tomography.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:31.913583Z",
     "start_time": "2018-09-21T15:06:30.304538Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fri Sep 21 11:06:31 2018\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.mlab as mlab\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from matplotlib import cm\n",
    "from scipy.optimize import curve_fit\n",
    "import scipy.signal as spy\n",
    "import glob\n",
    "import time\n",
    "import seaborn as sns\n",
    "sns.set()\n",
    "sns.set_style(\"whitegrid\")\n",
    "import imageio\n",
    "import moviepy.editor as mpy\n",
    "print(time.ctime())\n",
    "#%matplotlib notebook\n",
    "%matplotlib inline\n",
    "#import mpld3\n",
    "\n",
    "#mpld3.enable_notebook()\n",
    "\n",
    "#This module takes all unsorted files (1296) in a single directory & sorts them, then calculates 81 S values\n",
    "# Once all those S values are saved, it outputs them in a graphic\n",
    "# You'll need that bash file, or your own method to download all the .csv files\n",
    "# Commenting does exist, but if something doesn't make sense, check the module it is referencing\n",
    "# I tried to institute Hungarian Notation, where the first letter of each variable relates to the datatype \n",
    "# it should be. (l_ means list, a_ means array, s_ means string) \n",
    "# This notation isn't used in a lot of the modules because they were written earlier in the summers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:31.941643Z",
     "start_time": "2018-09-21T15:06:31.926903Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "#functions\n",
    "def linear(x,a,b):\n",
    "    return x*a + b\n",
    "def exp_decay(x,a,b,c):\n",
    "    return a * np.exp(-b*x) + c\n",
    "def root(x,a,b):\n",
    "    return a *np.sqrt(x) + b\n",
    "def fitting(function,xdata,ydata,  guess,sigma = 0):\n",
    "    if type(sigma) is int:\n",
    "        fit_opt, fit_cov = curve_fit(function,xdata,ydata, p0=guess, maxfev = 2600)\n",
    "    else:\n",
    "        fit_opt, fit_cov = curve_fit(function,xdata,ydata, p0=guess, sigma = sigma, absolute_sigma = True)\n",
    "    return function(xdata,*fit_opt), fit_opt, fit_cov \n",
    "def gaussian(x, amp, cen, wid,const):\n",
    "    return amp * np.exp(-(x-cen)**2 / wid)+const\n",
    "def polygauss(x,amp,cen,wid,am2,cen2,wid2,const):\n",
    "    return gaussian(x, amp, cen, wid,const) + gaussian(x,am2, cen2, wid2,const)\n",
    "def polyexpo(x,a,b,c,d):\n",
    "    return exp_decay(x,a,b) + exp_decay(x,c,d)\n",
    "def expogauss(x,a,b,c,d, amp,cen,wid):\n",
    "     return polyexpo(x,a,b,c,d) + gaussian(x, amp, cen, wid)\n",
    "def sin(x,a,b,c,d):\n",
    "    value = a*np.sin((x*(2*np.pi/b)+(180/np.pi)*c))+d\n",
    "    return value\n",
    "def chi_square_red(x, y, yerr, modelfunc, p_opt):\n",
    "    numerator = (y - modelfunc(x, *p_opt)) ** 2\n",
    "    denominator = yerr ** 2\n",
    "    chi = sum(numerator / denominator)\n",
    "    chired = chi / (len(y) - len(p_opt))\n",
    "    return chired"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:31.961461Z",
     "start_time": "2018-09-21T15:06:31.952836Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def data_gather(filelist,directory):\n",
    "        delta = []                                                  # generates new empty lists\n",
    "        coi = []\n",
    "        fiber = []\n",
    "        for i in range(len(filelist)):\n",
    "            s_file = directory+filelist[i]\n",
    "            data = np.loadtxt(s_file, dtype= float, delimiter=\",\", usecols = (0,1,2))   # loads data\n",
    "            deltafill = data[:, 0]                                                          # places data in separate arrays\n",
    "            coifill = data[:, 1]\n",
    "            fiberfill = data[:,2]\n",
    "            # places arrays into a Python list\n",
    "            delta.append(deltafill)\n",
    "            coi.append(coifill)\n",
    "            fiber.append(fiberfill)\n",
    "        return delta,coi,fiber"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:31.990757Z",
     "start_time": "2018-09-21T15:06:31.973380Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def GaussianFit(delta,coi, fiber,name,pol,init = (2500,0,40,2500,0,60,150),init3 =  (1000,0,50,800,0,80,1500)):\n",
    "    param3 = init3\n",
    "    cutd, cutc, cutf = fiberselect(delta,coi,fiber,1)\n",
    "    newd, newc,newf = fiberselect(delta,coi,fiber,2)\n",
    "    if np.max(coi) < 50:\n",
    "        guess = (250,0,40,15)\n",
    "        print(pol, 'SINGLE GAUSS!')\n",
    "        fit = fitting(gaussian,delta, coi,guess =guess)\n",
    "    else:\n",
    "        fit = fitting(polygauss,delta, coi,guess =init)\n",
    "    #fit2 = fitting(polygauss,cutd, cutf,guess = init3)\n",
    "    #fit3 = fitting(polygauss,newd,newf, guess = init3 )\n",
    "    cutd = np.sort(cutd)\n",
    "    newd = np.sort(newd)   \n",
    "    # calculating integral and error propogation\n",
    "    param = fit[1]\n",
    "    #param2 = fit2[1]\n",
    "    #param3 = fit3[1]\n",
    "       # integrals\n",
    "    if np.max(coi) < 50:\n",
    "        total = param[0]*np.sqrt(np.pi*(param[2]))\n",
    "\n",
    "    else:\n",
    "        total = param[0]*np.sqrt(np.pi*(param[2])) + param[3]*np.sqrt(np.pi*param[5]) \n",
    "    #total2 =(param2[0]*np.sqrt(np.pi*(param2[2])) + param2[3]*np.sqrt(np.pi*param2[5])) /2 \n",
    "    #total3 = (param3[0]*np.sqrt(np.pi*(param3[2])) + param3[3]*np.sqrt(np.pi*param3[5]))/2 \n",
    "    # errors from covariance matrix\n",
    "    perr = np.sqrt(np.diag(fit[2]))\n",
    "    #perr2 = np.sqrt(np.diag(fit2[2]))\n",
