Samplers.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' \\nInstallation: pip install --upgrade pyDOE\\nhttps://pythonhosted.org/pyDOE/randomized.html\\n\\nimport sys\\n!{sys.executable} -m pip install --upgrade pyDOE\\n\\n'"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Implementation of 3 experimental designs:\n",
    "## 1. Latin hypercube \n",
    "## 2. Grid \n",
    "## 3. Uniform design\n",
    "\n",
    "''' \n",
    "Installation: pip install --upgrade pyDOE\n",
    "https://pythonhosted.org/pyDOE/randomized.html\n",
    "\n",
    "import sys\n",
    "!{sys.executable} -m pip install --upgrade pyDOE\n",
    "\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting pyDOE\n",
      "  Downloading https://files.pythonhosted.org/packages/bc/ac/91fe4c039e2744466621343d3b8af4a485193ed0aab53af5b1db03be0989/pyDOE-0.3.8.zip\n",
      "Requirement not upgraded as not directly required: numpy in /anaconda3/envs/tf_gpu/lib/python3.5/site-packages (from pyDOE) (1.15.2)\n",
      "Requirement not upgraded as not directly required: scipy in /anaconda3/envs/tf_gpu/lib/python3.5/site-packages (from pyDOE) (1.1.0)\n",
      "Building wheels for collected packages: pyDOE\n",
      "  Running setup.py bdist_wheel for pyDOE ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /Users/nramachandra/Library/Caches/pip/wheels/7c/c8/58/a6493bd415e8ba5735082b5e0c096d7c1f2933077a8ce34544\n",
      "Successfully built pyDOE\n",
      "\u001b[31mmkl-random 1.0.1 requires cython, which is not installed.\u001b[0m\n",
      "\u001b[31mtensorflow 1.10.0 has requirement numpy<=1.14.5,>=1.13.3, but you'll have numpy 1.15.2 which is incompatible.\u001b[0m\n",
      "\u001b[31mtensorflow 1.10.0 has requirement setuptools<=39.1.0, but you'll have setuptools 40.2.0 which is incompatible.\u001b[0m\n",
      "Installing collected packages: pyDOE\n",
      "Successfully installed pyDOE-0.3.8\n"
     ]
    }
   ],
   "source": [
    "# Python imports \n",
    "\n",
    "import sys\n",
    "!{sys.executable} -m pip install --upgrade pyDOE\n",
    "\n",
=======
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Python imports \n",
    "\n",
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "import pyDOE as pyDOE"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 10,
=======
   "execution_count": 3,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [],
   "source": [
    "# Input parameters \n",
    "\n",
    "num_para = [3, 4, 5, 6, 7][0]\n",
    "grid_evals = np.array([2, 3, 4, 5, 6, 7, 8])[6]\n",
    "num_evals = grid_evals**num_para  ## Total number of evaluations for training the emulator (based on grid)\n",
    "# num_evals = np.array([10, 100, 1000, 10000])[1]  ## Total number of evaluations (applicable for lhc and random)\n",
    "design_all = ['latin', 'random', 'grid']"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 11,
=======
   "execution_count": 4,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [],
   "source": [
    "# WithRange = False\n",
    "\n",
    "def paramsWithRange():\n",
    "    # Parameters defined with ranges\n",
    "\n",
    "    #### replace this and the limits #####\n",
    "    ####  para = np.linspace(lower_lim, upper_lim, total_eval)\n",
    "        \n",
    "    gridpara1 = np.linspace(0.0, 0.62036, grid_evals)  #OmegaM\n",
    "    gridpara2 = np.linspace(0.1841, 0.4159, grid_evals) #Omegab\n",
    "    gridpara3 = np.linspace(1.2886, 2.9114, grid_evals) # sigma8\n",
    "\n",
    "    allGrid = np.vstack([gridpara1, gridpara2, gridpara3])\n",
    "\n",
    "    para1 = np.linspace(0.0, 0.62036, num_evals)  #OmegaM\n",
    "    para2 = np.linspace(0.1841, 0.4159, num_evals) #Omegab\n",
    "    para3 = np.linspace(1.2886, 2.9114, num_evals) # sigma8\n",
    "\n",
    "    allPara = np.vstack([para1, para2, para3])\n",
