exercises.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to pyROOT\n",
    "\n",
    "**NOTE** - If you get an error about a \"missing kernel\". Close this notebook and run the `setupLibraries` notebook to create the required `cmsdas-preexercise` kernel.\n",
    "\n",
    "This exercise is intended to provide you with basic familiarity with pyROOT provides bindings for all classes within the ROOT libraries and allows for replacing the usual C++ with the often less cumbersome python. The goal is to obtain a general understanding of the syntax required to import and make use of the ROOT libraries within a basic python script. Various examples are provided in order to demonstrate TH1 histogram manipulation including; reading from a .root file, creating, binning, re-binning, scaling, plotting and fitting to a Gaussian.\n",
    "\n",
    "Whether you use python or C++ to complete your analysis is a personal preference however with the current lack of documentation on pyROOT many students stick with C++ in order to ensure their access to coding examples and experts. It is our hope that through providing you with this basic introduction and Github repository of example scripts, which you are encouraged to add to, that we can bring together the existing pyROOT community within CMS and foster its growth.\n",
    "\n",
    "## About Jupyter\n",
    "Jupyter is a novel way to both execute and teach Python libraries. This `notebook` contains many `cells`, each of with contain snippets of executable Python code. You ran a cell earlier to create the `cmsdas-preexercise` kernel. Executions of cells within the same notebook happen within the same Python environment, so notebooks can be used to gradually and incrementally perform data analysis. Execute your first Python cell by clicking the following cell and pressing `Shift+Enter`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This cell creates a new variable `x`. The next cell confirms that Python remembers the value you previously gave it. Jupyter puts the output for each cell directly after it, so you can see what code produced which output."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Our first pyROOT example\n",
    "Let's examine what a typical pyROOT script will look like. As you read through the documentation, make sure to execute each cell in order.\n",
    "\n",
    "All pyROOT scripts must install the ROOT libraries into the Python process. ROOT isn't a standard part of Python, so we load it from CMSSW via the kernel we installed. To load ROOT, execute the following cell"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import ROOT"
   ]
  },
  {
   "attachments": {
    "Screen%20Shot%202018-11-04%20at%209.19.43%20PM.png": {
