKyLspbA3pPtaHnmN98FlC0s9465ms8P6UcdZ57FyUzaraaI8NC2vKXax5p5BdiQSJGQuaOOuiavJLBlagC0dbvAOMyxjHuIxxg2JxlJzeA5g+XT0G2uvS4FjGjBnDOecoA+Pbb7+d3/3udykL1d69e5kwYUK7yesdd9zBQw891C5Un/nMZ9r/vfvuuwE455xz+PznP89NN93U3qajM11NTg12P7GhwhCq0U44rNLLN21SczMez8hO74FaB3v+eeXJ14deUDEZ52C4mvXBA7T2UqQbjVqoqslm/+FEBp871HMGHxwVKFvSKDar26zCNr2RtcGn2Rh6gagMMsV2Lkvcn6fYOrN9m5kLm7rspCslHNrtZc07hRzZm47NobFoWQ1nLhucBIlkirlZmJmcNZkSbwkO68B7XySn99xuOPtsmDy5V2eqE5sUIp+hYiBtOVJwCzru50ceeYS1a9eycuVK5s6dy5YuIsmSkhLKy4/OCFRUVFCUQv+20YAhVKOVxkY1vbdjh4qmRnp6D1Qk99xz8MknKpK77DLlxddLL6iwHmVvuJKNwQNEZIxss5eMbop0gyEbFVW5HCotRALetCBWSy/p2+1O5vaj7WC7SSxo0ar5NPg3toReRSPGdPtFnO3+HPmWKb1+/KSDxKfvKgfztPQoF6woY9659TicA2tQCIkU82grVpOVabnTKE4rxmoeWNdkTVMzs+Hw0em9khLDOWKoKSsrY82aNSxevJinn36aJUuWpPze008/nSNHjnDgwAEmT57ME088wfnnn9/++rPPPsv3v/99nn322fb1pYMHD7Jo0SIWLVrEq6++eowgJbn66qu59dZb+da3vkVVVRX79+9n4cKFA/+ww4AhVKOJWEw5NWzZogpibbaRszfqyN69SqDWrVOr7ddeq1pt9FJM06oF2R2qSNRASXLM6d3WQLW2qQy+iqoUM/gAIlGIhJW7+qRJPTqZN8ZL+ST4V3aE3wQEsx2Xc7brsynZHEXDJjavzmXt+wX4m+zkFAa58rOHmLmgEYt14Lk+SZNYp8XF7PzZFLgLBpxiHo2q6T1dh2nTVPZeCr0lDQaJadOm8de//pV//dd/ZcqUKdx11/HWW93hcDj4y1/+wo033tieTPHlL3+5/fVIJMKiRYvQdZ2nn34agO9+97vs378fKSUXXnghc+bMOS6qmjFjBjfddBPTp0/HYrHw0EMPYR7pa0uKGEI1GmhpUb5727erBYSRdI9IIqUaz3PPqXWxtDS49Va44opevXKa4q3sCJWyM1yOuYcaqI5ddOvqM7Fa42T21EU3SSSiHk6nmr9K99Kdk3ltbB+fBP/Krsh7WLAx33k9Z7luJz0Fm6M2v4X1H+Sz8cN8wkELYyf7uezmUibPaBmwgwRAIBYgFAvisacxr2D+oJjEJlu62+2qnnrqVCN7byQwmUw88sgjxz2/dOnSlNLAL7zwQjZv3nzc88lW9f/xH/9xzPMvvvhiSuO69957uffee1PadjRhCNVIEY+r7rlbt6ooymJRK9q9mLEOKR98AI8/rm7FLRY1xsxM+MIX4NJLe1zQOJpifohDkRocJlu3nXT7lcEHKnqKRNWVd+zYHlttVMS280ngMfZHP8YmXJzt+iyLXJ/Bbeq9lWxTnZ1P3y1k66c5aJrgtNnNLL6kmpIJgV7f2xvtLubxEBnOTGbmzRxwDZSUqp46EFAzxBdfrHJIRvKrZDDyOJ1O3njjDb70pS+1F/x2x7Jlyzh06BDWUfql6VfjxNHECdfmo6UFDh5UAhUOq+gkPX3kKyrffx8efFCJUxKLRbmZX3Sci1U7utSpiDWyPrCf2lgzLpODLLOny8XjpAffgSNFBIJO3K4QLufxyQc7rJ+wyvEcPtGoWsC3rmCmfx543KpZYVoaXTlySSk5EtvAx4G/UBrbiFN4Wei6hQXOG3Gaek/nrjriZs07hezekonZLJl9VgNnXVhDdn641/f2Rnubdz1CriuPSVmTyHRkDijrKh5XS5nRKEyYoDKnCwpG/qs0kpxobT6SjhUdeeCBB9odI05WhqVxokEfSa49bd+uoiizWfnSjIZFg6RR7MMPq5X3jsTj8PTTXQpVss3GhtABfPEA6WZ3tynmx3nwuUPdppjvsH7CSuejx7aA9z4BObnMzFxB1wKlsz/6MR8HHqMqvhOPKYeLPN9gnmMFNlPPaW1SwqFd6ax+W1kcOZxxzrmkmjOX1uJJH1iLd1AC5Y+oNhuF7iImZk0k3ZE+oH0mW2uAMvuYPn1k+lwaDJykY4VBzxhCNVRIqabQ9u1TvnuxmFp7Ki4eHbe84TC89ZayOWpo6H67Tq+F9Ah7w5VsCh4kImOqzUY3AhUM2SmryOVIeQFSgtfbewbfKsdz7SKVJCairNIeYybXHvO8LjV2R97jk8BfqdMOkGEq4rK0e5jjuAKL6KXhYiKDb807hdRVukjLiCqLo3PqsDsGZnGkxqa3t9ko8ZYwIWMCHntq5rXd0dampvicTlVTPWWK+tnA4GTHEKrBpq0NSkvV1J7PpxYKsrNHT7l/Wxu89hq8+qry4Zs5E772NTXt15VgJSpBfVqAnaEytodK1dM9uJj7/K6EB18OZnOKGXxICIXxebtpAa/Xtv8cl1G2h99gdfBxmrUKcszjucZ7HzPsF2MSvfSzipjYsjqXT987msF31edUBp/ZMvBpcE1q+MI+JJJxGeMYnzF+QEW6HdefcnJg+XK1PDdavk4GBsOB8XUfDKJRNaW3cyeUlaki04wMVbAyWmhqUtHTm2+qotgFC+CGG9S8EcDnPgcPPaTmlRJIux3/rdex1r+ZA5FqrMJMXjcZfLouaG7xcOBIEY3NXqyWeGoJEkgIhtQ0Y0YG6aY8fLLuuK3STflEZYgtoX+wJvgkrXodhZbTuSH9AU6znYfoJRkh0Gphwz/z2fDPfEIBC2MGOYOv3UVCwKSsiYzxjh1Qka6mqfsGY/3JwMAQqv6j61Bbq6b29u5VVxaPZ/RM7SWprlZtNt57T41xyRK4/np19etIMmX2iSeQDQ1o2ZlsvP4cNs4CV7SR4m5MYjXNRF1DOvsPFdMacOJ0RMnuzSQWVKgQSghUZhYUFoDTxbLQV1nZ+gtiHE1gsGCnyDKdBxtWEJQtjLXO48q0e5loW9RrMkJzg4217xWyZXUO8ZiZ0+Y0sfjiGkomtvX4vlSJ6TF8ER8Wk4Up2VMo8ZZgt/Q87dgTyfonULVPM2YYvZ8MDAyh6gvJdadDh2DXrqNuCHl5I1+U25lDh1Sjwk8+URHeRRepQt3Cwm7fEjnvHA4umsjG4EFa9RDpJhdjzF1PW0WjFqpqszhwuJho1ILHHSK3N4sjUOcwGFSimZ2twgTH0YWWZKv3VYGH8ek12HGjEWd39H0m287mHNfnGWOb0+thjvHgE5JZCxtZfHE1OQUDz+ADiGoRWqNtWEwWpudOH7CLRDCogt5k/dNpp53k9kanGKOhH1VjYyM33HAD69ev5/Of/3x7+w84CftRnXJIqa4gpaVKnFpb1bpTZuboc/KUUk0/Pv+88gZ0OuGaa9Qjq/v6Ib8WZHeonG3hI2hSJ8ucRoa16yrRYMhOWWUuR8oKlMWRJ4g3LYUOtR3bvefkKMcNR9eZAMFd5xNqqobpjxGxBCkOXs5lJbdQYJ3a68c/si+NNW8XcWh3woPvghoWXlCDN2PgGXxw1ObIbrEzK3/WgF0kfD7lwZeerpoTTphg1D+djIyGflQOh4Of/vSn7Nixgx07dhyzndGP6kQkuYJdWqoy9nw+FTFlZqorymhD15W90QsvqGnI9HS4/Xa4/HI1HdkFyQLdbaEjHIxUY8ZEjiUdazdddFv8bo6U5VNVk92HBAmOCpTUEwKVB/auBaohfpiVZc9SXvQqFAnY+ln45B7qWifRcOthCrowiE1+/D1bMlnzdiHVZR7c3ijLriln/rl1OFwD9+CDozZHLoubuflzyffk99tFQtfVvU8opALc889XM8bd+OcanGCM1n5UbrebJUuWcODAgcH5oMOIIVRJklePsjIlTn6/EqfRlhTRkVgMPvxQCVRFhZqC/Nd/VdN89q7XSWIyTlmkno2hgzTEfLhM9m6bFCYdJA4cKaKpxYPdlmqCBAmBCqjzmpenHvaukwuqYrv4JPhX9kb+qURs3ddgzbfBr857DNUavrOTeTwm2LY2hzXvFNJc7yAzN8zltx5m9qKGQfHgg0Sr91gAjz2N+YXzyXXn9dtFQtOgvl792aZOVY2P8/IGZZgGXfDNN7/Jlpotg7rPuQVz+e3y3/a63WjsR3UiYwiVrsOGDUqcgsGjkdNoFSdQ43zrLdVNt7FRtbT49rdVokQ3a2UBLcy+SCWbg4cI6zEyeijQjcXNykHicBGhkB2nM5La+hMkBKoNJJCXC7n5XYqmcpHYyOrAXzkcW4dDpLHE9Xk+vu+XEDz+6u1rsrX/HE500V2X6KJbOLaN67+0n9PmNg9aVNIWa1M2R/aBt3rvmCAxa5ZKkBiNgbnB4DEa+1GdyBhCFQ4rocrL63EdZ1TQ3KxqoF5/XRXWzJqlaqDmzesy01BKSX3cx45QKXvDlZiEiRyzl9xuinBCYRsV1TkcLi1A00x4PCn0gEqi6yqCkqj1p7xcsHYlUDp7ox+yOvB4wkUimwvdX2Oe81rsJg/bHen4uljySs+K4m+xsu79AjZ9nEc0rLronn3xQcZNbR2UREuJpC3aSiQeIcuVzZyCOWQ5svptcxQMqj+ZzWYkSIwEqUQ+Q8Vo7Ed1ImMIFajFgW6mykYFVVUqxfz991U691lnqRTzqV0nFySn9zaHDlEX8+EwWSmydj29B4kC3Yo8qqpViw2vJ4jFkqI7Q1KgQAlUbh5Ybcdtpsk4O8JvsSb4OA3aETJMxV26SCy7upyVT00gFj0aGVqsGulZER780RykFEyf38jii2ooGJNCEkcqH6GDD1++u4BJmRPJcPY/JzzZoNDrhWXLVIssI0Hi1GI09qM6kTGEajSzZ48SqE8/VVYEF1wAK1aolfcu8GtB9ocr2RI6QlSPJfz3us5M1HRBU4f1J5s1TlYq9U9JdA0CCaEoLISc3C6vxqpI9xU+DT6JX68lzzKZFd77mW6/sEsXieQ61KpXxuBrsmGx6sRjZqpK3cxbUs+iC6vJzBl4F11QAtUaaSWmxyhKK2Ji5kS89t7Na7uio4NEXh5ceaWaPTYSJE5NRmM/KlDp7X6/n2g0yssvv8zbb7/N9GTR/yjGEKrRRnLN7MUXVTq8260cJK68ssvKT13qVMea2RY6wpFoLWZMZJu92LqZ3otGLVTXZXLwSBHhsA2Xqw/rT3BUoIRQreezu25NEtR9bAg9x/rg3wlJH2Osc7k87R4m2c7ucRpE6mBz6Hgzo/ia7FhtksUXV7Lg/FrcafFu39cXdKnjj/qJ63HGeMcMyIdP09QyYSSilgovvVQFlqOp5ttg+Bmt/aiS7z/RGFahEkIsB/4HMAN/klL+stPrtwH3JH5tA+6SUm4dzjGOGNGo6gf18ssqgy83F770JdVcqAvn0aAe4VCkmi3Bw/i1EG6TnWJLdrci0BZwUF6VQ2nCIDYtLYjH3YfiV01Tiy7CBMUJgbIcL1A+rZa1wafYHP4HMRliiu1cznZ9ttciXS0u2LE+mzXvFtBQ7cKbFeGSG0qZe049NvvATWIh4cMX8SGl8uEblz4Ot61/XQXjcZXBp2mqg+7s2aN/idPg1OJk6kc1bEIlhDADDwEXAxXAeiHEK1LKXR02OwycL6VsFkJcBvwRWDRcYxwR/H544w1YuVL1qpo4UWXwnXPOcc6jydqnXaEy9kWqABLu5V1fbJP+e4fLCqhtyMBi1lKvf0pyjEAVq1qoLgpc6+OHWBP8GzvCbyKBGfaLOdv9OfIsk3rcfSRsYvPHeax9v4DWFht5RUGu+fxBps9v6ts4e6Ddhw8YnzmecenjcFr7ZzuebLEhBMyZo6wSe2l4bHCKMX78+OMKartjKPtR3Xzzzdx8880pbZusyxqtDGdEtRA4IKU8BCCEeAa4BmgXKinl6g7bfwqM4hzxAVJdrUxi331XRVPz5imLo9mzj5s3CukRjkTq2Bw6hE8L4BA2CrrpngsQi5mpqc/k4JFCgkEHDkeUnFTrn5IkBcpkUost2dldClR5dCurg0+wP/oRFuzMd17PItetZJi7t2oCaPNZE23e8wiHLIyb4ueK2w4zabpv0KbNkgIlhGBy1mTGpI/ptw9fR4ujxYtVHouj/56zBgaA0Y8qVYZTqIqBjqkoFfQcLX0ReGNIRzQS7N6tEiTWrlU1T+efrxIkxo07ZjNd6onoqZwDkWqklGRaPJRYu7dtags4qKjKobQiH003kebuQ3p5Ek1TGQEWC4wZA9lZYOoc2ensj37C6uDjVMS24RTpnOv6Eme6bsRlyuhx9421Dj59r4BtiTbvp89Rbd6Lxw+8zXuSjkaxU3OmUuItwWY+PhMxFZIZfIbFkYHByDGcQtXVfXKXcztCiGUooeoyp1MIcSdwJ8DYsWMHa3xDh6bBmjUqgtq7V9ka3XADXHHFcQsbrVqQQ5FatoeO4NdDOIWNfEtGt9GTpguamtM4XFpAfVM6FrOGNy3Q92mzeFyFDRYLjB0H2ZnHCZQmY4kU87/RoB0m3VTIpZ5vM8d5FTbR81Ra5WHV5n3PVtXmfc5ZDZx1UTVZeZEe39cX2n34zDZm5s6gMK2oX0axnTP4rrhCabaRwWdgMDIMp1BVAGM6/F4CVHXeSAgxG/gTcJmUsssuelLKP6LWr1iwYMHgLGQMBcEgvPOOalJYV6ecwu+8U1kcdZg3isk4VdEmdobLKIvWIxBkmT2M6cYYFiASsVJdm8XB0gIiEVv/pvcAYnEIBVWYMH68yizs5GEX0dvYFH6ZdcFnaNXrybdMYYX3fqbZL8TcQ6NCqcOBnemseaeQsgOJNu+XVqk2797ByeADCMfDtMVacVpczM6f3W+jWF1X60/hsApwL7nE6AFlYDAaGE6hWg9MEUJMACqBW4BbO24ghBgLvAh8Vkq5bxjHNrjU1ioHibffVs6j06erDL4zz2y3OJJS0hD3sz9Sxa5wOTEZxy0cFFm6d0KQUpnDllXmUlWTi0Di8QRJ8/SjdUUXArUj8g6rmh7Gp9eSbsrnLOdt+GUdm0IvEpEBxlsXJPpAndVjirkWF+zYkM2n7xRQX+3Cmxnh4htKmXt2/aC0eU8SioUIxNtwWz2cUTCPPHdev4xi43FlcRSPK/eIOXPUkpyBgcHoYNiESkoZF0J8DXgLlZ7+qJRypxDiy4nXHwF+DGQDf0hcCONSytHXHKUrpFTrT6+8ogp0hVDee1dfDVOmtG/m14IcidSxPXwEfzyE1WQmy5zWrWs5qOSI2voMDpUW0hpwYbPFyExvxWTqRzAZiynxtNnUgktGJphM7Ai9eUzDQp9ew1uB3wCC6fYLWey6nULrtB53HQmZ2PRxHutWdcjgu+Mg0xcMXgYfQCAWIBgLku5I58y8hf324evswTdzpnKTMDAYbEZDP6p33nmH73//+0SjUWw2G7/61a+44IILAKMf1TFIKV8HXu/03CMdfv4S8KXhHNOAicVUc8JXXoEDB9T604oVqkA38WUK61Eqog3sCJVRHW9CIMg0eyixqdv2otU7mPbcKpyNPkLZ6ey+cRmVi2fib3VRUZ1DeWUeui7weELkZvv6P85QCOw2mDBRucJ3WHR5v+0Px3TVTZImsrku/Wc97trfYmX9+wVs+jiXSNjCuKmDn8GnfPjaiGhhshzZzM6fTZazfz584bASKMODz2C4GA39qHJycnj11VcpKipix44dXHrppVRWVgJGP6qTF58P3nxT1UA1Nakao7vuUuZuDgdRPUZ1pI7d4QpKo3XoSLwm53FFuUWrdzDn0ZVYoqqxn6vRx+w/r2TfwWI+LjwTq1UjI72tf9ETQCwKoUQn4kkTIT1D1UQl0GWc3ZH38cvaLt/e2vUyIQD1VU7WvFvAjvXZSF0wbV4Tiy+upnDs4HjwQQcfPi1CrjuPyVnzyHBk9Eug2tpUkoTbrZItJ09WYmVwcvLNb8Jge7POnQu//W3P24zWflRnnHFG+88zZswgHA4TiUTa9zuaMYSqrxw6pNaf/vlPFaWccQZ8/etwxhnEhE5trIV9/r0ciNSgSQ23yUGBJbNbQ9hpz61qF6kk1liM5WueZde/nNX/cUajKnSwO2DSJJVf3UGgInqALeFXWBd8Fp9ejQkzOsc3GUw35R/zu5RQui+NT98t5MDODCxWbdA9+KCjD1+UQncRE7Mmku7oX2+MZBfdrCxlcTRu3HG11AYGg8po70f1wgsvcMYZZ5wQIgWGUKWGpqm6p1dfVa3e7XZVVHPVVcRKCqmNtXAwsJN94SriaDiFjTxLercp5UnCESvOxq6n8jLaGvo31qRAORIClZEOHB2HX6tlfeg5NoVeIiLbGGOdyyVpdxPVArze9sAx039WHCxzKzNNXYPdm7P49N1CqsvcuDwxzruyggXn1eHyDF4Gny51/BE/mtQo9hYzIWMCafa+Wz9IqQLdYFB55i5dqqwJjRTzU4feIp+hZDT3o9q5cyf33HMPb7/9dn8+2ohgCFVP+P0qc+/119WiRl4efOELRC9cSo0jzoFwFQcbdhFHwyFs5Fi8WHpIigDQNBNNLR7KK3Opqc/iAncu2YH647ZrSetj2llSoJwONaeV7qWjQFXH9rA2+DS7Iu8gkUyzX8Ai160UW2e0byOEiVWBo1l/y9x3MVVczrpVuax9vwBfo52svDCXfUZ10bXaBi9BIq7H8Uf9IGFsxljGZ4zHZe37wlFHk9iJE1XAm5dnpJgbDC+jtR9VRUUF1157LY8//jiTJvVsbzaaMISqKw4eVNN7H32kBGD2bMJf+gJVcyexN1ZFWWQtMqLjEDZyU4icpITWNhfVtZmUVeYRi1lwOKJkZ/p5/9zruebdR7HFj06bRS023j3nxtTGGolCJCFQUyYn0tZMiePq7I9+zKfBpymLbcImXJzpvIkzXTeRYS46fl/bb4VX7oEmG3pGlF1jA7x5wEs4aKFkYiuXXF/KlNktgxqVtPvwCZiUOZEx3rE4rH33JjJMYg1GE6OxH1VLSwtXXHEFv/jFL9qjvRMFQ6iSxGKwerUyh92zB2m3E1m6hPKLF7I9D2rjPghsxS3sPa45dSQUtlHXkE5peT6tASdWs4bHEzqmKeG2aeoLc9Enz5HR2khLWjbvnnNj+/PdEomoh9Ol0t+9aSQFKipDbAu9xrrQszRp5XhN+Vzk+TfmOq7BYeq6ncWOdVnHNCxsbbHT2mKjcGyAS24sY8ykthROYupE9Sj+iH/ANkcdU8znzFFt3g2TWIORZjT2o3rwwQc5cOAAP/3pT/npT38KwNtvv01eXt7AP/AQYwhVZaVKLV+zBlpaiBfkUf7ZK1lzVgl+h0Dgx6s7Ke6hELcj0aiFxuY0yivzaGxOQwhwu8I99nzaNu2c3oUpSSSipvg8Hhg79hiB8mu1bAg9z6bQy4SlnyLLdK71/ifT7Mu6bFKYREp454Wxx3TVVQiCbdZBFamjNkd2ZubOpDCtsF82R6GQmuKz22HRIpVi3kU3FAODEWE09qP64Q9/yA9/+MNetxuNnNpCFYshFy+GpkZq50xh09IllM0owWqykGH2UGxK7fTE4yaaW9Ior86hrj4TXQqcjghZma2DtzaSjKDcbmXd7fWStE+sjO1kXfAZdkfeQyI53b6Uha7PUGKZ1bODhCbYtSGLte8XEGjtOprxNQ1O/nbSRcJlcTMnfw4FnoJ+uUgkU8zT0o62eTdSzA0MjsfoR3WyYLUSeeh3/H3rU5innobH5KQ4RWXRNBPNPg9VNVlU12YT1004bLGB1Tx1RccIaty4xLyWQJNx9kRWsT74LBXx7diFmzOdN7PAdSOZXa0/dSAcNLP5k9x2B4mcghBOV4xQ8PgvaXrWwFLOA7EAoXgQr93L/NwF5Lpz++Ui0dICra3K2mj5cnUqzH3XOQODIcfoRzX4nNpCBejLLyFY+RrF5t4zzOJxEy1+D9V1mVRV56DpAptVI90bGFxxgmMF6rTT2gUqpPvYHPoH60PP0arXkWku4RLPt5jjuBK7qedutc0NNtavKmDL6lyiETPjT/Nxxa3KQWLnhmPXqACsNo1lVx+/KNsbykWilUg8Qpaz/y4Suq5SzEMhVU99wQUqxdzI4DM4WTD6UaXGKS9UvRGLmWnxeaiqy6KmNqtdnPrVSiMVOk7xdRCouvhB1gefZXv4TeJEGG89k8vSvsdk29mYekmJrzjkYe17BezZkokQkhkLmlh0YQ0FY446SMxc2ATAqlfG4GuykZ4VZdnV5e3Pp4KOqoGK6zHy3QVMypxIej9cJDRNJUhEoyrTfu5clWJuYGBwamIIVReEwjaaW9xU1ebQ0JiOLgV2W2xoIqckHQUqMcWnS539kQ9ZH/w7R2IbsGBnlmM5Z7puIs8yucfd6Rrs2aLWnyoPe3A44yy+uJoFS2vxZsS6fM/MhU19Eqb2Y0kdX8SHLnWKvcWMzxiP1953d9dYTKWYS6my92bOVF1HDAwMTm0MoQJ0XeDzu2hqSaOyJhu/3w0CnI4ImRmDmBDRFV0IVEj3syX4JBuCz+PTq/Ga8lnm/gpnOFfgMvVsIxQOmdnySS7rPsjH32QnMzfM8puPMHtRA7ZBbLEBHWqggHEZ4xiXMa5fRbqRiIqgzGaYN0/VQXm6zqI3MDA4BTnlhSoWg41bFmG3jEeYdFzOCNn9aUDYV7oQqNrYftb7f8+O8FvEiTDWOo+LPP/GafbzekwvB2husLN+VX77+tPYyX4uvWHwC3RBtXr3R/yYhInJWZMp8Zb0q0g3GFQp5i4XnHOOKgdz9H03BgYGJzmGUMUgFHZRNLb7OqdBpVMWn+ZxsCfyARuan6c8trV9em+B60byLVN63JWUUH5QrT/t3ZaJSUimz1frT4PpYN4+9EQNlNVkZVruNIrSivpVpOv3K6PY9HTV7HjCBNW/0cDAwKArTnmhGjY6JUn4XSE2h55lc+NLtOmNZJiKucjzb8xxXIXT1PP6jhYX7NyYxbr3C6gpd+N0xznnkmoWnF9LWjfrTwMhWQM1kFbvUqr6p0AA8vPh3HOhpMQwiTU4OamtreXuu+/m008/JTMzE5vNxve+9z2uvfZazGYzs2bNIh6PtztYdGwBUl5ezuc+9zlqamowmUzceeedfOMb3xjBTzPyGEI11CS9+Nxu5JgxlDoPsDH0C/Y0/hOJxmTb2Sxw3sAk22JEL/VFgVYLmz7KY+OHebT5Vf2TMohtxGob3PUniSQQCxCOhfA6+l8Dpetq/SkSUR3vL71UCZWRYm4wLLz4ItR23WutX+TnQy