docs/content/tutorial/tutorial_linfa_2d.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "5d0145ed",
   "metadata": {},
   "source": [
    "# LINFA Tutorial 1 - Two dimensions\n",
    "This tutorial will guide you through using the functionality of LINFA to set up inference tasks involving physics-based models. \n",
    "<br>\n",
    "\n",
    "* **What is LINFA?**\n",
    "<br> \n",
    "    LINFA is a library for variational inference with normalizing flow and adaptive annealing. LINFA accommodates computationally expensive models and difficult-to-sample posterior distributions with dependent parameters.\n",
    "\n",
    "* **Why should I use LINFA?**\n",
    "<br>\n",
    "    Designed as a general inference engine, LINFA allows the user to define custom input transformations, computational models, surrogates, and likelihood functions which will be discussed throughout the tutorial."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "3fee9867",
   "metadata": {},
   "source": [
    "### Tutorial outline\n",
    "In this tutorial we will:\n",
    "\n",
    "1. Define a physics-based model for a simple ballistic application.\n",
    "\n",
    "2. Perform the following inference tasks:\n",
    "\n",
    "    * Variational inference with the original model.\n",
    "\n",
    "    * Variational inference with a neural network surrogate.\n",
    "\n",
    "After going through this tutorial, users should be able to define and integrate their model with LINFA, and use its features to perform inference."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "909c3262",
   "metadata": {},
   "source": [
    "### Additional Resources\n",
    "\n",
    "#### Background theory and examples for LINFA\n",
    "\n",
    "* Y. Wang, E.R. Cobian, J. Lee, F. Liu, J.D. Hauenstein, D.E. Schiavazzi, [LINFA: a Python library for variational inference with normalizing flow and annealing](https://arxiv.org/abs/2307.04675)\n",
    "\n",
    "* Y. Wang, F. Liu and D.E. Schiavazzi, [Variational Inference with NoFAS: Normalizing Flow with Adaptive Surrogate for Computationally Expensive Models](https://www.sciencedirect.com/science/article/abs/pii/S0021999122005162)\n",
    "\n",
    "* E.R. Cobian, J.D. Hauenstein, F. Liu and D.E. Schiavazzi, [AdaAnn: Adaptive Annealing Scheduler for Probability Density Approximation](https://www.dl.begellhouse.com/journals/52034eb04b657aea,796f39cb1acf1296,6f85fe1149ff41d9.html?sgstd=1)\n",
    "\n",
    "#### More about LINFA library: \n",
    "\n",
    "* LINFA library [documentation](https://linfa-vi.readthedocs.io/en/latest/index.html).\n",
    "\n",
    "* LINFA GitHub [repository](https://github.com/desResLab/LINFA)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "18a16a2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Import libraries ##\n",
    "import os\n",
    "from linfa.run_experiment import experiment\n",
    "from linfa.transform import Transformation\n",
    "from linfa.nofas import Surrogate\n",
    "import torch\n",
    "import random\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import cv2"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "daa74064",
   "metadata": {},
   "source": [
    "### Problem definition\n",
    "* Our physics-based model **phys** consists of a simple ballistic model. We would like to compute the quantities: \n",
    "    * $x_{1}$: *final location (m)*\n",
    "    * $x_{2}$: *time of flight (s)*\n",
    "    \n",
    "    from the inputs:\n",
    "    * $z_{1}$: *starting position (m)*\n",
    "    * $z_{2}$: *initial velocity (m/s)*\n",
    "<br>\n",
    "\n",
    "The model is described by the following equations\n",
    "$$\n",
    "x_{1} = z_{1} + \\frac{z_{2}^{2}\\,\\sin(2\\cdot 60)}{g},\\,\\,\n",
    "x_{2} = \\frac{2\\,z_{2}\\,\\sin(60)}{g}.\n",
    "$$\n",
    "\n",
    "This corresponds to the situation where a target location is reached from an object launched at an horizontal angle of 60 degrees. "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "20bbf1a5",
   "metadata": {},
   "source": [
    "#### Model identifiability\n",
    "\n",
    "This model is **identifiable**."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "2f6b1e92",
   "metadata": {},
   "source": [
    "#### Implementation as a Python class\n",
    "\n",
    "* We now create a new **Phys** model class, having three member functions:\n",
    "\n",
    "    * `__init__` - A constructor.\n",
    "\n",
    "    * `genDataFile` - A member function to create synthetic observations.\n",
    "    \n",
    "    * `solve_t` - A function to perform forward model evaluations.\n",
    "\n",
    "*Please refer to the comments below for additional implementation details.* "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3af4586b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#### Implementation of the traditional trajectory motion physics problem ####\n",
    "class Phys:\n",
    "    \n",
    "    ### Define constructor function for Phys class ###\n",
    "    def __init__(self):\n",
    "        ## Define input parameters (True value)  \n",
    "        # input[] = [starting_position, initial_velocity, angle] = [1(m), 5(m/s)]\n",
    "        self.defParam = torch.Tensor([[1.0, 5.0]])\n",
    "        \n",
    "        self.gConst = 9.81   # gravitational constant\n",
    "        self.stdRatio = 0.05 # standard deviation ratio\n",
    "        self.data = None     # data set of model sample\n",
    "\n",
    "    ### Define data file generator function ###\n",
    "    # dataSize (int): size of sample (data)\n",
    "    # dataFileName (String): name of the sample data file\n",
    "    # store (Boolean): True if user wish to store the generated data file; False otherwise.\n",
    "    def genDataFile(self, dataSize = 50, dataFileName=\"data_phys_2d.txt\", store=True):\n",
    "        def_out = self.solve_t(self.defParam)[0]\n",
    "        self.data = def_out + self.stdRatio * torch.abs(def_out) * torch.normal(0, 1, size=(dataSize, 2))\n",
    "        self.data = self.data.t().detach().numpy()\n",
    "        if store: np.savetxt(dataFileName, self.data)\n",
    "        return self.data\n",
    "\n",
    "    ### Define data file generator function ###\n",
    "    # params (Tensor): input parameters storing starting position, initial velocity, and angle in corresponding order.\n",
    "    def solve_t(self, params):\n",
    "        z1, z2 = torch.chunk(params, chunks=2, dim=1) # input parameters\n",
    "        z3 = torch.Tensor([[60.0]])\n",
    "        z3 = z3 * (np.pi / 180)                           # convert unit from degree to radians\n",
    "\n",
    "        ## Output value calculation\n",
    "        # ouput[] = [final_location, total_time]\n",
    "        x = torch.cat(( z1 + ((z2 * z2 * torch.sin(2.0 * z3)) / self.gConst),             # x2: finalLocation \n",
    "            (2.0 * z2 * torch.sin(z3)) / self.gConst), 1)                                 # x3: totalTime\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "677c95a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Generate phys sample file ##\n",
    "\n",
    "# Define model\n",
    "model = Phys()\n",
    "\n",
    "# Generate Data\n",
    "physData = model.genDataFile()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "aaa209c9",
   "metadata": {},
   "source": [
    "Now that we have our model set up, we go on to our second step, i.e., *inference with the full model.*"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "4bed3e1d",
   "metadata": {},
   "source": [
    "### Variational inference with full model\n",
    "\n",
    "#### Definition of hyperparameters\n",
    "The first step is to define all options and hyperparameters for the inference task. \n",
    "\n",
    "Additional detail for each hyperparameter can be found in the [documentation](https://linfa-vi.readthedocs.io/en/latest/content/linfa_options.html) or in the definition of the [experiment](https://github.com/desResLab/LINFA/blob/master/linfa/run_experiment.py) class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f0b0c887",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Experiment Setting\n",
    "exp = experiment()\n",
    "exp.flow_type        = 'maf'        # str: Type of flow\n",
    "exp.n_blocks         = 5            # int: Number of layers                            \n",
    "exp.hidden_size      = 100          # int: Hidden layer size for MADE in each layer  \n",
    "exp.n_hidden         = 1            # int: Number of hidden layers in each MADE      \n",
    "exp.activation_fn    = 'relu'       # str: Actication function used                  \n",
    "exp.input_order      = 'sequential' # str: Input order for create_mask           \n",
    "exp.batch_norm_order = True         # boolean: Order to decide if batch_norm is used    \n",
    "exp.save_interval    = 5000         # int: How often to sample from normalizing flow\n",
    "\n",
    "exp.input_size    = 2               # int: Dimensionality of input                   \n",
    "exp.batch_size    = 200             # int: Number of samples generated             \n",
    "exp.true_data_num = 2               # double: number of true model evaluated      \n",
    "exp.n_iter        = 25001           # int: Number of iterations                      \n",
    "exp.lr            = 0.01            # float: Learning rate                              \n",
    "exp.lr_decay      = 0.9999          # float: Learning rate decay                \n",
    "exp.log_interval  = 100             # int: How often to show loss stat   \n",
    "\n",
    "exp.run_nofas          = False      # boolean: to run experiment with nofas\n",
    "exp.annealing          = False      # boolean: to run experiment with annealing\n",
    "exp.calibrate_interval = 1000       # int: How often to update surrogate model     \n",
    "exp.budget             = 260        # int: Total number of true model evaluation\n",
    "\n",
    "exp.surr_pre_it  = 30000            # int: Number of pre-training iterations for surrogate model\n",
    "exp.surr_upd_it  = 6000             # int: Number of iterations for the surrogate model update\n",
    "exp.surr_folder  = \"./\"\n",
    "exp.use_new_surr = True             # boolean: to run experiment with nofas\n",
    "\n",
    "exp.results_file = 'results.txt'      # str: result text file name\n",
    "exp.log_file     = 'log.txt'          # str: log text file name\n",
    "exp.samples_file = 'samples.txt'      # str: sample text file name\n",
    "exp.seed         = random.randint(0, 10 ** 9)  # int: Random seed used\n",
    "exp.n_sample     = 5000               # int: Total number of iterations\n",
    "exp.no_cuda      = True               # boolean: to run experiment with NO cuda\n",
    "\n",
    "exp.optimizer    = 'RMSprop'          # str: Type of optimizer\n",
    "exp.lr_scheduler = 'ExponentialLR'    # str: Type of scheduler\n",
    "\n",
    "exp.device = torch.device('cuda:0' if torch.cuda.is_available() and not exp.no_cuda else 'cpu')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "60327144",
   "metadata": {},
   "source": [
    "#### Define the transformation \n",
    "Now we define the trasformation of parameters and initialize the `exp.transform` variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "496ba260",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define transformation based on normalization rate\n",
    "trsf_info = [['identity',0.0,0.0,0.0,0.0],\n",
    "             ['identity',0.0,0.0,0.0,0.0]]\n",
    "trsf = Transformation(trsf_info)        \n",
    "exp.transform = trsf"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "981bc9c6",
   "metadata": {},
   "source": [
    "#### Model and surrogate definition\n",
    "We create an instance of the **Phys** model and assign `None` to the surrogate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d2431510",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define model\n",
    "model = Phys()\n",
    "exp.model = model\n",
    "\n",
    "# Get data\n",
    "model.data = np.loadtxt('./data_phys_2d.txt')\n",
    "\n",
    "# Run experiment without surrogate\n",
    "exp.surrogate = None"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "5403ba1a",
   "metadata": {},
   "source": [
    "#### Log-likelihood definiton"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "eb020c36",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Define log density\n",
    "# x: original, untransformed inputs\n",
    "# model: our model\n",
    "# transform: our transformation \n",
    "def log_density(x, model, surrogate, transform):\n",
    "    # x contains the original, untransformed inputs\n",
    "    np.savetxt(exp.output_dir + '/' + exp.name + '_x', x.detach().numpy(), newline=\"\\n\")\n",
    "    # Compute transformation log Jacobian\n",
    "    adjust = transform.compute_log_jacob_func(x)\n",
    "\n",
    "    # Get the absolute values of the standard deviations\n",
    "    stds = torch.abs(model.solve_t(model.defParam)) * model.stdRatio\n",
    "    Data = torch.tensor(model.data).to(exp.device)\n",
    "    \n",
    "    # Check for surrogate\n",
    "    if surrogate:\n",
    "        modelOut = exp.surrogate.forward(x)\n",
    "    else:\n",
    "        modelOut = model.solve_t(transform.forward(x))\n",
    "\n",
    "    # Eval LL\n",
    "    ll1 = -0.5 * np.prod(model.data.shape) * np.log(2.0 * np.pi)\n",
    "    ll2 = (-0.5 * model.data.shape[1] * torch.log(torch.prod(stds))).item()\n",
    "    ll3 = 0.0\n",
    "    \n",
    "    for i in range(2):\n",
    "        ll3 += - 0.5 * torch.sum(((modelOut[:, i].unsqueeze(1) - Data[i, :].unsqueeze(0)) / stds[0, i]) ** 2, dim=1)\n",
    "\n",
    "    negLL = -(ll1 + ll2 + ll3)\n",
    "    \n",
    "    # Return LL\n",
    "    return -negLL.reshape(x.size(0), 1) + adjust"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "977dd450",
   "metadata": {},
   "source": [
    "#### Launch inference task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f2b39abe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Temporary TEST: Physics Example - without NOFAS\n",
      "\n",
      "\n",
      "--- Running on device: cpu\n",
      "\n",
      "VI NF (t=1.000): it:     100 | loss: 5.415e+03\n",
      "VI NF (t=1.000): it:     200 | loss: 2.116e+03\n",
      "VI NF (t=1.000): it:     300 | loss: 1.017e+03\n",
      "VI NF (t=1.000): it:     400 | loss: 5.071e+02\n",
      "VI NF (t=1.000): it:     500 | loss: 2.040e+02\n",
      "VI NF (t=1.000): it:     600 | loss: 7.246e+01\n",
      "VI NF (t=1.000): it:     700 | loss: 3.559e+01\n",
      "VI NF (t=1.000): it:     800 | loss: 3.411e+01\n",
      "VI NF (t=1.000): it:     900 | loss: 3.163e+01\n",
      "VI NF (t=1.000): it:    1000 | loss: 3.339e+01\n",
      "VI NF (t=1.000): it:    1100 | loss: 3.128e+01\n",
      "VI NF (t=1.000): it:    1200 | loss: 3.222e+01\n",
      "VI NF (t=1.000): it:    1300 | loss: 3.135e+01\n",
      "VI NF (t=1.000): it:    1400 | loss: 3.143e+01\n",
      "VI NF (t=1.000): it:    1500 | loss: 3.133e+01\n",
      "VI NF (t=1.000): it:    1600 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    1700 | loss: 3.128e+01\n",
      "VI NF (t=1.000): it:    1800 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:    1900 | loss: 3.153e+01\n",
      "VI NF (t=1.000): it:    2000 | loss: 3.137e+01\n",
      "VI NF (t=1.000): it:    2100 | loss: 3.141e+01\n",
      "VI NF (t=1.000): it:    2200 | loss: 3.136e+01\n",
      "VI NF (t=1.000): it:    2300 | loss: 3.148e+01\n",
      "VI NF (t=1.000): it:    2400 | loss: 3.130e+01\n",
      "VI NF (t=1.000): it:    2500 | loss: 3.131e+01\n",
      "VI NF (t=1.000): it:    2600 | loss: 3.141e+01\n",
      "VI NF (t=1.000): it:    2700 | loss: 3.132e+01\n",
      "VI NF (t=1.000): it:    2800 | loss: 3.142e+01\n",
      "VI NF (t=1.000): it:    2900 | loss: 3.119e+01\n",
      "VI NF (t=1.000): it:    3000 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:    3100 | loss: 3.131e+01\n",
      "VI NF (t=1.000): it:    3200 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    3300 | loss: 3.122e+01\n",
      "VI NF (t=1.000): it:    3400 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:    3500 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:    3600 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:    3700 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:    3800 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:    3900 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    4000 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:    4100 | loss: 3.124e+01\n",
      "VI NF (t=1.000): it:    4200 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:    4300 | loss: 3.123e+01\n",
      "VI NF (t=1.000): it:    4400 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:    4500 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:    4600 | loss: 3.127e+01\n",
      "VI NF (t=1.000): it:    4700 | loss: 3.134e+01\n",
      "VI NF (t=1.000): it:    4800 | loss: 3.124e+01\n",
      "VI NF (t=1.000): it:    4900 | loss: 3.113e+01\n",
      "--- Saving results at iteration 5000\n",
      "VI NF (t=1.000): it:    5000 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:    5100 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:    5200 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:    5300 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    5400 | loss: 3.130e+01\n",
      "VI NF (t=1.000): it:    5500 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:    5600 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:    5700 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:    5800 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:    5900 | loss: 3.097e+01\n",
      "VI NF (t=1.000): it:    6000 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:    6100 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:    6200 | loss: 3.097e+01\n",
      "VI NF (t=1.000): it:    6300 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    6400 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:    6500 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:    6600 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:    6700 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:    6800 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    6900 | loss: 3.124e+01\n",
      "VI NF (t=1.000): it:    7000 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:    7100 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:    7200 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    7300 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    7400 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:    7500 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:    7600 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    7700 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:    7800 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:    7900 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:    8000 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:    8100 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:    8200 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:    8300 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    8400 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:    8500 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:    8600 | loss: 3.123e+01\n",
