{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4d50c9fa-0422-4c70-b5c5-49a90edbffe7",
   "metadata": {},
   "source": [
    "# XmlObjects的基本用法\n",
    "\n",
    "@Author: 吴炜坤\n",
    "\n",
    "@E-mail: weikun.wu@xtalpi.com"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71494982-1f97-412d-87ed-3a7150d91cb3",
   "metadata": {},
   "source": [
    "### 1. 为什么有XmlObject？\n",
    "在最早期，Rosetta的mover、filter等都没有为python api流程参数控制的接口，导致部分的组件无法直接调用，只能通过xml脚本读取参数，然后生成对应的对象。\n",
    "\n",
    "XmlObjects是在PyRosetta中直接使用xml脚本最直接的方法，特别是某些Mover没有做好Pyrosetta接口时特别有用。但是缺点是加载XmlObjects的速度并不理想，比纯粹的PyRosetta脚本启动要慢。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06da9647-eda0-4cbc-80a5-fb0ce6704f13",
   "metadata": {},
   "source": [
    "### 2. 如何使用XmlObject?\n",
    "最方便的方法有两种调用方式:\n",
    "1. create_from_string\n",
    "2. create_from_file"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "647d98fd-e38e-4c68-8843-8cd41ec39f26",
   "metadata": {},
   "source": [
    "#### 2.1 create_from_string\n",
    "从字符串文本中提取信息，返回XmlObjects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0743b33d-8a9b-441a-9d3b-db8f052fe7cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PyRosetta-4 2021 [Rosetta PyRosetta4.conda.mac.cxx11thread.serialization.python37.Release 2021.26+release.b308454c455dd04f6824cc8b23e54bbb9be2cdd7 2021-07-02T13:01:54] retrieved from: http://www.pyrosetta.org\n",
      "(C) Copyright Rosetta Commons Member Institutions. Created in JHU by Sergey Lyskov and PyRosetta Team.\n",
      "\u001b[0mcore.init: {0} \u001b[0mChecking for fconfig files in pwd and ./rosetta/flags\n",
      "\u001b[0mcore.init: {0} \u001b[0mRosetta version: PyRosetta4.conda.mac.cxx11thread.serialization.python37.Release r288 2021.26+release.b308454c455 b308454c455dd04f6824cc8b23e54bbb9be2cdd7 http://www.pyrosetta.org 2021-07-02T13:01:54\n",
      "\u001b[0mcore.init: {0} \u001b[0mcommand: PyRosetta -ex1 -ex2aro -database /opt/miniconda3/lib/python3.7/site-packages/pyrosetta/database\n",
      "\u001b[0mbasic.random.init_random_generator: {0} \u001b[0m'RNG device' seed mode, using '/dev/urandom', seed=-2025690693 seed_offset=0 real_seed=-2025690693 thread_index=0\n",
      "\u001b[0mbasic.random.init_random_generator: {0} \u001b[0mRandomGenerator:init: Normal mode, seed=-2025690693 RG_type=mt19937\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mGenerating XML Schema for rosetta_scripts...\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0m...done\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mInitializing schema validator...\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0m...done\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mValidating input script...\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0m...done\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mParsed script:\n",
      "<ROSETTASCRIPTS>\n",
      "\t<SCOREFXNS>\n",
      "\t\t<ScoreFunction name=\"sfxn1\" weights=\"ref2015\"/>\n",
      "\t</SCOREFXNS>\n",
      "\t<RESIDUE_SELECTORS/>\n",
      "\t<TASKOPERATIONS/>\n",
      "\t<FILTERS/>\n",
      "\t<MOVERS>\n",
      "\t\t<SetTorsion name=\"setTorsion\">\n",
      "\t\t\t<Torsion angle=\"perturb\" perturbation_magnitude=\"1.0\" perturbation_type=\"gaussian\" residue=\"1,2,3,4,5\" torsion_name=\"rama\"/>\n",
      "\t\t</SetTorsion>\n",
      "\t</MOVERS>\n",
      "\t<APPLY_TO_POSE/>\n",
      "\t<PROTOCOLS/>\n",
      "\t<OUTPUT/>\n",
      "</ROSETTASCRIPTS>\n",
