Training_v1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T07:51:15.823587Z",
     "start_time": "2020-04-13T07:51:14.731491Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# mpl.rcParams['figure.figsize'] = (6,6)\n",
    "# mpl.rcParams['figure.dpi'] = 100\n",
    "# mpl.rcParams[\"image.origin\"] = 'lower'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:16.180916Z",
     "start_time": "2020-04-13T10:01:15.799117Z"
    }
   },
   "outputs": [],
   "source": [
    "config = {\n",
    "    \"base_dir\":        \"/eos/home-d/dvalsecc/www/VBSPlots/DNN_archive/\",\n",
    "    \"plot_config\":     \"FullRun2\",\n",
    "    \"cut\":             \"res_sig\",\n",
    "    \"model_version\":   \"v31\",\n",
    "    \"model_tag\":       \"res_5depth_l2reg_v2\",\n",
    "    \"samples_version\": \"v12\",\n",
    "    \"cols\": ['mjj_vbs',\n",
    "             'vbs_0_pt',\n",
    "             'vbs_1_pt',\n",
    "             'vjet_0_pt',\n",
    "             'vjet_1_pt',\n",
    "             'vjet_0_eta',\n",
    "             'vjet_1_eta',\n",
    "             'deltaeta_vbs',\n",
    "             'deltaphi_vbs',  \n",
    "             'Lepton_pt',\n",
    "             'Lepton_eta',\n",
    "             'Zlep',\n",
    "             'whad_pt',\n",
    "             'Asym_vbs',\n",
    "             ],\n",
    "    \n",
    "    \"test_size\": 0.2,\n",
    "    \"val_size\":  0.0,\n",
    "    \"batch_size\": 2048\n",
    "}\n",
    "\n",
    "import os\n",
    "\n",
    "config_base_dir = os.path.join(config[\"base_dir\"], config[\"plot_config\"])\n",
    "\n",
    "# create the model directory\n",
    "model_dir   = os.path.join(config_base_dir, config[\"cut\"] , \"models\",  config[\"model_version\"])\n",
    "os.makedirs(model_dir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T07:59:17.155046Z",
     "start_time": "2020-04-13T07:59:16.806629Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(config[\"cols\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```.python\n",
    "# config examples\n",
    "\"cut\":             \"boos_sig_mjjincl\",\n",
    "\"cut\":             \"res_sig_mjjincl\",\n",
    "\"model_tag\":       \"res_4depth_v0\",\n",
    "\"model_tag\":       \"boost_5vars_v0\",\n",
    " ```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:19.361126Z",
     "start_time": "2020-04-13T10:01:18.672275Z"
    }
   },
   "outputs": [],
   "source": [
    "# load numpy\n",
    "samples_dir = os.path.join(config_base_dir, config[\"cut\"] , \"samples\", config[\"samples_version\"])\n",
    "import pickle\n",
    "signal = pickle.load(open(os.path.join(samples_dir, \"for_training/signal_balanced.pkl\"),     \"rb\"))\n",
    "bkg    = pickle.load(open(os.path.join(samples_dir, \"for_training/background_balanced.pkl\"), \"rb\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Samples preparation for NN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:23.737171Z",
     "start_time": "2020-04-13T10:01:23.419661Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "## using \n",
    "## source /cvmfs/sft.cern.ch/lcg/views/LCG_96python3/x86_64-centos7-gcc8-opt/setup.sh\n",
    "#!pip3 install --user imbalanced-learn==0.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:26.233394Z",
     "start_time": "2020-04-13T10:01:25.831315Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from imblearn.keras import balanced_batch_generator\n",
    "from imblearn.over_sampling import RandomOverSampler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:28.458404Z",
     "start_time": "2020-04-13T10:01:28.109563Z"
    }
   },
   "outputs": [],
   "source": [
    "## instead of reading the variables from file,\n",
    "## we now set the variables here, and then dump them\n",
    "# import yaml\n",
    "# yaml_vars = yaml.safe_load(open(os.path.join(model_dir, \"variables.yml\")))\n",
    "# print(\"yaml: \", type(yaml_vars), len(yaml_vars))\n",
    "# cols = yaml_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:31.216549Z",
     "start_time": "2020-04-13T10:01:30.502864Z"
    }
   },
   "outputs": [],
   "source": [