/u5FJKVqxYwR133MFTTz0FQGlpKa+88gqgCnGT9ka33XYbjzzyCN/61rfa32+xWPjNb37DvHnzaG1tZf78+Vx88cVMnz598D7HCYYhVENFu1msi/DkIrZbP2Jj6D4aWo7gFF4WuW5hnuM6siwlve6qptzFulX57NyQjRY3MWl6C1ddcIiJ0wZ/La29UaEeIduZw5yCOWQ5+l4D1dkkdtYsVaxrYDCs1Naq0H2wqKjodZP3338fm812jD/fuHHj+PrXv37ctueeey7btm075rnCwkIKCwsBSEtLY9q0aVRWVrYL1datW/n6179OQ0MDe/bsQUrJj3/8Y37yk58M5JONagyhGmw6tNuongSbTC+zI/wWsUiYIssMrkz7ITMcF2MVR7MGdoTePK69xnT7cvZtzWTdqnzKDnix2jTmnl3PmUtrySk4vmX8QNGljj/qJ67HKUorYkLGhH41Kkxm8Amh6p+mTzdMYg1OLXbu3Mm8efN63S4ej/PGG2+wfPnybrc5cuQImzdvZtGiRYBqqnjzzTfz+OOPs3DhQn70ox8RDoe57777Bmv4oxJDqAaLRMv3mEOwc8JBNvEGVdHdWLAz03Ep853XUWiddtzbdoTeZGXrL9obFvoCYV75KMTbG6cQbMokPSvChdeWMfecepyu4zvwDpRkHygpJeMyxjE2fSweW99zw4NB5SJht8PixcqO0DCJNTCAr371q3z88cfYbDbWr19PKBRi7ty5gIqovvjFL3b5vra2Nq6//np++9vf4vWqdet3332XefPmsXDhQgBmz57Nm2++2ecZjxMNQ6gGSiwGoRB1rgY2jdvIdvk+kXgbOeYJXOr5NrMcl+EwdR9SrAo8rESqbjqs/Tps/Rx63EVk/GpuuCGXqbObhyThoGOK+aTMSYzxjulXinkyg8/rVRZHEycaGXwGpzYzZszghRdeaP/9oYceoqGhgQULFgDHrlF1RywW4/rrr+e22247pmPvjh07mDVrVvvvmzZtSil6O9EZVqESQiwH/gcwA3+SUv6y0+unA38B5gH3Sil/PZzj6xOxGLGwj13ebWzOXUuF3IVZt3K6/QLmOa9lrHVur3c5ug6+nYth7dfgyAVgCcHsv8HC36MV7OD0vE8HfdgRLUJrxI/NbO93innHDL68PLjiChgzxsjgMzAAuOCCC/jBD37Aww8/3N6HKhhMvVxESskXv/hFpk2bdkySBUB2djbvv/8+APv27ePFF19k9erVgzf4UcqwCZUQwgw8BFwMVADrhRCvSCl3ddisCfg3YMWwDOrJJ8m85/v8rbICX6oNC+NxamP72Zy2hu0564gQINs0jouc/8ZsxxW4TBm9HjbYZmHL6lw2fpgHTS+CtwwuugfO+DO4GwFINxUMwgfscMxYkGA8oNpsFMwl353f5xTzjm3ex42DSy6BggIjg8/AoCNCCF5++WXuvvtu/uu//ovc3FzcbjcPPPBASu//5JNPeOKJJ5g1a1b7FOHPf/5zLr/8cj7zmc/wyiuvMHPmTHJycnj66afJPgWylIYzoloIHJBSHgIQQjwDXAO0C5WUsg6oE0JcMeSjefJJuPNOzIk7nczWRq5591GALsUqorWxk4/YkraaKtsRzCSjpxWMtZ6R0hxxTbmL9R+o7L14zMS4KX5Ou+ofbBr7/xE3B9q3s+JgmTv1jqDdoVzM24jEw2Q4MpmRN4NsV3afU8w7ZvCdfrpq834K/N8wOBnIz08pU69P+0uBwsJCnnnmmS5fa2vruRHpkiVLkLJrT1GPx8Orr76a0hhOJoZTqIqBjj0jKoBFw3j8Y7n3XpUB0AFbPMpFnzzXLlQSSaXYx2bze+xK20TMFCHXPIFLnHcz03FZr757oIpzd2/OZMM/86k4lIbVpjF7UQMLltaSVxQCxqN9/Fc2rVyI9BUj0iuZfcU6Zi4Z0++P1tnFfGLmRDL64WIeiagISgjV5t3I4DM44eil5sngxGA4haqrq2S/rMiFEHcCdwKMHTu2f6MpK+vy6YzWRgLCx3bLx2yx/pMGazVWHMywX8xc1wqKLTNTuuD7m61s+jiPzZ/kEvDbyMwNc/ENpcw5qwFHh+y9Heuy2Pb8fGSiD5T0jWHb80WMsR3us5O5JrV2F/Mx3jGMzxhPmr3vypL04HM4jmbwGR58BgYGI8VwClUF0DFMKAGq+rMjKeUfgT8CLFiwoH99N8aOhdLS9l/jJnhzMvzvmTZeT/s3dKFTLKdyhftzTHde2mtTQjUuOLLXy4Z/5rFveyZSwpQZLSw4/zATp/noasZt1StjjmlWCBCLmln1ypiUhSqZwSeEYHzGeMalj8XZDxdzn09l8RkefAYGBqOJ4RSq9cAUIcQEoBK4Bbh1GI9/LD/7Gdx5J7vcQf4yF56YA7UeSI+aWCivYk7GDeTap6a0q3DQzLZPc9j4UR6NtU6c7hiLL6pm3rl1ZGT33Mrd19R1xl13z3ckokVoi7ZhMVk4Ped0itKKsFvsKY05iZSq/ikYhMJCOP981U3XyOAzMDAYLQybUEkp40KIrwFvodLTH5VS7hRCfDnx+iNCiAJgA+AFdCHEN4HpUsrBtza/7TbiUmPpji/QbNO55JCVGaZLsMz/DmZbatFIVambjR/msXNDFvGYmeIJbVx9x0Gmz2vCYk0t0EvPiuJrOl5c0rO6F7iOJrGz8mf1yyQ22UU3FlOR0xlnqFRzI4PPwMBgtDGsdVRSyteB1zs990iHn2tQU4LDguX2z/F/23J47JF6Zi6YqGwVeiEaMbFrYxYbP8yjusyD1aYxa2Ej886t61drjWVXl7PyqQnHTP9ZbRrLri4/ZjuJJBANENZCpNszWJB7JjnunD5n8MViSqB0XXXQnTHD6KJrYGAwujnlnSnOGbeUZ9PX9ypS9VVONn2cy7a1OURCFnIKg1x60xFmLWrE4ey/tVFyHWrVK2PwNdlIz4qy7Ory9ud1dFojrcT0KLmuPOZmzSXTkdnnDL5wWAmU1Qrz56s0c3fvy24GBgYGI84pL1Q9EY8Jdm/KYtPHeZQfTMNs0Zl2RhPzzq1jzKS2QZsmm7mw6bjECU1q+CN+dKlT7C1mfMZ4vPae+1R1RVubcpFwu+G882Dy5JQCRwMDA4NRgyFUXdBQ7WDzJ3lsW5tDKGAhMzfMhdeWMfusBtxp8SE9dlyP44/4QcD4jPGMTR+Lq48ZfFJCS4sSqawsWL5cOUmYzb2+1cDAYBAwGicOLoZQJYhFBXs2Z7H5k1zKDngxmXVOm9PMvCX1jJ/q7zK1fDDp2OZ9as5UitOK+5zBp+uq/ikcVi14LrwQioqMBAmDU5cXd71IbWDwGifmu/O5brrROHG4OeWFavcuwcZ3JvHS7gLCQRU9XbCijDmLhz56gsFp8x6Pq/WneFwV586eDbm5QzRgA4MTiNpALSXewcvPqvAbjRNHglNaqOJxuP4aB41NRZx+RjNnnFPH+CmtQx49tWfwxZNt3ueT48rFbOrb3Fw0qgQKlDjNmKHabRgYGIwcRuPEweeUFiqLBf781wiPvbmR6acPvQVDxwy+HGcucwr71+a9Y5PCRYvgtNOMJoUGBqMVo3HiwDmlherJ7U/y/d3/TkVmBe81qBbwM53d3930l6Nt3mMUp5UwPmN8v9q8t7aqJAmjSaGBwejFaJw4+JyyRjlPbn+SO1+9k4rWchASn17DytZfsCP05qAdI67HaQo10RJuocRbwvnjljKnYE6fRCrZpLC8HGw21aTwM59RUZQhUgYGo48LLriAcDjMww8/3P7cYDZOTK5pJRsn3nLLLYMz8FHMKRtR3fvevQRjx355YoRZFXh4wFFVTI/hi/iwCAuTsyZT4i3pc5v3jhl848bBxRcbTQoNDE4EjMaJg88pK1Rlvq7bfPj0/qeyJlPM7WYbM3JnUJRWhNXct7AnmcGnaSqDb84co0mhgUF/yXfnp5Sp15f9pYLROHFwOWWFamz6WEo/Ohve+zn4xkJ6GVz4A9Lnrurzvjq2eZ+bP5d8T76RwWdgMArorebJ4MTglBWqy8N/4+FXz4BYwvDONx5e/T8mO96EJb2/f7BMYpMefDYbLFyo1p5cfW8lZWBgYHDScsoK1euPLIFYpydjbg68eSUs2drt+3R02iJtRPWISjHP7l+KeUcPvvPPVx58tt5bUBkYGBiccpyyQtVNJ/puGxYeTTGPU+guZGLWxH6lmCe76GZlwaWXwvjxhgefgYGBQU+cskLVqRN9O50bFiZdzDWpMdY7lvEZ4/HYPX06VjLFPBBQmXtLlyoPPqOLroGBgUHvnLJClehET8fyho4NC9tdzEm4mGf03cW8Y4p5sotufr6RYm5gYGDQF05ZobrtNvXv9/9dp6JctDcsPG1BLU2hVoQQTM6azJj0MX12MU+2eY9GVXLE3LlGirmBgYFBfzllhQqUWF16ZZh/++V6xpSY8Uf8BGNWTss9jeK0YmzmvmU3xONQX6+EatYs9Ujv+zKWgYGBgUEHjFWSBJF4hFl5s1g2fhkTMib0SaRiMaiqgro6FT199rOwZIkhUgYGpyq1tbXceuutTJw4kfnz57N48WJeeuklAMxmM3PnzmXmzJnceOONx9krlZeXs2zZMqb9/+3df4xUZxXG8e/DujKU2qABEVkoG0IBQVjTZZtoY6gUBUKKVlvBalD/ICRgayJJ1arVapNWEyMxtRQtSRtbiQqNJK6KikIbpeVHqG2hNYTasG1DcQnK0khZevzj3q3T7db9dWfundnnk0yYO3OZe97s7py877xzzuzZzJkzh40bN+YxhEIZ0TMqgDFvGcO8ifOYNe2SQX9J99y5ZAbV2AgLFsDs2a5iblYk27fDiez6JjJxIlzbz3eI3Tgxe1VNVJKWABuBBuAnEXFHr+eVPr8MeBn4bEQcrHBMjCu9nYZBzC17vqQ7enQyc7rssuS+mRXLiRNJt+usdAygGpMbJ2avaolKUgNwF7AY6AD2SdoREYfLTlsKzEhvVwB3p/8WQk8fqIsugquugunTXcHczF7PjROzV83PqNqAoxFxLCJeAbYCK3qdswK4PxJ7gXGSJlUxxj6dPZu02Th/HhYtSjZhzJrlJGVm/Vu3bh3z589nwYIFAK81TmxtbWXq1KnDbpx46tQpN07M0GTgeNlxB2+cLfV1zmTgxcqG1reeMkfjxrmKhJkNjBsnZq+aM6q+Un7vWvYDOQdJayTtl7T/5MmTmQRXrqsrmUFJsHQprFyZLPM5SZlZf9w4MXvVTFQdwJSy4ybghSGcQ0RsjojWiGidMGFCZgGeOfO/BLVsGVx/fVJRwqWOzGygehon7t69m+bmZtra2li9evWgGyfu2rWLlpYWWlpaaG9vB2DVqlV0dXUxd+5c1qxZ48aJFbAPmCGpGXgeWAl8qtc5O4D1kraSLAv+KyIqvuzX3Z0kqPHjYfnyZJeQk5NZ7Zs4cWA79QbzegPhxonZqlqiiohuSeuB35FsT98SEU9JWps+vwloJ9mafpRke/rnKh1XY2NS5mjmTCcos3rT33eerDZU9XtUEdFOkozKH9tUdj+AddWMqbERFi+u5hXNzGwwPH8wM7NCc6Iys7ryZp/vWDEM5efjRGVmdaNUKtHZ2elkVVARQWdnJ6VSaVD/b8QXpTWz+tHU1ERHRweV+H6lZaNUKtE0yAKMTlRmVjcaGxtpbm7OOwzLmJf+zMys0JyozMys0JyozMys0FTru2MknQSeG+bLjAf+mUE4ReXx1b56H6PHV9uyGN+lEdFn8daaT1RZkLQ/IlrzjqNSPL7aV+9j9PhqW6XH56U/MzMrNCcqMzMrNCeqxOa8A6gwj6/21fsYPb7aVtHx+TMqMzMrNM+ozMys0JyoUpK+Lelvkg5J2inp3XnHlCVJ35P0dDrGhySNyzumLEm6TtJTkl6VVDe7qyQtkfSMpKOSvpx3PFmTtEXSS5KezDuWSpA0RdKfJB1Jfz9vyjumLEkqSXpM0uPp+L5Vket46S8h6ZKI+Hd6/0bgPRGxNuewMiPpw8CutNPynQARcXPOYWVG0mzgVeAeYENE7M85pGGT1AD8HVgMdAD7gFURcTjXwDIk6YNAF3B/RMzNO56sSZoETIqIg5LeBhwAPlovP0NJAsZGRJekRuAR4KaI2JvldTyjSvUkqdRYoK4yeETsjIju9HAvMLjyxQUXEUci4pm848hYG3A0Io5FxCvAVmBFzjFlKiL2AKfyjqNSIuLFiDiY3j8DHAEm5xtVdiLRlR42prfM3zudqMpIul3SceAG4Bt5x1NBnwd+k3cQ1q/JwPGy4w7q6E1upJE0DXgf8GjOoWRKUoOkQ8BLwO8jIvPxjahEJekPkp7s47YCICJuiYgpwAPA+nyjHbz+xpeecwvQTTLGmjKQ8dUZ9fFYXc30RwpJFwPbgC/2Wr2peRFxISJaSFZp2iRlvoQ7ovpRRcTVAzz1QeDXwK0VDCdz/Y1P0mpgObAoavDDyUH8/OpFBzCl7LgJeCGnWGyI0s9utgEPRMT2vOOplIg4LenPwBIg080xI2pG9f9ImlF2eA3wdF6xVIKkJcDNwDUR8XLe8diA7ANmSGqW9FZgJbAj55hsENLNBvcCRyLi+3nHkzVJE3p2EEsaA1xNBd47vesvJWkbMJNk59hzwNqIeD7fqLIj6SgwGuhMH9pbZ7saPwb8EJgAnAYORcRHcg0qA5KWAT8AGoAtEXF7vhFlS9LPgIUk1bdPALdGxL25BpUhSVcCDwNPkLy3AHw1Itrziyo7kuYB95H8fo4Cfh4Rt2V+HScqMzMrMi/9mZlZoTlRmZlZoTlRmZlZoTlRmZlZoTlRmZlZoTlRmZlZoTlRmdUQSe1DbdEi6ZuSNmQcklnFjagSSma1LiKW5R2DWbV5RmVWAZI+nTaUOyTpHklXpE0rS5LGpk3m5kpaKGlP2szysKRNkt7071LSPySNlzQtbcb34/S1dqYlbJA0XdJvJR2Q9LCkWdUbuVn2nKjMMpY2cfwk8IG0qvQFkvJcO4DvAN8FfhoRPYU724AvAe8FpgPXDvBSM4C7ImIOSdmoj6ePbwa+EBGXAxuAHw1zSGa58tKfWfYWAZcD+5KapIwh6dVzG0mh2f8AN5ad/1hEHIPXat9dCfxyANd5NiIOpfcPANPSdhLvB36RXhuSGo9mNcuJyix7Au6LiK+87kHpXcDFJF1QS8DZ9KneBTcHWoDzXNn9CyQJcRRwOp3JmdUFL/2ZZe+PwCckvRNA0jskXUqyJPd1kqaVd5ad35a28hhFsmT4yFAvnDble1bSdem1JWn+UF/PrAg8ozLLWEQclvQ1YGeafM4DvwK6I+JBSQ3AXyR9iKT1w1+BO0g+o9oDPDTMEG4A7k5jaAS2Ao8P8zXNcuM2H2Y5krQQ2BARy3MOxaywvPRnZmaF5hmVWQFJepQ37tb7TEQ8kUc8ZnlyojIzs0Lz0p+ZmRWaE5WZmRWaE5WZmRWaE5WZmRWaE5WZmRXafwGtjnxySmidAAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#make a regular grid to plot our Emulator predictions\n",
    "n_plot_pts = 100\n",
    "sim_pnt = 10\n",
    "test_points = np.array([[true_values['alpha_s'],x,\n",
    "                 true_values['exponent_el'],true_values['scale_inel'],true_values['scale_el']] for x in np.linspace(prior_ranges['exponent_inel'][0],prior_ranges['exponent_inel'][1],n_plot_pts)])\n",
    "\n",
    "simulation_out = simulation(test_points[np.arange(0,n_plot_pts,sim_pnt),:],*pT_range)[:,pT_obs]\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),1], simulation_out[:,0],'ro', label='sim pT_obs[0]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),1], simulation_out[:,1],'go', label='sim pT_obs[1]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),1], simulation_out[:,2], 'bo', label='sim pT_obs[2]')\n",
    "\n",
    "\n",
    "#get the GP Emulator's predictions of both the mean and std. deviation\n",
    "gp_y0, gp_dy0 = emu_predict0(test_points)\n",
    "gp_y1, gp_dy1 = emu_predict1(test_points)\n",
    "gp_y2, gp_dy2 = emu_predict2(test_points)\n",
    "\n",
    "plt.plot(test_points[:,1], gp_y0[:,0], color='red', label='pT_obs[0]')\n",
    "plt.plot(test_points[:,1], gp_y1[:,0], color='green', label='pT_obs[1]')\n",
    "plt.plot(test_points[:,1], gp_y2[:,0], color='blue', label='pT_obs[2]')\n",
    "plt.fill_between(test_points[:,1], y1 = gp_y0[:,0] - 2.*gp_dy0, y2 = gp_y0[:,0] + 2.*gp_dy0, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='red')\n",
    "plt.fill_between(test_points[:,1], y1 = gp_y1[:,0] - 2.*gp_dy1, y2 = gp_y1[:,0] + 2.*gp_dy1, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='green')\n",
    "plt.fill_between(test_points[:,1], y1 = gp_y2[:,0] - 2.*gp_dy2, y2 = gp_y2[:,0] + 2.*gp_dy2, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='blue')\n",
    "\n",
    "plt.xlabel(r'exp_inel')\n",
    "plt.ylabel(r'$R_{AA}$')\n",
    "plt.title(r'GP Emulator and Model Training Points')\n",
    "plt.legend()\n",
    "plt.tight_layout(True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 1/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: step size becomes too small\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: larger nsteps is needed\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/ipykernel_launcher.py:32: MatplotlibDeprecationWarning:Passing the pad parameter of tight_layout() positionally is deprecated since Matplotlib 3.3; the parameter will become keyword-only two minor releases later.\n"
     ]
    },
    {
     "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": [
    "#make a regular grid to plot our Emulator predictions\n",
    "n_plot_pts = 100\n",
    "sim_pnt = 10\n",
    "test_points = np.array([[true_values['alpha_s'],true_values['exponent_inel'],\n",
    "                 x,true_values['scale_inel'],true_values['scale_el']] for x in np.linspace(prior_ranges['exponent_el'][0],prior_ranges['exponent_el'][1],n_plot_pts)])\n",
    "\n",
    "simulation_out = simulation(test_points[np.arange(0,n_plot_pts,sim_pnt),:],*pT_range)[:,pT_obs]\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),2], simulation_out[:,0],'ro', label='sim pT_obs[0]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),2], simulation_out[:,1],'go', label='sim pT_obs[1]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),2], simulation_out[:,2], 'bo', label='sim pT_obs[2]')\n",
    "\n",
    "\n",
    "#get the GP Emulator's predictions of both the mean and std. deviation\n",
    "gp_y0, gp_dy0 = emu_predict0(test_points)\n",
    "gp_y1, gp_dy1 = emu_predict1(test_points)\n",
    "gp_y2, gp_dy2 = emu_predict2(test_points)\n",
    "\n",
    "plt.plot(test_points[:,2], gp_y0[:,0], color='red', label='pT_obs[0]')\n",
    "plt.plot(test_points[:,2], gp_y1[:,0], color='green', label='pT_obs[1]')\n",
    "plt.plot(test_points[:,2], gp_y2[:,0], color='blue', label='pT_obs[2]')\n",