    "    \n",
    "    return allGrid, allPara\n",
    "\n",
    "# def Rescaledparams(num_para, grid_evals, num_evals):\n",
    "def Rescaledparams(num_para, num_evals):\n",
    "\n",
    "    ### Samples between 0-1 and rescaling will be taken care of later\n",
    "    \n",
    "    allPara = np.zeros(shape=(num_para, num_evals))\n",
    "#     allGrid = np.zeros(shape=(num_para, grid_evals))\n",
    "\n",
    "    for ii in range(num_para):\n",
    "        allPara[ii, :] = np.linspace(0, 1, num_evals)\n",
    "#         allGrid[ii, :] = np.linspace(0, 1, grid_evals) \n",
    "        \n",
    "    return allPara#, allGrid\n"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 12,
=======
   "execution_count": 5,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [],
   "source": [
    "def gridDesign(allGridParams, paramOutDir):\n",
    "    ## NEEDS TO BE CHANGED BASED ON num_para\n",
    "    print(grid_evals)\n",
    "    print(allGridParams.shape)\n",
    "    \n",
    "#     num_para = allGridParams.shape[0]\n",
    "    \n",
    "#     if (allGridParams.shape[1]**num_para != (num_evals) ):\n",
    "#         print('grid ', allGridParams.shape[1])\n",
    "#         print('num_params ', num_para)\n",
    "#         print('num_evals ', num_evals)\n",
    "#         print('Number of grid evaluations need to be k^num_para')\n",
    "#         return None\n",
    "    \n",
    "    if True:\n",
    "#         num_para = allGridParams.shape[0]\n",
    "#         num_evals = allGridParams.shape[1]\n",
    "\n",
    "        grd = np.array(np.meshgrid(allGridParams[0], allGridParams[1], allGridParams[2]) )\n",
    "\n",
    "        allCombinations = np.vstack( [np.ravel(grd[0]), np.ravel(grd[1]), np.ravel(grd[2])] ).T\n",
    "        \n",
    "        Outfile = 'grid_para' + str(allCombinations.shape[1]) + '_total' + str(allCombinations.shape[0])\n",
    "        paramOutfile = paramOutDir + Outfile + '.txt'\n",
    "\n",
    "        print(paramOutfile)\n",
    "        np.savetxt(paramOutfile, allCombinations)\n",
    "        return allCombinations     "
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 13,
=======
   "execution_count": 6,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [],
   "source": [
    "def latinHyp(allParams, paramOutDir):\n",
    "    np.random.seed(421)\n",
    "    num_para = allParams.shape[0]\n",
    "    num_evals = allParams.shape[1]\n",
    "    \n",
    "    lhd = pyDOE.lhs(num_para, samples=num_evals, criterion=None) # c cm corr m\n",
    "    # lhd = norm(loc=0, scale=1).ppf(lhd)  # this applies to both factors here\n",
    "    idx = (lhd * num_evals).astype(int)\n",
    "\n",
    "    allCombinations = np.zeros((num_evals, num_para))\n",
    "    for i in range(num_para):\n",
    "        allCombinations[:, i] = allParams[i][idx[:, i]]\n",
    "    \n",
    "    Outfile =  'latin_para' + str(num_para) + '_total' + str(num_evals)\n",
    "    paramOutfile = paramOutDir + Outfile + '.txt'\n",
    "    \n",
    "    \n",
    "    print(paramOutfile)\n",
    "    np.savetxt(paramOutfile, allCombinations)\n",
    "    return allCombinations, lhd"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 14,
=======
   "execution_count": 7,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [],
   "source": [
    "def randomDesign(allParams, paramOutDir):\n",
    "    ## CHECK if it's uniform in every direction\n",
    "    np.random.seed(313)\n",
    "    num_para = allParams.shape[0]\n",
    "    num_evals = allParams.shape[1]\n",
    "    \n",
    "    rnd = np.random.uniform(0, 1, (num_evals, num_para) )\n",
    "    \n",
    "    idx = (rnd * num_evals).astype(int)\n",
    "    \n",
    "    allCombinations = np.zeros((num_evals, num_para))\n",
    "    for i in range(num_para):\n",