    "    #perr3 = np.sqrt(np.diag(fit3[2]))\n",
    "    # error propogation\n",
    "    if np.max(coi) < 50:\n",
    "        se = total*np.sqrt(perr[0]/param[0]**2)+ param[3]           \n",
    "    else:\n",
    "        se =total*np.sqrt((perr[0]/param[0])**2+(perr[2]/(2*param[2])**2))\n",
    "    #se2 = total2*np.sqrt((perr2[0]/param2[0])**2+(perr2[2]/(2*param2[2])**2))\n",
    "    #se3 = total3*np.sqrt((perr3[0]/param3[0])**2+(perr3[2]/(2*param3[2])**2))\n",
    "    #if np.isinf(se3) == True:\n",
    "        #print(pol)\n",
    "        #print(param3)\n",
    "        #print(perr3)\n",
    "    # Plotting functionality\n",
    "    fig,ax = plt.subplots(figsize = (10,10))\n",
    "    #plt.scatter(delta,fiber, marker = \"+\",color = 'lightcoral' ,alpha = 0.5)\n",
    "    plt.scatter(delta, coi,marker = \"+\", color = 'cadetblue', )\n",
    "    plt.plot(delta, fit[0], color = 'm', label = 'Coincidences' )\n",
    "    #plt.plot(cutd[:np.int(len(cutd)/2)], fit2[0][:np.int(len(cutd)/2)],color = 'gold', label = 'Coincidences - Same Fiber' )\n",
    "    #plt.plot(newd[np.int(len(newd)/2+1):], fit3[0][np.int(len(newd)/2+1):],color = 'orange', label = 'Coincidences - Same Fiber2')\n",
    "    plt.title('Coincidences ' + pol)\n",
    "    #plt.ylim(0,3150)\n",
    "    plt.xlim(-50,50)\n",
    "    plt.xlabel('Delta T (ns)')\n",
    "    plt.ylabel('Coincidences')\n",
    "    plt.legend()\n",
    "    textstr = '$Coincidences=%.2f$'% (total)\n",
    "    props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)\n",
    "    ax.text(0.05, 0.95, textstr, transform=ax.transAxes, fontsize=14, verticalalignment='top', bbox=props)\n",
    "    #plt.savefig(name,  bbox_inches='tight')\n",
    "    return total, se, param, param3\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:32.010048Z",
     "start_time": "2018-09-21T15:06:32.002066Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def fiberselect(d, c, f,fiber):\n",
    "    #this is used in the GaussianFit module. It is how I fit the two separate fibers (+/- 0)\n",
    "    if fiber == 1:\n",
    "        index = np.where(d <= 0)\n",
    "        newindex = np.where(d < 0)\n",
    "        delta_part = d[index]\n",
    "        delta =np.append(delta_part, -1*d[newindex] )\n",
    "    if fiber == 2:\n",
    "        index = np.where(d >= 0)\n",
    "        newindex = np.where(d > 0)\n",
    "        delta_part = -1*d[index]\n",
    "        delta =np.append(delta_part, d[newindex] )\n",
    "    coi_part   = c[index]\n",
    "    fiber_part = f[index]\n",
    "    coi   =np.append(coi_part, c[newindex])\n",
    "    fiber = np.append(fiber_part,f[newindex])\n",
    "    return delta, coi, fiber"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:32.034273Z",
     "start_time": "2018-09-21T15:06:32.022903Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def fixer(parameters):\n",
    "    # this just switches signs of things so I have positive amplitudes.\n",
    "    #  It also plots just to check that the parameters I have make sense\n",
    "    x = np.linspace(60,300,60)\n",
    "    plt.figure(figsize =(20,10))\n",
    "    plt.subplot(121)\n",
    "    for i in range(4):\n",
    "        plt.plot(x,sin(x,*parameters[i*4:(i*4+4)]))\n",
    "    plt.show\n",
    "    print(parameters)\n",
    "    for i in range(0,4):\n",
    "        if parameters[i*4] < 0:\n",
    "            print(parameters[i*4])\n",
    "            parameters[i*4] = np.absolute(parameters[i*4])\n",
    "            parameters[4*i+2] = -parameters[((4*i)+2)]  \n",
    "            parameters[4*i+1] = -parameters[((4*i)+1)]   \n",
    "    print(parameters)\n",
    "    plt.subplot(122)\n",
    "    for i in range(4):\n",
    "        plt.plot(x,sin(x,*parameters[i*4:(i*4+4)]))\n",
    "    plt.show()                                              \n",
    "    return(parameters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:32.130848Z",
     "start_time": "2018-09-21T15:06:32.119314Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def Sval(a_total):\n",
    "    # Bell Inequality s values\n",
    "    E = np.zeros(4)\n",
    "    index = np.array([10,6,8,4,11,7,9,5,14,2,12,0,15,3,13,1])\n",
    "    # I wish I could explain the index better. Once you sort the names of the files 'alphabetically'\n",
    "    # then this sorts them into an order where each group of four represents an \"E\" to be calculated\n",
    "    # Which ones belong to E are determined by orthogonality & described in the booklet for the photon source\n",
    "    a_total = np.array(a_total)\n",
    "    C = a_total[index]\n",
    "    for i in range(len(C)):\n",
    "        if np.isnan(C[i]) == True:\n",
    "            print(i, 'C FUCKED UP') # pardon my french\n",
    "        if np.isinf(C[i]) == True:\n",
    "            print(i, 'C FUCKED UP INF')\n",
    "    for i in range(4):\n",
    "        num = C[4*i]-C[4*i+1]-C[4*i+2]+C[4*i+3]\n",
    "        den = (C[4*i]+C[4*i+1]+C[4*i+2]+C[4*i+3])\n",
    "        E[i] = num/den\n",
    "        #print(den)\n",
    "        #print(i)\n",
    "    #print(E, \"E\")\n",
    "    S = E[0]+E[1]-E[2]+E[3]\n",
    "    print(S)\n",
    "    max = 2*np.sqrt(2)\n",
    "    #print(max)\n",
    "    return S,C,E"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:32.467426Z",
     "start_time": "2018-09-21T15:06:32.447484Z"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def errorprop(C,E,errors):\n",
    "# Error Propogation for each C\n",
    "    coi_err = np.zeros(16)\n",
    "    E_err = np.zeros(4)\n",
    "    x = np.array([22.5,67.5,112.5,157.5])\n",
    "    index = np.array([10,6,8,4,11,7,9,5,14,2,12,0,15,3,13,1]) \n",
    "    errors = np.array(errors)\n",
    "    coi_err = errors[index]\n",
