{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ATLAS Test Beam Data\n",
    "\n",
    "### Authors: \n",
    "- Christian Michelsen (Niels Bohr Institute)\n",
    "- Troels C. Petersen (Niels Bohr Institute)\n",
    "\n",
    "### Date:    \n",
    "- 13-11-2018 (latest update)\n",
    "\n",
    "***\n",
    "\n",
    "Python script for analysis of ATLAS test beam data. \n",
    "\n",
    "The program reads an ASCII (i.e. text) data file, containting a large number of events, where a charged particle (electron or pion) passed through a slice of the ATLAS detector. Each passage is recorded by different detectors, boiling down to eleven numbers (some more relevant than others). The exercise is to separate electron and pion events based on these numbers, and in turn us this information to measure the interaction of pions and electrons seperately.\n",
    "\n",
    "NOTE: Though the data is from particle physics, it could in principle have been from ANY other source, and the eleven numbers could for example have been indicators of cancer, key numbers for investors, or index numbers for identifying potential costumors.\n",
    "\n",
    "For more information on ATLAS test beam: http://www.nbi.dk/~petersen/Teaching/Stat2018/TestBeam/TestBeamDataAnalysis.html.\n",
    "\n",
    "***\n",
    "\n",
    "First, we import the modules we want to use:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read the data file:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Write extensive output\n",
    "verbose = True\n",
    "N_verbose = 10\n",
    "N_read = 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Initialize lists of the eleven variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "Cher_electrons = []\n",
    "nLT_electrons  = []\n",
    "nHT_electrons  = []\n",
    "EM0_electrons  = []\n",
    "EM1_electrons  = []\n",
    "EM2_electrons  = []\n",
    "EM3_electrons  = []\n",
    "Had0_electrons = []\n",
    "Had1_electrons = []\n",
    "Had2_electrons = []\n",
    "Muon_electrons = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Open file, `DataSet_AtlasPid_ElectronPion_2GeV.txt`, and read in all the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cher:  634.0    nLT, nHT: 32,  8    EM:  0.12  0.47  0.87 -0.04   Had:  0.00  0.00  0.00    Muon: 417.0\n",
      "Cher:  816.0    nLT, nHT: 42, 12    EM:  0.05  0.87  0.77 -0.01   Had:  0.00  0.00  0.00    Muon: 388.0\n",
      "Cher:  943.0    nLT, nHT: 46,  8    EM:  0.19  0.53  0.96  0.01   Had:  0.00  0.00  0.00    Muon: 377.0\n",
      "Cher:  907.0    nLT, nHT: 33,  3    EM:  0.20  0.74  0.72  0.02   Had:  0.00  0.00  0.00    Muon: 372.0\n",
      "Cher:  775.0    nLT, nHT: 35,  4    EM: -0.01  0.82  0.57  0.02   Had:  0.00  0.00  0.00    Muon: 392.0\n",
      "Cher:  773.0    nLT, nHT: 39,  9    EM:  0.10  1.24  0.27 -0.03   Had:  0.00  0.00  0.00    Muon: 398.0\n",
      "Cher:  782.0    nLT, nHT: 30,  7    EM:  0.35  0.89  0.60 -0.04   Had:  0.00  0.00  0.00    Muon: 369.0\n",
      "Cher:  700.0    nLT, nHT: 39,  9    EM:  0.03  1.14  0.60  0.02   Had:  0.00  0.03  0.76    Muon: 408.0\n",
      "Cher:  542.0    nLT, nHT: 42,  1    EM: -0.03 -0.01  0.13 -0.05   Had:  0.00  0.00  0.00    Muon: 412.0\n",
      "Cher:  752.0    nLT, nHT: 40,  7    EM:  0.09  0.66  0.57  0.10   Had:  0.00  0.00  0.00    Muon: 387.0\n",
      "Total number of events read: 33125\n"
     ]
    }
   ],
   "source": [
    "\n",
    "with open('DataSet_AtlasPid_ElectronPion_2GeV.txt', 'r') as f:\n",
    "\n",
    "    # Loop over lines and extract variables of interest\n",
    "    header = f.readline()\n",
    "\n",
    "    for line in f:\n",
    "        line = line.strip()         # Remove spaces, tabs, and other 'whitespace' from the line from both ends until a character is met\n",
    "        columns = line.split()      # Split the line at whitespace character (spaces, tabs, etc.) into a list\n",
