{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a199a325",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "IPython.OutputArea.prototype._should_scroll = function(lines) {\n",
       "    return false;\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%javascript\n",
    "IPython.OutputArea.prototype._should_scroll = function(lines) {\n",
    "    return false;\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "efb81342",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import openturns as ot\n",
    "import numpy as np\n",
    "import openturns.viewer as viewer\n",
    "\n",
    "from matplotlib import pylab as plt\n",
    "from matplotlib.patches import Circle, Wedge, Polygon, Rectangle\n",
    "from matplotlib.collections import PatchCollection\n",
    "from matplotlib.font_manager import FontProperties\n",
    "\n",
    "import math\n",
    "from functools import partial\n",
    "from joblib import Parallel, delayed\n",
    "import shelve\n",
    "import os\n",
    "\n",
    "import otaf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "66dedf7b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# HELPER FUNCTIONS\n",
    "# For plotting\n",
    "x_min = -0.25\n",
    "x_max = 0.7\n",
    "n_points = int(1e4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84809625",
   "metadata": {},
   "source": [
    "# Reference Analytical Example in 2D\n",
    "\n",
    "To show the benefits of this approach compared to the usual ones, we apply it using a minimal isostatic example using only a reduced number of variables and having an analytical solution. As explained, the analysis consists of 6 steps : \n",
    " - The definition of the mathematical model from the the parts plans. \n",
    " - The construction of the deviation domain for each toleranced feature\n",
    " - The construction of the probabilistic model using some quality criteria (sigma/6t smthing)\n",
    " - The generation of the design of experiment\n",
    " - The propagation of the uncertainty through the model.\n",
    " - The extraction of the upper and lower envelope of the response of the model. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f074bd6a-8b07-4fb6-b2ab-7e55a1ebf361",
   "metadata": {},
   "source": [
    "## Exploring the imprecise probability space. \n",
    "\n",
    "For the male piece, we know that in 95% of the cases, the value of X1 will be between X1 - X_tot < X1 < X1 + X_tol ,\n",
    "but the uncertainty is originating from a combination of positional and orientation defects. We do  not know which is contributing more. \n",
    "\n",
    "Let's first explore how the sum of the defects behaves, when under the constraint of the 95%\n",
    "\n",
    "Let's consider that the real measure of X1 is based on a base value and a sum of random modes:\n",
    "\n",
    "$$\n",
    "X_{1} = X1 + \\mathcal{e}_{pos} + \\mathcal{e}_{ori} \\\\\n",
    "\\mathcal{e}_{pos} + \\mathcal{e}_{ori} <= X_{tol} \\ (Dans\\ 95\\%\\ des\\ cas)\n",
    "$$\n",
    "\n",
    "![schema](../Pictures/Female_Male_Part.jpg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "bc12c066",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "### Different measures of our problem\n",
    "X1 = 99.8  # Nominal Length of the male piece\n",
    "X2 = 100  # Nominal Length of the female piece\n",
    "X3 = 10.0  # Nominal width of the pieces\n",
    "j = X2 - X1  # Nominal play between pieces.\n",
    "T = 0.2  # Tolerance for X1 and X2. (95% conform)  (= t/2)\n",
    "t_ = T / 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "10feadf6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "Cm = 1\n",
    "\n",
    "sigma_e_pos_max = T / (6 * Cm)\n",
    "e_pos = ot.Normal(0, sigma_e_pos_max)\n",
    "e_pos.setDescription([\"e_pos\"])\n",
    "\n",
    "# Le défaut en orientation est piloté par une incertitude sur un angle. On suppose les angles petits << 1 rad\n",
    "theta_max = T / X3\n",
    "sigma_theta_max = (2 * theta_max) / (\n",
    "    6 * Cm\n",
    ")  # 2* theta max!!!! A cause de l'intervalle + / - !!!!!norma\n",
    "e_theta = ot.Normal(0, sigma_theta_max)\n",
    "e_theta.setDescription([\"e_theta\"])\n",
    "\n",
    "e_ori = (e_theta * X3) / 2\n",
    "e_ori = e_ori.abs()\n",
    "e_ori.setDescription([\"e_ori\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9ddf68f7-10b5-4884-a973-fd44b629b5aa",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def analytical_assembly_model_1_5_D(rdv, rnd=6):\n",
    "    \"\"\"\n",
    "    Calculate the minimum difference between the upper and lower gaps for a given set of defects.\n",
    "\n",
    "    Parameters:\n",
    "        rdv ot.Sample: random deviation vector in the form of a openturns Sample\n",
    "        rnd (int, optional): Number of decimal places to round the result (default: 6).\n",
    "\n",
    "    Returns:\n",
    "        list: A list containing the minimum play/gap for each point in the random deviation array.\n",
    "\n",
    "    \"\"\"\n",
    "    size = rdv.getSize()\n",
    "    desc = rdv.getDescription()\n",
    "    nzs1 = np.zeros((size, 1))\n",
    "    smpU1 = np.array(rdv.getMarginal([\"u_d_1\"])) if \"u_d_1\" in desc else nzs1\n",
    "    smpG1 = np.array(rdv.getMarginal([\"gamma_d_1\"])) if \"gamma_d_1\" in desc else nzs1\n",
    "    smpU2 = np.array(rdv.getMarginal([\"u_d_2\"])) if \"u_d_2\" in desc else nzs1\n",
    "    smpG2 = np.array(rdv.getMarginal([\"gamma_d_2\"])) if \"gamma_d_2\" in desc else nzs1\n",
    "\n",
    "    X1_tilde = np.asarray(\n",
    "        [X1 + smpU1 - (X3 / 2) * smpG1, X1 + smpU1 + (X3 / 2) * smpG1]  # haut\n",
    "    ) \n",
    "\n",
    "    X2_tilde = np.asarray(\n",
    "        [X2 - smpU2 - (X3 / 2) * smpG2, X2 - smpU2 + (X3 / 2) * smpG2]  # bas\n",
    "    )\n",
    "    jeu = X2_tilde - X1_tilde\n",
    "    jeu_min = np.expand_dims(np.squeeze(jeu.min(axis=0)), axis=1).round(rnd).tolist()\n",
    "    return ot.Sample(jeu_min)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "78db7201-357f-4242-98b9-a08990853d0d",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0013498980316301035\n",
      "0.0013498980316301035\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAGxCAYAAACeKZf2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABKu0lEQVR4nO3deXwTdf4/8FeSJul90fsGOctRoEAtoMjPQvFAcRdlUQ4RcRGqaAUFQcqxWteDxVUUFVlURFFRdBe+aCk3FCrlPgoUKIXSE+jdJmkyvz9qA6EHTZt0cryejwcPOpOZ5P0mbfpi5jOfkQiCIICIiIjIRkjFLoCIiIjIlBhuiIiIyKYw3BAREZFNYbghIiIim8JwQ0RERDaF4YaIiIhsCsMNERER2RSGGyIiIrIpDmIX0N50Oh2uXr0KNzc3SCQSscshIiKiFhAEAeXl5QgKCoJU2vyxGbsLN1evXkVoaKjYZRAREVErXL58GSEhIc1uY3fhxs3NDUDdP467u7vI1RhHo9Hg999/x8iRIyGXy8Uux+zYr21jv7aN/do2MfotKytDaGio/vd4c+wu3NSfinJ3d7fKcOPs7Ax3d3e7+eFhv7aL/do29mvbxOy3JUNKOKCYiIiIbArDDREREdkUhhsiIiKyKXY35qalamtroVarxS7DgEajgVwuR1VVld2c021LvwqFAg4O/BYnIrI3/OS/jSAIyMnJQXFxsdilNMrf3x9ZWVlil9Fu2tqvj48PwsLCOKcREZEdYbi5TX2wCQ4Ohqur6x0nCiLLpNPpUFFRgdzcXABAeHi4yBUREVF7Ybi5RW1trT7YBAQEiF0OtZGrqysAIDc3Fzdu3EBUVBSP4BAR2QEelrhF/Rib+l+KZP3q38v9+/fjyJEj4hZDRETtguGmETwVZTvq30uZTIb9+/db3CBxIiIyPVF/i+/atQujR49GUFAQJBIJNm7ceMd9duzYgf79+0OpVKJz585Ys2aN2esk6+fs7AyVSoXKykqxSyEiIjMTNdxUVlYiKioKK1asaNH2Fy9exEMPPYThw4fjyJEjeOmll/Dss8/it99+M3Ol1JSWhlKx1Y+1EQRB5EqIiMjcRA03DzzwAP7xj3/gsccea9H2K1euRMeOHfH++++jR48eSEhIwNixY/Gvf/3LzJVSU/Ly8vDAAw80u01+fj5eeOEFdOrUCUqlEqGhoRg9ejRSU1P120REREAikUAikcDJyQkRERF44oknsG3bNoPnys7O1m93658JEyaYpT8iIrI+VnW1VFpaGuLi4gzWxcfH46WXXmpyH5VKBZVKpV8uKysDUDdBnEajMdj29mVqmlqthkKhuONVZdnZ2RgyZAg8PT3x7rvvonfv3tBoNPjtt98wc+ZMZGZm6rddsmQJpk2bBrVajezsbKxduxZxcXFYunQp5s+fb/C8W7duRc+ePfXLTk5Ozdah0+mg0+lQW1trse9zfV2WWp+psV/bxn6tjyAIUNXqUKnWokpdiyqVFqpaHVS1Oqi1dX+rNFqoa3WoUmlwLF+CK7vOo1Yn+XM7LdRaAepaLVQaHRY+3B1ujqabcNaYf1urCjf5+fnw9/c3WOfv74+ysjJUV1c3+gsuOTkZixcvbrD+999/h7Ozs8E6uVze4PmthU6nwz//+U989tlnyM/PR9euXfHGG29g7NixLdp/586dmDNnDo4ePQpvb29MnjwZ//jHP/Qz/N53333o1asXHBwcsHbtWvTu3Rvbt2+HRCLBzz//jDFjxjT6vDNmzIBEIkF6ejpcXFz063v27IlnnnnGYFs3Nzd9WAoLC8O9996LwMBALFy4EGPHjkW3bt3023bo0MGoy/WvXr2K/Px8pKamNnjfLU1KSorYJbQr9mvb2K/5aXVAtRaoqgWqa4FqraTua/06CdRaQKWD/m+VVgKVFlDrANUtjwkwZroMGXDxfJOP9pVdhpey7f3Vq6qqavG2VhVuWmPevHlITEzUL5eVlSE0NBQjR46Eu7u7wbZVVVVWO/tvcnIy1q5di5UrV6JLly7YtWsXJkyYAF9fXwwbNqzZfXNzc/Hggw/i6aefxldffYXMzExMmzYNjo6OWLRokX67L7/8Es8//zz27t3bopquX7+OLVu24M033zQINvU8PT3v+ByzZs3C0qVL8csvv+DVV19t0es2JigoCA4ODrj//vvh5eXV6ucxJ41Gg5SUFIwYMcJubq/Bfm0X+229KnUtrlWqca1CjWuValy/5eu6ZQ2uVapRUqVGWU0tqtRaE3VhWkPuHYaIDg0/+1ur/sxLS1hVuAkICEBBQYHBuoKCAri7uzd5WkKpVEKpbBgd5XJ5g2/Apr4hR3+4B0XlqkYfMydfNyX++8LQO26nUqnw1ltvYevWrYiNjQUAdOrUCXv27MGnn356x3Dz8ccfIzQ0FB999BEkEgm6d++Oq1ev4rXXXsPChQv1l1N36dIF77zzTovrz8rKgiAI6N69e4v3uZ23tzf8/PyQnZ1tsH7w4MEGl+zv3r0b/fr1a/J5pFIppFIpHBwcLP6DtrHvTVvGfm0b+71JEASUVdcir6waeSU1uFpq+Hd+WQ3yS2tQrWmfsKKQSeGslMFZLoOz0gEuChmcFQ5wVtQtO8tlcFLIoHCQQqn/U7fsIBFw+uRxDIruB2elAkp53WNKB6l++xAvZ8gdTDe015jvI6sKN7Gxsdi8ebPBupSUFP0vdHMpKlchv6zGrK/RFllZWaiqqsKIESMM1qvV6mZ/4dc7ffo0YmNjDWbvHTJkCCoqKnDlyhWEhYUBAKKjo42qy1RXJgmC0GBm4fXr16NHjx765dDQUJO8FhFRW9RotMi+XoPsa1W4dK0Sl65V4dL1KuTeqEJeaY3JjrIoZFJ4Osvh4SSHu9Offzs63LZc97W7owNcHf8MLQoHuCgc9KGltTQaDTYXHsMDvQIsMryKGm4qKioMTgNdvHgRR44cgbe3N8LCwjBv3jzk5ubiq6++AgBMnz4dH330EV599VU888wz2LZtG77//nts2rTJrHX6upnwpKEZXreiogIAsGnTJgQHBxs81thRq9Zq7NRSc7p06QKJRGIwaNhY165dQ1FRETp27GiwPjQ0FJ07d2718xIRtUVxhQrnCiqQVViOswUVOFdQhsxcGWalpd555ya4Kh0Q4OEIH1cFOrgq4eNS93cHVwU6uCj16zu4KuCmdODtZJoharg5ePAghg8frl+uHxszefJkrFmzBnl5ecjJydE/3rFjR2zatAkvv/wyPvjgA4SEhGDVqlWIj483a50tOTUkpsjISCiVSuTk5NzxFFRjevTogQ0bNhgcIdm7dy/c3NwQEhLS6rq8vb0RHx+PFStW4MUXX2wQjkpKSu447uaDDz6AVCptcsAyEZE5qWq1OJtfgRNXS3HqahnOFJQjq7AC1ysbm+286bDhJJch0NMRQR5OCPBwRJCHIwI9nRDo4YhADycEejrC3YRXFtk7UcPNfffd1+ypi8ZmH77vvvtw+PBhM1Zlfdzc3DB79my8/PLL0Ol0GDp0KEpLS7F37164u7tj8uTJze4/Y8YMLF++HC+88AISEhJw5swZJCUlITExsc23olixYgWGDBmCQYMGYcmSJejTpw9qa2uRkpKCTz75BKdPn9ZvW15ejvz8fGg0Gly8eBFr167FqlWrkJyczKM0RGR2NRotTl4txYncMpzILcWJq2U4V1COWl3LTrG7OAjoEuiJiA4uCO/ggvAOzgjv4IKIDs7wdlHwSEs7sqoxN9S0pUuXwtfXF8nJybhw4QI8PT3Rv39/vP7663fcNzg4GJs3b8acOXMQFRUFb29vTJ06FQsWLGhzXZ06dcKhQ4fw5ptv4pVXXkFeXh58fX0RHR2NTz75xGDbhQsXYuHChfr5c+6++26kpqYaHN0jIjKVwvIaHLp0AxmXbuDgpRs4kVsKjfbOQcbPTYku/q7o4ueGLv6u6OrvhnAvJdJ2bMWDD8ZY5BgUe8NwYyMkEglmzZqFWbNmtWr/YcOGIT09vcnHd+zY0ej6lgwaDgwMxEcffYSPPvqoyW1uvxqqKREREbyFAhG1Sl5pNfacK0ba+Ws4eOkGcq43P2+KVAJ08XNDzyB39Az2QM8gd/QIcIeHc8PwYs2T99kihhsiIrJJpdUa7L9wDXuzirEnqxgXipq/cW4nHxf0D/dCVKgnegW5o3uAO5wUsnaqlkyJ4cYOTJ8+HWvXrm30sQkTJmDlypXtXBERkekJgoAzBeVIOVmA1MxCHLtSgqaGyygcpIgK8UB0uDeiw73QP8wTHVzFuTKWTI/hxg4sWbIEs2fPbvSx22dpJiKyJrVaHdKzr2PrqUKknM7H5evVjW4nk0rQL9QTQzr7YEhnH/QN9WzTPC9k2Rhu7ICfnx/8/PzELoOIyCQ0Wh12nS3CpmN5SM0sRGl14+Nduvq7YkhnHwzt7INBHb1NehNHsmwMN43Q6XRil0AmUv9echAykXUTBAGHcm7g58O52HQsDzeqGgYaB6kEd3fqgBGR/oiL9EewZ+O35SHbx3BzC4VCAaBuxl9XV1eRqyFTqJ+9mVcyEFmnrMJybDx8Fb8czW30lJOr0gH3dfPFiEh/3NfNDx5OPDpDDDcGHBwc4OPjg9zcXACAq6trmyexI3HodDpUVFQgNzcXJSUl0Got8665RNRQjUaLTcfy8M2BSziUU9LgcUe5FHE9/DGmbzDu6eoDpQOvaCJDDDe3qb9JZH3AIetWUlKCgoICaLVa/V3BicgyXSyuxLoDl/BDxhWU3HbaSSoBhnT2wZi+wYjvFQBXJX+WqWn87riNRCJBeHg4ysrKsHv3bri6uhp9w0hzEQQBOTk5CAsLs4tpvNvar0ajgU6ng06nw/Xr1xEQEGAx7yUR1anV6rD1dAHW7s/BnqziBo93D3DD4wNCMToqEH5ujiJUSNaI4aYJkZGRuHbtGtLT06HT6SxiQKpOp0NBQQGqq6vt4nSZqfoVBAH+/v544IEHIJPx8DWRJahWa/FDxmV8tusCrtwwHEujcJDiod6BmHB3GPqHednFf+bItBhumiCTyXDPPffA398fpaWlFnEFVW1tLQ4dOoT+/fvbxekVU/Xr5OSE0NBQ+Pj4mLA6ImqN0ioNvkrLxpp92bh22521wzs446mYMIyNDoW3i0KkCskW2P5vyDaQyWTo0aOH2GXoaTQa3LhxA7GxsXZxYzZ765fIluWX1uCLPRew7kAOKtWGA/zv6+aLZ4Z0xNDOPpBKeZSG2o7hhoiIzKagrAaf7MrE9wcvG9xxWyoBHu4ThOnD7kJkEGdKJ9NiuCEiIpMrrdbgv5ekeO3gHtRobp7WVzpI8cSAUEy7pxPCOjiLWCHZMoYbIiIymWq1Fmv2ZeOTHVkoq5ECqAs2rkoHTIoNx5QhHeHrxhtUknkx3BARUZtptDp8f/AyPth6DoXlKv16uUyCSbERmDm8MwcJU7thuCEiojbZl1WMhb+eRFZhhX6dVAIM9NHhnUnDEO7LMTXUvhhuiIioVfJLa/Dm5tP479GrButHRvrjpfvvwrmDuxDEm1eSCBhuiIjIKBqtDl/uy8a/Us4aXNYdFeqJhQ9HIjrcCxqNBudErJHsG8MNERG12P4L17DwlxM4W3DzFJSXsxxzH+iOx6NDOU8NWQSGGyIiuqOSKjWW/O8Ufjp086bCEgnw5KAwzInvBk9nDhYmy8FwQ0REzdpyIg8LNp5EccXNq6CiQjywdEwv9AnxFK8woiYw3BARUaOuV6rxxi8nsOlYnn6du6MD5j7QA38byFNQZLkYboiIqIHtmYWY8+Mxg6M1cT388OZjveHv7ihiZUR3xnBDRER6VepavLX5NNbuz9Gv83SWY/EjPfFIVBAkEh6tIcvHcENERACA03llmLnuEC4UVerX3dfNF+/8tQ/8eLSGrAjDDRGRnRMEAd8cyMGS/52CurbuXlCOcinmPxSJCTFhPFpDVofhhojIjpXVaDBvw3FsOn5z0HDPIHf8e3w/3OXrKmJlRK3HcENEZKfO5Jdj+toMXCy+eRrq6cERmPdgdygdZCJWRtQ2DDdERHbolyO5mLvhOKo1dbdPcHd0wDtjozCqV4DIlRG1HcMNEZEdqdXq8Obm0/jP3mz9up5B7lg5IRqh3s7iFUZkQgw3RER2orRKg5nrDmFPVrF+3djoEPxjTC84ynkaimwHww0RkR3IKqzAtK8O6sfXyGUSLHqkJ54cxKuhyPYw3BAR2bjd54ow45tDKK+pBQB0cFFg5cRoDIzwFrkyIvNguCEismHf/3EZr/98HLU6AQDQI9Adn0+KRogXx9eQ7WK4ISKyQYIgYFnKWXy4LUu/bkSkP5aP6wsXJT/6ybbxO5yIyMaoa3WYu+EYfjqcq183ZUgEFjwUCRnv5E12gOGGiMiGVKlr8fzaQ9h5tggAIJEAbzwUiWeGdhS5MqL2w3BDRGQjSqrUmLLmDxzOKQEAKB2k+OBv/TgxH9kdhhsiIhuQX1qDiV8cwLnCCgCAm9IBqyYPQEynDiJXRtT+GG6IiKzc5etVeHLVfly+Xg0A8HFV4qtnBiEyyF3kyojEwXBDRGTFLhZX4qnP9+NqaQ0AIMzbGV9PHYTwDi4iV0YkHoYbIiIrda6gHE+tOoDCchUA4C5fF3zz7N0I8HAUuTIicTHcEBFZobMF5Rj/2X5cq1QDALoHuOHrqTHwdVOKXBmR+BhuiIisTFZhOZ78/Gaw6RXsjq+fiYGXi0LkyogsA8MNEZEVOV9UgfGfH0BxRV2w6RPiga+nxsDDSS5yZUSWQyp2AURE1DIXiysx/rP9KPpzjE39ERsGGyJDDDdERFYgt6QaT32+Xz94ODLQHWunxsDDmcGG6HYMN0REFq6oXIWJqw7oL/fuHuCGb56Ngaczx9gQNYbhhojIgpVWaTBpdTouFFcCADr6uODrqRw8TNQchhsiIgtVpa7FlDXpOJ1XBgAI8nDE2md5uTfRnTDcEBFZoFqtDi+sO4xDf94Es4OLAmufjUGwp5O4hRFZAYYbIiILIwgC5v98AqmZhQDqboL51dRB6OTrKnJlRNaB4YaIyML8a+s5rD94GQCgkEnx6aRo9AzyELkqIusherhZsWIFIiIi4OjoiJiYGKSnpze7/fLly9GtWzc4OTkhNDQUL7/8MmpqatqpWiIi81p3IAf/Tj2nX37/iSgMvstHxIqIrI+o4Wb9+vVITExEUlISDh06hKioKMTHx6OwsLDR7detW4e5c+ciKSkJp0+fxhdffIH169fj9ddfb+fKiYhMb+fZIrzxywn98sKHIzE6KkjEioisk6jhZtmyZZg2bRqmTJmCyMhIrFy5Es7Ozli9enWj2+/btw9DhgzBk08+iYiICIwcORLjx4+/49EeIiJLdya/HDO/OQStTgAATLunI54Z2lHkqoisk2j3llKr1cjIyMC8efP066RSKeLi4pCWltboPoMHD8batWuRnp6OQYMG4cKFC9i8eTMmTpzY5OuoVCqoVCr9cllZ3SWVGo0GGo3GRN20j/p6ra3u1mK/to393lRUrsKU/6SjQlULABjRww+z4zpb9b8N31/bJka/xryWRBAEwYy1NOnq1asIDg7Gvn37EBsbq1//6quvYufOnThw4ECj+/373//G7NmzIQgCamtrMX36dHzyySdNvs6iRYuwePHiBuvXrVsHZ2fntjdCRNQGai3w4UkZciolAIBQFwEv9tRCIRO5MCILU1VVhSeffBKlpaVwd3dvdluruiv4jh078NZbb+Hjjz9GTEwMsrKyMGvWLCxduhRvvPFGo/vMmzcPiYmJ+uWysjKEhoZi5MiRd/zHsTQajQYpKSkYMWIE5HLbv58M+7Vt7Lfuku+XfziOnMp8AECghyO++3sM/Gxgkj6+v7ZNjH7rz7y0hGjhxsfHBzKZDAUFBQbrCwoKEBAQ0Og+b7zxBiZOnIhnn30WANC7d29UVlbiueeew/z58yGVNhxCpFQqoVQ2/KCQy+VW+w1ozbW3Bvu1bfbc78c7srDpeF2wcVHI8J8pAxHsbVtz2djz+2sP2rNfY15HtAHFCoUC0dHRSE1N1a/T6XRITU01OE11q6qqqgYBRiarO3Yr0tk1IqJW2XqqAO/+dka//K9xfdE9wLqOJhNZKlFPSyUmJmLy5MkYMGAABg0ahOXLl6OyshJTpkwBAEyaNAnBwcFITk4GAIwePRrLli1Dv3799Kel3njjDYwePVofcoiILN25gnK8tP4I6v9P9sqIrhjZs/Ej1kRkPFHDzbhx41BUVISFCxciPz8fffv2xZYtW+Dv7w8AyMnJMThSs2DBAkgkEixYsAC5ubnw9fXF6NGj8eabb4rVAhGRUcprNHju6wz9lVEP9Q5Ewv/rLHJVRLZF9AHFCQkJSEhIaPSxHTt2GCw7ODggKSkJSUlJ7VAZEZFp6QTg1Q0ncLG4EgAQGeiOdx/vA4lEInJlRLZF9HBDRGQvUq9KsDWnCADg4STHpxOj4azgxzCRqYl+bykiInuwJ+saNuXUfeRKJMAHf+uLUG/OtUVkDgw3RERmlltSjcQfjkFA3emnl+O64r5ufiJXRWS7GG6IiMxIo9UhYd0h3Kiqmzp+eDcfJAznAGIic2K4ISIyo3d/O4PDOSUAAG+lgHf/2htSKQcQE5kTww0RkZlsPVWAz3ZdAADIZRI83VULDyf7mb2WSCwMN0REZnDlRhVe+eGofvm1+K4It607KxBZLIYbIiITq9Xq8OK3h1FaXTfOJr6nPybdHSZyVUT2gxMsEBGZ2Aep53Doz3E2IV5OeGdsFDhPH1H74ZEbIiIT2n/hGj7angUAkEkl+HB8P46zIWpnDDdERCZSUqXGy7fcEDNxRFf0C/MStygiO8RwQ0RkAoIgYO6G48grrQEA3N3JG9OH3SVyVUT2ieGGiMgEfjh4BVtO5gMAPJ3l+Ne4vpBxPhsiUTDcEBG10ZUbVVjyv1P65bf/0geBHk4iVkRk3xhuiIjaQKcT8OqPx1ChqgUAjI0OwaheASJXRWTfGG6IiNrg6/2XsO/8NQBAkIcjFo6OFLkiImK4ISJqpYvFlUj+v9P65XfGRsHdkZd9E4mN4YaIqBW0OgGvfH8ENRodAGDi3eEY2sVH5KqICGC4ISJqlc93X9DPQhzewRlzH+gubkFEpMdwQ0RkpDP55Vj2+1kAgEQCvPd4FFyUvJsNkaVguCEiMoJGq0Pi90eg1tadjpp2TycMjPAWuSoiuhXDDRGRET7aloWTV8sAAJ39XJE4oqvIFRHR7RhuiIha6ERuqcFNMZc9EQVHuUzkqojodgw3REQtUKvVYe5Px6DV1d0Vc+bwzugT4iluUUTUKIYbIqIWWLMvGydy605HdfN3Q8LwziJXRERNYbghIrqDKzeq8P4tV0e99ZfeUDjw45PIUvGnk4ioGYIgYOEvJ1Gt0QIAJsSEIzrcS+SqiKg5DDdERM3YdDwP2zILAQB+bkrMGdVN5IqI6E4YboiImlBarcHi/57SLy95tCfvHUVkBRhuiIia8M8tmSgqVwEA4nr4I75ngMgVEVFLMNwQETXij+zrWHcgBwDgopBhyaM9IZFIRK6KiFqC4YaI6DbqWh1e/+m4fnl2fDcEeTqJWBERGYPhhojoNp/uPI9zhRUAgKgQD0yKjRC3ICIyCsMNEdEtLhRV4MNbbrHw1l96Qybl6Sgia8JwQ0T0J0EQsOi/p6Curbvj97NDO6JnkIfIVRGRsRhuiIj+tPV0IXadLQIABHk4YlZcF5ErIqLWYLghIgJQo9Fi6f9uzmkz/6FIOCscRKyIiFqL4YaICMCq3ReQc70KABDbqQMe7M05bYisFcMNEdm9qyXVWLH9PIC6QcSLHuGcNkTWjOGGiOzeW5tP62+MOfHucHQLcBO5IiJqC4YbIrJraeev4X/H8gAA3i4KvDyiq8gVEVFbMdwQkd2q1eqw+L8n9cuvxneDhxNvjElk7RhuiMhufXMgB5n55QCAPiEeeGJAqMgVEZEpMNwQkV26VqHC+7+f0S8veqQnpJyJmMgmMNwQkV167/ezKKupBQD8tX8I+od5iVwREZkKww0R2Z0TuaX47o8cAICr0gGvPdBN5IqIyJQYbojIrgiCgKX/OwVBqFuedX8X+Lk5ilsUEZkUww0R2ZXU04U4cPE6ACCigzMmD44QtyAiMjmGGyKyG7VaHZL/77R+ee4D3aFw4Mcgka3hTzUR2Y3v/riM80WVAIAB4V6I78n7RxHZIoYbIrILFapaLN96Vr8878EevH8UkY1iuCEiu/DZzvMorlADAB7sHYDocF76TWSrGG6IyObll9bgs90XAABymQSvxncXuSIiMieGGyKyectSzqBGowMATLg7HBE+LiJXRETmxHBDRDYtM78MP2RcAQC4OTrgxf/XReSKiMjcGG6IyKYlb87UT9g3c3hneLkoxC2IiMxO9HCzYsUKREREwNHRETExMUhPT292+5KSEsycOROBgYFQKpXo2rUrNm/e3E7VEpE12X2uCDvPFgEAgj2d8DQn7COyCw5ivvj69euRmJiIlStXIiYmBsuXL0d8fDzOnDkDPz+/Btur1WqMGDECfn5++PHHHxEcHIxLly7B09Oz/YsnIoum1Ql4a3OmfnlOfDc4ymUiVkRE7UXUcLNs2TJMmzYNU6ZMAQCsXLkSmzZtwurVqzF37twG269evRrXr1/Hvn37IJfLAQARERHtWTIRWYmNh3NxOq8MANAr2B2PRAWJXBERtRfRTkup1WpkZGQgLi7uZjFSKeLi4pCWltboPr/++itiY2Mxc+ZM+Pv7o1evXnjrrbeg1Wrbq2wisgLqWh3+dcuEfa8/2ANSKSfsI7IXoh25KS4uhlarhb+/v8F6f39/ZGZmNrrPhQsXsG3bNjz11FPYvHkzsrKyMGPGDGg0GiQlJTW6j0qlgkql0i+XldX9T06j0UCj0Ziom/ZRX6+11d1a7Ne2mbPfdQdycOVGNQBgaOcOGBjmIfq/K99f28Z+2+81W0IiCPXXEbSvq1evIjg4GPv27UNsbKx+/auvvoqdO3fiwIEDDfbp2rUrampqcPHiRchkdefOly1bhnfffRd5eXmNvs6iRYuwePHiBuvXrVsHZ2dnE3VDRJZCrQWWHpahTFN3pCaxdy3CXUUuiojarKqqCk8++SRKS0vh7u7e7LaiHbnx8fGBTCZDQUGBwfqCggIEBDR+M7vAwEDI5XJ9sAGAHj16ID8/H2q1GgpFw0s8582bh8TERP1yWVkZQkNDMXLkyDv+41gajUaDlJQUjBgxQj/myJaxX9tmrn5X7clGmabulNSIHn54/om+JnvutuD7a9vYr/nVn3lpCdHCjUKhQHR0NFJTUzFmzBgAgE6nQ2pqKhISEhrdZ8iQIVi3bh10Oh2k0rrhQmfPnkVgYGCjwQYAlEollEplg/VyudxqvwGtufbWYL+2zZT9ltdo8NnuiwAAiQSYHd/d4v4t+f7aNvZr3tdqKVHnuUlMTMTnn3+OL7/8EqdPn8bzzz+PyspK/dVTkyZNwrx58/TbP//887h+/TpmzZqFs2fPYtOmTXjrrbcwc+ZMsVogIguyek82blTVnZd/JCoI3QLcRK6IiMQg6qXg48aNQ1FRERYuXIj8/Hz07dsXW7Zs0Q8yzsnJ0R+hAYDQ0FD89ttvePnll9GnTx8EBwdj1qxZeO2118RqgYgsREmVGqv+vDmmTCrBy3FdRa6IiMQiargBgISEhCZPQ+3YsaPButjYWOzfv9/MVRGRtfl01wWUq2oBAI9Hh/DmmER2TPTbLxARtVVheQ3W7M0GAChkUrxwP2+OSWTPGG6IyOp9vP08qjV1k3k+GROGYE8nkSsiIjEx3BCRVcstqca6AzkAACe5DDOHdxa5IiISG8MNEVm1D1PPQa3VAQCeHhIBX7eGUz8QkX1huCEiq3WxuBI/ZFwBALgpHfD3ezuJXBERWQKGGyKyWh+mnoNWV3cHmWfv6QRP58Yn8yQi+8JwQ0RWKbu4EhuP5AIAPJ3leGZohLgFEZHFYLghIqu0YnsW/jxog2eHdoSbo/1MeU9EzWO4ISKrc/l6FX46XHfUxt3RAZMGR4hbEBFZFIYbIrI6H+/I0o+1eWZoR7jzqA0R3YLhhoisypUbVfjxliukpgzuKHJFRGRpGG6IyKqs3HkeGm3dUZunh0TAw5lHbYjIEMMNEVmN/NIafP9H3VEbF4UMzwzhURsiaojhhoisxsqd5/WzEU8aHAEvF85rQ0QNMdwQkVUoLKvBt+k37yH17FAetSGixjHcEJFV+GzXBahq647aTIwNRwdX3kOKiBpnVLiZNGkSysvL9ctHjx6FRqMxeVFERLcqrlBh7YFLAAClgxTT7uE9pIioaUaFm2+++QbV1dX65XvuuQeXL182eVFERLf6fPcF1Gjqjto8FRPOO38TUbOMCjeCIDS7TERkatcr1fg6re6ojcJBir8P41EbImoex9wQkUVbveciqtRaAMD4gaHwd3cUuSIisnQOxu5w6tQp5OfnA6g7cpOZmYmKigqDbfr06WOa6ojIrpXXaPBlWjYAQC6T4O/D7hK3ICKyCkaHm/vvv9/gdNTDDz8MAJBIJBAEARKJBFqt1nQVEpHdWncgB+U1tQCAv/QLQZCnk8gVEZE1MCrcXLx40Vx1EBEZUNVq8cWeus8ciQR4jmNtiKiFjAo34eHh5qqDiMjAz4dyUViuAgDERwbgLl9XkSsiImth9GkpADh37hx++eUXZGdnQyKRoGPHjhgzZgw6deL/rIio7bQ6AZ/uuqBfnn4fx9oQUcsZHW6Sk5OxcOFC6HQ6+Pn5QRAEFBUVYe7cuXjrrbcwe/Zsc9RJRHbkt5P5uFhcCQAYfFcH9A31FLcgIrIqRl0Kvn37dixYsADz589HcXEx8vLykJ+frw83c+fOxa5du8xVKxHZAUEQsHLnef3y8zxqQ0RGMurIzcqVK/Hss89i0aJFBuu9vb2xZMkS5Ofn45NPPsG9995ryhqJyI7sO38Nx66UAgB6BrljaGcfkSsiImtj1JGb9PR0TJw4scnHJ06ciP3797e5KCKyX5/sMDxqI5FIRKyGiKyRUeGmoKAAERERTT7esWNH/QR/RETGOn6lFHuyigEA4R2c8UCvQJErIiJrZFS4qampgUKhaPJxuVwOtVrd5qKIyD7dOtbm7/feBZmUR22IyHhGXy21atUquLo2Pt9EeXl5mwsiIvt0sbgSm0/kAQB83ZT4S/9gkSsiImtlVLgJCwvD559/fsdtiIiM9dmu86i/s8szQzrCUS4TtyAislpGhZvs7GwzlUFE9qywrAYbMnIBAG5KBzx1N/+TREStZ9SYm23btiEyMhJlZWUNHistLUXPnj2xe/dukxVHRPbhi70XodbqAAATYsPh7igXuSIismZGhZvly5dj2rRpcHd3b/CYh4cH/v73v2PZsmUmK46IbF+FqhbrDuQAABQOUkwZEiFuQURk9YwKN0ePHsWoUaOafHzkyJHIyMhoc1FEZD9+PJSL8ppaAMBf+gXDz81R5IqIyNoZPc+NXN704WIHBwcUFRW1uSgisg9aAfhy3yX98tShHUWshohshVHhJjg4GCdOnGjy8WPHjiEwkJNuEVHLHL8uwZWSGgDAfd180cXfTeSKiMgWGBVuHnzwQbzxxhuoqalp8Fh1dTWSkpLw8MMPm6w4IrJt26/e/Ah6dmgnESshIlti1KXgCxYswE8//YSuXbsiISEB3bp1AwBkZmZixYoV0Gq1mD9/vlkKJSLbcjinBNkVdTMQdw9ww5DOHUSuiIhshVHhxt/fH/v27cPzzz+PefPmQfhzxi2JRIL4+HisWLEC/v7+ZimUiGzLF3uz9V8/e08n3iCTiEzG6NsvhIeHY/Pmzbhx4waysrIgCAK6dOkCLy8vc9RHRDYo51oVUk4XAgB8XRUYHcWxekRkOkaHm3peXl4YOHCgKWshIjuxeu9F6P681cLEu8OgdOCtFojIdIwaUExE1Fal1Rp8f/AyAEAuFfC3gSEiV0REtobhhoja1bfpOahSawEAMb4CvJwVIldERLaG4YaI2o1Gq8OaPwcSSyTAsECduAURkU1iuCGidrPpWB7yy+rmybq/my/8nEQuiIhsEsMNEbULQRCwas8F/fKUIeEiVkNEtozhhojaxf4L13EitwwA0CfEAwPDOX0EEZkHww0RtYsvbjlqM3VoR07aR0Rmw3BDRGZ36VolUjPrJu0L9HDEg705aR8RmQ/DDRGZ3VdplyDUT9oXGw65jB89RGQ+/IQhIrOqVNXi+z/qJu1TOkjxt4FhIldERLaO4YaIzOqnQ1dQrqoFAIzpGwxvF07aR0TmxXBDRGaj0wlYsy9bvzx5cIRotRCR/WC4ISKz2ZNVjPNFlQCAmI7eiAxyF7kiIrIHFhFuVqxYgYiICDg6OiImJgbp6ekt2u+7776DRCLBmDFjzFsgEbXKrUdtpgyJEK0OIrIvooeb9evXIzExEUlJSTh06BCioqIQHx+PwsLCZvfLzs7G7Nmzcc8997RTpURkjIvFldj25+XfwZ5OiOvhL3JFRGQvRA83y5Ytw7Rp0zBlyhRERkZi5cqVcHZ2xurVq5vcR6vV4qmnnsLixYvRqVOndqyWiFrqq7Rs/dcTY8PhwMu/iaidOIj54mq1GhkZGZg3b55+nVQqRVxcHNLS0prcb8mSJfDz88PUqVOxe/fuZl9DpVJBpVLpl8vK6qZ/12g00Gg0beygfdXXa211txb7tV4Vqlp8f7Du8m9HuRR/7RvYoC9b6rcl2K9tY7/t95otIWq4KS4uhlarhb+/4eFqf39/ZGZmNrrPnj178MUXX+DIkSMteo3k5GQsXry4wfrff/8dzs7ORtdsCVJSUsQuoV2xX+uzK0+CSpUMANDPqxb7djTdky30awz2a9vYr/lUVVW1eFtRw42xysvLMXHiRHz++efw8fFp0T7z5s1DYmKifrmsrAyhoaEYOXIk3N2t68oNjUaDlJQUjBgxAnK5XOxyzI79WiedTsDyf+8FUPdBNP/xIegW4NZgO1vpt6XYr21jv+ZXf+alJUQNNz4+PpDJZCgoKDBYX1BQgICAgAbbnz9/HtnZ2Rg9erR+nU6nAwA4ODjgzJkzuOuuuwz2USqVUCqVDZ5LLpdb7TegNdfeGuzXumw/U4iL1+qCTWynDugV6t3s9tber7HYr21jv+Z9rZYSdYSfQqFAdHQ0UlNT9et0Oh1SU1MRGxvbYPvu3bvj+PHjOHLkiP7PI488guHDh+PIkSMIDQ1tz/KJqBFr9mbrv36al38TkQhEPy2VmJiIyZMnY8CAARg0aBCWL1+OyspKTJkyBQAwadIkBAcHIzk5GY6OjujVq5fB/p6engDQYD0Rtb/zRRXYebYIABDixcu/iUgcooebcePGoaioCAsXLkR+fj769u2LLVu26AcZ5+TkQCrlJaRE1uCrW2+1EBsBmVQiXjFEZLdEDzcAkJCQgISEhEYf27FjR7P7rlmzxvQFEZHRKlS12HAoFwDgJJfhiQE8TUxE4uAhESIyiY2Hc1FRf/fvfkHwcLafQZVEZFkYboiozQRBwNdpl/TLE+4OF7EaIrJ3DDdE1GZ/ZN/AmYJyAEB0uBd6BnmIXBER2TOGGyJqs6/33zxqM5FHbYhIZAw3RNQmheU12HIiDwDQwUWBB3o3nICTiKg9MdwQUZusT78MjVYAAIwbGAqlg0zkiojI3jHcEFGr1Wp1WJeeAwCQSIAnY8JEroiIiOGGiNogNbMQeaU1AID7u/shxMtZ5IqIiBhuiKgN1u7n5d9EZHkYboioVS4UVWD3uWIAQHgHZ9zbxVfkioiI6jDcEFGrrN2fo/96Qkw4pLyPFBFZCIYbIjJalboWP2RcBgAoHaQYGx0ickVERDcx3BCR0f579CrKa+ruIzU6KgheLgqRKyIiuonhhoiMIggCvkrjjMREZLkYbojIKIcvl+Dk1TIAQFSIB6JCPcUtiIjoNgw3RGSUtbz7NxFZOIYbImqxG5Vq/O943X2kPJzkGB0VJHJFREQNMdwQUYttOHQF6lodAGBsdAgc5byPFBFZHoYbImoRQRD095ECgPGDeB8pIrJMDDdE1CL7L1zHhaJKAEBMR2909nMVuSIiosYx3BBRi9x61IZ3/yYiS8ZwQ0R3dK1ChS0n6gYSe7soMKpXgMgVERE1jeGGiO7ox4wr0GgFAMDj0SFQOnAgMRFZLoYbImqWTifgWw4kJiIrwnBDRM1Ku3AN2deqAABDOndAhI+LyBURETWP4YaImrXuwC0DiQdxRmIisnwMN0TUpKJyFX47mQ8A8HFVYESkv8gVERHdGcMNETXph4zLqNX9OZB4QCgUDvzIICLLx08qImqUTifgu/TL+uXxAzmQmIisA8MNETVqT1Yxcq7XDSS+p4sPwjo4i1wREVHLMNwQUaO+OXBJ//VTnJGYiKwIww0RNVBQVoOtpwsBAL5uStzfgwOJich6MNwQUQPf/3EZ2j8HEo8bEAq5jB8VRGQ9+IlFRAZ0OgHf/VE3kFgiAf42KFTkioiIjMNwQ0QGdmcVI7ekGgAwrKsvQrw4kJiIrAvDDREZWP/HzRmJ/8bLv4nICjHcEJFecYUKKacKAAA+rkrc38NP5IqIiIzHcENEej8dugKNtm4g8djoEA4kJiKrxE8uIgIACMLNgcQAMG4gBxITkXViuCEiAED6xeu4UFQJALi7kzc6+riIXBERUesw3BARAGD9LUdtxg/iQGIisl4MN0SE0ioNNh3PAwB4OMkR3zNA5IqIiFqP4YaI8MvRXKhqdQCAx/oFw1EuE7kiIqLWY7ghsnOCIODb9JunpDgjMRFZO4YbIjt3PLcUp/PKAAB9Qz3RPcBd5IqIiNqG4YbIzhkcteHl30RkAxhuiOxYpaoWvx7JBQC4KGQYHRUkckVERG3HcENkxzYdy0OlWgsAeKRvEFyUDiJXRETUdgw3RHbs21tukjmON8kkIhvBcENkp87kl+NwTgkAoHuAG6JCPMQtiIjIRBhuiOzUd7cctfnbwFBIJBIRqyEiMh2GGyI7pKrV4ufDdQOJFQ5SPNYvROSKiIhMh+GGyA6lnCpASZUGAPBArwB4OMtFroiIyHQYbojs0PcHr+i/HjeAc9sQkW1huCGyM1dLqrH7XBEAIMTLCXd36iByRUREpsVwQ2RnNmRcgSDUff14dCikUg4kJiLbYhHhZsWKFYiIiICjoyNiYmKQnp7e5Laff/457rnnHnh5ecHLywtxcXHNbk9EN+l0An7IqDslJZEAYwdwIDER2R7Rw8369euRmJiIpKQkHDp0CFFRUYiPj0dhYWGj2+/YsQPjx4/H9u3bkZaWhtDQUIwcORK5ubntXDmR9Tlw8TpyrlcBAIZ29kGwp5PIFRERmZ7o4WbZsmWYNm0apkyZgsjISKxcuRLOzs5YvXp1o9t/8803mDFjBvr27Yvu3btj1apV0Ol0SE1NbefKiazPDwdv3iTzCQ4kJiIbJeqNZNRqNTIyMjBv3jz9OqlUiri4OKSlpbXoOaqqqqDRaODt7d3o4yqVCiqVSr9cVlYGANBoNNBoNG2ovv3V12ttdbcW+zWt8hoNNp/IAwB4ODlgeBdvUf9t+f7aNvZr28To15jXkghC/dDC9nf16lUEBwdj3759iI2N1a9/9dVXsXPnThw4cOCOzzFjxgz89ttvOHnyJBwdHRs8vmjRIixevLjB+nXr1sHZ2bltDRBZkb0FEnx/QQYAuCdAh7EddSJXRETUclVVVXjyySdRWloKd3f3Zre16lsAv/322/juu++wY8eORoMNAMybNw+JiYn65bKyMv04nTv941gajUaDlJQUjBgxAnK57U+6xn5N64tP9wOoO3L5ymOD0TNI3O9/vr+2jf3aNjH6rT/z0hKihhsfHx/IZDIUFBQYrC8oKEBAQECz+7733nt4++23sXXrVvTp06fJ7ZRKJZRKZYP1crncar8Brbn21mC/bXcmvxzHrtR9MEQGuqNvuOXMbcP317axX9vWnv0a8zqiDihWKBSIjo42GAxcPzj41tNUt3vnnXewdOlSbNmyBQMGDGiPUomsmuFAYl7+TUS2TfTTUomJiZg8eTIGDBiAQYMGYfny5aisrMSUKVMAAJMmTUJwcDCSk5MBAP/85z+xcOFCrFu3DhEREcjPzwcAuLq6wtXVVbQ+iCyVulZ38yaZMike7RssckVEROYlergZN24cioqKsHDhQuTn56Nv377YsmUL/P39AQA5OTmQSm8eYPrkk0+gVqsxduxYg+dJSkrCokWL2rN0IquwLbMQ1yrVAICRPf3h5aIQuSIiIvMSPdwAQEJCAhISEhp9bMeOHQbL2dnZ5i+IyIZwbhsisjeiT+JHROZTUFaD7WfqZvsO8nDEkM4+IldERGR+DDdENmzDoSvQ/TmT1djoEMh4k0wisgMMN0Q2ShAE/HDwin55bDRPSRGRfWC4IbJRBy/dwMXiSgBAbKcOCOvAGbmJyD4w3BDZqO//uGUg8UDObUNE9oPhhsgGVahqsel43U0y3ZQOGNUzUOSKiIjaD8MNkQ3afCwPVWotAOCRvkFwUshEroiIqP0w3BDZoO85tw0R2TGGGyIbc76oAgcv3QAAdPN3Q58QD5ErIiJqXww3RDbm1qM2jw8IgUTCuW2IyL4w3BDZEI1Whw0ZdTfJdJBK8Fg/3iSTiOwPww2RDdl5pgjFFSoAQFwPf3RwVYpcERFR+2O4IbIhP2bcnJH48QGc24aI7BPDDZGNuF6pRmpmAQDA102JYV19Ra6IiEgcDDdENuLXI7nQaOvukvmXfsFwkPHHm4jsEz/9iGzEj4dunpL6azRPSRGR/WK4IbIBp/PKcCK3DAAQFeKBrv5uIldERCQehhsiG7DhloHEY3nUhojsHMMNkZXTaHXYeKRubhuFTIrRUUEiV0REJC6GGyIrVze3jRoAMCLSH57OCpErIiISF8MNkZX7kaekiIgMMNwQWbHb57a5p4uPyBUREYmP4YbIinFuGyKihvhJSGTFOLcNEVFDDDdEVopz2xARNY7hhshKcW4bIqLGMdwQWSHObUNE1DSGGyIrxLltiIiaxnBDZIU4tw0RUdMYboisDOe2ISJqHsMNkZXh3DZERM3jpyKRleHcNkREzWO4IbIinNuGiOjOGG6IrAjntiEiujOGGyIrwbltiIhahuGGyEpwbhsiopZhuCGyEpzbhoioZRhuiKwA57YhImo5hhsiK8C5bYiIWo6fkERWgHPbEBG1HMMNkYXj3DZERMZhuCGycJzbhojIOAw3RBaMc9sQERmP4YbIgnFuGyIi4zHcEFkwzm1DRGQ8hhsiC8W5bYiIWofhhshCcW4bIqLW4aclkYXi3DZERK3DcENkgTi3DRFR6zHcEFkgzm1DRNR6DDdEFoZz2xARtQ3DDZGF2XWumHPbEBG1AcMNkYX56fBV/dc8JUVEZDyGGyILUqEBtp8pAsC5bYiIWovhhsiCHCqWcG4bIqI24icnkQU5UHTzR5Jz2xARtY5FhJsVK1YgIiICjo6OiImJQXp6erPb//DDD+jevTscHR3Ru3dvbN68uZ0qJTKfzPxyXKmUAODcNkREbSF6uFm/fj0SExORlJSEQ4cOISoqCvHx8SgsLGx0+3379mH8+PGYOnUqDh8+jDFjxmDMmDE4ceJEO1dOZFo/cyAxEZFJiB5uli1bhmnTpmHKlCmIjIzEypUr4ezsjNWrVze6/QcffIBRo0Zhzpw56NGjB5YuXYr+/fvjo48+aufKiUzjeqUan+w4j+8z6ua2kcsknNuGiKgNHMR8cbVajYyMDMybN0+/TiqVIi4uDmlpaY3uk5aWhsTERIN18fHx2LhxozlLvaNfj17Fvqxis76GTqfD5ctS7N14ElKp6LnU7Oyh39JqDVIzC6Gu1enXjezBuW2IiNpC1HBTXFwMrVYLf39/g/X+/v7IzMxsdJ/8/PxGt8/Pz290e5VKBZVKpV8uK6u7X49Go4FGo2lL+QYOXryG7/64bLLna5oUaYW57fA6lsK++o301GH+A51N+r1pqep7tIdeAfZr69hv+71mS4gabtpDcnIyFi9e3GD977//DmdnZ5O9Tna2FBZwlo+skLNMQIyfgKEBOvg4An/s2SF2Se0qJSVF7BLaFfu1bezXfKqqqlq8rajhxsfHBzKZDAUFBQbrCwoKEBAQ0Og+AQEBRm0/b948g9NYZWVlCA0NxciRI+Hu7t7GDm4aUK5CSZXaZM/XmNraWuxLS8Pg2Fg4ONh8LrWLfiUSCcK8nKCUy6DRaJCSkoIRI0ZALpeLXZrZsV/bxn5tmxj91p95aQlRf2MoFApER0cjNTUVY8aMAVA3ziI1NRUJCQmN7hMbG4vU1FS89NJL+nUpKSmIjY1tdHulUgmlUtlgvVwuN+kbEuwtR7C3yZ6uURqNBtnOQGSwl9388NhTv/VM/b1p6divbWO/tq09+zXmdUT/73BiYiImT56MAQMGYNCgQVi+fDkqKysxZcoUAMCkSZMQHByM5ORkAMCsWbMwbNgwvP/++3jooYfw3Xff4eDBg/jss8/EbIOIiIgshOjhZty4cSgqKsLChQuRn5+Pvn37YsuWLfpBwzk5OQZXygwePBjr1q3DggUL8Prrr6NLly7YuHEjevXqJVYLREREZEFEDzcAkJCQ0ORpqB07djRY9/jjj+Pxxx83c1VERERkjXh5DxEREdkUhhsiIiKyKQw3REREZFMYboiIiMimMNwQERGRTWG4ISIiIpvCcENEREQ2heGGiIiIbArDDREREdkUhhsiIiKyKRZx+4X2JAgCAONunW4pNBoNqqqqUFZWZhd3nWW/to392jb2a9vE6Lf+93b97/Hm2F24KS8vBwCEhoaKXAkREREZq7y8HB4eHs1uIxFaEoFsiE6nw9WrV+Hm5gaJRCJ2OUYpKytDaGgoLl++DHd3d7HLMTv2a9vYr21jv7ZNjH4FQUB5eTmCgoIglTY/qsbujtxIpVKEhISIXUabuLu728UPTz32a9vYr21jv7atvfu90xGbehxQTERERDaF4YaIiIhsCsONFVEqlUhKSoJSqRS7lHbBfm0b+7Vt7Ne2WXq/djegmIiIiGwbj9wQERGRTWG4ISIiIpvCcENEREQ2heHGgly/fh1PPfUU3N3d4enpialTp6KioqLZfWpqajBz5kx06NABrq6u+Otf/4qCgoJGt7127RpCQkIgkUhQUlJihg6MY45+jx49ivHjxyM0NBROTk7o0aMHPvjgA3O30qgVK1YgIiICjo6OiImJQXp6erPb//DDD+jevTscHR3Ru3dvbN682eBxQRCwcOFCBAYGwsnJCXFxcTh37pw5WzCKKfvVaDR47bXX0Lt3b7i4uCAoKAiTJk3C1atXzd1Gi5n6/b3V9OnTIZFIsHz5chNX3Tbm6Pn06dN45JFH4OHhARcXFwwcOBA5OTnmasEopu63oqICCQkJCAkJgZOTEyIjI7Fy5UpztmAUY/o9efIk/vrXvyIiIqLZ71Vj/w1NRiCLMWrUKCEqKkrYv3+/sHv3bqFz587C+PHjm91n+vTpQmhoqJCamiocPHhQuPvuu4XBgwc3uu2jjz4qPPDAAwIA4caNG2bowDjm6PeLL74QXnzxRWHHjh3C+fPnha+//lpwcnISPvzwQ3O3Y+C7774TFAqFsHr1auHkyZPCtGnTBE9PT6GgoKDR7ffu3SvIZDLhnXfeEU6dOiUsWLBAkMvlwvHjx/XbvP3224KHh4ewceNG4ejRo8IjjzwidOzYUaiurm6vtppk6n5LSkqEuLg4Yf369UJmZqaQlpYmDBo0SIiOjm7Ptppkjve33k8//SRERUUJQUFBwr/+9S8zd9Jy5ug5KytL8Pb2FubMmSMcOnRIyMrKEn755Zcmn7M9maPfadOmCXfddZewfft24eLFi8Knn34qyGQy4ZdffmmvtppkbL/p6enC7NmzhW+//VYICAho9HvV2Oc0JYYbC3Hq1CkBgPDHH3/o1/3f//2fIJFIhNzc3Eb3KSkpEeRyufDDDz/o150+fVoAIKSlpRls+/HHHwvDhg0TUlNTLSLcmLvfW82YMUMYPny46YpvgUGDBgkzZ87UL2u1WiEoKEhITk5udPsnnnhCeOihhwzWxcTECH//+98FQRAEnU4nBAQECO+++67+8ZKSEkGpVArffvutGTowjqn7bUx6eroAQLh06ZJpim4Dc/V75coVITg4WDhx4oQQHh5uUeHGHD2PGzdOmDBhgnkKbiNz9NuzZ09hyZIlBtv0799fmD9/vgkrbx1j+71VU9+rbXnOtuJpKQuRlpYGT09PDBgwQL8uLi4OUqkUBw4caHSfjIwMaDQaxMXF6dd1794dYWFhSEtL0687deoUlixZgq+++uqO9+NoL+bs93alpaXw9vY2XfF3oFarkZGRYVCnVCpFXFxck3WmpaUZbA8A8fHx+u0vXryI/Px8g208PDwQExPTbO/twRz9Nqa0tBQSiQSenp4mqbu1zNWvTqfDxIkTMWfOHPTs2dM8xbeSOXrW6XTYtGkTunbtivj4ePj5+SEmJgYbN240Wx8tZa73ePDgwfj111+Rm5sLQRCwfft2nD17FiNHjjRPIy3Umn7FeE5jWMZvOkJ+fj78/PwM1jk4OMDb2xv5+flN7qNQKBp82Pv7++v3UalUGD9+PN59912EhYWZpfbWMFe/t9u3bx/Wr1+P5557ziR1t0RxcTG0Wi38/f0N1jdXZ35+frPb1/9tzHO2F3P0e7uamhq89tprGD9+vOj37TFXv//85z/h4OCAF1980fRFt5E5ei4sLERFRQXefvttjBo1Cr///jsee+wx/OUvf8HOnTvN00gLmes9/vDDDxEZGYmQkBAoFAqMGjUKK1aswL333mv6JozQmn7FeE5jMNyY2dy5cyGRSJr9k5mZabbXnzdvHnr06IEJEyaY7TVuJXa/tzpx4gQeffRRJCUlif4/I2o9jUaDJ554AoIg4JNPPhG7HLPIyMjABx98gDVr1kAikYhdTrvQ6XQAgEcffRQvv/wy+vbti7lz5+Lhhx+2qEG2pvThhx9i//79+PXXX5GRkYH3338fM2fOxNatW8UuzebY3V3B29srr7yCp59+utltOnXqhICAABQWFhqsr62txfXr1xEQENDofgEBAVCr1SgpKTE4mlFQUKDfZ9u2bTh+/Dh+/PFHAHVX3ACAj48P5s+fj8WLF7eys8aJ3W+9U6dO4f7778dzzz2HBQsWtKqX1vLx8YFMJmtw1VpjddYLCAhodvv6vwsKChAYGGiwTd++fU1YvfHM0W+9+mBz6dIlbNu2TfSjNoB5+t29ezcKCwsNjq5qtVq88sorWL58ObKzs03bhJHM0bOPjw8cHBwQGRlpsE2PHj2wZ88eE1ZvPHP0W11djddffx0///wzHnroIQBAnz59cOTIEbz33nsNTmm1p9b0K8ZzGoNHbszM19cX3bt3b/aPQqFAbGwsSkpKkJGRod9327Zt0Ol0iImJafS5o6OjIZfLkZqaql935swZ5OTkIDY2FgCwYcMGHD16FEeOHMGRI0ewatUqAHUfpjNnzrS5foG6SxSHDx+OyZMn48033zR5j3eiUCgQHR1tUKdOp0NqaqpBnbeKjY012B4AUlJS9Nt37NgRAQEBBtuUlZXhwIEDTT5nezFHv8DNYHPu3Dls3boVHTp0ME8DRjJHvxMnTsSxY8f0P6dHjhxBUFAQ5syZg99++818zbSQOXpWKBQYOHAgzpw5Y7DN2bNnER4ebuIOjGOOfjUaDTQaTYNxjzKZTH8USyyt6VeM5zSK2YcsU4uNGjVK6Nevn3DgwAFhz549QpcuXQwujb5y5YrQrVs34cCBA/p106dPF8LCwoRt27YJBw8eFGJjY4XY2NgmX2P79u0WcbWUIJin3+PHjwu+vr7ChAkThLy8PP2fwsLCdu3tu+++E5RKpbBmzRrh1KlTwnPPPSd4enoK+fn5giAIwsSJE4W5c+fqt9+7d6/g4OAgvPfee8Lp06eFpKSkRi8F9/T0FH755Rfh2LFjwqOPPmpRl4Kbsl+1Wi088sgjQkhIiHDkyBGD91KlUonS463M8f7eztKuljJHzz/99JMgl8uFzz77TDh37pzw4YcfCjKZTNi9e3e793c7c/Q7bNgwoWfPnsL27duFCxcuCP/5z38ER0dH4eOPP273/m5nbL8qlUo4fPiwcPjwYSEwMFCYPXu2cPjwYeHcuXMtfk5zYrixINeuXRPGjx8vuLq6Cu7u7sKUKVOE8vJy/eMXL14UAAjbt2/Xr6uurhZmzJgheHl5Cc7OzsJjjz0m5OXlNfkalhRuzNFvUlKSAKDBn/Dw8HbsrM6HH34ohIWFCQqFQhg0aJCwf/9+/WPDhg0TJk+ebLD9999/L3Tt2lVQKBRCz549hU2bNhk8rtPphDfeeEPw9/cXlEqlcP/99wtnzpxpj1ZaxJT91r/3jf259ftBTKZ+f29naeFGEMzT8xdffCF07txZcHR0FKKiooSNGzeau40WM3W/eXl5wtNPPy0EBQUJjo6OQrdu3YT3339f0Ol07dHOHRnTb1M/o8OGDWvxc5oT7wpORERENoVjboiIiMimMNwQERGRTWG4ISIiIpvCcENEREQ2heGGiIiIbArDDREREdkUhhsiIiKyKQw3REREZFMYboiIAEgkEmzcuFHsMojIBHhXcCIiAHl5efDy8hK7DCIyAd5+gYjsmlqthkKhELsMIjIhnpYiIouj0+mQnJyMjh07wsnJCVFRUfjxxx9btO/OnTsxaNAgKJVKBAYGYu7cuaitrdU/ft999yEhIQEvvfQSfHx8EB8fD4CnpYhsCU9LEZHFSU5Oxtq1a7Fy5Up06dIFu3btwoQJE+Dr64thw4Y1uV9ubi4efPBBPP300/jqq6+QmZmJadOmwdHREYsWLdJv9+WXX+L555/H3r1726EbImpvPC1FRBZFpVLB29sbW7duRWxsrH79s88+i6qqKqxbt67JfefPn48NGzbg9OnTkEgkAICPP/4Yr732GkpLSyGVSnHfffehrKwMhw4dMthXIpHg559/xpgxY8zSFxG1Hx65ISKLkpWVhaqqKowYMcJgvVqtRr9+/Zrd9/Tp04iNjdUHGwAYMmQIKioqcOXKFYSFhQEAoqOjTV84EVkMhhsisigVFRUAgE2bNiE4ONjgMaVSaZLXcHFxMcnzEJFlYrghIosSGRkJpVKJnJycZsfXNKZHjx7YsGEDBEHQH73Zu3cv3NzcEBISYo5yicgCMdwQkUVxc3PD7Nmz8fLLL0On02Ho0KEoLS3F3r174e7ujsmTJze574wZM7B8+XK88MILSEhIwJkzZ5CUlITExERIpbw4lMheMNwQkcVZunQpfH19kZycjAsXLsDT0xP9+/fH66+/3ux+wcHB2Lx5M+bMmYOoqCh4e3tj6tSpWLBgQTtVTkSWgFdLERERkU3hcVoiIiKyKQw3RGQ1pk+fDldX10b/TJ8+XezyiMhC8LQUEVmNwsJClJWVNfqYu7s7/Pz82rkiIrJEDDdERERkU3haioiIiGwKww0RERHZFIYbIiIisikMN0RERGRTGG6IiIjIpjDcEBERkU1huCEiIiKbwnBDRERENuX/A5u4LWpcWVxdAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(1 - e_theta.computeCDF(theta_max))\n",
    "print(1 - e_pos.computeCDF(T / 2))\n",
    "\n",
    "ot.Show(e_ori.drawCDF())\n",
    "ot.Show(e_theta.drawCDF())\n",
    "ot.Show(e_pos.drawCDF())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a1e951ee-dc13-4797-b8c1-a8ed6bb6b64b",
   "metadata": {},
   "source": [
    "## Only defects on one feature of one part (no interaction between parts yet)\n",
    "\n",
    "### Direct method on distributions using openturns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "384076f2",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "lambda_1, lambda_2 = 1, 0\n",
    "# Limit cases\n",
    "j_lim1 = X2 - e_pos - X1  # Cas limite 1, seulement erreur en position\n",
    "j_lim2 = X2 - e_ori - X1  # Cas limite 2, seulement erreur en orientation\n",
    "j_lim3 = (X2 - (np.sqrt(0.5) * e_pos + np.sqrt(0.5) * e_ori) - X1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0a33b9d5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1090x740 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "graph = j_lim1.drawCDF(x_min, x_max)\n",
    "graph = otaf.plotting.set_graph_legends(\n",
    "    graph, x_title=\"j\", y_title=\"P\", title=\"Cas limites de la distribution du jeu\"\n",
    ")\n",
    "graph.add(j_lim2.drawCDF(x_min, x_max))\n",
    "graph.add(j_lim3.drawCDF(x_min, x_max))\n",
    "graph = otaf.plotting.set_graph_legends(\n",
    "    graph,\n",
    "    legends=[\"POS 1 ORI 0\", \"POS 0 ORI 1\", \"POS 0.5 ORI 0.5\"],\n",
    "    colors=[\"red\", \"blue\", \"orange\"],\n",
    ")\n",
    "view = ot.viewer.View(graph, pixelsize=(1100, 750))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf04f351",
   "metadata": {},
   "source": [
    "## Sampling based estimation of the distribution with only defects on one feature, and an imprecise defect allocation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8ee91a84",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "size_lambda1 = 100\n",
    "size_MC1 = int(1e3)\n",
    "\n",
    "lambda_1_smp_base = np.array(list(range(size_lambda1 + 1))) / size_lambda1\n",
    "lambda_1_smp = ot.Sample(np.array([np.sqrt(lambda_1_smp_base), np.sqrt(1-lambda_1_smp_base)]).T) # sqrt cause multiplication of distribution with constant"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "efdebf12",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "ot.RandomGenerator.SetSeed(888)\n",
    "\n",
    "RandDeviationVect = otaf.distribution.get_composed_normal_defect_distribution(\n",
    "    defect_names=[\"gamma_d_1\", \"u_d_1\"],\n",
    "    sigma_dict={\"gamma_\": sigma_theta_max, \"u_\": sigma_e_pos_max},\n",
    ")\n",
    "\n",
    "sample_base_1Feature = RandDeviationVect.getSample(size_MC1)\n",
    "sample_composed_1Feature = otaf.sampling.compose_defects_with_lambdas(lambda_1_smp, sample_base_1Feature)\n",
    "sample_gap_1Feature = [analytical_assembly_model_1_5_D(_X) for _X in sample_composed_1Feature]\n",
    "distributions_1Feature = list(map(ot.UserDefined, sample_gap_1Feature))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "97805251",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1090x740 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors1 = [[\"grey\"]] * lambda_1_smp.getSize()\n",