    "    print(np.sum(coi_err), 'NAN SUM TEST')\n",
    "    if np.isnan(np.sum(coi_err)) == True:\n",
    "        print('Coincidence fucked up')\n",
    "    for i in range(4):\n",
    "        num = C[4*i]-C[4*i+1]-C[4*i+2]+C[4*i+3]\n",
    "        den = (C[4*i]+C[4*i+1]+C[4*i+2]+C[4*i+3])\n",
    "        num_err = np.sqrt(coi_err[4*i]**2+coi_err[4*i+1]**2+coi_err[4*i+2]**2+coi_err[4*i+3]**2)\n",
    "        den_err = np.sqrt(coi_err[4*i]**2+coi_err[4*i+1]**2+coi_err[4*i+2]**2+coi_err[4*i+3]**2)\n",
    "        if np.isnan(num_err) == True:\n",
    "            print( 'NUM FUCKED UP ')\n",
    "        if np.isnan(den_err) == True:\n",
    "            print( 'DEN FUCKED UP ')\n",
    "        E_err[i] = E[i]*np.sqrt((num_err/num)**2+(den_err/den)**2)\n",
    "        if np.isnan(E_err[i]) == True:\n",
    "            print( 'E FUCKED UP ')\n",
    "    #Error Propogation in S\n",
    "    print(E,E_err, 'E & error')\n",
    "    S_err = np.sqrt((E_err[0]**2)+(E_err[1]**2)+(E_err[2]**2)+(E_err[3]**2))\n",
    "    if np.isnan(S_err) == True:\n",
    "        print('S FUCKED UP')\n",
    "    print(S_err)\n",
    "    return(S_err)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:32.764840Z",
     "start_time": "2018-09-21T15:06:32.757284Z"
    }
   },
   "outputs": [],
   "source": [
    "def CoincidenceCount(delta,coi,fiber,pol):\n",
    "    index = np.where((delta < -100) | (delta > 100))\n",
    "    cutd, cutc, cutf = fiberselect(delta,coi,fiber,1)\n",
    "    newd, newc,newf = fiberselect(delta,coi,fiber,2)\n",
    "    sameindex1 = np.where((cutd < -100) | (cutd > 100))\n",
    "    sameindex2 = np.where((newd < -100) | (newd > 100))\n",
    "    bkgavg = np.mean(coi[index])\n",
    "    oneavg = np.mean(cutf[sameindex1])\n",
    "    twoavg = np.mean(newf[sameindex2])\n",
    "    total = np.sum(coi)-bkgavg*len(coi)\n",
    "    #total = np.sum(coi-bkgavg)\n",
    "    total1 = np.sum(cutf)-oneavg*len(coi)\n",
    "    total2 = np.sum(newf)-twoavg*len(coi)\n",
    "    '''\n",
    "    dexter = np.where((delta > -50) & (delta < 50))\n",
    "    dexter1 = np.where((cutd > -100) & (cutd < 100))\n",
    "    dexter2 = np.where((newd > -100) & (newd < 100))\n",
    "    othertot = np.sum(coi[dexter])\n",
    "    othertot1 = np.sum(cutf[dexter1])\n",
    "    othertot2 = np.sum(newf[dexter2])\n",
    "    '''\n",
    "    #print(total, pol)\n",
    "    return total, total1, total2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:33.287461Z",
     "start_time": "2018-09-21T15:06:33.149999Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1312\n"
     ]
    }
   ],
   "source": [
    "directory = \"/home/andrew/Documents/BNL 2018/Data/Scalc/\"\n",
    "l_file=os.listdir(directory)\n",
    "l_filesort = []\n",
    "l_label = []\n",
    "print(len(l_file))\n",
    "for i in range (3):\n",
    "    for j in range(3):\n",
    "        for k in range (3):\n",
    "            for l in range (3):\n",
    "                    str_i = str(i)\n",
    "                    str_j = str(j)\n",
    "                    str_k = str(k)\n",
    "                    str_l = str(l)\n",
    "                    s_labelfill = str_i+'x'+str_j+'_'+str_k+'x'+str_l\n",
    "                    l_label.append(s_labelfill)\n",
    "                    for m in range(len(l_file)):   \n",
    "                        if l_file[m].endswith(str_i+'x'+str_j+'_'+str_k+'x'+str_l+'.csv'):\n",
    "                            l_filesort.append(l_file[m])\n",
    "                        \n",
    "# This cell is rather messy, but it gives me labels and sorts all the file names \n",
    "# if you dumped all 1296 files into a single directory. I'll also supply the bash script that does that. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:40.036777Z",
     "start_time": "2018-09-21T15:06:35.340618Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.6970553501437102\n",
      "270.3525543286619 NAN SUM TEST\n",
      "[ 0.68521279  0.62289733 -0.63092891  0.75801632] [ 0.02577051  0.0348274  -0.04044299  0.02830271] E & error\n",
      "0.06567911627697287\n",
      "2.684001678060849\n",
      "542.8277940895467 NAN SUM TEST\n",
      "[ 0.67053826  0.655802   -0.6498186   0.70784282] [ 0.01617655  0.01928557 -0.01705086  0.01273411] E & error\n",
      "0.03296215189631693\n",
      "2.5801973793980366\n",
      "297.2143471462944 NAN SUM TEST\n",
      "[ 0.65523386  0.61636981 -0.64917959  0.65941413] [ 0.02410604  0.02817502 -0.03649305  0.03020625] E & error\n",
      "0.06015889917387968\n",
      "2.7501563861754637\n",
      "491.80934569039374 NAN SUM TEST\n",
      "[ 0.69848984  0.64638659 -0.66137348  0.74390647] [ 0.0174734   0.02376063 -0.0191153   0.00962106] E & error\n",
      "0.036439633854976984\n",
      "2.6958782330532243\n",
      "1043.6612666698452 NAN SUM TEST\n",
      "[ 0.68315006  0.66572172 -0.62869809  0.71830837] [ 0.00703948  0.01007606 -0.00979909  0.00696569] E & error\n",
      "0.01719373143802775\n",
      "2.6801393859053473\n",
      "597.8883863773505 NAN SUM TEST\n",
      "[ 0.663218    0.65738296 -0.64104582  0.7184926 ] [ 0.0145678   0.01156793 -0.01747653  0.0161171 ] E & error\n",
      "0.030186546262197144\n",
      "2.690612891149485\n",
      "257.2349982735415 NAN SUM TEST\n",
      "[ 0.7478334   0.7143615  -0.57356375  0.65485424] [ 0.03869917  0.04569007 -0.02803405  0.02705532] E & error\n",
      "0.07143603086284048\n",
      "2.680608480428527\n",
      "532.0194190466715 NAN SUM TEST\n",
      "[ 0.69014059  0.63500034 -0.63630733  0.71916023] [ 0.01331894  0.01794538 -0.0194254   0.01607301] E & error\n",
      "0.0336915192457887\n",
      "2.7760223317988473\n",
      "310.902127262917 NAN SUM TEST\n",