    "X_sig = signal[config[\"cols\"]].values\n",
    "X_bkg = bkg[config[\"cols\"]].values\n",
    "Y_sig = np.ones(len(X_sig))\n",
    "Y_bkg = np.zeros(len(X_bkg))\n",
    "W_sig = (signal[\"weight_norm\"]).values\n",
    "W_bkg = (bkg[\"weight_norm\"]).values\n",
    "Wnn_sig = (signal[\"weight_\"]).values\n",
    "Wnn_bkg = (bkg[\"weight_\"]).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:33.427720Z",
     "start_time": "2020-04-13T10:01:32.979988Z"
    }
   },
   "outputs": [],
   "source": [
    "X = np.vstack([X_sig, X_bkg])\n",
    "Y = np.hstack([Y_sig, Y_bkg])\n",
    "W = np.hstack([W_sig, W_bkg])\n",
    "Wnn = np.hstack([Wnn_sig, Wnn_bkg])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:36.516940Z",
     "start_time": "2020-04-13T10:01:35.274520Z"
    }
   },
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X)\n",
    "pickle.dump(scaler, open(f\"{model_dir}/scaler_model.pkl\", \"wb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-12T20:35:33.714071Z",
     "start_time": "2020-04-12T20:35:26.952418Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/cvmfs/sft.cern.ch/lcg/views/LCG_96python3/x86_64-centos7-gcc8-opt/lib/python3.6/site-packages/numpy/lib/histograms.py:824: RuntimeWarning: invalid value encountered in greater_equal\n",
      "  keep = (tmp_a >= first_edge)\n",
      "/cvmfs/sft.cern.ch/lcg/views/LCG_96python3/x86_64-centos7-gcc8-opt/lib/python3.6/site-packages/numpy/lib/histograms.py:825: RuntimeWarning: invalid value encountered in less_equal\n",
      "  keep &= (tmp_a <= last_edge)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## draw weights and input variables in order to spot any importing error\n",
    "_ = plt.hist(W, bins=100, range=(0,2))\n",
    "plt.title(\"weights\")\n",
    "plt.show()\n",
    "for idx,col in enumerate(config[\"cols\"]):\n",
    "    _ = plt.hist(X_scaled[:,idx], bins=100)\n",
    "    plt.title(col)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Balancing\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T08:58:59.408510Z",
     "start_time": "2020-04-13T08:58:58.058565Z"
    }
   },
   "outputs": [],
   "source": [
    "X_train, X_temp, y_train, y_temp, W_train, W_temp , Wnn_train, Wnn_temp = train_test_split(X_scaled, Y,  W, Wnn, test_size=config[\"test_size\"], random_state=42, stratify=Y)\n",
    "\n",
    "\n",
    "\n",
    "#X_val,   X_test, y_val,   y_test, W_val,   W_test = train_test_split(X_temp,   y_temp, W_temp, test_size=config[\"val_size\"]) ## test != val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T08:44:55.417564Z",
     "start_time": "2020-04-13T08:44:54.994317Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training   dataset:  (1234705, 12)\n",
      "Test + Val dataset:  (308677, 12)\n"
     ]
    }
   ],
   "source": [
    "print(\"Training   dataset: \", X_train.shape)\n",
    "print(\"Test + Val dataset: \", X_temp.shape)\n",
    "#print(\"Testing    dataset: \", X_test.shape)\n",
    "#print(\"Validation dataset: \", X_val.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create generators to balance signal and background"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-09T10:25:05.109479Z",
     "start_time": "2020-04-09T10:25:04.006069Z"
    }
   },
   "outputs": [],
   "source": [
    "training_generator,   steps_per_epoch_train = balanced_batch_generator(X_train, y_train, W_train, batch_size=config[\"batch_size\"], sampler=RandomOverSampler())\n",
    "#validation_generator, steps_per_epoch_val   = balanced_batch_generator(X_val,   y_val,   W_val,   batch_size=config[\"batch_size\"], sampler=RandomOverSampler()) ## test != val\n",
    "validation_generator, steps_per_epoch_val   = balanced_batch_generator(X_temp,  y_temp,  W_temp,   batch_size=config[\"batch_size\"], sampler=RandomOverSampler()) ## test == val"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Keras model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T08:59:00.752760Z",
     "start_time": "2020-04-13T08:59:00.392363Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T08:59:03.539687Z",
     "start_time": "2020-04-13T08:59:01.758239Z"
