Experiments/exp_ABRUPT_ENV_NEW.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import os\n",
    "os.chdir(\"..\")\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "import numpy as np\n",
    "import os\n",
    "from math import log\n",
    "import random\n",
    "import time\n",
    "import pickle\n",
    "from numpy.linalg import pinv\n",
    "from arm_class import ArmGaussian\n",
    "from LinUCB_class import PolicyLinUCB\n",
    "from dLinUCB_class import DynamicLinUCB\n",
    "from D_LinUCB_class import DLinUCB\n",
    "from SW_LinUCB_class import SWLinUCB\n",
    "from environment_class import Environment\n",
    "from simulator_class import Simulator\n",
    "from utils import plot_regret, scatter_abrupt, scatter_smooth, action_check, get_B_T_smooth, detection_sorted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# General parameters for all the experiments\n",
    "delta = 0.01 # Probability of being outside the confidence interval\n",
    "s = 1  # Bound on the theta_star parameter\n",
    "lambda_ = 0.1 # Regularisation parameter\n",
    "q = 5 # Diplaying the quantile (in %)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def experiment_policies(n_mc, option):\n",
    "    mab = Environment(d, theta, sigma_noise, verbose)\n",
    "    simulator = Simulator(mab, theta, policies, k, d, steps, bp, verbose)\n",
    "    if option == \"par\":\n",
    "        avgRegret, qRegret, QRegret = simulator.run_multiprocessing(n_process, steps, n_mc, q, t_saved)\n",
    "        return avgRegret, qRegret, QRegret\n",
    "    else:\n",
    "        avgRegret, qRegret, QRegret, timedic, theta_hat, action_check = simulator.run(steps, \n",
    "                                                               n_mc, q, n_scat, t_saved)\n",
    "        return avgRegret, qRegret, QRegret, timedic, theta_hat, action_check\n",
    "\n",
    "def data_from_experiment(n_mc, option):\n",
    "    if option == \"par\":\n",
    "        avgRegret, qRegret, QRegret = experiment_policies(n_mc, option)\n",
    "        data = [[policy, avgRegret[policy], qRegret[policy],\n",
    "                QRegret[policy]] for policy in avgRegret]\n",
    "        return data\n",
    "    else:\n",
    "        avgRegret, qRegret, QRegret, timedic, theta_hat, action_check = experiment_policies(n_mc, '')\n",
    "        data = [[policy, avgRegret[policy], qRegret[policy],\n",
    "                QRegret[policy]] for policy in avgRegret]\n",
    "        return data, theta_hat, action_check, timedic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "out_ABRUPT_NEW already exists\n",
      "out_ABRUPT_NEW already exists\n",
      "detection_out already exists\n"
     ]
    }
   ],
   "source": [
    "# The saved files should be located in the saved/ folder\n",
    "# The output images would be located in the out/ folder\n",
    "# If the folders are not created they must be created to run the code without bugs\n",
    "\n",
    "path = os.getcwd()\n",
    "out_dir = 'out_ABRUPT_NEW'\n",
    "saved_data_path = 'out_ABRUPT_NEW'\n",
    "detection_folder = 'detection_out'\n",
    "\n",
    "if not os.path.exists(path + '/' + out_dir):\n",
    "    os.mkdir(path + '/' + out_dir)\n",
    "    print('Creating the folder %s' %out_dir)\n",
    "else:\n",
    "    print(\"%s already exists\" %out_dir)\n",
    "    \n",
    "if not os.path.exists(path + '/' + saved_data_path):\n",
    "    os.mkdir(path + '/' + saved_data_path)\n",
    "    print('Creating the folder %s' %saved_data_path)\n",
    "else:\n",
    "    print(\"%s already exists\" %saved_data_path)\n",
    "    \n",
    "if not os.path.exists(path + '/' + detection_folder):\n",
    "    os.mkdir(path + '/' + detection_folder)\n",
    "    print('Creating the folder %s' %detection_folder)\n",
    "else:\n",
    "    print(\"%s already exists\" %detection_folder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_file(filename, var):\n",
    "    with open(saved_data_path + '/' + str(filename) + '.pkl', 'wb') as f:\n",
    "        pickle.dump(var, f)\n",
    "\n",
    "def load_file(filename):\n",
    "    with open(saved_data_path + '/' + str(filename) + '.pkl', 'rb') as f:\n",
    "        res = pickle.load(f)\n",
    "        return res\n",
    "\n",
    "def save_file_from_folder(folder, filename, var):\n",
    "    with open(str(folder) + '/' + str(filename) + '.pkl', 'wb') as f:\n",
    "        pickle.dump(var, f)\n",
    "        \n",