      "[ 0.68356343  0.74418202 -0.64822802  0.70004887] [ 0.02223018  0.02968554 -0.02928902  0.03386784] E & error\n",
      "0.058140257662649554\n",
      "2.725430616137718\n",
      "439.1055763772702 NAN SUM TEST\n",
      "[ 0.70434526  0.65480517 -0.64356531  0.72271487] [ 0.02634299  0.01504561 -0.01667403  0.02253458] E & error\n",
      "0.04130561803870654\n",
      "2.6887265329452967\n",
      "871.7937084672371 NAN SUM TEST\n",
      "[ 0.69084761  0.64474121 -0.63160986  0.72152786] [ 0.010254    0.01123859 -0.01031481  0.00892063] E & error\n",
      "0.020430939472468178\n",
      "2.7370025435699086\n",
      "473.82244321957876 NAN SUM TEST\n",
      "[ 0.72244825  0.64532234 -0.62241905  0.7468129 ] [ 0.01970305  0.01894055 -0.01975512  0.01679339] E & error\n",
      "0.037672769668638056\n",
      "2.720578078162677\n",
      "792.0079477128783 NAN SUM TEST\n",
      "[ 0.68862976  0.67420351 -0.63140512  0.72633969] [ 0.01127177  0.01314061 -0.00957838  0.01113944] E & error\n",
      "0.022705967450288207\n",
      "2.714672574299486\n",
      "1674.9360985987298 NAN SUM TEST\n",
      "[ 0.68391246  0.6721437  -0.62655654  0.73205988] [ 0.0053445   0.00522594 -0.00615755  0.00450923] E & error\n",
      "0.010682824372866124\n",
      "2.702303156590759\n",
      "962.0026950689426 NAN SUM TEST\n",
      "[ 0.68949994  0.66806499 -0.62338534  0.72135288] [ 0.00531409  0.00950774 -0.00999436  0.01174912] E & error\n",
      "0.01888294776969429\n",
      "2.7600664403223076\n",
      "401.03312222289117 NAN SUM TEST\n",
      "[ 0.70418137  0.69758701 -0.63055792  0.72774013] [ 0.02705778  0.02530299 -0.01702763  0.01924518] E & error\n",
      "0.0450852728951832\n",
      "2.71730558170897\n",
      "841.3002009755294 NAN SUM TEST\n",
      "[ 0.71817972  0.66075463 -0.6159398   0.72243143] [ 0.01052479  0.01072676 -0.0060872   0.01419365] E & error\n",
      "0.021548741004103054\n",
      "2.6859416567606376\n",
      "501.7739053308756 NAN SUM TEST\n",
      "[ 0.68995894  0.647728   -0.60525102  0.7430037 ] [ 0.02083286  0.01006086 -0.02212392  0.01569468] E & error\n",
      "0.03565136329244552\n",
      "2.5350015118975477\n",
      "221.70343507937528 NAN SUM TEST\n",
      "[ 0.69613223  0.61190899 -0.57059052  0.65636977] [ 0.03373138  0.03852624 -0.0385374   0.04920216] E & error\n",
      "0.08079642112735631\n",
      "2.659893096345999\n",
      "434.0002396486875 NAN SUM TEST\n",
      "[ 0.67506941  0.65315358 -0.60627364  0.72539646] [ 0.02099812  0.02092247 -0.01999376  0.02005416] E & error\n",
      "0.04099499944072932\n",
      "2.659283706094974\n",
      "237.35356309816814 NAN SUM TEST\n",
      "[ 0.75342218  0.66733756 -0.54420896  0.694315  ] [ 0.0489776   0.03758587 -0.03925579  0.0210569 ] E & error\n",
      "0.07613089566695527\n",
      "2.6576942004637045\n",
      "393.3576158287833 NAN SUM TEST\n",
      "[ 0.6655219   0.6651949  -0.62516596  0.70181144] [ 0.0122346   0.02511173 -0.02764015  0.02464501] E & error\n",
      "0.046385762444514346\n",
      "2.6835371384172397\n",
      "832.8480949015485 NAN SUM TEST\n",
      "[ 0.68377478  0.6718381  -0.59545782  0.73246643] [ 0.00568966  0.01026339 -0.01250637  0.01380271] E & error\n",
      "0.022014392440547554\n",
      "2.673414438816076\n",
      "473.4790390182672 NAN SUM TEST\n",
      "[ 0.68877903  0.60867211 -0.62872097  0.74724233] [ 0.01804727  0.01572997 -0.02409396  0.01789414] E & error\n",
      "0.038390817570994204\n",
      "2.721903275399105\n",
      "199.0156371359115 NAN SUM TEST\n",
      "[ 0.79861819  0.67518474 -0.55600187  0.69209847] [ 0.03953074  0.03729488 -0.04829846  0.05041648] E & error\n",
      "0.08847683420857147\n",
      "2.6013984387366835\n",
      "423.614171584301 NAN SUM TEST\n",
      "[ 0.7335007   0.63612878 -0.54819176  0.6835772 ] [ 0.02065778  0.01977058 -0.02060039  0.0236364 ] E & error\n",
      "0.04243436493317085\n",
      "2.7474448382758965\n",
      "242.91284179920203 NAN SUM TEST\n",
      "[ 0.70596667  0.67703932 -0.70603647  0.65840238] [ 0.03484397  0.03565314 -0.04544015  0.03218665] E & error\n",
      "0.07473979321291754\n",
      "2.724814457222515\n",
      "507.6467799368391 NAN SUM TEST\n",
      "[ 0.6984073   0.68303353 -0.59205029  0.75132333] [ 0.01766626  0.01875131 -0.01594448  0.01825143] E & error\n",
      "0.0353701819178847\n",
      "2.6877287745670753\n",
      "1019.5598327249935 NAN SUM TEST\n",
      "[ 0.68113986  0.65756747 -0.62205945  0.726962  ] [ 0.00878811  0.00686958 -0.01016379  0.00889752] E & error\n",
      "0.01751829021408943\n",
      "2.679308352187471\n",
      "564.7429172868451 NAN SUM TEST\n",
      "[ 0.69003397  0.66965506 -0.61252248  0.70709684] [ 0.01822103  0.01605668 -0.0157071   0.01279021] E & error\n",
      "0.031624754358191634\n",
      "2.7134200204018524\n",
      "931.7266309448494 NAN SUM TEST\n",
      "[ 0.70369858  0.66370702 -0.62306545  0.72294897] [ 0.00790543  0.01019032 -0.0116523   0.00782093] E & error\n",
      "0.019059945509212275\n",
      "2.727830977335824\n",
      "1956.285508928101 NAN SUM TEST\n",
      "[ 0.69009078  0.67866881 -0.62881639  0.730255  ] [ 0.00371505  0.00450518 -0.00431776  0.00568367] E & error\n",
      "0.009222008988497597\n",
      "2.7145157022773105\n",
      "1126.8604816582306 NAN SUM TEST\n",
      "[ 0.69281551  0.67196361 -0.62237922  0.72735736] [ 0.00740063  0.00939964 -0.00595145  0.00876248] E & error\n",
      "0.015978846406071234\n",
      "2.6982144482230774\n",
      "478.22295812077226 NAN SUM TEST\n",
      "[ 0.68193261  0.67837985 -0.61953773  0.71836426] [ 0.01664048  0.01586879 -0.02093901  0.02001957] E & error\n",
      "0.036985794347500184\n",