      "VI NF (t=1.000): it:    8700 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:    8800 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    8900 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:    9000 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:    9100 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:    9200 | loss: 3.123e+01\n",
      "VI NF (t=1.000): it:    9300 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    9400 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:    9500 | loss: 3.119e+01\n",
      "VI NF (t=1.000): it:    9600 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:    9700 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:    9800 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:    9900 | loss: 3.117e+01\n",
      "--- Saving results at iteration 10000\n",
      "VI NF (t=1.000): it:   10000 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   10100 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   10200 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   10300 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   10400 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   10500 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   10600 | loss: 3.097e+01\n",
      "VI NF (t=1.000): it:   10700 | loss: 3.132e+01\n",
      "VI NF (t=1.000): it:   10800 | loss: 3.123e+01\n",
      "VI NF (t=1.000): it:   10900 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:   11000 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:   11100 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   11200 | loss: 3.122e+01\n",
      "VI NF (t=1.000): it:   11300 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:   11400 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   11500 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   11600 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   11700 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   11800 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   11900 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   12000 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   12100 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   12200 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   12300 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   12400 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   12500 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   12600 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   12700 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   12800 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   12900 | loss: 3.119e+01\n",
      "VI NF (t=1.000): it:   13000 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   13100 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   13200 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   13300 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   13400 | loss: 3.095e+01\n",
      "VI NF (t=1.000): it:   13500 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   13600 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   13700 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   13800 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   13900 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   14000 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:   14100 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   14200 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   14300 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:   14400 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   14500 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   14600 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   14700 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   14800 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   14900 | loss: 3.112e+01\n",
      "--- Saving results at iteration 15000\n",
      "VI NF (t=1.000): it:   15000 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   15100 | loss: 3.133e+01\n",
      "VI NF (t=1.000): it:   15200 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   15300 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   15400 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   15500 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   15600 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   15700 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:   15800 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   15900 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:   16000 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   16100 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   16200 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   16300 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   16400 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   16500 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   16600 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   16700 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   16800 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   16900 | loss: 3.126e+01\n",
      "VI NF (t=1.000): it:   17000 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   17100 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   17200 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   17300 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   17400 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   17500 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   17600 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   17700 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   17800 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   17900 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   18000 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   18100 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:   18200 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   18300 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   18400 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   18500 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   18600 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   18700 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   18800 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   18900 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   19000 | loss: 3.099e+01\n",
      "VI NF (t=1.000): it:   19100 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   19200 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   19300 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   19400 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   19500 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   19600 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   19700 | loss: 3.096e+01\n",
      "VI NF (t=1.000): it:   19800 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   19900 | loss: 3.126e+01\n",
      "--- Saving results at iteration 20000\n",
      "VI NF (t=1.000): it:   20000 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:   20100 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   20200 | loss: 3.099e+01\n",
      "VI NF (t=1.000): it:   20300 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   20400 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:   20500 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   20600 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   20700 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   20800 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   20900 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:   21000 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   21100 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   21200 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   21300 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   21400 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   21500 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   21600 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   21700 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   21800 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   21900 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   22000 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   22100 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   22200 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   22300 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   22400 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   22500 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   22600 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   22700 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:   22800 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:   22900 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:   23000 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   23100 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   23200 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   23300 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   23400 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   23500 | loss: 3.095e+01\n",
      "VI NF (t=1.000): it:   23600 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   23700 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   23800 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   23900 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   24000 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   24100 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   24200 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   24300 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   24400 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   24500 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   24600 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   24700 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   24800 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   24900 | loss: 3.114e+01\n",
      "--- Saving results at iteration 25000\n",
      "VI NF (t=1.000): it:   25000 | loss: 3.109e+01\n",
      "\n",
      "--- Simulation completed!\n"
     ]
    }
   ],
   "source": [
    "## Run \n",
    "print('')\n",
    "print('--- Temporary TEST: Physics Example - without NOFAS')\n",
    "print('')\n",
    "\n",
    "# Experiment Setting\n",
    "exp.name = \"phys_full_2d\"        # str: Name of experiment\n",
    "exp.output_dir   = './' + exp.name    # str: output directory location\n",
    "\n",
    "# Assign logdensity\n",
    "exp.model_logdensity = lambda x: log_density(x, model, exp.surrogate, trsf)\n",
    "\n",
    "# Run VI\n",
    "exp.run()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "3769ed24",
   "metadata": {},
   "source": [
    "Notice that the model evaluation has been successfully completed by checking at the newly created **phys_nofasFree_2d** folder in our current directory.\n",
    "\n",
    "Note also that LINFA supports a post processing script to plot the results, including:\n",
    "\n",
    "* Loss profile.\n",
    "\n",
    "* Sample-based characterization of the posterior distribution.\n",
    "\n",
    "* Sample-based characterization of the posterior preditive distribution with observations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ad3b66ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plotting log...\n",
      "Plotting posterior samples...\n",
      "Plotting posterior predictive samples...\n"
     ]
    }
   ],
   "source": [
    "import linfa\n",
    "! python3 -m linfa.plot_res -n phys_full_2d -i 25000 -f \"./\" -p 'png' -d"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "9166120b",
   "metadata": {},
   "source": [
    "### Variational inference with neural network surrogate model\n",
    "\n",
    "LINFA enables the construction of an adaptively trained surrogate model to mitigate the cost of inference with computationally expensive models. \n",
    "\n",
    "By utilizing the surrogate model, model evaluations are replaced by interrogating the surrogate at a significantly cheaper cost."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "f1345650",
   "metadata": {},
   "source": [
    "#### Model and surrogate definition\n",
    "We create an instance of the **Phys** model and create an instance of a surrogate using our built-in Surrogate constructor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "921ab3ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Experiment Setting\n",
    "exp = experiment()\n",
    "\n",
    "exp.name             = \"phys_surr_2d\"      # str: Name of experiment\n",
    "exp.output_dir       = './' + exp.name    # str: output directory location\n",
    "\n",
    "exp.flow_type        = 'maf'        # str: Type of flow\n",
    "exp.n_blocks         = 5            # int: Number of layers                            \n",
    "exp.hidden_size      = 100          # int: Hidden layer size for MADE in each layer  \n",
    "exp.n_hidden         = 1            # int: Number of hidden layers in each MADE      \n",
    "exp.activation_fn    = 'relu'       # str: Actication function used                  \n",
    "exp.input_order      = 'sequential' # str: Input order for create_mask           \n",
    "exp.batch_norm_order = True         # boolean: Order to decide if batch_norm is used    \n",
    "exp.save_interval    = 5000         # int: How often to sample from normalizing flow\n",
    "\n",
    "exp.input_size    = 2               # int: Dimensionality of input                   \n",
    "exp.batch_size    = 200             # int: Number of samples generated             \n",
    "exp.true_data_num = 2               # double: number of true model evaluated      \n",
    "exp.n_iter        = 25001           # int: Number of iterations                      \n",
    "exp.lr            = 0.01            # float: Learning rate                              \n",
    "exp.lr_decay      = 0.9999          # float: Learning rate decay                \n",
    "exp.log_interval  = 100             # int: How often to show loss stat   \n",
    "\n",
    "exp.run_nofas          = True       # boolean: to run experiment with nofas\n",
    "exp.annealing          = False      # boolean: to run experiment with annealing\n",
    "exp.calibrate_interval = 1000       # int: How often to update surrogate model     \n",
    "exp.budget             = 5000       # int: Total number of true model evaluation\n",
    "\n",
    "exp.surr_pre_it  = 30000            # int: Number of pre-training iterations for surrogate model\n",
    "exp.surr_upd_it  = 6000             # int: Number of iterations for the surrogate model update\n",
    "exp.surr_folder  = \"./\"\n",
    "exp.use_new_surr = True             # boolean: to run experiment with nofas\n",
    "\n",
    "exp.results_file = 'results.txt'      # str: result text file name\n",
    "exp.log_file     = 'log.txt'          # str: log text file name\n",
    "exp.samples_file = 'samples.txt'      # str: sample text file name\n",
    "exp.seed         = random.randint(0, 10 ** 9)  # int: Random seed used\n",
    "exp.n_sample     = 5000               # int: Total number of iterations\n",
    "exp.no_cuda      = True               # boolean: to run experiment with NO cuda\n",
    "\n",
    "exp.optimizer    = 'RMSprop'          # str: Type of optimizer\n",
    "exp.lr_scheduler = 'ExponentialLR'    # str: Type of scheduler\n",
    "\n",
    "exp.device = torch.device('cuda:0' if torch.cuda.is_available() and not exp.no_cuda else 'cpu')\n",
    "\n",
    "# Define transformation based on normalization rate\n",
    "trsf_info = [['identity',0.0,0.0,0.0,0.0],\n",
    "             ['identity',0.0,0.0,0.0,0.0]]\n",
    "trsf = Transformation(trsf_info)        \n",
    "exp.transform = trsf\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "94c4c743",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warning: ./phys_surr_2d.npz does not found, please generate pre-grid.\n",
      "Suggestion: Use Surrogate.gen_grid(input_limits=None, grid_num=5, store=True)\n",
      "Warning: Surrogate model files: ./phys_surr_2d.npz and ./phys_surr_2d.npz could not be found. \n",
      "\n",
      "--- Pre-training surrogate model\n",
      "\n",
      "SUR: PRE: it:       0 | loss: 1.680e+00\n",
      "SUR: PRE: it:     500 | loss: 1.963e-02\n",
      "SUR: PRE: it:    1000 | loss: 2.407e-02\n",
      "SUR: PRE: it:    1500 | loss: 1.049e-02\n",
      "SUR: PRE: it:    2000 | loss: 7.563e-03\n",
      "SUR: PRE: it:    2500 | loss: 6.208e-03\n",
      "SUR: PRE: it:    3000 | loss: 7.349e-03\n",
      "SUR: PRE: it:    3500 | loss: 6.506e-03\n",
      "SUR: PRE: it:    4000 | loss: 4.128e-03\n",
      "SUR: PRE: it:    4500 | loss: 3.097e-03\n",
      "SUR: PRE: it:    5000 | loss: 2.454e-03\n",
      "SUR: PRE: it:    5500 | loss: 1.429e-03\n",
      "SUR: PRE: it:    6000 | loss: 1.274e-03\n",
      "SUR: PRE: it:    6500 | loss: 1.036e-03\n",
      "SUR: PRE: it:    7000 | loss: 5.934e-04\n",
      "SUR: PRE: it:    7500 | loss: 3.743e-04\n",
      "SUR: PRE: it:    8000 | loss: 7.861e-04\n",
      "SUR: PRE: it:    8500 | loss: 1.483e-03\n",
      "SUR: PRE: it:    9000 | loss: 7.878e-04\n",
      "SUR: PRE: it:    9500 | loss: 6.469e-04\n",
      "SUR: PRE: it:   10000 | loss: 8.619e-04\n",
      "SUR: PRE: it:   10500 | loss: 4.657e-04\n",
      "SUR: PRE: it:   11000 | loss: 7.060e-04\n",
      "SUR: PRE: it:   11500 | loss: 3.383e-04\n",
      "SUR: PRE: it:   12000 | loss: 3.328e-04\n",
      "SUR: PRE: it:   12500 | loss: 4.275e-04\n",
      "SUR: PRE: it:   13000 | loss: 2.478e-04\n",
      "SUR: PRE: it:   13500 | loss: 2.540e-04\n",
      "SUR: PRE: it:   14000 | loss: 2.588e-04\n",
      "SUR: PRE: it:   14500 | loss: 2.745e-04\n",
      "SUR: PRE: it:   15000 | loss: 3.259e-04\n",
      "SUR: PRE: it:   15500 | loss: 1.463e-04\n",
      "SUR: PRE: it:   16000 | loss: 1.860e-04\n",
      "SUR: PRE: it:   16500 | loss: 1.609e-04\n",
      "SUR: PRE: it:   17000 | loss: 1.595e-04\n",
      "SUR: PRE: it:   17500 | loss: 2.017e-04\n",
      "SUR: PRE: it:   18000 | loss: 1.159e-04\n",
      "SUR: PRE: it:   18500 | loss: 1.007e-04\n",
      "SUR: PRE: it:   19000 | loss: 7.205e-05\n",
      "SUR: PRE: it:   19500 | loss: 7.190e-05\n",
      "SUR: PRE: it:   20000 | loss: 6.536e-05\n",
      "SUR: PRE: it:   20500 | loss: 1.013e-04\n",
      "SUR: PRE: it:   21000 | loss: 6.878e-05\n",
      "SUR: PRE: it:   21500 | loss: 4.241e-05\n",
      "SUR: PRE: it:   22000 | loss: 3.532e-05\n",
      "SUR: PRE: it:   22500 | loss: 3.574e-05\n",