    "plt.fill_between(test_points[:,2], y1 = gp_y0[:,0] - 2.*gp_dy0, y2 = gp_y0[:,0] + 2.*gp_dy0, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='red')\n",
    "plt.fill_between(test_points[:,2], y1 = gp_y1[:,0] - 2.*gp_dy1, y2 = gp_y1[:,0] + 2.*gp_dy1, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='green')\n",
    "plt.fill_between(test_points[:,2], y1 = gp_y2[:,0] - 2.*gp_dy2, y2 = gp_y2[:,0] + 2.*gp_dy2, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='blue')\n",
    "\n",
    "plt.xlabel(r'exp_el')\n",
    "plt.ylabel(r'$R_{AA}$')\n",
    "plt.title(r'GP Emulator and Model Training Points')\n",
    "#plt.legend()\n",
    "plt.tight_layout(True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: step size becomes too small\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 1/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: larger nsteps is needed\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/ipykernel_launcher.py:32: MatplotlibDeprecationWarning:Passing the pad parameter of tight_layout() positionally is deprecated since Matplotlib 3.3; the parameter will become keyword-only two minor releases later.\n"
     ]
    },
    {
     "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": [
    "#make a regular grid to plot our Emulator predictions\n",
    "n_plot_pts = 100\n",
    "sim_pnt = 10\n",
    "test_points = np.array([[true_values['alpha_s'],true_values['exponent_inel'],\n",
    "                 true_values['exponent_el'],x,true_values['scale_el']] for x in np.linspace(prior_ranges['scale_inel'][0],prior_ranges['scale_inel'][1],n_plot_pts)])\n",
    "\n",
    "simulation_out = simulation(test_points[np.arange(0,n_plot_pts,sim_pnt),:],*pT_range)[:,pT_obs]\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),3], simulation_out[:,0],'ro', label='sim pT_obs[0]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),3], simulation_out[:,1],'go', label='sim pT_obs[1]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),3], simulation_out[:,2], 'bo', label='sim pT_obs[2]')\n",
    "\n",
    "\n",
    "#get the GP Emulator's predictions of both the mean and std. deviation\n",
    "gp_y0, gp_dy0 = emu_predict0(test_points)\n",
    "gp_y1, gp_dy1 = emu_predict1(test_points)\n",
    "gp_y2, gp_dy2 = emu_predict2(test_points)\n",
    "\n",
    "plt.plot(test_points[:,3], gp_y0[:,0], color='red', label='pT_obs[0]')\n",
    "plt.plot(test_points[:,3], gp_y1[:,0], color='green', label='pT_obs[1]')\n",
    "plt.plot(test_points[:,3], gp_y2[:,0], color='blue', label='pT_obs[2]')\n",
    "plt.fill_between(test_points[:,3], y1 = gp_y0[:,0] - 2.*gp_dy0, y2 = gp_y0[:,0] + 2.*gp_dy0, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='red')\n",
    "plt.fill_between(test_points[:,3], y1 = gp_y1[:,0] - 2.*gp_dy1, y2 = gp_y1[:,0] + 2.*gp_dy1, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='green')\n",
    "plt.fill_between(test_points[:,3], y1 = gp_y2[:,0] - 2.*gp_dy2, y2 = gp_y2[:,0] + 2.*gp_dy2, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='blue')\n",
    "\n",
    "plt.xlabel(r'scale_inel')\n",
    "plt.ylabel(r'$R_{AA}$')\n",
    "plt.title(r'GP Emulator and Model Training Points')\n",
    "#plt.legend()\n",
    "plt.tight_layout(True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: step size becomes too small\n",
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/scipy/integrate/_ode.py:1180: UserWarning:dopri5: larger nsteps is needed\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 1/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:78: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n",
      " /Users/dananjayaliyanage/git/parton_energy_loss_qcd_bjorken/solver_rk.py:73: IntegrationWarning:The integral is probably divergent, or slowly convergent.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working on 2/10 design\n",
      "Working on 3/10 design\n",
      "Working on 4/10 design\n",
      "Working on 5/10 design\n",
      "Working on 6/10 design\n",
      "Working on 7/10 design\n",
      "Working on 8/10 design\n",
      "Working on 9/10 design\n",
      "Working on 10/10 design\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " /Users/dananjayaliyanage/miniconda3/envs/parton_loss/lib/python3.6/site-packages/ipykernel_launcher.py:32: MatplotlibDeprecationWarning:Passing the pad parameter of tight_layout() positionally is deprecated since Matplotlib 3.3; the parameter will become keyword-only two minor releases later.\n"
     ]
    },
    {
     "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": [
    "#make a regular grid to plot our Emulator predictions\n",
    "n_plot_pts = 100\n",
    "sim_pnt = 10\n",
    "test_points = np.array([[true_values['alpha_s'],true_values['exponent_inel'],\n",
    "                 true_values['exponent_el'],true_values['scale_inel'],x] for x in np.linspace(prior_ranges['scale_el'][0],prior_ranges['scale_el'][1],n_plot_pts)])\n",
    "\n",
    "simulation_out = simulation(test_points[np.arange(0,n_plot_pts,sim_pnt),:],*pT_range)[:,pT_obs]\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),4], simulation_out[:,0],'ro', label='sim pT_obs[0]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),4], simulation_out[:,1],'go', label='sim pT_obs[1]')\n",
    "plt.plot(test_points[np.arange(0,n_plot_pts,sim_pnt),4], simulation_out[:,2], 'bo', label='sim pT_obs[2]')\n",
    "\n",
    "\n",
    "#get the GP Emulator's predictions of both the mean and std. deviation\n",
    "gp_y0, gp_dy0 = emu_predict0(test_points)\n",
    "gp_y1, gp_dy1 = emu_predict1(test_points)\n",
    "gp_y2, gp_dy2 = emu_predict2(test_points)\n",
    "\n",