     "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVQAAAA0CAYAAADVEcBbAAAYJGlDQ1BJQ0MgUHJvZmlsZQAAWIWVWQVUVU+3n3PPDbhcLt3d3Q3S3d0IXLq7sQiRUBEElBJBBUEFgxKxEEQQEVTAQCQMSgUFFAF5h9D/977vrffWm7XOOb+7Z8+e38zeM3P2PQCwMZMiIkJQ1ACEhsVE2RjqcDs5u3DjJgEeMAICgIA4yTs6QtvKygwg5c/zv5eVYUQPKc8ltmz9Z/3/Wmh8fKO9AYCsEOzlE+0diuBGANCs3hFRMQBgBhA5X3xMxBZeQjB9FEIQACzZFvbfwexb2GsHS2/r2NnoIlgPADICiRTlDwBxyz53nLc/YocYgdTRhvkEhiGq6QjW8A4g+QDA2onoiIeGhm/hBQQLe/2LHf//ZtPrr00Syf8v3hnLdiHTC4yOCCEl/j+n4/8uoSGxf/rgRS5CQJSRzdaYkXm7GBxuuoUJCG4P87KwRDAtgh8F+mzrb+HXAbFG9rv6897RusicIZ4GKOBD0jNFMDKXKMbYYHvtXSxLitpui+ijLAJjjO12sVdUuM2ufVRcWIiF2a6dzABf4z/4jG+0vu0fHb9AA2MEI5GGakwKsHPc4YnqjAt0sEAwEcED0cG2prttx5ICdC3+6ETF2mxx5kfwkl+Ugc2ODswcGv1nXLCkN2m7L2YEa8UE2BnttIWdfKOdzP5w8PHV09/hAPv4htnvcoOR6NKx2W2bERFitasPn/ENMbTZmWf4anSc7Z+2z2KQANuZB3gyiGRitcMfXomIsbLb4YZGAzOgC/QAN4hFLi8QDoJAYP98yzzya6fGAJBAFPAHvkBiV/KnheN2TRhytwVJ4DOCfEH033Y627W+IA6Rb/yV7twlgN92bdx2i2DwEcGhaFa0BloNbYbctZBLFq2MVvnTjpvqT69Yfawe1ghrgBX5y8MbYR2CXFEg8H+QmSJPX2R0W1zC/ozhH3uYj5hBzCRmCDOOeQUcwPttK7taHoGpUf/GnBuYg3HEmsHu6LwQmzN/dNCCCGsFtA5aHeGPcEczolmBBFoeGYk2WhMZmwIi/VeGsX+5/TOX/97fFut/Hc+unChKVNhl4fXXM7p/tf7diu6/zJEP8jT9d004E74Bd8P34R64HW4B3PBduBXug29v4b+R8H47Ev70ZrPNLRixE/hHR/qS9Iz0+n/0TtplELXtbxDjmxCztSB0wyMSowL9A2K4tZEd2ZfbOMxbUpxbVlpGGYCt/X1n+/hus71vQ4xP/5H5TgOwB4lx8oF/ZEEnAajrAoAp+x+ZoCsALOIAXHvmHRsVtyNDb90wyMlBhawMFsAJ+IAwMiZZoAjUgBbQBybAEtgBZ+COzHoACEVYx4N9IAVkgBxwAhSCElABzoGL4Aq4DlpAO7gPHoLHYAAMgTdIbHwAc2ABrIA1CIJwECVEB7FAXJAAJAbJQsqQBqQPmUE2kDPkCflDYVAstA9Kg3KgfKgEqoRqoWvQTeg+1AMNQq+gCWgG+gb9QsEoAooexYESREmhlFHaKFOUHWovyh8ViUpCpaOOo06jqlCXUc2o+6jHqCHUOGoOtQwDmAJmhHlgCVgZ1oUtYRfYD46CD8DZcBFcBdfDbYivn8Pj8Dy8isai6dDcaAkkPo3Q9mhvdCT6APoougR9Ed2M7kQ/R0+gF9C/MZQYdowYRhVjjHHC+GPiMRmYIkw1pgnThaydD5gVLBbLiBXCKiFr0xkbhE3GHsWWYxuw97CD2CnsMg6HY8GJ4dRxljgSLgaXgSvGXcbdxT3DfcD9JKMg4yKTJTMgcyELI0slKyKrI7tD9ozsE9kaOTW5ALkquSW5D3kieS75efI28qfkH8jX8DR4Ibw63g4fhE/Bn8bX47vwo/jvFBQUvBQqFNYUgRSHKE5TXKV4RDFBsUqgJYgSdAluhFjCcUIN4R7hFeE7JSWlIKUWpQtlDOVxylrKB5RjlD+JdERJojHRh3iQWEpsJj4jfqEipxKg0qZyp0qiKqK6QfWUap6anFqQWpeaRH2AupT6JvUI9TINHY0MjSVNKM1RmjqaHpppWhytIK0+rQ9tOu052ge0U3QwHR+dLp03XRrdebouug/0WHohemP6IPoc+iv0/fQLDLQM8gwODAkMpQy3GcYZYUZBRmPGEMZcxuuMw4y/mDiYtJl8mbKY6pmeMf1gZmPWYvZlzmZuYB5i/sXCzaLPEsySx9LC8pYVzSrKas0az3qGtYt1no2eTY3Nmy2b7Trba3YUuyi7DXsy+zn2PvZlDk4OQ44IjmKOBxzznIycWpxBnAWcdzhnuOi4NLgCuQq47nLNcjNwa3OHcJ/m7uRe4GHnMeKJ5ank6edZ4xXitedN5W3gfcuH51Pm8+Mr4OvgW+Dn4jfn38d/if+1ALmAskCAwCmBboEfgkKCjoJHBFsEp4WYhYyFkoQuCY0KUwprCkcKVwm/EMGKKIsEi5SLDIiiRBVEA0RLRZ+KocQUxQLFysUGxTHiKuJh4lXiIxIECW2JOIlLEhOSjJJmkqmSLZJfpPilXKTypLqlfksrSIdIn5d+I0MrYyKTKtMm801WVNZbtlT2hRylnIHcQblWuUV5MXlf+TPyLxXoFMwVjih0KGwoKilGKdYrzijxK3kqlSmNKNMrWykfVX6kglHRUTmo0q6yqqqoGqN6XfWrmoRasFqd2vQeoT2+e87vmVLnVSepV6qPa3BreGqc1RjX5NEkaVZpTmrxafloVWt90hbRDtK+rP1FR1onSqdJ54euqu5+3Xt6sJ6hXrZevz6tvr1+if6YAa+Bv8ElgwVDBcNkw3tGGCNTozyjEWMOY2/jWuMFEyWT/SadpgRTW9MS00kzUbMoszZzlLmJ+UnzUQsBizCLFktgaWx50vKtlZBVpNUta6y1lXWp9UcbGZt9Nt22dLYetnW2K3Y6drl2b+yF7WPtOxyoHNwcah1+OOo55juOO0k57Xd67MzqHOjc6oJzcXCpdll21XctdP3gpuCW4Ta8V2hvwt4ed1b3EPfbHlQeJI8bnhhPR886z3WSJamKtOxl7FXmteCt633Ke85Hy6fAZ8ZX3Tff95Oful++37S/uv9J/5kAzYCigPlA3cCSwMUgo6CKoB/BlsE1wZshjiENoWShnqE3w2jDgsM6wznDE8IHI8QiMiLGI1UjCyMXokyjqqOh6L3RrTH0yKtOX6xw7OHYiTiNuNK4n/EO8TcSaBLCEvoSRROzEj8lGSRdSEYneyd37OPZl7JvYr/2/soD0AGvAx0H+Q6mH/xwyPDQxRR8SnDKk1Tp1PzUpTTHtLZ0jvRD6VOHDQ9fyiBmRGWMHFE7UpGJzgzM7M+SyyrO+p3tk92bI51TlLN+1Pto7zGZY6ePbR73O96fq5h75gT2RNiJ4TzNvIv5NPlJ+VMnzU82F3AXZBcsFXoU9hTJF1Wcwp+KPTV+2ux0azF/8Yni9ZKAkqFSndKGMvayrLIf5T7lz85onamv4KjIqfh1NvDsy0rDyuYqwaqic9hzcec+nnc4331B+UJtNWt1TvVGTVjN+EWbi521SrW1dex1uZdQl2IvzVx2uzxwRe9Ka71EfWUDY0POVXA19ursNc9rw9dNr3fcUL5R3yjQWNZE15TdDDUnNi+0BLSMtzq3Dt40udnRptbWdEvyVk07T3vpbYbbuXfwd9LvbN5Nurt8L+Le/H3/+1MdHh1vHjg9eNFp3dnfZdr16KHBwwfd2t13H6k/au9R7bnZq9zb8ljxcXOfQl/TE4UnTf2K/c1PlZ62DqgMtA3uGbzzTPPZ/ed6zx++MH7xeMhiaHDYfvjliNvI+Eufl9OvQl4tvo57vfbm0ChmNPst9duiMfaxqnci7xrGFcdvT+hN9E3aTr6Z8p6aex/9fv1D+kfKj0WfuD7VTstOt88YzAzMus5+mIuYW5vP+EzzueyL8JfGr1pf+xacFj4sRi1ufjv6neV7zZL8Usey1fLYSujK2o/snyw/L64qr3b/cvz1aS1+Hbd+ekNko+236e/RzdDNzQhSFGn7VQBGLpSfHwDfagCgdAaADsnj8MSd/Gu3wNBW2gGAA6SP0oaV0cwYPJYMJ03mTJ6Gv0vAUpKILdR4mhDaXnoFhjImwBzM0s+myH6CY45LizuXZ5APz68i4CwYLBQq7CaiI8ohuij2ULxYIlhSXYpS6p10g8whWWs5HrnP8jcVDitaK7ErfVCuV0lQ1VbDqz3fU6buoyGu8U2zRWufto4OQeed7h29Ov1ygzzDA0YkY00TZpNF0z6zevNyi0rLdqspG4wtix2rPbUD7LDuuOYMXMhdiW6Ue9F7l90nPQY875FueFV7F/tk+yb6+fvbBegEygeJBvOEsIRShcFhS+GTEQORt6LORx+PORibEdeUgE70Tbq3D+wXPKB60PiQa0ps6vG0wvTkw/KHpzJyj1hlCmRRZIMc1FGaY8LHNXItTjjmueS7nHQqcCi0K7I+ZXHatNiwRKdUo0ylXO6MRIXoWelK06q0c+MXjKsv18zV0tQJXJK5rHZFr968wfGqx7WA6xE34hsPNKU2H27JbM25mdtWeKusvfp2452uuyP3xu8PdzQ88Otk7nzUVfQwvtvv0d4ex17rx6Z9hk+M+u2eRg6cHXz1nOKF1JDusPGI/kvlVwKvia9X30yPvnx7f+zcu7Rx/wn7SYsp8/eWHyw/mnxSmWaaHp/JnpWfHZ+7OJ/02egL2Zfar4ZfpxbOLSZ8c/9uuWS+HLTS8fPIr5YNvc3NXf/LwGh4Bj2OmcIukMHkivgAijLCOFGUKp76IS0LXSL9C0ZZplTmt6wKbBnsA5ysXE7ceTztvKN8y/wrArOCT4TOCUeJaIiSib4QqxAPklCQ+C35UOq4tKMMl8wn2Xq5OHl1BUihSzFbyVKZTnlYpVjVVY1DbRSJAjcNFo0RzVNartqC2ms6Q7rX9I7q+xrsMaQx/GjUblxoEmfqa+ZlHmARbhlq5WVtaaNmK2rHZk90QDmsOH5yGnZ+4FLvWuqWvTfJPdDDyVOPJOXF7A15z/oM+Xb6NflXBxQFpgeFBzuHaIUKhVEikTARMRa5FM0T4xFbHHc//mXCVOJ80uo+iv2cB4QPch/CHnqX0pSamxaV7n7YPsPpSGBmWlZ59pWcpqPNxxqPX8u9cqI270L+2ZOlBYWFuUVZp1JPJxaHl/iXBpYdKr9bIXL2YpXQufzzzy+s1hAvstby1YkicaB0RaNer8H8qvO1kOsZN8413mkabB5rmW793gbfYmoXu612R+uu0j2e+6j7kx3dD5o6a7pKH57oPvwoqSeqN+ZxVl97P+PT/QNvn7E+13xhN+Q3fGjkwsunr5be0I5KvDUbi3h3avzWxLPJsanJ93MfMYj3U2YG52jmpT8rfBH8SvX158LHxZFvvd9vLlUuH1xx+CH0Y+Vn+2rSL7U1wrrexsyu/yWhOVQ57I4WweAwi9gZ3CzZJPkiBZ4gQKlNdKFKob5MM0i7SS/AoM8YxHSYuYKlkbWL7RH7Q45bnJVcCdw63L94zvOa8s7xZfIL8XcIuAusChYISQv1CvuL4ERqRI1EP4lliAuLd0l4SwLJcqk9Ui+lY5G3mwZZM9lpuTR5TvlWBRuFecXDSlxKLchby7TKQVVG1Utq2mrP9njv+aKerIHTKNWU1xzWStLm1G7VsdR5pRugu6lXpW9lQG7wwHCfkbzRrHGViZsps+mwWaG5rQWVRY9lmpWa1ZJ1g02wrZDte7tK+70OLA4vHHOdjJw2nZtcQlz5Xd+6Fe212LviXuAh4NHoqe35mpTgxev1EtlHAnwN/ZT8VQKMA0lBocGkEM1Q6tDRsAvhoREKEeuRD6Kyo61iGGLexFbE+cQLxn9MOJOonziaFJJMn/x83639dw50Hnxw6GZKbWpRWlp6+GHXDP0jopmYzBdZxdkuOfw5a0fHjz05fjP37IkDea75qidZT64WDBdeLzp16tjp/OLKkhulD8tels+eWTtLWcldJXfO6LzbhfDqAzVZF4/WHqojXVK6TLz87crn+tWrhGuc12VvWDUmNzU2/2xVuRnRVnzranvr7Vt3eu4u3zfsuNlp27XcXdQj1/ui71i/54DxM+0XOsMhr4ijc5P9s8tLq1v+3/kfbqtgFQE4mYJkqBkA2GsCkNeJ5JlDSN6JB8CKEgA7FYAS9AMoQh+AVCf+nh8QctpgAQWgAcyACwgBaaCK5MaWwAX4ITlxCsgFZ0A9uAOeggmwhGSO7JAMZAh5QPFQHnQZegR9RGFRwigzVDSqHMnzNpG8Lg6+Cf9GG6JPoicxcphMzDusKrYYu4ZkWL1kSmQ15GzkeXgKfBYFnuIEgZVQQylP2U5UJ7ZRKVPdojaifkMTQ0tNe4VOj26Q3o5+kMGS4RmjB+NPpmJmdeYxlv2sbKxtbO7s5OztHHGc8pzfua5zR/Eo8KzzdvMV8QcI7BEkCo4L3RDOFPES1RYTFCeKr0l8kXwvNSTdJJMsKyM7JpcpryD/VaFVMV8pUdlHxUxVWo1pD1FdUqNUS0z7mE6