    }
   },
   "outputs": [],
   "source": [
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:39.413764Z",
     "start_time": "2020-04-13T10:01:38.722254Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "res_5depth_l2reg_v2\n",
      ">>> Creating model...\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_55 (Dense)             (None, 50)                700       \n",
      "_________________________________________________________________\n",
      "dropout_46 (Dropout)         (None, 50)                0         \n",
      "_________________________________________________________________\n",
      "dense_56 (Dense)             (None, 50)                2550      \n",
      "_________________________________________________________________\n",
      "dropout_47 (Dropout)         (None, 50)                0         \n",
      "_________________________________________________________________\n",
      "dense_57 (Dense)             (None, 50)                2550      \n",
      "_________________________________________________________________\n",
      "dropout_48 (Dropout)         (None, 50)                0         \n",
      "_________________________________________________________________\n",
      "dense_58 (Dense)             (None, 50)                2550      \n",
      "_________________________________________________________________\n",
      "dropout_49 (Dropout)         (None, 50)                0         \n",
      "_________________________________________________________________\n",
      "dense_59 (Dense)             (None, 50)                2550      \n",
      "_________________________________________________________________\n",
      "dropout_50 (Dropout)         (None, 50)                0         \n",
      "_________________________________________________________________\n",
      "dense_60 (Dense)             (None, 1)                 51        \n",
      "=================================================================\n",
      "Total params: 10,951\n",
      "Trainable params: 10,951\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# import local module that programmatically returns keras models\n",
    "import dnn_models\n",
    "\n",
    "print(config[\"model_tag\"])\n",
    "model = dnn_models.get_model(config[\"model_tag\"], X_train.shape[1])\n",
    "\n",
    "model.compile(optimizer=\"sgd\",\n",
    "            loss='binary_crossentropy',\n",
    "            metrics=['accuracy'])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:41.862570Z",
     "start_time": "2020-04-13T10:01:41.426032Z"
    }
   },
   "outputs": [],
   "source": [
    "## import and configure the plot loss callback\n",
    "\n",
    "import dnn_plot_loss\n",
    "\n",
    "data = {\n",
    "    \"X_train\": X_train,\n",
    "    #\"X_test\" : X_test, ## test != val\n",
    "    \"X_val\" : X_temp, ## test == val\n",
    "    \"y_train\": y_train,\n",
    "    #\"y_test\" : y_test, ## test != val\n",
    "    \"y_val\" : y_temp, ## test == val\n",
    "    \"W_train\": W_train,\n",
    "    #\"W_test\" : W_test, ## test != val\n",
    "    \"W_val\": W_temp, ## test == val\n",
    "    \"Wnn_train\": Wnn_train,\n",
    "\n",
    "    #\"W_test\" : W_test, ## test != val\n",
    "    \"Wnn_val\": Wnn_temp, ## test == val\n",
    "}\n",
    "\n",
    "plot_losses = dnn_plot_loss.PlotLosses(model, data, batch_mode=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:01:44.134949Z",
     "start_time": "2020-04-13T10:01:43.770014Z"
    }
   },
   "outputs": [],
   "source": [
    "import math\n",
    "def step_decay(epoch):\n",
    "    initial_lrate = 0.1\n",
    "    drop = 0.65\n",
    "    epochs_drop = 4.0\n",
    "    lrate = initial_lrate * math.pow(drop, math.floor((1+epoch)/epochs_drop))\n",
    "    if lrate < 0.002: return 0.002\n",
    "    return lrate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:32:53.478166Z",
     "start_time": "2020-04-13T10:01:46.120638Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x720 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## TRAINING without random sampling\n",