      "SUR: PRE: it:   23000 | loss: 5.258e-05\n",
      "SUR: PRE: it:   23500 | loss: 4.641e-05\n",
      "SUR: PRE: it:   24000 | loss: 3.958e-05\n",
      "SUR: PRE: it:   24500 | loss: 2.507e-05\n",
      "SUR: PRE: it:   25000 | loss: 2.378e-05\n",
      "SUR: PRE: it:   25500 | loss: 1.700e-05\n",
      "SUR: PRE: it:   26000 | loss: 1.842e-05\n",
      "SUR: PRE: it:   26500 | loss: 2.063e-05\n",
      "SUR: PRE: it:   27000 | loss: 2.230e-05\n",
      "SUR: PRE: it:   27500 | loss: 2.346e-05\n",
      "SUR: PRE: it:   28000 | loss: 1.618e-05\n",
      "SUR: PRE: it:   28500 | loss: 1.333e-05\n",
      "SUR: PRE: it:   29000 | loss: 1.124e-05\n",
      "SUR: PRE: it:   29500 | loss: 8.447e-06\n",
      "\n",
      "--- Surrogate model pre-train complete\n",
      "\n",
      "Success: [limits] loaded.\n",
      "Success: [pre_grid] loaded.\n",
      "Success: [grid_record] loaded.\n"
     ]
    }
   ],
   "source": [
    "# Define model\n",
    "model = Phys()\n",
    "exp.model = model\n",
    "\n",
    "# Get data\n",
    "model.data = np.loadtxt('./data_phys_2d.txt')\n",
    "\n",
    "# Define surrogate\n",
    "exp.surrogate = Surrogate(exp.name, lambda x: model.solve_t(trsf.forward(x)), exp.input_size, 2, dnn_arch = [128, 64],\n",
    "                          model_folder=exp.surr_folder, limits=torch.Tensor([[0, 2], [0, 10]]), \n",
    "                          memory_len=20, device=exp.device)\n",
    "surr_filename = exp.surr_folder + exp.name\n",
    "\n",
    "if exp.use_new_surr or (not os.path.isfile(surr_filename + \".sur\")) or (not os.path.isfile(surr_filename + \".npz\")):\n",
    "    print(\"Warning: Surrogate model files: {0}.npz and {0}.npz could not be found. \".format(surr_filename))\n",
    "    # 4 samples for each dimension\n",
    "    exp.surrogate.gen_grid(gridnum=4) \n",
    "    exp.surrogate.pre_train(exp.surr_pre_it, 0.01, 0.9999, 500, store=True)\n",
    "    \n",
    "# Load the surrogate\n",
    "exp.surrogate.surrogate_load()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "a7749e5a",
   "metadata": {},
   "source": [
    "#### Launch inference task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "78a7c2b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Temporary TEST: Physics Example - with NOFAS\n",
      "\n",
      "\n",
      "--- Running on device: cpu\n",
      "\n",
      "VI NF (t=1.000): it:     100 | loss: 6.041e+03\n",
      "VI NF (t=1.000): it:     200 | loss: 2.607e+03\n",
      "VI NF (t=1.000): it:     300 | loss: 6.170e+02\n",
      "VI NF (t=1.000): it:     400 | loss: 5.255e+02\n",
      "VI NF (t=1.000): it:     500 | loss: 5.003e+02\n",
      "VI NF (t=1.000): it:     600 | loss: 4.802e+02\n",
      "VI NF (t=1.000): it:     700 | loss: 4.736e+02\n",
      "VI NF (t=1.000): it:     800 | loss: 4.743e+02\n",
      "VI NF (t=1.000): it:     900 | loss: 4.734e+02\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "6.111e-03 -> 1.058e-01\n",
      "5.825e-03 -> 8.819e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.071e-02\n",
      "SUR: UPD: it:     500 | loss: 1.240e-03\n",
      "SUR: UPD: it:    1000 | loss: 8.859e-04\n",
      "SUR: UPD: it:    1500 | loss: 2.054e-04\n",
      "SUR: UPD: it:    2000 | loss: 4.302e-05\n",
      "SUR: UPD: it:    2500 | loss: 4.385e-05\n",
      "SUR: UPD: it:    3000 | loss: 9.119e-06\n",
      "SUR: UPD: it:    3500 | loss: 5.706e-06\n",
      "SUR: UPD: it:    4000 | loss: 1.156e-06\n",
      "SUR: UPD: it:    4500 | loss: 9.692e-07\n",
      "SUR: UPD: it:    5000 | loss: 1.727e-07\n",
      "SUR: UPD: it:    5500 | loss: 9.460e-08\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    1000 | loss: 5.575e+02\n",
      "VI NF (t=1.000): it:    1100 | loss: 1.842e+02\n",
      "VI NF (t=1.000): it:    1200 | loss: 8.191e+01\n",
      "VI NF (t=1.000): it:    1300 | loss: 3.252e+01\n",
      "VI NF (t=1.000): it:    1400 | loss: 3.258e+01\n",
      "VI NF (t=1.000): it:    1500 | loss: 3.238e+01\n",
      "VI NF (t=1.000): it:    1600 | loss: 3.317e+01\n",
      "VI NF (t=1.000): it:    1700 | loss: 3.371e+01\n",
      "VI NF (t=1.000): it:    1800 | loss: 3.308e+01\n",
      "VI NF (t=1.000): it:    1900 | loss: 3.260e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.216e-02 -> 1.054e-01\n",
      "2.007e-02 -> 1.384e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 9.643e-03\n",
      "SUR: UPD: it:     500 | loss: 4.002e-03\n",
      "SUR: UPD: it:    1000 | loss: 1.125e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.084e-03\n",
      "SUR: UPD: it:    2000 | loss: 3.880e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.975e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.411e-04\n",
      "SUR: UPD: it:    3500 | loss: 8.666e-05\n",
      "SUR: UPD: it:    4000 | loss: 5.408e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.771e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.647e-05\n",
      "SUR: UPD: it:    5500 | loss: 9.560e-06\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    2000 | loss: 1.604e+02\n",
      "VI NF (t=1.000): it:    2100 | loss: 7.059e+01\n",
      "VI NF (t=1.000): it:    2200 | loss: 6.960e+01\n",
      "VI NF (t=1.000): it:    2300 | loss: 6.971e+01\n",
      "VI NF (t=1.000): it:    2400 | loss: 6.940e+01\n",
      "VI NF (t=1.000): it:    2500 | loss: 6.923e+01\n",
      "VI NF (t=1.000): it:    2600 | loss: 6.874e+01\n",
      "VI NF (t=1.000): it:    2700 | loss: 6.854e+01\n",
      "VI NF (t=1.000): it:    2800 | loss: 6.824e+01\n",
      "VI NF (t=1.000): it:    2900 | loss: 6.720e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "5.759e-03 -> 2.602e-03\n",
      "3.762e-03 -> 8.234e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.198e-03\n",
      "SUR: UPD: it:     500 | loss: 6.127e-03\n",
      "SUR: UPD: it:    1000 | loss: 1.227e-03\n",
      "SUR: UPD: it:    1500 | loss: 7.419e-04\n",
      "SUR: UPD: it:    2000 | loss: 2.399e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.541e-04\n",
      "SUR: UPD: it:    3000 | loss: 9.047e-05\n",
      "SUR: UPD: it:    3500 | loss: 6.778e-05\n",
      "SUR: UPD: it:    4000 | loss: 5.160e-05\n",
      "SUR: UPD: it:    4500 | loss: 4.136e-05\n",
      "SUR: UPD: it:    5000 | loss: 3.497e-05\n",
      "SUR: UPD: it:    5500 | loss: 3.130e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    3000 | loss: 1.450e+02\n",
      "VI NF (t=1.000): it:    3100 | loss: 3.682e+01\n",
      "VI NF (t=1.000): it:    3200 | loss: 3.414e+01\n",
      "VI NF (t=1.000): it:    3300 | loss: 3.407e+01\n",
      "VI NF (t=1.000): it:    3400 | loss: 3.375e+01\n",
      "VI NF (t=1.000): it:    3500 | loss: 3.324e+01\n",
      "VI NF (t=1.000): it:    3600 | loss: 3.294e+01\n",
      "VI NF (t=1.000): it:    3700 | loss: 3.284e+01\n",
      "VI NF (t=1.000): it:    3800 | loss: 3.271e+01\n",
      "VI NF (t=1.000): it:    3900 | loss: 3.297e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "3.889e-03 -> 8.720e-02\n",
      "2.889e-02 -> 6.568e-03\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.959e-02\n",
      "SUR: UPD: it:     500 | loss: 4.191e-03\n",
      "SUR: UPD: it:    1000 | loss: 1.198e-03\n",
      "SUR: UPD: it:    1500 | loss: 3.682e-04\n",
      "SUR: UPD: it:    2000 | loss: 5.808e-04\n",
      "SUR: UPD: it:    2500 | loss: 3.874e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.417e-04\n",
      "SUR: UPD: it:    3500 | loss: 9.648e-05\n",
      "SUR: UPD: it:    4000 | loss: 7.210e-05\n",
      "SUR: UPD: it:    4500 | loss: 6.381e-05\n",
      "SUR: UPD: it:    5000 | loss: 5.444e-05\n",
      "SUR: UPD: it:    5500 | loss: 4.779e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    4000 | loss: 1.417e+02\n",
      "VI NF (t=1.000): it:    4100 | loss: 3.970e+01\n",
      "VI NF (t=1.000): it:    4200 | loss: 3.335e+01\n",
      "VI NF (t=1.000): it:    4300 | loss: 3.361e+01\n",
      "VI NF (t=1.000): it:    4400 | loss: 3.408e+01\n",
      "VI NF (t=1.000): it:    4500 | loss: 3.381e+01\n",
      "VI NF (t=1.000): it:    4600 | loss: 3.362e+01\n",
      "VI NF (t=1.000): it:    4700 | loss: 3.366e+01\n",
      "VI NF (t=1.000): it:    4800 | loss: 3.373e+01\n",
      "VI NF (t=1.000): it:    4900 | loss: 3.381e+01\n",
      "--- Saving results at iteration 5000\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "7.479e-03 -> 8.253e-02\n",
      "3.083e-03 -> 7.861e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 6.430e-03\n",
      "SUR: UPD: it:     500 | loss: 6.922e-03\n",
      "SUR: UPD: it:    1000 | loss: 2.836e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.018e-03\n",
      "SUR: UPD: it:    2000 | loss: 3.940e-04\n",
      "SUR: UPD: it:    2500 | loss: 5.492e-05\n",
      "SUR: UPD: it:    3000 | loss: 4.510e-05\n",
      "SUR: UPD: it:    3500 | loss: 3.963e-05\n",
      "SUR: UPD: it:    4000 | loss: 2.579e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.377e-05\n",
      "SUR: UPD: it:    5000 | loss: 2.022e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.840e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    5000 | loss: 7.616e+01\n",
      "VI NF (t=1.000): it:    5100 | loss: 3.364e+01\n",
      "VI NF (t=1.000): it:    5200 | loss: 3.445e+01\n",
      "VI NF (t=1.000): it:    5300 | loss: 3.379e+01\n",
      "VI NF (t=1.000): it:    5400 | loss: 3.368e+01\n",
      "VI NF (t=1.000): it:    5500 | loss: 3.400e+01\n",
      "VI NF (t=1.000): it:    5600 | loss: 3.400e+01\n",
      "VI NF (t=1.000): it:    5700 | loss: 3.401e+01\n",
      "VI NF (t=1.000): it:    5800 | loss: 3.376e+01\n",
      "VI NF (t=1.000): it:    5900 | loss: 3.397e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "6.581e-03 -> 1.361e-01\n",
      "6.170e-02 -> 9.207e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 9.361e-03\n",
      "SUR: UPD: it:     500 | loss: 5.632e-03\n",
      "SUR: UPD: it:    1000 | loss: 1.237e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.887e-03\n",
      "SUR: UPD: it:    2000 | loss: 2.834e-04\n",
      "SUR: UPD: it:    2500 | loss: 2.080e-04\n",
      "SUR: UPD: it:    3000 | loss: 3.206e-04\n",
      "SUR: UPD: it:    3500 | loss: 1.799e-04\n",
      "SUR: UPD: it:    4000 | loss: 1.234e-04\n",
      "SUR: UPD: it:    4500 | loss: 9.745e-05\n",
      "SUR: UPD: it:    5000 | loss: 8.481e-05\n",
      "SUR: UPD: it:    5500 | loss: 7.384e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    6000 | loss: 4.922e+01\n",
      "VI NF (t=1.000): it:    6100 | loss: 3.512e+01\n",
      "VI NF (t=1.000): it:    6200 | loss: 3.469e+01\n",
      "VI NF (t=1.000): it:    6300 | loss: 3.271e+01\n",
      "VI NF (t=1.000): it:    6400 | loss: 3.213e+01\n",
      "VI NF (t=1.000): it:    6500 | loss: 3.186e+01\n",
      "VI NF (t=1.000): it:    6600 | loss: 3.198e+01\n",
      "VI NF (t=1.000): it:    6700 | loss: 3.192e+01\n",
      "VI NF (t=1.000): it:    6800 | loss: 3.210e+01\n",
      "VI NF (t=1.000): it:    6900 | loss: 3.201e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "3.214e-02 -> 3.774e-02\n",
      "2.174e-02 -> 2.128e-01\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.443e-03\n",
      "SUR: UPD: it:     500 | loss: 2.069e-02\n",
      "SUR: UPD: it:    1000 | loss: 1.371e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.054e-03\n",
      "SUR: UPD: it:    2000 | loss: 6.096e-04\n",
      "SUR: UPD: it:    2500 | loss: 4.239e-04\n",
      "SUR: UPD: it:    3000 | loss: 4.691e-04\n",
      "SUR: UPD: it:    3500 | loss: 3.989e-04\n",
      "SUR: UPD: it:    4000 | loss: 3.880e-04\n",
      "SUR: UPD: it:    4500 | loss: 3.762e-04\n",
      "SUR: UPD: it:    5000 | loss: 3.714e-04\n",
      "SUR: UPD: it:    5500 | loss: 3.666e-04\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    7000 | loss: 7.511e+01\n",
      "VI NF (t=1.000): it:    7100 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:    7200 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:    7300 | loss: 3.134e+01\n",
      "VI NF (t=1.000): it:    7400 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:    7500 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:    7600 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:    7700 | loss: 3.146e+01\n",
      "VI NF (t=1.000): it:    7800 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:    7900 | loss: 3.110e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "6.882e-02 -> 1.174e-01\n",
      "6.200e-02 -> 4.097e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 3.461e-04\n",
      "SUR: UPD: it:     500 | loss: 6.803e-04\n",
      "SUR: UPD: it:    1000 | loss: 9.166e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.278e-03\n",
      "SUR: UPD: it:    2000 | loss: 5.025e-04\n",
      "SUR: UPD: it:    2500 | loss: 3.834e-04\n",
      "SUR: UPD: it:    3000 | loss: 2.359e-04\n",
      "SUR: UPD: it:    3500 | loss: 2.276e-04\n",
      "SUR: UPD: it:    4000 | loss: 2.142e-04\n",
      "SUR: UPD: it:    4500 | loss: 2.102e-04\n",
      "SUR: UPD: it:    5000 | loss: 2.085e-04\n",
      "SUR: UPD: it:    5500 | loss: 2.079e-04\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    8000 | loss: 3.129e+01\n",
      "VI NF (t=1.000): it:    8100 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:    8200 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:    8300 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:    8400 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:    8500 | loss: 3.099e+01\n",
      "VI NF (t=1.000): it:    8600 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:    8700 | loss: 3.125e+01\n",
      "VI NF (t=1.000): it:    8800 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:    8900 | loss: 3.108e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "5.031e-03 -> 2.264e-01\n",
      "1.757e-03 -> 7.609e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.045e-04\n",
      "SUR: UPD: it:     500 | loss: 1.412e-02\n",
      "SUR: UPD: it:    1000 | loss: 3.641e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.259e-03\n",
      "SUR: UPD: it:    2000 | loss: 5.400e-04\n",
      "SUR: UPD: it:    2500 | loss: 3.578e-04\n",
      "SUR: UPD: it:    3000 | loss: 3.110e-04\n",
      "SUR: UPD: it:    3500 | loss: 2.807e-04\n",
      "SUR: UPD: it:    4000 | loss: 2.633e-04\n",
      "SUR: UPD: it:    4500 | loss: 2.607e-04\n",
      "SUR: UPD: it:    5000 | loss: 2.587e-04\n",
      "SUR: UPD: it:    5500 | loss: 2.577e-04\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:    9000 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:    9100 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:    9200 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:    9300 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:    9400 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:    9500 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:    9600 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:    9700 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:    9800 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:    9900 | loss: 3.102e+01\n",
      "--- Saving results at iteration 10000\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.570e-02 -> 7.066e-02\n",
      "1.223e-02 -> 1.902e-01\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.864e-04\n",
      "SUR: UPD: it:     500 | loss: 5.465e-03\n",
      "SUR: UPD: it:    1000 | loss: 3.172e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.786e-03\n",
      "SUR: UPD: it:    2000 | loss: 9.306e-04\n",
      "SUR: UPD: it:    2500 | loss: 2.891e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.961e-04\n",
      "SUR: UPD: it:    3500 | loss: 1.701e-04\n",
      "SUR: UPD: it:    4000 | loss: 1.629e-04\n",
      "SUR: UPD: it:    4500 | loss: 1.593e-04\n",
      "SUR: UPD: it:    5000 | loss: 1.563e-04\n",
      "SUR: UPD: it:    5500 | loss: 1.560e-04\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   10000 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   10100 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   10200 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   10300 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   10400 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   10500 | loss: 3.123e+01\n",
      "VI NF (t=1.000): it:   10600 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   10700 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   10800 | loss: 3.099e+01\n",
      "VI NF (t=1.000): it:   10900 | loss: 3.103e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.554e-02 -> 5.137e-02\n",
      "3.020e-03 -> 1.343e-01\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.552e-04\n",
      "SUR: UPD: it:     500 | loss: 6.740e-03\n",
      "SUR: UPD: it:    1000 | loss: 2.061e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.745e-03\n",
      "SUR: UPD: it:    2000 | loss: 4.061e-04\n",
      "SUR: UPD: it:    2500 | loss: 2.855e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.796e-04\n",
      "SUR: UPD: it:    3500 | loss: 1.924e-04\n",
      "SUR: UPD: it:    4000 | loss: 1.509e-04\n",
      "SUR: UPD: it:    4500 | loss: 1.487e-04\n",
      "SUR: UPD: it:    5000 | loss: 1.471e-04\n",
      "SUR: UPD: it:    5500 | loss: 1.466e-04\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   11000 | loss: 3.093e+01\n",
      "VI NF (t=1.000): it:   11100 | loss: 3.082e+01\n",
      "VI NF (t=1.000): it:   11200 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   11300 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   11400 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   11500 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   11600 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   11700 | loss: 3.092e+01\n",
      "VI NF (t=1.000): it:   11800 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   11900 | loss: 3.107e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "8.279e-02 -> 3.688e-01\n",
      "6.718e-02 -> 1.169e-01\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.714e-04\n",
      "SUR: UPD: it:     500 | loss: 3.535e-03\n",
      "SUR: UPD: it:    1000 | loss: 8.724e-04\n",
      "SUR: UPD: it:    1500 | loss: 4.861e-04\n",
      "SUR: UPD: it:    2000 | loss: 6.086e-05\n",
      "SUR: UPD: it:    2500 | loss: 1.960e-04\n",
      "SUR: UPD: it:    3000 | loss: 3.153e-05\n",
      "SUR: UPD: it:    3500 | loss: 3.220e-05\n",
      "SUR: UPD: it:    4000 | loss: 1.802e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.101e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.664e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.635e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   12000 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   12100 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   12200 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   12300 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   12400 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   12500 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   12600 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   12700 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   12800 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   12900 | loss: 3.125e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "5.867e-02 -> 3.129e-01\n",