      "\u001b[0mcore.scoring.ScoreFunctionFactory: {0} \u001b[0mSCOREFUNCTION: \u001b[32mref2015\u001b[0m\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0mStarting energy table calculation\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0msmooth_etable: changing atr/rep split to bottom of energy well\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0msmooth_etable: spline smoothing lj etables (maxdis = 6)\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0msmooth_etable: spline smoothing solvation etables (max_dis = 6)\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0mFinished calculating energy tables.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/hbonds/ref2015_params/HBPoly1D.csv\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/hbonds/ref2015_params/HBFadeIntervals.csv\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/hbonds/ref2015_params/HBEval.csv\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/hbonds/ref2015_params/DonStrength.csv\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/hbonds/ref2015_params/AccStrength.csv\n",
      "\u001b[0mcore.chemical.GlobalResidueTypeSet: {0} \u001b[0mFinished initializing fa_standard residue type set.  Created 984 residue types\n",
      "\u001b[0mcore.chemical.GlobalResidueTypeSet: {0} \u001b[0mTotal time to initialize 0.682139 seconds.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/fd/all.ramaProb\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/fd/prepro.ramaProb\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/omega/omega_ppdep.all.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/omega/omega_ppdep.gly.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/omega/omega_ppdep.pro.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/omega/omega_ppdep.valile.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/P_AA_pp/P_AA\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/P_AA_pp/P_AA_n\n",
      "\u001b[0mcore.scoring.P_AA: {0} \u001b[0mshapovalov_lib::shap_p_aa_pp_smooth_level of 1( aka low_smooth ) got activated.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/P_AA_pp/shapovalov/10deg/kappa131/a20.prop\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0mStarting energy table calculation\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0msmooth_etable: changing atr/rep split to bottom of energy well\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0msmooth_etable: spline smoothing lj etables (maxdis = 6)\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0msmooth_etable: spline smoothing solvation etables (max_dis = 6)\n",
      "\u001b[0mcore.scoring.etable: {0} \u001b[0mFinished calculating energy tables.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/PairEPotential/pdb_pair_stats_fine\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/InterchainPotential/interchain_env_log.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/InterchainPotential/interchain_pair_log.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/EnvPairPotential/env_log.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/EnvPairPotential/cbeta_den.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/EnvPairPotential/pair_log.txt\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/EnvPairPotential/cenpack_log.txt\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mshapovalov_lib::shap_rama_smooth_level of 4( aka highest_smooth ) got activated.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/shapovalov/kappa25/all.ramaProb\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/avg_L_rama.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/avg_L_rama.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/sym_all_rama.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/sym_all_rama.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/sym_G_rama.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/sym_G_rama.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/sym_P_rama.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/sym_P_rama.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/avg_L_rama_str.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/avg_L_rama_str.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/sym_all_rama_str.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/sym_all_rama_str.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/sym_G_rama_str.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/sym_G_rama_str.dat.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/rama/flat/sym_P_rama_str.dat\n",