    "plt.plot(test_points[:,4], gp_y0[:,0], color='red', label='pT_obs[0]')\n",
    "plt.plot(test_points[:,4], gp_y1[:,0], color='green', label='pT_obs[1]')\n",
    "plt.plot(test_points[:,4], gp_y2[:,0], color='blue', label='pT_obs[2]')\n",
    "plt.fill_between(test_points[:,4], y1 = gp_y0[:,0] - 2.*gp_dy0, y2 = gp_y0[:,0] + 2.*gp_dy0, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='red')\n",
    "plt.fill_between(test_points[:,4], y1 = gp_y1[:,0] - 2.*gp_dy1, y2 = gp_y1[:,0] + 2.*gp_dy1, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='green')\n",
    "plt.fill_between(test_points[:,4], y1 = gp_y2[:,0] - 2.*gp_dy2, y2 = gp_y2[:,0] + 2.*gp_dy2, \n",
    "                 interpolate=True, alpha=0.3, label=r'GP 2$\\sigma$', color='blue')\n",
    "\n",
    "plt.xlabel(r'scale_el')\n",
    "plt.ylabel(r'$R_{AA}$')\n",
    "plt.title(r'GP Emulator and Model Training Points')\n",
    "#plt.legend()\n",
    "plt.tight_layout(True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEHCAYAAABMRSrcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAew0lEQVR4nO3df7SdVX3n8feHNEigdbI6pFUCMRmb4jAFwbmEuuJqNYqgtQWss4B2ZtrpzLCYLga1I8s4Ouq042ooXaNdlSlNla7aqmgF06wBGzpGS4syzY1EKcHYFBdDwsyACChjgCR8549zLh5uzr33POc8+3n2Ps/ntVbWuudHzt3nnPs8372/+7v3o4jAzMy67bi2G2BmZu1zMDAzMwcDMzNzMDAzMxwMzMwM+IG2GzCOk08+OdauXdt2M8zMirJ79+5vRcSqYY8VGQzWrl3L7Oxs280wMyuKpAcWesxpIjMzczAwMzMHAzMzw8HAzMxwMDAzMxoIBpIulLRP0n5Jm4c8/mpJT0ja0//33tRtsvFtu/sgG7fsZN3mW9m4ZSfb7j7YdpPMrAZJS0slLQOuB84HDgC7JG2PiL3znvpXEfGmlG2xyW27+yDvuuUeDh0+CsDBxw/xrlvuAeDic1a32TRr2La7D3Ldjn089PghTlm5gmsuON1/A4VLPTLYAOyPiPsj4hngJuCixL/TErlux77nAsGcQ4ePct2OfS21yNow1yk4+Pghgu93CjxKLFvqRWergQcHbh8AzhvyvFdK+irwEPCOiLh3/hMkXQFcAbBmzZoETR1Nl3tEDz1+qNL9Np0W6xRM27HQpeM9dTDQkPvmX03nK8BLIuJJSW8EtgHrj/lPEVuBrQAzMzOtXJGn62mSU1au4OCQE/8pK1e00BprS52dgpxPtl073lOniQ4Apw3cPpVe7/85EfGdiHiy//NtwHJJJydu11hySZO0NYl7zQWns2L5sufdt2L5Mq654PRGfr/lYaHgX7VTkHu6KZfjvSmpg8EuYL2kdZKOBy4Dtg8+QdKLJKn/84Z+mx5N3K6x5JAmafMAuvic1fzmm89k9coVCFi9cgW/+eYzp7KXZAurq1OQ+8k2h+O9SUnTRBFxRNJVwA5gGXBjRNwr6cr+4zcAbwH+naQjwCHgssj0wsw5pEnaztdefM5qn/wLVkdaZu75k75O7ifbHI73JiXftbSf+rlt3n03DPz8YeDDqdtRh2suOP15OURoPk2S+wFk+aozB15HpyD3k20Ox3uTitzCui119YgmkfsBZPkaZ1SZcoI395PtQsc7wMYtOxs5BzQ5we5gUFHbaZLcDyDLV9VRZepqmhw6V0uZf7w3WWHUdDWTg0FhSjiARpFzSeG0qjqqbGJ+qu3OVVVNztk1PT/oYFCg0g6g+bpWv52LqqNKz08dq8nPpOnP37uWWuNyLymcVlVLg+taTzBNmvxMmv78PTKYgFMd43GPsz1VRpWenzpWk59J05+/g8GY6k51dCmwVMldd+lzyc20zE/VqcnPpOnPX5mu71rUzMxMzM7OttqGjVt2Dj2hrV65gjs3b6r0WvMDC/R6ANO6unfU99u1z8UsNUm7I2Jm2GOeMxhTnamOruXQR81dd+1zMWuT00RjqnPxVxdz6KPkrrv4uZi1xSODMdW5g6erNoar8rn4cpxmk3EwGFOdO3h6a+jhRv1cct8K2awEThNNoK7FX67aGG7Uz6WtnVxd6WQLKfFvw8EgE6WvKk4l17mFSUuLSzxZ2GhKXWHvNJGNLZc8fRtzLpNUOjmtNd1KrYJzMLCx5HRCa2POZZLRSKknCxtNqVVwDgY2lpxOaG1cjnOS0UipJ4sStTF6LbU60HMGNpbcTmhNz7lMsm+ML1DUjLZy96Xu6eSRgY2l1N5PXSYZjbiUuBltjV7bGKnWwSMDG0upvZ86jTsacSlxM4aNvha7v04lVgc6GCQ0zeWDPqFNpsSTRWmWSRwdshHnMqmF1uTPwSCRUmuNq/AJzXI2LBAsdn/Xec4gkZyqbcy6ZtvdB1mo/7+6I/NaVTkYJJJbtY1Zl1y3Yx/D+v+CTs1rVeFgkEjXq23M2rRQpyuYnjRt3RwMEnH5oFl7Fup0OUW0ME8gJ1Jatc1g5dPKE5cTAU8cOpx9u6fRNFehNcWlz9U5GCRUSrXN/Mqnx753+LnHprEKKmddqEJrQmmdsRw4GNjQyqdBTVwbwHraujbDNCqlM5YLBwMbqcLJVVDNcBVae7qenvMEso1U4eQqqGa4Cq0dOW3J3hYHg8Kk2JJ3WOXTIE+8NcdVaO3wIlGniYqSanJx/mSbq4na44nPduSanmsydeVgUJCUk4uebGvf/AP/g5ee7e+kITleY6LpyjKniQqSa+/FJuecdbtyTM81nbpKHgwkXShpn6T9kjYv8rxzJR2V9JbUbSqVJxenl3PW7crxgjRNd/6SpokkLQOuB84HDgC7JG2PiL1DnnctsCNle0rnVZXTy6O+9g2mSudSdm//1J7W5m2aTl2lHhlsAPZHxP0R8QxwE3DRkOf9e+Bm4OHE7Slajr0Xq4dHffnIJWXXdOoq9QTyauDBgdsHgPMGnyBpNXAJsAk4d6EXknQFcAXAmjVram9oKTzRO5086stHLqvAm64sSx0Mhl1fYv424x8C3hkRR7XI5egiYiuwFWBmZsaXKrKp4pLSfOSUsmuy85c6GBwAThu4fSrw0LznzAA39QPBycAbJR2JiG2J22aWFY/68pBjmWkTUs8Z7ALWS1on6XjgMmD74BMiYl1ErI2ItcBngF91IDCztuRYZtqEpCODiDgi6Sp6VULLgBsj4l5JV/YfvyHl77c8dH0DMCtLV1N2iigv/T4zMxOzs7NtN8NGMH8VJfR6Wa6CMmuepN0RMTPsMa9AtqS8mMqsDA4GllROlRlmtjBvVGdJdbUyw0bj+aR8eGRgSXW1MsOWlstK3zqluN5IUxwMLClvoWELmbb5pNKDm9NElpwXU9kw0zaflMs2FuPyyMDMWjFtm/OVHtwcDBIqOX9oltq0zSeVHtwcDBIpPX9oltok80k5drRKD26eM0ik9PyhVeMSyfGMM5/U9LWBR1X6NhadCQZNH6yl5w9tdLmenKZVzh2tkoslOpEmaiNlU3r+0EY3bSWSuXNHK41OBIM2DtbS84c2Op+cmuWOVhqdCAZtHKxebNUdPjk1yx2tNDoxZ9DW/jgl5w9tdL5+cbNKn6jNVSeCgQ9WS8knp+a5o1W/TgSDhQ5WgI1bdvoAton55GSl60QwgGMPVpcD2ji8nsCmVScmkIdxOaBV5VXlNs06MzKYz+WAVlXOi53m8wimLDl8X50NBr4Cl1VVSgfCKdCy5PJ9dTZN5Fplq6qU9QROgZYll++rs8HAi8Lak+OOk6MopQNRygjGenL5vjqbJgKXA7YhlyHxOEpZT+AUaFly+b46HQyseSVNwg4zaQeiiYlCL7IsSy7fl4NBXw6z+V2Qy5C4DU2NikoZwVhPLt+XgwFlpy6qyCHg5TIkbkOToyKnQMuSw/fV2QnkQbnM5qeUy4KpUiZhU+jyqMjy52BANw7SXAJel6u4SilNtW5ymohupC5yCng5DInbkMtEodkwHhnQjdSFe6Xt6/KoyPLnkQH5zOan5F5pHro6KrL8ORj0TftB2oWAZ2bjGzkYSHopcCAinpb0auAs4GMR8Xiaplndpj3gmZWs7dLvKnMGNwNHJf0Y8FFgHfCJJK0yM+uQHEq/qwSDZyPiCHAJ8KGIeDvw4qX+k6QLJe2TtF/S5iGPXyTpa5L2SJqV9KoKbTIzK14Opd9V5gwOS7oc+CXgZ/v3LV/sP0haBlwPnA8cAHZJ2h4Rewee9nlge0SEpLOATwMvq9AuM7Oi5VD6XWVk8K+AVwIfiIhvSloH/MkS/2cDsD8i7o+IZ4CbgIsGnxART0ZE9G+eBARmZh2SQ+n3yMEgIvZGxNUR8cn+7W9GxJa5xyXdPOS/rQYeHLh9oH/f80i6RNLXgVuBXxn2+yVd0U8jzT7yyCOjNtvMLHs5rHWqc9HZPxpyn4bcd0zPPyI+GxEvAy4GfmPYi0fE1oiYiYiZVatWTdRQM7Oc5LAgsc51BsPSOweA0wZunwo8tOALRNwh6aWSTo6Ib9XYtk5ouzTNzMbXdul36kVnu4D1/fmFg8BlwC8MPqFfqvr3/QnkVwDHA48mbtfUGXUbbgcMMxumzmBwTEooIo5IugrYASwDboyIeyVd2X/8BuDngX8p6TBwCLh0YELZRjTKXvlduW6DpeGOxHSrMxi8c9idEXEbcNu8+24Y+Pla4Noa29FJo5SmlX7JSWuPOxLTb+QJZEnrJX1G0l5J98/9m3s8Im5P08Sybbv7IBu37GTd5lvZuGVnshWFo5Sm5VDLbGXKYVGUpVWlmugPgd8DjgCvAT4G/HGKRk2LJpeYj1KalkMts+WjSkfFHYnpVyUYrIiIzwOKiAci4v3ApjTNmg5N9qZGKU3LoZY5B02N1nJWtaPijsT0qzJn8JSk44C/608KHwR+JE2zpkPTvamlStO8jbVz33Oqzh/5ehjTr0oweBtwInA1vYVhm+jtU2QLWHnich773uGh97el7VrmtnkSvadqR8Udiek3cjCIiF39H5+kt0+RLWGhAlkXzrbHue+eca773fWOxLSrUk3045L+QNLtknbO/UvZuNI9cejYUcFi91t6zn33eP7I5quSJvpT4AbgD4CjSzzXGK/3ZWk5991TYtrHi97SqhIMjkTE7yVryRQa5cTjP/BmlXgSTKWktI8n/tPTUjs/SPrh/o9XAw8DnwWenns8Ir6drHULmJmZidnZ2