    "legends1 = [\"\"] * lambda_1_smp.getSize()\n",
    "\n",
    "sup_data1, inf_data1 = otaf.distribution.compute_sup_inf_distributions(distributions_1Feature, x_min, x_max)\n",
    "\n",
    "graph_full_1 = otaf.plotting.plot_combined_CDF(distributions_1Feature, x_min, x_max, colors1, legends1)\n",
    "\n",
    "graph_full_1 = otaf.plotting.set_graph_legends(\n",
    "    graph_full_1,\n",
    "    x_title=\"j\",\n",
    "    y_title=\"P\",\n",
    "    title=\"Position and orientation for one feature without correlation\",\n",
    "    legends=[\"\"] * len(distributions_1Feature),\n",
    ")\n",
    "\n",
    "graph_full_1.add(ot.Curve(inf_data1, \"blue\", \"solid\", 1.5, \"lower envelope\"))\n",
    "graph_full_1.add(ot.Curve(sup_data1, \"red\", \"solid\", 1.5, \"upper envelope\"))\n",
    "\n",
    "view = ot.viewer.View(graph_full_1, pixelsize=(1100, 750))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7a72f8f7-67a5-4433-a4f2-2c5dcc42c04e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0 0.0\n"
     ]
    }
   ],
   "source": [
    "sup_data1\n",
    "inf_data1\n",
    "get_prob_0 = lambda X: X[np.squeeze(np.argwhere(np.abs(X[:, 0]) < 1e-4))]\n",
    "print(get_prob_0(sup_data1)[:, 1].max(), get_prob_0(inf_data1)[:, 1].max())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c5d7c88",
   "metadata": {},
   "source": [
    "## Defects defined for both parts/features. All defects still independent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "5f0cd3ec",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Model the imprecise space of defect allocation with 2 features and 4 DOFs\n",
    "# i corresponds to the first feature, j corresponds to the second feature\n",
    "size_lambda = 7\n",
    "idar = np.arange(size_lambda + 1)\n",
    "lambda_2_ar = np.vstack([idar / size_lambda, 1 - idar / size_lambda]).T\n",
    "\n",
    "# Generate all possible allocation combinations and compute their square roots\n",
    "ij_list = np.sqrt(np.array([[np.append(lambda_2_ar[i], lambda_2_ar[j]) for i in idar] for j in idar]).reshape(-1, 4))\n",
    "\n",
    "# Create a sample using OpenTURNS\n",
    "lambda_2_smp = ot.Sample(ij_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c17de2b0",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Monte Carlo sample size and random generator seed\n",
    "size_MC2 = int(1e6)\n",
    "ot.RandomGenerator.SetSeed(888)\n",
    "\n",
    "# Get the composed normal defect distribution\n",
    "RandDeviationVect = otaf.distribution.get_composed_normal_defect_distribution(\n",
    "    defect_names=[\"gamma_d_1\", \"u_d_1\", \"gamma_d_2\", \"u_d_2\"],\n",
    "    sigma_dict={\"gamma_\": sigma_theta_max, \"u_\": sigma_e_pos_max}\n",
    ")\n",
    "\n",
    "# Sample from the random deviation vector\n",
    "rdvx2 = RandDeviationVect.getSample(size_MC2)\n",
    "\n",
    "# Compose the defect samples with the lambda samples\n",
    "rdvxld_l2 = otaf.sampling.compose_defects_with_lambdas(lambda_2_smp, rdvx2)\n",
    "\n",
    "# Compute the gap list and corresponding distributions\n",
    "gapLst2 = [analytical_assembly_model_1_5_D(rdvxld) for rdvxld in rdvxld_l2]\n",
    "distributions2 = [ot.UserDefined(gap) for gap in gapLst2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "88c8d9c6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Helper functions for lambda position and formatting strings\n",
    "get_pos_lambda = lambda i, ar: [ar[i, 1], ar[i, 3]]  # Extract specific lambda positions\n",
    "get_lambda_str = lambda l: f\"Allocation position piece 1: {l[0] * 100:.1f}%, piece 2: {l[1] * 100:.1f}%\"\n",
    "\n",
    "# Define interesting pairs of lambda values and their colors\n",
    "pos_pairs = [[0.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.2, 0.8], [0.8, 0.2], [1.0, 1.0]]\n",
    "pos_pair_cols = [[\"#FF00CD\"], [\"green\"], [\"orange\"], [\"#00FFF3\"], [\"yellow\"], [\"black\"]]\n",
    "\n",
    "# Find the indices corresponding to the interesting lambda pairs\n",
    "pos_pair_idx = [i for i in range(len(ij_list)) if get_pos_lambda(i, ij_list) in pos_pairs]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "cf6a5174",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1090x740 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Initialize colors and legends for each lambda sample\n",
    "colors2 = [[\"grey\"]] * lambda_2_smp.getSize()\n",
    "legends2 = [\"\"] * lambda_2_smp.getSize()\n",
    "\n",
    "# Set colors and legends for interesting lambda pairs\n",
    "for i, k in enumerate(pos_pair_idx):\n",
    "    colors2[k] = pos_pair_cols[i]\n",
    "    legends2[k] = get_lambda_str(pos_pairs[i])\n",
    "\n",
    "# Compute the upper and lower envelopes of the distributions\n",
    "sup_data2, inf_data2 = otaf.distribution.compute_sup_inf_distributions(distributions2, x_min, x_max)\n",
    "\n",
    "# Plot the combined CDF with additional curves for the envelopes\n",
    "graph_full_2 = otaf.plotting.plot_combined_CDF(distributions2, x_min, x_max, colors2, legends2)\n",
    "graph_full_2 = otaf.plotting.set_graph_legends(\n",
    "    graph_full_2,\n",
    "    x_title=\"j\",\n",
    "    y_title=\"P\",\n",
    "    title=\"Orientation and position defects for two parts without correlation.\",\n",
    "    legends=legends2\n",
    ")\n",
    "\n",
    "# Add the upper and lower envelopes to the graph\n",
    "graph_full_2.add(ot.Curve(inf_data2, \"blue\", \"solid\", 1.5, \"lower envelope\"))\n",
    "graph_full_2.add(ot.Curve(sup_data2, \"red\", \"solid\", 1.5, \"upper envelope\"))\n",
    "\n",
    "# Display the final graph\n",
    "view = ot.viewer.View(graph_full_2, pixelsize=(1100, 750))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4c7ea17d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Probability-Box of the Gap G with Imprecise Allocation and Independence')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 960x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %matplotlib qt\n",
    "fig3 = plt.figure(dpi=150)\n",
    "\n",
    "ax3 = fig3.add_subplot(1, 1, 1)\n",
    "ax3.plot(sup_data2[:, 0], sup_data2[:, 1], color=\"tab:orange\", label=\"upper envelope\")\n",
    "ax3.plot(inf_data2[:, 0], inf_data2[:, 1], color=\"tab:blue\", label=\"lower envelope\")\n",
    "ax3.grid(True)\n",
    "ax3.fill_between(\n",
    "    inf_data2[:, 0], inf_data2[:, 1], sup_data2[:, 1], color=\"gray\", alpha=0.3\n",
    ")\n",
    "ax3.set_xlabel(\"G\")\n",
    "ax3.set_ylabel(\"P\")\n",
    "ax3.legend()\n",
    "ax3.set_title(\"Probability-Box of the Gap G with Imprecise Allocation and Independence\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3c04f949",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Lower probability of failure: 0 %\n",
      "Upper probability of failure: 0 %\n"
     ]
    }
   ],
   "source": [
    "print(\"Lower probability of failure:\", round(otaf.distribution.get_prob_below_threshold(inf_data2)), \"%\")\n",
    "print(\"Upper probability of failure:\", round(otaf.distribution.get_prob_below_threshold(sup_data2)), \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "a991db7a-6650-4dd2-92d5-f98d80d96b58",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Minimum failure probability: 6.00000e-06, Maximum failure probability: 2.90000e-05\n"
     ]
    }
   ],
   "source": [
    "print(f\"Minimum failure probability: {otaf.distribution.get_prob_below_threshold(inf_data2):.5e}, Maximum failure probability: {otaf.distribution.get_prob_below_threshold(sup_data2):.5e}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9dd1e86",
   "metadata": {},
   "source": [
    "## Adding Correlations Between defects within features\n",
    "\n",
    "Adding correlations between random variables introduces additional dimensions of imprecision. For `d` imperfections modeled and influenced by `λ` (lambda) coefficients, we can define `d-1` correlations if we wish to model them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9f432853",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Monte Carlo sample size and lambda/correlation settings\n",
    "size_monte_carlo = int(1e5)\n",
    "\n",
    "# Lambda sampling: creating a base array for the lambda space\n",
    "size_lambda = 5\n",
    "lambda_array = np.linspace(0, 1, size_lambda + 1)  # Simplified sampling of lambda space\n",
    "\n",
    "# Correlation sampling: evenly spaced between -1 and 1\n",
    "size_correlation = 5\n",
    "correlation = np.linspace(-1 + 1e-7, 1 - 1e-7, size_correlation)  # Avoiding perfect correlation values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "7e4f85f9",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Create a 4D grid of lambda and correlation combinations\n",