    "\n",
    "import time\n",
    "start_time = time.time()\n",
    "from keras import callbacks\n",
    "\n",
    "early_stopping = callbacks.EarlyStopping(monitor='val_loss', \n",
    "                            min_delta=0.001, patience=20, verbose=0, mode='auto', \n",
    "                            baseline=None, restore_best_weights=True)\n",
    "\n",
    "lr_scheduler = callbacks.LearningRateScheduler(step_decay, verbose=1)\n",
    "\n",
    "config[\"epochs\"] = 300\n",
    "\n",
    "history = model.fit(\n",
    "            X_train, y_train,\n",
    "            sample_weight = W_train,\n",
    "            epochs=config[\"epochs\"],\n",
    "            validation_data  = (X_temp, y_temp, W_temp),\n",
    "            callbacks=[plot_losses, early_stopping,lr_scheduler], #lr_scheduler\n",
    "            shuffle = True,\n",
    "            batch_size= config[\"batch_size\"]\n",
    "            #callbacks = [], \n",
    "            )\n",
    "\n",
    "config[\"train_time\"] = time.time() - start_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-09T14:07:30.977351Z",
     "start_time": "2020-04-09T14:07:30.665683Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'training_generator' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-112-060b3941404b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     14\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     15\u001b[0m history = model.fit_generator(\n\u001b[0;32m---> 16\u001b[0;31m             \u001b[0mtraining_generator\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     17\u001b[0m             \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"epochs\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m             \u001b[0msteps_per_epoch\u001b[0m  \u001b[0;34m=\u001b[0m \u001b[0msteps_per_epoch_train\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'training_generator' is not defined"
     ]
    }
   ],
   "source": [
    "## TRAINING with generators and random sampling\n",
    "\n",
    "import time\n",
    "start_time = time.time()\n",
    "from keras import callbacks\n",
    "\n",
    "early_stopping = callbacks.EarlyStopping(monitor='val_loss', \n",
    "                            min_delta=0.001, patience=5, verbose=0, mode='auto', \n",
    "                            baseline=None, restore_best_weights=True)\n",
    "\n",
    "lr_scheduler = callbacks.LearningRateScheduler(step_decay, verbose=1)\n",
    "\n",
    "config[\"epochs\"] = 100\n",
    "\n",
    "history = model.fit_generator(\n",
    "            training_generator, \n",
    "            epochs=config[\"epochs\"],\n",
    "            steps_per_epoch  = steps_per_epoch_train, \n",
    "            validation_data  = validation_generator, \n",
    "            validation_steps = steps_per_epoch_val,\n",
    "            callbacks=[plot_losses, early_stopping, lr_scheduler],\n",
    "            #callbacks = [], \n",
    "            )\n",
    "\n",
    "config[\"train_time\"] = time.time() - start_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:00:49.465439Z",
     "start_time": "2020-04-13T10:00:46.539170Z"
    }
   },
   "outputs": [],
   "source": [
    "## SAVE THE MODEL, ITS METADATA AND TRAINING INFORMATIONS\n",
    "\n",
    "# dump the variables list\n",
    "import yaml\n",
    "varfile = os.path.join(model_dir, \"variables.yml\")\n",
    "if os.path.isfile(varfile):\n",
    "    print(\"ACHTUNG! variables file already existing: old file renamed with '_old'\")\n",
    "    os.rename(varfile, varfile[:-4] + \"_old.yml\")\n",
    "with open(varfile, \"w\") as out_var_file:\n",
    "    out_var_file.write(yaml.dump(config[\"cols\"]))\n",
    "    \n",
    "# dump the config\n",
    "config[\"a__model_dir\"] = model_dir\n",
    "model_config_file = os.path.join(model_dir, \"model_config.yml\")\n",
    "if os.path.isfile(model_config_file):\n",
    "    print(\"ACHTUNG! model_config_file file already existing: old file renamed with '_old'\")\n",
    "    os.rename(model_config_file, model_config_file[:-4] + \"_old.yml\")\n",