aZ/7VgWO9nP/wOHXrBoerdCs9xt3LJz6Ch79coV3Lm5GxXudXQcJe2OiJlhj40yMthNb0fSub2Hrhl4LBi+dbX1Ldb7cmWL2Wi6PvHfxMhoyWAQEetq+U12jK7/gZuNquvzb010HKtUE50g6dck3SLpZklvk3RCLa3oKFe2mI2m69VPTXQcq2xH8THgnwC/C3wYOAPvTTSRrv+Bm40qhyuBtamJjmOVaqLTI+LlA7e/IOmrtbWkg1zZYja6kqqf6jA4YbzyxOUsP04cfvb7BT91dxyrBIO7Jf1kRNwFIOk84M7aWtJRXfsDN7OlzZ8wfux7h1m+TKxcsZwnDh1O0nGsEgzOo3dFsv/Vv70GuE/SPUBExFm1tcrMrMOGTRgfPhqc9IIfYM/7Xp/kd1YJBhcmaYEl54VtZmVpo9JwyWAwsOjsu8Meb2PRmY1u2lduThLoHCQtV22U0o6z6GxuBkN40Vn2pnlh2ySBbtqDpE2uzc5CG3toVVp01h8lrAe8vqAQ07ywbZJAN81B0ibXdmehjUrDkecMJP0b4K3AqcAe4CeBLwGvTdIyq0Vdw80cUyqTBLocg2SOn3FX5dBZaLrSsMqis7cC5wIPRMRrgHOAbyVpldWmjoVtVa+X25RJFuLktvo718+4q3LsLKRWJRg8FRFPAUh6QUR8HfBS2czVsXJzsV5SmyYJdLmt/s71M+6q3DoLTahSWnpA0kpgG/AXkh4DHkrRKKvXpMPNXHtJk+RVc1v9netn3FVdvAhSlWsgX9L/8f2SvgD8A+DPk7TKspLzjpGTBLqcVn/n/Bl3UW6dhSZUGRk8JyL+su6GWL662Etqmj/j/OTUWWjCWMHAJlNa1UgXe0lN82dsbVvyspc5Kumyl/P5Updm1pZJL3tpNcqhftmsBKWNoEvnYNAwV42YLa3tFcBdVGWdgdWgi/XLZlV53UXzHAwalttiJ7Nxbbv7IBu37GTd5lvZuGVnraulPYJuntNEDXPViE2D1GmcHNZddG3OwsGgBV2rX7bpk7oQou11F12cs0ieJpJ0oaR9kvZL2jzk8V+U9LX+vy9JennqNpnZZFKncerYU2sSXZyzSDoykLQMuB44HzgA7JK0PSL2Djztm8BPR8Rjkt4AbKV3veXGdG04aDapJtI4bY6guzhnkXpksAHYHxH3R8QzwE3ARYNPiIgvRcRj/Zt30bteQmO8dbBZdSUWQlSZ8O5i1V/qYLAaeHDg9oH+fQv518Dnhj0g6QpJs5JmH3nkkdoa2MXhoNmk2k7jVFW101disJtU6glkDblv6P4Xkl5DLxi8atjjEbGVXgqJmZmZ2vbQ6OJw0KwOJRVCVJ3w7mLVX+pgcAA4beD2qQy5BoKks4CPAG+IiEcTt+l5cihhM7O0xun0lRTs6pA6TbQLWC9pnaTjgcuA7YNPkLQGuAX4FxHxjcTtOUYXh4NmXdPFOYCqkgaDiDgCXAXsAO4DPh0R90q6UtKV/ae9F/iHwH+TtEdSo9uRlpb7NLPq3OlbmrewNrNOcAm5t7A2M+vcHEBVDgbWKvfWzPLgYGCt6eL+L2a58hbW1hov+DPLh4OBtcYL/szy4WBgrXHtt1k+HAysNa79NltcyqvJzecJZGtNF/d/sXaVVL3WdIGFg4G1yrXf1pTSqtdSX01uPgcDM6tNzj3vpk+uk2q6wMLBwIqV84mni3LveZdWvdb0jsqeQLYi+Qp1+Wly3cg4E6ulVa81XWDhkYEtKOeed2lD/rY18V021fMedwRyzQWnP+//Qd7Va00XWDgY2FAe8k+PUb7LOoJFU2mNcTsCJVavNVlg4WBgQ+Xe8/YV6ka31HdZV+Bvquc9SUfA1WsL85xBYk0uGqlT7j1vL1gb3VLfZV25/qYuFFVa7r8UHhkklHuqZTG597znD/lXnricCHj7p/Zw3Y592Q//m7TUd1ln4G+i511a7r8UHhkkVPKunCX0vC8+ZzV3bt7EBy89m6cOP8vjhw67smiIpb7L0nravlRtGh4ZJJR7qmUxJU225T6/0balvssSe9rO/dfPwSCh3FMtSynlgCs56DZlse+ypMBv6TgYJFRij2uYnNcbQPlBNwelBH5Lx3MGCU1DbrOElb4lzG+Y5c4jg8RK73GVkI93msNscg4GtqhS8vGlB12ztjlNZIsqrezQzMbjYGCLcj7erBlt71bQ+TRR7pUybXM+3kbh42gyOexWoIho5BfVaWZmJmZnZyd+nflfAPR6vaVV/Ji1ycfR5DZu2Tm0PHr1yhXcuXlTbb9H0u6ImBn2WKfTRKm2i2h7uGfWpJK3XclFDoUanQ4GKb6AEuryzeqUw4msdDkUanQ6GKT4AtxLsq7J4URWuhwKNTodDFJ8Ae4lWdfkcCLL3VKp4xx2K+h0NVGKShnvk2Nd44qzxY1aKdT2wslOBwM49g95Lp0z7pcyLZvTmVXR9oksZyVs6QIOBrXX97qXZGaDSkkdJw8Gki4EfgdYBnwkIrbMe/xlwB8CrwDeHRG/nbpNg1JEbfeSbJAXZHVbKanjpBPIkpYB1wNvAM4ALpd0xrynfRu4Gmg0CMwpJWpbmVxqbKVMsKeuJtoA7I+I+yPiGeAm4KLBJ0TEwxGxCzicuC1DuSzOUnKpseVQKTSK1Gmi1cCDA7cPAOeN80KSrgCuAFizZs3kLevzhK8NU1dqxyNPgzJSx6lHBhpy31ibIUXE1oiYiYiZVatWTdis7yslaltz6kzteORppUg9MjgAnDZw+1TgocS/s7ISorY1p86iAo88bVxNFx6kDga7gPWS1gEHgcuAX0j8O80qmX/QDav8gPFSOy41tnG0saV10mAQEUckXQXsoFdaemNE3Cvpyv7jN0h6ETALvBB4VtLbgDMi4jsp22YGww86MTyXOW5qxyNPq6qNhWrJ1xlExG3AbfPuu2Hg5/9DL31k1rhhB13AMQHBqR1rUhuFB53eqM5soYMrwEUF1po2Cg86vx2FVTNtq2kXmiOo+wpTOZu273QatFF44JGBjayN1bSprxpXyurQVLxCOg/z/86BxkveO30NZKumqeu0zmnq2rpd7hk3/Z3asZq8hvRi10B2mshG1vSkVlMVFV2u9vEK6eYs1OnIZYtrBwMbWdO7L/pElV4pO2qWbrF1A7n8nXvOwEbWdH49dUVF6vmIEnR9zqQpi/X+c9myxMHARtb0Pk4pT1SeOO3x3lzNWKz3n0tAdprIKmkyv55yK4dc8rQ56PKcSVMWS8flsmWJg4FlLdWJKpc8rXXDUusGBv/O5yaa3/6pPY0GBgcD6yRPnFqTRu39t7FB3RwHgxGVUIueuo0lfAaj8tbS1rRRRrltpi87FwzGOaG1Ga1HlbqNJXwGVeSSp50W09RRaFOb6ctOBYNxT2glTDambmMJn0FVnjitR9XjyoFjYW2mLztVWjruxclLmGxM3cYSPgNrR5XjyiW9i2uzzLRTwWDcE1oui0IWk7qNJXwG1o4qx9W4HbKuaHPdR6eCwbgntFwWhSwmdRtL+AysHVWOK48wl3bxOau5c/MmvrnlZ7hz86bGUmidCgbjntBKWKWZuo0lfAbWjirHlUeY+ercFtaevDKr36jHVZPbNduxFtvCunPBwMza5Q5Ze3w9AzPLhkt689SpOQMzMxvOwcDMzBwMzMzMwcDMzHAwMDMzCi0tlfQI8MCY//1k4Fs1NqdNfi95mpb3Mi3vA/xe5rwkIlYNe6DIYDAJSbML1dmWxu8lT9PyXqblfYDfyyicJjIzMwcDMzPrZjDY2nYDauT3kqdpeS/T8j7A72VJnZszMDOzY3VxZGBmZvM4GJiZWfeCgaTfkPQ1SXsk3S7plLbbNC5J10n6ev/9fFbSyrbbNC5J/0zSvZKelVRkCaCkCyXtk7Rf0ua22zMuSTdKeljS37bdlklJOk3SFyTd1//7emvbbRqXpBMk/Y2kr/bfy3+u9fW7Nmcg6YUR8Z3+z1cDZ0TElS03ayySXg/sjIgjkq4FiIh3ttyssUj6x8CzwO8D74iIoi5YIWkZ8A3gfOAAsAu4PCL2ttqwMUj6KeBJ4GMR8RNtt2cSkl4MvDgiviLph4DdwMWFfi8CToqIJyUtB/4aeGtE3FXH63duZDAXCPpOAoqNhhFxe0Qc6d+8Czi1zfZMIiLui4iSr4q+AdgfEfdHxDPATcBFLbdpLBFxB/DttttRh4j43xHxlf7P3wXuA4q8mEL0PNm/ubz/r7bzV+eCAYCkD0h6EPhF4L1tt6cmvwJ8ru1GdNhq4MGB2wco9KQzrSStBc4B/mfLTRmbpGWS9gAPA38REbW9l6kMBpL+h6S/HfLvIoCIeHdEnAZ8HLiq3dYubqn30n/Ou4Ej9N5PtkZ5LwXTkPuKHXVOG0k/CNwMvG1edqAoEXE0Is6mlwXYIKm2NN5UXvYyIl434lM/AdwKvC9hcyay1HuR9EvAm4DXRuYTQBW+lxIdAE4buH0q8FBLbbEB/fz6zcDHI+KWtttTh4h4XNIXgQuBWib6p3JksBhJ6wdu/hzw9bbaMilJFwLvBH4uIr7Xdns6bhewXtI6SccDlwHbW25T5/UnXT8K3BcR/7Xt9kxC0qq5ikFJK4DXUeP5q4vVRDcDp9OrXHkAuDIiDrbbqvFI2g+8AHi0f9ddBVdGXQL8LrAKeBzYExEXtNqoiiS9EfgQsAy4MSI+0G6LxiPpk8Cr6W2V/H+B90XER1tt1JgkvQr4K+Aeesc8wH+MiNvaa9V4JJ0F/BG9v6/jgE9HxK/X9vpdCwZmZnaszqWJzMzsWA4GZmbmYGBmZg4GZmaGg4GZmeFgYGZmOBiYNUrSSkm/usRzTpH0mQl+xxdL3Qbc2uNgYNaslcCiwSAiHoqItzTTHLMeBwObGpL+ef/iH3sk/b6k8/oX/jlB0kn9C4L8hKRXS7qjf0GgvZJukHRc/zUul3RPfwO9awde+8n+brdflXSXpB/t379K0s2SdvX/bezf//7+RWK+KOn+/rUzALYAL+238boF3sfauQvLSPplSbdI+nNJfyfptwae93pJX5b0FUl/2t+MzWwsDgY2FfoXx7kU2Njf1fEovW1HtgP/Bfgt4E8iYm5Trw3AfwDOBF4KvFm9q95dC2wCzgbOlXRx//kn0dvu4+XAHcC/7d//O8AHI+Jc4OeBjww062XABf3f9b7+hmmbgb+PiLMj4poR397Z/fd2JnCpelfvOhl4D/C6iHgFMAv82oivZ3aMqdy11DrptcA/BXb19iZjBb0933+d3iZyTwFXDzz/byLifnhuL55XAYeBL0bEI/37Pw78FLANeAb47/3/u5veFc2gt1nYGf3fCfDC/hW1AG6NiKeBpyU9DPzomO/t8xHxRL9Ne4GX0Es3nQHc2f/dxwNfHvP1zRwMbGoI+KOIeNfz7pReBPwgvatCnQD8v/5D8zflCoZfk2DO4YEtwo/y/WPnOOCVEXFo3u8FeHrgrsH/U9Ww1xG9i5tcPuZrmj2P00Q2LT4PvEXSjwBI+mFJLwG2Av+J3oV/rh14/ob+dtPH0UvB/DW9K2D9tKST1bum8eXAXy7xe29n4AJJks5e4vnfBX5oieeM4i5go6Qf6//eEyX9eA2vax3lkYFNhYjYK+k9wO39E/xh4M+AIxHxif7J/UuSNtHbyvjL9CZzz6Q3B/DZiHhW0ruAL9Dred8WEX+2xK++Grhe0tfoHU93AAtuIx4Rj0q6sz9B/LkK8wbzX+cRSb8MfFLSC/p3vwf4xjivZ+YtrK1zJL0aeEdEvKnlpphlw2kiMzPzyMCsLZLOBP543t1PR8R5bbTHus3BwMzMnCYyMzMHAzMzw8HAzMxwMDAzM+D/AwJYGa1/K5mPAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "design = gp_array[0].X\n",