      "3.974e-02 -> 8.893e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.651e-05\n",
      "SUR: UPD: it:     500 | loss: 8.846e-03\n",
      "SUR: UPD: it:    1000 | loss: 2.439e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.040e-03\n",
      "SUR: UPD: it:    2000 | loss: 8.042e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.604e-04\n",
      "SUR: UPD: it:    3000 | loss: 7.391e-05\n",
      "SUR: UPD: it:    3500 | loss: 5.874e-05\n",
      "SUR: UPD: it:    4000 | loss: 3.408e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.803e-05\n",
      "SUR: UPD: it:    5000 | loss: 2.545e-05\n",
      "SUR: UPD: it:    5500 | loss: 2.401e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   13000 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   13100 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   13200 | loss: 3.123e+01\n",
      "VI NF (t=1.000): it:   13300 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   13400 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   13500 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   13600 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   13700 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:   13800 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   13900 | loss: 3.110e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "6.867e-02 -> 9.649e-02\n",
      "9.803e-02 -> 8.442e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.225e-05\n",
      "SUR: UPD: it:     500 | loss: 9.911e-03\n",
      "SUR: UPD: it:    1000 | loss: 2.221e-03\n",
      "SUR: UPD: it:    1500 | loss: 8.999e-04\n",
      "SUR: UPD: it:    2000 | loss: 4.299e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.834e-04\n",
      "SUR: UPD: it:    3000 | loss: 6.646e-05\n",
      "SUR: UPD: it:    3500 | loss: 3.524e-05\n",
      "SUR: UPD: it:    4000 | loss: 1.881e-05\n",
      "SUR: UPD: it:    4500 | loss: 1.630e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.680e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.486e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   14000 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   14100 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   14200 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   14300 | loss: 3.101e+01\n",
      "VI NF (t=1.000): it:   14400 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   14500 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   14600 | loss: 3.097e+01\n",
      "VI NF (t=1.000): it:   14700 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   14800 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   14900 | loss: 3.106e+01\n",
      "--- Saving results at iteration 15000\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.110e-02 -> 9.629e-02\n",
      "4.179e-02 -> 9.732e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 4.031e-04\n",
      "SUR: UPD: it:     500 | loss: 1.200e-02\n",
      "SUR: UPD: it:    1000 | loss: 2.011e-03\n",
      "SUR: UPD: it:    1500 | loss: 6.918e-04\n",
      "SUR: UPD: it:    2000 | loss: 3.562e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.094e-04\n",
      "SUR: UPD: it:    3000 | loss: 9.383e-05\n",
      "SUR: UPD: it:    3500 | loss: 3.949e-05\n",
      "SUR: UPD: it:    4000 | loss: 3.547e-05\n",
      "SUR: UPD: it:    4500 | loss: 1.806e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.695e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.533e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   15000 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   15100 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   15200 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   15300 | loss: 3.113e+01\n",
      "VI NF (t=1.000): it:   15400 | loss: 3.095e+01\n",
      "VI NF (t=1.000): it:   15500 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   15600 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   15700 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   15800 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   15900 | loss: 3.106e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "2.624e-02 -> 9.448e-02\n",
      "1.401e-02 -> 5.393e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.422e-05\n",
      "SUR: UPD: it:     500 | loss: 5.328e-03\n",
      "SUR: UPD: it:    1000 | loss: 7.122e-03\n",
      "SUR: UPD: it:    1500 | loss: 4.442e-04\n",
      "SUR: UPD: it:    2000 | loss: 4.223e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.169e-04\n",
      "SUR: UPD: it:    3000 | loss: 5.565e-05\n",
      "SUR: UPD: it:    3500 | loss: 2.981e-05\n",
      "SUR: UPD: it:    4000 | loss: 3.061e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.245e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.881e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.849e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   16000 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   16100 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   16200 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:   16300 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   16400 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   16500 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   16600 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   16700 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:   16800 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   16900 | loss: 3.120e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.507e-02 -> 3.806e-02\n",
      "9.023e-03 -> 1.233e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.642e-05\n",
      "SUR: UPD: it:     500 | loss: 9.687e-03\n",
      "SUR: UPD: it:    1000 | loss: 3.433e-03\n",
      "SUR: UPD: it:    1500 | loss: 9.443e-04\n",
      "SUR: UPD: it:    2000 | loss: 5.775e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.633e-04\n",
      "SUR: UPD: it:    3000 | loss: 7.060e-05\n",
      "SUR: UPD: it:    3500 | loss: 2.870e-05\n",
      "SUR: UPD: it:    4000 | loss: 1.850e-05\n",
      "SUR: UPD: it:    4500 | loss: 1.465e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.300e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.217e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   17000 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   17100 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   17200 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   17300 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   17400 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   17500 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   17600 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   17700 | loss: 3.115e+01\n",
      "VI NF (t=1.000): it:   17800 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   17900 | loss: 3.105e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "2.570e-02 -> 1.190e-01\n",
      "1.993e-02 -> 6.132e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.152e-05\n",
      "SUR: UPD: it:     500 | loss: 5.580e-03\n",
      "SUR: UPD: it:    1000 | loss: 1.578e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.647e-03\n",
      "SUR: UPD: it:    2000 | loss: 4.854e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.530e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.576e-04\n",
      "SUR: UPD: it:    3500 | loss: 1.219e-04\n",
      "SUR: UPD: it:    4000 | loss: 4.829e-05\n",
      "SUR: UPD: it:    4500 | loss: 4.021e-05\n",
      "SUR: UPD: it:    5000 | loss: 3.776e-05\n",
      "SUR: UPD: it:    5500 | loss: 3.609e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   18000 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   18100 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   18200 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   18300 | loss: 3.121e+01\n",
      "VI NF (t=1.000): it:   18400 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:   18500 | loss: 3.127e+01\n",
      "VI NF (t=1.000): it:   18600 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   18700 | loss: 3.119e+01\n",
      "VI NF (t=1.000): it:   18800 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   18900 | loss: 3.109e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "3.665e-02 -> 2.060e-02\n",
      "3.909e-02 -> 4.580e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 3.132e-05\n",
      "SUR: UPD: it:     500 | loss: 6.733e-02\n",
      "SUR: UPD: it:    1000 | loss: 1.541e-03\n",
      "SUR: UPD: it:    1500 | loss: 6.137e-04\n",
      "SUR: UPD: it:    2000 | loss: 3.736e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.979e-04\n",
      "SUR: UPD: it:    3000 | loss: 9.405e-05\n",
      "SUR: UPD: it:    3500 | loss: 7.938e-05\n",
      "SUR: UPD: it:    4000 | loss: 7.483e-05\n",
      "SUR: UPD: it:    4500 | loss: 5.132e-05\n",
      "SUR: UPD: it:    5000 | loss: 3.633e-05\n",
      "SUR: UPD: it:    5500 | loss: 3.246e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   19000 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   19100 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:   19200 | loss: 3.098e+01\n",
      "VI NF (t=1.000): it:   19300 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   19400 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   19500 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   19600 | loss: 3.097e+01\n",
      "VI NF (t=1.000): it:   19700 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   19800 | loss: 3.120e+01\n",
      "VI NF (t=1.000): it:   19900 | loss: 3.114e+01\n",
      "--- Saving results at iteration 20000\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "5.500e-02 -> 6.023e-02\n",
      "6.493e-02 -> 9.413e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.807e-05\n",
      "SUR: UPD: it:     500 | loss: 6.306e-02\n",
      "SUR: UPD: it:    1000 | loss: 3.708e-02\n",
      "SUR: UPD: it:    1500 | loss: 2.994e-03\n",
      "SUR: UPD: it:    2000 | loss: 2.967e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.214e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.085e-04\n",
      "SUR: UPD: it:    3500 | loss: 1.072e-04\n",
      "SUR: UPD: it:    4000 | loss: 9.212e-05\n",
      "SUR: UPD: it:    4500 | loss: 8.520e-05\n",
      "SUR: UPD: it:    5000 | loss: 8.675e-05\n",
      "SUR: UPD: it:    5500 | loss: 8.206e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   20000 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   20100 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   20200 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   20300 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   20400 | loss: 3.104e+01\n",
      "VI NF (t=1.000): it:   20500 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   20600 | loss: 3.099e+01\n",
      "VI NF (t=1.000): it:   20700 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   20800 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   20900 | loss: 3.114e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "2.027e-02 -> 1.130e-01\n",
      "4.196e-04 -> 9.385e-03\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 7.135e-05\n",
      "SUR: UPD: it:     500 | loss: 8.615e-03\n",
      "SUR: UPD: it:    1000 | loss: 2.416e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.949e-03\n",
      "SUR: UPD: it:    2000 | loss: 5.796e-04\n",
      "SUR: UPD: it:    2500 | loss: 2.990e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.299e-04\n",
      "SUR: UPD: it:    3500 | loss: 4.247e-05\n",
      "SUR: UPD: it:    4000 | loss: 2.681e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.063e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.689e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.242e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   21000 | loss: 3.131e+01\n",
      "VI NF (t=1.000): it:   21100 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   21200 | loss: 3.110e+01\n",
      "VI NF (t=1.000): it:   21300 | loss: 3.096e+01\n",
      "VI NF (t=1.000): it:   21400 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   21500 | loss: 3.116e+01\n",
      "VI NF (t=1.000): it:   21600 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   21700 | loss: 3.117e+01\n",
      "VI NF (t=1.000): it:   21800 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   21900 | loss: 3.104e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "4.258e-02 -> 1.241e-01\n",
      "3.361e-02 -> 1.448e-01\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 4.185e-06\n",
      "SUR: UPD: it:     500 | loss: 2.770e-02\n",
      "SUR: UPD: it:    1000 | loss: 3.364e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.270e-03\n",
      "SUR: UPD: it:    2000 | loss: 1.012e-03\n",
      "SUR: UPD: it:    2500 | loss: 1.679e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.023e-04\n",
      "SUR: UPD: it:    3500 | loss: 5.261e-05\n",
      "SUR: UPD: it:    4000 | loss: 3.341e-05\n",
      "SUR: UPD: it:    4500 | loss: 2.590e-05\n",
      "SUR: UPD: it:    5000 | loss: 2.229e-05\n",
      "SUR: UPD: it:    5500 | loss: 1.942e-05\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   22000 | loss: 3.102e+01\n",
      "VI NF (t=1.000): it:   22100 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   22200 | loss: 3.095e+01\n",
      "VI NF (t=1.000): it:   22300 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   22400 | loss: 3.097e+01\n",
      "VI NF (t=1.000): it:   22500 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   22600 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   22700 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   22800 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   22900 | loss: 3.098e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.858e-02 -> 1.058e-01\n",
      "1.554e-02 -> 1.148e-01\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 1.150e-05\n",
      "SUR: UPD: it:     500 | loss: 2.360e-03\n",
      "SUR: UPD: it:    1000 | loss: 6.171e-03\n",
      "SUR: UPD: it:    1500 | loss: 2.671e-03\n",
      "SUR: UPD: it:    2000 | loss: 1.605e-03\n",
      "SUR: UPD: it:    2500 | loss: 1.426e-04\n",
      "SUR: UPD: it:    3000 | loss: 1.423e-04\n",
      "SUR: UPD: it:    3500 | loss: 2.730e-05\n",
      "SUR: UPD: it:    4000 | loss: 1.561e-05\n",
      "SUR: UPD: it:    4500 | loss: 1.060e-05\n",
      "SUR: UPD: it:    5000 | loss: 8.279e-06\n",
      "SUR: UPD: it:    5500 | loss: 7.292e-06\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   23000 | loss: 3.114e+01\n",
      "VI NF (t=1.000): it:   23100 | loss: 3.105e+01\n",
      "VI NF (t=1.000): it:   23200 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   23300 | loss: 3.108e+01\n",
      "VI NF (t=1.000): it:   23400 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   23500 | loss: 3.103e+01\n",
      "VI NF (t=1.000): it:   23600 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   23700 | loss: 3.100e+01\n",
      "VI NF (t=1.000): it:   23800 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   23900 | loss: 3.113e+01\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "5.308e-02 -> 3.697e-02\n",
      "3.758e-02 -> 9.549e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 4.552e-06\n",
      "SUR: UPD: it:     500 | loss: 2.124e-02\n",
      "SUR: UPD: it:    1000 | loss: 6.420e-04\n",
      "SUR: UPD: it:    1500 | loss: 2.259e-03\n",
      "SUR: UPD: it:    2000 | loss: 7.607e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.122e-04\n",
      "SUR: UPD: it:    3000 | loss: 4.939e-05\n",
      "SUR: UPD: it:    3500 | loss: 2.542e-05\n",
      "SUR: UPD: it:    4000 | loss: 1.158e-05\n",
      "SUR: UPD: it:    4500 | loss: 4.708e-06\n",
      "SUR: UPD: it:    5000 | loss: 2.826e-06\n",
      "SUR: UPD: it:    5500 | loss: 2.281e-06\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   24000 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   24100 | loss: 3.109e+01\n",
      "VI NF (t=1.000): it:   24200 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   24300 | loss: 3.106e+01\n",
      "VI NF (t=1.000): it:   24400 | loss: 3.118e+01\n",
      "VI NF (t=1.000): it:   24500 | loss: 3.107e+01\n",
      "VI NF (t=1.000): it:   24600 | loss: 3.111e+01\n",
      "VI NF (t=1.000): it:   24700 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   24800 | loss: 3.112e+01\n",
      "VI NF (t=1.000): it:   24900 | loss: 3.095e+01\n",
      "--- Saving results at iteration 25000\n",
      "\n",
      "--- Updating surrogate model\n",
      "\n",
      "Std before inflation -> Std after inflation\n",
      "1.051e-02 -> 7.590e-03\n",
      "1.186e-03 -> 4.647e-02\n",
      "\n",
      "SUR: UPD: it:       0 | loss: 2.387e-06\n",
      "SUR: UPD: it:     500 | loss: 5.112e-03\n",
      "SUR: UPD: it:    1000 | loss: 2.201e-03\n",
      "SUR: UPD: it:    1500 | loss: 1.043e-03\n",
      "SUR: UPD: it:    2000 | loss: 6.130e-04\n",
      "SUR: UPD: it:    2500 | loss: 1.447e-04\n",
      "SUR: UPD: it:    3000 | loss: 4.833e-05\n",
      "SUR: UPD: it:    3500 | loss: 2.324e-05\n",
      "SUR: UPD: it:    4000 | loss: 1.961e-05\n",
      "SUR: UPD: it:    4500 | loss: 1.144e-05\n",
      "SUR: UPD: it:    5000 | loss: 1.022e-05\n",
      "SUR: UPD: it:    5500 | loss: 9.295e-06\n",
      "\n",
      "--- Surrogate model updated\n",
      "\n",
      "VI NF (t=1.000): it:   25000 | loss: 3.107e+01\n",
      "\n",
      "--- Simulation completed!\n"
     ]
    }
   ],
   "source": [
    "## Run \n",
    "print('')\n",
    "print('--- Temporary TEST: Physics Example - with NOFAS')\n",
    "print('')\n",
    "\n",
    "# Assign logdensity\n",
    "exp.model_logdensity = lambda x: log_density(x, model, exp.surrogate, trsf)\n",
    "\n",
    "# Run VI\n",
    "exp.run()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "95a4e1ce",
   "metadata": {},
   "source": [
    "Plot results in the newly created **phys_surr_2d** folder in our current directory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3315575f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plotting log...\n",
      "Plotting posterior samples...\n",
      "Plotting posterior predictive samples...\n"
     ]
    }
   ],
   "source": [
    "import linfa\n",
    "! python3 -m linfa.plot_res -n phys_surr_2d -i 25000 -f \"./\" -p 'png' -d"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "2c8caabe",
   "metadata": {},
   "source": [
    "#### Results\n",
    "The plots below offer a comparison between variational inference with the full model and with the neural network surrogate. "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "0b383b11",
   "metadata": {},
   "source": [
    "##### Loss profile plots\n",
    "\n",
    "The first set of plots represent the profile of the loss function as iterations increase."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "08cd8616",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Phys. without Surrogate')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## log-loss plots\n",