    "        allCombinations[:, i] = allParams[i][idx[:, i]]\n",
    "        \n",
    "    Outfile =  'random_para' + str(num_para) + '_total' + str(num_evals)\n",
    "    paramOutfile = paramOutDir + Outfile + '.txt'\n",
    "\n",
    "    \n",
    "    print(paramOutfile)\n",
    "    np.savetxt(paramOutfile, allCombinations)\n",
    "    return allCombinations, rnd"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 15,
=======
   "execution_count": 8,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Designs/latin_para5_total100.txt\n"
     ]
    },
    {
     "data": {
<<<<<<< HEAD
      "image/png": "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\n",
=======
      "image/png": "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\n",
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
      "text/plain": [
       "<Figure size 432x288 with 15 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Testing lhc design\n",
    "\n",
    "design = design_all[0]   \n",
    "\n",
    "allPara = Rescaledparams(num_para=5, num_evals= 100)\n",
    "lhdParams, lhd = latinHyp(allPara,  'Designs/')\n",
    "        \n",
    "    \n",
    "##\n",
    "f, a = plt.subplots(allPara.shape[0], allPara.shape[0], sharex=True, sharey=True)\n",
    "plt.subplots_adjust(left=None, bottom=None, right=None, top=None, wspace=None, hspace=None)\n",
    "plt.rcParams.update({'font.size': 8})\n",
    "plt.suptitle(design + ' design')\n",
    "for i in range(allPara.shape[0]):\n",
    "    for j in range(i+1):\n",
    "#         print(i,j)\n",
    "        if(i>j):\n",
    "            a[i, j].scatter(lhd[:, i], lhd[:, j], s=5)\n",
    "            a[i, j].grid(True)\n",
    "            \n",
    "        if(j<i):\n",
    "            plt.delaxes(a[j, i])\n",
    "        else:\n",
    "            # a[i,i].set_title(AllLabels[i])\n",
    "            # a[i, i].text(0.4, 0.4, AllLabels[i], size = 'xx-large')\n",
    "            hist, bin_edges = np.histogram(lhd[:,i], density=True, bins=8)\n",
    "            # a[i,i].bar(hist)\n",
    "            a[i,i].bar(bin_edges[:-1], hist/hist.max(), width=0.2)\n",
    "            plt.xlim(0,1)\n",
    "            plt.ylim(0,1)\n",
    "\n",
    "Outfile =  'latin_para' + str(num_para) + '_total' + str(num_evals)\n",
    "plt.savefig('Plots/' + Outfile + '.png', figsize=(10, 10))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 16,
=======
   "execution_count": 9,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Designs/random_para5_total10.txt\n"
     ]
    },
    {
     "data": {
<<<<<<< HEAD
      "image/png": "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\n",
=======
      "image/png": "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\n",
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
      "text/plain": [
       "<Figure size 432x288 with 15 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### testing uniform random design\n",
    "\n",
    "design = design_all[1]\n",
    "        \n",
    "# Outfile = design + '_para' + str(num_para) + '_total' + str(num_evals)\n",
    "allPara = Rescaledparams(num_para=5, num_evals= 10)\n",
    "rndParams, rnd = randomDesign(allPara,  'Designs/')\n",
    "\n",
    "##\n",
    "f, a = plt.subplots(allPara.shape[0], allPara.shape[0], sharex=True, sharey=True)\n",
    "plt.subplots_adjust(left=None, bottom=None, right=None, top=None, wspace=None, hspace=None)\n",
    "plt.rcParams.update({'font.size': 8})\n",
    "plt.suptitle(design + ' design')\n",
    "for i in range(allPara.shape[0]):\n",
    "    for j in range(i+1):\n",
    "#         print(i,j)\n",
    "        if(i>j):\n",
    "            a[i, j].scatter(rnd[:, i], rnd[:, j], s=5)\n",
    "            a[i, j].grid(True)\n",
    "        if(j<i):\n",
    "            plt.delaxes(a[j, i])\n",