      "2.68527728430133\n",
      "998.7407511366048 NAN SUM TEST\n",
      "[ 0.68673757  0.66772824 -0.61858941  0.71222207] [ 0.00809857  0.01107887 -0.00730509  0.00932956] E & error\n",
      "0.018130997761186048\n",
      "2.730793817594534\n",
      "586.5628850339519 NAN SUM TEST\n",
      "[ 0.67287626  0.70204816 -0.61814534  0.73772405] [ 0.01469574  0.01287119 -0.0162415   0.0170135 ] E & error\n",
      "0.030575771775913696\n",
      "2.692604171879398\n",
      "865.412903417921 NAN SUM TEST\n",
      "[ 0.6675597   0.65460729 -0.65899382  0.71144335] [ 0.01010515  0.01205679 -0.00542876  0.0124056 ] E & error\n",
      "0.02075693850790331\n",
      "2.7219120826286343\n",
      "1721.7555866553653 NAN SUM TEST\n",
      "[ 0.69053333  0.68286061 -0.62805473  0.72046341] [ 0.00516098  0.00680213 -0.00389715  0.00443669] E & error\n",
      "0.010381553357477124\n",
      "2.6951603819102044\n",
      "956.9842605442685 NAN SUM TEST\n",
      "[ 0.66882873  0.6809986  -0.62727193  0.71806112] [ 0.00761416  0.00874774 -0.01147102  0.00948323] E & error\n",
      "0.01886833635141485\n",
      "2.7278029316209964\n",
      "1566.379908658792 NAN SUM TEST\n",
      "[ 0.68896591  0.67360961 -0.63397235  0.73125506] [ 0.00581354  0.00579292 -0.0042602   0.00689491] E & error\n",
      "0.011534482769115291\n",
      "2.7246965254499895\n",
      "3300.8387420631384 NAN SUM TEST\n",
      "[ 0.69461101  0.67721479 -0.62231852  0.73055221] [ 0.00280217  0.00263871 -0.00313803  0.00214065] E & error\n",
      "0.005407826649655218\n",
      "2.7278177057122646\n",
      "1901.8808121670831 NAN SUM TEST\n",
      "[ 0.69367415  0.67684642 -0.6269299   0.73036723] [ 0.00431621  0.00457039 -0.00500497  0.00501473] E & error\n",
      "0.009471817723931514\n",
      "2.7113444724882676\n",
      "811.6598261272492 NAN SUM TEST\n",
      "[ 0.66852208  0.68724449 -0.61130981  0.7442681 ] [ 0.01161296  0.01215318 -0.00825964  0.01201056] E & error\n",
      "0.022249403159014133\n",
      "2.7358571778025165\n",
      "1675.4959413629078 NAN SUM TEST\n",
      "[ 0.69505734  0.68367664 -0.6260205   0.7311027 ] [ 0.00618509  0.00634447 -0.00390628  0.00462552] E & error\n",
      "0.010731365283456981\n",
      "2.7177895329180464\n",
      "999.5743579976563 NAN SUM TEST\n",
      "[ 0.67807149  0.67284013 -0.64660047  0.72027744] [ 0.00843856  0.00477953 -0.01077294  0.01106098] E & error\n",
      "0.018233337846653347\n",
      "2.6922234291097684\n",
      "446.0851371522747 NAN SUM TEST\n",
      "[ 0.68801299  0.6437494  -0.6284196   0.73204143] [ 0.01946084  0.02224256 -0.01908238  0.01872072] E & error\n",
      "0.03985044874577054\n",
      "2.7008406712947703\n",
      "879.4017941840459 NAN SUM TEST\n",
      "[ 0.69968567  0.66387218 -0.61232787  0.72495495] [ 0.00960325  0.0096423  -0.0087177   0.0128192 ] E & error\n",
      "0.020628292197769994\n",
      "2.6785587310192773\n",
      "493.34044192078363 NAN SUM TEST\n",
      "[ 0.67748074  0.68280597 -0.62018441  0.6980876 ] [ 0.01460269  0.01960816 -0.02154464  0.01616222] E & error\n",
      "0.03637453619061592\n",
      "2.731304983591543\n",
      "804.1607321570425 NAN SUM TEST\n",
      "[ 0.69443715  0.70765833 -0.62611617  0.70309333] [ 0.00858121  0.00852758 -0.01274758  0.01408022] E & error\n",
      "0.022519107948167305\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.7328421964648504\n",
      "1690.7965173392863 NAN SUM TEST\n",
      "[ 0.69387793  0.66917489 -0.63608721  0.73370218] [ 0.00510877  0.00585026 -0.00454059  0.00562288] E & error\n",
      "0.010609374863693993\n",
      "2.726355072356631\n",
      "975.891094986627 NAN SUM TEST\n",
      "[ 0.67652949  0.6848271  -0.6233741   0.74162438] [ 0.01019737  0.00695706 -0.01140682  0.00777025] E & error\n",
      "0.01851700454179151\n",
      "2.6876694734392825\n",
      "416.94633195665665 NAN SUM TEST\n",
      "[ 0.70437422  0.66920249 -0.65421004  0.65988272] [ 0.01624155  0.02163073 -0.02167799  0.02585523] E & error\n",
      "0.043244704229605004\n",
      "2.7571888106002502\n",
      "852.5542819618202 NAN SUM TEST\n",
      "[ 0.6998425   0.69062184 -0.6412578   0.72546667] [ 0.01044494  0.00787336 -0.01378443  0.00925835] E & error\n",
      "0.02113797593448899\n",
      "2.6960699274271946\n",
      "510.885967880781 NAN SUM TEST\n",
      "[ 0.71448687  0.65354185 -0.61067576  0.71736545] [ 0.01588678  0.01745554 -0.01771797  0.01934996] E & error\n",
      "0.03529069501202615\n",
      "2.545942633786867\n",
      "307.3950600044403 NAN SUM TEST\n",
      "[ 0.60194487  0.65155515 -0.59378018  0.69866244] [ 0.02213578  0.0346817  -0.0316872   0.02592868] E & error\n",
      "0.05804470504107021\n",
      "2.7207055258946045\n",
      "599.6907066553972 NAN SUM TEST\n",
      "[ 0.68841565  0.68515796 -0.61894954  0.72818238] [ 0.01543218  0.01194244 -0.0146426   0.01715532] E & error\n",
      "0.029824228219816833\n",
      "2.7320808258536804\n",
      "335.23123287009355 NAN SUM TEST\n",
      "[ 0.71779648  0.61470693 -0.62203503  0.77754238] [ 0.02155144  0.03342158 -0.02030816  0.02978974] E & error\n",
      "0.05367789872450654\n",
      "2.714921905949674\n",
      "550.5842165023287 NAN SUM TEST\n",
      "[ 0.69467659  0.66914754 -0.60903528  0.74206249] [ 0.01339571  0.02362817 -0.01103615  0.0162605 ] E & error\n",
      "0.033525149888051754\n",
      "2.7341815482449934\n",
      "1156.5311570037557 NAN SUM TEST\n",
      "[ 0.68612102  0.68359715 -0.63510924  0.72935414] [ 0.008948    0.00684683 -0.0083116   0.00641622] E & error\n",
      "0.015401175085068197\n",
      "2.7070064920068675\n",
      "669.5603855930385 NAN SUM TEST\n",