    "lambda_indices = list(range(size_lambda + 1))\n",
    "correlation_indices = list(range(size_correlation))\n",
    "\n",
    "# Generate all combinations of lambda and correlation pairs\n",
    "ijkz_list = np.array([\n",
    "    [\n",
    "        [\n",
    "            [\n",
    "                np.append([lambda_array[i], lambda_array[j]], [correlation[k], correlation[z]])\n",
    "                for i in lambda_indices\n",
    "            ]\n",
    "            for j in lambda_indices\n",
    "        ]\n",
    "        for k in correlation_indices\n",
    "    ]\n",
    "    for z in correlation_indices\n",
    "])\n",
    "\n",
    "# Reshape the resulting array to create a combined lambda-correlation sample\n",
    "ijkz_list = ijkz_list.reshape(-1, 4)  # Reshaping into a 2D array with 4 columns (λ₁, λ₂, correlation₁, correlation₂)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "3c174b82",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_gap_with_lambdas_and_correlation(\n",
    "    lmbd1, lmbd2, cor1, cor2, \n",
    "    lmb_arr=lambda_array, correlation=correlation, \n",
    "    SEED=888, N=size_monte_carlo\n",
    "):\n",
    "    \"\"\"\n",
    "    Compute the gap between two tilde values, influenced by lambda coefficients and correlations.\n",
    "    \n",
    "    Parameters:\n",
    "    -----------\n",
    "    lmbd1 : float\n",
    "        Lambda value for the first sample.\n",
    "    lmbd2 : float\n",
    "        Lambda value for the second sample.\n",
    "    cor1 : float\n",
    "        Correlation coefficient for the first sample.\n",
    "    cor2 : float\n",
    "        Correlation coefficient for the second sample.\n",
    "    lmb_arr : np.array\n",
    "        Array of lambda values (default: lambda_array).\n",
    "    correlation : np.array\n",
    "        Array of correlation values (default: correlation).\n",
    "    SEED : int\n",
    "        Random seed for reproducibility.\n",
    "    N : int\n",
    "        Monte Carlo sample size.\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    ot.Sample\n",
    "        A sample of the computed gaps between X1_tilde and X2_tilde.\n",
    "    \"\"\"\n",
    "    \n",
    "    # Set random seed for reproducibility\n",
    "    ot.RandomGenerator.SetSeed(SEED)\n",
    "    \n",
    "    # Generate correlated samples for both sets of imperfections\n",
    "    smp1 = otaf.distribution.generate_correlated_samples(\n",
    "        sigma1=sigma_e_pos_max, sigma2=sigma_theta_max, N=N, corr=cor1\n",
    "    )\n",
    "    smp2 = otaf.distribution.generate_correlated_samples(\n",
    "        sigma1=sigma_e_pos_max, sigma2=sigma_theta_max, N=N, corr=cor2\n",
    "    )\n",
    "    \n",
    "    # Extract positional and angular deviations\n",
    "    e_pos_X1, e_theta1 = np.squeeze(smp1[:, 0]), np.squeeze(smp1[:, 1])\n",
    "    e_pos_X2, e_theta2 = np.squeeze(smp2[:, 0]), np.squeeze(smp2[:, 1])\n",
    "    \n",
    "    # Compute X1_tilde and X2_tilde based on the lambda coefficients and deviations\n",
    "    X1_tilde = np.array([\n",
    "        X1 + np.sqrt(lmbd1) * e_pos_X1 - np.sqrt(1 - lmbd1) * (X3 / 2) * e_theta1,\n",
    "        X1 + np.sqrt(lmbd1) * e_pos_X1 + np.sqrt(1 - lmbd1) * (X3 / 2) * e_theta1,\n",
    "    ])\n",
    "    \n",
    "    X2_tilde = np.array([\n",
    "        X2 - np.sqrt(lmbd2) * e_pos_X2 - np.sqrt(1 - lmbd2) * (X3 / 2) * e_theta2,\n",
    "        X2 - np.sqrt(lmbd2) * e_pos_X2 + np.sqrt(1 - lmbd2) * (X3 / 2) * e_theta2,\n",
    "    ])\n",
    "    \n",
    "    # Compute the gap (jeu) between X2_tilde and X1_tilde\n",
    "    jeu = X2_tilde - X1_tilde\n",
    "    jeu = np.expand_dims(np.squeeze(jeu.min(axis=0)), axis=1)\n",
    "    \n",
    "    # Return the result as an OpenTURNS sample\n",
    "    return ot.Sample(jeu)\n",
    "\n",
    "\n",
    "# Create a partial function with fixed parameters for lambda_array, correlation, and Monte Carlo size\n",
    "compute_gap_func = partial(\n",
    "    compute_gap_with_lambdas_and_correlation, \n",
    "    lmb_arr=lambda_array, correlation=correlation, \n",
    "    SEED=168406047, N=size_monte_carlo\n",
    ")\n",
    "\n",
    "# Apply the function to each combination in the lambda-correlation grid\n",
    "ijkz_results = Parallel(n_jobs=-1)(delayed(compute_gap_func)(i, j, k, z) for i, j, k, z in ijkz_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dd17187a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create distributions from the results of the gap function\n",
    "#distributions3 = list(map(ot.UserDefined, ijkz_results))\n",
    "distributions3 = Parallel(n_jobs=-1)(delayed(ot.UserDefined)(result) for result in ijkz_results)\n",
    "\n",
    "# Initialize color and legend settings for each result\n",
    "colors3 = [[\"grey\"] for _ in ijkz_results]\n",
    "legends3 = [\"\" for _ in ijkz_results]\n",
    "\n",
    "# Compute the supremum and infimum data from the distributions\n",
    "sup_data3, inf_data3 = otaf.distribution.compute_sup_inf_distributions(distributions3, x_min, x_max)\n",
    "\n",
    "# Create a combined CDF graph for the distributions    odel.\n",
    "\n",
    "Exploring the imprecise probability space.¶\n",
    "graph_full_3 = otaf.plotting.plot_combined_CDF(distributions3, x_min, x_max, colors3, legends3)\n",
    "\n",
    "# Set the title and legends for the graph\n",
    "title = \"Probability-Box of the Gap G with Imprecise Allocation and Correlation\"\n",
    "graph_full_3 = otaf.plotting.set_graph_legends(\n",
    "    graph_full_3, \n",
    "    x_title=\"G\", \n",
    "    y_title=\"P\", \n",
    "    title=title, \n",
    "    legends=[\"\"] * len(ijkz_results)\n",
    ")\n",
    "\n",
    "# Add the lower and upper envelopes (infimum and supremum) to the graph\n",
    "graph_full_3.add(ot.Curve(inf_data3, \"blue\", \"solid\", 1.5, \"lower envelope\"))\n",
    "graph_full_3.add(ot.Curve(sup_data3, \"red\", \"solid\", 1.5, \"upper envelope\"))\n",
    "\n",
    "# Display the graph\n",
    "view = ot.viewer.View(graph_full_3, pixelsize=(1100, 750))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5331f5eb-d879-4df1-9304-264618de6eb9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# %matplotlib qt\n",
    "fig31 = plt.figure(dpi=150)\n",
    "\n",
    "ax31 = fig31.add_subplot(1, 1, 1)\n",
    "ax31.plot(sup_data3[:, 0], sup_data3[:, 1], color=\"tab:orange\", label=\"upper envelope\")\n",
    "ax31.plot(inf_data3[:, 0], inf_data3[:, 1], color=\"tab:blue\", label=\"lower envelope\")\n",
    "ax31.grid(True)\n",
    "ax31.fill_between(\n",
    "    inf_data3[:, 0], inf_data3[:, 1], sup_data3[:, 1], color=\"gray\", alpha=0.3\n",
    ")\n",
    "ax31.set_xlabel(\"G\")\n",
    "ax31.set_ylabel(\"P\")\n",
    "ax31.legend()\n",
    "ax31.set_title(\"Probability-Box of the Gap G with Imprecise Allocation and Correlation\")\n",
    "# otaf.plotting.save_plot(filename=\"PBoxGapGImpreciseAllocationCorrelation\", ax=ax31, dpi=600)\n",
    "plt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0100aacf-7d17-47d9-8f57-1d4d90d49e70",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"Lower probability of failure:\", round(otaf.distribution.get_prob_below_threshold(inf_data3) * 100, 9), \"%\")\n",
    "print(\"Upper probability of failure:\", round(otaf.distribution.get_prob_below_threshold(sup_data3) * 100, 9), \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7977c01-d934-4b76-903a-eed81ffc99a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the new figure\n",
    "fig_overlap = plt.figure(dpi=600)\n",
    "ax_overlap = fig_overlap.add_subplot(1, 1, 1)\n",
    "\n",
    "# Plot the upper and lower envelopes for both correlation and independence\n",
    "ax_overlap.plot(sup_data2[:, 0], sup_data2[:, 1], color=\"tab:green\", label=\"upper envelope (Independence)\")\n",
    "ax_overlap.plot(inf_data2[:, 0], inf_data2[:, 1], color=\"tab:green\", linestyle='--', label=\"lower envelope (Independence)\")\n",
    "\n",
    "ax_overlap.plot(sup_data3[:, 0], sup_data3[:, 1], color=\"tab:red\", label=\"upper envelope (Correlation)\")\n",
    "ax_overlap.plot(inf_data3[:, 0], inf_data3[:, 1], color=\"tab:red\", linestyle='--', label=\"lower envelope (Correlation)\")\n",
    "\n",
    "# Grid\n",
    "ax_overlap.grid(True)\n",
    "\n",
    "# Define arrays for gap values and bounds\n",
    "g_vals = sup_data2[:, 0]\n",
    "li = inf_data2[:, 1]  # Lower bound for Independence\n",
    "lc = inf_data3[:, 1]  # Lower bound for Correlation\n",
    "ui = sup_data2[:, 1]  # Upper bound for Independence\n",
    "uc = sup_data3[:, 1]  # Upper bound for Correlation\n",