    "\n",
    "        Cher_i = float(columns[ 0])\n",
    "        nLT_i  = int  (columns[ 1])\n",
    "        nHT_i  = int  (columns[ 2])\n",
    "        EM0_i  = float(columns[ 3])\n",
    "        EM1_i  = float(columns[ 4])\n",
    "        EM2_i  = float(columns[ 5])\n",
    "        EM3_i  = float(columns[ 6])\n",
    "        Had0_i = float(columns[ 7])\n",
    "        Had1_i = float(columns[ 8])\n",
    "        Had2_i = float(columns[ 9])\n",
    "        Muon_i = float(columns[10])\n",
    "\n",
    "        Cher_electrons.append(Cher_i)\n",
    "        nLT_electrons.append(nLT_i)\n",
    "        nHT_electrons.append(nHT_i)\n",
    "        EM0_electrons.append(EM0_i)\n",
    "        EM1_electrons.append(EM1_i)\n",
    "        EM2_electrons.append(EM2_i)\n",
    "        EM3_electrons.append(EM3_i)\n",
    "        Had0_electrons.append(Had0_i)\n",
    "        Had1_electrons.append(Had1_i)\n",
    "        Had2_electrons.append(Had2_i)\n",
    "        Muon_electrons.append(Muon_i)\n",
    "\n",
    "        if (verbose and N_read < N_verbose):\n",
    "            print(f\"Cher: {Cher_i:6.1f}    nLT, nHT: {nLT_i:2d}, {nHT_i:2d}    EM: {EM0_i:5.2f} {EM1_i:5.2f} {EM2_i:5.2f} {EM3_i:5.2f}   Had: {Had0_i:5.2f} {Had1_i:5.2f} {Had2_i:5.2f}    Muon: {Muon_i:5.1f}\")\n",
    "        N_read += 1\n",
    "        \n",
    "print(f\"Total number of events read: {N_read:d}\")\n",
    "\n",
    "Cher_electrons = np.array(Cher_electrons)\n",
    "nLT_electrons  = np.array(nLT_electrons)\n",
    "nHT_electrons  = np.array(nHT_electrons)\n",
    "EM0_electrons  = np.array(EM0_electrons)\n",
    "EM1_electrons  = np.array(EM1_electrons)\n",
    "EM2_electrons  = np.array(EM2_electrons)\n",
    "EM3_electrons  = np.array(EM3_electrons)\n",
    "Had0_electrons = np.array(Had0_electrons)\n",
    "Had1_electrons = np.array(Had1_electrons)\n",
    "Had2_electrons = np.array(Had2_electrons)\n",
    "Muon_electrons = np.array(Muon_electrons)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Alternative Numpy version (notice the difference in lenght between the cell above and the one below):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(33125,)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = np.loadtxt('DataSet_AtlasPid_ElectronPion_2GeV.txt', skiprows=1, unpack=True)\n",
    "Cher_electrons, nLT_electrons, nHT_electrons, EM0_electrons, EM1_electrons, EM2_electrons, EM3_electrons, Had0_electrons, Had1_electrons, Had2_electrons, Muon_electrons = data\n",
    "Cher_electrons.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Your analysis:\n",
    "\n",
    "This is where your analysis should go:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "\n",
    "\n",
    "# This is where your analysis should go:\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And plot the results, cuts or whatever type of analysis choices you have made:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(ncols=2, nrows=2, figsize=(16, 10))\n",
    "\n",
    "ax[0,0].hist(Cher_electrons, bins=200, range=(400, 1400), histtype='step', label='Cherenkov')\n",
    "ax[0,0].set(xlabel=\"Cherenkov\", ylabel=\"Frequency\")\n",
    "\n",
    "ax[1,0].hist(EM1_electrons , bins=200, range=(-0.5, 2.0), histtype='step', label='EM1')\n",
    "ax[1,0].set(xlabel=\"EM1\", ylabel=\"Frequency\")\n",
    "\n",
    "#                                                (binx, biny), ((x1, x2),     (y1, y2)) or \"binary\"\n",
    "h = ax[0,1].hist2d(Cher_electrons, EM1_electrons, (100, 100), ((400, 1400), (-0.5, 2.0)), cmap=\"Reds\")\n",
    "plt.colorbar(h[3], ax=ax[0, 1]) # z-scale on the right of the figure\n",
    "ax[0,1].set(xlabel=\"Cherenkov\", ylabel=\"EM1\")\n",
    "\n",
    "# You can add your own fourth plot\n",
    "fig.delaxes(ax[1,1])\n",