    "with open(model_config_file, \"w\") as out_var_file:\n",
    "    out_var_file.write(yaml.dump(config))  \n",
    "\n",
    "# save figure with training summary\n",
    "plot_losses.save_figure( os.path.join(model_dir, \"model_train.png\") )\n",
    "\n",
    "# save keras model\n",
    "model.save( os.path.join(model_dir, \"model.h5\") )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-13T10:00:56.146000Z",
     "start_time": "2020-04-13T10:00:50.736175Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0413 12:00:55.513631 140348349679424 deprecation.py:323] From <ipython-input-106-39c8d219d3f3>:53: convert_variables_to_constants (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use `tf.compat.v1.graph_util.convert_variables_to_constants`\n",
      "W0413 12:00:55.515126 140348349679424 deprecation.py:323] From /cvmfs/sft.cern.ch/lcg/views/LCG_96python3/x86_64-centos7-gcc8-opt/lib/python3.6/site-packages/tensorflow/python/framework/graph_util_impl.py:270: extract_sub_graph (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use `tf.compat.v1.graph_util.extract_sub_graph`\n"
     ]
    }
   ],
   "source": [
    "## Export:\n",
    "## * keras model to tensorflow model\n",
    "## * tf metadata\n",
    "## * scaler\n",
    "\n",
    "args = {\n",
    "    \"dir\": model_dir,\n",
    "    \"input\": \"model.h5\",\n",
    "    \"output\": \"model.pb\",\n",
    "    \"tf_metadata\": \"tf_metadata.txt\",\n",
    "    \"input_scaler\": \"scaler_model.pkl\",\n",
    "    \"output_scaler\": \"scaler.txt\",\n",
    "}\n",
    "\n",
    "\n",
    "from keras import backend as K\n",
    "# This line must be executed before loading Keras model.\n",
    "K.set_learning_phase(0)\n",
    "\n",
    "from keras.models import load_model\n",
    "model = load_model(os.path.join(args[\"dir\"], args[\"input\"]))\n",
    "\n",
    "from keras import backend as K\n",
    "import tensorflow as tf\n",
    "\n",
    "def freeze_session(session, keep_var_names=None, output_names=None, clear_devices=True):\n",
    "    \"\"\"\n",
    "    Freezes the state of a session into a pruned computation graph.\n",
    "\n",
    "    Creates a new computation graph where variable nodes are replaced by\n",
    "    constants taking their current value in the session. The new graph will be\n",
    "    pruned so subgraphs that are not necessary to compute the requested\n",
    "    outputs are removed.\n",
    "    @param session The TensorFlow session to be frozen.\n",
    "    @param keep_var_names A list of variable names that should not be frozen,\n",
    "                          or None to freeze all the variables in the graph.\n",
    "    @param output_names Names of the relevant graph outputs.\n",
    "    @param clear_devices Remove the device directives from the graph for better portability.\n",
    "    @return The frozen graph definition.\n",
    "    \"\"\"\n",
    "    from tensorflow.python.framework.graph_util import convert_variables_to_constants\n",
    "    graph = session.graph\n",
    "    with graph.as_default():\n",
    "        freeze_var_names = list(set(v.op.name for v in tf.global_variables()).difference(keep_var_names or []))\n",
    "        output_names = output_names or []\n",
    "        output_names += [v.op.name for v in tf.global_variables()]\n",
    "        # Graph -> GraphDef ProtoBuf\n",
    "        input_graph_def = graph.as_graph_def()\n",
    "        if clear_devices:\n",
    "            for node in input_graph_def.node:\n",
    "                node.device = \"\"\n",
    "        frozen_graph = convert_variables_to_constants(session, input_graph_def,\n",
    "                                                      output_names, freeze_var_names)\n",
    "        return frozen_graph\n",
    "\n",
    "\n",
    "frozen_graph = freeze_session(K.get_session(),\n",
    "                              output_names=[out.op.name for out in model.outputs])\n",
    "\n",
    "# Save to ./model/tf_model.pb\n",
    "tf.train.write_graph(frozen_graph, args[\"dir\"], args[\"output\"], as_text=False)\n",
    "\n",
    "## save tensorflow metadata\n",