    "def load_file_from_folder(folder, filename):\n",
    "    with open(str(folder) + '/' + str(filename) + '.pkl', 'rb') as f:\n",
    "        res = pickle.load(f)\n",
    "        return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "B_T value: 5.414213562373095\n",
      "Sigma value for the experimenxt: 1\n",
      "gamma: 0.9941173985691094\n",
      "tau: 169.99281895061446\n"
     ]
    }
   ],
   "source": [
    "# PARAMETERS\n",
    "d = 2  # dimension of the problem\n",
    "k = 50  # number of arms at each step\n",
    "steps = 6000  # number of steps for the experiment\n",
    "n_scat = 1000 # How frequently should we plot the true parameter\n",
    "t_saved = None\n",
    "\n",
    "# When saving only a part of the trajectory rather than all the points uncomment the 2 following lines\n",
    "# number_t_saved = steps//10\n",
    "# t_saved = np.int_(np.linspace(0, steps - 1, number_t_saved))\n",
    "\n",
    "alpha = 1\n",
    "sigma_noise = 1 # Square root of the variance of the noise\n",
    "verbose = False\n",
    "q = 5 # 5 percent quantiles used\n",
    "\n",
    "### Description of the experiment\n",
    "theta = np.array([1,  0]) # Starting point of the true unknown parameter\n",
    "bp = {1001: np.array([-1,  0]), 2001: np.array([0,  1]), 3001: np.array([0,  -1])} # Descrition of the breakpoints\n",
    "\n",
    "B_T = 4 + np.sqrt(2) # Computed manually\n",
    "print('B_T value:', B_T)\n",
    "print('Sigma value for the experimenxt:', sigma_noise)\n",
    "\n",
    "gamma  = 1 - (B_T/(d*steps))**(2/3) # Optimal Value to minimize the asymptotical regret\n",
    "tau = (d*steps/B_T)**(2/3) # Optimal Value to minimize the asymptotical regret\n",
    "print('gamma:', gamma)\n",
    "print('tau:', tau)\n",
    "\n",
    "policies = [DLinUCB(d, delta, alpha, lambda_, s, gamma, '', sm = False, sigma_noise = sigma_noise, verbose=verbose),\n",
    "            SWLinUCB(d, delta, alpha, lambda_, s, tau, '', sm=False, sigma_noise=sigma_noise, verbose=verbose),\n",
    "            DynamicLinUCB(d, delta, alpha, lambda_, s, tau=200, name='', filename = 'abrupt_exp_final',\n",
    "                          sm=True, sigma_noise = sigma_noise, delta_2=0.01, tilde_delta=0.002, verbose=False, \n",
    "                          omniscient=False),\n",
    "            PolicyLinUCB(d, delta, alpha, lambda_, s, '', sm = True, sigma_noise = sigma_noise, verbose=verbose),\n",
    "            PolicyLinUCB(d, delta, alpha, lambda_, s, '-OR', sm = True, sigma_noise = sigma_noise, verbose=verbose,\n",
    "                        omniscient = True)\n",
    "           ]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unparallelized experiments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  0%|          | 0/2 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 D-LinUCB [0. 0.] [1 0]\n",
      "1000 D-LinUCB [1.0336647  0.15032362] [1 0]\n",
      "2000 D-LinUCB [-0.99439429  0.1253892 ] [-1  0]\n",
      "3000 D-LinUCB [0.18665829 0.82470159] [0 1]\n",
      "4000 D-LinUCB [ 0.14962014 -0.97342344] [ 0 -1]\n",
      "5000 D-LinUCB [-0.09219099 -0.96700548] [ 0 -1]\n",
      "0 SW-LinUCB [0. 0.] [1 0]\n",
      "1000 SW-LinUCB [ 0.94207346 -0.11918458] [1 0]\n",
      "2000 SW-LinUCB [-0.81413941  0.09369063] [-1  0]\n",
      "3000 SW-LinUCB [0.00383771 1.05865697] [0 1]\n",
      "4000 SW-LinUCB [-0.26719374 -0.96224329] [ 0 -1]\n",
      "5000 SW-LinUCB [-0.14277934 -0.95955255] [ 0 -1]\n",
      "0 dLinUCB [0. 0.] [1 0]\n",
      "1000 dLinUCB [1.04543044 0.01047125] [1 0]\n",
      "Creation of a new model at time:1143\n",
      "2000 dLinUCB [-1.03135246  0.03413425] [-1  0]\n",
      "3000 dLinUCB [-0.59847738  0.10386267] [0 1]\n",
      "4000 dLinUCB [-0.49082773 -0.07334713] [ 0 -1]\n",
      "5000 dLinUCB [-0.49697686 -0.16282064] [ 0 -1]\n",
      "0 LinUCB [0. 0.] [1 0]\n",
      "1000 LinUCB [ 0.99906161 -0.03878663] [1 0]\n",
      "2000 LinUCB [0.08682585 0.00278432] [-1  0]\n",
      "3000 LinUCB [0.15907871 0.54730433] [0 1]\n",
      "4000 LinUCB [0.06069322 0.02766634] [ 0 -1]\n",
      "5000 LinUCB [ 0.21365131 -0.11175687] [ 0 -1]\n",
      "0 LinUCB-OR [0. 0.] [1 0]\n",
      "1000 LinUCB-OR [0.91485193 0.21864987] [1 0]\n",
      "2000 LinUCB-OR [-1.01365626 -0.08995285] [-1  0]\n",
      "3000 LinUCB-OR [0.03454116 1.04176173] [0 1]\n",
      "4000 LinUCB-OR [-0.08598569 -1.00636173] [ 0 -1]\n",