    "\n",
    "# Masks for the different regions\n",
    "# Green: Independence dominates (li < lc) or (ui > uc)\n",
    "mask_green_lower = li < lc\n",
    "mask_green_upper = ui > uc\n",
    "\n",
    "# Red: Correlation dominates (lc < li) or (uc > ui)\n",
    "mask_red_lower = lc < li\n",
    "mask_red_upper = uc > ui\n",
    "\n",
    "# Gray: Overlapping regions (li < uc) and (lc < ui)\n",
    "mask_gray = (li < uc) & (lc < ui)\n",
    "\n",
    "# Fill regions based on the masks\n",
    "# Green regions\n",
    "ax_overlap.fill_between(g_vals, li, lc, where=mask_green_lower, color=\"green\", alpha=0.3)\n",
    "ax_overlap.fill_between(g_vals, uc, ui, where=mask_green_upper, color=\"green\", alpha=0.3)\n",
    "\n",
    "# Red regions\n",
    "ax_overlap.fill_between(g_vals, lc, li, where=mask_red_lower, color=\"red\", alpha=0.3)\n",
    "ax_overlap.fill_between(g_vals, ui, uc, where=mask_red_upper, color=\"red\", alpha=0.3)\n",
    "\n",
    "# Gray regions (overlap)\n",
    "ax_overlap.fill_between(g_vals, np.maximum(li, lc), np.minimum(ui, uc), where=mask_gray, color=\"gray\", alpha=0.3)\n",
    "\n",
    "# Set labels and title\n",
    "ax_overlap.set_xlabel(\"G\")\n",
    "ax_overlap.set_ylabel(\"P\")\n",
    "ax_overlap.set_title(\"Overlapping Probability-Box of the Gap G with Imprecise Allocation\")\n",
    "\n",
    "# Add legend\n",
    "ax_overlap.legend()\n",
    "\n",
    "# Show the plot\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6137e8e",
   "metadata": {},
   "source": [
    "## Modify the way we create the experimental design to explore the imprecise space (lambda and correlation)\n",
    "\n",
    "### We will use Latin Hypercube Sampling (LHS) for this purpose.\n",
    "\n",
    "#### First, we define distributions to represent the uncertainties."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7e1f5ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define uniform distributions for lambda and correlation values\n",
    "lambda_12_dist = ot.Uniform(0, 1)\n",
    "lambda_34_dist = ot.Uniform(0, 1)\n",
    "correlation_dist12 = ot.Uniform(-1, 1)\n",
    "correlation_dist34 = ot.Uniform(-1, 1)\n",
    "\n",
    "# Combine these into a composed distribution\n",
    "composed_dist = ot.ComposedDistribution(\n",
    "    [lambda_12_dist, lambda_34_dist, correlation_dist12, correlation_dist34]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92fad9ed",
   "metadata": {},
   "source": [
    "#### We generate the LHS in the normal space and use the isoprobabilistic transformation to map it back to the original space."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3d24e415",
   "metadata": {},
   "outputs": [],
   "source": [
    "ot.RandomGenerator.SetSeed(123456789)\n",
    "SIZE_LHS = 100\n",
    "lhsExp = ot.LHSExperiment(composed_dist, SIZE_LHS, False, True)\n",
    "lhs_sample = lhsExp.generate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6e49174f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Modified mini_func4 without storing in shelve\n",
    "def compute_gap_with_lambdacor(id_, lambdacor, N=size_monte_carlo):\n",
    "    lb1, lb3, C1, C2 = lambdacor\n",
    "\n",
    "    # Set the random seed for reproducibility\n",
    "    ot.RandomGenerator.SetSeed(999)\n",
    "\n",
    "    # Generate correlated samples for the first and second sets of imperfections\n",
    "    smp1 = otaf.distribution.generate_correlated_samples(\n",
    "        sigma1=sigma_e_pos_max, sigma2=sigma_theta_max, corr=C1, N=N\n",
    "    )\n",
    "    smp2 = otaf.distribution.generate_correlated_samples(\n",
    "        sigma1=sigma_e_pos_max, sigma2=sigma_theta_max, corr=C2, N=N\n",
    "    )\n",
    "\n",
    "    # Extract positional and angular deviations\n",
    "    e_pos_X1, e_theta1 = np.squeeze(smp1[:, 0]), np.squeeze(smp1[:, 1])\n",
    "    e_pos_X2, e_theta2 = np.squeeze(smp2[:, 0]), np.squeeze(smp2[:, 1])\n",
    "\n",
    "    # Compute X1_tilde and X2_tilde based on the lambda coefficients and deviations\n",
    "    X1_tilde = np.array([\n",
    "        X1 + np.sqrt(lb1) * e_pos_X1 - np.sqrt(1 - lb1) * (X3 / 2) * e_theta1,\n",
    "        X1 + np.sqrt(lb1) * e_pos_X1 + np.sqrt(1 - lb1) * (X3 / 2) * e_theta1,\n",
    "    ])\n",
    "    \n",
    "    X2_tilde = np.array([\n",
    "        X2 - np.sqrt(lb3) * e_pos_X2 - np.sqrt(1 - lb3) * (X3 / 2) * e_theta2,\n",
    "        X2 - np.sqrt(lb3) * e_pos_X2 + np.sqrt(1 - lb3) * (X3 / 2) * e_theta2,\n",
    "    ])\n",
    "\n",
    "    # Calculate the gap (jeu) between X2_tilde and X1_tilde\n",
    "    jeu = X2_tilde - X1_tilde\n",
    "    jeu = np.expand_dims(np.squeeze(jeu.min(axis=0)), axis=1)\n",
    "\n",
    "    return ot.Sample(jeu)\n",
    "\n",
    "# Partial function with fixed parameters\n",
    "compute_gap_part4 = partial(compute_gap_with_lambdacor, N=size_monte_carlo)\n",
    "\n",
    "# Generate the samples using compute_gap_part4\n",
    "samples4 = [compute_gap_part4(id_, lambdacor) for id_, lambdacor in enumerate(lhs_sample)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e778ff39",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create UserDefined distributions based on the samples\n",
    "distributions4 = list(map(ot.UserDefined, samples4))\n",
    "\n",
    "# Generate colors and legends for the distributions\n",
    "colors4 = [[\"grey\"] for _ in samples4]\n",
    "legends4 = [\"\" for _ in samples4]\n",
    "\n",
    "# Compute the supremum and infimum data for the distributions\n",
    "sup_data4, inf_data4 = otaf.distribution.compute_sup_inf_distributions(distributions4, x_min, x_max)\n",
    "\n",
    "# Create a combined CDF graph for the distributions\n",
    "graph_full_4 = otaf.plotting.plot_combined_CDF(distributions4, x_min, x_max, colors4, legends4)\n",
    "\n",
    "# Set the graph titles and legends\n",
    "graph_full_4 = otaf.plotting.set_graph_legends(\n",
    "    graph_full_4,\n",
    "    x_title=\"j\",\n",
    "    y_title=\"P\",\n",
    "    title=\"Orientation and position defects for two parts. With correlation. Imprecision driven by LHS\",\n",
    "    legends=[\"\"] * len(samples4),\n",
    ")\n",
    "\n",
    "# Add the lower and upper envelopes to the graph\n",
    "graph_full_4.add(ot.Curve(inf_data4, \"blue\", \"solid\", 1.5, \"lower envelope\"))\n",
    "graph_full_4.add(ot.Curve(sup_data4, \"red\", \"solid\", 1.5, \"upper envelope\"))\n",
    "\n",
    "# Display the graph\n",
    "view = ot.viewer.View(graph_full_4, pixelsize=(1100, 750))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c660ff9b",
   "metadata": {},
   "outputs": [],
   "source": [
    "graph_all = ot.Graph(\n",
    "    \"\"\"Superposition of the different cases\"\"\", \"j\", \"P\", True, \"\", 1.0, 0\n",
    ")\n",
    "#graph_all.add(ot.Curve(inf_data1, \"orange\", \"solid\", 1.5, \"lower envelope 1\"))\n",
    "#graph_all.add(ot.Curve(sup_data1, \"orange\", \"solid\", 1.5, \"upper envelope 1\"))\n",
    "graph_all.add(ot.Curve(inf_data2, \"green\", \"solid\", 1.5, \"lower envelope 2\"))\n",
    "graph_all.add(ot.Curve(sup_data2, \"green\", \"solid\", 1.5, \"upper envelope 2\"))\n",
    "graph_all.add(ot.Curve(inf_data3, \"red\", \"solid\", 1.5, \"lower envelope 3\"))\n",
    "graph_all.add(ot.Curve(sup_data3, \"red\", \"solid\", 1.5, \"upper envelope 3\"))\n",
    "#graph_all.add(ot.Curve(inf_data4, \"blue\", \"solid\", 1.5, \"lower envelope 4\"))\n",
    "#graph_all.add(ot.Curve(sup_data4, \"blue\", \"solid\", 1.5, \"upper envelope 4\"))\n",
    "graph_all.add(ot.Curve([0, 0], [0, 0.3], 'black'))\n",
    "graph_all.setLegends(\n",
    "    [\n",
    "        \"lower envelope 1\",\n",
    "        \"upper envelope 1\",\n",
    "        \"lower envelope 2\",\n",
    "        \"upper envelope 2\",\n",
    "        \"lower envelope 3\",\n",
    "        \"upper envelope 3\",\n",
    "        \"lower envelope 4\",\n",
    "        \"upper envelope 4\",\n",
    "    ]\n",
    ")\n",
    "view = ot.viewer.View(graph_all, pixelsize=(1100, 750))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f127b63",
   "metadata": {},
   "source": [
    "### Let's plot some deviation domains with different types of allocations and correlations. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "41351a9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "SIZE_LHS = 9\n",
    "composed_dist_uiq = ot.ComposedDistribution([lambda_12_dist, correlation_dist12])\n",
    "lhsExp_uiq = ot.LHSExperiment(composed_dist_uiq, SIZE_LHS, False, True)\n",
    "lhs_arr = np.array(lhsExp_uiq.generate())\n",
    "lmbds, corr = lhs_arr[:, 0], lhs_arr[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "13157f00-223c-40fa-bb4a-cdc206b1109b",
   "metadata": {},