      "[ 0.6751447   0.68737534 -0.61213713  0.73234932] [ 0.01283481  0.01602467 -0.01209342  0.01245614] E & error\n",
      "0.02688733120423801\n",
      "2.7362367268470464\n",
      "290.60183968993636 NAN SUM TEST\n",
      "[ 0.64092767  0.78836248 -0.59208829  0.71485828] [ 0.02554099  0.03729611 -0.02927471  0.03218138] E & error\n",
      "0.06273748843390715\n",
      "2.7966764602110183\n",
      "582.0292142524368 NAN SUM TEST\n",
      "[ 0.68399264  0.711919   -0.64790872  0.7528561 ] [ 0.01453771  0.01658005 -0.01446202  0.0162601 ] E & error\n",
      "0.030980381902423717\n",
      "2.715638419794592\n",
      "354.60512843238405 NAN SUM TEST\n",
      "[ 0.71527659  0.63967413 -0.63870656  0.72198114] [ 0.02559714  0.02581995 -0.02471438  0.02490636] E & error\n",
      "0.05052732795284392\n",
      "2.633532518486417\n",
      "576.8540384124814 NAN SUM TEST\n",
      "[ 0.66684432  0.6650089  -0.58326624  0.71841306] [ 0.01703906  0.01390742 -0.01160744  0.01855108] E & error\n",
      "0.031026137656799492\n",
      "2.7262868737193955\n",
      "1141.2732259500835 NAN SUM TEST\n",
      "[ 0.68132349  0.69313009 -0.62898477  0.72284853] [ 0.00763156  0.00640682 -0.00889179  0.00829803] E & error\n",
      "0.01572288891542491\n",
      "2.740075812166258\n",
      "627.0122041095733 NAN SUM TEST\n",
      "[ 0.70366672  0.67303402 -0.62778634  0.73558873] [ 0.01121395  0.01580368 -0.01641359  0.01293381] E & error\n",
      "0.028499094026367827\n",
      "2.7073321034495423\n",
      "1037.1192594497556 NAN SUM TEST\n",
      "[ 0.68993765  0.66855607 -0.62942353  0.71941485] [ 0.00906045  0.0101445  -0.00634125  0.00900146] E & error\n",
      "0.017499723076381302\n",
      "2.7251217062772173\n",
      "2177.4325017442234 NAN SUM TEST\n",
      "[ 0.69035486  0.68221308 -0.62913903  0.72341474] [ 0.00376167  0.00457934 -0.00455099  0.00347291] E & error\n",
      "0.00823972634187354\n",
      "2.7631371370459146\n",
      "1254.9859091905953 NAN SUM TEST\n",
      "[ 0.69750411  0.69329242 -0.6360936   0.73624701] [ 0.00523922  0.008363   -0.00708074  0.00772299] E & error\n",
      "0.01439341987091507\n",
      "2.6677684248678553\n",
      "540.7790487224368 NAN SUM TEST\n",
      "[ 0.68536342  0.65420213 -0.59967514  0.72852773] [ 0.01505913  0.01679909 -0.01560588  0.01853799] E & error\n",
      "0.033108715769462886\n",
      "2.7371212064407278\n",
      "1106.477823154004 NAN SUM TEST\n",
      "[ 0.68962582  0.69776028 -0.63170081  0.7180343 ] [ 0.00838371  0.00889719 -0.00877847  0.00607448] E & error\n",
      "0.016229830154834066\n",
      "2.6982502768532064\n",
      "669.2383883826877 NAN SUM TEST\n",
      "[ 0.66691912  0.70365585 -0.6237228   0.70395251] [ 0.01511013  0.01593083 -0.01039505  0.01142429] E & error\n",
      "0.02684545607471135\n",
      "2.7537604867889742\n",
      "289.7368620145458 NAN SUM TEST\n",
      "[ 0.73699354  0.66974908 -0.63153738  0.71548049] [ 0.01716791  0.02989076 -0.03869703  0.03315232] E & error\n",
      "0.061520168049346485\n",
      "2.687297319153862\n",
      "576.1759950676218 NAN SUM TEST\n",
      "[ 0.68084699  0.65795641 -0.58801138  0.76048254] [ 0.01571071  0.01262175 -0.01722295  0.0162874 ] E & error\n",
      "0.031113410424652727\n",
      "2.6339757615257398\n",
      "323.7680373205736 NAN SUM TEST\n",
      "[ 0.6512096   0.69296177 -0.65180436  0.63800002] [ 0.01942953  0.02530504 -0.03504452  0.03072223] E & error\n",
      "0.05647854311834216\n",
      "2.657466086664978\n",
      "529.880437036473 NAN SUM TEST\n",
      "[ 0.68325169  0.65164739 -0.58851655  0.73405046] [ 0.01721926  0.01671948 -0.01831996  0.01472308] E & error\n",
      "0.03359216784570271\n",
      "2.7132015744053266\n",
      "1106.3649578300874 NAN SUM TEST\n",
      "[ 0.67205338  0.68570614 -0.63325159  0.72219046] [ 0.00881893  0.00676166 -0.00678739  0.00972967] E & error\n",
      "0.01625511164549253\n",
      "2.7547052724143892\n",
      "634.9171172451906 NAN SUM TEST\n",
      "[ 0.69604401  0.70742599 -0.62682664  0.72440864] [ 0.01134885  0.01647255 -0.01382453  0.01453934] E & error\n",
      "0.028331103164685966\n",
      "2.692224123065056\n",
      "267.92838257644604 NAN SUM TEST\n",
      "[ 0.63028163  0.69417002 -0.60728258  0.76048989] [ 0.02722956  0.03723711 -0.02296832  0.04522411] E & error\n",
      "0.0685624873439295\n",
      "2.7602293821088075\n",
      "560.3696563294684 NAN SUM TEST\n",
      "[ 0.72594747  0.6759569  -0.61452936  0.74379565] [ 0.01524508  0.01706268 -0.01823968  0.01293718] E & error\n",
      "0.031993809091188034\n",
      "2.760110090107376\n",
      "331.72242721495996 NAN SUM TEST\n",
      "[ 0.65497547  0.69497472 -0.63635759  0.77380231] [ 0.02939638  0.02540265 -0.02210611  0.03014356] E & error\n",
      "0.05391434269501016\n"
     ]
    }
   ],
   "source": [
    "l_S = []\n",
    "l_S_error = []\n",
    "#for i in range(40,41):\n",
    "for i in range(0,len(l_filesort)//16):\n",
    "    l_filelist = l_filesort[i*16:16*i+16]   \n",
    "    a_delta, a_coi, a_same = data_gather(l_filelist,directory)\n",
    "    #print(len(a_delta))\n",
    "    l_pol = []\n",
    "    for k in range(len(l_filelist)):\n",
    "        s_named = l_filelist[k]\n",
    "        l_pol.append(s_named[4:14])\n",
    "    #a_test = np.column_stack((l_filelist, l_pol))\n",
    "    #print(a_test)\n",
    "    a_total= np.empty(len(a_delta))\n",
    "    a_error = np.empty(len(a_delta))\n",
    "    a_first = np.empty(len(a_delta))\n",
    "    a_firsterror = np.empty(len(a_delta))\n",
    "    a_second = np.empty(len(a_delta))\n",
    "    a_seconderror = np.empty(len(a_delta))\n",
    "    a_param = (200,0,50,20,4,40,30)\n",
    "    a_param3 = (220,0,40,220,0,60,30),\n",