P7ld9MgMGQxYjdmN+E3lTC7NI89MWnZbfrPlsHG2P23U7oB31nDKc+1wZ3bz21rm/98SSaLywXsveH3xGfWf9qQJMAwuDPoXsCS0I+xJhElkXTYiJjH0db5DQmiSRXL2f+0DpIcaUvDR8esrh5SNBmXPZOUdDjzfl0ZxkLfhcVHvao4SxdKD8WIXh2eWq3PP0FzKrVy4G1367dOKKfgPN1cXrHxunm+daP7VNtS/eZbqv+8C9y7PbtkfzsdQTkaeKg2HPf46gX5OPVryjm7jzgTi9b077c8PXtW+KSwYr+B/HfvauTv/6sPZqvXHjxG+vTent/WPL/zhAALSABfAAUSAH1IERsAOeIBQkgyxQDGrBTfAYvAULEAZihaS3vZ8IFUBXoX7oM4oKJYdyQaWhrqM+wFywB3wenkcrotPRQxgRTApmFPF9KQ7gAnBDZPpkreRS5HV4EfxlCnmKuwQrwhRlApGcWEjFQ3UVyV/f0MTTMtK20DnQfabfz4BnOM0owdjLFM7MxHyPJZCVnvUeWzg7P/soRzGnExcz1yvuch4fXmk+wPeC/5JAuqCbkDySy82K9IneQE6xXIk0yX1SMdLeMlqyBNl+uWx5UwUmhUXFV0rdys0qVapH1ZL2xKlnabRq/tCW0/HRzdGr1m82uGV4y+i2cY/JhBnKXNTCwfKwVYv1vC2/nYd9ucOYE69zkEuzG26vo3uJR5fnIKnDq9Y70yfQ18bPyN85IDXwXjBliFdoezhrRFLk22idmNo4qviIhMdJPMlx+wYOKBw8n8KWWpCOP5ycMZ9JyprMSTomnYs68Tb/WkFckfypb8XXSmPLVc/8OltdJXuu/PynaqGagItX65gulV1Rr/98tfi6yo3+JlLzWmtVm3U7uF171+zeYkdFp9dD1Uc8vejHT57EPcUOZD8jPK8a8hgxfxXypubtp3GuSav3KR/vzDDNnfgiuPDke8HK0VXjNdn1Mxvvfy/u+h8NyAE1svp5gBhQBLrACrgjvt+PrPxK0AgegTFk3RMgQUgL2gslQ6XQbWgCRY54nYQqRA3ADLAvfBvNjj6EnsU4Y55gdbG3ceq4+2RmZG/Jo/FU+KsUDgSY0EIZSZQh/qTqoi6miaV1pjOmN2GwZjRhUmIWYVFg9WBLZI/h8OK047LgNucx5zXjM+e3EfAQjBY6Jlwn8kh0RpxSQknST6pEeliWVc5HvkFhTclK+Ylq1h5nDYzmCa11HVPdNMSDLQbthneM+o3XTE3Nmi0kLS9bS9o02+naDzuGOuNdLrs5uNN4Unh5+Lj6vvdXC8gJ/BhsE9IXZh7+LNI1ajomOY4zfizxYfK9/eUH7Q/9Sq1Md8jgOrKQdTvn6DG/XMM8lvzHBX6FK6fSimlKqsoUy59U+FVCVWXnlS8M1cTWstU9unyw3vCq1HWDxoPNVa25bc7tTLdH7pbed36A67zwUL77Vo9+70hfQr/UADy48Hx6aHAk75XQ6/I3v9/qj2W/ezxBNWk/dfb9zEeZT8HTZ2cezc7OYz6zf5H+qrfguEj65vPdaol3aXn52Ar7St0PlR8lP1Z/Ov5sXmVcjVptXl37pfUr/VfPGnHNdu3U2sA62brWesL6tfWZDZ4N5438jd6Njd8yv31+n/r9+PfvTZlN383Tm31b/o/2k5PdPj4ggg4AmLHNze+CAODyAdjI29xcq9rc3DiHJBujANwL2fm2s33WUANQtvWNB/TylP7HN5b/Arp4xn8K/dDZAAABnGlUWHRYTUw6Y29tLmFkb2JlLnhtcAAAAAAAPHg6eG1wbWV0YSB4bWxuczp4PSJhZG9iZTpuczptZXRhLyIgeDp4bXB0az0iWE1QIENvcmUgNS40LjAiPgogICA8cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjItcmRmLXN5bnRheC1ucyMiPgogICAgICA8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIgogICAgICAgICAgICB4bWxuczpleGlmPSJodHRwOi8vbnMuYWRvYmUuY29tL2V4aWYvMS4wLyI+CiAgICAgICAgIDxleGlmOlBpeGVsWERpbWVuc2lvbj4zNDA8L2V4aWY6UGl4ZWxYRGltZW5zaW9uPgogICAgICAgICA8ZXhpZjpQaXhlbFlEaW1lbnNpb24+NTI8L2V4aWY6UGl4ZWxZRGltZW5zaW9uPgogICAgICA8L3JkZjpEZXNjcmlwdGlvbj4KICAgPC9yZGY6UkRGPgo8L3g6eG1wbWV0YT4KxRPPwQAADjVJREFUeAHtnQlYU1cWx/8JEAJuQAfFFau4IKOiFQEVKy5U64BdtNiquNX2s/20jtbWWjvTVv2cUVtrdZxa96V1KZ1SlVopKrgiKIuKFAUVRIOIgqiQBJLMve8FZLMmGuiLnvt9JG8579xzf/fln3PvfYDMwAqoEAEiQASIwGMTkD+2B3JABIgAESACAgESVLoRiAARIAIWIkCCaiGQ5IYIEAEiQIJK9wARIAJEwEIESFAtBJLcEAEiQARIUOkeIAJEgAhYiAAJqoVAkhsiQASIAAkq3QNEgAgQAQsRIEG1EEhyQwSIABEgQaV7gAgQASJgIQIkqBYCSW6IABEgAiSodA8QASJABCxEgATVQiDJDREgAkTAlhBIg4BWq0VmZiY0Go00AqIoiAARMJsACarZyOrmgoyMDLi6uqJRo0Z1UwF5JQJEoM4J0JC/zhGbVgHPUElMTWNFVkRAqgRIUKXaMxQXESACVkeABNXquowCJgJEQKoESFCl2jMUFxEgAlZHgATV6rqMAiYCRECqBEhQpdozFBcRIAJWR4AE1eq6jAImAkRAqgRIUKXaMxQXESACVkeABNWKuiw/biO8v34LBzKp26yo2yjUp4iAZH5TqjgrFbGZdwT09go5NFq9sN2mx3PwcrazaJfknjuBpFwDZAYbdOnvgza1uL+XHYdDWa4IDGgPZbXaDbiL5Ohk5MptYefihn7ebWvYVLvEIrtF2ZuQjjicLVqJgVBYxKelnZz6YQbeL/DBlslj0crGYJZ7eWEaomKvQOGshFqtFq61b+IGD+/uaGl/35dBrsKxHTsREZ8DmV0nDBs/CoGdmlSp60ZyDMJ37sHFUhk8fEIwMrQ/njEYYJAV4GxMPFQamWCvVN6vy2DvhoAB3eB4v6oqPmmHCDyMgGQE9daFC1ix5B6cmumQlQLc7mGDntd08PpHO3j1a/qwdphxvgTxE9OwuFiJ/i81wTP+eiaolTM+LfOlgO7yRSwer0HgpfaARgfY21SqoxBXjuZibXgxbsoysSyxLXrXA8lmHtPx7t1XEeBqz2KR5qf+smo14gwqFOi4oFZCZsJmEev4T0dPxClm2ylkHPz0Bdi0Zw8aIQSRuTvh08QA+e0kvOvWB9sQhHlrRkEWOwUvek/DWz9fwPKgVkItqeFh6DXuB7hNW4r5HYsxZXwQ/jV+NvZp5sNDlovPhr6EyKFBGKxXITrqDMC2ByEKcb8GIfL2LvgqpMnWBIRk8icTkBlY+ZNjqFK9HhfxpudxzDgzBt0qiVROdAQ+2KRGY8+GaJdXiJ8P6mE7qDnWzh+EDg0qC2IVd7XslGDvqzuwYMJAHA1uU+N81qb1GLPUEYNe0SA6QoYeg4HkKAXWJ41Gx0rx8AuvRm5D6BwZVrFzlWMVnJaqcCgyBZeVzyBoqA/catRU9UBycjI6d+5c9WD5nq4AP66ZiW22rdDWoEb7trMxdYj4JXM3aTsmHdsPPc+yy3IRqXXExIZtkHo3BvHoiR+DVuAF7MfomPUsI3dCyybNcCA/CudlMkz8yydYMDoYLsZMMiduBxYlbsF6raiEE9ymYN7LwWJ2qFNhy7p3EAFb2Di+h5W+ZVgctRT/0e0H7NbiWthw5Kem4NujQ7HSpjs2en0Lz8YlkMnboutzzcpb8tB3m+xtcO60HdHaXfDWGaCXX8an9p7Yu/oMEsI8cGScA8aFi+LYUS/eulxA/cedx+7ceARqwmHvPqaKwEKbhGlN+mDtZ4ehmZANF/dNgv9uOIfRig8xV7sb5duzbu8mQX1oL5HBgwhUk4gHmdXfcXkZywZZEQd89+t17uSCu8mXkXX6Fq6/+ywWhKix7D0VAntdRM4bHvcNTdyy5VlnLaVlUAAWNEvDxx/yTBU41agVlmxphza1kJJpuY9aTrCj+bGH8NE/+V+OysNtdy9M83Rk249YdLnYod6OSOP3RvuCtzEVoqBqrx7Dbt1WoLw5zGZDMatHsD2LD87OxvNt0rCnbI9Y+a3yc8zu5ihsWPUdbk5/BZd2/g29VPtFG2M9G/OisHHlQlyYMQutdIVYU/IrEvhIWb0Hu/eKpsJr6Qqkn8tHYNwcQNDiFExI9a0wiPVUo7eJ42hNScVlwoZNgQqX2JYjG5pDm4bV4cCw7dNQLqbcyCvk7yxf7YNDybfhnrWOkWGZ/NDWgFFwoeiBWav9sO/tQ1C97AItrkOox0GowritxgHIMJcfK2cpnqZXImAyAeNHx2T7P82wQWs/zOiqw/nXvPDTWwEIHDgE33yuh/uaqzXE93GCtG3eAV6O4vwa91OSbcPm8MyfI23QomFFGK6NzBz7Vlxp3FB4YueMEtwIWCUccK503iV4MdZwATCEIGFirrhduhQZE5MQzKahM2+ehsx/Jk42HyRc5dk4GiXTS1A2/BfRi2EMItPysDFHFNO5XueF85oJiZjM/dp9jCUxeWwWxBOxM65hTTka5VZkTi5E9oCDiOn7M/y7hiGhy3ahTqA7vvCIQlSXfdjnk2KymIoBAffY8NtfoYSvry+UbgMR0XE6Fr/UGnomdlxv/TtXnS/VOyjBdfB03nXoHYuh8fNFq2oDL72DI5v5Toeq8+tITYyokYXK9Z5ISvq6ilCXx0PvRMBUArWnV6ZeXa92Ova3QmXowjLVuiwGllHGzLkJu08GImZYAfr2SUHslTKEtjYPlUOX4Th49DbUioZorHhMQTU2mC+g1FYcuHu9E3QNlXDm2zZt0aKB0daY8TnYiDI8s3cAN0aphz9YXoux7OdaUR5O8K9WfRhGBPJpEDbUbuyJmS0HYl3uARxkQgW4svzNAQ58lG0IQuzYkWjBs85ufuwMvwL46+ARCD0H7NZ3R7+h/dFNzo8+Wln5WyycT3yCMaeBBZvniwJobMsVDR+/1FyU09krIS8EnOJUTHhlcKw2z2wHJ7bgpERrz1omYPjxzu5iQx4tZLqKCIgDQylx0NuK4lNTOmygZFlHWaXFIYO9eSJnSjtlbMA46tgExIxkwtKgO46mhJktpkI99y5g55e/YeHywzhfZkrND7cRxOxBZrJCNGCiIJSyB/+R6i/jDwsmdhnHBTHlO+2cjeIk34wzZ8Vr+Wr4/msHBNvAps2E9/svytpXwmXGiRq28FNYKKayd9OTceIUF2TTij1LNfkiVEB/X7z8/nZs7QN85bcAV+Qy8FX44czN/C/ENpR7NCTGgc0EYIR3G7h2CkYW3seerPJUmp1gcSV9cwB3B/avNQPldVIhApYgIJkhv+ZeDk6fy0BGcjbybdnQPjWT/WQg5y4bt/JyLxN7z9jg0slsZJfyRCIPaZGlkF+/haTrxg+yaCmJ1/zjifjvj8U4sjUb+y7wSc1HL+e+D4PD136wjZ4kODmZ25XtO6DX2mjkJ2wQhVG2C8O/24ocRQc2rTsGMyKOiBWy7XXxRRWVpxUNFq61jXzReD4cgzp2xCKnlsL+mzFNMHXLFwhd1gLT+d3BstGxPk1hl3kQXb9yrqjLZ50vHJb7VvFdUYksCi9sCUbflQ5w3eePAUdHIL64ksBVGFbdkKkLcCL+BO5gF07GpbMs0xmvff8LcrEEYxfvRu4tF7wWsxDYORJvrDmCWyxTvZESgVC/d9DE7zuMctfD0W8i+MTIR52m4EA6mystVCFmcQjGHgMWLhtSUeGV39NxIfk0VLiO1OMpOP+76aJf4YQ2iEA1ApZP8apVYOruzagYTP3UKJ7sg7ws7KhwafPlDggf0BKqXadwlK3rPBtxEbMGeGCbxxn88wjLZu3uYNKiJCR95V8vz4Ka2h6njk2Z6RXB3MOl5vDUVD/crkR/lb2yZ8mqldTSIsh0979MlPzZTYaEFzX70mlqHHGX8LnQ8iILw3Cba4gsu4FJLgsw//UhLNs0oM+EBER//zkG3/gGGwvmGRe1ZiMyeB58GjNHqgJkVNFEFg/bVxXx+huL3lmGPGLo/+Cx9xVmG4VE3p1sbnd97xUmzaPeSFiEgHGrBV+Tn++O0nQNxrcJROqO6fAKHYWQpnylfyYy9rFnS18Ygp+mi9V2CV2LhM2vwlFYhHLGxNuJKA3tieHemwUDOzanuzImC5M7iv2gt0nD3B49hayWG7z5vC97CmM2DrHHqlqXL2SJrumVCJhFQHKPTZkV/SMZ//FjU+a4/MPHpsxxxGz/8LEpM31VN5eVqXHyh1HodyMa/+6Thfe6N4XBrrqVuC8vUUNtnDFQONWceKn9qppH5Wy+s4RNddjX1X90YcP4wgI1W6hSwoXNndZWtAWF4ImxwpnPndZmQceIgGUJ8EHdU1dKc1iTPz6MuZ/sRUKxMSs2g4IBOdj12Q5MnVPKfvNGZ9GnDMwIwzRTJjxblzkLYsov+PCYO5SrQnFGWyXdrPDFBYoL6eOIKXfGV+TrTEx5BSwbdnJyeqCYchMupNyGxJTToFIfBCQz5K+Pxop1OKD3Bk/Myy2FBvZoVuW3pEyNoiFa922JqWzBXN+wec2H+k11Ux92THi8OyzC1Ds6uLKhPV9sksuHoT3/VU7K2uqjB6iOp4jAUzjkl2bv1uWQX5otpqiIwJNH4Kkc8j953UgtIgJEQAoESFCl0AsUAxEgAk8EARLUJ6IbqRFEgAhIgQAJqhR6gWIgAkTgiSBAgvpEdCM1gggQASkQIEGVQi+wGBQKBe7cEf9jgURCojCIABEwkwA9NmUmsLoy12q1yMjIAH+nQgSIgHUSIEG1zn6jqIkAEZAgARryS7BTKCQiQASskwAJqnX2G0VNBIiABAmQoEqwUygkIkAErJMACap19htFTQSIgAQJkKBKsFMoJCJABKyTgK3nZvExnfmOu6yzBRQ1ESACREAiBP4P4A1P3j+7rNYAAAAASUVORK5CYII="
    },
    "Screen%20Shot%202018-11-04%20at%209.21.48%20PM.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When you run this cell, you probably notice that the cell temporarily changed to look like the following ![Screen%20Shot%202018-11-04%20at%209.19.43%20PM.png](attachment:Screen%20Shot%202018-11-04%20at%209.19.43%20PM.png) before finally changing to this. ![Screen%20Shot%202018-11-04%20at%209.21.48%20PM.png](attachment:Screen%20Shot%202018-11-04%20at%209.21.48%20PM.png)\n",
    "\n",
    "While a computation is occuring in the background, the status symbol for that line will change to `[*]`. Once the computation is complete, the status will change to `[3]`, where the number in the brackets is the execution counter. Each time a cell is executed, the counter is incremented to be one larger. This is helpful to keep track of the order cells were executed.\n",
    "\n",
    "Now that ROOT is loaded, we will want to open our input and output files, and load a histogram from the input."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "theInfile = ROOT.TFile(\"samples/infile.root\",\"READ\")\n",
    "theOutfile = ROOT.TFile(\"outfile.root\",\"RECREATE\")\n",
    "theHist = theInfile.Get(\"AK8MHist16\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you're familiar with ROOT, you'll notice that this syntax is very similar to the C++ ROOT syntax. In fact, the pyROOT commands have a one-to-one correspondance to the C++ versions (with few exceptions). \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One great advantage to Jupyter is that plots can be displayed directly in the browser, without any additional steps or complicated configuration. Any `TCanvas` that is written to will be showed immediately after the code that produced it. Execute the following cell to see an example Jet Mass plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ROOT.gStyle.SetOptFit(1111)\n",
    "theCanvas = ROOT.TCanvas('theCanvas','')\n",
    "theHist.Draw('hist')\n",
    "theCanvas.Update()\n",