      "5000 LinUCB-OR [-0.06355498 -1.01395648] [ 0 -1]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 50%|█████     | 1/2 [00:22<00:22, 22.69s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1152\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "100%|██████████| 2/2 [00:45<00:00, 22.59s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-- Building data out of the experiments ---\n",
      "--- Data built ---\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "data, theta_hat, action_check_val, timedic = data_from_experiment(n_mc=2, option = '')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'D-LinUCB': 10.487666845321655, 'SW-LinUCB': 9.115904331207275, 'dLinUCB': 8.719751834869385, 'LinUCB': 8.368414402008057, 'LinUCB-OR': 8.357692241668701}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot on a single experiment\n",
    "print(timedic) # Computation time for the different policies\n",
    "action_check(action_check_val, 0) # To see the available actions at one step\n",
    "plot_regret(data, t_saved, filename = None, log=False, qtl=True, loc=2, font=13, bp = bp, bp_2 = {})\n",
    "scatter_abrupt(theta_hat, None, theta, bp, loc=0, font=10, circle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  0%|          | 0/100 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 D-LinUCB [0. 0.] [1 0]\n",
      "1000 D-LinUCB [ 1.10028461 -0.00923249] [1 0]\n",
      "2000 D-LinUCB [-0.92883273 -0.01456212] [-1  0]\n",
      "3000 D-LinUCB [-0.05819116  0.97228247] [0 1]\n",
      "4000 D-LinUCB [-0.19413356 -0.97887389] [ 0 -1]\n",
      "5000 D-LinUCB [-0.11933417 -0.91318628] [ 0 -1]\n",
      "0 SW-LinUCB [0. 0.] [1 0]\n",
      "1000 SW-LinUCB [0.98908074 0.02045452] [1 0]\n",
      "2000 SW-LinUCB [-1.04158366 -0.07015873] [-1  0]\n",
      "3000 SW-LinUCB [-0.17663803  0.83564732] [0 1]\n",
      "4000 SW-LinUCB [-0.04960634 -1.07285185] [ 0 -1]\n",
      "5000 SW-LinUCB [ 0.02966317 -1.03225244] [ 0 -1]\n",
      "0 dLinUCB [-0.47191378 -0.21313744] [1 0]\n",
      "1000 dLinUCB [ 0.99477346 -0.03438142] [1 0]\n",
      "Creation of a new model at time:1207\n",
      "2000 dLinUCB [-0.9691038  0.0059536] [-1  0]\n",
      "3000 dLinUCB [-0.63967061  0.75574878] [0 1]\n",
      "4000 dLinUCB [-0.56415817  0.11048285] [ 0 -1]\n",
      "5000 dLinUCB [ 0.17834485 -0.20157163] [ 0 -1]\n",
      "0 LinUCB [0. 0.] [1 0]\n",
      "1000 LinUCB [ 0.97834109 -0.0019178 ] [1 0]\n",
      "2000 LinUCB [ 0.08975521 -0.00361395] [-1  0]\n",
      "3000 LinUCB [0.1822434  0.49731152] [0 1]\n",
      "4000 LinUCB [0.06539652 0.03826924] [ 0 -1]\n",
      "5000 LinUCB [ 0.22426423 -0.12227789] [ 0 -1]\n",
      "0 LinUCB-OR [0. 0.] [1 0]\n",
      "1000 LinUCB-OR [ 0.98539197 -0.06680445] [1 0]\n",
      "2000 LinUCB-OR [-0.93411743 -0.02569446] [-1  0]\n",
      "3000 LinUCB-OR [-0.00301195  1.01652797] [0 1]\n",
      "4000 LinUCB-OR [-0.09846715 -0.97435587] [ 0 -1]\n",
      "5000 LinUCB-OR [-0.0795424  -1.01546476] [ 0 -1]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  1%|          | 1/100 [00:23<38:51, 23.55s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1130\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  3%|▎         | 3/100 [01:11<38:33, 23.85s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1169\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  4%|▍         | 4/100 [01:35<37:56, 23.72s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1198\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  5%|▌         | 5/100 [01:59<37:59, 24.00s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1161\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  6%|▌         | 6/100 [02:23<37:30, 23.94s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1178\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  7%|▋         | 7/100 [02:47<36:54, 23.82s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1151\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  8%|▊         | 8/100 [03:11<36:35, 23.87s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1135\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  9%|▉         | 9/100 [03:36<36:55, 24.35s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1201\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 10%|█         | 10/100 [04:01<36:37, 24.42s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1102\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 11%|█         | 11/100 [04:25<36:13, 24.43s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1117\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 