TMVA_Keras_Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://oproject.org/img/ROOT.png\" height=\"30%\" width=\"30%\">\n",
    "<img src=\"http://oproject.org/img/tmvalogo.png\" height=\"30%\" width=\"30%\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr style=\"border-top-width: 4px; border-top-color: #34609b;\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ROOT import TMVA, TFile, TTree, TCut\n",
    "from subprocess import call\n",
    "from os.path import isfile\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "TMVA.Tools.Instance()\n",
    "TMVA.PyMethodBase.PyInitialize()\n",
    "\n",
    "output = TFile.Open('TMVA.root', 'RECREATE')\n",
    "factory = TMVA.Factory('TMVAClassification', output,\n",
    "                       '!V:!Silent:Color:!DrawProgressBar:Transformations=D,G:AnalysisType=Classification')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not isfile('tmva_class_example.root'):\n",
    "    call(['curl', '-O', 'http://root.cern.ch/files/tmva_class_example.root'])\n",
    "\n",
    "data = TFile.Open('tmva_class_example.root')\n",
    "signal = data.Get('TreeS')\n",
    "background = data.Get('TreeB')\n",
    "\n",
    "dataloader = TMVA.DataLoader('dataset')\n",
    "for branch in signal.GetListOfBranches():\n",
    "    dataloader.AddVariable(branch.GetName())\n",
    "\n",
    "dataloader.AddSignalTree(signal, 1.0)\n",
    "dataloader.AddBackgroundTree(background, 1.0)\n",
    "dataloader.PrepareTrainingAndTestTree(TCut(''),\n",
    "                                      'nTrain_Signal=4000:nTrain_Background=4000:SplitMode=Random:NormMode=NumEvents:!V')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(64, activation='relu', input_dim=4))\n",
    "model.add(Dense(2, activation='softmax'))\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='categorical_crossentropy',\n",
    "              optimizer=\"adam\", metrics=['accuracy', ])\n",
    "\n",
    "model.save(\"model.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "factory.BookMethod(dataloader, TMVA.Types.kFisher, 'Fisher',\n",
    "                   '!H:!V:Fisher:VarTransform=D,G')\n",
    "factory.BookMethod(dataloader, TMVA.Types.kBDT, 'BDT',\n",
    "                   '!H:!V:NTrees=500:MaxDepth=3:BoostType=AdaBoost:nCuts=20')\n",
    "factory.BookMethod(dataloader, TMVA.Types.kPyKeras, 'PyKeras',\n",
    "                   'H:!V:VarTransform=N:FilenameModel=model.h5:NumEpochs=10:BatchSize=32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "factory.TrainAllMethods()\n",
    "factory.TestAllMethods()\n",
    "factory.EvaluateAllMethods()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%jsroot on\n",
    "canvas = factory.GetROCCurve(dataloader)\n",
    "canvas.Draw()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}