      "\u001b[0mcore.scoring.ramachandran: {0} \u001b[0mReading custom Ramachandran table from scoring/score_functions/rama/flat/sym_P_rama_str.dat.\n",
      "\u001b[0mprotocols.jd2.parser.ScoreFunctionLoader: {0} \u001b[0mdefined score function \"sfxn1\" with weights \"ref2015\"\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mDefined mover named \"setTorsion\" of type SetTorsion\n",
      "\u001b[0mprotocols.rosetta_scripts.ParsedProtocol: {0} \u001b[0mParsedProtocol mover with the following settings\n"
     ]
    }
   ],
   "source": [
    "from pyrosetta.rosetta.protocols import rosetta_scripts \n",
    "from pyrosetta import init\n",
    "\n",
    "# 初始化脚本:\n",
    "init()\n",
    "xml = rosetta_scripts.XmlObjects.create_from_string('''\n",
    "<SCOREFXNS>\n",
    "  <ScoreFunction name=\"sfxn1\" weights=\"ref2015\"/>\n",
    "</SCOREFXNS>\n",
    "\n",
    "<RESIDUE_SELECTORS>\n",
    "</RESIDUE_SELECTORS>\n",
    "\n",
    "<TASKOPERATIONS>\n",
    "</TASKOPERATIONS>\n",
    "\n",
    "<FILTERS>\n",
    "</FILTERS>\n",
    "\n",
    "<MOVERS> \n",
    "\t<SetTorsion name=\"setTorsion\"> \n",
    "\t\t<Torsion residue=\"1,2,3,4,5\" torsion_name=\"rama\" angle=\"perturb\" perturbation_type=\"gaussian\" perturbation_magnitude=\"1.0\" /> \n",
    "\t\t</SetTorsion> \n",
    "</MOVERS>\n",
    "\n",
    "<APPLY_TO_POSE>\n",
    "</APPLY_TO_POSE>\n",
    "\n",
    "<PROTOCOLS>\n",
    "</PROTOCOLS>\n",
    "\n",
    "<OUTPUT />\n",
    "''')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "837479ff-3402-4018-b646-0b94dcf4902e",
   "metadata": {},
   "source": [
    "**提取Filter、Mover、Selector、SimpleMetric、TaskOperation的语法语句**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5f7fa7a8-13fb-4286-99f9-848277443082",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<pyrosetta.rosetta.protocols.simple_moves.SetTorsion at 0x7fc5f0380bf0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get a mover:\n",
    "mover_ = xml.get_mover(\"setTorsion\")\n",
    "mover_\n",
    "\n",
    "# 如此类推:\n",
    "# filter_ = xml.get_filter(name)\n",
    "# selector_ = xml.get_residue_selector(name)\n",
    "# score_ = xml.get_score_function(name)\n",
    "# sm_ = xml.get_simple_metric(name)\n",
    "# tf_ = xml.get_task_operation(name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ea7c9597-8eaf-4964-9404-f3f5001cc26c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "vector1_std_string[ParsedProtocol, null, setTorsion]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#列表式所选内容。\n",
    "xml.list_movers()\n",
    "\n",
    "# 如此类推:\n",
    "# xml.list_filters()\n",
    "# xml.list_residue_selectors()\n",
    "# xml.list_score_functions()\n",
    "# xml.list_simple_metrics()\n",
    "# xml.list_task_operations()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "707259f6-6bef-416f-9fdf-20ea673f425a",
   "metadata": {},
   "source": [
    "#### 2.2 create_from_file\n",
    "从字符串文本中提取信息，返回XmlObjects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b4e9c8e4-7d98-4a65-94a0-6ff971cfd88e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mGenerating XML Schema for rosetta_scripts...\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0m...done\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mInitializing schema validator...\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0m...done\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mValidating input script...\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0m...done\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mParsed script:\n",
      "<ROSETTASCRIPTS>\n",
      "\t<SCOREFXNS>\n",
      "\t\t<ScoreFunction name=\"molmech\" weights=\"mm_std_fa_elec_dslf_fa13\"/>\n",
      "\t\t<ScoreFunction name=\"r15_cart\" weights=\"ref2015\">\n",
      "\t\t\t<Reweight scoretype=\"pro_close\" weight=\"0.0\"/>\n",