    "\n",
    "# create figure\n",
    "fig = plt.figure(figsize=(10, 7))\n",
    "plt.style.use('dark_background')\n",
    "\n",
    "# setting row and column variables\n",
    "rows = 1 \n",
    "columns = 2\n",
    "\n",
    "# reading images\n",
    "logLoss_phys = cv2.imread('./phys_surr_2d/log_plot.png')\n",
    "logLoss_physFree = cv2.imread('./phys_full_2d/log_plot.png')\n",
    "\n",
    "# Adds a subplot at the 1st position\n",
    "fig.add_subplot(rows, columns, 1)\n",
    "\n",
    "# Add subplots corresponding to Phys with Surrogate\n",
    "plt.imshow(logLoss_phys)\n",
    "plt.axis('off')\n",
    "plt.title(\"Phys. with Surrogate\")\n",
    "\n",
    "# Add subplots corresponding to Phys without Surrogate\n",
    "fig.add_subplot(rows, columns, 2)\n",
    "# showing image\n",
    "plt.imshow(logLoss_physFree)\n",
    "plt.axis('off')\n",
    "plt.title(\"Phys. without Surrogate\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "da631ffb",
   "metadata": {},
   "source": [
    "##### Plots of predictive posterior distribution and observations\n",
    "\n",
    "The second set of plots are represents the compatiblity between model outputs and observations. The output values corresponding to the true model parameters are: <br>\n",
    "\n",
    "$x_1$: *3.2070* <br>\n",
    "$x_2$: *0.8828* <br>\n",
    "\n",
    "Samples from the posterior predictive distribution are represented with blue dots. \n",
    "\n",
    "We expect to observe these samples to be compatible with the available observations (in red)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "96a027f2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Phys. without Surrogate')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## output plots\n",
    "\n",
    "# create figure\n",
    "fig = plt.figure(figsize=(10, 7))\n",
    "plt.style.use('dark_background')\n",
    "\n",
    "# setting row and column variables\n",
    "rows = 1\n",
    "columns = 2\n",
    "\n",
    "# reading images\n",
    "phys1 = cv2.imread('./phys_surr_2d/data_plot_phys_surr_2d_25000_0_1.png')\n",
    "physFree1 = cv2.imread('./phys_full_2d/data_plot_phys_full_2d_25000_0_1.png')\n",
    "\n",
    "# Add subplots corresponding to Phys with Surrogates\n",
    "fig.add_subplot(rows, columns, 1)\n",
    "plt.imshow(phys1)\n",
    "plt.axis('off')\n",
    "plt.title(\"Phys. with Surrogate\")\n",
    "\n",
    "# Add subplots corresponding to Phys without Surrogate\n",
    "fig.add_subplot(rows, columns, 2)\n",
    "plt.imshow(physFree1)\n",
    "plt.axis('off')\n",
    "plt.title(\"Phys. without Surrogate\")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "dbc48711",
   "metadata": {},
   "source": [
    "##### Plots of posterior density\n",
    "\n",
    "The third set of plots show samples from the posterior distribution. The true parameter set was initially selected as:\n",
    "\n",
    "$z_1$: *1* <br>\n",
    "$z_2$: *5* <br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "8f705415",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Phys. without Surrogate')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjwAAADECAYAAACbUfvHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAADSxElEQVR4nOy9d7RdV3Xv/1m7nH16ub3fK131LlluspFtDDY2xqaXYSCPJISMQF54YZAXv5AfLyEBkoxAQnsvIaHZPDAhTgCDIdiWi9xUbPUuXV3dXk9vu6z1++NKx76WbMtGtnTl/fHYHtr77LL2Pmd/71xzzTmXABQ+Pj4+Pj4+Phcx2vlugI+Pj4+Pj4/Pq41v8Pj4+Pj4+Phc9PgGj4+Pj4+Pj89Fj2/w+Pj4+Pj4+Fz0+AaPj4+Pj4+Pz0WPb/D4+Pj4+Pj4XPT4Bs9vyKZNm/id3/md892M0+js7CSfz6NpL/wVK6Xo7e19DVvl4+PzWuDrko/P6fgGz1nQ19dHqVQin88zOjrKt7/9bSKRyPlu1osyMDBALBZDSgn85gKYSCT413/9V0ZGRsjlchw8eJD/+T//57lq7qvKZz/7We68887z3Qwfn3OKr0svj7PRgauuuorHHnuMTCbD1NQUmzdvZv369a9J+35T+vr6uP766893My5ofIPnLHnb295GLBZj3bp1rF+/ns985jPnu0mvKV/+8peJRqMsXbqURCLBrbfeypEjR17RuXRdP23bi/X4fHx8zszrXZfOJbFYjHvvvZevfvWr1NXV0d7ezl/8xV9QrVZf0fmEELPWz6R7Pq89yl9efOnr61PXX399bf1v//Zv1c9+9jMFqE2bNqm//Mu/VJs3b1a5XE796le/UvX19QpQ9957r/rEJz4x61w7d+5Ub3/72xWgvvSlL6mxsTGVzWbVrl271PLly1+yLf/7f/9v9ZWvfEUByjAMVSgU1N/+7d8qQAWDQVUul1UqlVLd3d1KKaV0XVd/9Vd/pVzXVeVyWeXzefXVr35VAUoppT72sY+pQ4cOqXQ6rb72ta+94HV3796tbrvttjN+9txrndq2adMm9Tu/8zsKUL/1W7+lNm/erL70pS+pyclJ9bnPfU59+9vfVt/4xjfUz3/+c1UoFNT111+vlixZojZt2qTS6bTas2ePetvb3lY7X11dnfrpT3+qstms2rJli/rc5z6nHn300drn//AP/6BOnDihstms2rZtm7r66qsVoG688UZVrVaVbdsqn8+rHTt2KEDF43H1L//yL2p4eFgNDg6qz33uc0rTtPP+W/MXfznb5fWoS0II9Wd/9mfq+PHjamxsTH33u99V8XhcAeqaa65RAwMDZ3xGL6QDz10uueQSlU6nX/AeP/vZz6o777yztv583du0aZP6q7/6K7V582ZVKpVUb2+vUkqpP/iDP1CHDh1Sx44dU4D63d/9XXX48GE1NTWlfvKTn6jW1tbaOd/85jerAwcOqEwmo77+9a+rhx56qKaj8+fPVw888ICanJxUExMT6q677lKJREIB6nvf+57yPE+VSiWVz+fVpz/9aQWoyy+/XD322GMqnU6rHTt2qGuuuea8/27P83LeG3DBL88Vlo6ODrVnzx71l3/5lwpmfuRHjhxRCxcuVMFgUG3atEl94QtfUIB6z3veo5588snaeVatWqUmJyeVaZrqhhtuUNu2bav9YJcsWaJaWlpesi3XXXed2rVrlwLUlVdeqY4cOVK7xnXXXVd7kc/0Mp56cU4tSin1s5/9TCUSCdXZ2anGx8fVjTfeeMbrfvOb31R79uxR/+2//Te1YMGCWZ+djcHjOI76xCc+oXRdV8FgUH37299WmUxGbdiwQQkhVDQaVYcPH1Z33HGHMk1TXXfddSqXy6lFixYpQP3gBz9QP/jBD1QoFFJLly5VJ06cmGXw3H777aqurk7puq7++I//WI2MjCjLshScLlSAuueee9T//b//V4XDYdXY2Kieeuop9Xu/93vn/bfmL/5ytsvrUZc+8pGPqMOHD6t58+apSCSi/v3f/11973vfU/DiBg+cWQeeu8RiMTU5Oam+853vqLe85S0qmUzO+vxsDJ7+/n61bNkypeu6MgxDKaXUf/3Xf6lUKqWCwaC67rrr1MTEhFq7dq0KBALqK1/5inr44YcVoOrr61U2m1XveMc7lK7r6r//9/+ubNuuPZ/e3l71pje9SQUCAdXQ0KAefvhh9eUvf/mM9wqotrY2NTk5qW666SYlhFBvetOb1OTkpGpoaDjvv93ztfjjCGfJf/7nf5JOp9m8eTMPP/wwn//852ufffvb3+bw4cNUKhV+9KMfsWbNGgB++tOfsmjRIhYsWADAhz70Ie6++24cx8FxHGKxGEuWLEEIwYEDBxgdHX3JdjzxxBMsXLiQuro6Nm7cyL/+67/S3t5OJBLhmmuu4eGHH35Z9/XFL36RbDbLwMAAmzZtqrX9+fzhH/4h3//+9/nEJz7Bvn37OHz4MG95y1vO+jrDw8N87Wtfw/M8KpUKAD/5yU94/PHHUUqxZs0aotEoX/ziF3Ech02bNnHvvffygQ98AE3TeNe73sVnP/tZyuUy+/fv57vf/e6s83//+99nenoaz/P40pe+hGVZLF68+IxtaWpq4uabb+aTn/wkpVKJiYkJvvzlL/P+97//rO/Hx+dC4PWmS7fffjtf+tKX6Ovro1gscscdd/D+97//nAwX5fN5rr76apRSfPOb32RiYoKf/OQnNDU1nfU5vvOd77Bv3z48z8N1XQC+8IUvkE6nqVQq3H777XzrW9/imWeewbZt7rjjDq688kq6u7u5+eab2bt3L//xH/+B53l85StfmfXsjx49yv33349t20xOTvKlL32Ja6655gXb8sEPfpBf/OIX3HfffSiluP/++9m2bRs333zzK39Icxzf4DlL3v72t5NKpejp6eHjH/947Y82MOtHWSqViEajAFSrVe6++24++MEPIoTgAx/4QC1obtOmTXzta1/j61//OuPj4/zTP/0TsVjsJdtRqVTYtm0b11xzDRs3buThhx/m8ccf56qrrnpFwvJCbT/Tdb/whS+wfv166uvr+dGPfsS//du/kUqlzuo6AwMDL7qtra2NgYEBlFK1bf39/bS3t9PY2IhpmrP2f/75PvWpT7Fv3z4ymQzpdJpEIkFDQ8MZ29Ld3Y1pmoyMjJBOp0mn0/zTP/3TyxI2H58LgdebLrW1tdHf31/7rL+/H9M0aW5uflnnfyEOHDjARz7yETo7O1mxYgVtbW38wz/8w1kffzY699z2F4tFpqamaG9vr2ngcxkcHKz9u6mpiR/84AcMDg6SzWa56667XlDjYEbn3vOe99Q0Lp1Oc/XVV9Pa2nrW93Ox4Rs8rzLf/e53uf3227n++usplUo8+eSTtc+++tWvsn79epYtW8aiRYv49Kc/fVbnfPjhh3njG9/I2rVr2bp1Kw8//DA33ngjl112GY888sgZj3muIfGbks/n+fznP080GmXevHkUi0UAwuFwbZ+WlpaXvP5ztw0PD9PZ2Tkr0K+rq4uhoSEmJiZwHIeOjo7aZ52dnbV/X3311fzJn/wJ733ve0mlUqRSKbLZbO1cz7/2wMAA1WqVhoaG2v6JRIIVK1a8ksfh4zPnmKu6NDw8THd3d229q6sLx3EYGxujWCzO0iBN02hsbHzF1zp48CDf+c53arrw/PM/X+Ne6BrP17nntj8cDlNfX8/Q0BAjIyOzNA6Ytf75z38epRQrV64kkUjUDNYXuvbAwAB33nlnTeNSqRTRaJS/+Zu/OdtHcNHhGzyvMk8++SRSSv7+7/9+Vkrk+vXrueyyyzAMg2KxSKVSqaVqvhQPP/wwH/7wh9m3bx+O4/DQQw/xu7/7u/T19TE5OXnGY8bGxpg/f/4rvo/PfOYzrF+/HtM0sSyLP/qjPyKdTnPw4EEmJycZHBzkgx/8IJqm8ZGPfORl19F46qmnKJVK/Mmf/AmGYXDNNdfwtre9jR/+8IdIKbnnnnv43//7fxMKhVi8eDEf/vCHa8fGYjFc12ViYgLDMPjzP/9z4vH4rHvv6empicPo6Cj/9V//xd///d8Ti8UQQjB//nw2btz4ip+Pj89cYq7q0g9+8AP+x//4H/T09BCJRPj85z/P3Xffjed5HDp0iGAwyM0334xhGHzmM5/BsqxZ13quDjyfxYsX88d//Me0t7cDM8bGBz7wgZoxuGPHDjZu3EhnZyfxeJw77rjjrNv93PZ/5CMfYfXq1QQCAT7/+c/z1FNP0d/fz89//nNWrlzJbbfdhq7rfPzjH59lVMViMQqFAtlslra2ttMM0ec/y7vuuou3ve1t3HDDDWiahmVZXHPNNbX7ez3iGzyvAd/73vdYtWoVd911V21bPB7nm9/8Jul0mv7+fqampvi7v/s7AO644w5+8YtfvOD5Hn/8cUKhUK3XtG/fPiqVygv2ogD+8R//kXe/+91MT0/zj//4jy/7HpRSfPvb32ZycpLh4WHe/OY389a3vrXm3fnoRz/Kpz/9aaampli+fDmPP/74yzq/4zi87W1v46abbmJycpJvfOMbfPjDH+bgwYMAfOITnyCRSDA6Osqdd97JD37wg1q66K9+9St++ctfcujQIfr7+6lUKrNcw//2b/8GwNTUFNu3bwfgwx/+MIFAgH379pFOp/nxj3/8unb1+rz+mIu69K1vfYs777yTRx55hL6+PiqVCn/4h38IQC6X4w/+4A/4l3/5F4aGhigWi7OGhM6kA88ln89z+eWX89RTT1EoFHjyySfZs2cPn/rUpwC4//77ufvuu9m1axfbt2/n3nvvfcn2Pp8HHniAP//zP+ff//3fGRkZobe3txY7ODU1xXve8x7+9m//lqmpKZYtW8a2bdtqOvcXf/EXrFu3jmw2y89//nPuueeeWef+whe+wGc+8xnS6TSf+tSnGBwc5LbbbuN//a//xcTEBAMDA3z6059+3ZcAOe+R0xf78qEPfWhWRpG//ObLF7/4RfWd73znvLfDX/xlri6+Ll3YixBCDQ0NqWuvvfa8t+UiWs57Ay7qJRQKqSeeeEJ96EMfOu9tmcvL4sWL1cqVKxWgLr30UjUxMfGCdYH8xV/85cUXX5cuzOWGG25QiURCBQIB9Wd/9mdqeHhYBYPB896ui2g57w24aJcbbrhBFQoF9Z//+Z+zatT4y8tf1q9frw4fPqyKxaI6duyY+tM//dPz3iZ/8Ze5uPi6dOEun/3sZ9Xk5KTK5XLqySefVJdddtl5b9PFtIiT//Dx8fHx8fHxuWh5fUcv+fj4+Pj4+Lwu8A0eHx8fHx8fn4se3+Dx8fHx8fHxuejxDR4fHx8fHx+fix7f4PHx8fHx8fG56PENHh8fHx8fH5+LHt/g8fHx8fHx8bno8Q0eHx8fHx8fn4se3+Dx8fHx8fHxuejxDR4fHx8fHx+fix7f4PHxOU8IIWb9+7nrPj4+Phc6c03DjPPdAB+f1xtCCK644go6Ozv50Y9+RCKRYMOGDXiex/Hjxzl06ND5bqKPj4/PCzJXNcz38Pj4vAJisRimadbWdV2noaGB+vp6NO3FXyulFGNjY3R1dSGE4Oqrr6ZSqbB7925uvfXWV7vpPj4+Pq9LDfM9PD4+Z0DTNKSUp63rus6VV17Jhz70Ib7+9a+za9cuhBDcdtttFIvFmoDce++9dHZ2sm7dupqbt1wu89BDD1GtVimVSiilAAiHw+RyOaSUhMPh1/5mfXx8Ljp8DTsd3+Dx8XkegUCAd7/73dx3332k02lWrFhBU1MTDz74IJ7nsXnzZm655RZ0XQegvr6eN77xjXzyk58kGo3y13/919x///0MDAwwODhYO+8pcRBCMH/+fFpaWqirq2PLli2sW7cOTdPYvHnzeblnHx+fiwdfw86Mb/D4+DwP27Z54okneO9738vu3buZN28e99xzz6x9Tr34AMlkkoaGBlzXxbZtIpEI9fX1DA0NzdrvuccePHiQr3zlKxSLRaampkin0+i6Ti6Xe9Xvz8fH5+LG17Az48fw+Picgb6+Pvr7+/n4xz/OQw89RLlcfsF9XdfFcZxaloLneVSr1Rc9/9TUFAMDA1QqFQByuRzpdBrP887pffj4+Lw+8TXsdHyDx8fnDKxYsYLGxkY+97nPccstt5BKpWqfaZqGYRgYhoEQgtHRUYaHh2lsbKSxsZHx8XHS6fR5bL2Pj8/rHV/DTkcAp/urfHxexwQCAW644QYeeOAByuUy8+bNo7Ozk0ceeQRN07jiiiu4/vrrGR4e5te//jUnTpxg3rx5rFixAiklO3bsYGho6Hzfho+Pz+sUX8POjG/w+PicASHErLHr56/7+Pj4XMj4GnY6ftDySyCEYPXq1bS0tOC67nlrh1LqrKpYnu1+54rfpF0vp62/6X2dOv6l2nEun9+psfBHHnnkBX87TU1NrFq16rTthUKBLVu2zEor9fF5Jfgadm6u52vY3Ncw3+B5CTRN4xOf+ARvWLOBQwNHcDlzQJaJQUJEmVY5JKd/wRYmKRFnSmVxOH+i4/PaEYlEaGtr46qrriKbzZ5xn40bN/IPP7ibfc6zPa+wgOiRA2y4/DJKpdJr1Vyfi5RTGrZmzRoGBgZe9vERQsRFmHGVxjuDtr0YdcRxcMkz+3esoREkQJkqyh9kuGC52DTMN3heAqUUtm3z91/9Et++67s1Y8ZAR0ejigNAAINurZUTcowq9qxz6Gg0iCRLtB62yf0U1QtHy59CABYBXLwXNLJ8Lmzmz5/PnXfe+aK9aiEET9qS/5V99lvu0QWff22a6PM64JSGffWrX+Wuu+562cfXiTidoom9su9laZFAcIVYQZYC+9XxWYZNEynWiEVsVjspUSHATLE7+6Se+lwYXGwa5hs8L8GpMc+SV8Hxnn0ZF4gOmkSKx+UuPCQaAbpoYlpmKT3PoJknWujSWths78DGQUOjSSSZUrmatyf4POMmSICrtFVMkeNpecDvBc1BzjY9UyKpKgd58ju2lZ886XPuOKVhnufVfpMmOs5ZGi8TpJkk/YoU6Cn2IFF4z7vWJBm2s58iJSSKxSxAR+NpfK27kLjYNOzCbNUcYEiNc0Aer7l4i5TZ7O1iUmVO23dETbNX9tV6LyEs1hlLSIooMOMBulRfyiKts3aMg0uZClJ56GgsEJ3EuHBLdvu8cpSSSGw85eApB+n3cn1eRYIEeIO+hkaSZ33MKzVBHNzTjJ1T28dJ1/5A9jHEUQZ8Y2eOMlc0zPfwvEIKlCkw25NTonLGfYuUZw1jFSnzkPM0FWYKO3lI9nvHqTxnKCxKmAYtyZg3jY5Gi6ijoErkKaGjoaOzkgUMMsYY06/CHfq8VigkUtk1qVdK52z+xASDQRYuXMjU1BTDw8OnfR4KhVi1ahWmabJ//36mpqbObcN95iQuHifkGEVeemj9tSJL4Xw3wec3YK5omO/hOUt0dDpEExbmS+98FpwyjsyTNuck2ZoBpSGwsdkvj5Mmj47Ok3IPY8x82au0hSwX86li+/E9FwUSpexnF5yz6ucuWrSIj370oyxfvvy0zzRN4/bbb8d1Xfr7+/n93/99LMs6901/nWKaJqlUCl3XSSaTtTmJnk88HicQCPxG10qlUjQ1Nc2a2ToajdLU1EQkEnnZ53PxOKaGKFGlWaRIidiL7h8TYSIi+LKvI5jxXvu8HpgbGnZePDwdHR284x3v4NChQzQ0NPCTn/yEQuFZC1/TNObNm8fVV1/No48+yrFjx17ynLquI6WcVWegrq6ODRs2ADA4OMjOnTsxTZNrrrkG0zSpVCovmm73XEIiwFKtG1s6jKuZCpQdohEHlzH1yipSdunN9BptPFLdVYvl0dG51FzMkDfBkJrgCmMFEzLNDu9w7bhhOYFEveJxdZ8LDQnYPKd7RDQa4c1vfnOtbHs6nWbLli2zjtq1axf9/f1nTEENh8Ns3LiR73//+ziOQ11dHR0dHRw9evRVvpfXB4sWLeLLX/4yIyMj7Nmzh69+9auz4h0Mw+Daa6/lfe97H//n//wfnn766Zc8Z2trK5lMZtYUABs2bKCjo4PBwUFWrFjBd77zHZqbm7npppt4+umnufzyy7n77ruZnJw867Y3kqRXb+cZ7zDz9DaKqoInJUVVPi0LSyBYY/ZSweagM4AmBBlZPKvrtGoNLDDaeMLe52emXvTMDQ07L+Z3KpViyZIlbNy4kf37988ydk4hpeSqq66ivr7+rM557bXX0tzcXFvXNI3f/u3f5vjx4/z617/mve99Ly0tLdx4442kUil++ctfsnjxYq644oqzOn9RVXjU28nEyRgdATRrdTRoydo+BjpREXrBc0REkEYtgYYgKAJMyxx97khtjLtFr6Nei5FXJSo49OptCCE56J2YdZ4xppl4nrGjoxEnwmtXvcLnXCHw0JT97IJDKpniIx/5CB/96Ef56Ec/ytve9raXdU7DMGhoaJjVCairq3s1mv+6RAjBl7/8Zf7u7/6Or371qzVRP4Xrutx///1ks9lZ3h8hBIFAYJa35hRXXHEFDQ0NtXXTNLn99tt5+OGHeeKJJ2hvb2fZsmW8973v5dChQ2zbto3x8XFuvvnml9V2F4+SquDhsd09SL8c4RpzNc0iddq+CsV25zB7neMsCXSxKjB/pm0YdBqNbAguo9dsQ3vOn5L5RiuXBBZRVlUmZfaMZTp8Li7mioadFw+PlJLvf//7lMtlRkZGzvh5X1/fabOudnZ2smDBAjzPY+vWrbN6QvF4fJaIWJbFxo0b+eY3v0m1WiWbzXLppZdy00038a1vfQspJUeOHOHGG288bTr7UxOowYzhJIRAoWbF2IQIss87XktLB+jWm5lvtPFQdQcOLgEMXDwkCoGg12ylWa/jkDPAisA8tlQP0BpIMVXNkpUllpidGEJnU3kHHpKcLHJcjGLjEBNhklqEQW/yjIF9jSLJKm0Bj3m7L6ixeZ+XRiiJrqo1Y1VTMDg0wIc+9KFaDYuXUyFVCIHjOIyPjxMIBPA8D03TmJiYeBVa//pESkk4HKauro4bb7yR++67D9u2X/QYwzB45zvfSalUor6+nq1bt7Jv374X3N+yLJYsWUK5XEYpRT6fZ8WKFaxbt46HHnoIgLGxMd7znvfwve99b9axqVSq1lnUNI1EIgFAAwnma208Iw/XhsNd5fGks5dGPYmUilE5OyawpGaMuZ3Vo2hiprPWZTayyppHSdo0GQmG3UnKaub+HeVSVQ5plSfrFggLCwcPW7kvGJSsn6zLU3yBOEifC5u5omHnxeBJp9PEYjEGBgb45Cc/ybe+9a2XHLZqamri05/+NH/xF3/BkiVL+OAHP8g3v/nN2ueaNttZZRgGsVis9pBd16WhoYG6urpa5cdyuTzLK3SK2267jXe/+93AzIO/9NJLeeyxx2qf62hcZixjXE2zzzte2z7sTZJVRRxcTAyuD69jv93PcXeMeWYzbWY9+6r9pL0Cu+xjlKWNfI4EnHDHWWH1oKPhIXHxWBdcwJA7hSs9eowWxrw09hncw5Mqy3Z58AUDp59LlBBNpDjOqN/7ugAQeBiqinfyt6orQBmz0ojPRFNTE62treRyOZLJJJlMhssuu4zW1lZ++tOf8utf/5oNGzYwOjrKyMjIGYMCfV4Zx48fZ2hoiHw+z5//+Z9TKBR44IEHXvSYJUuWcNttt3HHHXfQ1tbGrbfeSjabrXl1FixYQKlUqmnUyMjIaa5+0zRP07ozxQi9/e1v5/d+7/dq6/PmzeO+++7DQ1LFmWV4KBSTKss80YYjXEZPbhcI6vUYWa+IgzfTuVOwIbQUS5g8WNxJTpYwhIYpDJJahJKsMibTTMgMAPV6nOvCq/CQPFk+wJB75qDTZq2OtcZCHrS3U+bFDUefC4+5omHnxeBJJBIcPnyYqakp0uk011xzzUsaPEuXLqWzs5PVq1cTDAYZGxvj2muvZc2aNQCsXr2alStXkslksG2bn/3sZ7Nic0zTpFAoUKlUaiISDAbJ5/OnXWvfvn3827/9GzBj8DzfjeYh2ekdpqpmv5hlbMrSpkNvIKKFOOIM0WM2U1JVMl6BMTfNFeEl7K4cZ789U/H0ifL+2vElKgRNg6AWwJYuEskRZ5i8LJGVRTKqwNWhFWyvHib7vHF0F48pNVMJUwBJEaOgyrWx8+dOmhYjQptoZFBNYPsGz3lHUxJd2YiTYmEonbN5NYPBIP/xH/+BlLL2R+/w4cMMDQ2hlOLuu++mt7cXTdP4+te//pIeCJ+z5wMf+ABPPvkke/fupVAo0NXV9ZLHnPKyBINBpqen+f73v084HK7NYh0MBkkkElSrVTzPY3h4mFwuh67rCCEIh8M888wzTE5O1rzZyWTyjNp55513cvfddwMzncEvfelLAKTJk5ana54Ctjr7Ucx06ALCREfj2vAqHi/vZ8SdYqnViYtkT7UfgahpkKc81oV7aQ/UU5JVDKHjKo/7cztIewUeK+8jLCwy3gvH/kzKDFuc/VTOkM5souPi+fGKFzBzRcPOi8Fz/fXX8+CDDzI9PU0ymWTnzp1omlYLzjvlgXnu0NLQ0BCTk5Ns3rwZpRRdXV0cPXq05tp9xzvewbZt22ql0w3DYPfu3TQ0NFAsFqmrq+PHP/4x9fX1tLW1AdDT08MjjzxyWvsOHTrEoUOHam244YYbTtsnrU4XjVPEtDDLrW62VQ5RxcEUOkk9TL0ZJawHWBHq4oQ7QVHO9sZMujkeyu+iIGeGpBQzXp9TFFWFjFfAVS+emWURYKO5mv1eP4e8AaKEWK0vYKd3hAJlRpliQqX9DK8LhJnekV2rSaIrA3jhWLBTnDhxghMnZsd3pdNp0umZIHrXdTl48OA5b68P9Pf3o5Sivr6e1tZW/vM//5NQKMT73/9+fvSjH1EsFgmHw4TDYWKxGIFAgD179jA6OorrupTLZXp6enjiiSc4fHgmISGVSs3SMCEEP//5z1m7di27d+8mFArx9NNP4zgOa9euZf/+/SxevJhf/OIXp7XPdd1ah08Igeu6CAQBTFzc2m+tRUvRqTez1+0jhMW0yrHAbGNxoJMJL8OW8kHG3DQRLcjSYCd99hg5OXuqAAUcqgxjK4eCrFCVLnEjRIMZZ8ienuXVEQgCQqfZTDHt5imc1EAbtxYfeQoTnaCwuNxYyn6vnyF59oHZPq8tc0XDzkvQ8o9//GM6Ozu55ZZbOHDgAI8++ijhcJiPfexjxGIxdF3nDW94A42NjWzYsIH58+dz9OhR7rnnHt75zndy7bXXvuR4oOu6/PM//zNXX301N910E4899hh9fX3cfffdNDQ08Ja3vIVqtcr999//G99PnYixwGhDOzmCedgZZNCdQCJ5qnyAcS/D4mAHIc3igD3IMXuU5aFOolqQhB7GEDNBjS4e41629qMBiGhB2gIpdDQ0IShpZarqxYs62Tgc9gbIyplgcA9J/jkZGArlGzsXEBoSQ9m1Rcfxg88vcB566CEsy2LlypV87Wtf4+jRo3iex+DgIK7romkaK1asYNu2bSSTSVpbW8lms3zhC1+gq6uL9vZ2duzY8aLXUEpx1113US6XWbp0KV/72tdIp9Ns3ryZJ554gtWrV3PvvfeyZ8+es2qzJUw26mtoE88GRnsoHOXSrNVxhbWsNsVDVdm0minq9RgekkYjQVizGHFm4nsMoZPUn02SSHsFjlRHWBppIyMLSCTXJpfRaMZYFekiroeoN6L0WI3cmrqUSyMLaDPrWBrsYE1oHgExu+8tgFVmL6vNXo56w2Tkmev06Gg0izoMzlwWwOe1Ya5o2Hnx8IyOjjIxMYGmabiui1KKarXKZz/72Vqv5IknnmDLli0opWr73HfffRjGTJOfn0qezWZxnNmGwIEDBzhy5Eith6OUYnp6mu9+97tomobneedkJteEFqVNb+C4O4bEw0XyVOUAAsH8QAvzrWaSZphhZ5q2QIpd5eNcEVpEUDdYFG5jT/EEEskzheM4z/PezA82cUV8IdvyR5FKsTDYxriTZdo7XQCa9SRBLUC/M84+r7+2vUyVY3KI1cYCdrlHqGATJkiRsu8mvgAQSj6vd2Tyymvb+rwWuK7LM888M2ubbdv8+te/rq1v2bLltDTcycnJmlf6+ezcuZNMJjNrW7lc5oknnpi1TUr5ksbSmXCUy355nGn1bDLIhMwwITMY6EzLHFUcDjlDHHWGWR7sJqgFZmJ5jBjjXoZxd2bYfHGojdWRbg6VhxFC0GBGOVGdAgGO8jheHWdtvJs31a0gpAVwlcdl8V42Zw6ys9TPsD1NRTqsjyxgebiTSTfHoPOsJ0gBx9wRNCGYks+210AnJALkTxZyjRBirbaIbd5+Jjnz5JY+rz5zRcPOW6XlMwUzPdeIea5L9oX2eS6PPPLIGYOjzrT/SwVSnQntRezV494oJ7yxWTUsFDM9qnXhXhSKKXdmFvXN+f20WkkKqkzEsCh4ZabdAvNCjRhCn2XwaAgOV0ZQKJqsOD1WIxm3yMpYJw9n9mMIjUXBNgbtKYpelRYjRUwPc8KZQKHQELUfoKr9Bw0iySXGEh51dpxWLdrntUd7njvYUL9ZoTqfucnZ1Bv7TfCQDKkzZ7m4eGTVTIzNjAdYsbvST0wLsTjYji40Nhf24SERCHqCDYzYM7OnB4VJxi0R0HQsQ0cAbcEklikIKovjlUkSVpCn8kfor07SZiUxNA1XemRVgSIldE1gCA1XSXoCTQQ1k4OV4dM8+R16I8vNbn5deRobhzxFHvV2UDpZtf6FECf/709d8eowVzTsopla4myKB/4myBd5URTqjCF1rvKY8nIMOVN4SEJagHE3i4dHQNOxlctxOcHh8iiD1SlWxjsouBWihsXe/DBX1S2gKl0azRhBEeBwZYTGQBTbdbA0g3arjmvqlvBo+iCdwXp2F06wt3QChSKqBbkyvJQtpYPkZImgbrLTO0wFB0e5bHcPnFVGl8+rj0BiqGrNXNZfYsjSx+fVJiQCdJtNrAh3k/EK6JpGyohStKvoQvBIdj+O9E4Or4taB2t/cZi4EeSahkWEdYvxao7uSIqI3spINc2oneHK1AIKXoVNUwdwlIuNzY0NK/nJ+Ham3SIrw90UZYUBe5KloQ4OlYfxkBRllRFv6mQyxsw7ouCsUtkX0oWG4CD9vsnzKjBXNOyiMXguNKyTWQ4P5fecTD1XNR/RvFAjq2Nd/GJqB8PVDABhI8DSaAv95SnaQkkOFkaRShEzLQar0xwpjNMcjLMi3saoneGW5tUIJRi0pxiyp7E0k4Ks1GJzPCVJe3kc5RHWglwfWcPT5SMccUZmYoVOVocWfq/nvDNTw8KpZTjMiIX/nficHxr1BGuD84loQQbtScqqStQIMs9qYsLJck1qKYPVafYVh04eMfNblSgq0sGxPX41voeNjQsYs7M0aTH2FYZptuJ0hlMYAYWqSjbUzSegGYxWM4QCBlJIXCV5NL+XinQxhU6zmUQqRW+whZ+lt2IInYIqndXboaMRJUyOIlVsNDT/rXqVmCsa5hs8LxMDnUuMxRzzhk/LKjiFANaG5hPRLB4s7J7xuOhBBJAyIuwvDpN2iyyINCGRjFfzLI42U1Y2E06OVXVtdEVSJMMWzYE4k06ea1sX8l8j+/nh8BYcNZNlkXMq7C8Mo4DDlWHagklCjslINUtZ2WwrHyGqWVyXWEFQM8gUCzQZCXoDrewq92EIg3WBhTxdPVxzZ/u89pwK+JO1lE4HceFphc9FjI6GgU4Vh7Cw0NF5oLiTLquB1eF5PJTbzbibRaGoSgdXvnBIgETSHk3QFI5SwaEnniLtlfiv8b18qPdSwprFuJ1je3YAHY2EGeSKSA/vC13KM+kTPD51FAVUFPwi8zSWMBl1ZobP3hRfzZiTYWvxCDCjx81Gkkkvh61mz99ULxJcqi/lYfcZ+mvVhXxeDeaKhvkGzysgQoiEiDKhMliYKBQ2LhqClVYPk16OfZUBdKGhC42YFmRDYjFSSRqtGL+c3kmdFWZ9qoeViXYemTrEvvwIx0uTLIo301ecZLJa4MpgDw+OHiBkmMQDFgqFrsGtHat5cuI4hwrjtRe8M1TH9Y1LGa1meWjyIBXPIedWaDDjtAdT7C8NMe0VaDKSdJj1hDSLrcVDpPQonWYjWds3eM4XQkkM+dyAvwvTHexzcWJhssTsIqXF2FzdDQICwkCiiGlhlFLE9BBRw2LITvNw5sCLnk8BRddGaZKqslFCMlbJEtRNSl6V/uIUmqFY39SOpRlsGj7K9nQ/7aEkJc8moBl0R+pImRGOF6eYtosMO2mSRph6K8Kk92xwckwLcXlwMRXlcNwZY5/9bIrzlMqy2d3lV29+DZgrGuYbPC8TiaSgykRFCA3BWnMhLh7bnIMIBJYwMYVBSAtgK5fuQCNXxBZSUhW2FPpQJbi+fikxM4gmBJoQXNXQy89GdhExTVbUtfD/jm2jIh10QxG1AoR0k9FqllV1rVzS2EFIN+mpplhS18DuqRGO5dMcLIxS9mxual/GO9rXcKw4yabxg6ApDFPREoxj5nVipsUkGQ5VR9E0wQTpWbV+nktCRKgqZ9aUGj7nHg0PU1We7R3JKheiO9jn4kMgWGn0khRR9jp9eHiMummelAeoSpuoFkQpxaJwKw1mjPszuyl4ldPOEdB0FieaGC3nGa/k2Z0ZZkGyjsV19Zg6eEIyWS3yH/27+eDitUQMC0d6WLrBQCnNoxNH0IUgZJhc3tTFing7RduhO1yHqzweGj/MjU3LGShNcKj87HREGVngV8WnaTXqKMrZCRgekixnTmef1XZMqr7G/UbMFQ3zDZ6XiUTxjHcIhUKiOOQOIFGYGMS1MNsqRwDFtdGVZL0ieysDbC0c4Q3JJayOdvNo9gDHy5OsCrQjNIWHy9HCOJPVAgE9TtGr4CiP7liK+lCIxal6LFPn1yeOcFlzJ3Vhi8OZKarK4fKGdrriCU4UMkxXyhxJTyFMyXglx4JEPXXBNYyXCjjCoTEc4qamlTgOdAXryckieccibgRxlEujESfjlXDUTPC3hsblgWWMelPsc/tPBmb7VZlfDTTlzeodGRdo78jn4sNAIy1zHPOGa+nqVeUw5mUAeKy4n3o9RtSweKbQx6A9U4dHR9QSNZbEm1mZbMXQNZpDMUxD8MzkEJquCAcNDqUnmJ9K0JYI05fJ8MDgEfLOzPDHW+cv4pqubhLhAHsmxlld38rahnb+4+heBgoZbm5fTk+8jrWynTorzJHSGNPujBGTNMKsiXXzZPYIR5xhlgY7SJoR9lcG0RBEtRB5WWbmD++ZYxUbSLBGLGaz2kH5JTK9fF6YuaJh56Xw4FzHw6sFB0+rPFlVpElPcnVwBUERQKJ4rLifPZUTJI0wE26OjFvE1HRSZpjucD178sNsS/dTkQ4rkq20huJUpcvm8WNc0drF8vomKtKhLB064jE+sGwly5oaEKZkpJqhOR7CNRwMQ9AciaIUaJpgyi6wfXqAh8YOM1Utsis3xJPZI2g6xAMWeZHnmUIfK2KdGLrgv3I70ITGG+OraDWTtXvUEFiawQKznRus9aww5gEQI0yDSJ7hqfi8UsTJsuyGnFkuVHewz8VHnUiwVJ/HPL2Vpue810ktQkQEKckqEkmLkcJRLgk9RMII8Y7WdXQEZ6bESNslhstZQobO0lQjHZEEy+qaWN7QiBSSZyaHCZoGrZEYsUCAQ5lJ0naJ7roo7akQjXGLK9raWdrQwOWdbcQjOr11Sda3ttGSDCE1jyubu0hEDK5qmceq1EylfE1oWJpZKxliajoBzUBD0KQnuCm6nqQWYbHZxeXWkjOWFslSZJ86NmsSaJ+Xz1zRMN/D8zIJEmCdtpj98jhp8rSIOtr0Bg64/TxS2U355MzCVeUQ0gJcHVvKmJMB4PHsYQpemYOFUQ4URihLhy3pAO/oWsU1rb3Uh0MITTFdLTFVLbJtYpBnJoZpT0aIWCZTdp5cxeaSjhYe6x+guz7KluFBjkxnsHSNVa1NdDaEaChYtIRjOJ5LRzzGFQ2d7Ej3k6tUaQkleHL6IP2VSdJOCUd5SCXZXekj6xXR0ZAnU+ifsY+Q1KJERBB1MgKtS28mJWI84eZ8j885QiAxpDNr/FtcgO5gn4uPCZVhs7uTxXoXIWGBmunsXBJcSMYrsL16BEd5VJTD1YklSCQPZ/cjEAT1marMo5Uc9cEwTeEok9Ui9w8dYnF9A2PlIk+PDXMsk8Y0BGvbmuioi6L0Zq7sbqUlEkUgcKVLOApvWNCKdB2EoWEYgjd1zCOgGeweHycWbkDpgvFSjornoAuBocO2whFKcmY4anepn5AW4Ir4QiwCBE2D5kCCrFPAeYGZ2m0c0uSpI+YXLvwNmCsa5hs8LxMPSYFnJ+WEmTo8S81uPCVPlkCf+aIr0uaZYh9viC9lU24P006e+eEmGgIxAtoEQsDCWCNjlRzrmzrIuxWeHh9kf3qCda0tdCSjDBRD1EUDBEyNvKt4ZKCflliYeDhAXdzkjYs6eKPqxHElQ9ki0vBIhAMEAwJZFbyhs4sH+o5xIp/Bli5NsTBLko1cVj+PzWPHOJAbI6gHuDQ5j7Wqi8lqkadzx7k+sYqHc3vZ4/SxOjCflB5FdzQOeifQ0X1j5xwiOJnSefKZavLVrSnl43MKiSSjCmx199f+WEkUR+0RFgbasIRJ2ivgKQkKWqwUTWaCn4zuIGpYtAbjdEfryDtltkwMsDc9ymg5T9QyWdXcSH8+QywYYHVbI0ub6/CUZHlrCqlmJlv2lGQoU2BZax3C0dFCgieODVLBoapXKLmCroYwebvIE0PDHJyeZEGqjne2LCNGmPFykQdHDlNyZ+oBtQQTdARTjFazVESFBjPGqJNmzE2/4J/fLq2ZBVoHD7rb/dpkr5C5omG+wfMycXDZJY/U1kfVNKPuNO1aw2k1bRqNBEusDobtaQpehaAe4LJELxVVJaiZXJrqZk19G6YhMDVBKhjkstZ2EkGL8Wqe3roEa9uakJqHrTyGSzmunN/MsrY6ihWXrF0hGjQZTBf4ya5jLGpOsrAlzvWLO5BK8fTgOMeKE4yUs/TUxUhaIda0NrF12KUuYlEfDtLpJQjqJpszh6g3o6Ag55V5NLePKXdmgtR9dj+6eNbIOTWc99wZ2H1eOZqS6NKu9YjO1h2s6zoLFiwgEonQ19dXm3DvubS2ttLS0kI+n+fYsWPnZCoVn4uP5xdWzcgiOVnCRGdhsJMmMwFCUPAq5LwStzStpi2cwNAFhgn3Du2m7DmETZNuI0l/LsuTw0PUhS1uXDiPh48PUBcNYOgamWqFJ/pGWNic4OoFrZSUgS0cbN0jFgywpCuGjmBH/xjd9TG64wmOjee4bF4Tl3U30xKMky3bM4ZSUz1hy+A/ju0lpJm8uWUJZc+hRUuQ1IMMO/DWunX8V2YXE07ujPc+Iidp0BKIC3L2p7nBXNEw3+A5R5xpJl9HeYy7WXaW+3CVR5uVYmv2GKlAmIXRJhZFm3ClJFMuk5cVFqTqGCkW6ErGWR9vJmAI5pkxtgyOoDTJFfNa0HQQmsIKmBwYn8YuSgKGxmS5TDCjs+nYCbrroyxqTnLlwiYOjmRZ6zXy5gU9CE9nqlDhYGYSlEAi+cCiNeQqNvf278O0wmyfGkTTBJ7m1gKYlYA2K8lQdRqBxsJAG1mvRJtez/bq4VneLp+Xz2lFu86yd3TDDTcghGDfvn187GMf4xvf+Aa53LOi3trayoYNG3jiiSe47LLLCIfD7Nq161W5B5+Li4ws8FTlIBFh0WbUsa1whJ5QIwkjTMYtUTk5fCGEBiiWJ1tpi0WJh+ahhOS+/kPsn5rg1qULmCqXMU3YPTZJplxlbU8da3vq2Xp8nMsW1iOFYufIBGvn1ZGplGipCzKSKdHdHqQ7GWLvwASu8GiIJImbQQzhYeJx8MQEx3NpRotFLm/pIGyYPDRxiICmc7w4jY5GQDMYsdOk3dPLblhiJv4nq4o87u5+0Wr6Pi/OXNEw3+B5FUl7BbaWDhPXw7SZKS5NzJ+ZM8aAKbfA4cIELeEog+U0a5vaQMGuyVFaYxH6choT5QLLW+pY3JokbOlougJNgaFAkyzujLN/MMNYvkRDzGJlb4xkKMjC5iia7jE6WaS5MUBHcxOG4eFUXSwFQpdsGjrG5W3tRCKCbdOjRAIBVtS1ADA/0kBcD/O941vIuxWimsW66HxKnk3Zc+gyGzngDeIqD543Z5fPy0cg0eWzY96acgiFgqxbt45yeSbVtlAocPDgwdoxuq5z66238pd/+ZeMjIxgmiaLFy9m69attX0CgQDvfOc7mZqaQtd1qlU/C8VnNgZ6zWN7Joqqyq/yz+DhUSjODPc0BmIcLIwxXE3TGIxRpUpvrIG8U8U0NKTuccuChUjdwwoIRksFbl42Ywg9dGQQQ9cYny6QrzoM54s8fnSMdMEmU61w+eJ6CDo0NRk0eQE86TFazvHGZe0IqaEpl0JJEo7pXL2oicf7RkjGDBJhk5FciZwscm3bfJa6DVRdSUsgyb70KCcqU1yWnMfh4hhpp4RA1IrDPlDY5evXb8hc0TDf4HmZaAi6aGGaHDnOrlhfq5liYbCN/aUh1sV62JXrZ09hkPd2XQIolqdaCOkGI6Ucw8Ucb5rfw2ODA4yXi7ytaR5Fx2HLiQmuXNiEpoNpzhQgtKXH4YkMwaDGuzZ20FEX5sR4mb50hu6mCDlKpIsCrworupPsGJhCepCMmTTE6umoD2IbVd7Q007JcSiUXRrDEeYlUziepDEUJp+vkPFK3J/ZhanpRAyTrdWDTLg5FCMktDDrrIXstfsZP5nK6vPyEGq2WOjKpbm5ib/+67+uTXK7e/du/uiP/ujZY4Sgo6OjNodctVqlra1t1nnHx8fZu3cvf/qnf0o2m2Xz5s2v0R35zAVCWFyqLWO/7GOCzBn3MTGYF2hm0Jnk0uhCqtLBNASe8jANjWXRVo6Wx8k6FVqtKAOFDAvrUwR0ncOZNMOlHJd1NxM0NYbyJa5e0sxDB4ZRuuSNq5sIhBTvu6aLqnSRSGzHI+tUEAi2H06Tiga4fFUK3fR4eOcIC5oSaEKjNRKhQQtw5fwWWqIRJgoVMk6Z23tXgKuBpxHVgxQrHiPVDIsSjfRGGhmqZEg7JVJGhJ5wA0fKo+hofkzib8hc0TDf4HmZ6Gj0iDZaVD1b2feSL4qBxtHKKP3VcWzlgpjZVvJsHp04zBubF/HU+AmsgMaVrZ1c1tJBtlJlIJdjZVs9lqFzaHKarsYIgYAAQzJRqLCjf4o3rGzkqpUNNNdZCA0ms1UOTaRZ0BbF1m1CIQ2pFF5UsfnoCPsH8nzgDT0s6IwQ1ixCWoBM3uHYRIYlDQ1ErZnpKmzNxtJM3tixgIGD2/GkZFWijVXxLpSrMVTJcO/kM7jKY36ghfZAHbuqz8707M/P9fIQSqErF6GeDfjr7z/BrbfeSqlUAjht3FopxdjYGIHAzKzElmUxPj5TQFIIgVKK6667jieeeIIvf/nLvP/97+dd73oX3/jGN17DO/O5kLFxGVYTL1qJOKgFWBLoYMrNcbwyxqJQGzEjyBO5gwgNFkabOJwfZ1V9K3WBMEpIlAKpFEXbYf/EFIuak8SiOo3xIFpAcf3KVnRLEQpqCE0xliuRjJsEgwa2KxmcKNHaZPGGy5KgIGDoKOlx6boYj++cpC4UoqnBwAzpjE5k2bxzhI0LW7luSRvjmRLPDI9z08JeDM8jpMG8ZJJ1qS6ylSr2pENrKE7erpL1ClyS6GHYmWbYPj125Ez4cYtnZq5omG/wvEwcPA6o4/SKdnT14tlKQRHgjeHVPFM5yog3TZ0eRUlFV7iBrDczl5YmdKKGxdbJftY2tXBpcztGANa3tbCsNUW2UmVFex1Zu3LyWpJf7Rhi94kMiYTOygVxKsrB0CCShOuvbEAX4Ek4Ml6iKWXRFgvS0WYSjkMgIhG6wDA8KnaFeJ3OvHCQI6OTVMuK1voI47kM8xN1NAeCXNHWzrHpLOua2pkqF9iXHWdNopMFkUaSehTHkRxzRsjIIp1GA3V6jCAWh50hpmX+Nfte5jLiZMCfdlIsdOmAkjiOg+OcOfjP8zx+/vOfc8kllyCEwDAMDhw4QDAY5H3vex8//vGPKRQKNDc3E4vF2LdvH7FY7LW8LZ8LHA+PI2rwRffJyxK/LGzHwcOQOiFhERIBOq0G9pUHyNtVyp5DsxXH8RQjxQLHi1VaohEkcM38DroSccZyBaIxAY4iEIKAqXF4NMfi7igtTRaekIxmyySTOgXXxhUmoaCOVOCdnLYnZRnccHUDxwbLHJ7MsrIrSTgiaGs1STRAWEBXNEAk1kTQELiuTcWVGAHYNtXPwkQjK+qbWRHr4Ncj+7E1h6BhsDreRbgYYLiarqW4PxcTHQePFDG6tGb2yj5cvJlis0SYJve67+DNFQ3zCw++AkaZ4gm1G/slilW5yuWoM0JeltDRWB9dQFBYPJjey+pYFz2hRg5mx9iRHmJdXQfK07B0EwOdXMXmu9v38c0n9tA/nUMTMwJlGIK1vSkWdkRoaggwPF1G6RIzBKGwhm4qxjJVtIBk/bo4re0mB0ayhOIal65KEEwoHj08wvb+CVSkihn1CCclZsxBxGz0qEM4AkbERZoOSxsaCZgaynCJhHSmZJbDpVGuaVjM4mgLQ+40T5QO4iEJaxZRLURSi9Jm1AMzdYt69Bb0kz81Pw/idAQeuuegyZlFl2eX4fCzn/2M/v5+Wltb+cY3vkE6ncZxHPbt24fjOGzevJndu3fT29tLOp1m06ZNr/Kd+FxsxESIBUY7661FZNwiP8ts5WB5mCZjZhbz4+UpHOmRrVbYOTXM/UOHOJHPkrJCtIQj7B2douQ6NMWDhE2TeNggZOp4UpIt2UghQVcIUxKNCwwLVi6NUsXhmaMZHNNBD0s8y0VaLlpQsXBhkO5eEy3m0D3fpKHe5Oh0lqHKTMX5vF1lqJJGi7hE44KexjA9jTGakxaXt7VTocKa+nYGq9Mcr46RskKsj82nwTz9j2lUhLjGWk2zlmKlMZ8mLVXL5mogyXp9KRbma/21XHDMFQ07Lx6eWCxGV1cXg4ODtLa2cvz4cSqV2W5VwzCYN28eExMTZDKZV3QdwzBYsGABwWCQ48eP187T3d1NKpViYmKCoaGhV3Tuswlyc5EctJ/tQT2eP4CrPBzlcqQ0zu7CIL3RRpbFWllf38nh3DglVaUzGeOytnb2T01RsB2+9+Qhbr+ql/q6OFJJOppCXG7W0d0S5r4nR7GFw7JFUdAVxZLL43unuOGaemKGgeO4HB0usGJZBASYYcF1G1NoroaOQkoXXdNobbIIaSbTOQcrLpDBKpPZPDumpzANQSgKh8cmeUv3IhxXUXUq/Gx8N2mnVHsSB+0hDjLEcquLNrOeffYJElqExUYnEzJDlDC9ejtb3f3Uk5iZKZ6zcyVfzAil0JRTcwcL9cJBpM/FdV127tw5a5vnebOC/g4cePGJHn1eGfPmzeMjH/kIhw4dIhKJ8P3vf59C4dl5m3RdZ/ny5bzxjW/kvvvumxWs+UJYloXjOLNc/21tbVx55ZU4jsP09DSPPfYYoVCIG264Adu20XWdX/3qV9j2qzMXVKOWpNdsI6uK9ASakUqytzTAgD6FhsYTmZkSHf828DQAt3atoK8wxQ8P7qHoVLmiqxXL0NHFTHapOBnlYZqC9UtSaIYEQ6FpEI3pMwkZQqJZilACAlEFmodyFBVbEQxpaJogbCiU5pAIa8yPmBSmFYZQfPdnR4lFdK7obaYrFmTHyBTLmus5MVlAhB1s6ZFKGsiSQ3o0T314HkXb4dfp/eTcMi1WgsXBNiwCbC8eI4CJLRyKqkJa5RmXmVpW6jjTPObt8ucaZO5o2Hnx8PT09PCVr3yFb33rW2zYsKEWtHQKwzC46qqr+P/+v/+PhQsXntU5ly9fTiKRmLXt1ltvpbu7m2w2yx/8wR8QjUZZvXo1N9xwAxMTE7z3ve9l/vz55+y+XoqCrFBRDh6Kbbk+DhZHGK/mUAp2Tg/zwPARKq5LsWqzZXiYJY11/MEVa1jUmGLvYJrpfJVdx9MYJiztiVOxPXo6QixZGCFXcpjM2sQSGrfeUkcsBcpysTWX696UQE+4DEwXqUgXLSiZrJQRMZexaoEpVcCqd/HCVZJNkErq5JwK89sjvHVNJ++7fD7ROFzS1UgqbtKeCqMFPG7uXsL8WP2se2wJJBn1pnmktHvGoJEZHqruoKgqVHHIqiISSatooFnUn5XHp4V6mqh7db6UCwChJJp0Zy0+FzbRaBSAuro6HnzwwVnGDszEK4yOjrJo0SLi8fhZnfO6666jpaWltq5pGr/927/N7t27ue+++7j55pvp6OjglltuQQjBfffdR0tLCxs3bjx3N/Y8hr0pHqrsYHNlNyY6MS1MUVZJGRGuja/AFDoroh20WSls6bE7PULRrXLbvGWsa2pncV0Dw5kix6ZyTBWqJ2e1mpk0WQNcTwEnM091heN5SF1S16ixeHEIEZC4uodruhgJiQx7OIaLGZOIsIcIehgRyVN7p3ls9xTXXVnHb9/WRUOzxva+CcbtHLZeYUFXmEBEooVtQpYgFbbQdcFPhnbx09GdKCFZHGvhna2XsDLeQcqMoCHotVpAl5RUhd3eMcbUdO3ZzBShLb1qz34uMVc07LwYPFJK/vRP/5RPfepT3HXXXacZPK7r8vDDDzM2NjZreyAQoL29nebm5tPO+XxhCQQCvP3tb2fr1q0cP34cTdNYu3Yt73nPe9i7dy9DQ0McOXKEW2655dW5ybPAVRJbeqyJdxEVIcJagPtPHOHfD+3DFDobu7toiUZ415pe3rK8i4hl0tscZ2S6jKkLjg0VaWwIoGkwPl0FQ5Ir2WgmKEOBJTETilSrBiFFS6/OvoEc2w5kSHWBa7jUt2s0tGtYEbASEhV02Tc9yZH8JFlRIJSUhBMKEXSJxgSxuCAYUXQ0hHCUR955Nk1QQ3BpYh494UaqJ2v4RDSLlaFuYlqIjMqzz+vDQ7JTHWavOnZWI99N1NFI8qIdDhMo9NPE4vUdE3ChI6XkoYceYuvWrWcshKaUYnx8vBaweYolS5Zw0003cfPNN59mCIVCIXRdr61blsWGDRsYGxvDcRwymQyXXnopN9xwAwMDAyilOHbsGG9605tOu76maRiGgWEYmKaJpp291FsEWCA6CWCy3JjHCn0+EsVe+wQ7qkcBOFGd5FhljDojytpYN23WzLxa+7Nj9OXTjJcLrG1sJVd2GUgX2TUwhZRwdCTPdL5KseKCENiORAmF60nQJbsPFth7uECp6jGVt5GWRAQVnq4QQQVhiYhJZFgigx6j+QoV4bJgbYArr42wen2QaqDKiWyeYKPH9Vc0EqhzEUEPIlWam0yOVyeYUBmuXdBOMmoQs0ze1buSSxs7EJrE1qoMyQkSZoiOcJKdlb6XlbYuEKSIYb6OQmTnioadl2/EcRw6OzvxPI9bbrmF733ve7OKDZ2JUCjExz/+cR577DF6enoYGxvjwQcffMH9LcuqpbwppSiXy/T29rJgwYJaLn8mk2Hp0qWnHbt48eLadiEE8+bN4/HHH/8N7viFqXgOT2f66Yqm2NA4n4QVIBkJEAubTJVyBM0wBeWQChrYrsN4qYxuCRCwfEEMApKxqTJmYCYzajLrEKoPgAlKV1iWQAmFEmCGBCsuD+FJMA2BcBRCekjpoXQPQ2hYQY1VKyIY6DPpnY5HbqpKf6bMolQ9ViiAXfYYyRbIiByJiMn4ydFIhWJr7ijLY21EdYuSZ9NsJpkXaKbJSLKnfIIOo4EtlYMvWvvj+ezhyMnzX5yc6h1xqiz7WbqDfc4fExMT6LpOX18fd9xxB3feeSf79+9/0WPa29v55Cc/yac//WmWLFnCb/3Wb/H1r38dIU5Ofmma6LpeM3pM0yQUCqFOFnPzPI9kMkk0Gq1ts22burrTvZ/vf//7+fCHPwzMaNjSpUvPOqU3RIBu0UKZCie8MeTJ36U6GZq7LNBFRLNotZJYmsmD0/u4PNHLmlg3pjlT0PRnfQcI6joKxRt7u4kGTKq25LHDY1y3shnN1Il4golpm5QhiCZ0pFR0dlhUhYMIQDyhkSu6hJMamqFQAcH4mEOi3sCpeESiOuNVm7aUyeL1QbSKQNiKcBiuSEZBCmTRJYAGFZtAVQfXpr5Jo8VIoqSgqz5KKS9oDIZ5eniMXw0d4KqGXtY3dLJ9aoCnC30U5elZbAJBm15PRhYoqtmfW5hcJpazRx1liImzeuZznbmiYefF4BkeHubEiRNUKhVuvPFG3vSmN3HPPfe86DErV67kyiuv5MEHH2RiYoJVq1aRTqdZsGABAFdeeSXNzc1MTU3hOA5PPfVUTRRgpscjpazVBICZcfbn7nOKpUuX8u53v7u2/moNe5lCZ0GwhSPlMX42sotliRZawp0zmQkuZMtVIgGTH+49yI0rOkiEA/zwieO89fJWUALPg7HRCmYUWlsttIAkmADdBFdTaAFQmsITAmWAQoAl0JTC0wSFKY+APmNIBBMKqSTShVxVkgpa7NlTYnpM8oaVDaxNxXFtB81VTNol8prNFfOb6chEGSsWiRtBrmztIqVHqVbgmayBJUwuT/SyPdeHKyVVZSORNBoJJtzsWRs9F31RMKXQ5HNTOi9MsfB5lp6eHk6cOEE+n2dycpINGza8pMGzcOFCmpqauOSSSzBNk71793L11VezatUqANasWcPixYtJp9PYts0vfvELbNtGCIEQgkAgQDabpVwu14ykUCh0xnL8jzzyCMePHwdmDJ5PfepTZ31vGQpsk/u5Ul/OLnmUYTW7inxVOTRrSQIywEBlkikvz4STJ2pY5GQRpcD2PNrDcaKBAPceOErA1GhJhHFdxcHhHJcnZoy0joYwObcyM5O5BvUNBgQ0XCHRTAgHNZQOekBH6pDsDIACK6bjCcG8dRGcgoe0QFoKzfUQLmi2hKpAMxWqqoEmKdoOsXCAttYAOFVU2aA+qJMMG3hVBzOgiIYMhOVysDDCcDVNUyhOXbCHQ4UxKp5TqzAdwGB1oJeiqnDCHaPPHa09nwo2j6ldlHkdFfqcIxp2Xgyej33sY9x///3s3LkTpVQtD980zRdNYcvlchw7doxjx46xa9cuKpUKo6MzP7R4PM7OnTsZGhpCKUU+n2dkZATLshBCEI/HeeSRR1i8eDGRSASAxsbGM5ap/slPfsJPf/pTYMZQ+upXv3rO7n2e0YJE0e+O4SqPPYVBStLGkR57s6MENZOr23rYPznK3uwI1wfmkQyEiBgWhbJDMKCzuC2BXVGYEWhvDpF3bDQBlarEjAjQQTNA6SB1gTIElYqi5Aji9QaOpzB0EEkNZQp0oZCOxPFA8yDZpoHn0tAjGE87FIwyqqRjGhpBU9LeYtFRFwEbFobi3Kbm45Q0liWaGczmeWDiIBVlE9WD7Cj1cbQ6iiF0WswEBUpcG17JhJvlodLul+XpuVgRSqFJD3WBi4XPs6xatYpsNsv4+DiNjY1s2bIFXddZsmQJBw8erA3TnzJWAPr6+piammLLli14nkdvby/79u3jkUceAeAd73gH27ZtY2BgAJjpkG3bto3W1lZKpRL19fVs27aNaDRKT08P27dvp7e394ye7sHBQQYHB2ttuP3221/W/eUo8pS3j/wZYlSOOiOMuWkuCS3gyshS/iv/NFsyR1kT6wGlkQqEuLqhl2UNjbi4HMpMsqyhkaBmkqs46Jogapk4toceFKQiJiNjZSIpiCd1ylVJIAboIAzBticL9K4JE2sM4CoBmoZSoJkaypPopoGjCTQp8QIS3ZYoTaFpCmUpMARaSCNkarglm0JVcaSvQmcqiqnrZPKKjkSMRe1xeurimBgcHstyfWo+u0cnqHoub25chiVMDhTG2Jo9SlAzyYgstpTEtfBpz+j1FtszVzTsvBg8DzzwAKFQiFWrVhGJRHjggQeIRqP8yZ/8CX//939PLpeju7ub1tZWFixYwNGjR9m7dy8HDhyoCU0wGOSpp56qDYVNT08zNjbGyMhI7Tp333031113HQcOHKBYLLJ7925yuRxvectbGB8fp6enhx/+8IentU8pVfP8PPff54JWvY6yshkTaSrKZsLJcWNqNa5wGXEyrEi0kqlUOJCeIOPMuErf3DsPXUl+fXiAhkgQpMCuSoyQQEhFJKij6R7RkI4yJRJvxqNjAJpAaQItZhA1dTwlsF1FoSSplBV1dRbCc5GWQnc9pCsRAYluQ7ABRFQy4ZV4akuBDWtSzGuNQFXhliRKGQghCQUFrZEw+8dHefB4P+ub2+iJ1VEoSsbLBQ4URlgd62ZNtJtncsc56o6wxOqg2U4w5E7Pej5NIoWBflqv8uUSJEAVZ07UxxBINOkhT03KqryLNl7pYuGee+7h0ksvZdmyZTz++OM8+eSTBINB3vnOd/KP//iPFItFNm7cSCwW46qrriKdTnP48GG+853vcNtttzE9Pc2ePXte9Bqe5/HP//zPXHXVVSxevJhf/OIXDAwMcM8993DLLbdwyy23MDY2xkMPPXTO70+hSPPCdbQKqsK28mGazCS2cjGERr0ZpSDLNFox2iJxwobOYCWNANqjMQ6OZciWq4xnqlRK8MxAmo62AN3tQUKGgSkUKIFpaFTKMzEggYBGot4gl1cEGwSYOo7UUZaBNEykMPAcScCa+aNrVksoz57Jhq241Dfo6JZEehIbKJYhEDV4enCSX24bIWKZhEyD37l2EQlLp6TDoaEMJwo52psTrGitZ//UJBVRpT2URC/NeMSTRpRloU6OVsfYXT5+zp//XGOuaNh5MXh27NhBe3s7kUiEL37xixSLRTRN49vf/nYt20EIwT/8wz/gOA5CCCqVCl/+8pfp7u6mWq3Wei+nkFKeZpj8/Oc/p6OjA8uy+MpXvkKpVGL//v1MT0+TSCT413/9V6anZ//BfTUJYhITERzlcV1oNZvLeyhLm8PlUVZGO1kRjiJtxYPTB5myC1Sly48P7uOj69YykMnx5iWddNRHGEvnaGkxwFVM5x2GCwWWLA1hmCfjXOSz9UCVAKlrlIoSETYRQZNqKEChOuMGrloWoNCURPNcypNlwmFBMGATNuHKt8bQPUi1xIlqAll20VDohkAGXUTVpNu0yGaqTE5mWNKW4JL2JnTXZKw0TtjSWZhoYF9hkOFqmoJXYV1kPlLzuCy6iF9kt1N9zsy6nVoTrVoD007uFad7BrHYwCr2cpQxXrvv95UilEJIt1a060LtHfk8y/T0NL/61a9mbXNdl8997nO19U2bNtWMkVPa9Nhjj/HYY4+d8ZwTExOnzRXU399Pf3//rG35fJ4f/OAHv+kt0EkzcSLs5+yCcnU0WrUGxmUaG4eiqtJnzySWtBl1TFdKbC0eoddupCW4HOlqlCoutpT8/MgRdCH4wJoldNVHsJRGd12MxoQGriAWMjk2lCdeFdQ3GgRjGiqgkEoxf2UIF41cEaymIMWiwLMNgmELNB1laVRcD93UqIgg6C5m3MVqcah4Dka+gulKAik1EwRdcrn9A03s3FXgyS05Biar3LPtOG9Y2kwkbOKGyzw9PsxoscD7Fq/gstZ2lKMzkc/hMGPcNYeijHrTRLQAIS3AuXZUt1BPIyn2cqwWQ3UhM1c07LyFkT+//o2Ukr6+vtp6X1/frHWYCdA7fPjwGc+3efPmM6aGnjhx4rR9x8bGTssAey2o4rLXOU5JVZiWecrKxsWj6rkYyiAgDMpulZJr89b2FRia4N9P7OD/7d7L5R1ttAXrkI7HdMbh4HiadLnCqsVxuhujSEeCBcoDTBByxuRRAqo2DJ9wCbWESPWE0M3ATJCgMKgoUEKfCTgLahhWGdetUvQ8DM/BrNrIqktUF+hSIas2qqiTH4NYyEQzQQhwCpJrl7WAbWJ4YFccFjWmsLCYsotEDYv54UZSeozpaolfT+4kpUdOTkBKbQLS/V4/4yr9kkUdXwwbhwMcJ/MiPdQLCqXQpKy5g4WUcA69ij7nj5fjHX788cfPqTf5pfCQuHhn7QO1CLBS72Wb2s+EytS2G+h0mg1oYiYTbKA8zaNTh7kk1UWzFcfSDEKmwRu75pEyIuwbnWaRHkfagsGxCvPbQ2iGzvHjVbRJlze2pFCemNGwk1nrVVthhE3QdaykiRMI42oGrmaihI4TMDE0iQoohBBIt4T0ynhOFdsMYThVzKqNaTnoIYWowhVviBEICg4cKFN1Koy5WRbXpQhGYEFrjKhpcDA7xtK6JgwLtkyNklclGq0ol9Z1c//4PuqNOGWqaAjmBZqZcHPk5LkZzppTsYtzRMMumry5qamp892El0ShGPRmovafO+1Cvz1BOl2kKRAjoltcmVxASo9imrChaT4NEYv5yTierSGFS09dgsKow6PHJoiEdea3RygXbcyQwJEQCAg8b+bHJl3FnqcKLNjYgJ6IIHUdR7fw9CBS6NhaEFl1UShMU8PQQ+iijBlwcT0bz6wiwi5epYJVqCCRmEFFpeyiu5JYOIBlaLSaOoV8kaqnMTEhWdiYJKzrDI9nyTgl2kMpGsIhitUq7eEEq5JtPDh5AA9Ju1lHr9XCE8WDlFSF0smsiFMVTV9oWGoe7dg4DDFOlBBBLCbJIJEMz6HsCKEUmude8OPfPq8uZ0pvfzV5ue9IiQoPOtuwebaMSJAAXWYjnUYTW8oHmW+1MOxMsSM7gKUZLI214ilJPGAxP1FHulzi4SPDDOcLvGlZJ/+xa4imVDtxS6MpGcQL2ghPIG1BvuzimJJ4s0E4rGGHNVyhoXQdpc3omGOEcTUTqZnYSqGhEEpS0UIEvDKmWcXwbHTPxtNzOKEAVrGEHpRojmTVhhBrVkcRZR3h6Gi2ZHkgxrL2JG5V45EDI/zkyAHeu3Ip183r5PhUnv3jUwSCkpKsMs8IYWkGcT3MFdHFbCkexrU9YlqIoqxQUC88T9mLMcoUY0zPiSF5mDsadtEYPHOZsrQpS5sxJ0NvsJmFkWaeSZ8gaBq0huI0BC0qlZkZ0o9n87i6w7yGGLdf1UtrcwDhSWIhE2yPw31lepYEEFVFesqlsSfA6o0xZFhQkSA1A6k0PN3C0SxsM4IX0PEkZCfzBOuSqFAAMzeJ6eQIB4IEnBJEwlT1Inq5DMIj2SUoDrp4YYFeNtAkUJFoYUVDmwaaS8VxsGKS9T2NVKuKiKYzWM4wXMwSDwQJ6yZ5t0pQC9AVbJipZmqXqCqHIXeKZUY3AWHyjHO6V08gCBOcye4A2mminiRpcnNv5mOlTvaIntM78vG5AKk+z/PaZTSxzlxI1XWI62EujSzkmdJRtpePcqgwxnh1pmOXq9h8Z88zONLj8s42VrQ0EJA6bbEoOhpuWRAwNIJWEGkLFALN0WaGStTMlDSG4+CaAZTQAIWHjqtbOHoYV5h4wsDTg6A8DOVQli4Br4TpVdHcKpVJnYZGwfjQKCmzQtQSGEKC5aHlQVTA0xSaUoiAopKG1fNT7B/oJ+MUiQdCPDR0DF3p5GSZrFvmV1O7UEB7oI6KqDDuZtgYXU6rUc+uch/bK0de8bOeK8YOMGc0zDd4LjBMzWCglOapXB9vqF9IQJj88sQh6sJB3tg1n1zBZX5zHLts09OUwBNVNCmoFiUjExUSQQuvAvGohhmDySGbcKuGbrnoAZCuR6kqITTTO/KEiasHqGoRZFs9mUIZK2SRVgahSAo76GB4ZYJOnrBpgGXh2EU03ebAcIlLlmvkqy4nhmwmJx2WtaVobA5xtC8NCk5M21w9rx1V1dl+fJzDxUnaAkmWNnSQDK1l08hhYsIiYgZodBIEVZCCLDPkTpGVRUyhn/aM6ojTLOo4oI7jnRw8P8ogxxmZe8YOpzIcJLImFnNI6Hxe1/S5I4x5aQLCoNNoor8yTtorsDLcRUQLsrVwhN5oA1fW9/LvQ0/j4rEo2UDMsBiayrN7aJr6lMnqeUkihknY1MilXZLNGrGIgbIElZxE1CkIKLSTZUWU0EFoSKEjhYajh6nqM3omhY7mVEFXmKaHUIqglyfQmySjbIr1QaqZUcx8lajhUh/10C2JkXPxdNCCLoaro+swfKDMikURnhwZIqwFGCkUuKKtg3hMZ1VTM1tHh6h6HsN2mnunn8ZRLgeqg0S1IBNe9jx/O68dc0XDfIPnAuNgaZjDjKALQcqIsmWyn+PFaTyZ4tHB45Q9h/b6EJ11cZTj4mmCo4NF0CWBsE5Xm4UKOQhHYpoKJyeR9R66J5G2h6cZhFIWVemilMBROo4WxtbCVEsOMhTFNQ1oq6PqlHGVi0mVnF0l4U4QMjIUIo0ES2larwxgWw6YVRaus1iqgsicopyr0tajEdECxMMmgZCLqiia6wN01vUwOFHGDCratRi3GMvZPjHI3ko/2/LHcaRX69kMyjO73AOYxEQY8Zx3ysWbsynughmB0C7w3pGPz/Nx8MiqIiEC1OtxKrJKSouR9vJYwuSaxDL2lQf49dg+qp7LvFgdYRVCOYKEEeZtS3pJRQy2H5kmHBO0N0axlcAtemiehickVkRDViXSkLiOjdCdmfm3pIuS4BkGSggcLYinmTgiiNSTSKWh2SWwLIJegZCWIyRz6Avi6NkglpNF1xyK2WlCpo3tOMiKJBYGVRAE4xprV0cRVYNN2yfZcXAKoUHWLZNK6KyRLRzLpLGlx2R5xjPdE2pgcaSZrflDDDovP8zCIkA7jZxgdE7p2VzRsHNm8MRiMTo6Ojh69CiJRAIhBOPj4+fq9K8bFAoPhSUsspUqg+UMjcEYN7Us52BhjCfGB1CapDG0gJHpImOVPBPlEjdd2YRuQXqqQlQHQxPgCjrag7imwnU8hF0gl7exwgk0AxAglY6thXD0IF4ygaNZlDFn3MZ2CdOQ6MLDMkqUvToiYpSYVgQUgU4LtzyNZpmoYhXluKigS9m1aQgHESWPpnadgwemmZ9KUdego1V0uvUQfdlxqmVBT6yehBWk2UxwpDRG1injKMmacDf91XGm3AI6GobQa9lco0wxJufO+PZLohTi1JxCnH3vSAhBR0cH4XCYwcFBisXiafsYhkFnZydCCE6cOHHaNC4+z+Jr2NkjEEQJUaSMRFFWNg+Un6bLaCSoBeizxxlzMlwSnU/Zs8naZeZF6rmtYxX96TSpqElDLEh3IkF/LsMvnxni8sUNdNfFuGfrMEuXhLlsXRwnLwhYCmGA7njoEgxZQXoWntRQ5swfW3VyaNvDoKpFcMVMgUIvnERTkmJWYe8bonFNGwFLofJB6oKT1DFJVmikIhUCRgEzUMF1PYTuYDsakQgUbJeFXRGuWFLPN+/tw7IEjlll29QA1/f2ECTAt3Y/jSY0moJRFkabaLOS/Gh0C0VvdtadjvaiXugwFl2ihTE1jUv51fwKzy1zRMPOicETCoX4oz/6I5LJJP39/WzatInbb7+dO+6441yc/qJGAI0iRUGVKfFsgNup2BQBDJUybJ48ypJEM73xhpkJOjNlnhmf4K2rejiWzjA55dKQNNEMHVmAvsEiJ6aKlF2PziUB5q+NEIoJGgOKilehqsKYskLVjM6kWQG2o+EFQzjCxBUWMppEq5SwTA+bGAFZwjOCuJUJih5YiQARY4xYdQLPqhIs5jF0l1S3gJKHYyv0sKS1RyOgOcRDgtx0leZwmLpQkAcODuLlJQcKI8io5J0da9mXHeHJqT7qjRhTbh4dnWY9xXyzlaJX5enqYaZl/uIxdgDU7KwGcZbZDRs3bqS+vp5Dhw7xsY99jH/6p3+aJRiWZfHBD36QHTt2sHLlSg4cOMCTTz75qtzCXMfXsJdHjDAb9FVs9fYxxczQjQJG3ZmkgRY9xbiX4ZHcTPVpS8x4ZbPlKlEtxJbhE4xWs4QDBgvqU3z4kuXUJ02ODk7TP1rmytV1iIpByAJVdUDMzIYeDEhsxyWglchXFKFgASUDuCKAUBKJjouBK0xsLYKjBWfeq4Z6tKu6mRJVojKNaKnDdBJU3WYS1giF0hQi6xKvt1DVEoZuI5SLtKFc8qhrtwhJRVd7EFHSCMUUN63ppJBRbDo0RNg0eWv3EjqCdUznS1SUPK0OTZ0W47LQIh4r7yMvz2zMpMmzWe2YU94dYM5o2DkxeGKxGI899hgPPfQQy5YtY/HixYTDp1ef9DkdDY3VRi/HvBGOyiGCIkCznmLQnWBb/hhV6SJRRESQCCE21PcSDgnypTLL6psYmS6zfWCcN0Xb6SuUaW0zMVyNllSI5maToueQatfAU6iqi6GD5RSRThBXC6Arh4BXpKJFwDDxlEZFBZEigGOEUdF6Kk4RAxvdlAS0EpZeT8RsJOaMIw0HoQvMSgGp6QT1PIbhoJQHmsQrS4yKBgEHU5jE6wSa7WAYggWdYcbTZd7UMw9DmeweG2RvbpSmYIySXkShuCG1mu25Y2S8Ao1acmaOr9OeoaCeBGnyc08oYCYo8zmdPiFnJr+dP38+5fKMMFYqlVmlHDRN413vehd/8zd/w9DQEO9617tYtGgRzzzzTG2fyy+/nK6uLp588km2bdvGsWPHXrN7mmv4GvbyKFBmm7evVvqhTTQQEUEWGZ0c8PpZZnXzcHkneVWi3aynJ9jIilgHmycO0BgOcyKfYayaJxUMcduiJSQiBhWvwuKGOt66TtKRCDM6VqW5xcTJehgohOEREBJRKlP1dBIJi7KsYjl5XBFAQ6LhIVAzhs/J4S11Ml1e6ApXVbFVBENV8TwIBpPYepioDJLLuyyoq+IZFtnxHJM7xlnYG6C+zYSyh3Jsrr48yZHDZapGhaHpKj/e0k+uYvPmefNZ3lLHoYkM9wzuw5USqRRdoTqGKxlcJXFxGXKmahMrvxAvpWE6GgJxQWndXNGwc2LwFItFkskk1113HQ8++CCmabJx48ZzceqLHg/JZmdXLdUzqUVZZnYz7eXYEF7CuJtle/koO3MD1FkR6gMRdoyNMlhJ4wkP2/MouTaWMNk1PM1oscjGVY1UypBqChCJazOTg5YkhnAIBSS4FVy3jNJ0bBHE0ywsWcS2FSocRiFwzDCuFkR6iqrRhBAQVFkqWpygzFM1ouS9RqLmBLJyjLCWRlQFrm6hk2Xw6UlWLLGQuJRCihN9FZojYZLBIJrhoukG85tizEvpPHl4gtX1rbQ4Fp5oYH6kiYlCibpAGE9zWRBumalVhCBphKm4NgVZrtWpCBPkEm0p2+Q+JnnhQEGBoIV68pQuuNLvQvFs70hCe1s73/rWt2pzv+3cuZPf//3fr+2vaRo9PT3Y9kxxxnK5TFdX1yyxWLNmDfF4nPHxcd73vvdRX1/Pww8//Nrd1BzC17CXh0TW3jUNQZdooUSFXe5RRuU0Y+U0tnJYay2k2YyDUEgJQRXgkfEjhI0AhtDRgJFsieGCSzio09UQoW+oyNHxHD2tYZqSQaplgWVpSF2iKYWhPAyniAhqOEaQgNDxnBxFswFd8wioChURBwVKaLiehnQEhCNoSGy3jNAU5WCMiMpQ1uM4epDAFS1M2eMkqqMEF4SR04KsnidWrxHIKWTJJdGgszYWIT1SwUpKGupNMoM2GbfE7swIB6YyNEfD9GUyNFlx3ti4hF+M7abiumyIL2JL/ihhEWCl1c2uSh/OKzBaFtNDhBDb2HdBebrngoadM4PnwQcfJBaLATOVlD//+c+fi1O/Lig/p6LwmJdmU2UHFWXzROkg9sneQM6tcDA3yrxIA4siLezNjFDwqtw6fwmj5TyHh/McnchxS2cH2axHMGby1NPTdPWatHYbiKpA12fM8EChiNQMEALLM5F6AEMPEw6a2JqHoUFV01FCx9MMvHwVLVVPkTBCOVS0BCEvjaFVUcLAMSKIg3tpjRnEog4RwyS+UlCkgJ12iTfpSM3lRF+WiuvSlgrjIRlNV2k0YxBy2DU1yqJkAwsiTeQqNvtzWabdEs8U+lgS7MCWNkfsYS6PLMH1PO4vPsOENzOtSJEKm+VOSi8x5q2jMZ8Ohhi/sAyeU+Pfp7RLKvr7+7n55psplWba+dxJb2cOUaTTaQxj5hUOBAKnVQ0vlUrs27ePiYkJTpw4wcaNG32D5wXwNeyVIRAoFNvkfiSqVhXYVS5RESIuQuwsH2dK5livelkYauNYdYy3NK9kR/YE+wrD3LVvB02RCB9ZtYZS0aNUlnQnw6ztakDzPOKWBbbD/kNVtLjHohVBoiGJWy2hhEZ+tILosAibUzPFCDEwtTKuZiLwUErD83SECKFEAE8PgyZxVRRbxSl5WWwRJGgUKJr1eFqAsJGmY4NBuJzGrhQRqgR4CCQB26TJMNAqFqumEghP44ZLmzk+UiI8bXBFRzt37d6FEhJpOMRMi4hukTTDNASilGyHVrOOvdV+nFcwq/gwE5gYF5SxM1c07JwFLWezWbLZGYs/Ho9z/fXXs3fvXo4dO3ZaBWSfF0ahqKgZAyjjzYxlCgRtZh0nSmkOFkcJ6QHKno0mBP35DPWhEAviDTx6fJCtxyZxcLl6eSNa1WR6CBpTOlUkbkESbtSwgh7SKYNt4FkGuq1TNSIooWPqVSxVxJZhXD0ImglBEyUCCDRAzsy0LmLoqoSLRVhaBBevY8KepKwViJtxtLog+366j4WLDEL1BvVhaGwy8QoKWXCwHZfGZp2A43Lp/AY27x8nEJGcmJognfe4rq2XiuMxWSqza2qIoXKGoDBZEelib2WQtFeY9czOxoBx8djKHtwLLXVdAZKaWAg5U4SuVCrVxOL5eJ7Hr371K1auXEm1WiUUCrF//34sy+Lmm2/ml7/8Jffffz8f+tCHSCaTdHR08PTTT79mtzQX8TXs5bOSXhw89jO7Kv5SvYdevZ0jziDD3ky20s7icfaVBsh6JTZPHSLvzsQsKmCyVOL/7d2DIz06klEub29HVV2UKRAIJqZt6qMWsaSGLHmUbZdgo8DQq8RwcZ00ug1VPYIyNRwZoqJFMZSLY4bRDAuUgdBMhGEhAKlsPBWiRADPMwmnDyFSUfKBRmzjVPwPeIaFIwVhUcEMOMiii2YoVECybFmIbNYlEtEwQpKRSpbGssVbl/RybKxAYyTMm9sWU3QcJvIF2q0UDeEEjidfcQmNC7KC/BzRsFclLV1KWZvV/JT15vPSmBjMTP05+0UICIN1oV72VwY4bo8hlEABnlJsHRvisuYOvIpOnRVmKu0yVsqzuqOe9