12%|█▏        | 12/100 [04:51<36:17, 24.74s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1128\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 14%|█▍        | 14/100 [05:44<36:52, 25.73s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1160\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 15%|█▌        | 15/100 [06:08<35:43, 25.21s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1142\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 16%|█▌        | 16/100 [06:31<34:31, 24.66s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1141\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 17%|█▋        | 17/100 [06:55<33:46, 24.41s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1119\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 18%|█▊        | 18/100 [07:19<33:18, 24.37s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1131\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 19%|█▉        | 19/100 [07:43<32:32, 24.11s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1127\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 20%|██        | 20/100 [08:06<31:49, 23.86s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1120\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 21%|██        | 21/100 [08:29<31:10, 23.68s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1154\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 22%|██▏       | 22/100 [08:52<30:33, 23.50s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1141\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 23%|██▎       | 23/100 [09:16<30:06, 23.46s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1192\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 24%|██▍       | 24/100 [09:39<29:45, 23.49s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1161\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 25%|██▌       | 25/100 [10:02<29:16, 23.42s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1129\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 26%|██▌       | 26/100 [10:26<28:49, 23.37s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1180\n",
      "Creation of a new model at time:3197\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 27%|██▋       | 27/100 [10:50<28:52, 23.73s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1171\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 28%|██▊       | 28/100 [11:14<28:25, 23.69s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1114\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 29%|██▉       | 29/100 [11:37<27:59, 23.66s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1117\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 30%|███       | 30/100 [12:01<27:28, 23.55s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1136\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 31%|███       | 31/100 [12:25<27:24, 23.84s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1138\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 32%|███▏      | 32/100 [12:49<27:05, 23.91s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1103\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 33%|███▎      | 33/100 [13:14<27:07, 24.30s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1109\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 34%|███▍      | 34/100 [13:38<26:20, 23.94s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1166\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 35%|███▌      | 35/100 [14:03<26:15, 24.24s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1122\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 36%|███▌      | 36/100 [14:25<25:23, 23.80s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1209\n",
      "Creation of a new model at time:3182\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 37%|███▋      | 37/100 [14:49<24:52, 23.69s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1181\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 38%|███▊      | 38/100 [15:12<24:21, 23.58s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:3459\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 39%|███▉      | 39/100 [15:37<24:14, 23.84s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1156\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 40%|████      | 40/100 [16:00<23:49, 23.83s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1211\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 41%|████      | 41/100 [16:25<23:36, 24.00s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1157\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 42%|████▏     | 42/100 [16:50<23:25, 24.24s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1197\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 43%|████▎     | 43/100 [17:13<22:50, 24.05s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1141\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 44%|████▍     | 44/100 [17:36<22:07, 23.71s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1193\n",