    "for i in range(0,5):\n",
    "    for k in range(0,i):\n",
    "        fig, ax = plt.subplots()\n",
    "        ax.scatter(design[:,i],design[:,k])\n",
    "        ax.set_xlabel(f'{list(prior_ranges.keys())[i]}')\n",
    "        ax.set_ylabel(f'{list(prior_ranges.keys())[k]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the loglikelihood function\n",
    "def logLikelihood(theta, x, obs, obserr):\n",
    "  \n",
    "    # Model parameters\n",
    "    theta = np.array(theta)\n",
    "    \n",
    "    #alpha_s, mD_factor, exponent_inel, exponent_el = theta\n",
    "    alpha_s, exp_inel, exp_el, sc_inel, sc_el = theta\n",
    "    design = np.array([alpha_s, exp_inel, exp_el, sc_inel, sc_el]).reshape(1,-1)\n",
    "    \n",
    "    # Model predictions given parameters\n",
    "    \n",
    "    model = np.array([emu_predict0(design),emu_predict1(design),emu_predict2(design)],dtype=object)\n",
    "    modelerr = model[:,1]\n",
    "    model = model[:,0]\n",
    "    \n",
    "    # Likelihood of data given model parameters\n",
    "\n",
    "    term2 = 0.5*(2*np.pi*np.sqrt(np.sum(obserr**2+modelerr.flatten()**2)))\n",
    "    return -0.5*np.sum((obs.flatten()-model.flatten())**2/(obserr**2+modelerr.flatten()**2)) - term2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1.0042884042986624"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logLikelihood(truth.flatten(),x,obs,obserr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the logprior function\n",
    "def logPrior(theta):\n",
    "    \n",
    "    # Model parameters\n",
    "    theta = np.array(theta)\n",
    "    alpha_s, exp_inel, exp_el, sc_inel, sc_el = theta \n",
    "    \n",
    "    # Probability of model parameters: flat prior\n",
    "    if prior_ranges['alpha_s'][0] < alpha_s < prior_ranges['alpha_s'][1] \\\n",
    "        and prior_ranges['exponent_inel'][0] < exp_inel < prior_ranges['exponent_inel'][1] \\\n",
    "        and prior_ranges['exponent_el'][0] < exp_el < prior_ranges['exponent_el'][1]\\\n",
    "        and prior_ranges['scale_inel'][0] < sc_inel < prior_ranges['scale_inel'][1] \\\n",
    "        and prior_ranges['scale_el'][0] < sc_el < prior_ranges['scale_el'][1]:\n",
    "        return 0.0\n",
    "    return -np.inf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define logprobability function: l(D|theta) * p(theta)\n",
    "# Note: use this for emcee, not approxposterior!\n",
    "def logposterior(theta, x, obs, obserr):\n",
    "        \n",
    "    lp = logPrior(theta)\n",
    "    if not np.isfinite(lp):\n",
    "        return -np.inf\n",
    "    return lp + logLikelihood(theta, x, obs, obserr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running burn-in phase\n",
      "Running MCMC chains\n",
      "Done MCMC\n",
      "Mean acceptance fraction: 0.301 (in total 250000 steps)\n"
     ]
    }
   ],
   "source": [
    "ntemps = 25\n",
    "Tmax = 1e3\n",
    "\n",
    "nwalkers = 50 #guidance is 10*d\n",
    "ndim = 5\n",
    "nburnin = 2000\n",
    "niterations= 5000\n",
    "nthin = 1#0 # only record every nthin iteration\n",
    "\n",
    "nthreads = 4\n",
    "\n",
    "min_theta = np.array([prior_ranges['alpha_s'][0],prior_ranges['exponent_inel'][0],prior_ranges['exponent_el'][0],prior_ranges['scale_inel'][0],prior_ranges['scale_el'][0]])\n",
    "max_theta = np.array([prior_ranges['alpha_s'][1],prior_ranges['exponent_inel'][1],prior_ranges['exponent_el'][1],prior_ranges['scale_inel'][1],prior_ranges['scale_el'][1]])\n",
    "\n",
    "# Data and uncertainties that we use to condition our model\n",
    "args = (x, obs, obserr) # recall x = np.array(pT_obs)\n",
    "\n",
    "ptsampler=ptemcee.Sampler(nwalkers, ndim, logLikelihood, logPrior, ntemps, \n",
    "                  threads=nthreads, Tmax=Tmax, loglargs=args)\n",
    "\n",
    "# Making the sampling multi-threaded is as simple as adding the threads=Nthreads \n",
    "# argument to PTSampler. \n",
    "\n",
    "# First, we run the sampler for 1000 burn-in iterations:\n",
    "pos0 = np.random.uniform(min_theta, max_theta, (ntemps, nwalkers, ndim))\n",
    "\n",
    "print(\"Running burn-in phase\")\n",
    "for p, lnprob, lnlike in ptsampler.sample(pos0, iterations=nburnin,adapt=True):\n",
    "    pass\n",
    "ptsampler.reset()\n",
    "\n",
    "print(\"Running MCMC chains\")\n",
    "#Now we sample for nwalkers*niterations, recording every nthin-th sample:\n",
    "for p, lnprob, lnlike in ptsampler.sample(p, iterations=niterations, thin=nthin,adapt=True):\n",
    "    pass \n",
    "\n",
    "print('Done MCMC')\n",
    "\n",
    "mean_acc_frac = np.mean(ptsampler.acceptance_fraction)\n",
    "print(f\"Mean acceptance fraction: {mean_acc_frac:.3f}\",\n",
    "      f\"(in total {nwalkers*niterations} steps)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "chain = ptsampler.chain[0, :, :, :].reshape((-1, ndim))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 849.6x849.6 with 25 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = corner.corner(chain, quantiles=[0.16, 0.5, 0.84], truths=truth.T,\n",
    "                    labels=[r\"$\\alpha_s$\", 'exponent_inel', 'exponent_el','scale_inel','scale_el'], show_titles=True, scale_hist=True,\n",
    "                    plot_contours=True, range=(prior_ranges.values()));\n",
    "plt.tight_layout()\n",
    "plt.savefig('corner-lhsgp-ptemcee.png',dpi=600)\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}