QvqsZWLk7fZcahEfafG4lgIicSoOkAeKTSU0Ihq4xCEsKNTGDcxF0ZxVBBPBBBWCDWVQW9oR5wcD5cYOGi4ZpiyXkfYmwIjgOaOoywIJT3mv1sj5k7hqSJVUycgquhCoQmPsKkhS4KBvhyP7UqTCFkEQopQSPDve48wUslhKoM2K8X8eB2ucDlWmOSh/B7GnMwrNlpsXIIESBJnjKkLopcklEBIMTMPyMn1s+Gee+5h3bp1rFy5kn/5l39hcnIS0zQpFAooNdPD+uEPf8i6devYunXrLFexz4vja9jZMf0ChT4zssAxhhmVM8ZOTAtxfXgtW8oHSVMkoUe4Or6Ue6eeJu+VmR9pIG9X2NjVTWcyzvB0ifJUlUVdUQwh0WyDeFLD0gReweXEQJmF6wUBUSEZMCh6RTRbp04MkKEdV1h4wiSHwhUBpBZAiZPT2AgLgUIIAyUtpG5SDsaoRg0co4AtQqScQYpGknLGpsmaphpQaIEcnlYm4NnopocmIVav0dKt44TKtLaYfGhjL05Zo1qCR48P8vT0ABuaexidzLG/NMTVdQtJiihHihPI5wX21mtxNAQTMkuDSBIWFifkaz8F0ithrmjYOXmT161bx549e2pjcYVCgaeeeupcnPp1g4bGZdoyxlWaw2pg1mdV5fBgYScV6dAbbGFVuJt709uonEzT7s9lGI0XWN/cRiJkMeXl6U3FUQU4kS5C2KGchygWlDWEUJhFF01KPL2EEgKlaQih4QUD6M0hJt3MTBEvzQBNQ8ajgEIXM/nsmpwZk8cwkWgUhZgJEFQKW4TIRaPUmWFQCURxAEsrgKbw3CqRiIMW0CnbLnpU8Y4bm7A8C1H2iHqKlkSIq+e1s6nvOP1Vlw31vYR0E0d6ZJ0ijvIwhF6bh+vlUkeCJfQwTfY3mrPrnKFAeDzrDj7L26pUKjz++OOztjmOw69//eva+uHDh19w/jmfZ/E17JUxyJnT9ofVJMPeZG3dVR5SKVYEerCrDuN2joNimLK0MYTO6mQ7GadMvRnjmeExNo8c56al8xgbd6iTgnjIws45mJrCC0BnIoxZBCU8ZBQsq4imPFwtQJIRdOkigwJbBEFouCKIiwkqiBIeGiY6JmgCSRBXFZGJVoqqhKuFcG1JVC/gNCwh7B4hqhsU7CAWaYTIokouhapNXdBi3ZoolCEc0pEFgSzqENN4w6JmclkIhhVmUHJtWy+tZoKRQoEpN3daZ2u+0YohdCarWepFnKQWY0COXxCdspdkjmjYOTF4li1bhhCC7du3A6Dres0i8zk7JJJjcvgF41CKskpIBCi7Do/lD9Rq0rRZSdqtJPmKi61c7j9xlJZYhHzFxgxJpstlrl3bxLwVUcyIQhUdNA2y2Sp60iMkCxiuTXWyhF4viKd0XNVAo3kCTSg0XZLHRBkRlHKQGBjCQrOiSBWeGbXXA3gySDljMyabqItEiEQMSgEPNVGFSAdRM41QowxMOKSUQ9KQxFMmAUvDrABVDykFjYkgN6/rYPfxEZpjYVakWtgxMsjxTJa3tq5kqloiX7bJe1UeTR9AMeMBUyisk8XPBtxxBAL5vOn3IoSAmTo+F4yxAyfFQtTm2hMX2Ijb6wFfw149DHRaRQNPlvezzOompoUZcafJuoO0mXUENIMHxg6xINZIlAiVqiIRCLIw0sTh4UkKrkZdNEh9s4lmS0ygLqJTyto8vTXH2qsiUMkSqqsgpEfJquBqJg0VF1cECagKUhkUjSYcgkipoWkaAh1NGGjoaErgaAae1CjpBk4siOsOEyCKmSlSCUVJaKMI5aFcj+zRKRqbBI500PI642NV4lqQQFBSLjgkQiEuW9SAmzfwypLlDY0oRydXrvDryb2sinXwZmsZe/MjDFZm4lZ22EdP1hSCw3IAIUXN2BEIGkmRo0DlOTGfFwxzRMPOicHz4x//mNWrV9PU1ERTUxMf+MAHqKur88XiZTLC5Kz15xep6jSamG+2cn9pZhxzeaiTjlAK3YDxUoGhcpqxUpEr2jpZmKjnJ0f38dbVXYQ9C6dUIV92seIQwWB8sEx9tyBiSZTmkgwKQnKCfM5j5PAo8SuX4QkT/aRhlRMCV0sghY6nBEKE0JWGRCGEidIiiFQ3jhcm7WWoiiCOFsKuN1EqgzQslOPSukqgKjbZE3kiEYmmC2YGfz36BopMjSnWz2ugPRVhekLy1LEhdk6Ms6a+HcsURJVJ2XaZquRRQFCYXBFZQklWGbezLDG7mPJyXG4s46B3olatWQCLRBcagu3qwIUlGifrddXcwGfpDvY5d/ga9uoRwGSB3sl2dz+PlnejUFwZWsb/395/B0l2nffd+OecGzqHyXlmdzbNRixyDiRhEAQtSqT8UpRE23wtWyw56Q/JP6nKJZVVKpecJFZZZsmyaJmilUiKfCWSoAiAAIlEEMDmnGd2cu7cfeM5vz96d3YHuwAWRNpZ3M/Wrdq+93b3ubenn37Oc57n+xgIbNMgEAHb7T6WggrfHj9CfyrPA52bCH1Bu5ll75kJbh1ux9FQDhoUPYfePou+7jjdqQRxZWCHmqDhEYQ1/DkPsz9OPAdt7nmW4kNkxTyG71HRCt9oJdQShMBCIpGIAKSdxndCnDMnENs3U9M+jp3CzLn4QQFXxfCxSOoYBSugqzUkdB2kDsn2aaxGgKE0pmeiVYCQBqGnCUPF3slFbu3sx9aghaIeeuTNFElprdwn/7KmrAqNhVwJlFiY7JAbOK3HmdDX4TLXGrFh74jDY9s2hw8f5pd+6ZfYsWMHf/mXfxlpfrxN0iS4SW7mkDpN5UJC7qg/y0ywREDYbJ4pY0w1ihiG4LbcOiZqBYQQnFoscKawzPlimcePnOf2dR2UPZcdw1msuElQC9nUlyaQAXghYtnDalWEWhJWPAYGu/H9EqG0myXqMo4KJXX0BVFCkBcSAA1smjqnktAwCYWmevgk/kCOMNtNmJCIwMRSDZaKDdKzSwxtiJPeYuCWamjlUyz59OYS9Kwz6Ww1CJTHckFREj77Z+f4mZHNZKw4ZwtzqFAwHZQ5W58jbcR4uGUn441lzrsLFMIqU8ESeZmm3cxyOLgUwdHASX2eTtFCmgTV60jFtCnLLtAX178VV4iWRby7RDbs3eNil/XLdWOOu+P0Wq1spo+T3gTDsW5sy2RjsotcymLRL7NYdVifbWHSLvHcyRm6MymsuCYZt2iRKWKBgakdxk66tPUZJIRm4ViRV191uO9fxFBWgEpIlkerxDZsJkyYBHoZGfq4aAKRQEiBKWyk3czrMRNZYpt2o4WBK30MilTivYRhCmGa2G6ANm16b/VwgzratSCs4FcCUh0KUROkhCKsejSKgieOzDOz6LBQcQgCuLO/l0zc4qXCOZKGRdm/ejf1vExzV2wrLzpHqOgGHj4vqUNXNG+9XlgrNuwdcXg+/vGPU6/XOXz4MHv37uXll1/Gtu134qU/sASElHR1ldfvE+BfKFPXaF6tNTvxPpAb4WR1ljm3TE+8qf3wd5MH8ZWi2gjI6jTaEOStBLWSx+h4iV1bMjgVRULD2KzDhpsF6UQNyzaoGSXcejPt3gwd0BoVMzB0QE0H+LIVT/oYJDBkAlNYCCSGMNEigbluByqjqIkaShhoIbGUQ3xznEoYUBWLJBIhKLD9MmHGR+UCkpaBtkC5AZPnqmzvbuVzD2xifsFj78wyD/YNkzbi1BqdbMl2UXE8OowUi0GZWsO5UBYbUFQVnnMPsaTLq+6pi8eA6MLG4pQef88+yzdFC1CXhYHV9WgqbmwiG/bu8lqRvGVVoeTWmPAX8bSPQtNqpinVq1Bphgs2ZNr51tJxxqsFPrZpmN5cmlm/SCIuiQuLsAGqbBJTBmZJEIqQjrRk200x8kaVpcmA9u46XkUQhElKvkLqkJrhoYXAlW34SqFFGikspDAQQmDE04TaB5knEIo6CjQoW2JqD18m0ImArL9IaDQVnuNWHb8Yog1NLCZZWHRJ2Da5FoGrDXwsqqKGKzykhO0tXdzW0c93x09gC5M5p4ItLDqsDOca8zjaYzyYX0lduHgPtxnrmFDz9NHBgi6wQPG9/SBfjzViw94xHZ7BwUHuueceNmzYwPHjx1leXuYLX/jCVc83DAPbtnFdl1gshuu6qNc0GzMMg0QiQRiGK0qNPwmJRALDMGg0Gis6ALFYDMuy8DxvJUnxesPB44g+u2qfRLLVGGJOLbOky8SwcPB4sXQKjUah6bbzVByPkWw3N7X0c6wyTegLNra0US56PHVqmo5WC+FY+J5PKmmwsTOD0QgJTYUtQlS1BGhaVUg11oapHKTQ1I0skgBH16jTSSgCtNZo4hjCQmuNFCZWWw+hqqBCj0DEqMs8NbMVqUOsTcM0PBulGoTCxNAhHXEDqiFKNp05Q5rctC2NHcA6K4FlCWJ2lkxGgBcyWVxmulGn286RtAx2yD5O12dxPB9LNHVIDSnoFDlmwgIGEhuTBh6vqKPXZ1l6eKnCASUuJf9FvCe8VRvW2trK9u3bGR0dpa+vj0OHDl1hp/r7+xkcHMRxHA4fPozvv/HsXAiBfk3lTiwWY8eOHcTjcc6ePcvs7CxSSkZGRsjn8ywuLnL69OkrnrcWCFGUVFN6oxjU2RLr53BlnFYrxeZ0F1mZ5ExxiaRt0hXLIUNBhiQddhyrrlDKpzUZpxw2sIWJ5ZnYKcFASiALdTqSAfUgTl+fTbVxHiFA6KaWjkBhCBfPzOLLAEkSgxjywk9isxrVRpPAk83zRRhStHoQOiSRyOFJD6TECN1maf3pAn3tBsoIiLcqnILHrTtyZIw4QU1iuwmcBmzvbeWVcwuMVpb52MAIrUaa70+fouaGbE/1MeEs4WiPY/75VffLQJITaRYoEhc2lr6OqgfXiA17R+7Yd7/7XYIgwDAM0uk0/f393Hfffa97/qZNm/jd3/1dzp8/z9zcHF/84hdX1erbts2nP/1pJicnGRwc5OWXX+bkyZNvOIbW1laq1eoqB2bXrl1s27aNxcVFOjo6+PrXv05raysf//jHOXv2LJs2beJv/uZvVrQ3rnckgqxIUhQV2kWWO+ytPO8eoqwv3bujlWlG64vk7DgPdW/i0cw2pAo5s1DhufExpAEPb+vDrYER2AQVhZ0VyIpAeyFhJoByDdmiSGfrSBVg6gClDWKxVnwRR5garQU10yRUCiU8lEgghLyg1QOIGErmcbRJiI2hPMIQcnFB0RTkG1MIOyRmVZAqwA80RhBipkJ0CEbKRNdDtNL0dsToTKSQoaJUbjDrlujP5+mxspQbHqEQ3Jof4vml03wkv5MkcYzA4pQ7xWxYoN/oYKs5xA/c/ddnSFhfGQ6OeG95qzasr6+P3/zN36RSqfA3f/M3V+T6dHR08E/+yT/hj//4j7n77rtZv3493/jGN95wDLfccgujo6Mr4mtCCD796U9z/vx5Tp48ya/8yq/whS98ga1bt7Jt2za++c1v8rnPfY4nn3ySY8eOvf2b8D4yGxT4QeUwtyQ3sCHRyZIq8/dzR3FVgOP4/PWJI/Sk0rh47OhpYwM5hAsThYBXJ5b4xIe7aO82+NGrVXrWmfQmJFaHB9UCfjyB1pJA2lhhnVRQoGj3UDXbKAqNxkCLoLnEJWwMYaMv/qt56LiNKzOIC20rTMvFUi5S+cjQo9ywmHyuQjYOfXkT5ShypkSmTFRDIqsa0wCqAbWGx5G5RabrZRadGoaQLBhVKqHDlFdizFlYEZt9LQ4ePw6OEKJY0IXXlGS8z6wRG/a2HR4hBKlUilKphFKKcrnM0aNHGR0dfd3nmKbJN7/5TSYnJzl48OAVwkS7du2ir6+Pv/iLv2BkZIRf/MVf5Hd+53cIwxBxodHla2c0999/P3v37mVycnLlPf7pP/2n/MEf/AEzMzP8zu/8Dps2beKhhx7izJkzPPfcc/T39/PII4/w9a9//e3ehveEgJDT4SR5MizpMgf9s9T16m68jvJxlE8lcKg6AdPVCmfqs5R9l7Lf4FMjI/gVi5OLZUbWpbFdjWh4IATClYQNhS99UpYLRoBG4geQ9zwa08u0DAiMZB+mqmLFHBpGCy5ZQlMhsGi6ZRedHonGJpAJamYbCInhK4RWVGMdIAQy3ZxxEVaRWqFcheco7BAqNUU6biG1xIgrxmeLdMfyfGjjAE5NcGphnu+MnmIk081IpodOO0tHLM2R8hRFv4FGk5Qx5sMCrvZX2ndcjRYyBIQr+VLvKVqAEpcS/q7TcPCNyk9iw5RS/NZv/RaFQoGZmZkrItR33XUXjuOwtLTEgQMH+O3f/m2+853vYBgGAwMDeJ7H2NjYKjs2ODjI/Pz8isNj2zYf//jH+ZVf+RWKxSK+73Prrbfy6KOP8s1vfpNiscipU6f4+Mc//oYOz0WbeT3TLrO0GzmyMsm4s0Qdh3vSW3CkS8Gv0pFM4qiAFxfP0hJLsqOtCxPF5lab4Z4UKVtDNcB2bWKeSRITVQ2JawcjE2C6DqbfYNnqxMq6SB1gaYeEKrNoDeEaWTyRIxRplKgjiAEC7TrIZBJNs6GyJ5O4RppCrA8tJG5gMvP8KMO7++lraRB6VYTwaZQD4gZYCY32BeNzdSbHfUIfPrqrj11LbdjaZmy+yuZcH5uzHTw+fpzjpdVl/rZo/kRfdIIuFrF0iDwN3FWT3feVNWLD3pEIzwMPPMDY2Bhnz57lZ37mZ/je9753hUT05WitaTQaK6Wfzz///CrZ6V27duH7PlprFhcX2bFjB6lUirvuuotkMkk6neb555/n/PlLIT8p5aovdiwWY9euXVQqFZRSlEolbr75Zu655x5eeeUVAKampvjUpz51hcOTTCZXGgdKKUkkEu/EbXpHSJGgXeQY1TNMhgurjl0sabxI1XMZjLezIdWBj893Z44wX6vTm08y0tKBaIRoGTblzUOFnTCIBTGW6j75vGb8QJ2JUo3ukRrtGzLYsRrFkiIeVnHjOUpBg5quURyv01Bp5OadgEaJGGChdYAQoJEoYeLLBDWzFSVMhABJgAw9SHUxfbLO+l4TmxA7qyjVXEIMYobAjDcF7Af6bVTD45lDc2xOd5BMSNwwoCeVoSOV4AG5kYpfZ8pf5p7MFhI6zga/myfK+5hVr//3KBCMsI4KdY5wdkUu/z3jauHgiPeUt2rDHMdZaYz4q7/6q3zpS19iaWlp5XhfX9+KM9NoNOjt7aWtrY1f/uVf5hvf+Abbt29nx44dfPvb3155zmsdE9u26erqIgxDtNY4jsPQ0BCDg4Mry2PlcpkNGzZcMb5bb72VO+64Y+V1t2/ffoXeyftNO3k0miVKtMs8eZHiqcp+QhQJaZOUNmkrRi30wFDc17aJZbeGDAwqZQ2+T0vaxlSC+qKHSIfctbkDHQ8QFR8ZCIQKEV6AnwhJ6RC/WqawUKRzYBkr4VKzW7BCh5LVTV2WcI08gYihhIUSMWRLHHSAWlxA521CEcPHxhBx6kYOI+HS8ejNpBJFGn4FsxogNOh0SOgqjEDz/ItFXt5TZrkQsKkny10buxACJmZcxmsFFmt1PrlhOyOtHRR9h5n6pdzD21IbMYTghcqJFZtkIBkxh1hQRY6FY+/Tp/ca1ogNe9sOj2ma7Nixg0ajQaVSoVKpcN999/Gtb33rdZ8zMTHB0tISi4uL/PZv/za+7/Piiy+uHLcsa8VYKKWIxWJs3ryZX/zFX+TXfu3XWLduHT/3cz/H//pf/4t0Og1AV1cXPT09zZwSranX6xjGpc7aWmtisdgq1VSlFPF4/IrxfeYzn1lpciaEYHBwkKeffvrt3ah3iHE9x6SeX/H0jQtllRrNbnsD48E886qI1prTlQXQgmHZTj6epCOWYbHsIFviVJVGxxR5ZRO6CjMw8HwPGYMuO8W5QxV6BmxmRmu0xgIsp05agJCaZKlMjX6SYZmy1UE8YVCKmahwDJdUM6lP2CgMhL448xWEwiSQMQIdwzHSWEYDy25g6ICWXT14tWXKi2Xak5DtMDh2vkF2wMQpaVKxpnZGpeFjWtDZYXHubIFtne2UgwYhIZYpsGWM4XQ7Y+482xIDqODNA78azV6aBqWdPP10coSz7203Yn3Zmvd1Gg6+UflJbFihUODpp5+mXq/ziU98gocffpivfvWrK8d931+ZNBmGQRiG7Nixg127dvHkk09Sr9cZGhrirrvuYuvWrQDccccdrF+/nuXlZXzf5wc/+MGqyJFhGPi+TxBcilRalrXq8UXy+Tzr168HmjYsm82+vZv0DmMg2S03U6TCsipzOpikRWQYlF2cDqcohXUaysMJA0IRstSweHlplF25fjAUL0yOY1iaewb6OTo+y66BVggMdKiI+SZuI6QQuLR2S6QFE/M1hm9NkktBKlakUjeIVYFOTTyoYgU1GnYLVdVO2ewgEAk0DZQwUdpAnT0BI8PojIUnk5jax5dxPDNFqrONwHGRBHjxNEoYJNMB0lZQhzvvSbNuKE5pXlOtKA5PLzHQmmKgK0FbdpD/b/8ok40iW1o6yNtJ/vrMAWLSpBZ4nHam6bLy9Nh5pr0C0JygnVPTKwUs1w1rwIa9bYfHMAy6uro4cOAAIyMjdHV1vWlzr0cffZQDBw4wOztLsVhkZGRklcNTKpVWHJlkMsni4iJtbW3EYjEGBgZQSvF3f/d3DA0NrcxuhoeHUUrR399PEAS8/PLL1Ov1lchPIpFgdnaWUqm04ghls9lV7eov8td//dcrMy8pJf/pP/2nt3ub3jE0mvCyn/CNsp8u2corwTEMYZCRCeLCJhABtyY38oPiYY7VpuiLt6CU5sGuTRjaYO/UJNm0yQ6jlbhpIxqKo+fLZNtg3foYLTKJGSruvCkLSlGbbSCyAam8JhFrYNcVdauFMGlgptPEzWWKYQwpAwzloTEIhU0Ygjc2iRjuQqAJhYVGEGLhGSkcM4tUCito0CDGUkXSmtegFet32CjPpzgvSLWY4BrEsxLTAixFf1uSu/v7eH5skv979lWUhi3pbhLS4nBtAkNITtWnsaRBqN74G3hRhPDiLOo9XR3XNJVJr/PZ0Y3KT2LDfumXfoknn3ySgwcPYhjGitORSCRoNBqMjY1x++23I4SgpaWFM2fOsLy8TKFQYP/+/WityeVyFItFjhw5AoDruuzbt29lWT4IAsbHx0kmk1QqFXK5HE899RTr169fcWC6u7tXxBIv5+mnn+aZZ54Bmjbsf/yP//GO3a93AoXihB6jousIBLvkRuLCwpQG58IZFIohq5OtsQHOh3MMxTpYCksoz+CZ4nEWvSqfHt7NkallWtIZTk3WWHAr3L6hnYYF+2eLPHBbG9UFh7YOk63dAlEO8XwXnfDJBT7VmVnyZgU3kScml3FUOzFRQtoNMC1qRitKSNCC8LZNgMZUDaQOCJEomgrOridxjDRCBfhGHB0TeAmXmHBQYchixaetLUZbRrL/QJVjs0ssFB0e3t7H3LLHJ3dtYDCZp1EP6SLFR/o2sj7RwUsLY+xfnmRHaoBukVtxeLbZQ+yw1nHWn2HOK1wfSsxrxIa9bYfHcRz+63/9rytOxsDAAK7rvuFzLh5PJpP09fXxxBNPEIvFePTRR3niiSf40Y9+xOc//3mSySQjIyO88MIL7Nu3j0984hNMT0/jeR6bN2/m5Zdf5uDBg0AzgrNnzx4mJpptGaSUPP3002zdupXDhw+Ty+XYs2cPWmt27tzJgQMH2LJlC9/73veuGN/lDc+EEG+rSuzdpIUMPaKDc+EkLh4vu8fZaPZyX2InR71R/r64j6pqoAJNwavz6e47+PHceabcZdww4KeHtzG/EKBrLoNtKba0t+FJj8qiTyYepzDRwGoNaDNNKnM+RpsiZ0LoNyBUWPEGMb9CKdlH3Cw39S2sNJ5M4MgMvogR+hqrOomlMkihkDpAi6axCJXErQZYlsn8eZfegSw9uzReo4ItfJI5QW0xJJO3INaMtpiBoKVNUtcu/R1J7AB6cinsGYNcLM6B0jhaQ4uVxDQhacR4IL+NpwqHKIfNzzGGhU9w1aS/JUos8R4nsV9Y/14xEtepsbhR+Uls2Le+9S26urq45557APj+979PJpPht37rt/iP//E/8tJLL7F7927uuOMOtm7dyte+9jUOHjzI3r17eeihhygWi7iuy9zc3Mp7NRoNarXaqkalX/nKV3j44Yc5d+4cs7OzHDt2jGKxyE//9E9TqVTo6uriL//yL686xsuj5NdbFZcGJnUzX0UgaOAyrRaZD5dXvpfn/XkWghJV1WDUnecj2Z04vmLRrWFJEwKJDgwOL8wzW6viqYBNnTmGO7IEdQO/JmhNp8DzMQ0TVRUUp1zS3Qq7rcFQdwzHXUYYAmm4WKYmLkokRA1PpTFMF99IEmATaomUGhBIHWCg0LIZ/fE8gR+3Ma0UbuAgVUAYi+ErhV9p0DVsMXXaISts7rsnx3Bfku/9cIla6NLbHidMG6i64vziMnHiWNIkG4tzd8d6pJYMJlp4vnBq5d4FBJwKJjnsja5SYm4RGcq69t5Gpi+yRmzYO5LDc3mU5Otf/zqWZb3B2fDEE09w6623smHDBr785S9z7NgxYrEYyWQSIQQTExN89atf5YEHHsDzPP7iL/6Cer3Of/kv/4Xbb7+dcrnMgQMH3vA9lFL86Z/+KXfeeSf33Xcff/Inf8LCwgJPP/00d999Nx/60Id44YUX2LNnzztxC943yrrKrF5a+ekeDWZZqpdpKA8Xb2V/2ogTBDBeW+au1mE6kynyMkbclBxdqBLXAe25GPUgwK0p+nttZCNg3ytl/sGDbfTlJToWopcVftbDrxeQeR8j5qGlgRkUEHYNnUpTkW3UZJbATuOaScT2flAlpA5ACEzlIi4YDB1PEgpIDLbjGQ1My0GHDqHQoHzsnGD//jI3DeeQSuF7il0bcuw/VWJy0udjI+sY7sjw/yZ24bqaF8cn6U3kOLAwQ3+ihbO1eY7UxhEIuq08haDGA/FdjAWznPavjO69L6yR9e8bmbdqw06ePMni4iKxWIzf//3fx3VdhBD8t//231byBr/4xS/S2trK6dOnV/KB/viP/5jOzk6CIGBxcbWyuu/7VyQ///CHP6S9vR3TNNmzZw+e5zE6Osr/+T//h1QqxZe+9KU138ldozmuxq7Y7+lgJVk3K5KkdIJSWCUmTe5v2YLrwtHyLLe2DlBpTPLIQD9HRheZKtQRSmJU4yA0OhRo5SES0JNJoFXIwqiLyvpkuj1s1yGwLCr1Co1kK1pLfLNOPCjTMLM4Mk2gTXwzCQjkhbbgVlhH6gCZiOHUKlhJE9dM4i8VEYaBkAZmyiB0Q7oGLOxAQkPh6ZCffaSblKnBC5ChQMcDRnryjC/UGa0uMpxtaQrKtg3xwtxZ5rzmJCwubEYSfeypn8HFX0lpsLG4w9zGgeA0s3rpinv5rrNGbNg7XsgfBMFV15Qvx3GcVUtYF/f91V/91crjQ4cOcejQoVXnnD9/flWi8uWcOnWKSqWyal+pVOLJJ59ctc/3fZ577rk3vY61QIEKRXVqVUgzRFFQVSxMbrE2cyaYoqRrzPslHl/eRzGos+zVyJgJNA6tJBhOdeB4Dc4vVMllDTIk0RUD4YTMzPicH/NY15dA4KMJsEuKc6eqmAMenVsySBWSMBOkvGUc0Ykdb9CiJY7KUjNb8DxBYMSxLYXCwND+xYAwhi1pzDewdQiWxLcTWH4dqUKwQ8xYSDUI+Nb3F3j0wTYSWYEMFb3dNmcmaoSWR0zbLJYdcvEEH12/gdHlMkt+lYCQDekOztUX6bVbGUn08d3CPk7443TIPOvMLsaC60Cm/eLs6CfQsGhvbycWi7G4uPi6UQkhxMryScSbcy02DFiVpAzNiMr8/KUqG8dxmJ6eXnVOGIbMzMxc9fWeffZZHGe18q7WmoWFhSvOLZVKa0ZO483opo1WspzgPOp1kj+KqsYr9dNM+oskpE2nmWO0scCA3YYbhPQnc/TE83QmMvxo9jzDXRnMwMateZwYrdDRYdHdY2OkQhCa6gx876tLfOzTeXqHNUHCxDIdXFUh7RepJDvxjCS2apAwKngyju/HCGQCJQwECks3l7dMW2LLOKECpSDWkkF4PsrXCBUgtcJIgmhoKmWfllaTpBBoXyFsReD5iNDCkwGtaZuOXIyGcGlNpDlZX2Sssch9rRs5Xp1hslHg1fppSmFTv2iT1U+f2c6zjYO8FByheqFqy8R473MQ14ANu46Ui94eR48efb+H8L7weuu3EkFCxDAulIgHWrHgVxAIztcLtFs5XiqM05lI0ZlMc6AwSagVwtCk4gafuW0TcSNGfyrHX3xnks98tJ8Nw3GELyEekMHk3P4afR0GZtzDSzgYdox41cPQAa6VxtIuVtggMOJ4oYUWVjOyIyQgsMIG0nOIt8QwghRBWMdAEBoWoQqQZgC25KFHs5RnFbEAYqZm+nyd83MuD9/SRSwGZdXgpelJ7u4eoCuRZX1rjva5JE/OH2Uw0cqH2rbw7dlDTLpLuNqnoV02x3s55rzxj5pAMEwvy5QpUHnDc98WmtcYi2t72u23387w8DBjY2N84hOf4Mtf/vJVl1+3bdvGP//n/5zf+I3fuG6FNiO4YsL2QcFAYr3JT1FVNTjnzXJzbCPrYp08XzhJIaxxZ34DhwqTlFWdOadCfybLRwY2gKFolEOeGJ1ioV7lF9qG8cqaeCgJqoLTx4p05Wz60jGMhkIqj8A2QJXRfogV1PHMJLVkBw0zi5IWSlr4IoYvLEIZa4oXah9LORjaxwwaBFUXkRRoaaCkILAshB+wsNRg6rjH9g0pcm0G+OCVFZ4XgqUxlMGLY1Ocm6mxo6OTfMri+NwcXckMUgpqoUdS2tycHUSHBrvsIZ4uHWYmWKKiaivK/NCs5L3NGOFQeObdtVuXs0Zs2A3j8ESsxsXnBe/wFfv7rDZui2/m1cIpdmb60WiemDrGI70jDKVbmHFLfGf8KI8fmMAwYa5WY7ns8/yeZWq1HFZCM7PssG17gps3ZLCWNaQ9TNfHTNpgVbHqZdx0K4EZo+rHCTJ5/FpA2U2Q7M2jhIGpvKbqqXKRpmhq8ugALS2UaROGPhYCbQiEKWhpN6EqwINEDjo6DVI5TWG5QTxm8Q+2DbBc9CkHDRAGThiQsiw25Ft4dvEktjTIWQmqDYc2M4tlSM74zdm3jUlGJllWlVUOpETSLvJ4OnhvHJ7L1r8N06Sjo2Mll8z3/VWzGykln/nMZ/jCF77A5OQkH/vYx9i4cSOHD6/+zNva2ti0aRPpdHpN6LFEfPCYYoEproxivZa4sOgyWjjQOMuoP49G8+TioZUijqLXIAg1g/E2npk7iU/IQr3Oo1vWMTHp89L0JJv709y3o4N7N3WxZ3SR6ROK/nU2RiZEpEJEI8QwA4KMi5nwMJRPLFbFN+MEMkEoDAIjhhIGIAgDhSE1lnIxCLAyBtpzUEKikWgpUaaBpwVuqJhedulMQ9wwODXusqUvi8Si4Xis70lhWwYyUAg7JJuwWPKqIDSHKhNsSHYwnOjkdHWejBEnKW2Wg+pKtOciPj5zevm9bZC8RmxY5PB8wKiEdRaCEoN2B20yx6g3y+2ZDexdnOBQcQrQ3N25jtt6ejAMTcGvcyQ5z+auNK2hjRGGKMti794ijzzYDpUQ3w+w02D4PkEiQMZDZCHAT6WJGxKvViE0Y2Skhe96zaakQiK0wsRD+AFm6OLXfCzLo9mMFEJhoJVPqDQWmpoT0pKAfItJLmEh6gIjLymVPGZrPoY0SaclY/MVbu7qZnO6i5yZYrJeIkuaTakuvjd/mGl3iRdqPo5uGoQuo4Vd1jBPO/tXGQlFyD594t1vRaHFFevfg4ODfOMb31jJ6Thw4AC//Mu/vPIUKSUbN25clfC6fv36VcbCtm1uueUWXnrpJR555JHrLnE1IuKtUNUOT9b34BNgYnBTbAPzYRF9oVNlVdV5sGUE7StKjkfJb7Ax14bbgN7WPPXKBM8cmGdDWx7TFMSESVLFqBUUOWkiwhAsBUaAqerYnkPMruHHyvhWDCeWIzDjhNImkBZhKDj/7AQ9N3dg55qLSEboYeJzKden6Qt091v095jU50MSSlAthGzZnEQ5imMTBUrLmvs2d7GhI0dQNTk+vcz3z07wwMA6PrftFsq1kCDQnFheYHuuh6cWDlMIVudu5WSKQaOTc8EMJ9TVUz/eNdaIDYscng8AA6KTvMxwJDxLSdWZdJcYtDp5sngQRciu9BAFr07d8bCkwb3tG1ksebRlLTRw31Af+bSNkQqYqpRpjSXwfI0uWwhbMnbSYd1WExnT1IVPuiNECA/hOBjZOLas41sJYqaN53goDUpYaMMAITCUD2HA7LE5EkMWMgVCqWbvG9tAao1XC8m2GPiOwq0HGFpSXPKpljQLSz63DXcgGhZGINncnaM3mePUfJFRp8hdHev4zvgxhlPtDCRaGW0sUvEaSAQpGWc2XKaunVWNWgE2iH5iwuLYVRIq33H0ZcZCC8bPj/OpT31q1exo1elaU61WVyQWDMOgXF7dLPXee+/F9316e3tpaWmhr6/vDdWDIyKudy6ppQviwmaD3U2XnaeOw9PlgxwoTrA908fNmSHG3AU+MbCTsfoik4Uad/cMUlUu2jFpy8exWyVuKeD0eJF7b21BNEKEEXL8XJmGDLj1njQxy0GmQgzbI96oEFg2jXgehEALA71BYss6saCZOmCEHlL5GKG/okGm/UuZSamcgeFCcdJjasFny7o0A30x1ndZIBSuozEsQSphsKEjz0hHK69MztBnt9ASS9KWTOAon3tbN/Kt2QMremzQTOzeZPVRVnVq4dW7sL+rrAEbFjk8HwBe6xO72mc5rDAflNBoni4eWTmWFDFKjstQuoW909M8P3eWfCLGRzdswNM+r85O89O3DnJTdwfC89BCUl5SmI4FHpihwIhrhKGZnqzRuSEglrYwbA/PsJHJZqWWBhACJUwQILSia12cmBUQOiEWasVgGFpjJAQEzQWnUCkySZOjpQrT0x63DrXjhD5HxgrcOdCDNAxemj3Pc6NT2NLgeGKO8XqB78weohhckmLvt9u5LbWRp0oH2GIPUFENDnuXvkweflM19d3WubgiHNxMbl1cXLyi7cpFwjDkmWeeYcuWLVQqFbLZLCdOnMCyLO6//35efPFFTpw4QWtrKz09PaRSqauKbEZErAVkM/tv5bFPwAvOEdpklvVBNyeDCerKJSnihAEMxTpAaMZLZZ6YOcVt3b30Z7I4yqdYCuhOWGQtybxbZXt7B0EJZhsOPT02hRlN3dU4w5BshZjvY9k+YVwg/QDLcXHsBPUG2LWAdLyGJUy0kGghkaGPVCFeLSCsOLSkNIYWEGqEACQkM5Jv/90SA90JUmkTOzAJnZC5ikNPMktbJsbmjhxHl+c4tDDLTSM91BoB/ak8+xemsKW5ytkBmAwXWG5UaOj3IU9vjdiwyOH5ADCp55kML1WPTIdLTIdXli4aSO7NbaHYqFPxZkkZCbSGpXqDuXKdguswtljhO/sm+Kd3jVBcDlnwK2xqb8GuGDjCx4qDURA42kOVNKIYElcaZfsY0kGFLvWGIJaxELK5bKWkiedqzv9wnt4NNvkWAyMlUL7CUAqtobAY0JG3mJ/z6W9N4jmKRkPxUw92Iz1J2DAYL5VJGBa7ujuxbUlrIs5DA+uxMInNGuzM9fPq0jjnqs1y4Hm/yKvV09SVy0SwgKtXz0DG9XtUwaVpJvldtF/XWNL51a9+lbvvvpv77ruP//t//y+zs7PYtk1bWxuGYTAzM8Pi4iKDg4O8+OKLZDKZd+sKIiLeNXKkuVls5qA+fUUunY1FTqRwlY8G9lbPcbQ+yYO5bSRJIn0LXylemB6n3U6x6NXY3tbBumQ7VlxwbqbBbVtaoCLIAaqguWNjG4kMlJc86q4ingMzqxGhxjA9QjSG7WIrgchZuL5i5qxPx/okMVuD1sgwoLbQQPohMmE0m9VcaDGogfY2iw8/mCeXNcAXaBHi1wW2LTgxWyAbs4nZJuemF/BUyJJbI2Ha/GDqLAOxNhKGhSkk21J9TDjLFIIaFiY3x4c57k2wFF6ZcxjDoku0MaUXCN/pCq41YsMihydihRDF3so5bkquY32sg2VVxpYmPfEch+cWsKSBIQTnlyp89dUzzFcbbO7OsbOrA1ULMQyB1JJKw0MJk854krQSiIZCehpthGi/ju9oEtIktExMQ4LWnHmxRkuLSe+AhdAKEYY4FR9bK2xb0N5hInyIJyShUFiWwc27Mkip8ZTCMg36OxIgQ3zDozUZ45du20WKBA03ZDDZQuhJGkGz2uFYeQZH+8wEy81+PmEZ+SYJcSYGIeqdj/hoAaG8TJb92oxFtVrlqaeeWrXP87xVveF83+fv//7v+fu///t3arQREe8pHj4mJltZzx6OrWoCPKuWWHCLKyXYdeXhKJ9pt4CnA0r1Gp4KaIuleKxnJ9+fO85NrX0I16Lg1chZcWhYqEBT8kL2jxW5Y1sLhg8pBCoIsOsCdIghNSqu8EJFLBliJkA3Qgwt6E8rWPbQtiQeEyAFHW0CqQwIFEJpRACEzZ6HppRs3ZpEqmYEWSooOi5ff2GK/+fO9fTnMpyZKZOyLT5/x03EdAx8k7t6+hkrlGkELgaSrele+uKt/GD5GErp15FTbZImyRYxyKIuUn8ThydJHB//Dc9ZxRqxYZHDEwHABqMXhWYxLNImcxS8OhNegbQZ46G2Lcz5RXqTOabcZeqhRzZpUahMsbG1BeUaBJ5GmhaGb5BISEIZIizFuRNVevstEgkT21YoX5MhRNZCTCskNCVaCjatF5w5WkMOC0xTgNZkbRBKoAONDsDQ0NpmIn3wHUU6YaJ8jWVLhKe5bVMbtjZ59dQCbYkMubTJwnKNmIix6FURvklvKodSgsF4GxP1AhvjPSilSegEjdDn8fIrV01SNjG4Q25jUs+/85Gf14p2hdeuYRERcaPTwGWvPs7dYidtOs8MlwQbNVyhN9NvdCB8k7P+FJ1mDolk2atzcHkKqQ2OLizQaqXQwkIrk+l5j2RScHCyQj4bY25SkWsTpDMmUggKFZd0XmLGBDIISdkCVVNoXyNjGkNorIQApQjcEFMLtGxWnqpAYyiNUwlJJSSCC99tTXOyp5oaM16oaMla/PT9PaRMiTQ1g20ZBtN5RGhwZHKR4VQ7vgrJWnFCrQh0SE059CdaSBo2y6rGjxsngGZFW6DVyr1pERkUiuN69E2rtyxM7mAH41xdL+qqrBEbFjk8EUAzdydEYSiDV6qnmQ+LNJRLiOZr03vIWjH6462sT7Uz0VgmUD6WNDg0sUSp6tOVTTLQnqRYcnj+/Ay5rEkubfHMvgV+/h/2MDhogCORSUXgK0QDrJjAtjROGJIw4JbbEghPIS9O4C76HR4szwbk0iblxZDO1hgacDyF11CcPF2mJRZnfUeG0NUMtKfIywQ1z+Pxs2dYrDQT+LoSaaarFdww5CPtW9mdG2ChUSMh4sz5Baa85detyFIoZvQSZV2nkxZcfEq8Qwq3WqBVcwPQ+tpmRxERHxQKVHhW73vTH2sDySarD40mZyTZER9iLihSVQ3azAybkt2YhuRbY8dAaO7rW8e5+QrxuODOngHiSc35UomGo9jQm8DVPkfOl1EyxMHnkXvb0KamFno4oaK9xwRTIR3QFhhSoD1Aa1w3xLYFUgjSMYNqMWRpNsQ2JL3tMYTWKKWRGnxfk4yZbOrLEDoS7SueOz3L7JJDZzLF1rYOTEvQlopzrjTDUqPOllwXDeVSDRxSRoyiX0ehkQjuS2xnLixy1D2PBm4yN6K1JkOKYrCPKq/fLsknYD8ncHDpY+DaPqA1YsMihycCgCPBGCYGD8R20WZk2KtPc1Y1dWqqoUM9dPnm9D58HeIoH6U1PzWwne5UmlcWx8nZcYRrc3xynpeOL7GpK8fWnhgPbe+mL5lFlxQypkGFGJ7gzGyVLVvjSC1ISIEONTighSYIdVNsqxYSj0lihqSz1UYoQbzDRASCmCnRtqBeCTl4qsyOIUFrOsb+k0UWCz67ezvY0NLKbYOd7Du/yKPrNhIXNi9OTnB0aY4fF87xWNdOFlWZg6WjBDq8YkIimivvACg0o3oaieQOuY2Srq4Ifb1trqJhERERsZoab155FKJ4yT1GoENCFMNmnW6jldNqih8sH8cQksF4GzFiuNpn7/QMi16Fm7u7OTgzz7rWLK4jsLXgRycW2DKQYWtPnkRSUKx7SM/i7FiVrm6LzlSc0WNVYjlNT1cMXysMG6ShQQtsDUYzpgNaoOoK7QiMuESEknOjDXIxm/a8wcySQ0KGLC4GbO/PIyVs7s0wlM/iOhCPS5ZrddK2zYf6NiC1getowkBDYBKXFvfmtlAOGpxz5jjmjTNgdnJTbJj97lle9o+h0ZgY1N7A2bnIW57MrREbFjk8EUBzHdknYK93ipiwKKhLSW8mBhkjQcV3cS4k9grgUGGaStDGbK1COBOQknHGl6uMtLXRmY2xqTNPOm0gXYHUCgghkNimhKqJWrYQKYG2QpaKLtmciWVJYgZopWnMhUyVXPxQ09luYxkCU0tySQMRAkqQSZr87Id6yMVjCN/ggZ2duDUwQpPQD6m5Pj+9dSNpGcf34PaOQVJGjCcnT+Eoj3XJNvYWx9BA3kgR6JCqcpAIbrI3EMPiiD9GVTeNhEKxV51YkcDvpBWNYoHi27j7V2pYRERE/GTUdVPXRQCzwTJV1fzuejoADZYw2Zrqpc1OY5ow6xX59vhBPBVyrzPAXX39JGyD83N1nl9eIp0w2T6Qo68rQ3HWoSORRNUUJUcxOeZz0/YMoi6x0PhOSCgV8YTAoqmojASlNdm4RW7ARiJAgdeAeNpkdt5jacnnpg1pxiYaLBRdYkJxZLzIhzb388LJOc7Ol3hgYB1H5uYJXcGmXAcHCjPEiLEj04+vA3wVsD3ZT0La/LhymriwiYlmT7gGb9wM9+2zNmxY5PBErKKoq6vWXgWCYbubO5NbOONN8+P6CfSF/TEVo1O08JJ7nslaiSML8wznWvmHG0b4v8cOELhw78YekthUlUs19AiMgLY2g5H+LIYCVdVgCSxXIx2BQVNNWQhNb6tJNmFx4kSdwBIoQ9DWbRO4YEmBVAJTCmIWGEKA0SyLD6RgermGpSyOzy+TEDbb2mIYhiRpG6zPtrKrtQf/wvr2QKKVGafEnZlN1EOXUWee+aCEp326jVbSIrHi8AArej0CQa9oJ9Th23N41sjsKCJiLaGBk/7kFfuP1iY525hjONHBbbn1tMg0O9MD7K+M0WnmiQUJar7LzvZu2tNxpB0i4yFhPcQI4/z4xAytWZvj00ViMcFdW23cgiIWl/gOeBrsVkG9ocgkLZRQOG5IPGYgDVheDsjETbYMphFasrwUML/oodfB3dvaWSp4uF5Aoe7y1LFJhlvytHUlSNsG29rbEcrg2Pwi/ak8M/Uy3507zLRbZKyxxJF6s0UQwDl/9r292WvAhkUOT8TrIhDsNjdiYvBc7QilsLbiC7WYKe7KbmLCWaQRNruye0qRMeMUqj4bcm3c3N5FqeKRs+MkYnGScYtXRuf46rOzJBOSB27qwLKblVdD3UnwFJ4bUqn5TC057BzJYAQm63qTtLZYzC547D9Q5aYtORxXUS55nDpXY7AjRaLdJvQVUwsOzx5c5O4N3cQsk+H2HBXPQ0goeA3GC1VGst3c07mOb44dYTjRwce7dvGN6b28WD7BUKyDD2V38t3SXo765zntT79utYJGc1if5W1n52lWr39fp8YiIuJGQKNpKI+jtSlGGwusS7RTCz1CrXlm9jQzTol7O4dJGjEqNY9kIDk0Mc9ko8iDG/tYLvncub6LoVyWVEYi6xDgo0KwhAVKUpwJmVrwGByQWJZAKYkQEi0gY9hYWqIUuI6mJWnz0bu6sDAQF6Ik2bjFp+4Y4odH5mhPxbGkwXS5QrURMlYoc3RhgZ5Elu5Ellm3hKeakzdH+QQ6xES++wrxq2/qmrBhkcMT8bpoNMu6TBCETKnFVcdKYZ2nl48y6xUJtMISBoEOObg0w8niAhvzbZSqAe2pFLpuUav7vDAzRaAUj25az4n5AqdPO2TTJhXPR3mCofY02hBkhEltyUHVJLaQ2LaF9DQpYRFHUSko6vVmr5mEaSFCidSSuhviufDQth7aEgkmFurkYzE2trbgBSEz1SpJy0LLkMPFGQSQtC0W3AqGkFSVQ6uZZiEo4amAjEywK7aefc5ZGvrqIeF3Qs9CX6Wk8zoscIiIuO6RCIbpY5EixWvIQ6krj2O1Zq6iAKq+x8vz4wShJtAh6ZjNnV0DLJQdcnYKajHmig1ml1wScYlhSCxlogXMLDfYP77EvTvaSdkW3SlJMjCRCITUSF/jeRpbSkpVn1otpLslQcwWCNWMCI1N16jVQm5Z185y3aMvm6Yrm8TzNfVGwN+dOkXRcelOZvjY0BYMZZKzEvzd+GHarAx35ofZXxpnR3KQlyunWQquLRdHIjAx8d5KKfplrBUbFjk8EW/IRDiPRXMduEe20S1bORtOUdF1Rp1mw7+EtHk4v5N91VGmvGU8FXKuVOCB9s1khUXoKrTQLBV8OrJxNre0s6m9FYyQSujwxLFxvv38LP/iH2wmnrCRBtw02IaqNde+3TDkh/vn2bEhy9ahDIaWtMYESkHcbNCeiaNCQdyw2NQdh1CgfUl/Po3MmOybmKc7niVhmSSwURp6U1mCQNOTzPKdySPszgxiCoNXq2fYnVrPg9ntnKxPk5A27/pc5UJJp768pDMiIuItI5B0ihbq2n1Th6dNZllndnEmmKakagzFOtmS6OWHpSMcKkzzUNsIZ8pzPOOMcrw8S086TcFxmChV+ZMXjnLf4AD5lE02J1nflkH4NrpuYngWtrRoi5kYKAih3giYmmgw3JvGdRTTUz4betMoV2AaEs9TiFAy3J7hW+emGMhniEmTjZ05GkGA7wOimQD9seGNTJYrTFSL2FisT7fTFk9jKklfvIW8kcbUBrrSdOIuOh6X//+19IoONooBfqQO/WROzxqxYfL9HkDE9Ym88DM/IDq5x9iJjYVE0CIzPGTfTFokV871dciYs0BcWjyc30lMWDQCn2LdpVpTlCsK3xH81NA2bmvvZ77g8vjhUby65KWTCxw8v4zUBkFd4lYk9arG9C1M10Y5Eu0Y4Bm8crBEtQDaN3AbmtGpOsfO1fAcEIHE0AaOFzJbcKg7Aa+MzvO942Ms1xzma3UGMlmm62UOLc5Q8z2GMi0cWJ6k3U7TYqX4SNs2bs4MMe4tcLh+npuTw5z2p1eSIN81LpR0EjY3ra5PDYuIiOudkJCX9VGm36T7uo3JXfZWeo127Avz/nLQ7DMYaIXWGk8FtJgpbs2tI2smGC0VOL24TFxaWNJgR0s32jP4231jHBkrkZcpbhvs4sVDS9RKoBsmqmbhVw1Mz2Z+JqS4EGKHNlu68khlIJWB9gXKk5RKIWEgcFzF+bkaGTvGQrXBD05OUa57dCSS/NSWTWxr76AtnmAw00J3KoMhBfe0r6cUNnhy8QgHK+OUwgZJafPR3M3sTqxne2yIjyR3kxYJLEwyIrHqfizrMmf0BMFr+gleM2vEhkURnogrSBLnJrGJI/osC7qEr0J8AqbUIgtekTaZpa4vlYgGOuRwfZxWM02P8tFoAhVyuDhFzk4QM0y0EZIwLPYXJvnE8Fa6Y3mWiwEnZ4s8NLiO1nSM+QWfvnyKvdNzDHYk6G1L8r39cyxUG/TmUmRjkuWiT8aMYSmTdlvyiVszmIZAeQLPV+hQEvqQiNns7uvgldF5bu/rQYQGppT0p7PohGSiVGYo3YIfaBIixlOzx7k9O8zOzABnG/PUlMuz1cPXHBJ+W7xm/ft6TfiLuIRlWaRSKSqVCul0mmq1ShiuXt60bZtcLofv+xSLxZ/4vfL5PJZlUSwWVxowptNpEokEtVrtdXsVfVB5bY+py0kQwyPAI+AF9wiu9nAuRDSWwyrLjeb3XWk4X13mrtwm9i6NU/Bq3Jpfxx2tQ0y7RV5YPMOJpSXWZ1oZSOY5tVBkulRjqlzln966jXhoUSk1iNkGM+UGPW1xbl/XiRQCvxGihMaWJlXHRwpBxfFpTyUQCD556zoq9YB6I+THZ+ZZqDhsaWvD8UO6kxliwmK2XuW74ydotVLc0zHMUKqVdalWjpRm0GhO1WfYlOjB1T7r7E5MDKa9IlIIhoxONph9PO3su6RQjbPKpr9l1ogNe18cnv7+fj7xiU9w8uRJOjs7+fa3v021eumHRQjBLbfcQmdnJ1JKnnvuOSqVK3uDXI5hGCilVrWPb2lp4Z577gFgYmKCw4cPrzQms22bRqPB888/f4Wh+qAToqjqBgEhDdxVXwSPgFBrMiRXhYwFsBxUebHcVPrst9oZqxZYDscBwUCyhY92bafknOVb547zoZ6NHJ8t0AgCRtraiZsmGcugUHY5NllguCXP3LzPiYky7ekEd6/r4ekTk7xwcJneu7KYpiRl2kwV6vi+ojuX4lv7zvMPtg6Ss+MEISxVXDpTSb574hy3dfexLmtT9T1mKjW67CwSSUjI0co0c26ZKbeAqwMCrfhwbidnnVnmghJwcY3bWCVr/07RXP++VOGgr9NwcMQlNm/ezO///u8zPT3N8ePH+cM//MNVdiSRSPCP//E/5tChQ2zZsoWDBw9y4MCBN3zN7u5uisUijnPp+3bXXXcxODjI9PQ0IyMjfOUrX6Gzs5PHHnuMAwcOcNttt/HVr36VpaUre+NFrMbE4D55E3N6mcP6LCVde/1zhcm2xCBTToFT9TkyZoLdmSFOlOfoT+W5vWUdaWJ0WFk+3LOZb44e4mhjibhpUKj4EBp869R5WtMxtna3IhwbwxKEMsDxBOcWqjRCn2NTBfpb0sQMg/ZNaU7OFVnXmmNsrshsqc5HNg1xaGqRasNnMGNxcrlAzkoSas3J5VkGUnl25nvxAsUdretwgoAztXkMYbAp0c1L5dMseGVMYbA9NsTO2Dr2Nc5QUNUrFKrfDmvFhr0vS1otLS3s3LmThx9+mNOnT69ydgCGhoZ47LHH+P73v0+pVOKzn/0s4k36HD344IN0dXWtPJZS8s/+2T9jfHyc73//+3zmM5+hq6uLRx55hPb2dp544gm2bdvGXXfd9a5c41rGxeMwZ66q3SCATbKfPtmxsq9HtnGPvQMTA4D1Zje3JjYyHOtmR2KIhLSZahT5+uQe3DDA8zTnKwX8QIES1KqQ1AmEZ/PUyXEyRhzpWlQrmp5sml+8dSsiMBidr7K9p5VyWTE15zA172Ark7yZRISS2aLDC6dmKdZ8CCT92Qw7Oju4o7+XsWKR00vLLDbq7Jmd4gfTZzi4PE2rneLe9vV0xbMs+VX6Yi18vH03+2vnKAV1huzmdW60erkvsbOprfFOo0EHAh3KC9v1aSwiLiGE4L//9//OF77wBf7wD/9wlZMCcPPNN5PL5Xj55Zf50Y9+xM///M9jmiZCCGzbxrKsK17z7rvvpqPj0vfKsiw++9nP8uyzz/Liiy8yMDDA1q1b+fSnP83p06d55ZVXWFxc5OMf//i7fr03AgEh43qWVpF90++xrwOeLh/ixfIJXBUgteQbM3v44eJJvjm1n7OVRXpjLUyWK5yvFAi1pi2WQiAJXRO3LtiU7aBNZHnqyBQzyw2cOiwthcTCOJtaWtnR1smHh4e4Z6iPe9f3MrFY58nDU8wVHG7r72Z3bxeTy1Vu6u5kpKONlydm8INm9LwtnqQ7meFn1u0AAfNuhZPleW7OD2IJg0rY4DvL+5j2lvEIqGuX8/4c5/xZGngsqfLrXruBZKfYQDv5a7+5a8SGvS8Oj1KKr3zlK3z1q19lYmLiiuN33XUXtVoN3/c5e/Ys9957L4lEgv7+fh566CHuv//+K9rE53K5VUYkFovx4IMPMj4+jud5lMtlbr/9dj7+8Y9z7tw5lFKcPn2aj370o1e8vxACKSVSSgzDeFNn64OEBl4Jj3NSja/sc7VHSdWaTfCQbLT7OOlMc8adYdDuICXj9Fot5ESawUQbj7bdxGixhPIl/6BzG1XPZ65Sp1ILmShVkEhOzRdJ6QQ7WrrQjknYMPB8xVLZo1pXpGSc+YLH44cmsLE4OVVmd1cnQ5k8swWHhqOoNkIqjYCZUo0whKRps392hru7BnmkfzOdiRTnKksESvNo5w5uyQ/yYvE09dBjfaKDdivLhng3BpKCqjLqzxAQvvNJzBdl2UPQIZfEu66BdDpNS0sLpnn1YG08Hqetre2K70vE20MpRSKRoLW1lUceeeQKB2bbtm2EYYjWmkKhwJYtW0in0/zsz/4sjzzyCD//8z/PyMjIG75HLBZjZGSERqOB1ppKpcLOnTu59dZbKZebP1izs7PcfvvtVzw3n88zPDzM8PAwGzZsIJvNvnMXv4Y5rSf4kTp8TZHahvIIUPRZrdyT3IqvQtqNLH6o2BDrplB3+M7EUX4wdZacmeBnB3azI9vD2GKFx8+cASXZ2trJY+s3g2vh1SWjczWK5ZC/PTjKxHKNhGlhawsRmBydKDJZrPHVvaeZWqqTMmwGMjn2jM/xv/ce4qmzo/zg/BijhRL9qSxuGPDc7Dkmq0U6YmkOliZ5bvEUH2nfRquVwrnQQf4ii2GZmWD5iusUCIzLXAGBwBLmqn1vyhqxYe/LklaxWCSbzTI1NcWv/uqv8r//9/9mdHR05Xg+n19ZmvI8j5aWFnp6evi3//bf8ru/+7uMjIzw2c9+li996Usrz5Fy9YdjmiaZTAatNVprfN+no6OD1tZWlGqu8TYaDbq7u68Y3yc+8Ql+9md/duXxnXfeyYsvvviO3oO1jP8aY7GsKywHl5Ycf1A/QIhCoXmucpTdsWGSlk1FNWgoF0vZDMXbGUq2Ugxq/GDuJMOZNu7uWE93PMutuSG+cf4Q05UKu7t60J6B70m0hu+fHOfZM1P8wy0bUEoQx0b5gp54lnVpgzPLBTa3tiFDydhykZyZ4JbOXg7MztEZz/DRoc2cXl4mZcap+z5HSzMcK81ye34dB0uTGBiUwzqn6jPMeiUETQNwS3wDY/48INhhraeqGkyHi7g/YRnnKi6Idml14W84bL7rm7Fjxw527drFzMwMg4OD/NVf/RWed6nXUHt7O/fffz+u63LPPffwR3/0R0xNTb398UYwNjbG1NQUlUqF3/qt36JarfLMM8+sHJdSrtgwrTWmabJ161Y++clP8pu/+Zv09fXxMz/zM3zlK1+hra0NgA0bNlCtVsnn8yilmJubu8KuWZZ1xT7btq8Y3yc/+Uk+//nPrzxet24d3/ve996x61+rKPQ1VyHFsdlureN8MMfztaMkZZxHMrv5YfUw816FY7Up1sU7uDO/gTONWfYtTbIr30dLLMmOfC/LfpWvnThKXzrLHV0DHJtZoBEGpIhzZ+cg3z91jvlGjc/sHCFmGahQIBH4gabhKPAVi40at/f0oUJBi5WiK5lmONtKGGju7h7iu+dPcL5SJGfFqQYOCWmxLtnOoldludT8TW0zm7+D2+ODnHPnmApWL38OyE4GZCc/Do4S0mw4ul+dQqMZvtYozxqxYe+Lw5NOpzl58iTLy8sUCgUefPDBVQ6P4zgkk80qIMuycByHLVu2MDg4yK5du4jH4ywsLPDQQw+xe/duAHbv3s3OnTspFot4nse3v/1tguDSD7NlWdRqNRzHWYnYxOPxK5bTAE6ePMnf/u3fAs1oT2dn57t0J9YGXbTSTTtHOPO6CYEGcuWYT4iJQatIU9cOAYpXq2eoaYf1sU72VM5xuD7Gj0uS/lgr7WaWE+V5ZhsVaoHL4eVZ6r7PntkZji4usLGljZwdIwgVWoMbhCyUm697R08vtoqhCNg3OcuTZ8+xs7OTj2/cxLpkG7ZhcHxxgd5klqfGzzBaKpC3k2TMOJvSXeh2wbcnD/Od2UNsSHZyZ24jk84ys15p5XoEguPeBKXwUnJom8wyYg7yrHuQ+jX0+HkjtAYVSvQFR1yrZoQxlUqt/K2GYbhq2URKyS/+4i/yP//n/2R8fJz/8B/+Axs2bOD48eMr59x5551IKfnud7/Lli1beOyxx/iTP/mTtzXWiCaf+cxnePnllzl69CjVapWhoaFVxwuFwkpUJZVKMTc3RyqVAiCZTFIsFvnLv/xL0un0yjJWMpmkra2NMAwJw5DZ2VnK5fJKlDmZTHLgwAGWlpZWIkq5XG6V7bzIn//5n/O1r30NaNqwL3zhC+/avbgRyZHGEgadMs90uMhMWMLC4NnqEeb8IpP+EkkRo648XiicRAP3tWxmz9IoC0GRjniaku8wWllmtlbB0Aa3tPaz5NVw64KOeJq+dI7RYpGvHznJvb2D3NbRR7URcGt3LzaSFivByxMz2NpiS0s7XlZRdFwKjQY9ySynigtooODVKXhN27TgVXi1MMpwspNDlQk8FXJnehOFoEZKJLgnuZVnagdZCi9NUKu6waIurvQNBFb9/1pYKzbsfXF4HnnkEZ5++mkKhQK5XI6DBw8ipaSvr4+pqSmOHDnCxz72MaSU9PT0cOjQIc6ePcvi4iIvvPACSikGBwc5d+4cP/zhD4HmjGbPnj0rS2SmaXL48GHa29up1Wq0trbyjW98g9bWVnp6eoBmrtBzzz13xfhOnDjBiRPN5FshBB/5yEfemxtznaLQBFzZXPMi3bSxSQ7wsjqyEipuFVnusrbxtLeXlxrHyYkUH47v5kfOsZVkQaEVbWaGrJGkFrp0mTkGMq3kYjFKSZd5r8KuXB+vLI/RGkvy4d5NVH2PRbfKSLaHAwtTpESSSkXR0AFZM47SmmMLi2zLdzGca6HmBjw7fp6trR0cX16g6Do0/IBbWwYxhGSuUVm5LikkY84CrxTPsTnRQ0xaHKiNodGM+5fKXA/754hjs81aR0LYb6+6AZrh4JBVKqVDQ0M8/vjjK9HIgwcPrpqxSynZunUr9XodrTX1ep2NGzeuMhbPPfccUkpM06SlpYVTp069vXFGrDAxMYHWmra2Nnp6evi7v/s7EokEP/dzP8fXvvY1XnjhBW666SZaW1u56aabeOaZZ9izZw+PPfYYnufhOA7r1q3jpZdeWvlccrncKhsmhOC73/0uu3fv5vDhwyQSCfbu3Yvnedx8880cO3aMkZERvvvd714xPt/3Vyq6hBAr/494cySS24wRJtQcT7v7VmyaT8iE3xRgNTG4L7mNqWCZo7Xz5IwkRbdOGCom6kUm60Xua91ER0uOlwtneWn2PHkzyYLbFMeRDdia6aZzfZbR6hKb8u3sm50mY8XoS2WZq9VQITw6tBmJ5JXZSV6cPs/PDu/kbHmZAwuz1Lwrl+UkkimngKMCAqXIGUlsYXLKmcbG5MHkTtZb3ascnmVdZjm8MqdHvtUlrTVgw94Xh+frX/86N910E8PDw5w+fZrnn3+eZDLJr/zKr/Cf//N/5uDBg2zatIlHH32U9vZ2/uzP/ozz58/zt3/7t3zyk59keXmZc+fOveF7BEHAn/zJn3Dfffexbds2fvSjH3Hu3Dm++tWv8slPfpJHH30U3/d56qmn3qOrXrssUGCRwus6PDUaTOuFVVLmS7rED/391HGJYdEh80wECzi6Ga6MC5tQh5ytz/Nw7iYey93KKW+K8cYSCc8GJZFCYAhJVzzD/R0bOVqZJmlaPNKzlUOFSdwwJAzAkialeo1zxRIa6E5m6E3k2Dc3y+nSIn6oeHV2inrYNPqd8Qz9yTwFp0E98DAQtNsZ+mJ5Dlcm8fWV1QsSgbrsDjh4xITFerOHZb9y1RmRuNZsn5WSzgsPFUxNTvK5z32ORqPZw+u1SbEX911c3pBSXlGeXKlUMAyDj3zkI5w6dYof/OAH1zaeiDflhz/8ITt37mTXrl188YtfZHR0FNu2mZmZIQxDZmZm+NM//VN27drF/Pw8TzzxBJ7n8Xu/93vs3LmTSqXyplVbWmv+/M//nJtvvpnt27fzxS9+kUKhwIsvvkilUuGWW27h8ccfX/UDEfH2USheDo/i4V+R62NikBAxqrrOIWeM2oXJTims87eLewi0QgBxaTPnVgh0c6JYD30enzzOTdl+njh/mmWvyt2d67i9fRBbmvzFsUMIBP/Phl1MFCrM1iv8sDLKSL6D9Zk2hBbsbu1jplrheGGOj/Vuo+g1+M5k01ExhGRLups2K83GZBdPLx6jy86xKzVEzkyyNd7PmDvPc/UjOMpb0VQrqioh6gpRwhxpdshhite6ZL9GbNj74vDMzs6yuLiIEIIgCNBa47ouv/3bv72yDPXXf/3XmKaJ1npl3+OPP45lWav2XaRcLl8xizl+/DinT59e9T7Ly8t8+ctfRkpJGIYr3mfEG/NGAc4KdSp69R9qiKJ8YV+LyLBR9vO8f5AQxVZziI1WL2PBLEqG6BDmwzIHqmPUlMsdmQ20mGnGnSW+P3cChGbJq1Hw6mSseLP0srLIolvlTGmJfziwjeFsG2XXY6+YwpIGKpCsS7Sxf36Gz264nfHqMt+eOAqAUhrXD1l26+zKDtAba2GqXmQk1cuMW2LRr3K8MYUAMjKBJQx2xtcx6s2tivQc8s+yy9xAm8xSVnWGZS/n1exKdZuNeU1Oj9bNMk6tmndZhQLP8xkbG3tdjZUwDHnhhRdYv349y8vL5PN5Tp48iWma3HTTTRw6dAiAf/SP/hFCCM6dO8euXbv48Y9//KbjiXhzfN9n3759q/Z5nscTTzyx8vjcuXNXTMwWFhZW5fpczqFDh67Q66nX61fkD4ZhyP79+9/G6CPejCqNq+7vlx1sM9fxfW8vs2Fh1THvwkQpbyR5JHczz5QPsRhUkAgGYm3UlUtPLMecU6YRBjh+yLligWfmT1L0HD7Wu5WaE/C3E8coeQ2U1izU6jzLGCWvwR3tg0zVG9jCpOK5vLJ4noRh8WD7ZnwVMpToYN4ts+BW+Ejbdg6UxklIG61hON7FnF/ktDtDViabArKJm3iucYhQazaZfRzwz6zkJCoUFibyGkvX14oNe9+UloMgwPf9Vbo5lzsxFxONX+vYXG0fwLPPPsvc3Nw1vU8Yhvi+Hzk77xFzusCz/n5qOJgYDBgdJIhjYHDSneJb5Zf5XmkfngpIiwR7Kuc4Vpvio/mbabcyhFqz7NXRwJZkD1vjA9zfuhkpJBXf5YnJkxxamOXY0jxKa0bLy4xXSqSMBDfl+/nx/Bi9sTwDyRa2ZLp4rGcnbhjyxOwxnl04RdlvMO9VeL5wijP1+ZVxa+CW1DC7k8OUVZ2G9lZdl6t9FAoDiY1Jn2znDmMrvaK9eRz/mtbCtRboUKAubNfSeO/i7L+7u5tHH32Ub3zjG0xNTWGaJjfffPNKhc/999/Pvffey2c/+9lIb+o65+zZs2+qNxbx3pAhyTp6VhTnLzKtFnnJP/q6DYUBasplb+0s1dBBAFkjyYfy2+mxW/jW7EHGGs2k4VeXxvnRwih35zfRF89T9wNaRJY7W9ZjCoNHurfxc4O30h3LEmrNjxfOkxAxPtazA60FS26dhGHTYqVIyBjHytPsLZ0HLUgQJ2+m+NbSHr5VeJVvF/Zw2p0hIWw+ktzNeruLHzvHqegGGkWIusJWJUWchIhd0/1aKzbshlFavpoTFHF9oNE0aDoLDh7H/XHuMLexGJYICLG1yUajD0MKBq0unq7vpxTUeKVymkKwWhys4NcJQ0iRZHdmiL3lUZa9OmXXY2OqizPVBRSa70+fZEeul+3ZPnQIcWJ4KqAWuBS9BoEK0RomnQIlv8GOdD8ddpbR+gKbEz2ca8zhaJ89tTM0O/QI+sx2ClwS7AoIecU/sWIoXggOsUUOvuWEPzSrjMS1algUi8WV5PqLOI6zUr14+PBh/uW//JdvbSwRERFkSdEvupjSC6jLlrU8Ajx9dae0mfOiCbRizitxZ2KEunJwtMczhaPMB6VVy+K+DokJi5yZAASTtSIxYTHRKGAIScV3GK8VmG40xU8Thk098HhlaYzT1XksYTR7/WlJQtoYQlILXE7VZ0iJOBqNQpOQNurCWpOjfY565/FVyIg1yGy4xFH/PHv8k6uupUqdH4b76NX913bD1ogNu2Ecnoi1w5IqsS84xYxqznSSIs6Q2cVe/xQLqrxS9n7eWyApYtjCoqYdNJqx+gI7EoO02Em2J/s5UZ1GozlQnMBR/opBKfoN9hYm0Epwc3aQuUaVHruF0foCR4ozbEp28cnO25hyC/THWjlWm+KppcNkjAR3ZzYTFxbH61Oss7s47U7TbmRZb3cx4S8QXJbjc7lz4+JzWJ37CSscLhmLKPAYEfH+Ms0Cs3ppVVVqlhQxbBYoXPU5t5ibCAhJyjij4TSO8sjIJP1GB49XX8HTqyflbWaG7YlBDhSmmHNKPJAf4XhplnXJDvrybbxaOkcldFZyCj/UNkItdHl26RRDiTYeaN3Ck4tHOFSexMJgONHFxvYe9lXG+PvCQUIdotFsjPXg6YCFoEyX2cJ0sExVNZoyIa/TJ1DTbDeh3qBNx6rz14gNixyeiPecOi6janrl8aIu8bS774qSdwHcHt9CViapa4eD7jm00Fja4rnlEyQMG1+H3J8bIdSKvdVzbE71cq4xR125bIx3sSXex6nqHOcbS9yZG2Z3ZoikEQMNngrJm0n2lEc57yzSUD6uCnh8eR+10CEhY2yI9TDpLTIbFCjVDnNbbAtn/Cmmwiul/LMk2Wau51BwhvpVVKpfD60EKgR1wY9SQSR0GRHxfqK5sidXHx3kRYZlfUmyootWQhSLFFlSZTQaW1h0i1ZiKoaNzX7vLErr5qTJ6uaEO4GLz73pEdqNDDommHaXaSgfgWS8schHWnYw0NHO40v7mfGKALxSHMUSBhsTXcz7JV5cPk23mafiNyeDZauBZZp023mO1icILkR1flw7idZN6ZCb4usZ9xY47U2zqErv3P1aIzYscngi3jIX17XVW126eQOupu+jgWPueWLCottqIdAhvg5Z9mrM+iVqbjPrf191FK017WaG+7Ij7E6t45niEerKZcopkDFS5KXL6eoCOStBWsZZDqos+zWO1CZQWiMQdJo5diWHeKFyHFcH5A2TpysH2WYPkpZxJvxFQhRbrAEWwtIVFRwKjaeDt35XdNNQXCrp5LrsNBwR8UHmNBNILVfZqj7Ziad9FnWRUTUDwJiaZUB2AIKSrqG05sPxmzkdTJKXaR5O3sKLzlGeqxwjLizSRpwH0juYC4rcmtzIk6X9HCxPYEmDJf+STtyCV6HLznFrZj3fLxzBwGBDohuN5jtL+9iWHEApmHSWV5wdYFXV6bO1w9jC4uHkzex3z6xKvI5jExD+ZD221ogNixyeiLfMVoZRKE4w+q7/TS9cmIVMhk39ixaZJkmcGBY1HDIiSS1w8Qjwdcj+yhgBAeWwwbTXYNxZYlOim7yZIiFtXi6d5a7MJvZXzlMMamhgJN5Ht93C0fo45bBBSsbZHetlyO7kxeoxlsMKWZkkIxMc98bZZPWtGqMANsoBarrB/vDkW74nl8LBFyoclLgebUVExAeapiOw2hk4qE5d9bs6r4rUlcsyZVIiQZI4GZI8Vz9Mt9lKKaytVETJQBBqRTlsMBcUqCsPqSUDRidTZoG5oEjWSJI1Eox585ytzZOTKW5KDZGUMcbdRQIdcrI+TZuZoRo2J4IC6DFbqagGVdVMoG5oD1+HnPan6JZttIs8HgHjwRx3WttYVmUOh28s+XI11ooNixyeiLdMkTKa98eBL6gqT9X24WofE4P7EzsY9Wc56U/QUB4/rq4WpvJ1yLH61EqLCIXmicLBVdGpQlgDD5aCCgtBmU4zR4uZ5qXaCeaCInFhY2Jw9kIvrWW3/JprF6RFotlMQze1Ot7KLEm/ZnakomKqiIg1wespz/eKdjbJAX4Q7qOqG/zYO4arfVx8zgerq4kVmlGvWR1aCKtkZYpD9fN0GDkeyuxkzi+SMRIsqTIT3hJZI8VSUOU7y/vosvJUwgYKzWlnhtPMrLyuKQx2JtYx6S2itCYjE7zqnCIg5Kw/wzZziLxMkxAxpsNFTgbj7DCGKcgqk2qet8JasWGRwxPxlpli4c1P+gkZEj2kiHNcj61K/k0So1d2MKZmcLRPTqQRwI+cpiG5xd5EgOKAdwaAQbMDS5ic82dWnLOLrzdsd9Np5nm5fpIQxZxfZM4vrrzXfFDiyfL+FWPWKVMMmp2cD+ZWhMQuR6M5FDbbbmRIcqexjVfDaxeDu2gsLib6Xa8JfxEREdfGhJ5nPiyuFGDMqiubdkIzPeBy25SXaR5M7OJHzjFqgUc9XOS0N02f1YanAtqNLD8oH0ahsYXJJquXo+H4ymttsHsohFUWwzK+DnmueoRAh3SbrRdKz8HCZKe1nnPBDEeDsZUJWh2HSbXwuonMb8RasWHvmw5PRMTV0KirVgYkRYIho3tFzG+z0c9mY4CiqtLQLnNhgYWw2PzSGz30mx20GGmu1sCurtyV7u4XEQj6zDaSF3QnUjLOFqsfA8l0uMTzzhH064S08iLNA+ZuksRw8RjXczh4Vz/5atesIQxXbxEREWuXgJDa64gXXkQAN5ub2GZc6sNWVDUOuufYZPRx3J3g5fpJxv0FXqmfotPIc0t840rLh1ArZoPm0hk0HZ5+q53bE5sZNJv92XwdsjO2Hkd5nPKmVs7rkHlGzEEkYiUardCcUOdZ0m89mXmt2LDI4Ym4rhjXc5zU568o7S7rGvNBERDNiEpwloPBmZXjE+ECU+EiJgbDZi+T/iL7nDNXLRGfDpY56oyvWtayMLg5voEuswWArEgyYHZiCgONZsDo4MHYTVhXCYqaGPgEBCg8Ak6pibfURV2rpoFQl20RERE3NhpYVCUK+lJiskKBhg7ZQlnVqWoHgaBNZgmVZiEorUwIQxSHnTGKqoZEkBBxjjsT5EjTamRok1kMJDFhYwpj5T1cfI4HTRt7ze1v3uxa1ogNi5a0Iq47rhZIMZDkRRpbW9RxXteh8Aj4oXuA4IJyaELE6JcdjIbN/BtxwWG62vOerh3E083XnQwXmWksryxrTYdLVFWDgCsFLjtECxYW4WV5O2+l8Z5mbYSDIyIifnIG6CJNguOMrew7r67sDlBTDqPB7IqyewyLu+xtHPPPcz6Yu2p1bJvMcm98Bwe8M4z5s1RDhw/Hb+YZZz8vOceuOH88nGc8fGt5Om/EWrFhkcMTsSZo4PIjdeh1kwQvx7/M8UiLOOvMbqbUIpY2uMPexmH/LMtXUUt97dr15e9V085Ko8DXckZNMKZmMJBIJEni3CJHWBRvHNK+yMrs6KKxCLkuSzojIiJ+ci4WTrwZc7rAXHCpXNzB42l3H472Xtf+lVWdl5xjLKgSY8EcvbINXwdXtMZ4t1grNixa0opYM1yLs/NalNbs9U5R1w4hioIqX6GfcxEDyU5jmDaRveJYm8jSI9uu+jyfEAeXO43t7Jab8PCZ0vP4+tranVxa/9bNTb3l5hQRERHXOePMcZzRn+i5jvZJ0MwvtDDpFe0YF36+TQxus0aIE1tZ7ppRy/x945VrEhfMkCRN4ica10XWig2LHJ6IG5Z2cuyUG+kXzQQ+j4BDwTmq+srIiwC6RRtdshVbWFcc75XtDMnu152haWBCzTOh52jgckqP411jHo/WEASawG9uYdQWLiLihuStOgEmButEL8Oil/uN3cSxyYs0u83NJImTIIZGM68KVPSlruQafaF58RsjEGw31zNyWeL05aRE/JqiUmvFhkVLWhFrhtfLv7kaFia75RYm1Rxn9dS1nW9uYlotrPT4upxj4diFEtLXf/8xPfO6x94IpTRBcCnRLwyux7lRRETEe42NxXrRy0k9xsvhUVpFlm7RynP+fixM7jZ28Gp4nNPh5Ju+VhybODZFLiVJazQHg6sXd1iYbBcbMDGuOPZa1ooNixyeiDWBRLKTjSxTYoIrE/1ei0/AS+oQLv41iQD6BOwNTlw1+gPN5bR3q/BA66aBUBdUSsPrNOEvIiLi2kgQI0WCRYpv63XqOLygDhDQbFnTRSsBijoOEskRdY4K9Td9HYAh2U23aOPF8NAqm9jAJY59hWBqlhRpkbymSeZasWGRw/MmCNEM55mmiWVdudQR8c5gYJAiQZX6VXV4JPJC5RVYXNvn4BEikFjXuHK7RDOR+Vpf/yJ5kWFIdHNMjaLQSAQ+wbX/vQiBMAzEhXEK+eYzqoiIayWyYe89g/TSJzp5SR+6amXn5ZiYhIRv6FiYF2xSgSoFqhhGU48sFJAnS+my0naAmLCIE6OsL+mNTbHIgigj5WqbaGFyj7GLab3AaTWxsr+Kw37j5LXlTq4RGxY5PG+CEALTNPl3/+7f8bnPfe79Hs47ghACwzAIgutnodVAkiJJldo72pT0cizLwvevXR/nWolhkyROiSoCseLwxGIxkskkpvn6XzOlFFsfuon/3+P/cWWB34rb1MyrV4RFRLxVIhv23mMgsTCvSYDUxLiggvzmds8wmo5EeEHZL0MShaLGanthYxHHpkztmsabJYWHf8V4Y7EYiWTihrFhguuyeOz6YuPGjeTz+ZU/srXOhg0b+Pf//t/z+c9//l1xAK5H0uk0f/RHf8Sv/dqvMT//zulPvBEXZ9YHDx583b+dfD7P+vXrr9jvOA4nT55EXa+CFhFrisiG3Rj8i3/xL/B9ny9/+cvvyfvdaDYsivBcA2fOnHnzk9YQvu9TrVY5ePAgrvvW+6asRXK5HLVajaNHjzI5+eYJfu8VxWKR/fv3v9/DiLjBiWzYjcHMzAye511XNmMt2bCoLD0iIiIiIiLihieK8HwACcOQer2Ofr1umDcgWmtqtdp1FV6NiIj4yfgg2jAA13XxvGtvTByxmiiH5wOIaZq0tLSwsLDwfg/lPUMIQXt7O8vLyzdMHkNExAeVD6INA8hkMmitqVarb35yxBVEDk9ERERERETEDU+UwxMRERERERFxwxM5PBERERERERE3PJHDcwNj2zaDg4P8wi/8AvF4/KrnbNmyhUceeYRHHnmEdDr9Ho/wnUdKSW9vL5/61Kfo6em56jnt7e3ceeed3Hvvve/x6CIiIt4KH0QbBk1tm507d/LYY49d9XgsFmPdunX8wi/8QqSe/RaIHJ4bGNM06ejo4DOf+cxVjUVHRwe/8Au/wIsvvki5XOZzn/vcitDUWkUIQUtLC5/61Kdob2+/6jmZTIY77riD22+//T0eXURExFvhg2jDAFKpFNu3b+fhhx++6nHLsujq6uLnfu7nIofnLRA5PDcw9Xqdw4cP02hcvSHmLbfcglKKWq3G6dOnueuuu9b8DCkMQ44ePUqxWHzdc0ZHR5mYmPjAlbRGRKw1Pog2DGBqaoqzZ8++roxGtVrlyJEjHyjRxXeCyOH5ANPW1rYyG3Jdl5aWFmzbfp9HFREREXFtRDYs4q0QOTwfYFzXXdVJOQiCSJgvIiJizRDZsIi3QuTwfMC4KMAnpeT06dOYpolhGLS3t3P69OkbVtAqmvlFRNwYfJBs2OX5SLZt09LS8j6OZu0TOTw3MMlkkoceeoiWlhY+9KEPkc/nicVi/Pqv/zptbW0cO3aMU6dO8fDDD3P33Xfzla98Zc13HjZNk3vuuYfu7m7uvfde+vv7Afj85z/PyMgIACMjI+zatYvNmzdz8803v5/DjYiIeAM+iDYMYGhoiDvuuIO+vj7uuusupJRs3bqVX/qlX8IwDDKZzMp9+fCHP0wmk3m/h7wmiJSWb2CEECSTSSzLIggC6vU6SikSiQSO46C1xjAM4vE4YRjiOM77PeR3hIvXrJTCcRx83ycej+P7PmEYEovFiMViCCFwXfeGue6IiBuND6oNsyyLeDyOlBLf96nX6xiGgWmauK6LlJJkMrmyjHfxvkS8MZHDExEREREREXHDEy1pRURERERERNzwRA5PRERERERExA1P5PBERERERERE3PBEDk9ERERERETEDU/k8ERERERERETc8EQOT8QbYts2GzZswLbtG6JHTURExAeLyIZFXCRyeCLekJ/6qZ/iX/2rf8W//tf/mjvvvBMhBD09Pfze7/0en/jEJ5BSsnv3bn7913+dnp6eVc/N5/OsW7fu/Rl4REREBJENi1iNjrZoe70tlUrpvr4+/cgjj+h4PK4BHY/H9Ze+9CXd2tqqN2/erO+8806dTqdXPS+TyehPfepT+jd+4zfe92uItmiLtg/uFtmwaLu4RRGeiDeku7uboaEhXnjhhRX58p6eHjzP47HHHuOBBx7g1VdfvaJ/TaVS4fDhw2it349hR0RERACRDYu4ROTwRLwu69at41Of+hR33HEH/+bf/JuVvlQjIyPUajX27NnDgw8+SC6Xe59HGhEREXElkQ2LuBzz/R5AxPXL0tISf/AHf0Bvby9CCCYmJjAMg9tuu41vfetbnDhxgj179vAzP/Mz/Nmf/RlKKXbv3s2pU6doNBrE43FisdhKv5eIiIiI95LIhkVcThThiXhdKpUKYRgyMTHB+Pg4Wmu6urpYXFzEcRyEELz66qvE43FaWloA2LhxI4ODgySTSYaGhpibm2N4ePh9vpKIiIgPIpENi7icqHloxDvK9u3bmZ6eplAovN9DiYiIiHjLRDbsxiVyeCLeUbLZLJVKJUr0i4iIWJNENuzGJXJ4IiIiIiIiIm54ohyeiIiIiIiIiBueyOGJiIiIiIiIuOGJHJ6IiIiIiIiIG57I4YmIiIiIiIi44YkcnoiIiIiIiIgbnsjhiYiIiIiIiLjhiRyeiIiIiIiIiBueyOGJiIiIiIiIuOGJHJ6IiIiIiIiIG57I4YmIiIiIiIi44YkcnoiIiIiIiIgbnsjhiYiIiIiIiLjhiRyeiIiIiIiIiBueyOGJiIiIiIiIuOH5/wM7C/uMkVIFBwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## parameter estimation plots\n",
    "\n",
    "# create figure\n",
    "fig = plt.figure(figsize=(10, 7))\n",
    "plt.style.use('dark_background')\n",
    "\n",
    "# setting row and column variables\n",
    "rows = 1\n",
    "columns = 2\n",
    "\n",
    "# reading images\n",
    "phys1 = cv2.imread('./phys_surr_2d/params_plot_phys_surr_2d_25000_0_1.png')\n",
    "physFree1 = cv2.imread('./phys_full_2d/params_plot_phys_full_2d_25000_0_1.png')\n",
    "\n",
    "# Add subplots corresponding to Phys with Surrogates\n",
    "fig.add_subplot(rows, columns, 1)\n",
    "plt.imshow(phys1)\n",
    "plt.axis('off')\n",
    "plt.title(\"Phys. with Surrogate\")\n",
    "\n",
    "# Add subplots corresponding to Phys without Surrogate\n",
    "fig.add_subplot(rows, columns, 2)\n",
    "plt.imshow(physFree1)\n",
    "plt.axis('off')\n",
    "plt.title(\"Phys. without Surrogate\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "2831ac40",
   "metadata": {},
   "source": [
    "As shown in the picture, use of the full model or its surrogate lead to the same posterior distribution.\n",
    "\n",
    "For more information, please refer to our [paper](https://arxiv.org/abs/2307.04675) *Appendix B. Detailed numerical benchmarks* where a few more examples are discussed."
   ]
  }
 ],
 "metadata": {
  "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}