      "Creation of a new model at time:3174\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 45%|████▌     | 45/100 [17:59<21:34, 23.53s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1092\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 46%|████▌     | 46/100 [18:23<21:20, 23.71s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1222\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 47%|████▋     | 47/100 [18:47<21:03, 23.85s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1181\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 48%|████▊     | 48/100 [19:10<20:24, 23.54s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1196\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 49%|████▉     | 49/100 [19:33<19:54, 23.42s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1152\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 50%|█████     | 50/100 [19:57<19:35, 23.52s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1165\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 51%|█████     | 51/100 [20:21<19:23, 23.75s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1123\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 52%|█████▏    | 52/100 [20:46<19:08, 23.93s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1284\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 53%|█████▎    | 53/100 [21:10<18:55, 24.15s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1144\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 54%|█████▍    | 54/100 [21:34<18:25, 24.04s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1144\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 56%|█████▌    | 56/100 [22:23<17:50, 24.33s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1069\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 57%|█████▋    | 57/100 [22:47<17:22, 24.24s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1123\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 58%|█████▊    | 58/100 [23:12<17:03, 24.36s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1153\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 59%|█████▉    | 59/100 [23:36<16:27, 24.10s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1110\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 60%|██████    | 60/100 [24:00<16:06, 24.16s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1119\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 61%|██████    | 61/100 [24:24<15:37, 24.03s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1187\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 62%|██████▏   | 62/100 [24:47<15:05, 23.84s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1184\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 63%|██████▎   | 63/100 [25:10<14:35, 23.67s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1113\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 64%|██████▍   | 64/100 [25:34<14:08, 23.58s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1189\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 65%|██████▌   | 65/100 [25:58<13:50, 23.73s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1174\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 66%|██████▌   | 66/100 [26:23<13:41, 24.16s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1163\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 67%|██████▋   | 67/100 [26:47<13:20, 24.25s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1199\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 68%|██████▊   | 68/100 [27:10<12:45, 23.93s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1128\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 69%|██████▉   | 69/100 [27:33<12:13, 23.66s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1166\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 70%|███████   | 70/100 [27:59<12:04, 24.16s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1104\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 