    "        else:\n",
    "            # a[i,i].set_title(AllLabels[i])\n",
    "            # a[i, i].text(0.4, 0.4, AllLabels[i], size = 'xx-large')\n",
    "            hist, bin_edges = np.histogram(rnd[:,i], density=True, bins=8)\n",
    "            # a[i,i].bar(hist)\n",
    "            a[i,i].bar(bin_edges[:-1], hist/hist.max(), width=0.2)\n",
    "            plt.xlim(0,1)\n",
    "            plt.ylim(0,1)\n",
    "\n",
    "            \n",
    "\n",
    "Outfile =  'random_para' + str(num_para) + '_total' + str(num_evals)\n",
    "plt.savefig('Plots/' + Outfile + '.png', figsize=(10, 10))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 17,
=======
   "execution_count": 10,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8\n",
      "(3, 9)\n",
      "Designs/grid_para3_total729.txt\n"
     ]
    },
    {
     "data": {
<<<<<<< HEAD
      "image/png": "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\n",
=======
      "image/png": "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\n",
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### testing grid design\n",
    "\n",
    "design = design_all[2]\n",
    "\n",
    "# num_para = 5\n",
    "# num_evals = 27\n",
    "# grid_evals = 4\n",
    "\n",
    "\n",
    "allGrid = Rescaledparams(num_para=3, num_evals= 9)\n",
    "gridParams = gridDesign(allGrid,  'Designs/')\n",
    "\n",
    "##\n",
    "f, a = plt.subplots(allGrid.shape[0], allGrid.shape[0])\n",
    "plt.subplots_adjust(left=None, bottom=None, right=None, top=None, wspace=None, hspace=None)\n",
    "plt.rcParams.update({'font.size': 8})\n",
    "plt.suptitle(design + ' design')\n",
    "for i in range(allGrid.shape[0]):\n",
    "    for j in range(i+1):\n",
    "#         print(i,j)\n",
    "        if(i>j):\n",
    "            a[i, j].scatter(gridParams[:, i], gridParams[:, j], s=5)\n",
    "            a[i, j].grid(True)\n",
    "        if(j<i):\n",
    "            plt.delaxes(a[j, i])\n",
    "        else:\n",
    "            # a[i,i].set_title(AllLabels[i])\n",
    "            # a[i, i].text(0.4, 0.4, AllLabels[i], size = 'xx-large')\n",
    "            hist, bin_edges = np.histogram(gridParams[:, i], density=True, bins=8)\n",
    "            # a[i,i].bar(hist)\n",
    "            a[i,i].bar(bin_edges[:-1], hist/hist.max(), width=0.2)\n",
    "\n",
    "## Plot looks like 9 points, but they're are coincided on the grids\n",
    "\n",
    "Outfile = 'grid_para' + str(num_para) + '_total' + str(num_evals)\n",
    "plt.savefig('Plots/' + Outfile + '.png', figsize=(10, 10))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
<<<<<<< HEAD
   "execution_count": 18,
=======
   "execution_count": 11,
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4\n",
      "(5, 4)\n",
      "Designs/grid_para3_total64.txt\n"
     ]
    }
   ],
   "source": [
    "### testing grid design\n",
    "\n",
    "design = design_all[2]\n",
    "\n",
    "num_para = 5\n",
    "# num_evals = 27\n",
    "grid_evals = 4\n",
    "\n",
    "\n",
    "allGrid = Rescaledparams(num_para=num_para, num_evals= grid_evals)\n",
    "gridParams = gridDesign(allGrid,  'Designs/')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_evals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 5)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lhdParams.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, 5)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rndParams.shape"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
<<<<<<< HEAD
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
=======
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
<<<<<<< HEAD
   "version": "3.5.6"
=======
   "version": "3.5.2"
>>>>>>> 7d15d4438faa8c87839a01d1569cb38e1472271c
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}