    "\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we show a simple, quick Seaborn function `jointplot` and how it can be very useful when visualizing 2D-distributions:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x7efdad763860>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the joint distribution using kernel density estimation\n",
    "g = sns.jointplot(Cher_electrons, EM1_electrons, xlim=(400, 1400), ylim=(-0.5, 2.0), kind=\"kde\", height=10, space=0) # also try kind=\"hex\" or \"kde\"\n",
    "g.set_axis_labels(xlabel='Cher_electrons', ylabel='EM1_electrons')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Questions to be answered:\n",
    "\n",
    "Generally, this analysis is about separating electrons and pions (and determining how well this can be done), followed by a few questions characterizing the detector\n",
    "response to each type of particle.\n",
    "\n",
    "Below are questions guiding you, some/most of which your analysis should cover, but you do not have to follow them blindly (I've put \"Optional\" on those that are not\n",
    "essential). \n",
    "\n",
    "Start by considering the data, and get a feel for the typical range of each variable. Plot the variables, both in 1D and also 2D! From considering these plots,\n",
    "guess/estimate an approximate knowledge of how electrons and pions distribute themselves in the variables above, and how to make a selection of these.\n",
    "\n",
    "As described on the webpage introducing the data, the three detectors, Cherenkov, TRT and Calorimeters are each capable of separating electrons and pions. As they are\n",
    "INDEPENDENT (three separate detectors), they may be used to cross check each other, and this is what you should use, in fact the essential part of the analysis!\n",
    "\n",
    "\n",
    "Questions:\n",
    "----------\n",
    "1. Find for each of these three detector systems one variable (i.e. a combination), which seem to separate electrons and pions best. For example, start with the\n",
    "    Cherenkov, which is only a single number, and assume/guess that the large peak at low values is mainly from pions, while the upper broad peak is from electrons.  Now plot the TRT and Calorimeter distributions when the Cherenkov selects a pion and afterwards an electron. This should give you a hint of how to separate pions and electrons using the TRT and Calorimeters.\n",
    "\n",
    "    Hint: Sometimes variables from a single detector are more powerful, when they are combined, e.g. taken ratios of. For the TRT this may be somewhat obvious, but for the EMcalo, it is not as simple. While one may simply use the second layer, involving this layer in a ratio may enhance the separation power.\n",
    "\n",
    "2. Next you should try to see, if you can make a selection, which gives you a fairly large and clean electron and pion sample, respectively. The question is, how can you know how clean your sample is and how efficient your selection is?  This can actually be measured in the data itself, using the fact that there are three independent detectors. For example, start by making an electron and a pion selection using two of the three variables, and plot the third variable for each of these selections. Now you can directly see, how electrons and pions will distribute themselves in this third variable. Are you worried, that there are pions in your electron sample, and vice versa? Well, there will probably be, but so few, that it won't matter too much, at least not to begin with. Why? Well, let us assume that for each detector, 80% of electrons pass your requirement, but also 10% of pions do. Assuming an even number of electrons and pions (which is not really the case), then with two detector cuts, you should get a sample, which is: $\\frac{0.8\\cdot0.8} {0.8\\cdot0.8 + 0.1\\cdot0.1} = 98.5\\%$ pure.\n",
    "\n",