71%|███████   | 71/100 [28:23<11:44, 24.28s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1284\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 72%|███████▏  | 72/100 [28:47<11:16, 24.17s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1201\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 73%|███████▎  | 73/100 [29:13<11:04, 24.62s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1165\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 74%|███████▍  | 74/100 [29:37<10:37, 24.50s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1181\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 75%|███████▌  | 75/100 [30:02<10:14, 24.58s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1124\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 76%|███████▌  | 76/100 [30:27<09:51, 24.66s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1118\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 77%|███████▋  | 77/100 [30:51<09:26, 24.63s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1117\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 78%|███████▊  | 78/100 [31:16<09:04, 24.74s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1099\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 79%|███████▉  | 79/100 [31:41<08:39, 24.76s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1183\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 81%|████████  | 81/100 [32:30<07:50, 24.74s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 82%|████████▏ | 82/100 [32:54<07:19, 24.44s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1123\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 83%|████████▎ | 83/100 [33:18<06:54, 24.38s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1153\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 84%|████████▍ | 84/100 [33:44<06:36, 24.75s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1149\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 85%|████████▌ | 85/100 [34:08<06:08, 24.57s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1116\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 86%|████████▌ | 86/100 [34:33<05:44, 24.58s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1183\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 88%|████████▊ | 88/100 [35:21<04:53, 24.45s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1150\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 89%|████████▉ | 89/100 [35:45<04:27, 24.34s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1071\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 90%|█████████ | 90/100 [36:09<04:02, 24.21s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1137\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 91%|█████████ | 91/100 [36:33<03:36, 24.01s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1227\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 93%|█████████▎| 93/100 [37:22<02:50, 24.30s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1176\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 94%|█████████▍| 94/100 [37:46<02:25, 24.22s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1139\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 95%|█████████▌| 95/100 [38:11<02:01, 24.40s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1160\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 96%|█████████▌| 96/100 [38:35<01:37, 24.34s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1158\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 97%|█████████▋| 97/100 [38:59<01:12, 24.24s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1121\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 98%|█████████▊| 98/100 [39:23<00:48, 24.16s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1112\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 99%|█████████▉| 99/100 [39:47<00:24, 24.05s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1124\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "100%|██████████| 100/100 [40:11<00:00, 23.90s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-- Building data out of the experiments ---\n",
      "--- Data built ---\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# 100 indepent experiments without parallelization, should take around half an hour.\n",