   "outputs": [],
   "source": [
    "seed = 3641641648\n",
    "fig, ax = plt.subplots(ncols=3, nrows=3)\n",
    "idx = 0\n",
    "for i in range(3):\n",
    "    for j in range(3):\n",
    "        ot.RandomGenerator.SetSeed(seed)\n",
    "        sample =  otaf.distribution.generate_correlated_samples(sigma1=sigma_e_pos_max, sigma2=sigma_theta_max, corr=corr[idx], N=3355)\n",
    "        \n",
    "        # Compute and print variance for diagnostic purposes\n",
    "        var_1 = np.var(sample[:, 0])\n",
    "        var_2 = np.var(sample[:, 1])\n",
    "        print(f\"Sample {idx}: Variance in X1 = {var_1}, Variance in X2 = {var_2}, Correlation = {corr[idx]}\")\n",
    "        \n",
    "        otaf.plotting.print_sample_in_deviation_domain(\n",
    "            ax[i, j], sample[:, 0], sample[:, 1], \n",
    "            np.sqrt(lmbds[idx]), np.sqrt(1 - lmbds[idx]), \n",
    "            t_, theta_max, ratio=1, remove_ticks=False\n",
    "        )\n",
    "        idx += 1\n",
    "\n",
    "fig.set_dpi(100)\n",
    "print(\"sigma_e_pos_max**2\", sigma_e_pos_max**2)\n",
    "print(\"sigma_theta_max**2\", sigma_theta_max**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3bba9300-579b-4d87-bbb0-4b2e4ba08896",
   "metadata": {},
   "outputs": [],
   "source": [
    "sigma_e_pos_max**2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "75caa2fb-8ad0-4218-84d3-9e2a594c5307",
   "metadata": {},
   "outputs": [],
   "source": [
    "seed = 3641641648\n",
    "fig, ax = plt.subplots()  # Only one plot for testing\n",
    "\n",
    "# Set the first element of the sequences for testing\n",
    "idx = 0  # First element of sequences\n",
    "ot.RandomGenerator.SetSeed(seed)\n",
    "\n",
    "# Generate one sample with the first values of sigma and correlation\n",
    "sample = otaf.distribution.generate_correlated_samples(\n",
    "    sigma1=sigma_e_pos_max, sigma2=sigma_theta_max, corr=0, N=3355 #\n",
    ")\n",
    "\n",
    "sample_pos = e_pos.getSample(3355)\n",
    "sample_theta = e_theta.getSample(3355)\n",
    "\n",
    "# Compute and print variance for diagnostic purposes\n",
    "var_1 = np.var(sample[:, 0])\n",
    "var_2 = np.var(sample[:, 1])\n",
    "print(f\"Sample {idx}: Variance in X1 = {var_1}, Variance in X2 = {var_2}, Correlation = {corr[idx]}\")\n",
    "\n",
    "# Plot the sample in the deviation domain using the first parameters\n",
    "otaf.plotting.print_sample_in_deviation_domain(\n",
    "    ax, sample_pos, sample_theta, \n",
    "    np.sqrt(lmbds[idx]), np.sqrt(1 - lmbds[idx]), \n",
    "    t_, theta_max, ratio=1, remove_ticks=False\n",
    ")\n",
    "\n",
    "# Adjust figure DPI for better resolution\n",
    "fig.set_dpi(100)\n",
    "\n",
    "# Print the squared sigma values for diagnostic purposes\n",
    "print(\"sigma_e_pos_max**2\", sigma_e_pos_max**2)\n",
    "print(\"sigma_theta_max**2\", sigma_theta_max**2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddadf9c3",
   "metadata": {},
   "source": [
    "### Let's plot the same without correlation at all "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cca4f10b",
   "metadata": {},
   "outputs": [],
   "source": [
    "SEED=265\n",
    "N8PTS = 3350 #10000#\n",
    "ot.RandomGenerator.SetSeed(SEED)\n",
    "sample_pos = e_pos.getSample(N8PTS)\n",
    "sample_theta = e_theta.getSample(N8PTS)\n",
    "lmbds2 = [0.5, 0.98, 0.02]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "767d981f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ensure 'osifont' is correctly recognized\n",
    "custom_font = FontProperties(family='osifont')\n",
    "\n",
    "def plot_rect_part(ax, scaleFactor=100, rect_length=100, rect_height=10, aspect=8, custom_font=custom_font):\n",
    "    \"\"\"\n",
    "    Plots a rectangular part with hatching, maintaining the given aspect ratio and applying\n",
    "    optional custom font and size scaling.\n",
    "    \"\"\"\n",
    "    # Create the main rectangle with hatching\n",
    "    main_rect = Rectangle((0, 0), rect_length, rect_height, linewidth=2, edgecolor='black',\n",
    "                                  facecolor='none', hatch='//')  # Diagonal hatching\n",
    "    ax.add_patch(main_rect)\n",
    "    \n",
    "    # Maintain the aspect ratio\n",
    "    ax.set_aspect(aspect)\n",
    "\n",
    "    # Add the tolerance zone as a green transparent area centered around the nominal value\n",
    "    tolerance_zone_x = np.array([rect_length - scaleFactor * t_, rect_length + scaleFactor * t_])\n",
    "    tolerance_zone_y = np.array([rect_height, rect_height])\n",
    "    ax.fill_between(tolerance_zone_x, tolerance_zone_y[0], y2=0, color='green', alpha=0.3)\n",
    "\n",
    "    # Adding arrows with proper alignment for the tolerance zone\n",
    "    ax.arrow(rect_length, rect_height + 0.5, -t_ * scaleFactor, 0, head_width=0.5, head_length=1.0,\n",
    "             overhang=0.3, linewidth=2, color=\"r\", length_includes_head=True)\n",
    "    ax.arrow(rect_length, rect_height + 0.5, t_ * scaleFactor, 0, head_width=0.5, head_length=1.0,\n",
    "             overhang=0.3, linewidth=2, color=\"r\", length_includes_head=True)\n",
    "    ax.text(rect_length - 5, rect_height + 0.75, f\"t*{scaleFactor}\", color=\"r\",\n",
    "            fontsize=\"x-large\", fontweight=\"bold\", fontproperties=custom_font)\n",
    "\n",
    "    # Add label 'A' centered and aligned on the left\n",
    "    ax.text(-11, rect_height * 0.5, \"A\", ha=\"center\", va=\"center\",\n",
    "            fontsize=\"xx-large\", fontweight=\"bold\", fontproperties=custom_font,\n",
    "            bbox=dict(facecolor=\"none\", edgecolor=\"black\", boxstyle=\"circle\", linewidth=1.5))\n",
    "\n",
    "    # Arrow pointing to the rectangle surface\n",
    "    ax.arrow(-7, rect_height * 0.5, 6.5, 0, head_width=0.45, head_length=1.5,\n",
    "             overhang=0, linewidth=1.5, color=\"k\", length_includes_head=True)\n",
    "\n",
    "    # Removing the grid, ticks, and bounding box\n",
    "    ax.set_xticks([])#[0, rect_length])\n",
    "    ax.set_yticks([])#[0, rect_height])\n",
    "    ax.tick_params(left=False, bottom=False, labelsize='large')\n",
    "    ax.spines['top'].set_visible(False)\n",
    "    ax.spines['right'].set_visible(False)\n",
    "    ax.spines['left'].set_visible(False)\n",
    "    ax.spines['bottom'].set_visible(False)\n",
    "    ax.grid(False)\n",
    "\n",
    "    return ax\n",
    "\n",
    "def plot_deviated_surfs(ax, sample_pos, sample_rot, lambda_, scaleFactor=100, rect_length=100, rect_height=10,):\n",
    "    \"\"\"adds on the rectangular part a ensemble of deviated lines, to\n",
    "    see the geometrical distribution of the defects and the impact\n",
    "    of the allocations.\n",
    "    The defects are multiplied by a scaleFactor\n",
    "    \"\"\"\n",
    "    sample_pos = np.squeeze(sample_pos)\n",
    "    sample_rot = np.squeeze(sample_rot)\n",
    "    x_line = np.array([rect_length, rect_length])\n",
    "    y_line = np.array([rect_height, 0.0])  # stays constant\n",
    "\n",
    "    assert len(sample_pos) == len(sample_rot), \"Mismatch in position and rotation lengths.\"\n",
    "    for i in range(len(sample_rot)):\n",
    "        x_ltemp = x_line + (np.sqrt(lambda_) * sample_pos[i]) * scaleFactor\n",
    "        x_ltemp[0] = x_ltemp[0] - (np.sqrt(1 - lambda_) * sample_rot[i] * 0.5 * rect_height) * scaleFactor\n",
    "        x_ltemp[1] = x_ltemp[1] + (np.sqrt(1 - lambda_) * sample_rot[i] * 0.5 * rect_height) * scaleFactor\n",
    "        ax.plot(x_ltemp, y_line, \"b-\", linewidth=1.1)\n",
    "\n",
    "    return ax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fe6b5c7a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# %matplotlib qt\n",
    "ncols, nrows = (3, 2)\n",
    "fig, ax = plt.subplots(ncols=ncols, nrows=nrows)\n",
    "scale_factor = 100\n",
    "for i in range(ncols):\n",
    "    axDD = otaf.plotting.print_sample_in_deviation_domain(\n",
    "        ax[0, i],\n",
    "        np.array(sample_pos),\n",
    "        np.array(sample_theta),\n",
    "        np.sqrt(lmbds2[i]),\n",
    "        np.sqrt(1 - lmbds2[i]),\n",
    "        t_,\n",
    "        theta_max,\n",
    "        ratio=1,\n",
    "        r=None,\n",
    "        remove_ticks=True\n",
    "    )\n",
    "    axRP = plot_rect_part(ax[1, i], scale_factor)\n",
    "    axRP = plot_deviated_surfs(axRP, sample_pos, sample_theta, lmbds2[i], scale_factor)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8d2d2ba-9753-4ea3-987a-db3368059465",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Raw Cell Format",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.20"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}