      "\t\t\t<Reweight scoretype=\"cart_bonded\" weight=\"0.625\"/>\n",
      "\t\t</ScoreFunction>\n",
      "\t</SCOREFXNS>\n",
      "\t<RESIDUE_SELECTORS/>\n",
      "\t<TASKOPERATIONS/>\n",
      "\t<FILTERS/>\n",
      "\t<MOVERS>\n",
      "\t\t<MinMover bb=\"1\" cartesian=\"F\" chi=\"true\" name=\"min_torsion\" scorefxn=\"molmech\"/>\n",
      "\t\t<MinMover bb=\"1\" cartesian=\"T\" chi=\"true\" name=\"min_cart\" scorefxn=\"r15_cart\"/>\n",
      "\t</MOVERS>\n",
      "\t<APPLY_TO_POSE/>\n",
      "\t<PROTOCOLS>\n",
      "\t\t<Add mover=\"min_cart\"/>\n",
      "\t</PROTOCOLS>\n",
      "\t<OUTPUT scorefxn=\"r15_cart\"/>\n",
      "</ROSETTASCRIPTS>\n",
      "\u001b[0mcore.scoring.ScoreFunctionFactory: {0} \u001b[0mSCOREFUNCTION: \u001b[32mref2015\u001b[0m\n",
      "\u001b[0mcore.mm.MMLJLibrary: {0} \u001b[0mMM lj sets added: 105\n",
      "\u001b[0mcore.mm.MMTorsionLibrary: {0} \u001b[0mMM torsion sets added fully assigned: 1039; wildcard: 48 and 1 virtual parameter.\n",
      "\u001b[0mprotocols.jd2.parser.ScoreFunctionLoader: {0} \u001b[0mdefined score function \"molmech\" with weights \"mm_std_fa_elec_dslf_fa13\"\n",
      "\u001b[0mprotocols.jd2.parser.ScoreFunctionLoader: {0} \u001b[0mdefined score function \"r15_cart\" with weights \"ref2015\"\n",
      "\u001b[0mprotocols.jd2.parser.ScoreFunctionLoader: {0} \u001b[0m setting r15_cart weight pro_close to 0\n",
      "\u001b[0mprotocols.jd2.parser.ScoreFunctionLoader: {0} \u001b[0m setting r15_cart weight cart_bonded to 0.625\n",
      "\u001b[0mcore.energy_methods.CartesianBondedEnergy: {0} \u001b[0mInitializing IdealParametersDatabase with default Ks=300 , 80 , 80 , 10 , 80\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/bondlength_bondangle/default-lengths.txt\n",
      "\u001b[0mcore.energy_methods.CartesianBondedEnergy: {0} \u001b[0mRead 759 bb-independent lengths.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/bondlength_bondangle/default-angles.txt\n",
      "\u001b[0mcore.energy_methods.CartesianBondedEnergy: {0} \u001b[0mRead 1434 bb-independent angles.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/bondlength_bondangle/default-torsions.txt\n",
      "\u001b[0mcore.energy_methods.CartesianBondedEnergy: {0} \u001b[0mRead 1 bb-independent torsions.\n",
      "\u001b[0mbasic.io.database: {0} \u001b[0mDatabase file opened: scoring/score_functions/bondlength_bondangle/default-improper.txt\n",
      "\u001b[0mcore.energy_methods.CartesianBondedEnergy: {0} \u001b[0mRead 529 bb-independent improper tors.\n",
      "\u001b[0mprotocols.minimization_packing.MinMover: {0} \u001b[0mOptions chi, bb: 1, 1 omega: 1\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mDefined mover named \"min_torsion\" of type MinMover\n",
      "\u001b[0mprotocols.minimization_packing.MinMover: {0} \u001b[0mOptions chi, bb: 1, 1 omega: 1\n",
      "\u001b[0mprotocols.rosetta_scripts.RosettaScriptsParser: {0} \u001b[0mDefined mover named \"min_cart\" of type MinMover\n",
      "\u001b[0mprotocols.rosetta_scripts.ParsedProtocol: {0} \u001b[0mParsedProtocol mover with the following settings\n",
      "\u001b[0mprotocols.rosetta_scripts.ParsedProtocol: {0} \u001b[0mAdded mover \"min_cart\"\n"
     ]
    }
   ],
   "source": [
    "from pyrosetta.rosetta.protocols import rosetta_scripts \n",
    "xml = rosetta_scripts.XmlObjects.create_from_file('./data/Example2-MinMover.xml')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4ab1ba4-c717-4a15-9b5c-579f02bd225c",
   "metadata": {},
   "source": [
    "### 3. RosettaScript所有的API\n",
    "\n",
    "详见: https://new.rosettacommons.org/docs/latest/scripting_documentation/RosettaScripts/RosettaScripts"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c7b732c-36bd-45d7-9075-bf7614dcb043",
   "metadata": {},
   "source": [
    "#### 结语: \n",
    "RosettaScript大而全，Python API灵活速度快！"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6db9e0f5-c610-45eb-840d-499887a4ae37",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}