    "data_100, theta_hat_100, action_check_val_100, timedic_100 = data_from_experiment(n_mc=100, option = '')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'D-LinUCB': 564.9769353866577, 'SW-LinUCB': 488.5641691684723, 'dLinUCB': 465.18879771232605, 'LinUCB': 445.79935932159424, 'LinUCB-OR': 446.4877917766571}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot average on 100 independent experiments\n",
    "print(timedic_100) # Computation time for the different policies\n",
    "plot_regret(data_100, t_saved, filename = None, log=False, qtl=True, loc=2, font=13, bp = bp, bp_2 = {1157:''})\n",
    "scatter_abrupt(theta_hat_100, None, theta, bp, loc=0, font=10, circle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For simplifying the plot and removing unwanted policies\n",
    "theta_hat_100_partial = {}\n",
    "theta_hat_100_partial['D-LinUCB'] = theta_hat_100['D-LinUCB']\n",
    "theta_hat_100_partial['SW-LinUCB'] = theta_hat_100['SW-LinUCB']\n",
    "theta_hat_100_partial['dLinUCB'] = theta_hat_100['dLinUCB']\n",
    "theta_hat_100_partial['LinUCB'] = theta_hat_100['LinUCB']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fbe085a7d30>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_regret(data_100, t_saved, filename = None, log=False, qtl=True, loc=2, font=13, bp = bp, bp_2 = {1157:''})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fbe087aec18>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# when removing the LinUCB-OR policy\n",
    "scatter_abrupt(theta_hat_100_partial, None, theta, bp, loc=0, font=13, circle=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Parallelized experiments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creation of a new model at time:1104\n",
      "Creation of a new model at time:1152\n",
      "Creation of a new model at time:1125\n",
      "Creation of a new model at time:1191\n",
      "Creation of a new model at time:1136\n",
      "Creation of a new model at time:1095\n",
      "Creation of a new model at time:1123\n",
      "Creation of a new model at time:1091\n",
      "Creation of a new model at time:1122\n",
      "Creation of a new model at time:1136\n",
      "Creation of a new model at time:1150\n",
      "Creation of a new model at time:1179\n",
      "Creation of a new model at time:1064\n",
      "Creation of a new model at time:1135\n",
      "Creation of a new model at time:1179\n",
      "Creation of a new model at time:1114\n",
      "Creation of a new model at time:1167\n",
      "Creation of a new model at time:1174\n",
      "Creation of a new model at time:1094\n",
      "Creation of a new model at time:1139\n",
      "Creation of a new model at time:1185\n",
      "Creation of a new model at time:1198\n",
      "Creation of a new model at time:3205\n",
      "Creation of a new model at time:1145\n",
      "Creation of a new model at time:1105\n",
      "Creation of a new model at time:1090\n",
      "Creation of a new model at time:1139\n",
      "Creation of a new model at time:1068\n",
      "Creation of a new model at time:1139\n",
      "Creation of a new model at time:1116\n",
      "Creation of a new model at time:1104\n",
      "Creation of a new model at time:1181\n",
      "Creation of a new model at time:1171\n",
      "Creation of a new model at time:1186\n",
      "Creation of a new model at time:3257\n",
      "Creation of a new model at time:1170\n",
      "Creation of a new model at time:1311\n",
      "Creation of a new model at time:1168\n",
      "Creation of a new model at time:1108\n",
      "Creation of a new model at time:1197\n",
      "Creation of a new model at time:1123\n",
      "Creation of a new model at time:1125\n",
      "Creation of a new model at time:1153\n",
      "Creation of a new model at time:1217\n",
      "Creation of a new model at time:1113\n",
      "Creation of a new model at time:1162\n",
      "Creation of a new model at time:1143\n",
      "Creation of a new model at time:1105\n",
      "Creation of a new model at time:1138\n",
      "Creation of a new model at time:1135\n",
      "Creation of a new model at time:1159\n",
      "Creation of a new model at time:1098\n",
      "Creation of a new model at time:1162\n",
      "Creation of a new model at time:1152\n",
      "Creation of a new model at time:1136\n",
      "Creation of a new model at time:1156\n",
      "Creation of a new model at time:1172\n",
      "Creation of a new model at time:1119\n",
      "Creation of a new model at time:1160\n",
      "Creation of a new model at time:1189\n",
      "Creation of a new model at time:1174\n",
      "Creation of a new model at time:1149\n",