    "theCanvas.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, like the C++ version, we need to write and close any open objects and files. The following cell implements this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "theHist.Write()\n",
    "theCanvas.Write()\n",
    "\n",
    "theOutfile.cd()\n",
    "theOutfile.Write()\n",
    "theOutfile.Close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Great! We just looked at a very simple pyROOT example. Let's see a more complicated example which builds on the first but also illustrates some commonly used histogram manipulations such as; creating, binning, re-binning, scaling, fitting to a Gaussian and saving to various file types.\n",
    "\n",
    "# pyROOT Example Two\n",
    "Like before, we'll want to load our input file and histogram. The `sumw2()` call tells ROOT to properly propagate the $w^2$ for each histogram, which causes the errors to be produced properly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "theInfile = ROOT.TFile(\"samples/infile.root\",\"READ\")\n",
    "\n",
    "histName = \"AK8MPt400To500Hist15\"   \n",
    "theHist = theInfile.Get(histName)\n",
    "# Ensure proper error propogation\n",
    "theHist.Sumw2()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The pyROOT library lets you call any function defined in ROOT. Compare these two calls with their documentation:\n",
    "* `scale()` - [doc](https://root.cern.ch/doc/master/classTH1.html#add929909dcb3745f6a52e9ae0860bfbd)\n",
    "* `rebin()` - [doc](https://root.cern.ch/doc/master/classTH1.html#aff6520fdae026334bf34fa1800946790)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Scale the histogram vertically -- this is useful for things like\n",
    "# scaling a Monte Carlo sample to match the theorertical predictions\n",
    "theHist.Scale(100)\n",
    "\n",
    "# Change the number of bins from the value the histogram was initialized with\n",
    "theHist.Rebin(10)"
   ]
  },
  {
   "attachments": {
    "Screen%20Shot%202018-11-04%20at%2010.19.34%20PM.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once you run the cell, you should notice the following output\n",
    "![Screen%20Shot%202018-11-04%20at%2010.19.34%20PM.png](attachment:Screen%20Shot%202018-11-04%20at%2010.19.34%20PM.png)\n",
    "This is a convenience method to preent you from having to explicitly print the values. Jupyter will print the value of the last line in the cell\n",
    "\n",
    "Next we will want to fit a function to our data. In this case, we will use a regular Gaussian distribution, but other distributions (e.g. Poisson) are also available. In ROOT, we do this via a [TF1](https://root.cern.ch/doc/master/classTF1.html) object. We can call the C++ functions described in the docs via pyROOT."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fitting the histogram to a Gaussian\n",
    "theFitFunc = ROOT.TF1(\"theFitFunc\", \"gaus\", 110, 210)\n",
    "theFitFunc.SetLineColor(1)\n",
    "theFitFunc.SetLineWidth(2)\n",
    "theFitFunc.SetLineStyle(2)\n",
    "# Perform the fit with our histogram we extracted from simulation\n",