    "with open(os.path.join(args[\"dir\"], args[\"tf_metadata\"]), \"w\") as f:\n",
    "    f.write(str(model.inputs[0].name) + \" \" + str(model.outputs[0].name) + \"\\n\")\n",
    "\n",
    "\n",
    "## Export \n",
    "## * tf tensor input name and output name\n",
    "## * scaler mean_ and scale_ (where scale_ = np.sqrt(var_)) for each variable\n",
    "##\n",
    "import os\n",
    "import pickle\n",
    "\n",
    "import yaml\n",
    "yaml_vars = yaml.safe_load(open(os.path.join(args[\"dir\"], \"variables.yml\"), \"r\"))\n",
    "\n",
    "scaler = pickle.load(open(os.path.join(args[\"dir\"], args[\"input_scaler\"]), 'rb'))\n",
    "with open(os.path.join(args[\"dir\"], args[\"output_scaler\"]), \"w\") as f:\n",
    "    for var, mean, scale in zip(yaml_vars, scaler.mean_, scaler.scale_):\n",
    "        f.write(var + \" \" + str(mean) + \" \" + str(scale) + \"\\n\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Retrieve graphs history\n",
    "## plot \n",
    "# n = 30\n",
    "# plt.hist(plot_losses.dnn_score_plot[n][0][1][:-1], bins=plot_losses.dnn_score_plot[n][0][1], weights=plot_losses.dnn_score_plot[n][0][0], histtype=\"step\")\n",
    "# plt.hist(plot_losses.dnn_score_plot[n][1][1][:-1], bins=plot_losses.dnn_score_plot[n][1][1], weights=plot_losses.dnn_score_plot[n][1][0], histtype=\"step\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Columns List"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```.python\n",
    "# resolved\n",
    "    \"cols\": ['mjj_vbs', \n",
    "             'vbs_0_pt', \n",
    "             'vbs_1_pt', \n",
    "             'deltaeta_vbs',\n",
    "             'deltaphi_vbs', \n",
    "             'mjj_vjet', \n",
    "             'vjet_0_pt', \n",
    "             'vjet_1_pt', \n",
    "             'vjet_0_eta', \n",
    "             'vjet_1_eta', \n",
    "             'Lepton_pt', \n",
    "             'Lepton_eta', \n",
    "             'Lepton_flavour', \n",
    "             'PuppiMET', \n",
    "             'Zvjets_0', \n",
    "             'Zlep', \n",
    "             'Asym_vbs', \n",
    "             'Asym_vjet', \n",
    "             'A_ww', \n",
    "             'Mtw_lep', \n",
    "             'w_lep_pt', \n",
    "             'Mww', \n",
    "             'R_ww', \n",
    "             'R_mw', \n",
    "             'Centr_vbs', \n",
    "             'Centr_ww'\n",
    "         ]\n",
    "\n",
    "# boosted\n",
    "    \"cols\": ['mjj_vbs', \n",
    "             'vbs_0_pt', \n",
    "             'vbs_1_pt', \n",
    "             'deltaeta_vbs',\n",
    "             'deltaphi_vbs', \n",
    "             'mjj_vjet', \n",
    "             'vjet_0_pt', \n",
    "             'vjet_1_pt', \n",
    "             'vjet_0_eta', \n",
    "             'vjet_1_eta', \n",
    "             'Lepton_pt', \n",
    "             'Lepton_eta', \n",
    "             'Lepton_flavour', \n",
    "             'PuppiMET', \n",
    "             'Zvjets_0', \n",
    "             'Zlep', \n",
    "             'Asym_vbs', \n",
    "             'Asym_vjet', \n",
    "             'A_ww', \n",
    "             'Mtw_lep', \n",
    "             'w_lep_pt', \n",
    "             'Mww', \n",
    "             'R_ww', \n",
    "             'R_mw', \n",
    "             'Centr_vbs', \n",
    "             'Centr_ww'\n",
    "         ]\n",
    "\n",
    "         \n",
    "# OLD\n",
    "    \"cols\": ['mjj_vbs', 'deltaeta_vbs', \n",
    "            'mjj_vjet', \n",
    "            'Lepton_pt', 'Lepton_eta' ]\n",
    "    \"cols\": ['mjj_vbs', \n",
    "             'vbs_0_pt', 'vbs_1_pt', \n",
    "             'deltaeta_vbs', 'deltaphi_vbs', \n",
    "             'mjj_vjet', \n",
    "             'vjet_0_pt', 'vjet_1_pt', \n",
    "             'vjet_0_eta', 'vjet_1_eta', \n",
    "             'Lepton_pt', 'Lepton_eta', 'Lepton_flavour', \n",
    "         ]\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.6.9"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "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()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}