      "Creation of a new model at time:1179\n",
      "Creation of a new model at time:1169\n",
      "Creation of a new model at time:1164\n",
      "Creation of a new model at time:1161\n",
      "Creation of a new model at time:1129\n",
      "Creation of a new model at time:1135\n",
      "Creation of a new model at time:1143\n",
      "Creation of a new model at time:1172\n",
      "Creation of a new model at time:1136\n",
      "Creation of a new model at time:1431\n",
      "Creation of a new model at time:3213\n",
      "Creation of a new model at time:1095\n",
      "Creation of a new model at time:1180\n",
      "Creation of a new model at time:1083\n",
      "Creation of a new model at time:1122\n",
      "Creation of a new model at time:1132\n",
      "Creation of a new model at time:3277\n",
      "Creation of a new model at time:1137\n",
      "Creation of a new model at time:1148\n",
      "Creation of a new model at time:1187\n",
      "Creation of a new model at time:1092\n",
      "Creation of a new model at time:1196\n",
      "Creation of a new model at time:1103\n",
      "Creation of a new model at time:1217\n",
      "Creation of a new model at time:1164\n",
      "Creation of a new model at time:3163\n",
      "Creation of a new model at time:1166\n",
      "Creation of a new model at time:1119\n",
      "Creation of a new model at time:1215\n",
      "Creation of a new model at time:1166\n",
      "Creation of a new model at time:3130\n",
      "Creation of a new model at time:1185\n",
      "Creation of a new model at time:1190\n",
      "Creation of a new model at time:1211\n",
      "Creation of a new model at time:1121\n",
      "Creation of a new model at time:1224\n",
      "Creation of a new model at time:1117\n",
      "Creation of a new model at time:1153\n",
      "Creation of a new model at time:1106\n",
      "Total Time Experiment:  299.2663106918335\n",
      "Conversion time: 1.1676139831542969\n",
      "Post-Processing Time: 0.027048110961914062\n"
     ]
    }
   ],
   "source": [
    "n_process = 11\n",
    "\n",
    "file = 'abrupt_par_exp_final'\n",
    "policies = [DLinUCB(d, delta, alpha, lambda_, s, gamma, '', sm = False, sigma_noise = sigma_noise, verbose=verbose),\n",
    "            SWLinUCB(d, delta, alpha, lambda_, s, tau, '', sm=False, sigma_noise=sigma_noise, verbose=verbose),\n",
    "            DynamicLinUCB(d, delta, alpha, lambda_, s, tau=200, name='', filename = file,\n",
    "                          sm=True, sigma_noise = sigma_noise, delta_2=0.01, tilde_delta=0.002, verbose=False, \n",
    "                          omniscient=False),\n",
    "            PolicyLinUCB(d, delta, alpha, lambda_, s, '', sm = True, sigma_noise = sigma_noise, verbose=verbose),\n",
    "            PolicyLinUCB(d, delta, alpha, lambda_, s, '-OR', sm = True, sigma_noise = sigma_noise, verbose=verbose,\n",
    "                        omniscient = True)\n",
    "           ]\n",
    "\n",
    "n_expe = 100\n",
    "data_par = data_from_experiment(n_expe,'par')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "95\n",
      "0\n",
      "6\n"
     ]
    }
   ],
   "source": [
    "with open(\"detection_out/\" + file, 'r') as f:\n",
    "    detect = list(map(lambda x:x.strip(),f.readlines()))\n",
    "\n",
    "detect = [int(el) for el in detect]\n",
    "list_1 = [el for el in detect if el < 2000]\n",
    "list_2 = [el for el in detect if el < 3000 and el >= 2000]\n",
    "list_3 = [el for el in detect if  el >= 3000]\n",
    "\n",
    "# Number of detections\n",
    "print(len(list_1))\n",
    "print(len(list_2))\n",
    "print(len(list_3))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of detection first breakpoint: 95.0\n",
      "Percentage of detection second breakpoint: 0.0\n",
      "Percentage of detection third breakpoint: 6.0\n"
     ]
    }
   ],
   "source": [
    "print('Percentage of detection first breakpoint:', len(list_1)/n_expe*100)\n",
    "print('Percentage of detection second breakpoint:',len(list_2)/n_expe*100)\n",
    "print('Percentage of detection third breakpoint:',len(list_3)/n_expe*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1150.93684211\n",
      "3207.5\n"
     ]
    }
   ],
   "source": [
    "print(np.mean(list_1)) # Average detection step of the first breakpoint\n",
    "print(np.mean(list_3)) # Average detection step of the third breakpoint"
   ]
  }
 ],
 "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}