    "theHist.Fit(theFitFunc,'S')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once the fit is performed, ROOT provides the values of your chosen equation. Since this is a gaussian, the equation is in the form of\n",
    "\n",
    "$ y = \\frac{c}{\\sqrt{2 \\pi \\sigma^2}} e^{ - \\frac{(x - \\mu)^2}{2 \\sigma} } $\n",
    "\n",
    "where $ \\sigma $ is the sigma (or width), $ \\mu $ is the mean (or peak) value, and c is a scaling constant. The parameters output above are the parameters that cause the gaussian to best fit the data.\n",
    "\n",
    "Let's now plot our data distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ROOT.gStyle.SetOptFit(1111)\n",
    "c = ROOT.TCanvas('c-' + histName,'')\n",
    "\n",
    "theHist.Draw('hist')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To overlay the fit over the data, we do `Draw()` again, but with the `same` parameter. This tells ROOT to draw on top of the old plot, without clearing it first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "theFitFunc.Draw(\"same\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we plot, we want to make sure to change the axes to be sensible values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set the maximum value for the y axis\n",
    "max_y = theHist.GetMaximum() \n",
    "theHist.SetMaximum(max_y * 1.2)\n",
    "theHist.GetXaxis().SetRangeUser(80 , 250)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, let's show the result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "c.Update()\n",
    "c.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that ROOT helpfully puts the fit parameters in the legend.\n",
    "\n",
    "When using pyROOT in Jupyter, plots are displayed inline, but it's sometimes useful to output plots to files. The following two lines will output this plot in both PDF and PNG formats."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "c.Print('plots/'+ histName + '_' + 'GaussianFit.png', 'png' )\n",
    "c.Print('plots/'+ histName + '_' + 'GaussianFit.pdf', 'pdf' )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, it's good practice to close any open files once the script is done. In particular, if any objects were made, it's good to explicitly close (and write) files. This can help prevent file corruption"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "theInfile.Close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "This extremely brief introduction to pyROOT and Jupyter just scratches the surface of what's possible with either. Other DAS exercises will build on these lessons to more advancd use-cases.\n",
    "\n",
    "For CMSDAS@CERN2020 please submit your answer at the [Google Form sixth set](https://docs.google.com/forms/d/e/1FAIpQLSf7lcwh16qamvBDll1tyiSRsZNjSn1oHf5HC6bQ30WK_cHZkQ/viewform).\n",
    "\n",
    "Question 19.1: What is the mean value of the Gaussian fit of the jet mass spectrum for jets of pt 300-400 GeV?"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "cmsdas-preexercise",
   "language": "python",
   "name": "cmsdas-preexercise"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14+"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}