    "    # parameters are of form (amplitude, center, width, amp2, cent2, wid2, offset)\n",
    "    # this means the offset is actually double this parameter, because each gaussian uses it\n",
    "    # might get more functionality if each gaussian had different offsets? \n",
    "    \n",
    "    # commented out is the fitting function for the data. With the new results the hope is that this should \n",
    "    # be used to fit the Gaussians. Remember that I have currently commented out the same fiber fitting as well, \n",
    "    # so if we wanted that, it would need to be put back in\n",
    "    # these fits seem pretty sensitive to initial parameters\n",
    "    '''\n",
    "    for i in range(len(a_delta)):\n",
    "        if len(a_param) < 5:\n",
    "            a_param = (200,0,40,190,0,60,20)\n",
    "        a_total[i], a_error[i],a_param,a_param3 = GaussianFit(a_delta[i],a_coi[i],a_same[i],'/home/andrew/Documents/BNL 2018/Coding/Images/0x0_0x0'+np.str(i)+'.png',l_pol[i],a_param,a_param3)\n",
    "        #print(a_total[i])\n",
    "    totalcts= np.empty(len(a_delta))\n",
    "    first = np.empty(len(a_delta))\n",
    "    second = np.empty(len(a_delta))\n",
    "    #print(coi[0])\n",
    "    '''\n",
    "    for j in range(len(a_delta)):\n",
    "        a_total[j],a_first[j],a_second[j] = CoincidenceCount(a_delta[j],a_coi[j],a_same[j],l_pol[j])\n",
    "        a_error[j] = np.sqrt(np.abs(a_total[j]))\n",
    "    a_totalsort = [x for _,x in sorted (zip(l_pol,a_total))]\n",
    "    a_errorsort = [x for _,x in sorted (zip(l_pol,a_error))]\n",
    "    f_S,C,E = Sval(a_totalsort)\n",
    "    S_error = errorprop(C,E,a_error)\n",
    "    index = np.array([10,6,8,4,11,7,9,5,14,2,12,0,15,3,13,1]) \n",
    "    # index already explained in S_val module\n",
    "    l_pol = np.array(sorted(l_pol))[index]\n",
    "    l_S.append(f_S)\n",
    "    l_S_error.append(S_error)\n",
    "    data = np.column_stack((l_pol,np.array(a_totalsort)[index]))\n",
    "    #print(data)\n",
    "    #print(i)\n",
    "    #print(l_label[i])\n",
    "    np.savetxt('Coincidence'+l_label[i]+'.txt', data,fmt='%s' )\n",
    "a_datamap = np.column_stack((l_label,l_S,l_S_error))\n",
    "#np.savetxt('S_Valuesint.csv',a_datamap,fmt='%s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:40.237949Z",
     "start_time": "2018-09-21T15:06:40.227176Z"
    }
   },
   "outputs": [],
   "source": [
    "\n",
    "a_datamap = np.column_stack((l_label,l_S,l_S_error))\n",
    "np.savetxt('S_Values.csv',a_datamap,fmt='%s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:40.563302Z",
     "start_time": "2018-09-21T15:06:40.547819Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['A45_B22,5_' 'A135_B22,5' 'A45_B112,5' 'A135_B112,' 'A45_B67,5_'\n",
      " 'A135_B67,5' 'A45_B157,5' 'A135_B157,' 'A90_B22,5_' 'A0_B22,5_W'\n",
      " 'A90_B112,5' 'A0_B112,5_' 'A90_B67,5_' 'A0_B67,5_W' 'A90_B157,5'\n",
      " 'A0_B157,5_']\n"
     ]
    }
   ],
   "source": [
    "l_pol = []\n",
    "i = 0\n",
    "l_filelist = l_filesort[i*16:16*i+16]  \n",
    "for i in range(len(l_filelist)):\n",
    "    s_named = l_filesort[i]\n",
    "    l_pol.append(s_named[4:14])\n",
    "#a = [x for _,x in sorted (zip(l_pol,a_total))]\n",
    "b = sorted(l_pol)\n",
    "#print(l_pol[:16:4])\n",
    "#c = [x for _,x in sorted (zip(l_pol[:16][6:8],a_total))]\n",
    "#print(b)\n",
    "b = np.array(b)\n",
    "index = np.array([10,6,8,4,11,7,9,5,14,2,12,0,15,3,13,1])\n",
    "print(b[index])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:48.633985Z",
     "start_time": "2018-09-21T15:06:40.845805Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['$\\\\sigma = $0.066', '$\\\\sigma = $0.033', '$\\\\sigma = $0.06', '$\\\\sigma = $0.036', '$\\\\sigma = $0.017', '$\\\\sigma = $0.03', '$\\\\sigma = $0.071', '$\\\\sigma = $0.034', '$\\\\sigma = $0.058', '$\\\\sigma = $0.041', '$\\\\sigma = $0.02', '$\\\\sigma = $0.038', '$\\\\sigma = $0.023', '$\\\\sigma = $0.011', '$\\\\sigma = $0.019', '$\\\\sigma = $0.045', '$\\\\sigma = $0.022', '$\\\\sigma = $0.036', '$\\\\sigma = $0.081', '$\\\\sigma = $0.041', '$\\\\sigma = $0.076', '$\\\\sigma = $0.046', '$\\\\sigma = $0.022', '$\\\\sigma = $0.038', '$\\\\sigma = $0.088', '$\\\\sigma = $0.042', '$\\\\sigma = $0.075', '$\\\\sigma = $0.035', '$\\\\sigma = $0.018', '$\\\\sigma = $0.032', '$\\\\sigma = $0.019', '$\\\\sigma = $0.009', '$\\\\sigma = $0.016', '$\\\\sigma = $0.037', '$\\\\sigma = $0.018', '$\\\\sigma = $0.031', '$\\\\sigma = $0.021', '$\\\\sigma = $0.01', '$\\\\sigma = $0.019', '$\\\\sigma = $0.012', '$\\\\sigma = $0.005', '$\\\\sigma = $0.009', '$\\\\sigma = $0.022', '$\\\\sigma = $0.011', '$\\\\sigma = $0.018', '$\\\\sigma = $0.04', '$\\\\sigma = $0.021', '$\\\\sigma = $0.036', '$\\\\sigma = $0.023', '$\\\\sigma = $0.011', '$\\\\sigma = $0.019', '$\\\\sigma = $0.043', '$\\\\sigma = $0.021', '$\\\\sigma = $0.035', '$\\\\sigma = $0.058', '$\\\\sigma = $0.03', '$\\\\sigma = $0.054', '$\\\\sigma = $0.034', '$\\\\sigma = $0.015', '$\\\\sigma = $0.027', '$\\\\sigma = $0.063', '$\\\\sigma = $0.031', '$\\\\sigma = $0.051', '$\\\\sigma = $0.031', '$\\\\sigma = $0.016', '$\\\\sigma = $0.028', '$\\\\sigma = $0.017', '$\\\\sigma = $0.008', '$\\\\sigma = $0.014', '$\\\\sigma = $0.033', '$\\\\sigma = $0.016', '$\\\\sigma = $0.027', '$\\\\sigma = $0.062', '$\\\\sigma = $0.031', '$\\\\sigma = $0.056', '$\\\\sigma = $0.034', '$\\\\sigma = $0.016', '$\\\\sigma = $0.028', '$\\\\sigma = $0.069', '$\\\\sigma = $0.032', '$\\\\sigma = $0.054']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# hasn't been modularized, but generates the graphic for S values. \n",
    "a_sval = np.loadtxt('S_Values.csv',dtype = \"float\", usecols = (1))\n",
    "a_test = np.loadtxt('S_Values.csv',dtype = \"str\")\n",