    "    Now with this sample based on cuts on the two other detectors, ask what fraction of electrons and pions passes your electron selection. The fraction of electrons, that are not selected as electrons will be your TYPE I errors, denoted alpha, while the fraction of pions, that do pass the cut will be your TYPE II errors, denoted beta. Measure these for each of the two cuts in the three detector types, and ask yourself if they are \"reasonable\", i.e. something like in the example above. If not, then you should perhaps reconsider adjusting your cuts.\n",
    "\n",
    "    By now, you should for each detector have 6 numbers:\n",
    "    - The electron cut value above which you accept an electron.\n",
    "    - The efficiency (i.e. 1-alpha) for electrons of this cut.\n",
    "    - The fake rate (i.e. beta) for pions of this cut.\n",
    "    - The pion cut value below which you accept a pion (may be same value as above!).\n",
    "    - The efficiency (i.e. 1-alpha) for pions of this cut.\n",
    "    - The fake rate (i.e. beta) for electrons of this cut.\n",
    "\n",
    "\n",
    "3. Given the efficiencies and fake rates of each cut, try to combine these (again assuming that they are independent) into knowledge of your sample purities and also the total number of electrons and pions in the whole sample. Do the sum of estimated electrons and pions added actually match the number of particles in total? This is a good cross check!\n",
    "\n",
    "4. If the number of pions was suddenly 1.000 that of elections, would you still be able to get a sample of fairly pure (say 90% pure) electrons? And if so, what would the efficiency for these electrons be?\n",
    "\n",
    "5. One of the purposes of the testbeam was to measure the response of the TRT detector to exactly electrons and pions. Consider for example only events that has 33 TRT hits (i.e. `nLT` $= 33$). As the High-Threshold probability (i.e. probability of passing the High-Threshold, given that the Low-Threshold was passed), is assumed to be constant in the TRT detector (but quite different for electrons and pions), what distribution should the number of High-Threshold hits (`nHT`) follow? And is that really the case, both for electrons and pions?\n",
    "\n",
    "6. Expanding on problem 2), try now to calculate ROC curves for each of the three detectors. These are obtained by making a clean selection using the two other detectors for electrons and pions seperately, and then integrating over these two distributions, using the running (normalised) integral of each as x and y coordinate in a (ROC) graph. If you do not manage on your own, perhaps consider the ROC calculator example, which is posted along with this exercise.\n",
    "\n",
    "\n",
    "7. __(Optional)__: Still considering `nLT` $= 33$, and given that there are both electrons and pions in the sample (each with a different HT probability), `nHT` should in the unselected data be a combination of two distributions. Try to fit the number of HT hits with two distributions combined. Do they fit the data? And can this fit be used to estimate the fraction of pions and electrons in the sample? And does that estimate match you previous estimate? Perhaps retry with other values for the number of TRT hits.\n",
    "\n",
    "8. __(Optional)__: Try to select pions using three different (mutually exclusive) techniques:\n",
    "    1. Passing only a hadronic calorimeter requirement (e.g. that the sum of the three HCal values is above some minimum energy).\n",
    "    2. Passing only Cherenkov AND EMcalo requirements.\n",
    "    3.  Passing both A) and B).\n",
    "\n",
    "    Try to measure the HT probability (i.e. fraction of High-Threshold hits) for each of these three pion samples. Do they agree with each other?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "executable": "/usr/bin/env python",
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  },
  "main_language": "python"
 },
 "nbformat": 4,
 "nbformat_minor": 4
}