    "a_svalerr = np.loadtxt('S_Values.csv',dtype = \"float\", usecols = (2))\n",
    "#print(a_test)\n",
    "a_s = a_sval\n",
    "title = []\n",
    "a_subplot = [0]*9\n",
    "a_errors = [0]*9\n",
    "a_errnum = [0]*9\n",
    "for i in range(3):\n",
    "    for j in range(3):\n",
    "        title.append(np.str(i)+'x'+np.str(j))\n",
    "fig, axn = plt.subplots(3, 3, sharey=True, sharex = True, figsize = (10,10))\n",
    "cbar_ax = fig.add_axes([.91, .3, .03, .4])\n",
    "plt.rcParams[\"font.family\"] = \"Times New Roman\"\n",
    "label = [0]*81\n",
    "for i in range(81):\n",
    "    #label[i] = np.str(np.around(a_s[i],2)) + \"+\\-\" + np.str(np.around(a_svalerr[i],2))\n",
    "    label[i] = '$\\sigma = $' +  np.str(np.around(a_svalerr[i],3))\n",
    "print(label)\n",
    "for i in range(9):\n",
    "    a_subplot[i] = np.row_stack(((a_s[i*9:9*i+3]),a_s[i*9+3:9*i+6],a_s[i*9+6:9*i+9]))\n",
    "    a_errors[i] = np.row_stack(((label[i*9:9*i+3]),label[i*9+3:9*i+6],label[i*9+6:9*i+9]))\n",
    "    a_errnum[i] = np.row_stack(((a_svalerr[i*9:9*i+3]),a_svalerr[i*9+3:9*i+6],a_svalerr[i*9+6:9*i+9]))\n",
    "    #print(a_subplot[i],i)\n",
    "#print(a_subplot)\n",
    "#print(a_errors)\n",
    "\n",
    "for i, ax in enumerate(axn.flat):\n",
    "    sns.heatmap(np.around(a_subplot[i],2), annot=a_errors[i], fmt='',annot_kws={\"size\": 10}, ax=ax,cbar=i == 0,\n",
    "                vmin=2.60, vmax=2.80,       # these min & maxes should change when we get a max of 2.82\n",
    "                cbar_ax=None if i else cbar_ax,cmap=\"YlGnBu\" ) # feel free to change the color\n",
    "    ax.set_title(title[i],fontweight=\"bold\")\n",
    "    for label in ax.get_yticklabels():\n",
    "        label.set_weight(\"bold\")\n",
    "    for label in ax.get_xticklabels():\n",
    "        label.set_weight(\"bold\")\n",
    "plt.savefig('tomography.eps', bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:48.855731Z",
     "start_time": "2018-09-21T15:06:48.839021Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.06567912 0.03296215 0.0601589 ]\n",
      " [0.03643963 0.01719373 0.03018655]\n",
      " [0.07143603 0.03369152 0.05814026]]\n"
     ]
    }
   ],
   "source": [
    "def f(x, y):\n",
    "    return np.sin(np.sqrt(x ** 2 + y ** 2))\n",
    "\n",
    "x = np.linspace(0, 2, 3)\n",
    "y = np.linspace(0, 2, 3)\n",
    "\n",
    "X, Y = np.meshgrid(x, y)\n",
    "Z = a_subplot[0]\n",
    "sigma = a_errnum[0]\n",
    "print(sigma)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:49.275389Z",
     "start_time": "2018-09-21T15:06:49.091894Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 1. 2.]\n",
      "[[1.]]\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "too many indices for array",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-16-9e608f1ba679>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mtest1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmeshgrid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m: too many indices for array"
     ]
    }
   ],
   "source": [
    "print(X[1])\n",
    "#print(x[1])\n",
    "test1, test2 = np.meshgrid(x[1],y[1])\n",
    "print(test1)\n",
    "print(X[1,2,2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:49.277308Z",
     "start_time": "2018-09-21T15:06:38.013Z"
    }
   },
   "outputs": [],
   "source": [
    "print(len(a_subplot))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:49.280340Z",
     "start_time": "2018-09-21T15:06:38.317Z"
    }
   },
   "outputs": [],
   "source": [
    "x = np.empty((3,9))\n",
    "y = np.empty((3,9))\n",
    "Z = np.empty((3,3,9))\n",
    "sigma = np.empty((3,3,9))\n",
    "X = np.empty((3,3,9))\n",
    "Y = np.empty((3,3,9))\n",
    "for i in range(3):\n",
    "    x[:,i] = np.linspace(i*3, 2+3*i, 3)\n",
    "    y[:,i] = np.linspace(i*3, 2+3*i, 3)\n",
    "    fill1, fill2 = np.meshgrid(x[:,i], y[:,i])\n",
    "    X[:,:,i]= fill1\n",
    "    Y[:,:,i]= fill2\n",
    "    Z[:,:,i] = a_subplot[i]\n",
    "    print(Z[:,:,i])\n",
    "    sigma[:,:,i] = a_errnum[i]\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:49.283509Z",
     "start_time": "2018-09-21T15:06:38.654Z"
    }
   },
   "outputs": [],
   "source": [
    "z = np.empty((9,9))\n",
    "for i in range(9):\n",
    "    z[:,i] = Z[:,:,i].flatten()\n",
    "print(z[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-21T15:06:49.286739Z",
     "start_time": "2018-09-21T15:06:39.004Z"
    }
   },
   "outputs": [],
   "source": [
    "from matplotlib.ticker import MaxNLocator\n",
    "fig = plt.figure(figsize = (8,8))\n",
    "ax = plt.axes(projection='3d')\n",
    "for i in range(9):\n",
    "    ax.scatter(X[:,:,i], Y[:,:,i], Z[:,:,i], s = 40, marker = 'o', c = z[i],\n",
    "                cmap='YlGnBu', edgecolor='none' )\n",
    "\n",
    "\n",
    "#ax.scatter(X, Y, Z_poserr, \n",
    "#                cmap='YlGnBu', edgecolor='none', alpha = 0.5)\n",
    "#ax.scatter(X, Y, Z_negerr, \n",
    "#                cmap='YlGnBu', edgecolor='none', alpha = 0.5)\n",
    "ax.set_title('surface');\n",
    "ax.set_xlabel('x')\n",
    "ax.set_ylabel('y')\n",
    "ax.set_zlabel('z');\n",
    "ax.xaxis.set_major_locator(MaxNLocator(integer=True))\n",
    "ax.yaxis.set_major_locator(MaxNLocator(integer=True))\n",
    "#ax.view_init(20, 55)\n"
   ]
  }
 ],
 "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": "3.6.5"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "oldHeight": 471,
   "position": {
    "height": "40px",
    "left": "656px",
    "right": "-28px",
    "top": "72px",
    "width": "683px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "varInspector_section_display": "none",
   "window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}